bovine viral diarrhea virus

http://labibliotecademaverick.blogspot.com/

http://labibliotecademaverick.blogspot.com/

Bovine Viral Diarrhea VirusDiagnosis, Management, and Control

2550_Goyal_Bovine Viral Diarrhea Virus_FM 3/18/05 9:46 AM Page i

Bovine Viral Diarrhea VirusDiagnosis, Management, and Control

Edited by Sagar M. Goyal and Julia F. Ridpath

2550_Goyal_Bovine Viral Diarrhea Virus_FM 3/18/05 9:46 AM Page iii

Sagar M. Goyal, BVSc, MSC, PhD, Professor, De-partment of Veterinary Population Medicine, Collegeof Veterinary Medicine, University of Minnesota, St.Paul.

Julia F. Ridpath, PhD, Virus and Prion Diseases ofLivestock Research Unit, NADC/ARS/USDA, Ames,Iowa.

©2005 Blackwell PublishingAll rights reserved

Blackwell Publishing Professional2121 State Avenue, Ames, Iowa 50014, USA

Orders: 1-800-862-6657Office: 1-515-292-0140Fax: 1-515-292-3348Web site: www.blackwellprofessional.com

Blackwell Publishing Ltd9600 Garsington Road, Oxford OX4 2DQ, UKTel.: +44 (0)1865 776868

Blackwell Publishing Asia550 Swanston Street, Carlton, Victoria 3053, AustraliaTel.: +61 (0)3 8359 1011

Authorization to photocopy items for internal or per-sonal use, or the internal or personal use of specific

clients, is granted by Blackwell Publishing, providedthat the base fee of $.10 per copy is paid directly to theCopyright Clearance Center, 222 Rosewood Drive, Dan-vers, MA 01923. For those organizations that have beengranted a photocopy license by CCC, a separate systemof payments has been arranged. The fee code for usersof the Transactional Reporting Service are ISBN-13:978-0-8138-0478-1; ISBN-10: 0-8138-0478-7/2005$.10.

First edition, 2005

Library of Congress Cataloging-in-Publication Data

Bovine viral diarrhea virus : diagnosis, management,and control / edited by Sagar M. Goyal and Julia F.Ridpath.— 1st ed.

p. cm.Includes bibliographical references and index.ISBN 0-8138-0478-7 (alk. paper)1. Bovine viral diarrhea virus. 2. Bovine viral diar-

rhea. I. Goyal, Sagar M., 1944- II. Ridpath, Julia F.

SF967.M78B68 2005636.2�08963427—dc22

2004029262

The last digit is the print number: 9 8 7 6 5 4 3 2 1

To our families: Krishna, Vipin, Kavitha, Dinesh, Sarina,Harold, Lance, Reid, and Grant.

2550_Goyal_Bovine Viral Diarrhea Virus_FM 3/18/05 9:46 AM Page iv

Contents

Contributors viiPreface ix

1 Introduction and History 3Dirk Deregt

2 Risk Assessment 35Hans Houe

3 Classification and Molecular Biology 65Julia F. Ridpath

4 Virus Replication 81S. K. Hietala and B. M. Crossley

5 Virus Transmission 91Mark C. Thurmond

6 Clinical Features 105James F. Evermann and George M. Barrington

7 Pathogenesis 121E. M. Liebler-Tenorio

8 Reproductive Disease and Persistent Infections 145Kenny V. Brock, Daniel L. Grooms, and M. Daniel Givens

9 Immunity and Immunosuppression 157Sanjay Kapil, Paul Walz, Melinda Wilkerson, and Harish Minocha

10 Hosts 171Trevor R. Ames

11 Interactions of Virus and Host 177John D. Neill

v

2550_Goyal_Bovine Viral Diarrhea Virus_FM 3/18/05 9:46 AM Page v

12 Diagnosis 197Sagar M. Goyal

13 Vaccines 209Robert W. Fulton

14 Management Systems and Control Programs 223Robert L. Larson

15 Conclusions and Future Research 239Julia F. Ridpath and Sagar M. Goyal

Index 245

vi Contents

2550_Goyal_Bovine Viral Diarrhea Virus_FM 3/18/05 9:46 AM Page vi

Trevor R. AmesDepartment of Veterinary Population MedicineCollege of Veterinary MedicineUniversity of MinnesotaSt. Paul, MN

George M. BarringtonDepartment of Veterinary Clinical SciencesCollege of Veterinary MedicineWashington State UniversityPullman, WA

Kenny V. BrockDepartment of PathobiologyCollege of Veterinary MedicineAuburn UniversityAuburn, AL

B. M. CrossleyCalifornia Animal Health and Food Safety

LaboratoryUniversity of CaliforniaDavis, CA

Dirk DeregtAnimal Diseases Research Institute Canadian Food Inspection AgencyLethbridge, Alberta, Canada

James F. EvermannDepartment of Veterinary Clinical Sciences andWashington Animal Disease Diagnostic LaboratoryCollege of Veterinary MedicineWashington State UniversityPullman, WA

Robert W. FultonDept of Veterinary PathobiologyCollege of Veterinary MedicineOklahoma State UniversityStillwater, OK

M. Daniel GivensDepartments of Pathobiology & Clinical SciencesCollege of Veterinary MedicineAuburn UniversityAuburn, AL

Sagar M. GoyalDepartment of Veterinary Population MedicineCollege of Veterinary MedicineUniversity of MinnesotaSt. Paul, MN

Daniel L. GroomsDepartment of Large Animal Clinical SciencesCollege of Veterinary MedicineMichigan State UniversityEast Lansing, MI

S. K. HietalaCalifornia Animal Health and Food Safety

LaboratoryUniversity of CaliforniaDavis, CA

Hans HoueDepartment of Large Animal SciencesSection for Veterinary EpidemiologyThe Royal Veterinary and Agricultural UniversityFrederiksberg, Denmark

Sanjay KapilDepartment of Diagnostic Medicine-PathobiologyKansas State UniversityManhattan, KS

Robert L. LarsonVeterinary Extension and Continuing EducationUniversity of MissouriColumbia, MO

E. M. Liebler-TenorioFriedrich-Loeffler InstitutFederal Research Center for Animal Health Jena, Germany

vii

Contributors

2550_Goyal_Bovine Viral Diarrhea Virus_FM 3/18/05 9:46 AM Page vii

Harish MinochaDepartment of Diagnostic Medicine-PathobiologyKansas State UniversityManhattan, KS

John D. NeillNational Animal Disease CenterAgricultural Research ServiceU.S. Department of AgricultureAmes, IA

Julia F. Ridpath National Animal Disease CenterAgricultural Research ServiceU.S. Department of AgricultureAmes, IA

Mark C. ThurmondDepartment of Veterinary MedicineCollege of Veterinary MedicineUniversity of CaliforniaDavis, CA

Paul WalzDepartment of Clinical SciencesCollege of Veterinary MedicineKansas State UniversityManhattan, KS

Melinda WilkersonDepartment of Diagnostic Medicine-PathobiologyKansas State UniversityManhattan, KS

viii Contributors

2550_Goyal_Bovine Viral Diarrhea Virus_FM 3/18/05 9:46 AM Page viii

The disease caused by bovine viral diarrhea virus(BVDV) was first described in 1946, along withvirus isolation from sick cattle. In the intervening 60years, many important advances have been made inunderstanding this virus and the disease it produces.These developments include the recognition ofBVDV biotypes and genotypes; development ofmonoclonal antibodies (Mabs) to study strain varia-tion; defining the mechanism of development of per-sistent infections, mucosal disease, and immune tol-erance; BVDV-induced immunosuppression; thediscovery of problems caused by the presence ofBVDV and anti-BVDV antibodies in fetal bovineserum used in the production of cell cultures; se-quencing of viral genome; and the development ofrapid methods for virus detection. These achieve-ments have culminated in the development of con-trol and eradication programs in several Scandina-vian countries with some successes. Many other

countries, including the U.S., are contemplating era-dication and/or control programs.

In spite of these developments and several sym-posia dedicated to discussing the pathogenesis,transmission, diagnosis, and molecular virology ofthe virus, BVDV and BVDV-induced diseases arenot completely understood, as exemplified by theappearance of severe hemorrhagic disease caused byBVDV 2 in the early 1990s in Canada and then inother countries. It is with this in mind that we haveattempted to collate information on the current stateof knowledge on the diagnosis, management, andcontrol of the multifaceted diseases caused by BVDviruses. An internationally renowned team of ex-perts has been assembled to contribute chapters onvarious aspects of this problem.

We thank Dede Pedersen of Blackwell Publishingfor keeping the book on track.

ix

Preface

2550_Goyal_Bovine Viral Diarrhea Virus_FM 3/18/05 9:46 AM Page ix

In 1996, Cornell University held an internationalsymposium at the College of Veterinary Medicine.This event was a celebration of 50 years of bovineviral diarrhea virus (BVDV) research marked fromthe time of the first publication on BVDV, “An ap-parently new transmissible disease of cattle,” byCornell University researchers (Olafson et al.,1946). Presenting at the meeting were two Cornellpioneers: Dr. Francis Fox, one of the coauthors ofthe original publication, and Dr. James Gillespiewho, with his colleagues, had propagated the firstisolates of BVDV in tissue culture. Also presentingat this meeting was another BVDV pioneer, Dr.Bernd Liess of the Hannover Veterinary School,who recounted his early experiences and interac-tions with Cornell University researchers. Their his-torical perspectives on the discovery of the disease,the virus, and the scientific contributions of CornellUniversity are recorded (Fox, 1996; Gillespie, 1996;Liess, 1996). These personal accounts are highlyrecommended reading. The meeting was a blend of“old” and “new.” Classical (type 1) BVDV was now50 years old, but what dominated the discussion wasthe newly discovered genotype of BVDV (type 2BVDV) and the disease it causes.

From its emergence as a new virus in 1946 to thepresent day, BVDV has proven to be multifaceted inthe disease it produces and is arguably the mostcomplicated bovine virus in its pathogenesis. Duringthe past half-century and more, a multitude of ad-vances have been made in BVDV research and diag-nostics. These advances have led to the developmentof numerous diagnostic tests and the design of suc-cessful management and control strategies forBVDV. The discovery of persistent infection andhow this state is produced in cattle was key to bothour understanding of how BVDV maintains itselfand to developing rational control and eradicationstrategies.

Vaccines were first mass produced in the 1960s todeal effectively with the acute disease but were alsofound to occasionally precipitate a severe manifesta-tion of BVDV called mucosal disease (MD). Thepathogenesis of spontaneous and vaccine-inducedmucosal disease in persistently infected cattle cameto be understood by important discoveries in the1960s and 1980s. The decade of the 1980s also sawthe first complete sequencing of a BVDV genomeand the production of monoclonal antibodies to thevirus. These developments led eventually to new nu-cleic acid- and monoclonal antibody-based diagnos-tic methods.

In the 1990s, virulent strains of a new BVDVgenotype, type 2 BVDV, devastated many NorthAmerican herds and presented yet another challengefor researchers, veterinarians, and vaccine compa-nies. At the same time, several European countriesembarked on eradication programs for BVDV with-out vaccination. These programs have been highlysuccessful to date. Thus, in the beginning of the newmillennium, countries and regions with endemicBVDV will need to consider what type of controlprograms, if any, they will embark on and what thecosts and risks are. In the absence of national or re-gional programs, veterinarians and cattle producersneed to know how to apply proper management andcontrol strategies for individual herds. The objectiveof this book is to provide detailed information onour current knowledge of the virus and the diseasesit causes, the diagnostic methods available, andmanagement systems.

In writing this chapter on the history of BVDV,the author has divided the historical developmentsinto specific time periods. However, to avoid toomuch fragmentation of thought, some related devel-opments, that either overlap the time periods (earlieror later) or may be too brief to mention individuallyin their time period, may be brought together in a

3

1Introduction and History

Dirk Deregt

2550_Goyal_Bovine Viral Diarrhea Virus_Chap 01 3/18/05 9:46 AM Page 3

single context in the period in which the most signif-icant findings or a majority of the research tookplace. These events are referenced with their date ofpublication and, thus, should not be confusing to thereader. In addition to providing a history and intro-duction to the virus and disease, another objective inwriting this chapter is to provide an introduction (inhistorical context) of subjects that are further devel-oped in subsequent chapters.

THE EARLY YEARS, 1946–1969

THE VIRUS AND THE DISEASE

The new transmissible disease in cattle, described byCornell University researchers Olafson, MacCallum,and Fox (1946), was characterized by leukopenia,high fever, depression, diarrhea and dehydration,anorexia, salivation, nasal discharge, gastrointestinalerosions, and hemorrhages in various tissues.Initially, the disease was observed in Ithaca, NewYork, in a “one-cow herd” by Dr. Francis Fox, whoinitially considered the disease to be classical winterdysentery in his recounting of the event (Fox, 1996).Subsequently, outbreaks of the disease occurred inother herds in the area. In five initial herds, the mor-bidity ranged from 33–88% and mortality was 4–8%.In addition to other signs, milk production was di-minished and fetal abortions occurred 10 days to 3months following infection. Some of the cows in oneherd developed pneumonia. The severe leukopeniaseen in clinically affected animals was considered tobe indicative of a viral etiology. The attempted treat-ment involved blood transfusions, but often the ap-parently healthy donor animals of the herd had amore severe leukopenia than the clinically affectedanimals (Fox, 1996). This indicated that the donorswere infected and that BVDV infections could alsobe subclinical in nature.

The lesions of BVD resembled those of rinder-pest, an exotic disease for the U.S. However, the dis-ease observed by Olafson et al. (1946) did not be-have as rinderpest. Since the U.S. had a susceptiblecattle population, rinderpest would have presented amuch more devastating clinical picture with hightransmission and mortality rates. However, it wasthought that a mild form of rinderpest could havebeen responsible for the lesions observed and, thus,it was important to rule out this disease. Subse-quently, Walker and Olafson (1947) demonstratedexperimentally that sera from cattle that recoveredfrom BVD did not neutralize rinderpest virus andthat these recovered cattle did not show resistance toinfection with the rinderpest virus.

After the report of acute BVD in New York, a sim-ilar but more severe disease in cattle was reported inCanada by Childs (1946). This report was consid-ered by Pritchard (1963) in his review of those earlytimes as probably the first description of mucosaldisease in cattle. Mucosal disease was characterizedby fever, anorexia, depression, profuse salivation,nasal discharge; gastrointestinal hemorrhages, ero-sions, and ulcers; and severe diarrhea with wateryfeces that were sometimes mixed with blood.Lesions of the gastrointestinal tract were quantita-tively much more severe in MD than those observedin BVD. Furthermore, MD usually affected only afew animals in a herd but had a very high case fatal-ity rate. In 1953, Ramsey and Chivers (1953) re-ported on MD in the U.S. and gave the disease itsname. These authors noted that MD had some of thecharacteristics of BVD but was apparently not trans-missible experimentally, and only fever was ob-served in cattle inoculated with the agent. Thus,based on the differences in lesions of the gastroin-testinal tract, the low morbidity and high case fatal-ity rates, and the nontransmissibility of MD, it wasthought to be a disease distinct from BVD.

It was discovered in 1957 that the viral agent iso-lated in acute BVD did not cause cytopathology invitro, meaning that infected cells in culture appearednormal (Lee and Gillespie, 1957). In the same year,a cytopathic virus had been isolated from MD(which was thought at the time to be caused by a dif-ferent virus than the one causing BVD) (Underdahlet al., 1957). After the isolation of two noncyto-pathic viruses from BVD cases, including the refer-ence NY-1 strain (Lee and Gillespie, 1957), Gilles-pie et al. (1960) reported the first isolation of acytopathic strain of BVDV, designated OregonC24V. The discovery of cytopathic strains allowedthe development of serum neutralization and plaqueneutralization assays. Cytopathic strains could bemore easily studied in tissue culture than noncyto-pathic strains, and their neutralization with antiseraallowed characterization of the antigenic relatednessof viruses from cases of BVD and MD. Studies em-ploying virus neutralization (Gillespie et al., 1961;Kniazeff et al., 1961; Thomson and Savan, 1963)determined that the viral agents isolated from BVDand MD in North America and Europe were, indeed,the same and that BVD and MD were actually dif-ferent disease manifestations caused by the sameagent. Thus, several years later, the disease becameofficially known as bovine viral diarrhea-mucosaldisease (BVD-MD) (Kennedy et al., 1968).

In an early review of BVD-MD, Pritchard (1963)

4 BVDV: Diagnosis, Management, and Control


described BVD as a separate syndrome from MDand stated that BVD occurs in three forms: severeacute BVD, mild acute BVD, and chronic BVD(now recognized as a form of MD). Severe acuteBVD had been prominent a decade earlier, but by1963, a very mild clinical form of BVD was mostcommonly observed. Pritchard (1963) noted thatone of the common early clinical signs of the severeacute form of BVD was a harsh, dry cough in manyof the animals. In some herds, lameness attributedto laminitis was also prominent. In the chronicform, affected cattle failed to grow at a normal rateor lost weight. Many of these animals became ema-ciated and developed continuous or intermittentdiarrhea.

Although BVD could be reproduced experimen-tally, MD could not. Only fever or a milder form ofBVD could be produced with the virus obtainedfrom cases of MD (Pritchard, 1963). An early studyof field cases of BVD and MD indicated that therewas an immune component to MD (Thomson andSavan, 1963). In seven cases, the cattle that becamediseased and died were serologically negative to thevirus. One animal, which was clinically ill for 2months, remained serologically negative immedi-ately prior to death when viable virus was still iso-lated from its blood. Based on these observations,Thomson and Savan (1963) suggested that cattlethat did not recover from their illness may have beenincapable of producing an immune response.

By the end of the 1960s, it had become evidentthat cattle with MD had persistent viremia and oftenfailed to produce neutralizing antibodies to the virus(Malmquist, 1968). It was also observed that fetalbovine serum frequently contained BVDV (Malm-quist 1968). This finding and the finding of BVDVinfections in newborn and 1-day-old calves were in-dicative of intrauterine infections (Bürki andGermann, 1964; Romváry, 1965; Malmquist, 1968).Thus, one mechanism proposed to explain persistentinfection, and the failure of cattle with MD to pro-duce antibodies to BVDV was the development ofimmune tolerance in these cattle during intrauterineinfection (Malmquist, 1968). Exposure to BVDVbefore the age of immune competence could lead tofailure of the fetus to recognize the virus as foreignand, thus, a failure to produce antibodies to thevirus. This would result in the eventual birth ofcalves infected with BVDV but without the capabil-ity to clear the infection. An alternative mechanismproposed by Malmquist (1968) for the lack of neu-tralizing antibodies in cattle with MD was the de-struction of immunologically competent cells by the

virus. It was noted at the time that, if the operatingmechanism was immune tolerance, the lack of im-mune response would be expected to be BVDV-specific, whereas if immune cell destruction was op-erating, there would also be a depression of the im-mune response to unrelated antigens (Segre, 1968).The hypothesis of immune tolerance was eventuallyproven in the 1980s, a decade in which the precipi-tating event in MD was also identified and in whichMD was experimentally reproduced.

Fetal abortions were associated with BVDV in-fection from the time of the first outbreaks of BVD(Olafson et al., 1946). Evidence of a causative rolein abortion included the isolation of noncytopathicBVDV from aborted fetuses and the occurrence ofabortions following experimental exposure and afternatural outbreaks (Olafson et al., 1946; Baker et al.,1954; Gillespie et al., 1967). Cerebellar hypoplasiaand ocular defects (cataracts, retinal degeneration,and optic neuritis) in newborn calves were also ob-served after BVDV infection of pregnant cows.These BVDV-induced congenital defects were firstreported by Ward et al. (1969). For the dams of threecalves with cerebellar hypoplasia, Kahrs et al.(1970) reported that infection occurred at an esti-mated 134–183 days of gestation. Experimentally,cerebellar and ocular defects were produced incalves by inoculation of their dams with a noncyto-pathic BVDV at 79–150 days of gestation (Scott etal., 1973).

Early in the 1960s, it became established thatBVDV was antigenically related to hog choleravirus, now more commonly known as classical swinefever virus (CSFV) (Darbyshire, 1962). Soon after,live BVDV vaccines were proposed for immuniza-tion of swine against CSFV in the U.S., but this waslater abandoned in 1969 when the Department ofAgriculture issued a notice against their use for thispurpose (Fernelius et al., 1973). Swine had originallybeen thought to be dead-end hosts for BVDV, mak-ing the virus attractive for immunization of swine.However, this was proven to be false when BVDVwas isolated from naturally infected swine (Ferneliuset al., 1973). Later, serological evidence indicatedthat the agent causing border disease in sheep wasalso related to BVDV and CSFV (Plant et al., 1973).

DIAGNOSIS AND CONTROL

The serum neutralization test developed after thediscovery of cytopathic strains of BVDV became animportant diagnostic assay and is still widely usedtoday. Early studies in the 1960s using this testproved that BVDV was present worldwide at a high

Introduction and History 5


seroprevalence (usually about 60%) in most adultcattle populations and that the majority of BVDV in-fections were subclinical in nature.

During the 1960s, the use of cell lines such asMadin-Darby bovine kidney (MDBK) cells sim-plified the study of the virus and its diagnosis(Marcus and Moll, 1968). Also of importance wasthe development of the fluorescent antibody tech-nique (FAT) for the detection of BVDV in inocu-lated cell cultures (Fernelius, 1964). Prior to the de-velopment of FAT, identification and titration ofnoncytopathic BVDV was difficult, often involvingcalf inoculation, an interference test in which thecytopathic effect of cytopathic BVDV was inhibitedby noncytopathic virus (Gillespie et al., 1962;Diderholm and Dinter, 1966), or agar gel-diffusionprecipitin tests. With FAT, the detection of noncyto-pathic BVDV from biological specimens was mademuch easier, although the interference test was stillused experimentally as late as 1985 (McClurkin etal., 1985).

Soon after the discovery of the cytopathic BVDVstrain, Oregon C24V, Coggins et al. (1961) reportedon its attenuation by serial passages in primarybovine kidney tissue culture. This led to its massproduction as a modified-live vaccine for BVD in1964 (Peter et al., 1967). However, soon after itsuse, reports began to appear of a few sick animals insome herds following immunization. It became ap-parent that the vaccine produced an MD-like diseasein a small number of animals in a few herds (Fuller,1965; Peter et al., 1967; McKercher et al., 1968).The animals that became sick with MD-like symp-toms usually died. The discussion of possible causesof postvaccinal MD (pvMD), as it became known,included insufficient attenuation of the OregonC24V strain and contamination of the vaccine witha virulent BVDV field strain. Further, since pvMDwas first observed after the introduction of multiva-lent vaccines against both BVD and infectiousbovine rhinotracheitis (IBR), it was considered pos-sible that a synergistic effect between the twoviruses may have been responsible for the postvac-cinal syndrome. The possible role of the vaccine asa stressor in cattle already in the acute stage of in-fection with a BVDV field strain was also suggested(McKercher et al., 1968). Peter et al. (1967), how-ever, noted that animals that died from pvMD failedto develop antibodies to BVDV but could produceantibodies to IBR virus. Thus, they suggested thatcattle succumbing to MD or pvMD were uniquelysusceptible to BVDV and that this susceptibility in-volved a failure of the immune system.

BVDV was recognized as a highly contagious dis-ease that was easily spread from herd to herd(Pritchard, 1963). The original outbreaks in NewYork were described as explosive in character, inwhich practically all the animals in a herd camedown with the disease within a few days (Olafson etal., 1946). Thus, before a vaccine became available,control procedures were limited to protecting thenoninfected herd by prevention of direct and indirectcontact with infected animals.

By the mid- and late 1960s, modified-live vac-cines comprising the attenuated Oregon C24V andNADL strains were in use (Bittle, 1968; Gutekunst,1968). The latter strain was isolated in 1962 at theNational Animal Disease Laboratory and attenuatedin porcine kidney cell culture. Although modified-live vaccines were considered efficacious in prevent-ing acute BVD, their use was thought to be some-what risky due to the occurrence of pvMD. The riskof pvMD was somewhat a matter of perception andpersonal experience. Bittle (1968), describing theOregon C24V vaccine, stated that the incidence ofpostvaccinal problems reported to the USDA com-pared to the number of doses used had been ex-tremely small (less than 1 in 10,000) and he encour-aged its use. Likewise, Gutekunst (1968) stated that,following 8 months of field use of 350,000 doses ofthe NADL vaccine, no reports of postvaccinal reac-tions had been received. In contrast, a commentatorat the meeting at which Drs. Bittle and Gutekunstspoke remarked that he had personally observed 40cases of pvMD in a 3-month period in approxi-mately 1,000 vaccinated cattle (Clark, 1968). Subse-quently, it was discovered that pvMD occurs only inpersistently infected cattle and although these ani-mals comprise a small portion of the total cattle pop-ulation, they could comprise a significant portion ofanimals in an individual herd.

The modified-live vaccines were contraindicatedfor use in pregnant animals because of the possi-bility of inducing abortions, and thus, emphasiswas placed on vaccination of calves and heifers.Kahrs (1971) also advocated protecting naive cat-tle from contact with potentially infected cattle,particularly those from auction markets or shippingcenters and, probably because of the risk of pvMD,considered this the “best control method.” Forfeedlot cattle, Fuller (1965) stated that it was pooreconomy to vaccinate already stressed cattle imme-diately upon arrival to feedlots. Instead, he recom-mended acclimatizing calves to their new sur-roundings for about 3 weeks before vaccinatingthem against BVDV.



THE DECADE OF THE 1970s

PERSISTENT INFECTIONS IN SICK ANDAPPARENTLY HEALTHY CATTLE

By the 1970s, it was established that calves withcongenital BVDV infections were unthrifty and usu-ally died within a few months, and that survivingcalves often suffered from chronic disease, werepersistently infected with the virus, and were defi-cient in serum neutralizing antibodies againstBVDV (Bürki and Germann, 1964; Malmquist1968; Johnson and Muscoplat, 1973). These obser-vations were made in sick and unthrifty animals. In1978, Coria and McClurkin (1978) reported on thepersistent infection and immune tolerance of an ap-parently healthy bull. The bull was continuouslyviremic and a noncytopathic BVDV was isolatedfrom its blood leukocytes repeatedly from birth to2.5 years of age. The virus was also isolated repeat-edly from the bull’s semen. The bull remainedseronegative for BVDV antibodies during this time.When challenged with the equivalent of three dosesof a killed vaccine, it failed to produce a significantimmune response. It was suggested that the bull hadacquired the infection early in gestation before thedevelopment of its immune system and had thus ac-quired a specific immune tolerance to BVDV.

A year later, McClurkin et al. (1979) describedthe reproductive performance of four healthy cattlepersistently infected with BVDV: the bull previouslyidentified and three newly identified pregnant cows.As an aid to the identification of apparently healthy,persistently infected cattle in a herd, they noted thatcattle that remain seronegative while in contact withseropositive cattle were likely to be immune-tolerantand persistently infected. They further noted that,after three vaccinations of the herd, all cattle exceptthese three cows had serum neutralizing antibodytiters of 1:16 or greater against BVDV. As in the per-sistently infected bull, a noncytopathic BVDV wasconsistently isolated from the three persistently in-fected cows, and these cows gave birth to calves thatbecame ill: Two died within the first week postpar-tum and one calf was euthanized at a few weeks ofage. A noncytopathic BVDV was isolated from theblood leukocytes of all three calves indicating ma-ternal transmission.

When seropositive cows were bred by the persist-ently infected bull, normal calves were born butservices per conception averaged a high 2.3. For fiveseronegative heifers bred by the bull, services perconception averaged 2.0. All seronegative heifers se-roconverted to BVDV and had high antibody titers

(�1:128) 6 weeks after breeding. Four heifers gavebirth to normal calves and one heifer aborted at 6months of gestation but no BVDV was isolated fromthe fetus. None of the calves produced by eithergroup showed evidence of intrauterine infection.However, the study indicated that BVDV may playa role in repeat breeding problems. It was suggestedthat losses due to repeat breeding and from neonataldisease might be prevented by ensuring that all cat-tle have high antibody titers to BVDV before breed-ing (McClurkin et al., 1979).

Also of interest in this study was the observationthat BVDV antibody titers in the herd (38% of theanimals had titers >1:256) were generally muchhigher than expected (usually 1:16 to 1:64) for cat-tle vaccinated with the killed vaccine. It was sur-mised that virus shedding by the persistently in-fected animals had constantly challenged the othercattle in the herd and boosted their antibody titers.The idea that the level of BVDV antibodies in a herdcould be used to predict which herds contained per-sistently infected animals was later developed in the1990s and recently evaluated for herds of unknownBVDV status (Pillars and Grooms, 2002), as will bediscussed later in this chapter.

Lesions in the persistently infected, healthy cattlewere found to be microscopic, primarily in the brainand kidney (Cutlip et al., 1980). However, immuno-fluorescence staining of tissues demonstrated wide-spread distribution of viral antigen in brain andspinal cord neurons, renal glomeruli, renal tubules,lymph nodes, spleen, small intestine crypts, testicu-lar tubules, and endothelial cells.

DIAGNOSIS AND CONTROL

The observation that fetal bovine sera frequentlycontained BVDV and neutralizing antibodiesagainst BVDV (Kniazeff et al., 1967; Malmquist,1968) was an increasing concern in the 1970s forcell culture work because contaminated culturescould have undesirable consequences for researchand vaccine production. Tamoglia (1968) found that8% of licensed live IBR vaccines were contaminatedwith BVDV, raising concerns that such vaccinesmight give rise to fetal abortions. Experimentalstudies, like those examining the effects of cyto-pathic BVDV, could be compromised if cell culturesused to propagate the cytopathic virus were contam-inated with noncytopathic BVDV. Commercial fetalbovine sera were contaminated with noncytopathicBVDV as a result of pooling sera from infected andnoninfected fetuses. Commonly, sera of 500 fetuses



were pooled and, although the incidence of fetal in-fection was unknown at the time, Nuttall et al.(1977) calculated that contamination of a batch offetal bovine serum required only a 0.2% incidenceof infection. At the time, the interference test (Gil-lespie et al., 1962), which relied on the visible inhi-bition of cytopathic BVDV, was used to screen fetalbovine sera and bovine cells for noncytopathicBVDV. Nuttal et al. (1977) believed that the interfer-ence test was not sensitive enough to detect low-level contamination of fetal bovine serum with non-cytopathic BVDV and advocated the use of FAT forregular screening of sera and cells for noncytopathicvirus.

For serological diagnosis of BVDV infection,Lambert et al. (1974) recommended the use of theserum neutralization assay on paired serum samplescollected 2 or 3 weeks apart, with a rising titer indi-cating active infection. To prevent BVD in neonatalcalves, they recommended vaccination of openheifers and cows 30–60 days prior to breeding andthe consumption of antibody-rich colostrum bycalves.

Because of the concern for pvMD, abortions, andin utero infections, a number of studies were done inthe early and mid-1970s on the efficacy of killedvaccines for BVDV. However, these efficacy studieswere often limited in scale and involved challengewith homologous virus. Lambert et al. (1971) evalu-ated a killed BVDV-NADL vaccine and found it ef-ficacious against homologous virus challenge incalves aged 5–11 months. McClurkin et al. (1975)examined the use of killed vaccines using inacti-vated cytopathic NADL and Singer strains to pre-vent fetal infection and found these vaccines effica-cious. However, these researchers used homologouschallenge, and a heterologous or noncytopathicBVDV challenge was not attempted. Challenge withnoncytopathic virus would prove to be essential inevaluating fetal protection when it was later discov-ered that only noncytopathic viruses caused persist-ent infections.


EXPERIMENTAL PRODUCTION OFPERSISTENT INFECTION AND MUCOSALDISEASE

In 1984 and 1985, a number of important advanceswere made in BVDV research. McClurkin et al.(1984) described the production of persistently in-fected, immune-tolerant calves in five experimentsinvolving 44 cows in the first trimester of pregnancy

(42–125 days). Cows or their fetuses were inocu-lated directly with one of five different BVDV iso-lates. Four of the isolates were noncytopathic strainsand the fifth was the cytopathic NADL strain. For 38pregnant cows or fetuses inoculated with noncyto-pathic strains, there were 10 abortions, 1 stillborncalf, 4 weak or unsteady calves, and 23 calves thathad a normal and vigorous appearance at birth. Allweak calves and 22 of the 23 normal appearingcalves were persistently infected and seronegativeand were the result of inoculations occurring at days42–125 of gestation. One calf infected at 125 daysof gestation was immune-competent, virus-negative,and seropositive at birth.

Of interest, but not fully appreciated at the time,was the failure by McClurkin et al. (1984) to pro-duce any persistently infected calves with the cyto-pathic NADL strain. Prior to this study, Done et al.(1980) had infected 15 pregnant cows at 100 days ofgestation with a mixture of 10 cytopathic strains ofBVDV. Virus was recovered from eight live-borncalves, but in each case these viruses were noncyto-pathic BVDV. These researchers also did not appearto fully recognize the significance of this finding,stating that “the absence of cytopathogenicity in allreisolates of virus is remarkable, but perhaps nomore than a reminder of the genetic variablity ofviruses in general and of the pestiviruses in particu-lar.” It is almost certain that noncytopathic BVDVcontaminated their cytopathic virus stocks used forinfection. Later, Brownlie et al. (1989) infectedpregnant cattle with a cytopathic virus and could notproduce persistent infection. Thus, it became gener-ally accepted that only noncytopathic BVDV couldproduce persistent infections.

McClurkin et al. (1984) also followed the fate ofthe persistently infected calves they had produced.All four weak calves either died or were euthanizedwithin 4 months after birth. Of the 22 apparentlyhealthy calves, 6 developed diarrhea and/or pneu-monia within 5 months of birth and died, and 1 be-came unthrifty and remained small. Ten of the ap-parently healthy calves remained healthy at 6months of age or as yearlings (most of these calveswere then used in other experiments). Three animalsremained healthy as 2-year-olds and were bred. Twoof the persistently infected cows produced appar-ently healthy, persistently infected calves, whereasthe third cow lost her calf and developed MD at 28months of age. This study demonstrated severalcharacteristics of persistent infection: that persist-ently infected calves may be born weak or appar-ently normal; that some may live to breeding age;



and that persistently infected families can arise bybreeding persistently infected cattle.

Soon after the study by McClurkin et al. (1984),Brownlie et al. (1984) and Bolin et al. (1985c) ex-perimentally reproduced MD in persistently infectedcattle. Brownlie et al. (1984) noted that whilehealthy, persistently infected cattle were infectedwith only noncytopathic BVDV, both noncytopathicand cytopathic BVDV could be isolated from per-sistently infected cattle that were clinically ill withMD. The latter finding was also observed byMcClurkin et al. (1985). From their observations,Brownlie et al. (1984) developed a hypothesis forthe induction of MD, which stated that cattle be-come persistently infected with noncytopathicBVDV after in utero infection and postnatally suc-cumbed to MD when superinfected with a cyto-pathic BVDV. To test the hypothesis, Brownlie et al.(1984) used a cytopathic isolate from an animal suf-fering from MD to inoculate two healthy persist-ently infected herdmates. Both animals came downwith MD, supporting the hypothesis. In anotherstudy, Bolin et al. (1985c) inoculated persistently in-fected cattle with noncytopathic or cytopathicBVDV. The cattle inoculated with noncytopathicvirus did not develop clinical signs of disease,whereas MD developed in all cattle inoculated withcytopathic BVDV.

Although pvMD was known to be a relativelycommon phenomenon, failure to consistently induceMD in persistently infected cattle with cytopathicBVDV vaccines (Bolin et al., 1985b) suggested thatthe induction of MD was somewhat more compli-cated than simple superinfection with a cytopathicBVDV in an animal persistently infected with non-cytopathic BVDV. Subsequent studies indicated thatthe noncytopathic and cytopathic viruses (called avirus pair) from individual MD cases were antigeni-cally similar. Howard et al. (1987) compared fivevirus pairs from separate MD outbreaks and foundthat virus pairs from the same outbreak were anti-genically indistinguishable when tested in cross-neutralization tests with antisera. Corapi et al.(1988) using a panel of monoclonal antibodies alsoshowed that noncytopathic and cytopathic viral pairsfrom MD had a high degree of antigenic similarity.This led to the conclusion that in natural outbreaksof MD the likely origin of the cytopathic BVDV wasvia mutation of the noncytopathic BVDV infectingthe persistently infected animal. Thus, the hypothe-sis for the induction of MD was refined to includeantigenic similarity: MD was induced in a persist-ently infected animal by “superinfection” with a cy-

topathic BVDV with antigenic similarity to the non-cytopathic BVDV. Superinfection could occur bymutation in spontaneous MD or, in the case ofpvMD, by vaccination with a vaccine virus antigeni-cally similar to the noncytopathic BVDV infectingthe persistently infected animal.

The above hypothesis suggested that inoculationof persistently infected animals with a cytopathicBVDV antigenically different (heterologous) fromthe noncytopathic persisting virus would not resultin MD; but rather, the animal would produce anti-bodies to the cytopathic virus and clear the infec-tion. This apparently was the reason for the failureof superinfection with cytopathic BVDV (e.g., byvaccination) to produce MD consistently. In supportof this hypothesis, Moennig et al. (1990) showedthat persistently infected animals superinfected withclosely related cytopathic BVDV developed MDwithin 14 days of infection but those that were su-perinfected with heterologous cytopathic BVDV didnot develop MD within a 2–3-week time frame.Instead, these animals developed antibodies to thesuperinfecting virus. Other studies, however,showed that the inoculated cytopathic BVDV couldsometimes be heterologous and yet precipitate MDin a persistently infected animal. Westenbrink et al.(1989) inoculated 14 clinically healthy, persistentlyinfected animals with three heterologous cytopathicviruses. Twelve of these animals developed MD,some within the expected time frame for MD of 2–3weeks postinoculation, and several others after sev-eral months (so called late-onset MD). The actualrange when MD began was 17–99 days. Interest-ingly, neutralizing antibodies were produced againstthe inoculated cytopathic virus but, as expected, notto the persisting noncytopathic virus. However, in10 of 12 cases, the neutralizing antibodies did notneutralize the cytopathic virus recovered at necropsyfrom the intestines. This suggested that mutationalchanges had occurred and, as a result, cytopathicvirus antigenically similar to the persisting noncyto-pathic virus arose and induced MD.

Shimizu et al. (1989) also produced MD in per-sistently infected cattle with inoculation of heterolo-gous cytopathic BVDV. Similar to the findings ofWestenbrink et al. (1989), they found that the cyto-pathic viruses recovered from blood early after in-fection were antigenically similar to the challengecytopathic virus but that the cytopathic viruses iso-lated from the carcasses at necropsy were antigeni-cally different from the challenge virus but similarto the noncytopathic persisting virus. Neutralizingantibodies were produced to both the challenge



virus and cytopathic virus isolated early in infectionbut not to the persisting noncytopathic virus nor tothe cytopathic virus recovered from the carcasses atnecropsy. These results again suggested that muta-tional changes or selection of antigenic variants inthe challenge virus occurred, which resulted in a ho-mologous cytopathic virus responsible for precipi-tating MD. However, subsequent studies wouldshow that the arising homologous cytopathic viruswas often the result of RNA recombination betweenthe cytopathic challenge virus and the persistingnoncytopathic virus.

ACUTE AND PERSISTENT INFECTIONS IN THEBULL

Whitmore et al. (1978) examined BVDV sheddingin the semen of nine bulls after acute infection andfound detectable virus in 4 of 70 semen samples.They were also able to recover virus from the testi-cle of one bull. The positive samples were obtainedin the first 10 days postinoculation from four of thenine bulls but the amount of virus in the semen wasnot quantified. Treatment of the bulls with dexam-ethasone at 28 or 56 days postinoculation did not re-sult in additional virus shedding in semen. More re-cently, Kirkland et al. (1991) studied five bullsduring an acute infection with BVDV. Semen sam-ples were collected between 7 and 14 days after in-fection on four occasions from each bull. Virus wasisolated from three bulls and from 9 of 12 batches ofsemen from these bulls. Titers of virus were lowranging from 5–75 TCID50/ml of semen. They alsonoted that acute infection did not appear to affect thequality of semen.

Subsequent to Coria and McClurkin’s (1978) re-port, discussed earlier, Barlow et al. (1986) reportedon virus shedding in a persistently infected bull.Virus was shed at high titer (104.0–105.5 TCID50/0.2ml) in the semen. This animal had a sight defect,which was discovered postmortem to be due to reti-nal atrophy, but otherwise the bull appeared normal.Upon examination, the semen had been of acceptablequality and the bull had successfully sired a calf.Thus, had the bull not had a vision defect, the risk itposed of transmitting BVDV might have gone unde-tected. Revell et al. (1988) reported on two bulls thatwere persistently infected. In contrast to the findingsof Barlow et al. (1986), the quality of the semen ofthese bulls was consistently poor, as measured bydensity and motility. Gross abnormalities of spermheads (“collapsed heads”) was seen in 28–45% ofspermatozoa from one of the bulls (Cy 105). Paton etal. (1990) inseminated six seronegative and six

preimmunized, seropositive heifers with semen fromthis bull (Cy 105). Both groups had poor rates ofconception but eventually all but one heifer con-ceived after repeated inseminations. Eleven appar-ently normal calves were born and none were persist-ently infected. Paton et al. (1989) also used virusisolated from the serum of Cy 105 to infect fourFreisen bulls. Virus was isolated from the semen ofone of the bulls from four collections taken 7–14days postinoculation. The highest titer obtained was101.4 TCID50/ml. Semen quality of the bull wasfound to deteriorate following infection and showeda reduction in both sperm density and motility.

Meyling and Jensen (1988) were the first to reportthe production of a persistently infected calf siredfrom the semen of a persistently infected bull.Twelve seronegative heifers were inseminated withsemen from this bull containing 105–107.5 TCID50/ml of BVDV. All 12 heifers became infected as indi-cated by seroconversion and all heifers gave birth toclinically normal calves. Only1 of the 12 calves waspersistently infected. Later, Kirkland et al. (1994)described the results of widespread field use ofsemen from a persistently infected bull. Approxi-mately 600 doses of semen had been distributed to97 dairy farms for sire evaluation purposes and 162cows were inseminated. The first service conceptionrate was only 38%. A subsequent study of 61 calvessired by the bull revealed that only two of thesecalves were persistently infected. From these stud-ies, it is apparent that production of persistently in-fected offspring via BVDV-contaminated semen isan uncommon event.

RESPIRATORY DISEASE

During the 1980s, research into the contribution ofBVDV to respiratory disease was initiated. This re-search followed the successful reproduction of“shipping fever” pneumonia in cattle with aerosalsof infectious bovine rhinotracheitis virus andMannheimia haemolytica in the late 1970s (Jerichoand Langford, 1978). Shipping fever pneumonia, acause of significant mortality in cattle in feedlots,was thought to be precipitated by the stress of trans-porting cattle from the farm to the feedlot. Thepathogenesis of the disease is thought to involvedual infection of a pneumotropic virus and a colo-nizing bacterial species, most often M. haemolytica(Yates, 1982). As early as the first description ofBVDV herd outbreaks in 1946, the virus has beenimplicated in causing at least mild respiratory dis-ease with symptoms of nasal discharge and cough-ing. Potgieter et al. (1984, 1985) performed several



experiments examining the ability of BVDV to in-duce respiratory disease. BVDV infection of calveswithout subsequent bacterial superinfection resultedin mild respiratory tract lesions characterized bysmall, scattered areas of interstitial pneumonia in-volving 2–7% of the total lung volume. Infection ofcalves with M. haemolytica alone produced local-ized lesions involving about 15% of the lung. Incontrast to the mild disease produced by theseagents when given individually, inoculation ofBVDV followed by M. haemolytica produced a se-vere fibrinopurulent bronchopneumonia and pleuri-tis involving 40–75% of the lung volume. Potgeiter(1985) also found that BVDV strains may differ intheir pneumopathogenicity. That BVDV infectioncan precipitate severe respiratory disease is also sup-ported by the observation that pneumonia, includingshipping fever–like fibrinous pneumonia, was acommon finding in the severe type 2 BVDV out-breaks that occurred in Ontario, Canada, in the mid-1990s (Carman et al., 1998, van Dreumel, 2002).

THROMBOCYTOPENIA

Thrombocytopenia with hemorrhage associatedwith BVDV virus infection was first reported in1987 within a summary of case reports for dairyherds in the northeastern United States (Perdrizet etal., 1987). In one report from 1985, 6 of 30 first-calfheifers developed a high fever, had bloody and mu-coid diarrhea, and died within 2 weeks. Rebhun etal. (1989) reviewed case records of cattle admittedto the College of Veterinary Medicine at CornellUniversity for the years 1977–1987 and found thatthrombocytopenia was reported in about 10% ofclinically acute BVDV infections in adult cattle.Clinical signs included hemorrhages, red or orangebloody diarrhea, epistaxis, and abnormal bleedingfrom injection sites. Hemorrhages associated withBVDV infection in young veal calves were also ob-served with increasing frequency in the late 1980s inthe northeastern United States (Corapi et al., 1990b).

Corapi et al. (1989) experimentally reproducedthrombocytopenia in young calves with a BVDVisolate (CD-87) recovered from a severe outbreak.This outbreak involved 50% of a milking herd of100 holsteins in New York in which 20 animals died.Of the eight calves inoculated with the CD-87 iso-late, three developed severe thrombocytopenia(�5,000 platelets/µl). Two of these three calves de-veloped hemorrhages when their platelet counts fellto 2,000/µl of blood or less for a period of 24 hoursor longer. Hemorrhages were observed on the scleraof the eyes, inner surface of the eyelids, mucosal

surfaces of the cheeks, lower gingiva, tongue, andsoft palate. Both calves had prolonged bleedingfrom venepuncture sites. Internal hemorrhages werealso observed on the surfaces of various organs ofone calf at necropsy. In a later report, Corapi et al.(1990b) inoculated 8 veal calves with the CD-87isolate and 10 veal calves with CD-89, a BVDV iso-late recovered from a veal calf on a farm in Pennsyl-vania where hemorrhages were noticed in severalcalves. Ten in-contact calves were included in theexperiment. During the experiment, virus was iso-lated from all calves including in-contact calves.Severe thrombocytopenia was observed in 12calves, and 11 of these developed hemorrhages.Calves that had a pre-exposure virus-neutralizingantibody titer of >1:32 to the Singer strain of BVDVdid not develop severe thrombocytopenia. Fivecalves died during the course of the experiment, fourof which exhibited hemorrhages, and the others re-covered. Hemorrhages were observed in these ex-periments when platelet counts decreased below5,000 platelets/µl. Although the majority of severelythrombocytopenic calves recovered, there was noway to determine beforehand the calf’s fate sincethose that died often appeared to be in relativelygood physical condition just hours prior to death(Corapi et al., 1990b). A few years later, it was dis-covered that hemorrhagic syndrome, as the diseasecame to be known, was caused by a new type ofBVDV, genetically distinct from the classicalviruses used in BVD vaccines.

ADVANCES IN MOLECULAR BIOLOGY

In the late 1980s, significant advances were made inthe molecular biology of BVDV with the first ge-nomic sequencing of BVDV strains, the finding of amarker protein for cytopathic BVDV, and the firstevidence of RNA recombination in cytopathicstrains of BVDV. Monoclonal antibodies to the viruswere also first produced in the late 1980s as reagentsfor protein studies and diagnostic test methods (seethe following section, “Diagnosis”).

The first BVDV strains to be sequenced were twocytopathic strains, the North American NADL strainand the European Osloss strain (Collett et al.,1988b; Renard et al., 1987). The NADL sequenceshowed that the RNA genome of BVDV has onelong open reading frame (ORF). Thus, proteins areproduced by cotranslational and posttranslationalprocessing of a polyprotein (Collett et al., 1988a).The original Osloss sequence was at first shown tocontain two ORFs but was later corrected to consistalso of a single ORF (de Moerlooze et al., 1993).



The elucidation of the genomic organization ofBVDV and the pestivirus genus led to their taxo-nomic reclassification from the Togaviridae to theFlaviviridae.

At about the same time, protein studies revealedthat the two biotypes of the virus, cytopathic andnoncytopathic, could be distinguished at a molecu-lar level. It was found that cytopathic strains ofBVDV produce in infected cells one additionalnonstructural protein not observed in cells infectedwith noncytopathic BVDV (Donis and Dubovi,1987; Pocock et al., 1987). The protein, which is amarker protein of cytopathic BVDV, is designatedp80 or NS3. This protein is actually a smaller ver-sion of a larger nonstructural protein, p125 or NS2-3, that is present in all BVDV-infected cells. Severalmechanisms leading to expression of NS3 werelater found to occur: The NS3 protein can be gener-ated by proteolytic cleavage of NS2-3 or generatedas the result of genetic duplications or deletions inthe genomes of some cytopathic BVDVs (Meyersand Thiel, 1996).

Meyers et al. (1989) compared the BVDV Oslossand NADL genomic sequences with that of the pes-tivirus CSFV and found insertions in the NS2 geneof these cytopathic strains of BVDV. The Osloss in-sertion was 228 nucleotides long whereas the inser-tion of NADL was 270 nucleotides. The insertionswere found to be different and that of NADL wasunidentified. However, remarkably, the Osloss inser-tion was found to be derived from the cellular genecoding for ubiquitin. These findings led to a molec-ular model for pathogenesis of MD: In persistentlyinfected animals, the noncytopathic virus mutates toa cytopathic virus by the incorporation of cellularsequences during a recombination event. Subse-quently, it was found that genetic recombinationcould also occur between BVDV genomes as evi-denced by insertions of viral sequences in thegenomes of some cytopathic viruses.

DIAGNOSIS

Meyling (1984) described a microisolation test fordetection of BVDV in serum samples that used im-munoperoxidase staining of viral antigens ratherthan immunofluorescence. The immunoperoxidasemonolayer assay (IPMA) is still commonly usedtoday because of its ease and the ability to test manyserum samples for BVDV at a time. Prior to the de-velopment of the assay for serum, Bielefeldt-Ohmann (1983) had utilized the immunoperoxidasestaining technique for the detection of BVDV in tis-sues of infected animals.

Panels of monoclonal antibodies were producedto both BVDV and CSFV in the late 1980s and early1990s by several research groups. The specificity ofthese monoclonal antibodies was usually to the E2(gp53) protein, which was determined to be themajor neutralizing envelope protein of the virus(Donis et al., 1988), or to the nonstructural NS3(p80) protein or less frequently, to the Erns (gp48)protein, a second envelope protein, which was dis-covered later to have RNase activity (Schneider etal., 1993). Edwards et al. (1988) examined a total of38 monoclonal antibodies, of which 26 were againstBVDV antigens and 12 were against CSFV anti-gens. The monoclonal antibodies could be dividedinto three panels: those that were pan-pestivirus spe-cific; those that were CSFV-specific; and those thatwere selectively reactive with ruminant pestiviruses.Monoclonal antibodies to the NS3 protein tend to becross-reactive with all pestiviruses, because theamino acid sequence of this protein is highly con-served among pestiviruses. Monoclonal antibodiesto the envelope proteins E2 and Erns tend to be spe-cific for the viral species (i.e., BVDV or CSFV)used as the immunogen in their production (Ed-wards et al., 1991). Corapi et al. (1990a) examinedthe cross-reactivity of a panel of BVDV monoclonalantibodies to 70 BVDV isolates. They found that 12of 13 of their NS3 monoclonal antibodies reactedwith 100% of BVDV isolates tested, whereas the re-activity of E2 monoclonal antibodies varied from6–98%. Of two Erns monoclonal antibodies, one wasreactive to 100% of BVDV isolates tested, whereasthe other was less cross-reactive (57%). For the 70BVDV isolates, a total of 32 distinct patterns ofmonoclonal antibody reactivity were observed. Thisdemonstrated that considerable antigenic diversityexists among BVDV isolates.

CONTROL

To gain insight into the carrier state and for controlstrategies for BVDV, it was important to determinethe prevalence of persistently infected cattle in en-tire populations. Meyling (1984), used the immu-noperoxidase monolayer assay and found that ap-proximately 1% of slaughter cattle in Denmark wereviremic and apparently persistently infected. Subse-quently, Bolin et al. (1985a) made the first attemptto determine the prevalence of persistently infectedcattle in U.S. herds. The prevalence of persistent in-fection in a nonrandom population of 66 herds was1.7%. Since 50% of these herds were chosen be-cause of a history of BVDV infection, it was notedthat the prevalence figure for the entire U.S. cattle



population was probably somewhat lower. However,the figure of 1.7% turned out to be similar to theprevalence of persistently infected cattle of approxi-mately 0.5% to 2.0% obtained in later surveys con-ducted in different countries in the 1980s and 1990s(Houe, 1999). In their study on the prevalence ofpersistently infected cattle, Bolin et al. (1985a) alsotitrated the virus from the serum of persistently in-fected animals and found that these animals usuallyhad high BVDV titers (104–105 TCID50/ml).

At about the same time as the study of McClurkinet al. (1984) on the production of immune-tolerant,persistently infected calves, Liess et al. (1984) pro-duced congenital malformations (cerebellar hy-poplasia, hydrocephalus) and persistent infections incalves by inoculation of pregnant cattle with amodified-live vaccine and demonstrated the risk ofusing these vaccines in breeding animals. In bothstudies, persistently infected calves could be pro-duced by inoculation of pregnant cows to about 120days of gestation, after which BVDV infection gen-erated a fetal immune response.

By the late 1980s, modified-live vaccines, atemperature-sensitive mutant virus vaccine (Lob-man et al., 1984), and killed-virus vaccines wereavailable for use. Two elements of control were con-sidered essential: the detection and elimination ofpersistently infected carriers and immunization ofbreeding animals before their first conception(Radostits and Littlejohns, 1988). Immunization ofcalves was now considered to be less important bysome. Previously, before the pathogenesis of MDwas understood, it was thought by some that immu-nization of young cattle might prevent MD from oc-curring because the disease usually occurs in cattlefrom 6–24 months of age. Radostits and Littlejohns(1988) commented that pvMD gave vaccines a poorreputation and as a result they had not been used ona regular basis. There was also a concern that modi-fied-live BVDV vaccines might cause immunosup-pression and increase the risk of mortality in feedlotcattle (Martin et al., 1980; 1981).

Radostits and Littlejohns (1988) stated that therewas no substantial evidence to warrant the vaccina-tion of feedlot cattle. They further suggested that, ifvaccination of the dam prior to conception is a partof the control program, vaccination of calves may beunnecessary until they approach breeding age. Thus,there was a de-emphasis of vaccination for somegroups of cattle (calves and feedlot animals), at leastin some circles. Baker (1987), cited studies showingan association between BVDV vaccination and in-creased risk of mortality in feedlot animals and a re-

port on the immunosuppressive properties of amodified-live (Singer strain) vaccine (Roth andKaeberle, 1983) and believed that it would be ad-vantageous to vaccinate calves in a preconditioningprogram before their arrival at a feedlot. If calves arevaccinated on arrival at feedlots, he believed theyshould be vaccinated with a killed-virus vaccine.

Bolin (1990) also stated the concerns regardingthe use of modified-live vaccines: immunosuppres-sion, the potential to produce pvMD, and the poten-tial to adversely affect the fetus. During the late1980s, there was also a growing awareness of theantigenic diversity among BVDV isolates. Thus, an-other concern for vaccines in general was their effi-cacy in protecting against fetal infections with anti-genically variable field viruses. Bolin (1990)recommended using modified-live vaccines in largegrazing herds or when handling facilities for cattlewere poor, since only a single dose of vaccine is re-quired for immunization. Modified-live vaccineswere contraindicated for use in pregnant cattle andanimals in contact with pregnant cattle. Instead,Bolin (1990) recommended the use of killed-virusvaccines in dairy herds where pregnant cattle are al-ways present. Killed-virus vaccines were also rec-ommended for bulls in semen collection centers.

Although the identification and elimination ofpersistently infected cattle was considered an essen-tial element of control, it was also considered costlybecause of the amount of testing involved. Baker(1987) recommended testing for virus or viral anti-gen and determining antibody status, but stated thatscreening of whole herds may not be economicallyfeasible in all situations. He suggested an alternateapproach to reduce the cost of testing. Persistentlyinfected animals are viremic, usually antibody-negative, and often exhibit a poor response to vacci-nation. Hence, one possibility is to vaccinate all cat-tle greater than 6 months of age with a killed-virusvaccine followed by a booster, and then determinetheir antibody titers. Cattle that remained antibody-negative or respond poorly with low levels of anti-body would be suspected to be persistently infectedand could be tested for virus or viral antigen. Thiswas reminiscent of the observations of McClurkin etal. (1979) that cattle that remain seronegative whilein contact with seropositive cattle are likely to beimmune-tolerant and persistently infected, and thesecattle had low antibody titers after vaccination.However, because the immune tolerance is specificfor the persisting BVDV in these animals, screeningby this method may not be highly reliable, particu-larly when the vaccine is antigenically very different



from the virus persisting in the herd. In the lattercase, higher than expected levels of antibodies mightbe produced by persistently infected animals andobscure differences in antibody levels between per-sistently infected and normal cattle.

Baker (1987) noted several important factors thatshould be considered in herd screening, includingthat persistently infected calves may be seropositivebecause of colostral antibodies, and that passive im-munity may interfere with virus isolation. Further-more, any calves born in the 9 months after herdtesting potentially may be persistently infected. Healso noted that retesting along family lines may beworthwhile because of the possibility of persistentlyinfected families. After a herd is free of persistentlyinfected animals, he recommended isolating andtesting all new additions to the herd and that any ad-ditions that are pregnant should also have theircalves tested for virus at birth.


THE EMERGENCE OF TYPE 2 BVDVIn the early and mid-1990s, in addition to outbreaksof hemorrhagic syndrome (severe thrombocytopeniawith hemorrhages) (Ridpath et al., 1994), outbreaksof acute, severe BVD in which hemorrhagic syn-drome was either inapparent or was not a prominentclinical feature, occurred in Canada and the UnitedStates (Pellerin et al., 1994; Carman et al., 1998;Sockett et al., 1996; Drake et al., 1996). In Canada,these BVD outbreaks were most damaging to thecattle industry in the provinces of Quebec andOntario. In 1993, Quebec lost approximately 25% ofits veal crop (overall mortality: 32,000 out of143,000 calves) to these BVDV outbreaks (Pellerinet al., 1994). Although the Quebec outbreaks werenot described in detail from a clinical standpoint,Pellerin et al. (1994) stated that herds of calves thatlooked healthy one day could suffer a 10% deathloss the next day, and it was not a rarity to have amortality rate of 100%. Fever, pneumonia, diarrhea,and sudden death occurred in all age groups andabortions were frequent in the Ontario outbreaks of1993 to 1995 (Carman et al., 1998). The diseaseoften resembled MD. However, severe acute BVDcould be distinguished from MD because only non-cytopathic BVDV was isolated from animals suffer-ing severe acute BVD.

Pellerin et al. (1994) and Ridpath et al. (1994) de-termined that the BVDV isolates causing hemor-rhagic syndrome and acute severe BVD formed anew genetic group (genotype) of BVDV distinct

from early strains such as the Oregon C24V, NADL,and Singer strains utilized in vaccines. The newgroup was designated type 2 (BVDV 2) and thegroup comprising the early strains as type 1 (BVDV1). Pellerin et al. (1994) further subdivided BVDV 1into two subgroups: 1a, comprising such strains asNADL, Oregon, and Singer; and 1b, which includedNY-1 and Osloss strains.

In the Ontario BVDV 2 outbreaks, the initial clin-ical complaint was frequently of respiratory diseasein calves or adults (Carman et al., 1998). Diarrheaand abortion were also listed as initial clinical signs.Postmortem lesions were generally those describedfor MD: gastrointestinal erosions and ulcers.Pneumonia was the most common concurrent diag-nosis and was observed in all age groups.

Retrospective typing of Ontario isolates recov-ered from 1981–1994 proved that as early as 1981,BVDV 2 was already present in Ontario, Canada(Carman et al., 1998). Since outbreaks of severeacute BVD did not occur until much later (1993),this suggests that either earlier circulating strains ofBVDV 2 were not highly virulent and that somestrains had subsequently acquired virulence deter-minants or, alternatively, virulent strains of BVDV2 existed early but were harbored in seropositiveherds and only later caused outbreaks of severe dis-ease when naive populations became exposed andinfected.

PHYLOGENETIC STUDIES

Prior to the decade of the 1990s, BVDV, CSFV, andBDV were the three recognized pestiviruses, al-though it was uncertain whether BDV represented aunique viral species or whether border disease insheep was caused by BVDV. During the 1990s, inaddition to the segregation of BVDV into two geno-types, genetic characterization of isolates fromsheep showed that sheep could be infected by bothgenotypes of BVDV and by a unique pestivirus thatwas referred to as the “true” border disease virus(Becher et al., 1995). The first “BDV” isolate se-quenced was actually a BVDV 2 strain (Sullivan etal., 1994; Becher et al., 1995). It was also demon-strated that wildlife could be infected with pes-tiviruses. Genetic analysis of a pestivirus isolatedfrom a giraffe proved this virus to be a unique geno-type, whereas three deer isolates were found to be-long to the BVDV 1 genotype (van Rijn et al., 1997;Becher et al., 1997).

Baule et al. (1997) reported that southern Africanisolates consisted of four BVDV 1 subtypes, includ-ing the subtypes 1a and 1b previously described



(Pellerin et al., 1994; Ridpath et al., 1994). The twonew BVDV 1 subtypes from southern Africa weredesignated 1c and 1d. The latter BVDV subtype wasreported to be predominantly associated with respi-ratory tract disease and later shown experimentallyto be able to cause primary respiratory disease(Baule et al., 2001). Vilcek et al. (2001) examined76 BVDV 1 isolates from various countries and re-ported that these viruses could be separated into 11phylogenetic groups. Recently, Flores et al. (2002)identified six South American strains and one NorthAmerican BVDV 2 strain that cluster in a separategroup from other BVDV 2. Thus, two subgenotypesof BVDV 2 (2a and 2b) have now been identified.

GENETIC RECOMBINATION ANDSPONTANEOUS AND POSTVACCINALMUCOSAL DISEASE

After the findings by Meyers et al. (1989) of the cel-lular insertion of ubiquitin-coding sequences in theNS2 gene of cytopathic Osloss strain of BVDV andanother unidentified insertion (also identified laterto be of cellular origin) in the same gene of theNADL strain, much of the research activity in theearly and mid-1990s was devoted to the geneticanalysis of cytopathic and noncytopathic virus pairsfrom cases of MD. This analysis revealed a numberof genetic changes in cytopathic BVDV, includinginsertions and gene duplications and deletions,changes that resulted in the production of NS3 bycreation of a cleavage site at its amino terminus (re-viewed by Meyers and Thiel, 1996). In cases whereviral genes were duplicated, an additional NS3 genegenerated the NS3 protein. The genetic changes ob-served in cytopathic BVDV demonstrated that bothviral-cellular and viral-viral recombination were theoperating mechanisms for the generation of cyto-pathic BVDV from the persisting noncytopathicBVDV.

In 1995, Ridpath and Bolin (1995) characterized anoncytopathic (BVDV 2-125nc) and cytopathic(BVDV 2-125c) viral pair isolated from an animalsuffering from late-onset pvMD. This animal hadbeen vaccinated 3 months prior with a modified-liveBVDV 1 (NADL strain) vaccine. Genetic sequenc-ing of the BVDV 2 viral pair revealed that in com-parison with BVDV 2-125nc, the BVDV 2-125c iso-late contained a 366-nucleotide insertion in theNS2-3 gene (leading to the expression of the NS3protein). The inserted sequence was found to have a99% identity with sequences of the BVDV 1-NADLvaccine virus, indicating that the vaccine virus hadrecombined with the noncytopathic virus BVDV

2-125nc to produce the cytopathic BVDV 2-125c.The authors speculated that the prolonged time be-tween vaccination and pvMD may have been a re-flection of the time required for a single virus parti-cle, generated by the recombination event, toreplicate to a high enough titer to precipitate pvMD.The cytopathic BVDV 2-125c was one of the firstisolated and characterized cytopathic BVDV 2strains and has since been used by many laboratoriesas a challenge virus in serum neutralization assaysfor determination of neutralizing antibodies toBVDV 2.

Subsequently, Becher et al. (2001) also found, intwo independent cases of pvMD, that genetic re-combination between the persisting noncytopathicBVDV and a BVDV vaccine virus had occurred. Inone of these cases, the cytopathic component of thevirus pair was shown to have been generated by bothrecombination and deletion events and was actuallycomprised of five different cytopathic BVDV sub-genomes (genomic BVDV RNA containing large in-ternal deletions but able to express the NS3 protein).

LATE-ONSET MUCOSAL DISEASE

Several experimental studies were conducted onlate-onset (or delayed-onset) MD in the mid- andlate 1990s. Whereas in MD (or acute MD) the super-infecting cytopathic BVDV kills the persistently in-fected host within about 3 weeks, in late-onset MDthere may be several months between challenge withthe cytopathic virus and the onset of clinical signs ofdisease. Late-onset MD was first described byBrownlie et al. (1986), who infected six persistentlyinfected calves with homologous cytopathic BVDVand six with heterologous cytopathic BVDV. All sixanimals superinfected with the homologous virusdeveloped MD in the first 3 weeks, whereas none ofthe animals superinfected with a heterologous virusdeveloped MD in this time period. However, two ofthe animals infected with the heterologous virus de-veloped late-onset MD at 98 and 146 days postinoc-ulation. Three others remained healthy until eutha-nized at 59–209 days postinoculation. The sixthanimal appeared to be developing late-onset MD atabout 80 days with signs of intermittent diarrheaover a 4-week period. Later, Westenbrink et al.(1989) also inoculated clinically healthy, persist-ently infected animals with three heterologous cyto-pathic BVDV, and MD occurred as late as 99 dayspostinoculation. It should be noted here that sincespontaneous MD can occur in persistently infectedanimals that are not superinfected and are held inisolation (Brownlie and Clarke, 1993), it is possible



that some experimental cases of late-onset MD (andpvMD), could actually be cases of spontaneous MD.Thus, in experiments of this type, an analysis isneeded (preferably genetic sequence analysis), be-yond that of virus-neutralizing antibody responses,of the superinfecting input cytopathic BVDV, thepersisting noncytopathic virus, and the output orMD-associated cytopathic BVDV.

Moennig et al. (1993) inoculated a persistently in-fected bull with a heterologous cytopathic BVDV(TGAC), and MD was observed 15 weeks after in-oculation. In this case, the neutralizing antibody re-sponse to the cytopathic TGAC virus was delayed31 days postinfection and neutralizing antibodytiters increased to a maximum on day 114 posti-noculation (when the animal was euthanized), indi-cating prolonged persistence of the superinfectingcytopathic virus in this animal. However, only non-cytopathic virus was recovered from buffy coat cellsand nasal swabs. The cytopathic BVDV recoveredfrom fecal samples (designated cpX) at the onset ofdisease was analyzed and found to have the samephenotype (monoclonal antibody reactivity pattern)as the noncytopathic persisting virus but the samegenotype in the NS2-3 region as the TGAC cyto-pathic virus as observed by restriction enzyme di-gestion of PCR products (Fritzemeier et al., 1995).Thus, they concluded that cpX was a phenotypicallyaltered variant of TGAC. Subsequent nucleotide se-quencing (Fritzemeier et al., 1997) indicated thatcpX arose by recombination of persisting noncyto-pathic virus and TGAC. This finding was consistentwith the finding of Ridpath and Bolin (1995) ofviral-viral recombination for a case of late-onsetpvMD. Fritzemeier et al. (1997) also found that asuperinfecting cytopathic BVDV in experimentalacute MD (MD occurring in a 2–3-week time frameafter inoculation) remained genetically unchangedfrom inoculation to the onset of MD. They con-cluded that acute MD and late-onset MD may occurby two different mechanisms: In the former, it is thesuperinfecting virus that causes the disease; whereasin the latter, genetic recombination between the su-perinfecting cytopathic virus and the persisting non-cytopathic virus creates a new cytopathic virus thatcauses the disease. Subsequently, Fray et al. (1998)inoculated a persistently infected calf with culture ofan uncloned heterologous cytopathic BVDV ob-tained from a natural case of MD (thus also contain-ing a noncytopathic virus) and the animal developedMD at 145 days postinoculation. Of interest in thisstudy was the occurrence of a prolonged nasal shed-ding and viremia of cytopathic BVDV before the

onset of MD. No genetic analysis of viral isolateswas conducted, but neutralizing antibody data indi-cated that at least three antigenically distinct cyto-pathic viruses were isolated from the calf during theperiod between superinfection and postmortem ex-amination. The authors speculated that the noncyto-pathic virus in the heterologous challenge may haveinfluenced the survival of cytopathic BVDV in theanimal in some manner, possibly by influencing theimmune response of the host.

NORMAL CALVES FROM PERSISTENTLYINFECTED COWS

Calves born to persistently infected cows are like-wise persistently infected; thus these animals are un-suitable for breeding (McClurkin et al., 1984). Wen-tink et al. (1991), in an attempt to preserve thegenetic material of a highly valued heifer that wasdevelopmentally normal but persistently infected,transferred a fertilized embryo of the heifer to animmunocompetent recipient cow. Before transfer,the embryo was treated and washed according toroutine methods of the International Embryo Trans-fer Society (IETS). The result was a normal heifercalf with normal immunity to BVDV. Since then,several other researchers have repeated these resultsand have produced, by embryo transfer, normalcalves from persistently infected cows (Bak et al.,1992; Brock et al., 1997; Smith and Grimmer,2000).

It was noted however, that there was a lack of asuperovulatory response following hormone stimu-lation in persistently infected cows and only lownumbers of viable embryos could be obtained.Wentink et al. (1991) obtained only one viable em-bryo out of six, and Brock et al. (1997), who usedseven persistently infected donors, obtained onlynine transferable embryos from two of the femalesafter 45 individual uterine flushes. In the study ofBrock et al. (1997), only one pregnancy was ob-tained from the transfer of six embryos into seroneg-ative recipients. None of the recipient cows showedseroconversion to BVDV leading the authors to con-clude that neither horizontal nor vertical transmis-sion of the virus occurs when recommended IETSembryo washing procedures are followed.

ATYPICAL PERSISTENT INFECTION

In 1998, Voges et al. (1998) described for the firsttime a persistent infection of BVDV in an immune-competent animal. The bull in this case was notviremic but continuously shed virus in its semen overa period of 11 months until slaughtered. The bull ap-



peared healthy and its growth rate and testicular de-velopment were unremarkable. Semen quality wasconsidered good. Limited pregnancy data suggestedthat the in vivo fertility of the bull’s sperm was notcompromised. The animal had a consistently hightiter of serum neutralizing antibodies to the standardtest virus (�1:4096). Against the homologous virus,the animal had an extraordinarily high serum neutral-izing antibody titer of >1:100,000.

The virus titer in the semen was relatively low(<103.3 TCID50/ml) when compared to semen ob-tained from typical, persistently infected (viremic)bulls (approximately 105–107 TCID50/ml) (Barlowet al., 1986; Meyling and Jensen, 1988; Kirkland etal., 1991). On postmortem, BVDV was isolatedfrom testicular tissues but not from any other tissueexamined. This long-term shedding of BVDV in anon-viremic bull is reminiscent of the situation installions infected with equine arteritis virus(Timoney and McCollum, 1993). In this infection,stallions undergo a transient viremia resulting in animmune response that eliminates the virus exceptfor that in the genital tract. Chronic shedding of thevirus in the semen occurs in many infected stallionsand may persist for years, although most horseseventually clear the infection. These carrier stallionsalso have high serum neutralizing antibody titers.

As hypothesized by Voges et al. (1998), the bullwith the atypical persistent infection was likely in-fected near the age of puberty, shortly before theblood-testis barrier becomes fully functional. Thisallowed infection of testicular tissues but excludedensuing antibodies from the site. The high level ofserum neutralizing antibodies suggests continual ex-posure of the immune system to the virus. This atyp-ical persistent infection of BVDV in bulls appears tobe rare, but at least one other case has been reported(van Rijn, 1999). Nevertheless, this has changed thetesting requirements for bulls entering artificial in-semination (AI) centers. Previously, testing for vire-mia was all that was required to identify typical per-sistently infected bulls and to restrict their entry;semen was rarely tested. Now the World Organiza-tion for Animal Health (Office International desEpizooties) requires the testing of semen from eachbull before entry into an AI center.

ADVANCES IN MOLECULAR BIOLOGY

In the latter half of the 1990s, infectious cDNAclones were constructed for the type 1, cytopathicBVDV strains cp7, and NADL (Meyers et al., 1996;Vassilev et al., 1997; Mendez et al., 1998). Subse-quently, 40 years after its initial discovery (Gillespie

et al., 1960), an infectious cDNA clone was con-structed from the genome of the type 1, cytopathicOregon strain (Kümmerer and Meyers, 2000). Theavailability of infectious cDNA clones has begun tohave, and will continue to have, a large impact onour understanding of the molecular biology ofBVDV as these cDNA clones are manipulated andused in studies to decipher the function of genes andgenetic elements.

Meyers et al. (1996) removed an insertion of 27nucleotides in the NS2 gene of the infectious cloneof the cytopathic cp7, which is not present in thegenome of cp7’s noncytopathic counterpart, ncp7(the two viruses representing a homologous viralpair from a case of MD). In removing the insertion,a noncytopathic BVDV was recovered, proving thatthe insertion was responsible for the cytopatho-genicity of the cp7 virus. Similarly, Mendez et al.(1998) deleted a 270-nucleotide insertion in the NS2gene of the infectious clone of NADL and also pro-duced a virus that was no longer cytopathic. In bothcases, the altered infectious clones of cp7 andNADL failed to produce NS3, showing that these in-sertions led to processing at the NS2/NS3 site, pro-duction of NS3, and cytopathic effect.

Some strains of cytopathic BVDV, such as theOregon and Singer strains, do not contain insertionsor genetic rearrangements and, thus, did not arise byrecombination but by another mechanism. Küm-merer et al. (1998) and Kümmerer and Meyers(2000) used chimeric cDNA constructs in a transientexpression system and alterations in an infectiouscDNA clone to show that processing at the NS2/NS3 site in the Oregon strain was the result of pointmutations within the NS2 protein. They noted thatabout 40 cytopathic pestiviruses have thus far beenanalyzed and the majority of these have been gener-ated by recombination. The remaining strains haveapparently arisen by point mutations generated dur-ing replication. However, since these cytopathicviruses occur less frequently than those that arise byrecombination (itself a rare event) these researchershypothesized that this second mechanism of gener-ating cytopathic BVDV was not one of sequentialaccumulation of point mutations, but rather the si-multaneous introduction of a set of point mutations.

Recently, an infectious cDNA clone for a virulent,noncytopathic type 2 BVDV (New York 93 strain)was constructed (Meyer et al., 2002). Infectiousvirus derived from the cDNA clone retained viru-lence when used to infect cattle. These animals de-veloped fever, leukopenia, and clinical signs, includ-ing respiratory symptoms and gastrointestinal



disorders. Alteration of single histidine codons (atpositions 300 and 349) in the sequence of the Erns

gene in the infectious cDNA clone led to two newinfectious clones with RNase-negative phenotypes.Virus derived from the infectious clone having adeletion of the histidine codon at position 349 wasused in animal experiments and was found to havean attenuated phenotype. None of the calves in-fected with this mutant virus had body temperaturesabove 39.5°C, nor did they develop diarrhea; onlymild respiratory signs were observed. Leukopeniaoccurred but with an early recovery of leukocytenumbers. Viremia and nasal shedding also occurredbut for a shorter duration than in animals infectedwith virus derived from the unaltered infectiouscDNA clone. As demonstrated by this study, theavailability of an infectious cDNA clone in whichvirulent BVDV can be recovered allows new ap-proaches to the study of BVDV-induced disease andidentification of genetic markers of virulence andattenuation.

DIAGNOSIS BY VIRUS ISOLATION

An essential element of control of BVDV in herds isthe detection and elimination of persistently infectedcalves. The classical method of detecting persistentinfections is virus isolation in cell culture usingserum or leukocytes as the test sample. In testingyoung calves, a major concern with the use of virusisolation, especially from serum, was the presenceof colostral antibodies to BVDV, which could inter-fere with the sensitivity of the test. Palfi et al. (1993)studied the decline of BVDV colostral antibodiesand the detectability of BVDV in young, persist-ently infected calves and found that viremia was notdetectable in the serum of seven persistently in-fected calves with colostral antibody titers, whichwere as low as 1:16 to 1:24.

In a later study, only one of four persistently in-fected calves, which had the highest viral titer (106.5

TCID50/ml) before ingestion of colostrum, was de-tected as viremic after colostrum ingestion whenboth serum and leukocytes were tested (Brock et al.,1998). Palfi et al. (1993) found that the half-life ofcolostral antibodies in persistently infected calves (5to 11 days) was much shorter than in non-persist-ently infected calves (approximately 3 weeks).Presumably, the continuous high virus production inpersistently infected calves is responsible for therapid clearance of colostral antibodies. Palfi et al.(1993) found that clearance of these antibodies inpersistently infected calves may occur by 8 weeks ofage, at which time viremia can be detected.

However, it has been generally recommended that, ifcalves under 3 months of age have been tested byvirus isolation, they be retested at 3 months of age(Brock et al., 1998; Saliki et al., 1997). Dubovi(2002) recommended that, for young calves, in ad-dition to the standard washing of leukocytes to re-move colostral antibody, fresh, rather than freeze-thawed, leukocytes be used for virus isolation sinceinterference, presumably by residual antibodies,may occur upon freeze-thawing.

Although persistently infected cattle generallyhave very high BVDV titers (104–105 TCID50/ml ofserum), Brock et al. (1998) found that the level ofviremia in one of seven persistently infected animalsbecame undetectable over several test dates whenserum was tested. In this apparently rare case, theanimal developed neutralizing antibodies to the per-sisting virus although the infection was nevercleared. During periods when the virus was unde-tectable in serum because of neutralizing antibody,virus was isolated at low concentrations from bloodleukocytes.

Rae et al. (1987) examined the viability of BVDVin serum and plasma collected from persistently in-fected cattle. Storage of samples in the dark at roomtemperature (17–26°C) for up to 5 days had no sig-nificant effect on virus titer. They concluded thatsuccessful isolation of BVDV from persistently in-fected animals was unlikely to be compromised by adelay of up to 5 days between sample collection andtesting.

MONOCLONAL ANTIBODY-BASED TESTS

With the emergence of BVDV 2, existing panels ofBVDV 1 monoclonal antibodies were examined forcross-reactivity with these new isolates, and mono-clonal antibodies to BVDV 2 were also produced.Ridpath et al. (1994) tested 29 E2-specific mono-clonal antibodies produced against BVDV 1 strainsand found that most of these failed to react withBVDV 2 isolates. However, two E2-specific mono-clonal antibodies were reactive with all 15 BVDV 2isolates tested. Deregt and Prins (1998) also deter-mined that one BVDV 1 E2-specific monoclonal an-tibody was reactive to all 21 BVDV 2 isolates tested,whereas another E2-specific monoclonal antibodywas determined to be BVDV 1-specific. The epitopesof these two monoclonal antibodies were mapped,the first to an immunodominant, type-common epi-tope (Paton et al., 1992; Deregt et al., 1998a) and thelatter to a conformational epitope containing a criti-cal amino acid deleted in BVDV 2 isolates (Deregt etal., 1998a). E2-specific monoclonal antibodies



against BVDV 2 that were reactive with neutralizingepitopes in three antigenic domains were also pro-duced (Deregt et al., 1998b). Most of these mono-clonal antibodies were unreactive or only weakly re-active with BVDV 1. One monoclonal antibody hada type-common specificity, and two others were en-tirely BVDV 2-specific.

A number of monoclonal antibody-based testswere developed in the 1990s. Competitive (block-ing) enzyme-linked immunosorbent assays(ELISAs) for the detection of antibodies to BVDVwere constructed using NS3 monoclonal antibodies(Lecomte et al., 1990; Paton et al., 1991). Paton etal. (1991) used two NS3 monoclonal antibodies re-active to 157 different pestiviruses for developmentof a blocking ELISA. They examined the ability ofthe assay to detect antibodies against pestiviruses incattle, sheep, and swine. The relative sensitivity ofthe blocking ELISA compared to serum neutraliza-tion was high for bovine and ovine sera (94.7% and99.1%, respectively) but lower for swine (76%),whereas the specificity was high in each case(�96%).

Several capture ELISAs for detection of the NS2-3 protein in persistently infected cattle were also de-veloped and commercialized (Brinkhof et al., 1996).The NS3/NS2-3 proteins are highly immunogenicand the NS2-3 protein is produced in large amountsin persistently infected cattle. Brinkhof et al. (1996)evaluated four commercial capture ELISAs employ-ing NS3 monoclonal antibodies in which bloodleukocyte preparations were tested. These assaysdemonstrated high sensitivity (94–97%) and speci-ficity (100%). The capture ELISA was considered tobe the test of choice for eradication programs wheremany animals need to be screened and monitored. Inthe above study and in another study (Shannon et al.,1992), false negative results were obtained with cap-ture ELISAs for calves with high levels of colostralantibodies in their blood. Thus, as for virus isola-tion, it was recommended that if young calves under3 months of age are tested by capture ELISA forpersistent infections they be retested again at 3months of age.

Haines et al. (1992) developed a monoclonal anti-body-based immunohistochemical method for de-tecting BVDV antigen in formalin-fixed tissuesusing the monoclonal antibody, 15C5. This mono-clonal antibody, with specificity for the Erns protein,was the only monoclonal antibody out of 32 testedthat was reactive against antigen preserved in thesetissues, suggesting that most protein epitopes ofBVDV are conformational in nature. The mono-

clonal antibody 15C5 was found to be particularlyuseful for this application because its epitope wasshown to be highly conserved, reacting with all ofthe 70 BVDV isolates tested (Corapi et al., 1990a).

Pooled monoclonal antibodies were employed inimmunoperoxidase monolayer assays (IPMA)(Saliki et al., 1997; Deregt and Prins, 1998) as wellas in a monolayer ELISA (m-ELISA) for microtitervirus isolation (Saliki et al., 1997). Saliki et al.(1997) utilized the Erns monoclonal antibody 15C5and a NS3 monoclonal antibody, whereas Deregtand Prins (1998) utilized a pool of E2 and NS3 mon-oclonal antibodies for detection of BVDV 1 andBVDV 2. The m-ELISA uses a spectrophotometerfor reading samples and is a more rapid test than theIPMA. The microtiter IPMA and m-ELISA, com-pared to conventional virus isolation, were shown tohave a relative sensitivity of 100% for samples fromcattle greater than 3 months of age suspected ofbeing persistently infected and 85% when samplesof cattle with acute infections were included (Salikiet al., 1997).

Flow cytometry has also been investigated for useas a diagnostic assay. Qvist et al. (1990, 1991) eval-uated its use in identification of persistently infectedcattle and found it to be more sensitive than virusisolation. Using a fluorescence-activated cell sorterto analyze fluorescent antibody-bound, infectedleukocytes, BVDV-specific antigens were shown tooccur in 3–21% (mean 11%) of mononuclear leuko-cytes of persistently-infected cattle.

POLYMERASE CHAIN REACTION

The first of many reverse transcription-polymerasechain reaction (RT-PCR) assays for BVDV was de-veloped in 1990 (Schroeder and Balassu-Chan,1990). Although validation of this assay was limited,it was reported to be as analytically sensitive as theIPMA for virus detection. Since the development ofthis assay, many more sophisticated PCR assayswere developed for BVDV in the 1990s, whichmatched the development of PCR in general forother viruses. These included nested PCR assayswith two rounds of PCR to increase analytical sen-sitivity, multiplex assays with species-specificprimers for genotyping, and real-time PCR assaysthat utilize fluorescent signals from oligonucleotideprobes to detect amplification.

With the discovery of BVDV 2 and the existenceof two genotypes of BVDV in the mid 1990s, sev-eral PCR-based assays for typing BVDV were de-veloped. Harpin et al. (1995) used an indirectmethod, following RT-PCR with restriction endonu-



clease digestion of the PCR product for typing.Ridpath et al. (1994) utilized the specific amplifica-tion of BVDV 2 for typing, with a negative result in-dicating BVDV 1. Subsequently, these researchersmade improvements to their PCR assays, includinga positive test for BVDV 1 and specific primers forthe differentiation of BVDV 1 subtypes, 1a and 1b(Ridpath and Bolin, 1998). The earlier PCR assayswere followed by two multiplex PCR assays inwhich specific primers for BVDV 1 and BVDV 2are used in a nested PCR format in the second roundof amplification (Sullivan and Akkina, 1995; Gilbertet al., 1999). The advantage of the multiplex assay isthat a specific product of differing size is producedfor each genotype and differentiation occurs in a sin-gle assay. The assay of Sullivan and Akkina (1995)could type BDV as well as BVDV 1 and BVDV 2since BDV-specific primers were included in theassay. This is seen as an advantage for typing pes-tiviruses from sheep, which can be infected by allthree viruses, but in cattle BDV does not appear tobe readily infectious. BDV has not been isolatedfrom North American or European cattle (Paton etal., 1996; Ridpath, 1996) and only one bovine BDVisolate has ever been reported; its original isolationwas thought to have been made in the 1960s (Becheret al., 1997). The PCR assay of Gilbert et al. (1999),further described in Deregt et al. (2002) could beused with conventional RNA extraction, or directlywithout RNA extraction, by adding the sample to theRT-PCR mixture.

Radwan et al. (1995) developed a PCR assay fortesting bulk milk samples to identify dairy herds in-fected with BVDV. For validation of the assay, theseresearchers first determined the BVDV titers in milkfrom an experimentally (acutely) infected cow andtwo persistently infected cows. Virus titers of 102.5,106.5, and 105.5 TCID50/ml were present in milkfrom the acutely infected cow and persistently in-fected cows, respectively. The virus titers in the milkof persistently infected cows were higher than thosein their sera (approximately 104.5 TCID50/ml). ThePCR assay was found to be about 15 times moresensitive than virus isolation in detecting BVDV inmilk somatic cells. By testing bulk milk samples,BVDV was detected in 33 of 136 dairy herds by thePCR assay. In sharp contrast, virus isolation did notdetect BVDV in any of the bulk milk samples. It wassuggested that BVDV antibodies in the milk wereresponsible for the poor virus isolation results.

Drew et al. (1999) also applied PCR to bulk milksamples to identify BVDV infection among lactat-ing cows, but did not attempt virus isolation from

these samples. Their assay was able to detect BVDVshed by 1 persistently infected cow in a herd of 162lactating animals. Later, Renshaw et al. (2000) useda PCR assay, which employed the primers used inthe assay of Radwan et al. (1995), to detect BVDVin bulk milk samples and also attempted virus isola-tion from these samples. They found that BVDVcould be detected by both methods when milk froma single persistently infected animal was diluted1:600 with milk from a herd of BVDV-negative an-imals. Of 144 bulk milk samples from 97 farms, 24were BVDV-positive by either PCR or virus isola-tion: 20 were positive by PCR and 17 were positiveby virus isolation. Renshaw et al. (2000) suggestedthat the poor virus isolation rate of Radwan et al.(1995) may have resulted from using milk somaticcells that were frozen before virus isolation was at-tempted, and that successful virus isolation could beobtained when freshly prepared (not frozen) milksomatic cells are utilized. Furthermore, they recom-mended simultaneous PCR and virus isolation test-ing of bulk milk samples to ensure detecting BVDVthat may either not be amplified by primers cur-rently in use or which for unknown reasons may bedifficult to isolate.

McGoldrick et al. (1999) designed a real-time(Taqman) nested PCR assay for CSFV, which couldbe modified with different fluorescent probes toallow specific detection of BVDV 2 or BDV.Another fluorescent probe allowed detection of allpestiviruses except for some isolates of BVDV 2.The nested PCR assay was performed in a closed,single tube in which reagents for the second roundof PCR were maintained in the inner lid. After re-verse transcription and a primary round of PCR, thetubes were inverted to mix the second roundreagents in the lid with the first round products forinitiation of the second round of PCR. Later,Mahlum et al. (2002) designed a real-time (Taqman)PCR assay specifically for BVDV. The assay wasfound to be more sensitive than virus isolation andIPMA in detecting BVDV in sera, and more sensi-tive than virus isolation or immunohistochemistry indetecting BVDV in tissues.

CONTROL BY VACCINATION

Prior to the emergence of virulent BVDV 2 strains,most BVDV infections of nonbreeding animals werethought to be benign and emphasis was placed onthe prevention of fetal infections. Thus, it was rec-ommended that all breeding females be vaccinatedprior to conception (Baker, 1987; Radostits andLittlejohns 1988). Other animals in the herd were



often not vaccinated. In the 1990s, with the emer-gence of virulent BVDV 2, the failure to vaccinateall animals in a herd proved to be disastrous, withlosses of up to $40,000 to $100,000/herd when out-breaks due to these virulent viruses occurred(Carman et al., 1998). Fortunately, the vaccines atthat time, which contained only BVDV 1, did pro-vide a measure of protection against acute infectionswith BVDV 2 and outbreaks could be controlled(Carman et al., 1998).

Although immunization with some vaccines con-taining BVDV 1 alone were protective against acuteBVD caused by BVDV 2 (Cortese et al., 1998,Carman et al., 1998), apparent vaccine breaks didoccur (Ridpath et al., 1994) and incorporation ofBVDV 2 in vaccines was considered a priority. Inaddition to the prevention of acute disease by BVDV2, a continuing concern was the degree of fetal pro-tection against BVDV provided by vaccines in gen-eral. With the emergence of BVDV 2, this concernescalated with reports that vaccines containingBVDV 1 appeared not to be protective against fetalinfection with BVDV 2 (van Campen et al., 2000).

van Oirschot et al. (1999) reevaluated the resultsof fetal protection experiments that were conductedin the 1970s to 1990s by six different researchgroups. They concluded that no vaccine had shownfull fetal protection and that protection varied from33–86%. In only one study were none of the fetusesfrom the vaccinated cows infected after challenge;however, one fetus from six unvaccinated controlcows also remained uninfected (Brownlie et al.,1995). In this type of experiment it is important thatall fetuses in the control group become infected todemonstrate the ability of the challenge virus to in-fect the fetus under the defined experimental condi-tions. Thus, in the latter study the degree of protec-tion conferred by the vaccine was calculated to be86% because only five of six fetuses from the con-trol group became infected rather than 100% ofthem.

CONTROL WITHOUT VACCINATION

In the 1990s, eradication programs for BVDV on anational level were implemented in Sweden, Nor-way, Finland, and Denmark (Bitsch and Ronshølt,1995). These programs were conducted without vac-cination. In Sweden, the national program began in1993 as a voluntary program entirely financed byproducers (Alenius et al., 1996). In Denmark, theBVDV eradication program was initiated by dairyfarmers in 1994 followed by a government order tosupport the program in 1996 (Houe, 1996). Both the

Swedish and Danish programs involve classificationof herds by their BVDV status, removal of persist-ently infected cattle from infected herds, monitoringherd status, and prevention of infection in BVDV-free herds. Since these countries do not vaccinate,antibody-positive animals act as indicators of infec-tion. Thus, bulk milk testing for antibodies consti-tutes an important component of these programs andis used to classify and monitor herds as to their BVDstatus. In Sweden, bulk milk testing is done with anindirect ELISA and in Denmark with a blockingELISA (Alenius et al., 1996; Houe, 1996). Theseprograms significantly reduced the prevalence ofBVDV-positive dairy herds after only a few years.For Sweden, the prevalence dropped from 51% in1993 to 24% in 1995; for Norway, from 23% in1993 to 14.4% in 1996; and for Denmark, from 39%in 1994 to 9% in 1999 (Alenius et al., 1997; Bitschand Ronshølt, 1995; Bitsch et al., 2000; Waage etal., 1997). Finland, which began its program in 1994with a very low prevalence (only 1%) in dairy herds,further reduced the prevalence to just 0.4% in 1997(Nuotio et al., 1999).

Bitsch et al. (2000) reported that for the Danishprogram, legislation was needed in 1996 to ensurethat no persistently infected cattle were allowed oncommon pastures. Movement of cattle to other herdsor common pastures was allowed only after bloodtesting and certification that the animals were notpersistently infected. At first, pregnant (non-persist-ently infected) cows were not controlled. Instead, allbuyers of pregnant animals were advised to isolateand ensure a non-persistently infected status of theircalves before introducing them into a new herd.Since not all farmers heeded these recommendationsand herds became infected because of persistentlyinfected calves resulting from the purchase of preg-nant animals, legislation was modified to restrict themovement of all female cattle over 1 year of age,specifically that no such cattle could be moved fromnon-free herds to other herds or common pastures.

2000 TO THE PRESENT

RESPIRATORY DISEASE

Respiratory diseases have plagued the feedlot indus-try for many years and are collectively considered tobe the most significant cause of mortality for feedlotcalves. Several recent studies have linked BVDVwith bovine respiratory disease in feedlot cattle andanother study has identified BVDV strains that cancause primary respiratory disease experimentally.

Martin et al. (1999) determined the antibody titers



to BVDV, infectious bovine rhinotracheitis, parain-fluenza-3, bovine respiratory syncytial virus, andtwo mycoplasma strains in 32 groups of calves priorto entry, and 4–5 weeks after their arrival, at thefeedlot. They then looked for an association of theseinfectious agents (by increases in antibody titers)with the risk of respiratory disease. Of all the agents,BVDV had the most consistent association with ele-vated risk of respiratory disease and lower weightgains. In another study, Haines et al. (2001) exam-ined 49 cases of antibiotic-unresponsive, chronicdisease (most often respiratory disease and/or arthri-tis) in Canadian feedlot cattle. By immunohisto-chemistry, they found that Mycoplasma bovis andBVDV were the most common pathogens persistingin the tissues of these animals. BVDV was found inlung and/or joint tissues in 20 of the 49 (41%) cases.

Shahriar et al. (2002) reported on the prevalenceof pathogens in cases of chronic, antibiotic-resistantpneumonia with or without concurrent polyarthritisoccurring in feedlot cattle in western Canada. Theyexamined retrospective (1995–1998) and currentcases (1999) by immunochemistry of lung and hearttissue and found that Mycoplasma bovis was pres-ent in 44 of 48 cases of the retrospective group and15 of 16 of current cases; whereas BVDV was pres-ent in 31 of 48 retrospective and 9 of 16 currentcases. From four positive virus isolations in the cur-rent group, BVDV type 1b was isolated in two casesand BVDV 2 was isolated from the other two cases.These researchers suggested that a synergism be-tween Mycoplasma bovis and BVDV might occurin this syndrome of pneumonia with concurrentarthritis.

Later, Fulton et al. (2002) examined the preva-lence of BVDV in stocker calves with acute respira-tory disease and reported that BVDV type 1b wasthe predominant type involved. These researchersalso noted that vaccines in the U.S. primarily con-tained BVDV type 1a, that some vaccines also in-cluded BVDV 2, but that only one vaccine containedBVDV 1b. They suggested that, for effective vacci-nation against BVDV type 1b, there should bedemonstrated efficacy of current BVDV type 1avaccines against type 1b. Alternatively, new compo-nents of BVDV type 1b could be included in currentBVDV type 1a vaccines.

Baule et al. (1997, 2001) identified two new sub-types of BVDV 1 isolated in southern Africa anddesignated these as type 1c and type 1d. Type 1dviruses were found to be predominantly associatedwith respiratory disease. Two cytopathic viruses ofthis type were inoculated in calves intranasally or

intravenously. All inoculated calves developed res-piratory symptoms. One of the isolates producedmainly nasal discharge and fever in calves; however,the other isolate produced a more severe disease thatincluded ocular discharge, nasal discharge, fever,coughing, abnormal breathing, and oral erosions.Transient diarrhea was observed in only 2 of 10calves. There was a widespread distribution of virusin various tissues and organs of infected calves, in-cluding heart muscle, skin, bone marrow, and brain.However, lesions were mainly observed in the respi-ratory tract (focal catarrhal bronchopneumonia andatelectasis of the lung) and lymphoid tissues. In onecalf, virus was still present in tissues 31 days afterinfection in the absence of viremia. Furthermore,two calves were still shedding cytopathic BVDV innasal secretions at 21 and 31 days after infectionwhen virus was no longer detectable in blood.

MOLECULAR ACTIONS OF CYTOPATHIC ANDNONCYTOPATHIC BVDVIn recent years, the actions of cytopathic and noncy-topathic BVDV on cells have been studied more in-tensively. For cytopathic strains of BVDV, it hasbeen found that, like many other “lytic” viruses,they kill cells by triggering apoptosis (programmedcell death) rather than lysis and necrosis (Zhang etal., 1996; Adler et al., 1997; Hoff and Donis, 1997).Apoptosis is responsible for the elimination of cellsin normal developmental processes but can be trig-gered by many stimuli and is characterized by con-densation of chromatin, cell shrinkage, generationof apoptotic bodies, and fragmentation of chromo-somal DNA with the generation of typical oligonu-cleosomal fragments. The mechanism(s) by whichcytopathic BVDV triggers apoptosis is currently anactive area of research. BVDV is an ideal model sys-tem for viral-induced apoptosis because both cyto-pathic and noncytopathic forms of the virus exist. Inaddition to the production of NS3, which is corre-lated with cytopathic effect, it has recently been dis-covered that cytopathic viruses accumulate muchhigher levels of viral RNA in cells than do noncyto-pathic viruses.

Vassilev and Donis (2000) used an infectiouscDNA clone of the NADL strain to create, by dele-tion of the cellular insertion in the NS2 gene, an iso-genic, noncytopathic virus mutant. To study viralRNA accumulation in cells, they utilized both cyto-pathic and noncytopathic (mutant) forms of theNADL strain and several naturally occurring cyto-pathic and noncytopathic viral pairs. At a multiplic-ity of infection of one, viral RNA accumulation was



6.5 to 23 times higher when cells were infected withthe cytopathic virus than with the correspondingnoncytopathic virus. Thus, in addition to generationof NS3, a second factor—increased viral RNA con-centration in cells—may contribute to the ability ofcytopathic BVDV to induce the death of cells.

Studies spanning several decades have shown thatnoncytopathic strains of BVDV do not induce type 1(alpha/beta) interferons in cells (in vitro) and blockthe induction of interferons by other viruses(Diderholm and Dinter 1966; Nakamura et al., 1995;Adler et al., 1997). In contrast, cytopathic strains ofBVDV do induce type 1 interferons in cells in vitro(Adler et al., 1997). Recently, Schweizer andPeterhans (2001) showed that noncytopathic BVDVinhibits the induction of apoptosis and interferonsby a synthetic double-stranded RNA (poly IC). Sub-sequent work indicates that noncytopathic BVDVblocks an interferon regulatory factor (Baigent et al.,2002). These studies explain the basis of two some-what obscure diagnostic tests for noncytopathicBVDV. The first, described in 1968, was the ENDmethod (enhancement/exaltation of Newcastle dis-ease virus) (Inaba et al., 1968). Newcastle diseasevirus (NDV) induces, and is sensitive to, interferon.Co-infection of noncytopathic BVDV and NDVleads to an enhancement of the replication of NDV.In the END method, the presence of noncytopathicBVDV is thus shown by NDV enhancement due tothe suppression of interferon by noncytopathicBVDV. The second diagnostic test was based on thesuppression of an effect of poly IC (Maisonnave andRossi, 1982). Cells were inoculated with noncyto-pathic BVDV, and then treated with poly IC, andsubsequently inoculated with vesicular stomatisvirus (VSV). Cells infected with noncytopathicBVDV were not protected against cytopathic effectby VSV, whereas cells that were not infected withnoncytopathic BVDV, were protected. It is nowclear that the action of noncytopathic BVDV is viainhibition of interferon induction by poly IC.

The interference with interferon production invitro by noncytopathic BVDV has been suggested asan enabling factor in the ability of these viruses toestablish persistent infections in the early fetus(Schweizer and Peterhans, 2001). To determinewhether interference occurs in vivo, Charleston etal. (2001) inoculated the amniotic fluid of approxi-mately 60-day-old fetuses with noncytopathic or cy-topathic BVDV. Whereas cytopathic BVDV inducedinterferon production in the fetus, noncytopathicBVDV failed to do so. In contrast, the dams of thefetuses inoculated with noncytopathic BVDV did

produce interferons, as did calves in a subsequentstudy (Charleston et al., 2002). Thus, there appearsto be a marked difference in the interferon responsebetween the early fetus and immune-competent ani-mals upon infection with noncytopathic BVDV.

AN OLD VIRUS, ONCE LOST, TEACHESLESSONS

Of the old cytopathic BVDV strains—those thathave been used in laboratories for many years—theNADL strain that was isolated from a case of MD in1962 (Gutekunst, 1968) is still one of the mostwidely used in research and diagnostics and as acomponent in vaccines. As for other old cytopathicstrains, the noncytopathic counterpart of the cyto-pathic NADL strain had been lost. However, re-cently, this virus has made an unexpected appear-ance; it was found, through in vivo studies, to be stillresiding in a stock of cytopathic NADL virus origi-nally obtained from the American Type CultureCollection (Harding et al., 2002). Pregnant cows (at90–105 days of gestation) were inoculated withvirus originating from an infectious cDNA clone ofNADL (i-VVNADL), or the parental NADL virusstock (termed NADL-A), and 3 or 6 weeks after in-oculation the fetuses were harvested. Virus isolatedfrom cows inoculated with iVVNADL was alwayscytopathic in biotype and no virus was isolated fromfetuses from these cows. Surprisingly, viruses ofboth biotypes were initially isolated from cows in-oculated with NADL-A, but after 8 days postinocu-lation only noncytopathic virus was recovered. Viruswas isolated from 8 of 10 fetuses from dams inocu-lated with NADL-A. In each case, the virus wasnoncytopathic.

The nucleotide sequence of the NS2-3 and NS5Aregions of the genome of the contaminating noncy-topathic BVDV (termed NADL-1102) in theNADL-A stock was determined and the virus wasfound to lack the 270 base pair cellular insertion inthe NS2 gene of NADL-A. Aside from the lack ofthe cellular insertion, which had been shown pre-viously to be responsible for cytopathogenicity(Mendez et al., 1998), a greater than 99% homologyexists between sequences of NADL-1102 and pub-lished NADL strain sequences. This level of homol-ogy indicates that NADL-1102 is the ancestral non-cytopathic NADL virus from which the cytopathicNADL virus arose by genetic recombination. Theauthors of this report state that there had been nosystematic attempts to eliminate noncytopathicBVDV from this stock by the repository personnelor the depositors (Gutekunst and Malmquist, 1963).



This study illustrates well, and confirms the find-ings of others, that noncytopathic BVDV, but not cy-topathic BVDV, can infect the fetus and establishpersistent infections (McClurkin et al., 1984; Brown-lie et al., 1989). Prior to 1984, it was not appreciatedthat both BVDV biotypes could be isolated fromcases of MD. Thus, cytopathic BVDV isolates maynot have always been purified or adequately purified.For instance, plaque purification may not be ade-quate for purification of cytopathic BVDV withoutadditional immunochemical or immunofluorescentstaining to rule out the presence of noncytopathicBVDV. McClurkin et al. (1984), in their classic studyon the production of immune-tolerant and persist-ently infected calves, produced persistent infectionswith noncytopathic BVDV but not with the cyto-pathic NADL strain. Apparently, their NADL stockvirus was free of noncytopathic BVDV.

DIAGNOSIS

Within the past few years, several studies have ex-amined the use of immunohistochemistry (IHC) onskin biopsies to detect BVDV infections, a tech-nique that had been introduced earlier by Thür et al.(1996) for use on frozen specimens. The skin biopsytechnique is now being used by some laboratories toscreen herds for persistently infected cattle. Njaa etal. (2000) applied the technique on formalin-fixed,paraffin-embedded specimens and were able to de-tect positive staining in 41 of 42 skin samples frompersistently infected calves. Sections of skin fromall acutely infected animals were negative for stain-ing when they were infected with 105 TCID50 ofBVDV, and only 40% of acutely infected animalsshowed positive skin biopsies when infected with amuch higher dose (108 TCID50) of BVDV. It wasconcluded that immunohistochemical staining ofskin biopsies was an effective method for identify-ing persistently infected cattle.

Ridpath et al. (2002) used virus isolation as thestandard to evaluate both the skin biopsy methodand a serum PCR-probe test, another technique usedfor screening herds for persistently infected animals,for detection of acute BVDV infections. In thisstudy, 16 animals were inoculated with 106 TCID50of BVDV 1 or BVDV 2. Whereas virus was isolatedfrom all animals, only three (19%) animals werepositive by the PCR-probe test and none was posi-tive by immunohistochemical (IHC) staining of skinbiopsies. These researchers concluded that the skinbiopsy test would usually not confuse persistentlyinfected animals with acutely infected animals.However, both methods (IHC and PCR-probe) were

considered unreliable for detecting acute outbreakscaused by BVDV.

Recently, the use of skin biopsies as diagnosticspecimens for a BVDV antigen capture ELISA wasevaluated (Brodersen et al., 2002). The assay, basedon the detection of the Erns protein, was reported tobe equally sensitive to PCR, IHC, and virus isolationfrom leukocytes. The same capture ELISA had beenpreviously validated for the detection of BVDV inserum for screening herds for persistently infectedanimals and for screening BVDV in sublots of fetalbovine serum for the production of the commercial-ized product (Plavsic and Prodafikas, 2001).

The use of pooled samples for diagnostic testing,a trend that began in the 1990s, has continued in re-cent years. Weinstock et al. (2001) developed a sin-gle tube PCR assay for detection of BVDV inpooled serum. The assay was compared with a mi-croplate virus isolation method performed on indi-vidual sera. The PCR assay was sensitive enough todetect a single viremic serum sample in 100 pooledsamples. Muñoz-Zanzi et al. (2000) examined least-cost strategies for using PCR/probe testing ofpooled blood samples to identify persistently in-fected animals in herds. For a herd prevalence of1%, the least-cost strategy for pooled sample testingwas to test pools from 20 animals initially, and thenpools of five blood samples for repooled testing. Asherd prevalence increased beyond 3%, the competi-tive benefit of pooled testing diminished.

Based on earlier work of Houe (1992; 1994) andHoue et al. (1995), who reported that antibody lev-els can be used to predict the presence of persist-ently infected animals in a herd, Pillars and Grooms(2002) applied this technique to herds with un-known BVDV status. Specifically, they examinedwhether serological evaluation of five unvaccinated,randomly selected, 6- to 12-month-old heifers is avalid method for identifying herds that contain per-sistently infected cattle. A herd was classified aslikely to contain persistently infected animals whenat least three of the heifers had neutralizing antibodytiters of �1:128 to BVDV 1 or BVDV 2. Of the 14herds examined, 6 contained persistently infectedanimals. The sensitivity and specificity of the sero-logic method for determining which herds had per-sistently infected animals was 66% and 100%, re-spectively. In one herd with a false negative result,by chance three of the heifers examined serologi-cally were persistently infected animals and hadtiters of <1:4, whereas the other two animals hadtiters of >1:4,096. The authors noted that this herdwould have been rightly classified had virus isola-



tion been used in addition to the serological testing.In the other herd with a false negative result, a sin-gle persistently infected calf of 3 months of age wasmissed. This animal was reported not to have hadcontact with the target population. The authorsnoted that the overall sensitivity of the serologicevaluation of five unvaccinated heifers for determin-ing herds with persistently infected cattle could beimproved by periodic retesting of herds that hadnegative results. The use and testing of sentinel ani-mals in vaccinated herds has been recommended asan important element of a BVDV control program(Dubovi, 2002).

CONTROL BY VACCINATION

In recent years there have been two major trends invaccine research and development: the inclusion ofBVDV 2 strains in commercial vaccines and theevaluation of the efficacy of BVDV vaccines forprotection against fetal infection. The latter, an as-sessment of efficacy for fetal protection, has notbeen a requirement for the licensing of vaccines inthe U.S. and Canada. For licensing, the vaccines arerequired only to show protection against acute infec-tions. However, in recent years, efforts by both indi-vidual researchers and vaccine companies have beenintensified to evaluate BVDV vaccines for their effi-cacy in providing fetal protection. The ability todemonstrate and claim vaccine efficacy for fetal pro-tection should give those vaccine companies a com-petitive advantage over others that cannot make sucha claim for their products. This will be the drivingforce for new and better products in the absence ofnew regulations.

Recently, experiments involving immunizationwith BVDV-specific nucleic acids (DNA or RNA)have been performed. The first of these utilized arecombinant plasmid containing the BVDV E2gene with a human cytomegalovirus promotor forDNA immunization of mice (Harpin et al., 1997).Neutralizing responses against BVDV 1 strainswere generated, which persisted for 6 months afterthe last injection. In cattle, immunization with theplasmid produced both a neutralizing antibody re-sponse and a cellular immune response, but immu-nization was only partially protective when the cat-tle were inoculated with a challenge virus (Harpinet al., 1999).

Vassilev et al. (2001) used full-length, infectiousBVDV RNA derived from a recombinant, infectiouscDNA clone (NADL strain) to immunize cattle andsheep against BVDV. The RNA was coated ontogold microcarrier particles and delivered to the inoc-

ulation sites using a gene gun delivery system. Se-rumneutralizing antibody titers of >212 and >27

were generated in calves to the NADL strain ofBVDV 1 and to a BVDV 2 strain, respectively.

Muñoz-Zanzi et al. (2002) developed models topredict the age of dairy calves when colostrum-derived BVDV antibodies would decline to a levelthat would no longer offer disease protection or in-terfere with vaccination. The ages at which 50% ofthe calves were predicted to have low antibody titers(<1:16) were, for two herds, 107–111 days forBVDV 1 and 75–81 days for BVDV 2. The ages atwhich 50% of the calves were predicted to beseronegative (titer <1:8) were, for the same herds,139–143 days for BVDV 1 and 110–118 days forBVDV 2. These ages were considerably shorter thanearlier estimates for the persistence of colostral an-tibodies (6–10 months). The differences in these es-timates may be due, in part, to the large sample size(466 calves) used in this study or due to differencesin protocols for administering colostrum.

ERADICATION

The eradication programs without vaccination thatbegan in the 1990s in the Scandinavian countrieshave proven to be very successful in reducing thenumber of herds with BVDV infections. For in-stance, 52% of herds were estimated to be infectedin Sweden in 1993, whereas in 2002, 93% of dairyand 88% of beef herds were considered BVDV-free(Lindberg, 2002). Eradication of BVDV appears tohave been even more successful in Denmark where,in 2002, approximately 99% of dairy and 98% ofbeef herds were free of BVDV (Bitsch et al., 2002).

In Germany, an ongoing control program foreventual BVDV eradication is voluntary and par-tially subsidized by the state (Greiser-Wilke, 2002).Vaccination is practiced in Germany to reduce viruscirculation and to minimize economic losses. Akilled vaccine is used for basic immunization and anattenuated live virus vaccine is used to booster theimmune response. Screening for removal of persist-ently infected animals is performed using an antigencapture ELISA. If all animals up to 3 years of age ina herd are negative, the herd is classified as “BVDVunsuspicious.”

In 2002, the Academy of Veterinary Consultants(AVC), presented a position statement on BVDV foreventual eradication of the virus in North America ata USDA-sponsored meeting (Grotelueschen, 2002).The AVC, established in 1970, is an association ofveterinarians involved in beef cattle medicine, herdhealth programs, and consultation. Its members con-



sist of veterinarians from both the U.S. and Canada.The BVDV position statement reads:

The beef and dairy industries suffer enormousloss due to effects of bovine viral diarrhea virus(BVDV) infection. The highly mutable nature ofBVDV and the emergence of highly virulentstrains of BVDV contribute to limited success ofpresent control programs. Also persistently in-fected cattle are the primary source of infectionand effective testing procedures are available toidentify those infected carriers. Therefore, it isthe resolve of the Academy of VeterinaryConsultants that the beef and dairy industriesadopt measures to control and target eventualeradication of BVDV from North America.

The AVC position statement is likely only a firststep toward a more comprehensive program againstBVDV in North America, and the support of pro-ducer groups will be critical before such a programbecomes a reality. The consensus of those presentwas that any eradication program in North Americawill likely include vaccination, and that vaccinesthat provide maximum fetal protection will be cru-cial to the success of this type of program.

REFERENCESAdler B, Adler H, Pfister H, et al.: 1997, Macro-

phages infected with cytopathic bovine viral diar-rhea virus release a factor(s) capable of priminguninfected macrophages for activation-inducedapoptosis. J Virol 71:3255–3258.

Alenius S, Lindberg A, Larsson B: 1996, Control ofBVDV in Sweden. Int Symp Bovine Viral DiarrheaVirus: A 50 Year Review June 23–25. Ithaca, NewYork, pp. 127–129.

Alenius S, Lindberg A, Larsson B: 1997, A nationalapproach to the control of bovine viral diarrhoeavirus. In: Proc 3rd ESVV Symp Pestivirus InfectionsSeptember 19–20. Eds. Edwards S, Paton DJ, andWensvoort G, pp. 162–169. Lelystad, TheNetherlands.

Baigent SJ, Zhang G, Fray MD, et al.: 2002, Inhibi-tion of beta interferon transcription by noncytopath-ogenic bovine viral diarrhea virus is through an in-terferon regulatory factor 3-dependent mechanism.J Virol 76:8979–8988.

Bak A, Callesen H, Meyling A, Greve T: 1992, Calvesborn after embryo transfer from donors persistentlyinfected with BVD virus. Vet Rec 131:37.

Baker JA, York CJ, Gillespie JH, Grayson BM: 1954,Virus diarrhea in cattle. Am J Vet Res 15:525–531.

Baker JC: 1987, Bovine viral diarrhea virus: A review.J Am Vet Med Assoc 190:1449–1458.

Barlow RM, Nettleton PF, Gardiner AC, et al.: 1986,Persistent bovine virus diarrhoea virus infection ina bull. Vet Rec 118:321–324.

Baule C, Kulcsár G, Belák K, et al.: 2001, Pathogene-sis of primary respiratory disease induced by iso-lates from a new genetic cluster of bovine viral di-arrhea virus type 1. J Clin Microbiol 39:146–153.

Baule C, van Vuuren M, Lowings JP, Belák S: 1997,Genetic heterogeneity of bovine viral diarrhoeaviruses isolated in Southern Africa. Virus Res52:205–220.

Becher P, König M, Paton DJ, Thiel H-J: 1995,Further characterization of border disease virus iso-lates: Evidence for the presence of more than threespecies within the genus pestivirus. Virology209:200–206.

Becher P, Orlich M, Shannon AD, et al.: 1997, Phylo-genetic analysis of pestiviruses from domestic andwild ruminants. J Gen Virol 78:1357–1366.

Becher P, Orlich M, Thiel H-J: 2001, RNA recombi-nation between persisting pestivirus and a vaccinestrain: Generation of cytopathic virus and inductionof lethal disease. J Virol 75:6256–6264.

Bielefeldt-Ohmann H: 1983, Pathogenesis of bovineviral diarrhoea-mucosal disease: Distribution andsignificance of BVD-antigen in diseased calves.Res Vet Sci 34:5–10.

Bitsch V, Hansen K-EL, Larsen KB, et al.: 2002, Theeradication of BVD in Denmark. Proc Detectingand Controlling BVDV Infections April 4–5. Ames,Iowa, p. 46.

Bitsch V, Hansen K-EL, Rønsholt L: 2000,Experiences from the Danish programme for eradi-cation of bovine virus diarrhoea (BVD) 1994–1998with special reference to legislation and causes ofinfection. Vet Microbiol 77:137–143.

Bitsch V, Rønsholt L: 1995, Control of bovine viraldiarrhea virus infection without vaccines. Vet ClinNorth Am Food Anim Pract 11:627–640.

Bittle JL: 1968, Vaccination for bovine viral diarrhea-mucosal disease. J Am Vet Med Assoc 152:861–868.

Bolin SR: 1990, Control of bovine viral diarrheavirus. Rev Sci Tech Off Int Epiz 9:163–171.

Bolin SR, McClurkin AW, Coria MF: 1985a,Frequency of persistent bovine viral diarrhea virusinfection in selected cattle herds. Am J Vet Res46:2385–2387.

Bolin SR, McClurkin AW, Cutlip RC, Coria MF:1985b, Response of cattle persistently infected withnoncytopathic bovine viral diarrhea virus to vacci-nation for bovine viral diarrhea and to subsequentchallenge exposure with cytopathic bovine viral di-arrhea virus. Am J Vet Res 46:2467–2470.

Bolin SR, McClurkin AW, Cutlip RC, Coria MF:1985c, Severe clinical disease induced in cattle per-



sistently infected with noncytopathic bovine viraldiarrhea virus by superinfection with cytopathicbovine viral diarrhea virus. Am J Vet Res46:573–576.

Brinkhof J, Zimmer G, Westenbrink F: 1996, Com-parative study on four enzyme-linked immunosor-bent assays and a cocultivation assay for the detec-tion of antigens associated with the bovine viraldiarrhoea virus in persistently infected cattle. VetMicrobiol 50:1–6.

Brock KV, Grooms DL, Ridpath J, Bolin SR: 1998,Changes in levels of viremia in cattle persistentlyinfected with bovine viral diarrhea virus. J VetDiagn Invest 10:22–26.

Brock KV, Lapin DR, Skrade DR: 1997, Embryotransfer from donor cattle persistently infected withbovine viral diarrhea virus. Theriogenology47:837–844.

Brodersen BW, Huchzermeier R, Dubovi EJ, SalikiJT: 2002, Utilization of an antigen-capture ELISAon extracts of fresh skin (ACEsk) samples fordetection of BVDV. Proc Detecting and Control-ling BVDV Infections April 4–5. Ames, Iowa,pp. 32.

Brownlie J, Clarke MC: 1993, Experimental andspontaneous mucosal disease of cattle: A validationof Koch’s postulates in the definition of pathogene-sis. Intervirology 35:51–59.

Brownlie J, Clarke MC, Hooper LB, Bell GD: 1995,Protection of the bovine fetus from bovine viral di-arrhoea virus by means of a new inactivated vac-cine. Vet Rec 137:58–62.

Brownlie J, Clarke MC, Howard CJ: 1984, Experi-mental production of fatal mucosal disease in cattle.Vet Rec 114:535–536.

Brownlie J, Clarke MC, Howard CJ: 1989, Experi-mental infection of cattle in early pregnancy with acytopathic strain of bovine virus diarrhoea virus.Res Vet Sci 46:307–311.

Brownlie J, Clarke MC, Howard CJ, Pocock DH:1986, Mucosal disease: the dilemma of experimen-tal disease. In: Proc 14th World Congress onDiseases of Cattle, Vol 1. Eds. Hartigan PJ andMonaghan ML, pp.199–203. Dublin, Ireland.

Bürki F, Germann E: 1964, Letale pneumoenteritidenbei kalbern, verursacht durch die erreger der bovinevirusdiarrhoe. Berl Munch Tierarztl Wochenschr16:324–326.

Carman S, van Dreumel T, Ridpath J, et al.: 1998,Severe acute bovine viral diarrhea in Ontario,1993–1995. J Vet Diagn Invest 10:27–35.

Charleston B, Brackenbury LS, Carr BV, et al.: 2002,Alpha/beta and gamma interferons are induced byinfection with noncytopathic bovine viral diarrheavirus in vivo. J Virol 76:923–927.

Charleston B, Fray MD, Baigent S, et al.: 2001,Establishment of persistent infection with non-cytopathic bovine viral diarrhoea virus in cattle isassociated with a failure to induce type I interferon.J Gen Virol 82:1893–1897.

Childs T: 1946, X disease of cattle—Saskatchewan.Can J Comp Med 10:316–319.

Clark JG: 1968, In: Holper JC: Comments on bovineviral diarrhea-mucosal disease. J Am Vet Med Assoc152:868–870.

Coggins L, Gillespie JH, Robson DS, et al.: 1961,Attenuation of virus diarrhea virus (strain OregonC24V) for vaccine purposes. Cornell Vet54:539–545.

Collett MS, Larson R, Belzer SK, Retzel E: 1988a,Proteins encoded by bovine viral diarrhea virus:The genomic organization of a pestivirus. Virology165:200–208.

Collett MS, Larson R, Gold C, et al.: 1988b,Molecular cloning and nucleotide sequence of thepestivirus bovine viral diarrhea virus. Virology165:191–199.

Corapi WV, Donis RO, Dubovi EJ: 1988, Monoclonalantibody analyses of cytopathic and noncytopathicviruses from fatal bovine viral diarrhea virus infec-tions. J Virol 62:2823–2827.

Corapi WV, Donis RO, Dubovi EJ: 1990a,Characterization of a panel of monoclonal antibod-ies and their use in the study of the antigenic diver-sity of bovine viral diarrhea virus. Am J Vet Res51:1388–1394.

Corapi WV, Elliot RD, French TW, et al.: 1990b,Thrombocytopenia and hemorrhages in veal calvesinfected with bovine viral diarrhea virus. J Am VetMed Assoc 196:590–596.

Corapi WV, French TW, Dubovi EJ: 1989, Severethrombocytopenia in young calves experimentallyinfected with noncytopathic bovine viral diarrheavirus. J Virol 63:3934–3943.

Coria MF, McClurkin AW: 1978, Specific immunetolerance in an apparently healthy bull persistentlyinfected with bovine viral diarrhea virus. J Am VetMed Assoc 172:449–451.

Cortese VS, West KH, Hassard LE, et al.: 1998,Clinical and immunologic responses of vaccinatedand unvaccinated calves to infection with a virulenttype-II isolate of bovine viral diarrhea virus. J AmVet Med Assoc 213:1312–1319.

Cutlip RC, McClurkin AW, Coria MF: 1980, Lesionsin clinically healthy cattle persistently infected with the virus of bovine viral diarrhea—glomerulonephritis and encephalitis. Am J Vet Res41:1938–1941.

Darbyshire JH: 1962, Agar gel diffusion studies witha mucosal disease of cattle. II. A serological rela-



tionship between a mucosal disease and swinefever. Res Vet Sci 3:123–128.

De Moerlooze, Lecomte C, Brown-Shimmer S, et al.:1993, Nucleotide sequence of the bovine viral diar-rhoea virus Osloss strain: Comparison with relatedviruses and identification of specific DNA probesin the 5’ untranslated region. J Gen Virol74:1433–1438.

Deregt D, Bolin SR, van den Hurk J, et al.: 1998a,Mapping of a type 1-specific and a type-commonepitope on the E2 (gp53) protein of bovine viral di-arrhea virus with neutralization escape mutants.Virus Res 53:81–90.

Deregt D, Carman PS, Clark RM, et al.: 2002, A com-parison of polymerase chain reaction with andwithout RNA extraction and virus isolation for de-tection of bovine viral diarrhea virus in youngcalves. J Vet Diagn Invest 14:433–437.

Deregt D, Prins S: 1998, A monoclonal antibody-based immunoperoxidase monolayer (micro-isolation) assay for detection of type 1 and type 2bovine viral diarrhea viruses. Can J Vet Res62:152–155.

Deregt D, van Rijn PA, Weins TY, van den Hurk J:1998b, Monoclonal antibodies to the E2 protein ofa new genotype (type 2) of bovine viral diarrheavirus define three antigenic domains involved inneutralization. Virus Res 57:171–181.

Diderholm H, Dinter Z: 1966, Interference betweenstrains of bovine virus diarrhea virus and their ca-pacity to suppress interferon of a heterologousvirus. Proc Soc Exp Biol Med 121:976–980.

Done JT, Terlecki S, Richardson C, et al.: 1980,Bovine virus diarrhoea-mucosal disease virus:Pathogenicity for the fetal calf following maternalinfection. Vet Rec 106:473–479.

Donis RO, Corapi W, Dubovi EJ: 1988, Neutralizingmonoclonal antibodies to bovine viral diarrhoeavirus bind to the 56K to 58K glycoprotein. J GenVirol 69:77–86.

Donis RO, Dubovi EJ: 1987, Differences in virus-induced polypeptides in cells infected by cytopathicand noncytopathic biotypes of bovine virus diar-rhea-mucosal disease virus. Virology 158:168–173.

Drake TR, Moore DA, Whitlock RH, et al.: 1996, Anoutbreak of acute BVD in Pennsylvania cattle. IntSymp Bovine Viral Diarrhea Virus: A 50 YearReview June 23–25. Ithaca, New York, pp. 208.

Drew TW, Yapp F, Paton DJ: 1999, The detection ofbovine viral diarrhoea in bulk milk samples by theuse of a single-tube RT-PCR. Vet Microbiol64:145–154.

Dubovi EJ: 2002, Personal communication.Edwards S, Moennig V, Wensvoort G: 1991, The de-

velopment of an international reference panel of

monoclonal antibodies for the differentiation of hogcholera virus from other pestiviruses. Vet Microbiol29:101–108.

Edwards S, Sands JJ, Harkness JW: 1988, The appli-cation of monoclonal antibody panels to character-ize pestivirus isolates from ruminants in GreatBritain. Arch Virol 102:197–206.

Fernelius AL: 1964, Noncytopathogenic bovine viraldiarrhea viruses detected and titrated by immuno-fluorescence. Can J Comp Med Vet Sci 28:121–126.

Fernelius AL, Amtower WC, Lambert G, et al.: 1973,Bovine viral diarrhea in swine: Characteristics ofvirus recovered from naturally and experimentallyinfected swine. Can J Comp Med 37:13–20.

Flores EF, Ridpath JF, Weiblen R, et al.: 2002, Phylo-genetic analysis of Brazilian bovine viral diarrheavirus type 2 (BVDV-2) isolates: Evidence for asubgenotype within BVDV-2. Virus Res 87:51–60.

Fox FH: 1996, Historic clinical perspective. Int SympBovine Viral Diarrhea Virus: A 50 Year ReviewJune 23–25. Ithaca, New York, pp. 1–4.

Fray MD, Clarke MC, Thomas LH, et al.: 1998, Pro-longed nasal shedding and viraemia of cytopatho-genic bovine viral diarrhoea virus in experimentallate-onset mucosal disease. Vet Rec 143:608–611.

Fritzemeier J, Greiser-Wilke I, Haas L, et al.: 1995,Experimentally induced “late-onset” mucosal dis-ease—characterization of the cytopathogenicviruses isolated. Vet Microbiol 46:285–294.

Fritzemeier J, Haas L, Liebler E, et al.: 1997, The de-velopment of early vs. late onset mucosal disease isa consequence of two different pathogenic mecha-nisms. Arch Virol 142:1335–1350.

Fuller DA: 1965, When to vaccinate for IBR-VD.Mod Vet Prac 46:40–43.

Fulton RW, Ridpath JF, Saliki JT, et al.: 2002, Bovineviral diarrhea virus (BVDV) 1b: PredominantBVDV subtype in calves with respiratory disease.Can J Vet Res 66:181–190.

Gilbert SA, Burton KM, Prins SE, Deregt D: 1999,Typing of bovine viral diarrhea viruses directlyfrom blood of persistently infected cattle by multi-plex PCR. J Clin Microbiol 37:2020–2023.

Gillespie JH: 1996, Cornell’s contributions to bovinevirus diarrhea mucosal disease research. Int SympBovine Viral Diarrhea Virus: A 50 Year ReviewJune 23–25. Ithaca, New York, pp. 5–10.

Gillespie JH, Baker JA, McEntee: 1960, A cytopathicstrain of virus diarrhea virus. Cornell Vet 50:73–79.

Gillespie JH, Bartholomew PR, Thompson RG,McEntee: 1967, The isolation of noncytopathicvirus diarrhea virus from two aborted bovine fe-tuses. Cornell Vet 57:564–571.

Gillespie JH, Coggins L, Thompson J, Baker JA:1961, Comparison by neutralization tests of strains



of virus isolated from virus diarrhea and mucosaldisease. Cornell Vet 51:155–159.

Gillespie JH, Madin SH, Darby N: 1962, Cellularresistance in tissue culture induced by non-cytopathogenic strains to a cytopathogenic strain ofvirus diarrhea virus of cattle. Proc Soc Exp BiolMed 110:248–250.

Greiser-Wilke I: 2002, BVDV eradication programs inEurope. Proc Detecting and Controlling BVDVInfections April 4–5. Ames, Iowa, pp. 20.

Grotelueschen D: 2002, Position statement on bovinevirus diarrhea virus: Academy of VeterinaryConsultants. Proc Detecting and Controlling BVDVInfections April 4–5. Ames, Iowa, pp. 22.

Gutekunst DE: 1968, Comments on vaccination forbovine viral diarrhea-mucosal disease. J Am VetMed Assoc 152:865–866.

Gutekunst DE, Malmquist WA: 1963, Separation of asoluble antigen and infectious particles of bovineviral diarrhea viruses and their relationship to hogcholera. Can J Comp Med Vet Sci 27:121–123.

Haines DM, Clark EG, Dubovi EJ: 1992, Monoclonalantibody-based immunohistochemical detection ofbovine viral diarrhea virus in formalin-fixed,paraffin-embedded tissues. Vet Pathol 29:27–32.

Haines DM, Martin KM, Clark EG, et al.: 2001, Theimmunohistochemical detection of mycoplasmabovis and bovine viral diarrhea virus in tissues offeedlot cattle with chronic unresponsive respiratorydisease and/or arthritis. Can Vet J 42:857–860.

Harding MJ, Cao X, Shams H, et al.: 2002, Role ofbovine viral diarrhea virus biotype in the estab-lishment of fetal infections. Am J Vet Res63:1455–1463.

Harpin S, Elahi SM, Cornaglia E, et al.: 1995, The 5’-untranslated region sequence of a potential newgenotype of bovine viral diarrhea virus. Arch Virol140:1285–1290.

Harpin S, Hurley DJ, Mbikay M, et al.: 1999,Vaccination of cattle with a DNA plasmid encodingthe bovine viral diarrhoea virus major glycoproteinE2. J Gen Virol 80:3137–3144.

Harpin S, Talbot B, Mbikay M, Elazhary Y: 1997, Im-mune response to vaccination with DNA encodingthe bovine viral diarrhea virus major glycoproteingp53 (E2). FEMS Microbiol Letters 146:229–234.

Hoff HS, Donis RO: 1997, Induction of apoptosis andcleavage of poly (ADP-ribose) polymerase by cyto-pathic bovine viral diarrhea virus infection. VirusRes 49:101–113.

Houe H: 1992, Serological analysis of a small herdsample to predict presence or absence of animalspersistently infected with bovine viral diarrhoeavirus (BVDV) in dairy herds. Res Vet Sci53:320–323.

Houe H: 1994, Bovine viral diarrhoea virus: Detectionof Danish dairy herds with persistently infected ani-mals by means of a screening test of ten youngstock. Prev Vet Med 19:241–248.

Houe H: 1996, BVDV in Denmark—Comparisonwith Michigan. Int Symp Bovine Viral DiarrheaVirus: A 50 Year Review June 23–25. Ithaca, NewYork, pp. 121–126.

Houe H: 1999, Epidemiological features and econom-ical importance of bovine virus diarrhoea virus(BVDV) infections. Vet Microbiol 64:89–107.

Houe H, Baker JC, Maes RK, et al.: 1995, Applicationof antibody titers against bovine viral diarrhea virus(BVDV) as a measure to detect herds with cattlepersistently infected with BVDV. J Vet Diagn Invest7:327–332.

Howard CJ, Brownlie J, Clarke MC. Comparison bythe neutralization assay of pairs of non-cytopatho-genic and cytopathogenic strains of bovine virusdiarrhoea virus isolated from cases of mucosaldisease. Vet Microbiol 13:361–369.

Inaba Y, Tanaka Y, Kumagai T, et al.: 1968, Bovine di-arrhea virus. II. END phenomenon: Exaltation ofNewcastle disease virus in bovine cells infectedwith bovine diarrhea virus. Jpn J Microbiol12:35–49.

Jericho KWF, Langford EV: 1978, Pneumonia incalves produced with aerosals of bovineherpesvirus-1 and Pasturella haemolytica. Can JComp Med 42:269–277.

Johnson DW, Muscoplat CC: 1973, Immunologic ab-normalities in calves with chronic bovine viraldiarrhea. Am J Vet Res 34:1139–1141.

Kahrs RF: 1971, Differential diagnosis of bovine viraldiarrhea-mucosal disease. J Am Vet Med Assoc159:1383–1386.

Kahrs RF, Scott FW, de Lahunta A: 1970, Bovineviral diarrhea-mucosal disease, abortion, and con-genital cerebellar hypoplasia in a dairy herd. J AmVet Med Assoc 156:851–857.

Kennedy P, Collier JC, Ramsey F, et al.: 1968, Reportof the ad hoc committee on terminology for thesymposium on immunity to the bovine respiratorydisease complex. J Am Vet Med Assoc 152:940.

Kirkland PD, Mackintosh SG, Moyle A: 1994, Theoutcome of widespread use of semen from a bullpersistently infected with pestivirus. Vet Rec135:527–529.

Kirkland PD, Richards SG, Rothwell JT, Stanley DF:1991, Replication of bovine viral diarrhoea virus inthe bovine reproductive tract and excretion of virusin semen during acute and chronic infections. VetRec 128:587–590.

Kniazeff AJ, Huck RA, Jarrett WFH, et al.: 1961, An-tigenic relationship of some bovine viral diarrhoea-



mucosal disease viruses from the United States,Great Britain, and West Germany. Vet Rec73:768–769.

Kniazeff AJ, Rimer V, Gaeta L: 1967: Gamma-globulin in foetal bovine sera: Significance invirology. Nature (London) 214:805–806.

Kümmerer BM, Meyers G: 2000, Correlation betweenpoint mutations in NS2 and the viability and cy-topathogenicity of bovine viral diarrhea virus strainOregon analyzed with an infectious cDNA clone. JVirol 74:390–400.

Kümmerer BM, Stoll D, Meyers G: 1998, Bovineviral diarrhea virus strain Oregon: A novel mecha-nism for processing of NS2-3 based on point muta-tions. J Virol 72:4127–4138.

Lambert G, Fernelius AL, Smith RL: 1971,Immunogenicity in calves of a bovine viral diarrheavaccine inactivated with beta-propiolactone. Proc75th Annual Meeting U.S. Anim Hlth Assoc, pp.84–89.

Lambert G, McClurkin AW, Fernelius AL: 1974,Bovine viral diarrhea in the neonatal calf. J Am VetMed Assoc 164:287–289.

Lecomte C, Pin JJ, De Moerlooze L, et al.: 1990,ELISA detection of bovine viral diarrhoea virusspecific antibodies using recombinant antigen and monoclonal antibodies. Vet Microbiol23:193–201.

Lee KM, Gillespie JH: 1957, Propagation of virus di-arrhea virus of cattle in tissue culture. Am J Vet Res18:952–953.

Liess B: 1996, Cytopathogenicity discovery and im-pact in BVD research. Int Symp Bovine ViralDiarrhea Virus: A 50 Year Review June 23–25.Ithaca, New York, pp. 11–15.

Liess B, Orban S, Frey H-R, et al.: 1984, Studies ontransplacental transmissibility of a bovine virus di-arrhoea (BVD) vaccine virus in cattle. II.Inoculation of pregnant cows without detectableneutralizing antibodies to BVD virus 90–229 daysbefore parturition (51st to 190th day of gestation).Zbl Vet Med B 31:669-681.

Lindberg A: 2002, The Swedish BVDV scheme in thefinal phase—Measures needed to conclude theeradication and to maintain future freedom. ProcDetecting and Controlling BVDV Infections April4–5. Ames, Iowa, pp. 45.

Lobmann M, Charlier P, Florent G, Zygraich N: 1984,Clinical evaluation of a temperature-sensitivebovine viral diarrhea vaccine strain. Am J Vet Res45:2498–2503.

Mahlum CE, Haugerud S, Shivers JL, et al.: 2002,Detection of bovine viral diarrhea virus by TaqManreverse transcription polymerase chain reaction. JVet Diagn Invest 14:120–125.

Maisonnave J, Rossi CR: 1982, A microtiter test fordetecting and titrating noncytopathogenic bovineviral diarrhea virus. Arch Virol 72:279–287.

Malmquist WA: 1968, Bovine viral diarrhea-mucosaldisease: etiology, pathogenesis and applied immu-nity. J Am Vet Med Assoc 152:763–768.

Marcus SJ, Moll T: 1968, Adaptation of bovine viraldiarrhea virus to the Madin-Darby bovine kidneycell line. Am J Vet Res 29:817–819.

Martin SW, Meek AH, Davis DG, et al.: 1981, Factorsassociated with morbidity and mortality in feedlotcalves: The Bruce County beef cattle project, yeartwo. Can J Comp Med 45:103–112.

Martin SW, Meek AH, Davis DG, et al.: 1980, Factorsassociated with mortality in feedlot cattle: TheBruce County beef cattle project. Can J Comp Med44:1–10.

Martin SW, Nagy E, Armstrong D, Rosendal S: 1999,The association of viral and mycoplasmal antibodytiters with respiratory disease and weight gain infeedlot calves. Can Vet J 40:560–568.

McClurkin AW, Bolin SR, Coria MF: 1985, Isolationof cytopathic and noncytopathic bovine viral diar-rhea virus from the spleen of cattle acutely andchronically affected with bovine viral diarrhea. JAm Vet Med Assoc 186:568–569.

McClurkin AW, Coria MF, Cutlip RC: 1979,Reproductive performance of apparently healthycattle persistently infected with bovine viral diar-rhea virus. J Am Vet Med Assoc 174:1116–1119.

McClurkin AW, Coria MF, Smith RL: 1975,Evaluation of acetylethyleneimine-killed bovineviral diarrhea-mucosal disease virus (BVD) vaccinefor prevention of BVD infection of the fetus. ProcAnnu Mtg U.S. Anim Hlth Assoc 79:114–123.

McClurkin AW, Littledike ET, Cutlip RC, et al.: 1984,Production of cattle immunotolerant to bovine viraldiarrhea virus. Can J Comp Med 48:156–161.

McGoldrick A, Bensaude E, Ibata G, et al.: 1999,Closed one-tube reverse transcription nested poly-merase chain reaction for the detection of pestiviralRNA with fluorescent probes. J Virol Methods79:85–95.

McKercher DG, Saito JK, Crenshaw GL, BushnellRB: 1968, Complications in cattle following vac-cination with a combined bovine viral diarrhea-infectious bovine rhinotracheitis vaccine. J Am VetMed Assoc 152:1621–1624.

Mendez E, Ruggli N, Collett MS, Rice CM: 1998,Infectious bovine viral diarrhea virus (strainNADL) RNA from stable cDNA clones: A cellularinsert determines NS3 production and viral cy-topathogenicity. J Virol 72:4737–4745.

Meyer C, von Freyburg M, Elbers K, Meyers G: 2002,Recovery of virulent and Rnase-negative attenuated



type 2 bovine viral diarrhea viruses from infectiouscDNA clones. J Virol 76:8494–8503.

Meyers G, Rümenapf T, Thiel H-J: 1989, Ubiquitin ina togavirus. Nature (London) 341:491.

Meyers G, Tautz N, Becher P, et al.: 1996, Recoveryof cytopathogenic and noncytopathogenic bovineviral diarrhea viruses from cDNA constructs. JVirol 70:8606–8613.

Meyers G, Thiel H-J: 1996, Molecular characteriza-tion of pestiviruses. Adv Virus Res 47:53–118.

Meyling A: 1984, Detection of BVD virus in viremic cattle by an indirect immunoperoxidasetechnique. In: Recent Advances in Virus Diag-nosis. Eds. McNulty MS and MacFerran JB,pp. 37–46. Martinus Nijhoff Publishers,Boston, MA.

Meyling A, Jensen AM: 1988, Transmission of bovineviral diarrhoea virus (BVDV) by artificial insemi-nation (AI) with semen from a persistently infectedbull. Vet Microbiol 17:97–105.

Moennig V, Frey HR, Liebler E, et al.: 1990,Reproduction of mucosal disease with cytopatho-genic bovine viral diarrhoea virus selected in vitro.Vet Rec 127:200–203.

Moennig V, Greiser-Wilke I, Frey HR, et al.: 1993,Prolonged persistence of cytopathogenic bovineviral diarrhea virus (BVDV) in a persistentlyviremic cattle. J Vet Med B 40:371–377.

Muñoz-Zanzi CA, Johnson WO, Thurmond MC,Hietala SK: 2000, Pool-sample testing as a herd-screening tool for detection of bovine viral diarrheavirus persistently infected cattle. J Vet Diagn Invest12:195–203.

Muñoz-Zanzi CA, Thurmond MC, Johnson WO,Hietala SK: 2002, Predicted ages of dairy calveswhen colostrum-derived bovine viral diarrhea virusantibodies would no longer offer protection againstdisease or interfere with vaccination. J Am Vet MedAssoc 221:678–685.

Nakamura S, Shimazaki T, Sakamoto K, et al.: 1995,Enhanced replication of orbiviruses in bovine testi-cle cells infected with bovine viral diarrhoea virus.J Vet Med Sci 57:677–681.

Njaa BL, Clark EG, Janzen E, et al.: 2000, Diagnosisof persistent bovine viral diarrhea virus infection byimmunohistochemical staining of formalin-fixedskin biopsy specimens. J Vet Diagn Invest12:393–399.

Nuotio L, Juvonen M, Neuvonen E, et al.: 1999,Prevalence and geographical distribution of bovineviral diarrhoea (BVD) infection in Finland1993–1997. Vet Microbiol 64:231–235.

Nuttal PA, Luther PD, Stott EJ: 1977, Viral contami-nation of bovine foetal serum and cell cultures.Nature (London) 266:835–837.

Olafson P, McCallum AD, Fox FH: 1946, An appar-ently new transmissible disease of cattle. CornellVet 36:205–213.

Palfi V, Houe H, Philipsen J: 1993, Studies on thedecline of bovine virus diarrhoea virus (BVDV)maternal antibodies and detectability of BVDV inpersistently infected calves. Acta Vet Scand34:105–107.

Paton DJ, Brockman S, Wood L: 1990, Inseminationof susceptible and preimmunized cattle with bovineviral diarrhoea virus infected semen. Br Vet J146:171–174.

Paton DJ, Edwards S, Sands J, et al.: 1996, Antigenicvariation among pestiviruses. Int Symp Bovine ViralDiarrhea Virus: A 50 Year Review June 23–25.Ithaca, New York, pp. 61–64.

Paton DJ, Goodey R, Brockman S, Wood L: 1989,Evaluation of the quality and virological status ofsemen from bulls acutely infected with BVDV. VetRec 124:63–64.

Paton DJ, Ibata G, Edwards S, Wensvoort G: 1991,An ELISA detecting antibody to conserved pes-tivirus epitopes. J Virol Methods 31:315–324.

Paton DJ, Lowings JP, Barrett ADT: 1992, Epitopemapping of the gp53 envelope protein of bovineviral diarrhea virus. Virology 190:763–772.

Pellerin C, van den Hurk J, Lecomte J, Tijssen P:1994, Identification of a new group of bovine viraldiarrhea virus strains associated with severeoutbreaks and high mortalities. Virology203:260–268.

Perdrizet JA, Rebhun WC, Dubovi EJ, Donis RO:1987, Bovine virus diarrhea clinical syndromes indairy herds. Cornell Vet 77:46–74.

Peter CP, Tyler DE, Ramsey FK: 1967, Characteristicsof a condition following vaccination with bovinevirus diarrhea vaccine. J Am Vet Med Assoc150:46–52.

Pillars RB, Grooms DL: 2002, Serologic evaluation offive unvaccinated heifers to detect herds that havecattle persistently infected with bovine viral diar-rhea virus. Am J Vet Res 63:499–505.

Plant JW, Littlejohns IR, Gardiner AC, et al.: 1973,Immunological relationship between border dis-ease, mucosal disease and swine fever. Vet Rec92:455.

Plavsic MZ, Prodafikas G: 2001, Evaluation of a newsandwich enzyme-linked immunosorbent assay fordetection of bovine viral diarrhea virus in un-processed fetal bovine serum. J Vet Diagn Invest13:261–262.

Pocock DH, Howard CJ, Clarke MC, Brownlie J:1987, Variation in the intracellular polypeptide pro-files from different isolates of bovine viral diarrheavirus. Arch Virol 94:43–53.



Potgieter LND, McCracken MD, Hopkins FM, et al.:1984, Experimental production of bovine respira-tory tract infection with bovine viral diarrhea virus.Am J Vet Res 45:1582–1585.

Potgieter LND, McCracken MD, Hopkins FM, GuyJS: 1985, Comparison of the pneumopathogenicityof two strains of bovine viral diarrhea virus. Am JVet Res 46:151–153.

Pritchard WR: 1963, The bovine viral diarrhea-mucosal disease complex. In: Advances in Veterin-ary Science. Eds. Brandley CA and Jungherr EL,pp. 1–47. Academic Press, New York, NY.

Qvist P, Aasted B, Bloch B, et al.: 1990, Flow cyto-metric detection of bovine viral diarrhea virus inperipheral blood leukocytes of persistently infectedcattle. Can J Vet Res 54:469–472.

Qvist P, Houe H, Aasted B, Meyling A: 1991,Comparison of flow cytometry and virus isolationin cell culture for identification of cattle persist-ently infected with bovine viral diarrhea virus. JClin Microbiol 29:660–661.

Radostits OM, Littlejohns IR: 1988, New concepts inthe pathogenesis, diagnosis and control of diseasescaused by the bovine viral diarrhea virus. Can Vet J29:513–528.

Radwan GS, Brock KV, Hogan JS, Smith KL: 1995,Development of a PCR amplification assay as ascreening test using bulk milk samples for identify-ing dairy herds infected with bovine viral diarrheavirus. Vet Microbiol 44:77–92.

Rae AG, Sinclair JA, Nettleton PF: 1987, Survival ofbovine viral diarrhea virus in blood from persist-ently infected cattle. Vet Rec 120:504.

Ramsey FK, Chivers WH: 1953, Mucosal disease ofcattle. N Am Vet 34:629–633.

Rebhun WC, French TW, Perdrizet JA, et al.: 1989,Thrombocytopenia associated with acute bovinevirus diarrhea infection in cattle. J Vet Intern Med3:42–46.

Renard A, Dina D, Martial JA: 1987, Vaccines and di-agnostics derived from bovine viral diarrhea virus.European Patent Application No. 86870095.6.Publication No. 0208672.

Renshaw RW, Ray R, Dubovi EJ: 2000, Comparisonof virus isolation and reverse transcription poly-merase chain reaction assay for detection of bovineviral diarrhea virus in bulk milk tank samples. J VetDiagn Invest 12:184–186.

Revell SG, Chasey D, Drew TW, Edwards S: 1988,Some observations on the semen of bulls per-sisently infected with bovine virus diarrhoea virus.Vet Rec 123:122–125.

Ridpath JF: 1996, Sequence diversity and genotyping.Int Symp Bovine Viral Diarrhea Virus: A 50 YearReview June 23–25. Ithaca, New York, pp. 39–42.

Ridpath JF, Bolin SR: 1995, Delayed onset postvacci-nal mucosal disease as a result of genetic recombi-nation between genotype 1 and genotype 2 BVDV.Virology 212:259–262.

Ridpath JF, Bolin SR: 1998, Differentiation of types1a, 1b and 2 bovine viral diarrhoea virus (BVDV)by PCR. Mol Cell Probes 12:101–106.

Ridpath JF, Bolin SR, Dubovi EJ: 1994, Segregationof bovine viral diarrhea virus into genotypes.Virology 205:66–74.

Ridpath JF, Hietala SK, Sorden S, Neill JD: 2002,Evaluation of the reverse transcription-polymerasechain reaction/probe test of serum samples and im-munohistochemistry of skin sections for detectionof acute bovine viral diarrhea infections. J VetDiagn Invest 14:303–307.

Romváry J: 1965, Incidence of virus diarrhoea amongnewborn calves. Acta Vet 15:341–347.

Roth JA, Kaeberle ML: 1983, Suppression of neu-trophil and lymphocyte function induced by a vac-cinal strain of bovine viral diarrhea virus with andwithout the administration of ACTH. Am J Vet Res44:2366–2372.

Saliki JT, Fulton RW, Hull SR, Dubovi EJ: 1997,Microtiter virus isolation and enzyme immunoas-says for detection of bovine viral diarrhea virus incattle serum. J Clin Microbiol 35:803–807.

Schneider R, Unger G, Stark R, et al.: 1993,Identification of an RNA virus as a ribonuclease.Science 261:1169–1171.

Schroeder BA, Balassu-Chan TC: 1990, Specific se-quence amplification of bovine viral diarrhoea virusnucleic acid. Arch Virol 111:239–246.

Schweizer M, Peterhans E: 2001, Noncytopathicbovine viral diarrhea virus inhibits double-strandedRNA-induced apoptosis and interferon synthesis. JVirol 75:4692–4698.

Scott FW, Kahrs RF, De Lahunta A, et al.: 1973, Virusinduced congenital anomalies of the bovine fetus. I.Cerebellar degeneration (hypoplasia), ocular lesionsand fetal mummification following experimental in-fection with bovine viral diarrhea-mucosal diseasevirus. Cornell Vet 63:536–560.

Segre D: 1968, In: Gillespie JH: Comments on bovineviral diarrhea-mucosal disease. J Am Vet Med Assoc152:768–770.

Shahriar FM, Clark EG, Janzen E, et al.: 2002,Coinfection with bovine viral diarrhea virus andmycoplasma bovis in feedlot cattle with chronicpneumonia. Can Vet J 43:863–868.

Shannon AD, MacIntosh SG, Kirkland PD: 1992,Identification of pestivirus carrier calves by anantigen-capture ELISA. Aust Vet J 70:74–76.

Shimizu M, Satou K, Nishioka N, et al.: 1989,Serological characterization of viruses isolated



from experimental mucosal disease. Vet Microbiol19:13–21.

Smith AK, Grimmer AK: 2000, Birth of a BVDV-freecalf from a persistently infected embryo transferdonor. Vet Rec 146:49–50.

Sockett D, Bolin D, Ridpath J, Bolin S: 1996,Outbreak of severe acute bovine virus diarrhea(BVD) in Wisconsin. Int Symp Bovine ViralDiarrhea Virus: A 50 Year Review June 23–25.Ithaca, New York, pp. 207.

Sullivan DG, Akkina RK: 1995, A nested polymerasechain reaction assay to differentiate pestiviruses.Virus Res 38:231–239.

Sullivan DG, Chang GJ, Trent DW, Akkina RK: 1994,Nucleotide sequence analysis of the structural genecoding region of the pestivirus border disease virus.Virus Res 33:219–228.

Tamoglia TW: 1968, Laboratory evaluation of bovinerespiratory disease vaccines for safety. J Am VetMed Assoc 152:847–850.

Thomson RG, Savan M: 1963, Studies on virus diar-rhea and mucosal disease of cattle. Can J CompMed Vet Sci 27:207–214.

Thür B, Zlinsky K, Ehrensperger F: 1996,Immunohistochemical detection of bovine viral di-arrhea virus in skin biopsies: A reliable and fast di-agnostic tool. J Vet Med B 43:163–166.

Timoney PJ, McCollum WH: 1993, Equine viral ar-teritis. Vet Clin North Am Equine Pract 9:295–309.

Underdahl NR, Grace OD, Hoerlein AB: 1957,Cultivation in tissue-culture of cytopathogenicagent from bovine mucosal disease. Proc Soc ExpBiol Med 94:795–797.

van Campen H, Vorpahl P, Huzurbazar S, et al.: 2000,A case report: Evidence for type 2 bovine viral di-arrhea virus (BVDV)-associated disease in beefherds vaccinated with a modified-live type 1 BVDVvaccine. J Vet Diagn Invest 12:263–265.

van Dreumel T: 2002, Personal communication.van Oirschot JT, Bruschke CJM, van Rijn PA: 1999,

Vaccination of cattle against bovine viral diarrhoea.Vet Microbiol 64:169–183.

van Rijn PA: 1999, Personal communication.van Rijn PA, van Gennip HG, Leendertse CH, et al.:

1997, Subdivision of the pestivirus genus based onenvelope glycoprotein E2. Virology 237:337–348.

Vassilev VB, Collett MS, Donis RO: 1997, Authenticand chimeric full-length genomic cDNA clones ofbovine viral diarrhea virus that yield infectioustranscripts. J Virol 71:471–478.

Vassilev VB, Donis RO: 2000, Bovine viral diarrheavirus induced apoptosis correlates with increased

intracellular viral RNA accumulation. Virus Res69:95–107.

Vassilev VB, Gil LHVG, Donis RO: 2001,Microparticle-mediated RNA immunization againstbovine viral diarrhea virus. Vaccine 19:2012–2019.

Vilcek S, Paton DJ, Durkovic B, et al.: 2001, Bovineviral diarrhoea virus genotype 1 can be separatedinto at least eleven genetic groups. Arch Virol 146:99–115.

Voges H, Horner GW, Rowe S, Wellenberg GJ: 1998,Persistent bovine pestivirus infection localized inthe testes of an immuno-competent, non-viremicbull. Vet Microbiol 61:165–175.

Waage S, Krogsrud J, Nyberg O, Sandvik T: 1997,Results achieved by a national programme for theeradication of bovine virus diarrhoea. In Proc 3rdESVV Symp Pestivirus Infections September 19–20.Eds. Edwards S, Paton DJ, and Wensvoort G, pp.170–172. Lelystad, The Netherlands.

Walker RVL, Olafson P: 1947, Failure of virus diar-rhea of cattle to immunize against rinderpest.Cornell Vet 37:107–111.

Ward GM, Roberts SJ, McEntee K, Gillespie JH:1969, A study of experimentally induced bovineviral diarrhea-mucosal disease in pregnant cowsand their progeny. Cornell Vet 59:525–538.

Weinstock D, Bhudevi B, Castro AE: 2001, Single-tube single-enzyme reverse transcriptase PCR assayfor detection of bovine viral diarrhea virus inpooled bovine serum. J Clin Microbiol 39:343–346.

Wentink GH, Aarts T, Mirck MH, van Exsel ACA:1991, Calf from a persistently infected heifer bornafter embryo transfer with normal immunity toBVDV. Vet Rec 129:449–450.

Westenbrink F, Straver PJ, Kimman TG, de LeeuwPW: 1989, Development of a neutralizing antibodyresponse to an inoculated cytopathic strain ofbovine virus diarrhoea virus. Vet Rec 125:262–265.

Whitmore HL, Gustafsson BK, Havareshti P, et al.:1978, Inoculation of bulls with bovine virus diar-rhea virus: Excretion of virus in semen and effectson semen quality. Theriogenology 9:153–163.

Yates WDG: 1982, A review of infectious bovinerhinotracheitis shipping fever pneumonia and viralbacterial synergism in respiratory disease in cattle.Can J Comp Med 46:225–263.

Zhang G, Aldridge S, Clarke MC, McCauley JW:1996, Cell death induced by cytopathic bovine viraldiarrhoea virus is mediated by apoptosis. J GenVirol 77:1677–1681.



INTRODUCTIONMany biological variables can be used to quantifythe occurrence of bovine virus diarrhea virus(BVDV) infections, including clinical, pathological,virological, serological, and production measures.The importance of each measure depends on thepurpose of the investigation. In surveillance anderadication programs, the laboratory diagnoses areimportant to document where the infection is pres-ent or was present recently; from an economic pointof view it is important to combine virological andserological measures with information on clinicaland production variables; and from an animal wel-fare point of view the emphasis would be on theseverity of the clinical signs. However, it is not al-ways possible to fully describe the occurrence of adisease for all purposes, simply because of the vari-ations in the quality of the data.

For BVDV infections, good quality data are avail-able from many parts of the world on the presenceof persistently infected (PI) animals and antibodycarriers in addition to epidemiological data on theoccurrence of infection. However, when it comes tothe disease (clinical signs and reduced productiondue to BVDV infections), the data are often basedon selected cases, making it difficult to estimate thetrue occurrence of the disease in a population.Therefore, the calculation of the economic effect ofinfection is often based on the proportion of infectedanimals that are estimated to have suffered from thedisease, which are hereafter included in economicalmodels. Also the importance of risk factors is usu-ally related to the occurrence of infection rather thanthat of the disease. This chapter outlines the occur-rence of BVDV infections, the associated risk fac-tors for infection and the effect of infections on dis-ease and production including financial losses.

MEASURABLE AND QUANTIFIABLEEPIDEMIOLOGICAL VARIABLES

The relevant variables for describing the occurrenceof BVDV infection include the detection of virusand/or antibodies in the animal as well as data re-lated to disease and reduced production followinginfection. In addition, the occurrence of BVDVshould be described in relation to risk factors thatcan be host-, agent-, or environment-related. Thischapter emphasizes risk factors for determining theoccurrence and distribution of BVDV. An epidemio-logical framework for the necessary data for de-scribing BVDV occurrence is given in Figure 2.1.Here the variables are divided into risk factors onthe one hand and measurable output variables on theother. The list of epidemiological data in Figure 2.1is not exhaustive, but the boxes in the figure refer tothe following subchapters where more detailed in-formation is given.

There are two important mathematical measuresof disease occurrence (or disease frequency): preva-lence and incidence. The prevalence is a measure ofwhat is present here and now, whereas incidencemeasures the occurrence of new cases over time.Both measures are useful in describing BVDV in-fections. The occurrence of long-lasting antibodies(often lifelong) and the lifelong presence of virus inPI animals make prevalence studies more suitablefor BVDV infections than is the case for many otherinfections. Prevalence and incidence can be appliedboth to individual animals as well as herds.However, the definition of a herd infection mayoften be more uncertain and more complicated thanat the individual animal level.

35

2Risk Assessment

Hans Houe


PREVALENCE OF BVDVINFECTIONSThe prevalence of a disease or infection (or anyother condition) is defined as the proportion of thepopulation with the disease at a given point in time:

The prevalence can be interpreted as the probabil-ity that a randomly selected animal has the diseaseat time t. Thus, the prevalence of BVDV antibodycarriers is the proportion of animals having antibod-ies, no matter when they were acquired. The advan-tage of a prevalence study is that animals need to besampled only once. The disadvantage is that it is notknown when the antibodies were acquired, or how.However, for PI animals we know that they were in-fected in the first trimester of fetal life (althoughviremic animals need to be retested after 2–3 weeks,it is still considered a prevalence study), and henceprevalence studies on the occurrence of PI animalsare very useful.

Diagnostic methods for the detection of BVDV

and its antibodies are covered in detail in Chapter12. In the context of describing occurrence, it is im-portant to evaluate prevalence studies in relation tothe sensitivity and specificity of the diagnostic tests.If, for example, a serologic test has revealed a preva-lence of 45% antibody carriers (apparent preva-lence, AP) and the test is known to have sensitivity(Se) of 97% and a specificity (Sp) of 88%, the trueprevalence (TP) can be calculated from this formula:

where AP denotes apparent prevalence (the preva-lence of test-positive animals). This means that thetrue prevalence is 39%. These calculations are madewith the assumption that all serologically positiveanimals arose via natural infection. Antibodies titersthat are due to vaccination will throw off these cal-culations.

For the detection of both virus and antibodies inindividual animals, the sensitivities and specificitiesare usually very high and often close to 100%. Fordetection of antibodies, the Se and Sp of different

TPAP Sp

Sp Se= + −

+ −= + −

+ −=1

1

0 45 0 88 1

0 88 0 97 10

. .

. ..339

p = Number of diseased or infected animals at time t

Number of animals in the sample at ttime t


Figure 2.1. An epidemiological framework for the necessary data for describing BVDV occurrence, risk factors, andeffect.


ELISA tests as compared to serum neutralization testhave often been higher than 90% (Cho et al., 1991:Se = 97.6% and Sp = 100%; Rønsholt et al., 1997: Se= 96.5% and Sp = 97.5%; Canal et al., 1998: Se =97.5% and Sp = 99.2%; Kramps et al., 1999: Se =98% and Sp = 99%). Comparison of BVDV antigenELISA with conventional tissue culture isolationtechnique has also shown high sensitivity and speci-ficity (Rønsholt et al., 1997: Se = 97.9% and Sp =99.5%).

Many studies give estimates on the prevalence ofinfected herds. Here it is important to know that thedefinition of an infected herd can vary substantiallyand that the sensitivity and specificity can showhigher variation and will often be known with alower certainty than at the individual animal level.Correct classification of herds into being truly in-fected or noninfected has important implications forobservational studies on epidemiological aspects—e.g., prevalence, incidence, identification of risk fac-tors, and quantification of the effect of BVDV infec-tion on disease and production.

The term herd test or herd diagnosis is based ontesting samples from more than one animal. Thesample can consist of separate samples from indi-vidual animals (e.g., blood samples) or it can be apooled sample from several animals (e.g., pooledblood samples or a bulk tank milk sample). If thesamples consist of individual animals with a samplesize of n, and the herd is defined as positive if thereis at least one positive animal in the herd, the herdsensitivity (HSe) and herd specificity (HSp) can becalculated from the prevalence of test-positive ani-mals (AP) and from the specificity at animal level(Sp) as follows:

and

From the formulas it can be seen that increasing thesample size will result in increased herd sensitivitybut decreased herd specificity. Furthermore, a highprevalence would increase herd sensitivity butwould not affect herd specificity. These formulas aretrue only when definition of a herd is based on oneor more animals being positive (i.e., the criticalnumber is 1). Sometimes, a herd is defined positiveonly if a higher number of animals are positive andin these cases other approaches are needed (Martinet al., 1992). In many published epidemiologicalstudies the herd prevalence is given based on at least

one animal being positive for either antibody orvirus.

The herd diagnoses for the presence of PI animalsare often based indirectly on an interpretation ofserological test results of screening young stock,from antibody reaction in bulk tank milk or combi-nations hereof. The true status should then prefer-ably be based on examination of all individual ani-mals. In 26 herds where blood testing of all animalswas used as the gold standard, the HSe and HSp ofusing a small screening sample of young stock wascalculated as 0.93 and 1.0, respectively (Houe,1999). Others have reported HSe and HSp of 0.66and 1.0, respectively, using serologic results fromunvaccinated heifers among 14 dairy herds (Pillarset al., 2002). The reason for a low HSe can be thatthe PI animals are young (Houe, 1992b; Houe,1999; Pillars et al., 2002) or it has even occurred thatby chance the PI animals were selected for serologictesting giving negative results (Pillars et al., 2002).Therefore repeating the serologic testing of youngstock a few months later and also having them testedfor virus will increase HSe significantly.

The antibody level in bulk tank milk has showngood correlation with the prevalence of antibodycarriers in a herd (Niskanen, 1993; Houe, 1994;Beaudeau et al., 2001). Using a screening sample ofyoung stock as a gold standard, a blocking ELISAmeasuring antibody levels in bulk tank milk wasevaluated. Among 352 dairy herds the bulk tankmilk antibody test had a sensitivity of 1 and a speci-ficity of 0.62 at a blocking percentage (cutoff) of50% (Houe, 1999). Furthermore, at an estimatedherd prevalence of 26%, the positive and negativepredictive values were 0.48 and 1.00, respectively.This means that the bulk tank milk test is very goodat detecting herds with true PI animals. On the otherhand, many herds without PI animals may also havehigh antibody levels (e.g., due to a PI animal movedfrom the herd before the test), and therefore the testresult is false positive. In some areas, there may bepoor predictive values of antibody detection amonga few young stock or antibody level in bulk milk forherd diagnosis because of high prevalence of infec-tion and large variation of antibody-positive ani-mals in herds without PI animals (Zimmer et al.,2002).

Many studies use an increase in bulk tank milkantibody level as an indication of a recent infectionin the herd (Niskanen et al., 1995). Also PCR onbulk milk has been used to directly detect virus ex-cretion in the herd (Drew et al., 1999). Unfortu-nately, there are no international standards for clas-

HSp Spn=

HSe AP n= − −1 1( )

Risk Assessment 37


sification of herd infection status, and one should becareful when comparing studies that use differentdefinitions for an infected herd.

EPIDEMIOLOGICAL STUDIES FORESTIMATION OF PREVALENCE

Due to the development of faster and cheaper diag-nostic methods the number of prevalence studies hasincreased significantly in recent years. When com-paring different studies one should be aware of dif-ferences in study design, study period, samplingframe, sampling method, sampling units, samplesize, and the exact test variable being measured.

Tables 2.1–2.3 summarize a number of prevalencestudies including some keyword information on sam-pling frame and sampling methods. These studieswere selected to represent a broad variety of differentgeographical regions. Another criterion for selectionwas that the studies contained sufficient informationon study design and sampling strategies to be judgedfairly representative for the study area. Attemptswere made to include both older and newer preva-lence studies. Studies in countries with national con-trol and eradication programs are from before or atthe beginning of the eradication programs.

The sampling frame in the tables specifieswhether samples have been taken from certain re-gions in the country and whether there is informa-tion on the type of enterprise (dairy herds, beefherds, breeding herds, or artificial insemination cen-ters). The exact sampling methods are often rathercomplicated and for details the reader is referred tothe cited publications. The tables include informa-tion on the major criteria for sampling such as therandomness and representativeness of the samplesand whether only certain age groups are sampled.The sampling has often been done by initially se-lecting herds (primary sampling units) followed bysampling of individual animals (secondary samplingunits). But there are also examples where animalsare the primary sampling unit (e.g., sampling fromabattoirs) or where herds are sampled without fur-ther sampling of animals (e.g., testing for antibodylevel in bulk milk, as shown in Table 2.3).

The variables behind the prevalence measureswhen testing individual animals are either antibod-ies (Table 2.1) or virus (Table 2.2). The same studycan, therefore, be represented in both tables if ani-mals are tested for both antibodies and virus. Theprevalence can be given both at the animal level aswell as at the herd level (here defined as the preva-lence of herds with any antibody or virus-positiveanimal). In addition, many studies aim directly at

obtaining a herd infection status by serological test-ing of a few young stock, testing the bulk tank milkfor antibody or virus, or by various combinations ofthese methods. These studies provide prevalence es-timates of herds with any antibody carriers or pres-ence of any PI animals (or rather herds being sus-pected of having PI animals). As these studies areoften based on different interpretations of antibodylevel in bulk tank milk, they are presented separatelyin Table 2.3 (some keywords of the interpretationfrom the studies are given in the table). Altogether,the following six prevalence measures are presentedin Tables 2.1–2.3:

1. Prevalence of antibody-positive animals2. Prevalence of virus-positive animals3. Prevalence of herds with at least one antibody

carrier4. Prevalence of herds with at least one virus-

positive animal5. Prevalence of herds with antibody-positive ani-

mals as detected from screening samples (spottests) and bulk milk

6. Prevalence of herds with high probability ofvirus-positive animals as detected from screen-ing samples (spot tests), antibody level in bulkmilk, or PCR on bulk milk

One of the difficulties of interpreting serologicalresults is the very different use of vaccines. Whenthere were clear indications of the use of vaccine, ei-ther in the country or in the study herds, this infor-mation is provided in the tables. One should also becareful interpreting the prevalence of infected herdsif not all animals in the herd are tested.

Due to the various sampling strategies and inter-pretation of test results, formal statistical analyses ofTables 2.1–2.3, estimation of confidence intervals(CI) of the true occurrence, and testing for differ-ences between different areas are not straightfor-ward. Instead, many different studies are outlined inthe tables in order to provide the reader with a spec-trum of results from different areas, emphasizingsome selected trends. The reader should be aware ofthe sampling methods when comparing prevalence.If, for example, only younger animals were sampledfor the detection of PI animals, the prevalence wouldbe higher than if sampling was done among all ani-mals (simply because PI animals die faster thanother animals).

Considering the prevalence of antibody carriers atthe animal level, there seems to be a continuum ofprevalence levels from a minimum of 12% to a max-imum of 89% (Table 2.1). However, if all the pre-



39

Tab

le 2

.1.

Epi

dem

iolo

gica

l stu

dies

for

estim

atio

n of

pre

vale

nce

of a

nim

als

with

ant

ibod

ies

agai

nst

BV

DV

Sam

plin

g M

etho

dSa

mpl

e Si

zePr

eval

ence

Cou

ntry

/St

udy

Her

dA

nim

alR

egio

nPe

riod

Sam

plin

g Fr

ame

Her

dsA

nim

als

Her

dsA

nim

als

Lev

el (

%)

Lev

el (

%)

Vac

cina

tion

Ref

eren

ce

Eur

ope

Bel

gium

Den

mar

k

Net

herl

ands

Nor

way

Pol

and

Scot

land

Slov

enia

Spai

n

Swed

en

Swed

en

. . .

1988

. . .

1984

–198

6

. . .

1992

–199

3

1996

1997

. . .

1987

Sout

hern

Bel

gium

Bel

gian

Whi

te B

lue

and

Frie

sian

Hol

stei

nJu

tland

in D

enm

ark

Dai

ry h

erds

. . .

Wid

e ge

ogra

phic

repr

esen

tatio

nD

airy

her

dsB

ulls

at a

rtif

icia

lin

sem

inat

ion

cent

ers

Sout

h W

est S

cotla

ndB

reed

ing

bulls

on

dai

ry,b

eef

ofm

ixed

far

ms

(5 b

ulls

fro

mde

aler

s)5

regi

ons

Bre

edin

g he

rds

Ast

uria

s re

gion

Dai

ry h

erds

11 c

ount

ies

in d

if-

fere

nt p

arts

of

Swed

enC

ount

y of

Kop

par-

berg

D

airy

her

ds

42.5

% o

f he

rds

had

prio

r di

agno

sis

orw

ere

susp

icio

us

Rep

rese

ntat

ive

NPE

. . .

Geo

grap

hic

repr

e-se

ntat

ive,

NPE

— . . .

. . .

Ran

dom

/str

atif

ied

NPE

NPE

Ran

dom

Som

e va

ccin

atio

n(n

ot c

onsi

dere

dim

port

ant)

No

vacc

inat

ion

. . .

No

vacc

inat

ion

. . .

. . .

. . .

No

vacc

inat

ion

No

vacc

inat

ion

No

vacc

inat

ion

Schr

eibe

r et

al.,

1999

Hou

e an

d M

ey-

ling,

1991

Kra

mps

et a

l.,19

99

Løk

en e

t al.,

1991

Pola

k an

dZ

mud

zins

ki,

1999

McG

owan

and

Mur

ray,

1999

Gro

m e

t al.,

1999

Mai

ner-

Jaim

e et

al.,

2001

Ale

nius

et a

l.,19

86

Nis

kane

n et

al.,

1991

(con

tinu

ed)

All

anim

als

in h

erd

All

anim

als

in h

erd

Ran

dom

am

ong

ani-

mal

s in

BH

V1a

diag

nost

ics

and

fiel

d tr

ial

Ran

dom

>2

year

s

>6

mon

ths

old

Ran

dom

All

anim

als

in h

erd

20 h

erds

:all

anim

als

8 he

rds:

Ran

dom

,>

1-ye

ar-o

ldB

reed

ing

heif

ers

All

lact

atin

g co

ws

61 19 >10

0

187

— 78 274

28 114

15

100

100

— 28 — — . . .

86 — 73

66 64 65 19 86 78 17 21 41 46

9685

2570

1798

1133

175

109

6892

529

711

413


40

Swit

zerl

and

Swit

zerl

and

Swit

zerl

and

Uni

ted

Kin

gdom

Uni

ted

Kin

gdom

Nor

th,C

entr

al a

nd S

outh

Am

eric

aC

anad

a

Can

ada

Can

ada

1994

–199

5

1995

1993

–199

4

1974

–197

5

1985

–198

6

. . .

1990

–199

1

. . .

Can

ton

of S

t. G

alle

nD

airy

her

dsC

anto

n of

St.

Gal

len,

7 A

lpin

epa

stur

esSw

iss

Bra

unvi

ehca

ttle

Dai

ry h

erds

Dai

ry h

erds

Eng

land

and

Wal

es

Eng

land

and

Wal

es

Sask

atch

ewan

and

Alb

erta

Seve

ral b

reed

sE

spec

ially

bee

f

Nor

thw

este

rnQ

uebe

c,be

efhe

rds

(cow

cal

f)

3 M

ariti

me

prov

ince

s:N

ewB

runs

wic

k,N

ova

Scot

ia,P

rinc

eE

dwar

d Is

land

Dai

ry h

erds

Ran

dom

Invi

ted

by c

anto

nal

vete

rina

ry o

ffic

er

Ran

dom

(at

leas

t 5

cow

s)3

herd

s in

eac

hco

unty

— Ran

dom

>25

bre

edin

gfe

mal

es

Stra

tifie

d,tw

o st

age

rand

om s

ampl

e

. . .

. . .

. . .

. . .

. . .

Som

e va

ccin

atio

n.O

ne-t

hird

of

vacc

inat

ed h

adan

tibod

ies

Onl

y no

nvac

c.w

ithin

1 y

ear

incl

uded

her

e

Test

ed a

nim

als

not v

acci

nate

d

Bra

un e

t al.,

1997

Bra

un e

t al.,

1998

Stär

k et

al.,

1997

Har

knes

s et

al.,

1978

Edw

ards

et a

l.,19

87

Dur

ham

and

Has

sard

,199

0

Gan

aba

et a

l.,19

95

Van

Lee

uwen

et a

l.,20

01

Cow

s an

d he

ifer

s(A

ll)A

nim

als

prio

r to

past

ure,

98%

wer

e re

plac

emen

tca

ttle

NPE

All

cow

s

12 p

er h

erd

repr

e-se

ntin

g a

rang

e of

age

sSu

bmis

sion

s of

mor

e th

an 1

0sa

mpl

es to

Cen

-tr

al V

eter

inar

yL

abor

ator

y

Syst

emat

ic r

ando

mam

ong

preg

nant

.12

–25

cow

s pe

rhe

rd5

cow

s or

5 h

eife

rs>

6 m

onth

s ol

d

95 149

113

133

— 295

15 89

100

— 99 — — (67)

b

93 (66)

84 63 72 62 65 41 53 38

2892

990

1635

1593

18,7

59

1745

311

445

Tab

le 2

.1.

Epi

dem

iolo

gica

l stu

dies

for

estim

atio

n of

pre

vale

nce

of a

nim

als

with

ant

ibod

ies

agai

nst

BV

DV

(co

ntin

ued)

Sam

plin

g M

etho

dSa

mpl

e Si

zePr

eval

ence

Cou

ntry

/St

udy

Her

dA

nim

alR

egio

nPe

riod

Sam

plin

g Fr

ame

Her

dsA

nim

als

Her

dsA

nim

als

Lev

el (

%)

Lev

el (

%)

Vac

cina

tion

Ref

eren

ce

Seve

ral s

ourc

es:B

ruce

llosi

s ce

rtif

icat

ion,

bull

test

sta

tion,

rese

arch

sta

tion

and

othe

rs


41

Uni

ted

Stat

es

Uni

ted

Stat

es

Uni

ted

Stat

es

Uni

ted

Stat

es

Mex

ico

Bra

zil

(Arg

enti

na)

Chi

le

Ven

ezue

la

Afr

ica,

Asi

a an

d N

ew Z

eala

ndTa

nzan

ia

1976

. . .

1993

. . .

. . .

. . .

. . .

1997

1985

Lou

isia

na

Mid

wes

t and

wes

t-er

n U

.S.

Bee

f an

d da

iry

herd

sM

ichi

gan,

2 co

un-

ties

Dai

ry h

erds

17 s

tate

s B

eef

cow

cal

f op

er-

atio

nsD

airy

cat

tle:2

sta

tes

Bee

f ca

ttle:

11st

ates

Stat

e of

Rio

Gra

nde

¶ 1

farm

inA

rgen

tina

Sout

hern

par

t of

Chi

leB

eef,

dair

y an

dm

ixed

bre

ed5

dist

rict

s of

Apu

reSt

ate

Bee

f ca

ttle

18 r

egio

ns

Res

earc

h he

rds

App

r. H

alf

of th

ehe

rds

had

past

evid

ence

of

BV

DV

5 he

rdsc

2 he

rdsd

2 he

rdse

Rep

rese

ntat

ive

ofth

e na

tiona

l bee

fhe

rd. .

.

. . .

Lar

ge h

erds

Stra

tifie

d

Mul

tista

ge,r

ando

m

NPE

Age

not

rec

orde

d

No

vacc

inat

ion

. . .

Vac

cina

tion

used

in 4

her

ds

Vac

cina

tion

re-

port

ed in

137

herd

s. .

.

. . .

No

vacc

inat

ion

assu

med

. . .

No

vacc

inat

ion

Fulto

n an

d Se

ger,

1982

Bol

in e

t al.,

1985

Hou

e et

al.,

1995

b

Pais

ley

et a

l.,19

96

Suza

n et

al.,

1983

Can

al e

t al.,

1998

Rei

nhar

dt e

t al.,

1990

Oba

ndo

et a

l.,19

99

Mso

lla e

t al.,

1988

(con

tinu

ed)

>2

year

s ol

d br

eed-

ing

cattl

eB

etw

een

20 a

nd 3

12pe

r he

rd B

oth

calv

es a

nd a

dults

All

anim

als

in h

erd

Ow

ner

chos

e ag

egr

oup

Ran

dom

Adu

lts,A

batto

irs

(mos

tly)

Bet

wee

n 2

and

47fr

om e

ach

herd

Cow

s an

d bu

llsR

ando

m

Ran

dom

5 co

ws

per

herd

5 66 5c 2d 2e 256

. . .

20 40 123

. . .

100

. . .

80c

100d

100e

91 a

ll95

Vf

86 N

Vf

— — 90 100

— —

68 89 29c

76d

91e

69 a

ll78

V57

NV

71 63 56 74 81 36 12

444

3157

794c

428d

372e

3894

132

(dai

ry)

771

(bee

f)

430

948

cow

s11

6 bu

lls

615

419

Tab

le 2

.1.

Epi

dem

iolo

gica

l stu

dies

for

estim

atio

n of

pre

vale

nce

of a

nim

als

with

ant

ibod

ies

agai

nst

BV

DV

(co

ntin

ued)

Sam

plin

g M

etho

dSa

mpl

e Si

zePr

eval

ence

Cou

ntry

/St

udy

Her

dA

nim

alR

egio

nPe

riod

Sam

plin

g Fr

ame

Her

dsA

nim

als

Her

dsA

nim

als

Lev

el (

%)

Lev

el (

%)

Vac

cina

tion

Ref

eren

ce

“Sto

ck a

t Ani

mal

Dis

ease

s R

esea

rch

Inst

itute

,Dar

-es-

Sala

am”


42

Tanz

ania

Indi

a

New

Zea

land

New

Zea

land

Gen

eral

lege

nds

and

abbr

evia

tions

in ta

bles

:—

Inf

orm

atio

n no

t app

licab

le.

. . .

Info

rmat

ion

not a

vaila

ble

(the

rea

der

can

of c

ours

e ad

d in

form

atio

n on

the

use

of v

acci

ne f

rom

oth

er s

ourc

es).

NPE

:No

past

evi

denc

e,m

eani

ng th

at h

erds

wer

e no

t sel

ecte

d ba

sed

on p

ast e

vide

nce

of in

fect

ion

(unk

now

n B

VD

sta

tus)

.A

I:A

rtif

icia

l ins

emin

atio

n ce

nter

s.V

:Vac

cina

ted;

NV

:Not

vac

cina

ted.

a BH

V:B

ovin

e he

rpes

vir

us.

b Sam

ple

size

var

ied

cons

ider

ably

bet

wee

n fa

rms.

c Her

ds w

ithou

t use

of

vacc

inat

ion

and

with

out P

I an

imal

s.

d Her

ds w

ith u

se o

f ki

lled

vacc

ine;

no

PI a

nim

als

pres

ent.

e Her

ds w

ith u

se o

f ki

lled

vacc

ine

and

pres

ence

of

PI a

nim

als.

f V:P

reva

lenc

e am

ong

vacc

inat

ed; N

V:P

reva

lenc

e am

ong

nonv

acci

nate

d (h

erds

wer

e cl

assi

fied

acc

ordi

ng to

rep

orte

d us

e of

any

BV

DV

vac

cine

dur

ing

the

prev

ious

12

mon

ths)

.g N

ote

that

ther

e w

ere

only

two

sam

pled

ani

mal

s pe

r fa

rm.

1985

–198

6

1996

–199

7

. . .

1993

5 re

gion

s,11

dif

fer-

ent d

istr

icts

,N

orth

ern

Tanz

ania

16 s

tate

s

Sout

h Is

land

,Ota

godi

stri

ctN

orth

and

Sou

thIs

land

B

eef

cattl

e

NPE

Ran

dom

Two

abat

toir

s

14 a

batto

irs

. . .

. . .

. . .

. . .

Hye

ra e

t al.,

1991

Sudh

arsh

ana

et a

l.,19

99

Rob

inso

n,19

71

Pére

z et

al.,

1994

. . .

. . .

— 70

— — — (78)

g

34 15 66 63

938

327

64 140

Tab

le 2

.1.

Epi

dem

iolo

gica

l stu

dies

for

estim

atio

n of

pre

vale

nce

of a

nim

als

with

ant

ibod

ies

agai

nst

BV

DV

(co

ntin

ued)

Sam

plin

g M

etho

dSa

mpl

e Si

zePr

eval

ence

Cou

ntry

/St

udy

Her

dA

nim

alR

egio

nPe

riod

Sam

plin

g Fr

ame

Her

dsA

nim

als

Her

dsA

nim

als

Lev

el (

%)

Lev

el (

%)

Vac

cina

tion

Ref

eren

ce

“Und

er th

e A

uspi

ces

of th

e R

inde

rpes

tca

mpa

ign”

Nat

iona

l ser

um b

ank


43

Eur

ope

Bel

gium

Den

mar

k

Ger

man

y

Ger

man

y

Pol

and

Swed

en

Swit

zerl

and

Uni

ted

Kin

gdom

Uni

ted

Kin

gdom

. . .

1988

. . .

1993

–199

4

. . .

. . .

1995

1980

–198

5

1986

Sout

hern

Bel

gium

.B

elgi

an W

hite

Blu

e an

d Fr

iesi

anH

olst

ein

Jutla

nd in

Den

mar

kD

airy

her

dsN

orth

ern

Ger

man

y.B

reed

ing

anim

als

Low

er S

axon

y

Bul

ls a

t art

ific

ial

inse

min

atio

nce

nter

s11

cou

ntie

s in

dif

-fe

rent

par

ts o

fSw

eden

Can

ton

of

St. G

alle

n,7

Alp

ine

past

ures

Swis

s B

raun

vieh

cattl

e D

airy

her

ds. .

.

Eng

land

and

Wal

es

42.5

% o

f he

rds

had

prio

r di

agno

sis

orw

ere

susp

icio

us

Rep

rese

ntat

ive,

NPE

Exp

ortin

g he

rds

NPE

— NPE

Invi

ted

by c

anto

nal

vete

rina

ry o

ffic

er

— —

Som

e va

ccin

a-tio

n (n

otco

nsid

ered

impo

rtan

t)N

o va

ccin

atio

n

. . .

Som

e va

ccin

a-tio

n. .

.

No

vacc

inat

ion

. . .

. . .

. . .

Schr

eibe

r et

al.,

1999

Hou

e an

d M

ey-

ling,

1991

Lie

ss e

t al.,

1987

Frey

et a

l.,19

96

Pola

k an

d Z

mud

-zi

nski

,199

9

Ale

nius

et a

l.,19

86

Bra

un e

t al.,

1998

How

ard

et a

l.,19

86

Edw

ards

et a

l.,19

87

(con

tinu

ed)

All

anim

als

in h

erd

All

anim

als

in h

erd

Preg

nant

NPE

Up

to 3

yea

rs

>6

mon

ths

old

Bre

edin

g he

ifer

s

Ani

mal

s pr

ior

topa

stur

e,98

%w

ere

repl

acem

ent

cattl

e N

PE

Bee

f ca

lves

2–4

mon

ths

old

Cow

s 2–

3 ye

ars

old

Gno

tobi

otic

cal

ves

NPE

Subm

issi

ons

ofm

ore

than

10

sam

ples

to C

en-

tral

Vet

erin

ary

Lab

orat

ory

61 19 >10

00

329

— 114

149

— —

44 53 — 45 — — — — —

0.75

(PI

)

1.4/

1.1

0.9 (v

irae

mic

)2.

1 (P

I)

0.9

(PI)

1.7/

1.3

0.9

0.8/

0.4

1.8 (v

irae

mic

)

9685

2570

2317

20,2

53

219

711

990

924

3151

Tab

le 2

.2.

Epi

dem

iolo

gica

l stu

dies

for

estim

atio

n of

pre

vale

nce

of v

irus-

posi

tive

and

pers

iste

ntly

infe

cted

ani

mal

s

Sam

plin

g M

etho

dSa

mpl

e Si

zePr

eval

ence

Cou

ntry

/St

udy

Her

dA

nim

alR

egio

nPe

riod

Sam

plin

g Fr

ame

Her

dsA

nim

als

Her

dsA

nim

als

Lev

el (

%)

Lev

ela

(%)

Vac

cina

tion

Ref

eren

ce


44

Nor

th A

mer

ica

Can

ada

Can

ada

and

U.S

.

Uni

ted

Stat

es

Uni

ted

Stat

es

Uni

ted

Stat

es

Uni

ted

Stat

es

a If

two

perc

enta

ges

are

give

n,th

e fi

rst n

umbe

r is

pre

vale

nce

of a

nim

als

bein

g vi

raem

ic a

nd th

e se

cond

num

ber

is a

nim

als

prov

en to

be

PI.

b Am

ong

6 he

rds

whe

re v

irus

was

isol

ated

onl

y on

e he

rd h

ad b

een

sele

cted

bas

ed o

n pa

st e

vide

nce.

1991

. . .

. . .

1993

. . .

1996

Wes

tern

Can

ada

feed

lot c

alve

s

Nor

thea

ster

n an

dw

este

rn s

tate

s (2

AI

cent

ers

thro

ugho

ut U

.S.

and

Can

ada

(2 A

Ice

nter

s)M

idw

est a

ndw

este

rn U

S.B

eef

and

dair

yhe

rds

Mic

higa

n,2

coun

ties

Dai

ry h

erds

17 s

tate

s B

eef

cow

calf

ope

ratio

ns

Ala

bam

a,N

ebra

ska,

Nev

ada,

Nor

thD

akot

a,an

d O

hio

Bee

f he

rds

1 fe

edlo

t,bu

t ani

-m

als

wer

e ar

riv-

ing

from

man

ypl

aces

— App

r. ha

lf o

f th

ehe

rds

had

past

evid

ence

of

BV

DV

b

Ran

dom

fro

mD

HIA

list

Rep

rese

ntat

ive

ofth

e na

tiona

l bee

fhe

rd

A. R

ando

m,7

6he

rds,

min

20

and

max

500

bre

ed-

ing

fem

ales

B. S

uspe

cted

,52

herd

s

. . .

. . .

. . .

Vac

cina

tion

used

in 1

5 he

rds

Vac

cina

tion

re-

port

ed in

137

herd

s

92%

of

posi

tive

and

82%

of

nega

tive

herd

sus

ed v

acci

na-

tion

Tayl

or e

t al.,

1995

How

ard

et a

l.,19

90

Bol

in e

t al.,

1985

Hou

e et

al.,

1995

b

Pais

ley

et a

l.,19

96

Witt

um e

t al.,

2001

Syst

emat

ic r

ando

m,

ever

y 5t

h ca

lf a

tar

riva

l

Bul

ls a

t AI

isol

atio

npr

ior

to p

roge

nyte

stin

g. M

ostly

3–12

mon

ths

old

Bet

wee

n 20

and

312

per

herd

B

oth

calv

es a

ndad

ults

All

anim

als

in h

erd

Ow

ner

chos

e ag

egr

oup.

Onl

y A

b-ne

g te

sted

for

viru

sC

alve

s <

4 m

onth

sol

d (m

ost >

4m

onth

s ol

d)

— — 66 20 256

Tota

l:12

8A

:76

B:5

2

— — 9.1

15 Tota

l:10

.2A

:4B

:19

0.1

0.4

1.9/

1.7

0.13

/0.1

1

0.67

of A

bne

gativ

esan

d 0.

2 of

tota

lTo

tal:

0.3/

0.17

1029

1532

3157

5481

1201

Ab

neg.

out

of 3

894

18,9

31

Tab

le 2

.2.

Epi

dem

iolo

gica

l stu

dies

for

estim

atio

n of

pre

vale

nce

of v

irus-

posi

tive

and

pers

iste

ntly

infe

cted

ani

mal

s (c

ontin

ued)

Sam

plin

g M

etho

dSa

mpl

e Si

zePr

eval

ence

Cou

ntry

/St

udy

Her

dA

nim

alR

egio

nPe

riod

Sam

plin

g Fr

ame

Her

dsA

nim

als

Her

dsA

nim

als

Lev

el (

%)

Lev

ela

(%)

Vac

cina

tion

Ref

eren

ce


45

Tab

le 2

.3.

Epi

dem

iolo

gica

l stu

dies

for

estim

atio

n of

her

d le

vel p

reva

lenc

e ba

sed

on s

cree

ning

sam

ples

and

bul

k m

ilk s

ampl

es

All

Co

un

trie

s

Her

d Pr

eval

ence

,H

erd

PI A

nim

als

Sam

ple

Prev

alen

ce,

or S

uspe

ct o

f C

ount

ry/

Stud

ySa

mpl

ing

Size

Ant

ibod

yH

avin

g PI

Reg

ion

Peri

odSa

mpl

ing

Fram

eM

etho

d(H

erds

)Sa

mpl

eC

arri

ers

(%)a

Ani

mal

s (%

)aR

efer

ence

Aus

tria

1996

–199

8N

iede

r-ös

tere

ich

Step

wis

e:A

:5,0

24M

ilk. .

.1.

0R

ossm

anith

and

A

ll br

eedi

ng h

erds

A:m

ilk,B

:spo

t B

:512

Spot

test

(det

ectio

n of

PI

Dei

nhof

er,1

998

test

and

C:

C:1

54A

ll an

imal

san

imal

s)al

l ani

mal

sN

PED

enm

ark

1994

Dai

ry h

erds

All

herd

s16

,113

Bul

k m

ilk. .

.39

Bits

ch a

nd R

ønsh

olt,

(sus

pect

ed to

hav

e PI

)19

95E

ston

ia19

93–1

995

Dai

ry h

erds

with

R

epre

sent

ativ

e 32

8B

ulk

milk

. .

.46

Vilt

rop

et a

l.,20

0219

97–1

998

>=

20 c

ows

rand

om s

ampl

e36

3an

d/or

16

1999

–200

035

1yo

ung

18

stoc

k te

st(s

uspe

cted

to h

ave

PI)

Fin

land

1993

Dai

ry h

erds

All

herd

s (>

98%

)34

,115

Bul

k m

ilk1

. . .

Nuo

tio e

t al.,

1999

Eng

land

19

96D

airy

her

ds

Syst

emat

ic

1070

Bul

k m

ilk95

66Pa

ton

et a

l.,19

98an

d W

ales

>40

cow

sra

ndom

(OD

>=

0.13

5)(O

D >

=0.

9)N

orth

ern

1999

Dai

ry h

erds

From

the

larg

est

929

Bul

k m

ilk99

50G

raha

m e

t al.,

2001

Irel

and

milk

pro

cess

or(O

D >

0.04

)(O

D >

=0.

55)

Nor

way

1993

Dai

ry h

erds

All

herd

s26

,430

Bul

k m

ilk37

7.1

Waa

ge e

t al.,

1997

(OD

>=

0.05

)(O

D >

=0.

55)

Per

u19

98M

anta

ro V

alle

yPr

opor

tiona

l 60

Bul

k m

ilk96

27St

åhl e

t al.,

2002

geog

raph

ical

(O

D >

=0.

05)

(OD

>=

0.55

)di

stri

butio

nSw

eden

1993

Dai

ry h

erds

Maj

ority

of

dair

y 14

,463

Bul

k m

ilk. .

.51

Ale

nius

et a

l.,19

97he

rds

(OD

>0.

55)

Uni

ted

3-ye

ar

Mos

t her

ds:N

ew

. . .

97B

ulk

milk

—

12R

ensh

aw e

t al.,

2000

Stat

espe

riod

Yor

k,Pe

nnsy

lvan

ia,

PCR

and

(m

ost l

ikel

y PI

an

d M

ichi

gan

viru

san

imal

s am

ong

Rem

aini

ng f

rom

is

olat

ion

lact

atin

g an

imal

s)11

oth

er s

tate

s

a N

ote

that

the

ant

ibod

y de

tect

ion

met

hods

var

y be

twee

n co

untr

ies

as d

o th

e cu

toff

s w

hen

a he

rd i

s co

nsid

ered

to

have

ant

ibod

y ca

rrie

rs o

r PI

ani

mal

s. P

reva

lenc

es a

re t

here

fore

jus

t in

-di

cativ

e of

the

leve

l and

not

dir

ectly

com

para

ble

betw

een

coun

trie

s.


valence studies are ranked according to prevalencelevels there is a tendency to different plateaus. Espe-cially, a large number of studies have revealed pre-valence from 60–70% of antibody carriers. At thelow scale there are almost as many studies revealingprevalence in the range of 10–20% as there are from20–60%.

Turning to the prevalence of virus-positive and PIanimals it is evident that many studies find preva-lence from 0.75 to around 2%. However, anothergroup of studies has revealed prevalence in the rangeof 0.1–0.2% (Table 2.2). This is also reflected in theprevalence at the herd level where some studies havefound 40–50% of the herds to host PI animals (Table2.2) or being suspicious of hosting such animals(Table 2.3) whereas other studies have found only10–15% of the herds to host (or being suspicious ofhosting) PI animals (Tables 2.2 and 2.3). This couldlead one to speculate that areas can be divided intohigh-and low-prevalence areas.

Looking at the regional differences, many Euro-pean countries were found to have high prevalence.However, generalizations are not possible becausesome countries (e.g., Norway, Finland, and Austria)show low prevalence. It is noteworthy that preva-lence studies in North America have revealed a verylow prevalence of PI animals although the preva-lence of antibody carriers in nonvaccinated herds insome studies has been relatively high. The reasonfor these differences can only be speculated upon.Variation in vaccination programs and managementstyles may result in differences. It can also be spec-ulated that the mortality among PI animals may varyor that the outcome of fetal infection is different(i.e., the ratio of abortion versus birth of PI animalsmay vary). Furthermore, it would be reasonable toassume that some of the differences are due to vari-ations in demographic circumstances (see later sec-tion, “Risk Factors for Occurrence of BVDV Infec-tions”). In U.S. and Canadian studies it is alsocharacteristic that PI animals can occur in high num-bers in some herds.

Follow-up studies in some areas have shown thatthe infection can stay endemic in an area but with apronounced change in infection status among theunderlying herds (Houe et al., 1997). A study in Es-tonia showed a large decrease in herds suspected tohave PI animals from 1993 to 2000 (Viltrop et al.,2002). The change was believed to be due to de-creased trade of breeding animals. Further, it wasproposed that the reduction of BVDV in these herdswas facilitated by the facts that most large herdswere closed and the density of cattle farms was low.

In Sweden, self-clearance has been fostered by con-trolling introduction of new animals (Lindberg andAlenius, 1999). However, without such specific in-terventions, BVDV infections will usually maintainthemselves at high levels. In the Nordic countries,eradication programs were initiated in the begin-ning of the 1990s. In these countries the prevalenceof herds with PI animals has been significantly re-duced and is now approaching zero (Alenius et al.,1997; Bitsch et al., 2000; Valle et al., 2000a; Houe,2001).

EPIDEMIOLOGICAL STUDIES ONOCCURRENCE OF DIFFERENT GENOTYPES

The variation of BVDV strains is described in detailin Chapter 3. On most occasions, the variation inBVDV is described at a qualitative level, but quan-tification of the strains in different regions and theirrelationship to risk factors is largely unknown.Among 96 field isolates collected in Germany be-tween 1992 and 1996, 11 (11%) were identified asBVDV genotype 2 (Wolfmeyer et al., 1997). Anotherstudy of 61 field isolates collected between 1960 and2000 in Northern Germany identified 2 isolates asgenotype 2 (Tajima et al., 2001). It was also indi-cated that the virus population had been relativelystable over time. Among 28 field isolates fromBelgium, only 1 belonged to genotype 1a and the re-maining could be divided into genotypes 1b and 2(Couvreur et al., 2002). Other isolations of BVDgenotype 2 in Europe include findings from Slovakia(Vilcek et al., 2002) and United Kingdom where thefirst definitive genotype 2 isolate was obtained in2002 (Drew et al., 2002).

In the U.S., several strains gathered over the lastmany years have belonged to genotypes 1a, 1b, and2 (Ridpath et al., 2000). Among a sample of 203BVDV isolates from lots of pooled fetal bovineserum, 51 (25.1%) were identified as genotype 1a,64 (31.5%) as genotype 1b, 65 (32%) as genotype2, and 23 (11%) as a mixture of isolates (Bolin andRidpath, 1998). Of the 105 BVDV isolates fromdiagnostic laboratory accessions, 61% belonged togenotype 1 and 39% to genotype 2 (Fulton et al.,2000). In South America, analysis of 17 Brazilianisolates identified 13 isolates (76%) as genotype 1(4 as 1a and 9 as 1b) and 4 isolates (24%) as geno-type 2 (Flores et al., 2000). In summary, BVDVgenotype 2 seems much more prevalent in NorthAmerica than in Europe. There are little or no dataon prevalence in the African or Asian continentsand only very limited data from South America thus far.



INCIDENCE OF BVDVINFECTIONSThe incidence of an event (disease, infection, or anyother condition) is a measure of new cases within agiven time period. The incidence is usually calcu-lated relative to the population at risk. Two often-used measures are the incidence risk and the inci-dence rate. The incidence risk (Irisk) can be definedas follows (Toft et al., 2004):

The incidence risk can be interpreted as the proba-bility that a susceptible animal (i.e., animal at risk)will experience the event of interest in a given timeperiod. Thus the incidence risk can be interpreted atthe individual animal level. If the incidence risk hasbeen determined for a time period, the incidence riskfor n time periods of the same length, where we as-sume that the incidence risk is constant throughoutall individual periods, can be determined as follows:

The incidence rate (Irate) on the other hand is ameasure of the speed (rate) at which healthy animalsbecomes diseased or infected. It is expressed pertime unit (animal time at risk) and can mathemati-cally be defined as:

The animal time at risk is the sum of the time atrisk for each animal until it either becomes sick, iswithdrawn from the study or the study ends. The Irateis suitable for studies in dynamic populations whereanimals enter and leave the study during the studyperiod, because it is a measure per animal time andnot per animal, as is the case for the Irisk. If animalsenter and leave the study population uniformlythroughout the study period (i.e., on average theyenter and leave in the middle of the study period) thetime at risk can be estimated as follows:

Under the assumption that the incidence risk or in-cidence rate is constant during the study period,

the following relationship exists between the twomeasures:

EPIDEMIOLOGICAL STUDIES FORESTIMATING INCIDENCE

There are few studies with direct determination ofincidence of infection based on repeat sampling. Itis, therefore, relevant to determine the incidence in-directly from prevalence measures (see next section,“Prevalence and Incidence”). The incidence of in-fection within herds or group of animals will dependon whether only transiently infected animals arepresent or both PI animals and transiently infectedanimals are present. The incidence of infectionwhen only transiently infected animals are presentvaries from case to case. In a dairy herd where thelast PI animal was removed in April 1987, 7 of 41animals born in 1988 and 7 of 54 animals born in1989 had seroconverted by December 1990 (Barberand Nettleton, 1993). As we do not know the birthdates of calves, the exact incidence rate cannot becalculated. However, assuming an average time atrisk of 2 years for each animal, the total time at riskwill be 95 animals of 2 years—i.e., a total of 190 an-imal years at risk. A total of 14 seroconversions in190 animal years at risk gives an incidence rate of14/190 = 0.073 cases of seroconversions per animalyear corresponding to an annual incidence risk of0.067 (6.7%) using the shown formula for conver-sion between incidence risk and incidence rate.

In a long-term epidemiological study in a dairyherd where the virus was spreading slowly overmore than 2 years and where there was no directcontact with PI animals, a SIR (susceptible/infectious/removed) model was used to determinethe basic reproduction rate R0 (the number of sec-ondary cases produced by one infectious animalduring the whole period in which it is infectious).Among groups of freely mixing cattle with nocontact with PI animals, R0 was estimated as 3.3(CI: 2.6–4.1) (Moerman et al., 1993). There is noconsensus on whether the transient infection will bean ongoing process or will be self-limiting andaffect only a few animals (1992a). In some experi-mental studies, susceptible animals in direct con-tact with transiently infected animals did not getinfected (Niskanen et al., 2000; Niskanen et al.,2002).

When PI animals are present in a herd there willbe a continuous high infection pressure. Among 67antibody-negative animals from herds with PI ani-

I eriskI trate= − −1 *

Time at risk

Number of animals atrisk at s

=

ttart + Number ofanimals at risk at end

2t* iime period.

Irisk = Number of new cases in the study periood

Animal time at risk

I Irisk risk pn= − −1 1( )( )

Irisk =

Number of new casesin the study periood

Number of animals at riskat the start off the study period

Risk Assessment 47


mals, retesting 6 months later showed that 65 hadseroconverted (Houe and Meyling, 1991). This cor-responds to an incidence risk of 65/67 = 0.97 for a6-month period and an annual incidence risk of:1�(1�0.97)2 = 0.999 i.e., practically a 100% risk ofinfection. The study by Moerman et al. (1993) alsoshowed a 100% risk of seroconversion amonggroups of cattle having lived among PI animals.Among 22 cows that have had only a brief acciden-tal contact with a PI animal, only 12 had becomepositive 4 months later (Moerman et al., 1993).Therefore the incidence risk naturally depends onthe length of exposure to PI animals. Among 22heifers brought in contact with a PI animal, all sero-converted within 5 months, with the fastest serocon-version among heifers with the closest contact to thePI animal (Wentink et al., 1991). It is important tostress that in production systems in which subgroupsof the herd are segregated from PI animals, thesesubgroups of animals can stay uninfected for longperiods (Taylor et al., 1997). In addition to variationin the production system, there are considerable dif-ferences in the pasturing conditions among coun-tries. In Switzerland, animals are often pastured inthe Alps where incidence risk during the summer forreplacement animals has been calculated as 45%(Braun et al., 1998). On pastures with PI animals theincidence varied between 33–100% as compared to6–22% on pastures without PI animals.

Antibody levels in bulk tank milk were taken at 1-year intervals among dairy herds in Sweden. It wasfound that 5 of 43 herds and 7 of 91 herds with aninitial low level showed an increase in antibody level1 year later, indicating that these herds had experi-enced a new BVDV infection during the year (Nis-kanen, 1993; Niskanen et al., 1995). The correspon-ding annual herd incidence risk can be calculated as12% and 8%, respectively, for these herds. Among90 dairy herds in Northern Ireland, with an initiallow bulk tank milk antibody level, 12 had a substan-tial increase after 1 year, corresponding to an annualincidence risk of 13.3% (Graham et al., 2001). Ifherds with a smaller increase in antibody level werealso included, the annual incidence risk could havebeen as high as 47.7%, but it was not knownwhether some of these moderate increases in anti-body level were due to purchase of antibody-positive animals. In Denmark, repeated serologicalexamination of individual dairy cattle in 9 and 24herds revealed herd annual incidence risks of 52%and 26%, respectively (Houe and Palfi, 1993; Houeet al., 1997). Based on the annual tested antibodylevel in bulk tank milk, the average seroconversion

risk in Norwegian herds was estimated as 12% in1993 (i.e., at the start of the eradication program)declining to 2% in 1997 (Valle et al., 2000a).

The incidence of birth of PI animals will naturallyfollow the incidence of transient infection in suscep-tible pregnant cows. Based on the incidence of tran-sient infection and age distribution of cattle at thetime of conception in a high-prevalence area, the the-oretical incidence risk of fetal infection in the first 3months of pregnancy was calculated as 3.3% (Houeand Meyling, 1991). Because of fetal death and abor-tions, the incidence of PI animals born would besomewhat lower. Although the incidence risk may beassumed to be relatively constant in a larger popula-tion (i.e., the infection is endemic), there is oftenhigh variation over time in the individual herd.

Some possible reasons for this variation are dif-ferences among BVDV strains (e.g., differences inviral shed or differences in PI/abortion ratios) andsubsequent introduction of other BVDV that areantigenically divergent from the first BVDV. In ad-dition PI animals often occur in clusters. In 22dairy herds with a total of 129 PI animals, the inci-dence of PI animals born was closely related to thetime of birth of the first PI animal (i.e., oldest iden-tified PI animal) in the herd (Houe, 1992a). Thus26.3% of the PI animals were born within the first2 months after the oldest PI animal, whereas no PIanimals were born from 2 months until 6 monthsafter the oldest PI animal. Hereafter from 6–14months after birth of the oldest PI animal, a largergroup consisting of 52.7% of all identified PI ani-mals was born (note that these percentages arecompared to other PI animals and therefore not in-cidence risks, but a measure of relative incidenceof PI animals born).

This pattern of occurrence of PI animals seems toreflect two periods of transient infection amongcows: an initial period with transient infections ofshort duration (and there are probably no PI animalsaround) and a second period of transient infectionsof longer duration caused by continuous presence ofPI animals. It is not possible to calculate the levelsof incidence risk from these figures because thenumber of susceptibles at different periods was notknown. Taken together, these studies indicate that atthe herd level a common scenario concerning inci-dence could be the following:

• An initial transient infection period of relativelylow infection pressure and variable duration(sometimes it stops and sometimes it willcontinue)



• After the birth of PI animals from 6 months laterthe incidence increase to high levels stayingthere as long as PI animals are present.

PREVALENCE AND INCIDENCE

Most epidemiological studies on the occurrence ofBVDV infection have focused on determining theprevalence from cross-sectional or repeated cross-sectional studies. Studies on incidence are usuallymore expensive because it is necessary to take sam-ples at more than one time point. However, under cer-tain circumstances, the incidence can be calculatedfrom data on antibody prevalence. The assumption isthat the prevalence at a certain age is a measure of theincidence of infection throughout the cow’s life untilthat age (Houe and Meyling, 1991; Toft et al., 2003).We, therefore, need a formula to sum up the Irisk overthe animal’s life. As calculation of incidence over se-quential periods is dependent on the previous periods,it is easier to calculate the probability of avoiding in-fection over several periods and then finally subtractthat probability from 1. The probability of avoidinginfection in a given time period is 1 � Irisk(i) and thetotal probability of avoiding infection during n timeperiods is simply the product of each time period:

where ∏ denotes the products of all time periods.Hereafter the total incidence simply becomes the

following:

As the prevalence at a given age is a measure of totalincidence risk and assuming a constant incidenceover different time periods e.g., a constant annual in-cidence risk (Irisk(p)) the formula can be written asfollows:

where n is the number of time periods, each of thesame length p—i.e., in this case the age in years.This methodology was used by Houe and Meyling(1991) where age specific prevalence of 0.48, 0.65,0.75, 0.85, 0.95, and 0.96 for 1-year age groupsfrom 1–7 years was entered on the left side of theformula together with age, whereafter the annual in-cidence risks were calculated as 0.35, 0.34, 0.32,0.34, 0.42, and 0.39 for each age group, respectively.Thus it can be seen that the incidence was fairly con-

stant at around 0.34 for most age groups. It is an im-portant assumption that the population is in an en-demic situation with a constant infection pressure inthe population as a whole. Similar increases in age-specific prevalence of antibody carriers by increas-ing age has also been seen in other studies (Hark-ness et al., 1978; Braun et al., 1998; Luzzago et al.,1999; McGowan and Murray, 1999).

RISK FACTORS FOROCCURRENCE OF BVDVINFECTIONSA risk factor can be defined as a factor that is asso-ciated with increased probability of a given event oroutcome (infection, disease, or reduced productiv-ity). Sometimes risk factors are used for all factorsthat have a statistical association with the outcome,but often they are used only for factors where a cer-tain degree of causality can be anticipated. On theother hand, the term etiology is used for the imme-diate cause of the disease. The combination of etiol-ogy and risk factors defines the disease determi-nants, which are often described according to theirassociation with the host, the infectious agent, or theenvironment. As the host and agent are described indetail in other chapters, this chapter will focus onthe environmental determinants, which in this caseare environmental risk factors.

Risk factors can be viewed as either the risk of in-fection or the risk of disease. For example, the use ofcommon pasture will increase the risk of introducingthe infection, whereas high humidity in the calf barnwill increase the risk that calves develop pneumoniafollowing infection. This section covers risk factorsfor the occurrence of infection, and the next sectioncovers the effect of infection on disease and produc-tion (i.e., the risk of damages following infection).

When the outcome of risk factor studies is infec-tion, disease, or other dichotomous variables, theprevalence of a given outcome (e.g., infection) in anexposed group can be compared to the prevalence ina nonexposed group by setting up the data in a 2 � 2table:

Age specific prevalence = − −1 1( )( )Irsk pn

Irisk totall

= − −=

∏1 1( )( )Irisk ii

n

Probability of avoiding infection= (1 − Irisk(iii

n

))=

∏l

Risk Assessment 49

Infection StatusRisk Factor Status Infection No Infection Total

Exposed a b a + bNot exposed c d c + dTotal a + c b + d a + b +

c + d = n


The risk in the exposed group (i.e., prevalenceamong exposed: a/(a + b)) as compared to the riskin the unexposed group (i.e., prevalence among non-exposed: c/(c + d)) is measured as the relative risk(RR):

The RR is a measure of the relative importance orassociation of the risk factor with infection. If wewant to measure the additional importance of therisk factor we simply subtract the prevalence in non-exposed group from the prevalence in the exposedgroup obtaining the so called attributable risk (AR):

Another important measure is the odds ratio (OR).Rather than establishing the relation between preva-lence in exposed and nonexposed group as for RR,the OR establishes the relationship between the oddsin the exposed versus nonexposed group:

Note that if a and c are small (low prevalence of in-fection), the OR is a good approximation of the RR.Usually the RR is preferable to OR, but as the RR isnot a valid measure in case-control studies it be-comes an important measure.

Risk factor studies on BVDV infections haveoften been performed as observational studies wherethe main types are cross-sectional, cohort or case-control studies. In cross-sectional studies we samplewithout regard to infection or exposure status, in co-hort studies we sample according to risk factor sta-tus (only among noninfected), and in case-controlstudies we sample according to infection status. Incase-control studies where we have sampled accord-ing to infection, we are not allowed to establish theprevalence measures of infection. Instead, preva-lence measures or odds of risk factor relative to in-fection status can be calculated. Especially the oddsfor risk factor status are useful as the OR for riskfactors ends up being the same formula as for OR forinfection status:

Still, one should be more careful interpreting causal-ity from OR obtained in case-control studies as it

measures the ORs of exposure. Thus, if infected an-imals in a case-control study have an OR of 2.0 ofhaving been on the pasture, this is not the same as tosay that pastured animals have an OR of 2.0 of beinginfected, although we intuitively feel that in practiceit will almost be the same. There are several othermeasures on the importance of risk factors. Formore detail, the reader is referred to epidemiologicaltextbooks. Among the risk factor studies on BVDVthe OR seems to be the most frequently used meas-ure of effect.

The presence of PI animals in the vicinity of sus-ceptible animals poses the highest proven risk of se-roconversion in the population. In a cross-sectionalstudy, the mean antibody prevalence among cattlefrom herds with PI animals was 87% as compared to43% in herds without PI animals (Houe and Mey-ling, 1991). The RR of being infected according tothe risk factor of coming from a herd with PI ani-mals was not stated but it can be obtained simply asthe relationship between the two prevalences (87/43= 2.0) or to illustrate the use of the RR formula di-rectly from the distribution of antibody carriers inthe two herd categories:

Hereafter the RR is calculated as follows:

This may be an underestimation of the effect of PIanimals because many of the animals in herds with-out PI animals at the time of cross-sectional studymay have been exposed to PI animals previously.

The following description of other risk factorswill often be a reflection of the increased risk of di-rect or indirect contact with PI animals and to someextent with acutely infected animals or othersources. Some may claim that the presence of PI an-imals should be viewed as the direct etiology andnot a risk factor, and often the distinction between

RRa a b

c c d= +

+= =/ ( )

/ ( )

/

/.

1083 1249

572 13212 0

ORa c

b d

a d

b crisk factor = = ××

/

/

ORa b

c d

a d

b c= = ×

×/

/

ARa

a b

c

c d=

+−

+

RRa a b

c c d= +

+/ ( )

/ ( )


Antibody Status of Individual Animals

Risk Factor Status:Presence of PI Antibody- Antibody-Animals in the Herd Positive Negative Total

PI animals present 1083 166 1249PI animals not present 0572 749 1321Total 1655 915 2570


etiology and risk factor is unclear. However, this ismore of an academic discussion.

The importance of different risk factors will varysubstantially between regions. For example, the pur-chase of animals without testing will be a muchhigher risk factor in high-prevalence areas such ascentral Europe than in low-prevalence areas such asFinland. The following examples on risk factorsshould therefore be seen in the context of the preva-lence, as shown in Tables 2.1–2.3.

In a cross-sectional study, the prevalence of anti-body carriers was significantly higher among ani-mals kept in extensive production systems (76.8%)than those in intensive productions systems (70.9%)(Reinhardt et al., 1990). This was believed in part tobe due to a higher rate of animal movement in exten-sive production systems. In the same study there wasno difference in prevalence between beef, dairy, ormixed farming.

A case-control study of 314 herds in Norwaycompared risk factors between herds with antibody-positive young stock (case herds) and those withnegative bulk tank milk samples (control herds)(Valle et al., 1999). A high number of risk factorsoccurred in a higher percentage of case herds ascompared to control herds, including (percentage ofcase herds compared to control herds given in paren-thesis): heifers on common pasture (21.0 vs. 8.7),sheep in pasture with cattle (31.5 vs. 22.1), breakingthrough pasture fences (33.8 vs. 24.6), over pasturefence contact (50.0 vs. 31.2), mixing of herds in pas-ture (43.8 vs. 30.6), wild animals in pasture (92.2 vs.84.5), purchasing animals (67.3 vs. 48.7), not askinginformation about BVD when purchasing animals(73.4 vs. 40.9), other animal traffic including com-mon summer housing and exchange of calves (9.6vs. 0.6), veterinarian reusing needles between farms(7.1 vs. 1.3), and not using dairy advisors (25.0 vs.12.0). However, due to the long list of variables andoccasional low number of herds with the indicatedrisk factor, few risk factors were significant. In thelogistic regression models, purchasing animals (OR= 1.8), use of common pasture (OR = 5.1), over pas-ture fence contact (OR = 2.5), purchasing withoutBVD documentation (OR = 5.4), not using dairy ad-visors (OR = 4.1), being a younger farmer, and“other animal traffic” (OR = 28.6) were found to berisk factors, which explained 51% of the seroposi-tive herds.

In a repeat cross-sectional study in Denmark, 41dairy herds were classified into lightly infected (�3antibody carriers among 10 young stock) and highlyinfected (≥8 antibody carriers among 10 young

stock) in 1992 and 1994 (Houe et al., 1997; Houe,1999). Of the 24 herds being lightly infected in1992, 11 changed to highly infected in 1994 and 13remained lightly infected. The change in infectionstatus was associated with the purchase of new ani-mals within the past 3 years (P = 0.052) and moder-ately associated with pasturing cattle at a distance ofless than 5 m to cattle from other herds or havingother contacts with cattle from other herds (P =0.085).

Risk factors for the presence of PI animals in cat-tle herds have been studied including demographicinformation on herd sizes and information from ge-ographic information systems concerning distancesbetween herds. Multivariable logistic regressionwith data from more than 8,000 herds showed sig-nificant effect of the following risk factors (OR, CI,and p values given in parenthesis) on the occurrenceof PI animals: herd size measured as number ofcows (OR = 1.09 per change in herd size of 10 cows;CI = 1.06–1.11; P<0.001), mean distance to neigh-boring herds (OR = 0.87 per change in unit of 500m;CI = 0.81–0.93; P<0.001) and sum of infectedneighbors (OR = 1.54 for herds with >3 infectedneighbors as compared to no infected neighbors; CI= 1.17–2.02; P = 0.024 for the overall effect of sumof infected neighbors) (Ersbøll and Stryhn, 2000). Ina study in U.S. herds with PI animals were signifi-cantly (P<0.01) larger than herds without PI animals(Wittum et al., 2001). Thus, herds with PI animalshad a medium number of 245 calves born in thecalving season as compared to 89 in herds withoutPI animals. The OR for a herd of having high levelsof antibodies in bulk tank milk was shown to in-crease by a factor 1.003 for each additional cow inthe herd (Paton et al., 1998).

Several risk factors are confounded among eachother. If, for example, larger herds purchase moreanimals than smaller herds then an apparent effect ofherd size may in reality be due to purchase of ani-mals and not to the herd size. It is, therefore, impor-tant that risk factor studies are performed by includ-ing several risk factors. Furthermore, there is a needto address interactions between risk factors. For ex-ample, the risk of infection at pasture will be higherin a high-prevalence area as compared to a low-prevalence area.

An alternative to large-scale studies identifyingrisk factors for the presence of infection is to performintensive follow-up in recently infected herds.Follow-up investigations were performed in 67 pre-viously BVDV-free Danish herds that got infected in1998. Obvious explanations for re-infection were

Risk Assessment 51


identified in 74% of the cases. Of the 67 herds, 28%had purchased pregnant animals that later deliveredPI animals. In 36% of the herds, PI animals had beenpresent on neighboring pastures. Seven percent ofthe herds had had animals on common pasture and in3% of the herds there had been PI animals in neigh-boring farmhouses. In the remaining 26% of theherds, no obvious explanations could be identified(Bitsch et al., 2000). Because this is an explorativetype of study, probabilities cannot be given for theactual reasons apart from the given frequency distri-bution of the most obviously identified reasons.

As previously stated, the effect of risk factors willbe highly influenced by the prevalence of infectionin the area. Therefore the effect of different risk fac-tors will obviously change during an eradicationphase. It can be beneficial to combine different stud-ies in order to obtain sufficient observational unitsfor statistical analysis. Attempts have been made tocompare risk factors in a low-prevalence area(Michigan) with that in a high-prevalence area(Denmark) (Houe et al., 1995a). The actual preva-lence related to two previous studies (Houe andMeyling, 1991; Houe et al., 1995b; prevalence canbe seen in Tables 2.1 and 2.2) showed that the preva-lence of PI animals was 10 times higher in Denmarkthan in Michigan. When the effect of purchasing anyanimal versus no purchase of animals was analyzedseparately in the two areas there was no significanteffect of purchase on the occurrence of PI animals inindividual herds. However, when including bothareas in the analyses stratifying on area, a significanteffect of purchasing animals was found. The analy-ses also included a number of demographic vari-ables (either obtained at the country or state level orobtained for the studied herds), and all risk factorsexcept herd size were in favor of the lower preva-lence in Michigan. The following examples wherethe variables for Denmark versus Michigan aregiven in parentheses illustrate the differences: con-centration of cattle in the country or state (DK: 67per km2, MI: 27 per km2), concentration of herds inthe country or state (DK: 1 per km2, MI: 0.29 perkm2), herd size among the studied herds (DK: 135,MI: 274), use of pastures (DK: 79% of herds, MI:45% of herds), purchase and addition of animals tothe studied herds (DK: 18.5% of herds; MI: 12.6%of herds), and use of vaccination (DK: no vaccina-tion, MI: 75% of studied herds). On a regional basisa study in England and Wales showed a significantrelationship between antibody level in bulk tankmilk and cattle population density (Paton et al.,1998).

With the high number of prevalence studies per-formed in recent years it would be an obvious nextstep to gather all the information on prevalence stud-ies with information on cattle demographics (e.g.,cattle density, herd density, herd size, trade patterns,use of vaccination) from a number of regions andperform formal meta-analyses of possible risk fac-tors. Already some trends are obvious (Tables2.1–2.3). For example, low-prevalence areas such asNorway and Finland have lower cattle populationdensity and smaller herd size than high-prevalenceareas (or areas that used to be high-prevalence areasbefore an eradication program), such as Belgium,Germany, and Denmark.

Despite the intuitively obvious effects of many ofthese risk factors, the documentation in literature isoften scarce and the risk factors can explain only arelatively low percentage of infections. Some of thereasons for difficulties in establishing the risk fac-tors may be due to uncertainties of the time of in-fection. For example, presence of infection in aherd may be due to introduction of infection 3–4years ago as illustrated in the following study.Routinely recorded data (register data) from theDanish Cattle Database and the BVDV eradicationscheme were used to estimate the origin of the firstPI animal occurring in Danish cattle herds (Alban etal., 2001). The study showed that among herds par-ticipating in the milk recording scheme (i.e., dairyherds) the first PI animal was born to a dam that hadbeen in the herd the whole period of lactation in76.1% of the cases, PI animals were introduced di-rectly in 4.5% of the cases and they were introducedby purchase of a dam in late gestation in 4.3% ofthe cases. In the remaining cases the exact sourcecould not be identified. For comparison, amongherds not in the milk recording scheme (mostly beefherds), the first PI animal was born from a dam thathad been in the herd during the whole lactation in26.3% of the cases, whereas PI animals had beendirectly introduced in 33.9% of the cases andthrough dams in late lactation in 2.4% of the cases.It should be noted that some of the dams that hadbeen in the herds throughout their whole lactationcould have been infected by previously purchasedbut unidentified PI animals. Still, most of the PI an-imals in the herds were generated within the herd,meaning that the exact trace back for risk factorsoften needs to go several years back in time.Therefore, cohort studies linking the time of infec-tion with the time of occurrence of risk factorswould be useful to further quantify the importanceof risk factors.



QUANTIFICATION OF THEEFFECTS/CONSEQUENCES OFBVDV INFECTIONS ON DISEASEAND PRODUCTIONThe measures of consequences include the presenceof the disease as well as reduced production. Theclinical manifestations often used in epidemiologi-cal studies to measure the effect of transient infec-tion include diarrhea, reduced milk production, re-productive disorders, increased occurrence of otherdiseases, and death. Losses from fetal infection in-clude abortions, congenital defects, birth of weakand undersized calves, unthrifty PI animals, anddeath among PI animals (Baker, 1995). These con-sequences of BVDV infections have been docu-mented in several experimental and case studies,whereas quantification of the consequences in largerobservational studies is often less well documented.The validity of these data in relation to their verifi-cation of being BVD-related varies substantially.The losses following acute (transient) infection areespecially difficult to quantify. Losses such as respi-ratory disease, repeat breeding, or abortions haveseldom been measured together with a rise in anti-body titer, and therefore such losses are often ana-lyzed in retrospective studies, where the time periodof transient infections is only loosely estimated.Losses among the PI animals have been quantifiedwith a higher degree of certainty because the pres-ence of persistent infection together with clinicalsymptoms has often been documented. Measures ofprevalence and incidence together with quantifica-tion of disease and production parameters are thefoundation for calculating the effect of BVDV ineconomical terms. The following sections empha-size measures of consequences that have been quan-tified under field conditions followed by estimationof economic impact of BVDV infections. On someoccasions quantitative assessments from case stud-ies and experimental studies are also given whenthey are judged to have sufficient quantitative inter-pretation.

The effects of infection have often been investi-gated in case studies and experimental studies, butto a lesser extent in observational studies. All ofthese approaches have drawbacks associated withthem. For example, more severe outbreaks tend toattract attention and are liable to be investigated andpublished leading to overestimation of the effect ofBVDV infections. In experimental studies, it is dif-ficult to mimic natural circumstances and it hasoften been difficult to demonstrate clinical signs ofacute BVD, although reproductive disorders (e.g.,

congenital defects) have frequently been docu-mented. Observational studies often suffer from thelack of knowledge of how many animals actually se-roconverted and at what time points. Therefore, theeffects seen in observational studies are probablyunderestimated when considering the effect in theindividual seroconverting animal. However, obser-vational studies are useful for estimation of the ef-fect of infection in a partly immune population.

SUBCLINICAL INFECTIONS

The clinical spectrum of postnatal infection of im-munocompetent animals ranges from subclinical in-fection to severe disease followed by death. Severalyears ago it was estimated that 70–90% of transientinfections were mild or subclinical (Ames et al.,1986). Although the risk of disease following tran-sient infection has not been exactly quantified, epi-demiological investigations indicate that these esti-mates are good. The clinical consequences of acuteBVDV infection were studied in a large dairy herdwhere BVDV was circulating for approximately 2.5years (Moerman et al., 1994). Among 136 cattlewith transient infection, 129 (95%) showed no clin-ical signs of infection, 5 animals were slightly af-fected, and only 2 animals were severely ill. Further-more, many studies in the 1990s have shown thatBVDV infection was more widespread than previ-ously thought, indicating the presence of a largenumber of subclinical infections.

REDUCED MILK YIELD

Milk yield was studied in a longitudinal study of 54cows of which 22 seroconverted (Moermann et al.,1994). A gradual drop in the milk yield (measured asmoving average over 3 days) of 10% or more within10 days was seen in 18 out of 22 cows that serocon-verted and 9 out of 32 cows that did not seroconvert,giving an OR of reduced milk production of 11.5.On the other hand comparison of antibody level inbulk tank milk and milk yield did not reveal a signif-icant association (Graham et al., 2001).

OCCURRENCE OF OTHER DISEASESFOLLOWING TRANSIENT INFECTION INCOWS

Examination of bulk tank milk samples from 213Swedish dairy herds in 1988 and 1989 indicated thatrecent infection had occurred in 7 herds (Niskanenet al., 1995). Compared to 84 herds with continuouslow level of antibodies in bulk tank milk (i.e., unin-fected), cows in the 7 infected herds had increasedORs for several clinical conditions, with mastitis,

Risk Assessment 53


miscellaneous diseases, and retained placenta beingsignificant (OR and 95% CI for OR are given inparenthesis): mastitis (OR = 1.8; CI = 1.1–2.8),tramped teat (OR = 1.6; CI = 0.9–2.8), ketosis (OR= 1.3; CI = 0.6–2.2), miscellaneous diseases (OR =2.8; CI = 1.7–4.4) and retained placenta (OR = 2.8;CI = 1.6–4.7). Increased risk of retained placentamay also be seen among dams delivering PI calves(i.e., several months after the transient infection).Thus, 5 (41.7%) out of 12 PI dams had retained pla-centa compared to 7 (3.5%) cases of retained pla-centa in 198 non-PI dams (P<0.001) (Larsson et al.,1994).

In Norway, the national screening of all dairyherds for BVDV antibodies in bulk milk in 1992 and1993 was used to estimate the effect of transient in-fection on udder health (Waage, 2000). A total of404 herds with a significant rise in antibody level inbulk milk were compared to herds with continuouslow levels of antibodies. In infected herds, there wasa 7% (CI = 0.2%–11.4%) increase in the incidenceof clinical mastitis in the year of exposure as com-pared to negative control herds but there was no sig-nificant effect on the bulk milk somatic cell count orthe culling rate due to mastitis. However, a study byGraham et al. (2001) found a significant increase insomatic cell count with increasing antibody levels inbulk tank milk (P<0.01).

OCCURRENCE OF OTHER DISEASESFOLLOWING TRANSIENT INFECTION INCALVES

Transient BVDV infections have been associatedwith respiratory diseases and diarrhea in calves.However, due to the multifactorial nature of thesediseases it is difficult to quantify the exact role ofBVDV in their occurrence. Different field investiga-tions indicate that BVDV will often double the riskof these conditions. Among calves entering a feed-lot, 13 of 29 (45%) calves treated for respiratorydisease and 8 of 36 (22%) untreated calves had se-roconverted (Martin and Bohac, 1986). This corres-ponds to an OR of 2.8 for treatment following sero-conversion. The effect of BVDV infection on theoccurrence of respiratory diseases has also beendemonstrated indirectly by comparing calves receiv-ing colostrum from seronegative dams with calvesreceiving colostrum from seropositive dams (Moer-man et al., 1994). Among calves exposed to BVDVsoon after birth, 30 of 44 calves (68.2%) receivingcolostrum from seronegative dams developed mod-erate or severe bronchopneumonia whereas 35 of 86calves (40.7%) receiving colostrum from seroposi-

tive dams developed moderate or severe broncho-pneumonia. This corresponds to an OR of 3.1 for de-veloping bronchopneumonia if receiving colostrumfrom antibody-negative dams. In another herd 10 of61 calves (16.4%) born in the period when BVDVwas introduced into the herd, underwent veterinarytreatment for respiratory disease and/or enteritis be-tween 2 days and 4 months of age, whereas only 5out of 134 calves (3.7%) born before or after the in-troduction period received treatment (P<0.01)(Larsson et al., 1994). This corresponds to an OR of5.1 of being treated if born in the period of BVDVintroduction. In the same study the mortality was13.1% among calves born in the period of introduc-tion of BVDV as compared to 2.2% in the other pe-riods (P<0.01).

After strict closure of a dairy herd together witheradication of BVDV the incidence of diarrheaamong calves in the first 31 days of life decreasedsignificantly from 70.6% to 19.4% (Klingenberg etal., 1999). However, it was difficult to distinguishthe strict effect of BVDV compared to general im-provement in management practice including clo-sure of the herd.

In addition to having an effect on clinical dis-eases, BVDV also has an effect on the occurrence ofother infections. For example infectious bovinerhinotracheitis (IBR) virus was much more wide-spread in the respiratory tract of calves previouslyexposed to BVDV (Potgieter et al., 1984a). Further-more, there are many examples that BVDV will ag-gravate the severeness of other infections—for ex-ample, P. haemolytica (Potgieter et al., 1984b; alsosee Chapter 9 on immunity and immunosuppres-sion). However, the quantitative effect of BVDV onthe prevalence, incidence, and severity of other in-fections is not completely documented.

TRANSIENT INFECTION BY VIRULENTSTRAINS OR BVDV IN COMBINATION WITHOTHER PATHOGENS

For many years postnatal infection of immunocom-petent animals was characterized as mild or subclin-ical in most cases. In the 1990s there came increasedevidence of the importance of highly virulent strainswith much more severe consequences followingtransient infection (hemorrhagic syndrome and per-acute BVD) especially with BVDV genotype 2. Al-though there is no unambiguous connection betweengenotype and virulence, genotype 2 has predomi-nated as the cause of hemorrhagic syndrome andperacute BVDV (Bolin and Ridpath, 1996; Ridpathet al., 2000).



An outbreak of acute BVD in a herd also infectedwith Leptospira hardjo and Coxiella burnetii causedvery severe losses (Pritchard et al., 1989). In a dairyherd with 183 milking cows, the losses included 15dead cows, 20 culled cows (primarily because ofemaciation and infertility), 40 abortions, 18 stillborncalves, and 3 calves dying soon after birth. Othercase descriptions of acute outbreaks have revealedmorbidity of up to 40% and mortality up to 10%(Hibberd et al., 1993; David et al., 1994).

In 1993, BVDV genotype 2 was associated withsevere outbreaks of BVDV in Ontario, Canada. Astudy of seven herds (mean number of cattle: 93;range: 40–191) with outbreaks of acute BVD re-vealed a mortality among adults of 9% (range2–26%), a mortality among immature animals (<2years of age) of 53% (range 13–100%), and 44%(range 3–83%) mean abortion frequency based onthe number of breeding age females. Furthermore,the occurrence of respiratory diseases was high alongwith a number of sequelae (Carman et al., 1998). In1993, the mortality was estimated as 31.5% forgrain-fed calves and 17.1% for milk-fed calves, andthe overall mortality was 32,000 out of 143,000calves (Pellerin et al., 1994).

REPRODUCTIVE DISORDERS

Reproductive disorders are a very important clinicalmanifestation. These are outlined in more detail inChapter 8. This chapter emphasizes the quantitativeaspects of their occurrence. An often-used measurefor reproduction efficiency is the conception rate.Note in the following examples that although themeasure is usually defined as a proportion it is stilltermed conception rate (although it is mathemati-cally not a rate). In an experimental study, 15 heifers(Group I) were infected by contact with PI animals4 days after insemination, another 18 heifers (GroupII) were infected intranasally 9 days before insemi-nation, and 14 uninfected heifers acted as controlgroup (McGowan et al., 1993). The conception rates20 days after insemination (defined as proportion ofeligible heifers that had serum progesterone concen-tration of more than 2.0 ng/ml on day 20 after in-semination) were 60% (9 of 15) and 44% (8 of 18)for Groups I and II as compared to 79% (11 of 14)in the control group. At 77 days after inseminationthe conception rates (defined as proportion of eligi-ble heifers which were palpably pregnant 77 daysafter insemination) were 33% (5 of 15) and 39% (7of 18) for the infected groups; those for the controlgroup remained unchanged. At 77 days, conceptionrates in Groups I and II were both significantly

lower than the control group. A group of seronega-tive cows that were inadvertently exposed to a PI an-imal were bred before (9 cows), during (9 cows), orafter seroconversion (14 cows) (Virakul et al.,1988). The first service conception rates at 21 daysin these three groups were 22.2%, 44.4%, and78.6% (the difference between 22.2% and 78.6%being significant at P<0.05).

The severe effect on conception rates is also seenin observational studies, although the effect is some-what less pronounced. In five Danish dairy herds,the overall conception rate in a risk period of infec-tion (presence of a young PI animal spreading theinfection) was 38% as compared to a conceptionrate of 47% (P<0.001) in a post-risk period (pres-ence of older PI animals where the herd would beimmune) (Houe et al., 1993a). The increase in con-ception rate ranged from 16–64% in individualherds.

In seven Swedish herds with recent infection therewas an increased OR of estrus stimulating treat-ments (OR = 2.2; CI = 1.0–4.9); however, the num-ber of inseminations per service period was not sig-nificantly increased (Niskanen et al., 1995). InNorway the ORs for abortions in herds with recentinfection was calculated as 2.6 and 11.6, respec-tively for two different registration periods (P< 0.01)as compared to BVDV free control herds (Frederik-sen et al., 1998). A Swiss study showed that animalsseroconverting in midgestation (days 46–210) hadan OR of abortion of 3.1 (P = 0.02), whereas an ef-fect was not seen after infection in later stages ofgestation (Rüfenacht et al., 2001). In the study byMoerman et al. (1994), abortions were not more fre-quent in seroconverting cattle. A field investigationof the causes of abortions in the United Kingdomshowed that 39 out of 149 (26%) abortions in 54farms were associated with BVDV (Murray, 1990).

Although congenital defects, stillbirths and weak-born calves have been demonstrated in many exper-imental studies (e.g., Done et al., 1980; Liess et al.,1984; McClurkin et al., 1984; Roeder et al., 1986),they have seldomly been quantified in larger obser-vational studies. Some observational studies did notfind significantly more stillbirths, weak-born calves,or congenital defects (Moerman et al., 1994;Frederiksen et al., 1998; Rüfenacht et al., 2001).

LOSSES AMONG PI ANIMALS

PI animals are often born weak and undersized andwhen they grow up they are at risk of acquiring otherdiseases and being culled due to unthriftiness orthey may die from mucosal disease. In a study the

Risk Assessment 55


mean heart girth at 80 days of age among 8 male PIcalves was 96 cm as compared to 100.5 cm among13 male non-PI calves (P<0.05) (Larsson et al.,1994). The same figures at 180 days were 123.3 vs.130.2, respectively (P<0.05).

The incidence risk of dying or being slaughtereddue to unthriftiness within 1 year among 34 PI ani-mals in 10 Danish dairy herds was 0.28 and 0.31, re-spectively (Houe, 1993). The attributable risk for PIanimals of leaving the herd for any reason within 1year was calculated as 0.35. In eight Danish dairyherds with outbreaks of mucosal disease, 31 of 52 PIanimals died that were identified as PI either at out-break or following whole herd testing (Houe et al.,1994). However, as these were selected outbreaks,this mortality might have been higher than whatwould occur among PI animals in the whole cattlepopulation. On the other hand, had whole herd test-ing not been performed more PI animals wouldprobably have died before they were identified as PIanimals and slaughtered. Another way of judgingthe mortality among PI animals is to look at age spe-cific prevalence. Thus, the prevalence of PI animalsamong all cattle in 19 Danish dairy herds was 1.4%(Houe and Meyling, 1991). However, among cattleyounger than 1 year of age the prevalence was 2.9%indicating a high mortality (or at least culling) alsoamong younger PI animals. In a prevalence study byFrey et al. (1996), 66% of the PI animals wereyounger than 1 year, whereas only 10% were olderthan 2 years.

ECONOMIC IMPACT OF BVDVEconomic calculations and models depend on theunderlying data. The types of data necessary for cal-culation of economic impact have been outlined inprevious sections of this chapter. The significantvariation in occurrence, frequency, and damagingeffects following infection will affect the uncer-tainty in economic estimations. Also, there are indi-cations of a changing disease pattern of BVDV overtime (Evermann and Ridpath, 2002) and hence in-corporation of appropriate variables in the economicmodels should be a dynamic process.

There are two major functions of performing eco-nomic analyses. First, calculation of the pure costsof disease will give an indication of the relevance ofdisease as compared to other diseases in animal hus-bandry. In addition to economics, there are other cri-teria for ranking the relevance and importance ofdiseases, such as animal welfare and human safety.These criteria will not be discussed further althoughanimal welfare is an important aspect of BVDV in-

fections due to the severe suffering from mucosaldisease. The second purpose of performing an eco-nomic analysis is to guide decision makers in deter-mining whether it is cost-effective to control the in-fection and also to compare the cost effectiveness ofdifferent control options. The different control op-tions are discussed in Chapter 14.

The direct economic cost (C) of a disease is thesum of the production losses (L) and the expendi-tures (E) for treatment and prevention (McInerney etal., 1992; Bennett et al., 1999a):

C = L + E

The expenditures for treatment and prevention areoften relatively easy to obtain as they follow specificactions in the herd. The production losses on theother hand are often associated with much higheruncertainty due to biological variation. The produc-tion losses can be obtained if there is general infor-mation about the population size, the incidence ofinfection, damaging effect (including both the prob-ability and magnitude) on disease and production,and the value of each damaging effect:

The calculations need to be done for each damagingeffect and the associated value (drop in milk produc-tion, abortion, death, etc.).

The economic losses at the herd level are oftendetermined from what actually happened in a partic-ular herd or group of herds. In larger populationsand at the national level, it is more common to firstestablish the relevant variables (population size, in-cidence, effect, and values) and then include them ineconomic models. The following section gives someexamples of economic estimations that have beenpublished. The size of the economic estimations iscited as they appeared in the original publications,sometimes together with recalculations in other pub-lications. In addition, if standardization has beenmade (e.g., measuring economic values per cow),this is also included.

Present values (PV) can be calculated based onearlier values (EV) and the real rate of interest (r),where the real rate of interest is the difference be-tween the market rate of interest and the inflationrate (Huirne and Dijkhuizen, 1997). The formula forcalculating the present value based on a given earliervalue n years ago is

In addition, the exchange rate between currenciesshould also be considered.

PV EV r n= × +( )1

L population size incidence effect value= × × ×∑



ECONOMIC LOSSES AT THE HERD LEVEL

There are many ways to estimate the economic im-pact at the herd level in any given outbreak situation.The outbreak situation can consist of accumulatedclinical data of both reproductive disorders andlosses among PI animals over a longer time period,or they may only estimate the sudden losses of anoutbreak of mucosal disease over a few months. Inany case they usually reflect observable problems.Often only outright losses are included.

In an outbreak of abortion, neonatal death andsubsequent mucosal disease in a 67-cow dairy herdthe estimated economic losses varied between£1720 and £4115, depending on whether dead orculled animals were replaced (Duffel et al., 1986).In U.S. dollars this corresponds to approximately$40–95 per dairy cow. A study of 14 outbreaks in theNetherlands including losses due to abortions, still-births, various clinical lesions, and mucosal diseasehad losses in the range of 42Dfl. to 285Dfl.($24–161) per dairy cow (Wentink and Dijkhuizen,1990). In outbreaks of mucosal disease in eightDanish dairy herds, the losses due to mucosal dis-ease only were estimated to be $13–39 per animalcorresponding to $33–98 per dairy cow (Houe et al.,1994).

Case descriptions of severe outbreaks of acuteBVD have resulted in losses several times higher.An outbreak with a combined infection with BVDV,Leptospira Hardjo and Coxiella burnetii was, asso-ciated with death in adult cows, abortion, and neona-tal mortality. This resulted in a total cost of morethan £50,000 (Pritchard et al., 1989) correspondingto $410 per cow. Seven outbreaks of severe acuteBVD in Canada with high mortality and abortionrate caused losses in the range of $40,000–100,000per herd (Carman et al., 1998). These herds had be-tween 40 and 191 animals, so the losses per cowwould be higher than $400 per cow.

This means that whether we are dealing with amore “classic” outbreak with losses in the range of$50–100 per cow (note that these figures are per allcows being in the herd and not only diseased cows)or we are dealing with severe outbreaks of acuteBVD with losses of more than $400 per cow, thelosses make up a significant percentage of the totalvalue of the livestock. Due to the variation of theseestimations and the uncertainty of how representa-tive these cases are, we can only have rough indica-tions of the true losses or costs. Therefore BVDV isamong the diseases where an outbreak can be devas-tating for the economy of the individual farmer.

ECONOMIC LOSSES IN LARGERPOPULATIONS AND AT THE NATIONAL LEVEL

The losses or costs for the cattle industry may bebetter calculated using average values for variablesobtained from epidemiological studies rather thansumming up the losses described in case studies.There are significant differences in the virulence ofBVDV strains and both genotype 1 and genotype 2have high virulent strains. We only have sparse in-formation on the occurrence of strains with differentvirulence, so this information cannot be directly in-corporated in the models for calculating the eco-nomic impact. Therefore, many calculations havebeen performed anticipating an average effect of in-fections within the area, or the estimations havebeen made anticipating either a high-virulent or alow-virulent strain being present.

Using the formula for production loss, the eco-nomic losses in Denmark before initiation of theeradication campaign have been estimated as $20million per million calvings (Houe et al., 1993b;Houe, 1999). Including the variables on damagingeffects from virulent strains the losses were esti-mated as $57 million per million calvings (Houe,1999). These calculations were based on an annualincidence risk of 34% corresponding to very highprevalence of infection, and therefore the magnitudeof the losses should be adjusted when consideringareas with a lower prevalence (for details see Houe,1999). The losses in Norway (at the start of the erad-ication program) were estimated as 26 million NKrper year (Valle et al., 2000b) corresponding roughlyto $10 million per million calvings. This corre-sponds well with the lower prevalence seen in Nor-way as compared to the prevalence in Denmark. InGreat Britain, the direct costs associated withBVDV have been estimated as between 5 and 31million£ (1996 values) (Bennett et al., 1999b),which with an estimated 4.5 million calvingsroughly corresponds to $6 million per million calv-ings. In this case, the population at risk was consid-erably lower compared to the one used in the Danishmodel. In Canada, the total annual cost for an aver-age 50-cow dairy herd have recently been estimatedas $48 per cow (Chi et al., 2002).

It is noteworthy that in a high-prevalence area theaverage loss due to infection is almost half theamount of losses calculated in the outbreak situa-tion. But in many herds the losses were not very ob-vious because most herds were partly immune andabortions and deaths were more often seen as occa-sional cases than actual outbreaks. Therefore calcu-

Risk Assessment 57


lation of the economic losses has been an importantmotivator for considering control programs bothwhen farmers have experienced outbreaks and whenthere was a more continuous occurrence of variouslosses.

OPTIMIZING DECISIONS BASED ONECONOMIC CALCULATIONS

From an economical point of view the optimal situ-ation is when the total costs are minimized—i.e., thesum of losses and expenditures are minimized,which is the basic economical principle (McInerneyet al., 1992). Naturally, the more money we spendon treatment and prophylaxis the more the produc-tion losses will be reduced. However, we shouldonly continue to increase expenses as long as $1spent on treatment and prophylaxis will cause a re-duction in production loss of more than $1. In prac-tice, it is not always possible to change the size ofthe expenditures on a continuous scale because acontrol strategy (e.g., vaccination) will only makesense if it is performed to a certain extent. But stillcalculating losses and expenses for different controlstrategies makes them directly comparable from aneconomic point of view.

ECONOMIC EVALUATION OF CONTROLSTRATEGIES AT THE HERD LEVEL

Decision tree analyses have been used to support de-cision of selecting a blood testing strategy followingoutbreaks of mucosal disease (Houe et al., 1994).The calculations showed that testing a risk group ofanimals being no more than 3 months younger orolder than the age of the index case of mucosal dis-ease was most beneficial to the greatest number offarms. However, the most beneficial strategy was al-ways dependent on the individual farm. Further-more, decision tree models are not suitable for han-dling the long-term effect of control decisions. Theeffect of different control strategies has been evalu-ated in simulation models. A model simulatingDutch conditions showed that culling of PI animalswas unattractive (Pasman et al., 1994). However, theconclusion was highly dependent on the risk of rein-fection. For comparison, a Markov Chain model as-suming that reinfection could be avoided was infavor of eradication. Bennett (1992) described a de-cision support system showing that depending onthe farm-specific situation, the recommendationcould be any of the following: a do nothing strategy,a culling strategy, or a vaccination strategy. In con-clusion, a general optimal control strategy cannot begiven because it will depend on an evaluation of the

farmer’s capability to avoid reinfection and certainlyalso of the general infection status of the area thefarm is located in.

ECONOMIC EVALUATION OF NATIONALERADICATION PROGRAMS

In Norway, a cost-benefit evaluation of the controland eradication program from 1993 to 1997 wasmade by subtracting the program costs from thebenefits (Valle et al., 2000b). The 1993 net presentvalue for the 5 years of the program was calculatedas $6 million, and the program was cost-effective inthe second year. For comparison, it was estimatedthat an eradication program in France would take 15years to be cost-effective (Dufour et al., 1999).Before the eradication program was initiated, the an-nual losses in Denmark were estimated as approxi-mately $20 million. Thus, this amount is a possiblebenefit that can be obtained after total eradication.In the initial phases the cost of the program was ap-proximately $9 million per year (Bitsch and Røn-sholt, 1995), thereafter being reduced to $3.5 mil-lion per year. The costs for continuous monitoringhave not been established but will be significantlylower. Because the country is close to total eradica-tion, an annual benefit of almost $20 million is ob-tained and the program can therefore be consideredhighly cost-effective. As for the individual herdlevel, the calculations will be highly dependent onthe capability to stay free of infection. Another im-portant factor for making the Scandinavian programvery cost-effective is the low cost of using bulk tankmilk testing for antibody for the continued monitor-ing of dairy herds.

CONCLUDING REMARKS ANDPERSPECTIVESThis chapter summarizes results of epidemiologicalstudies on the occurrence of bovine virus diarrheavirus (BVDV) infections, identification of risk fac-tors, and effects of the infection on disease and pro-duction. The occurrence is described as prevalenceand incidence of antibodies and virus, both at the an-imal level and at the herd level. It is concluded thatlaboratory methods are sufficiently valid at the ani-mal level for their use in epidemiological studies. Atthe herd level the sensitivity and specificity showlarger variation. Studies show that in many areas theprevalence of PI animals is 0.75–2% and that60–70% of animals are antibody-positive. However,there may be regional differences in prevalence. Forexample, many studies in the U.S. have revealed alower prevalence of PI animals than those reported



in European studies. Other differences in prevalencemay be related to cattle density and managementstyle. For example, some areas with low cattle pop-ulation density and small herds seem to have rela-tively low prevalence.

Susceptible animals housed together with PI ani-mals will have a very high incidence rate of infec-tion, and almost 100% of the animals will be in-fected within a few months. In production systemsin which subgroups of the herd are segregated fromPI animals, these subgroups of animals can stay un-infected for long periods.

Risk factors for BVDV infections are often a re-flection of the risk of direct or indirect contact withPI animals. Risk factors include livestock trade, pas-turing of animals, use of common pasture, cattledensity in the area, herd size, number of infectedneighbors, fence breakout, other animal contact be-tween herds (e.g., exhibitions), insufficient hygienicprocedure, and occurrence of other species (sheep,wild life). The documentation in literature on theimportance of risk factors is often surprisingly lowas compared to the obvious effect one would antici-pate. Often the proven risk factors explain only a rel-atively low percentage of infections. Some of thereasons for the difficulties of establishing the impor-tance of risk factors may be due to uncertainties ofthe time of infection of herds. The improved surveil-lance systems for herds make larger risk factor stud-ies possible in the future with more precise identifi-cation of the herd infection time. Furthermore manyrisk factors are confounded among each other in-creasing the need for studies on a larger scale. Withthe high number of prevalence studies being per-formed, it would be recommended to gather infor-mation on risk factors in a uniform way betweenstudies in order to combine them in formal meta-analyses. These studies should also include infor-mation on cattle demographics (e.g., cattle density,herd density, herd size, trade patterns, use of vac-cination).

Postnatal infection of immunocompetent animalsis often subclinical in most animals. Some mayshow the typical clinical signs of acute BVD.Transient infection in cows may be followed by re-duced milk production, higher incidence of otherdiseases—such as mastitis—ketosis, and retainedplacenta. Transient infection in calves may be fol-lowed by increased frequency of respiratory dis-eases and diarrhea. The infection can have signifi-cant effect on reproductive disorders, such asconception rate, abortions, congenital effects, andweak-born calves. PI calves are often weak and un-

dersized and have a considerable increased risk ofeither being culled early due to unthriftiness ordying throughout their lifetime. Different observa-tional studies on the effect of infection show re-markable differences in the clinical consequences.Although some of the differences obviously seem tobe due to differences in virulence, a closer under-standing of the host-agent-environment triad seemsrelevant for the understanding of the variation.There is a high variation in the general disease levelin herds without BVDV infection, and the impor-tance of infection to herds with a general high healthstatus as compared to herds with low health statusneeds to be clarified. A lot of the variation seen inthe observational studies can also simply be due tochance. If for example the virus is spread at a timewhen there are many cattle in first trimester meansdifferent consequences compared to the situationwhere most cattle would be in the last trimester.Such differences would be most obvious in herdswith seasonal calvings. Furthermore many observa-tional studies are based on the fact that some infec-tion took place, but without knowing exactly howmany seroconversions occurred.

Estimations of economic impact are often at-tempted by combining all the information presentedin this chapter, including the occurrence of the in-fection (prevalence and incidence) and quantifica-tions on the different clinical and production conse-quences following infection. In order to improve theeconomic models, a closer description of distribu-tion of the different genotypes and especially thedifferences in virulence both within and betweengenotypes would be needed.

ACKNOWLEDGMENTSThank you to associate professor Annette KjærErsbøll and assistant professors Søren SaxmoseNielsen and Nils Toft for critical reading of themanuscript.

REFERENCESAlban L, Stryhn H, Kjeldsen AM, et al.: 2001, Esti-

mating transfer of bovine virus-diarrhoea virus inDanish cattle by use of register data. Prev Vet Med52:133–146.

Alenius S, Jacobsen SO, Cafaro E: 1986, Frequencyof bovine viral diarrhea virus infections in Swedenamong heifers selected for artificial insemination.Proc World Congr Diseases of Cattle 14:204–207.

Alenius S, Lindberg A, Larsson B: 1997, A nationalapproach to the control of bovine viral diarrhoeavirus. In: Proceedings of the Third ESVV

Risk Assessment 59


Symposium on Pestivirus Infections. Eds. EdwardsS, Paton DJ, Wensvoort G, 19–20 September 1996,pp.162–169. Lelystad, The Netherlands.

Ames TR: 1986, The causative agent of BVD: Its epi-demiology and pathogenesis. Vet Med 81:848–869.

Baker JC: 1995, The clinical manifestations of bovineviral diarrhea infection. Vet Clin North Am FoodAnim Pract 11:425–445.

Barber DML, Nettleton PF: 1993, Investigations intobovine viral diarrhoea virus in a dairy herd. Vet Rec133:549–550.

Beaudeau F, Belloc C, Seegers H, et al.: 2001,Informative value of an indirect enzyme-linked im-munosorbent assay (ELISA) for the detection ofbovine viral diarrhoea virus (BVDV) antibodies inmilk. J Vet Med B 48:705–712.

Bennett RM: 1992, Case-study of a simple decisionsupport system to aid livestock disease controldecisions. Agric Sys 38:111–129.

Bennett RM, Christiansen K, Clifton-Hadley RS:1999a, Estimating the costs associated with en-demic diseases of dairy cattle. J Dairy Res66:455–459.

Bennett RM, Christiansen K, Clifton-Hadley RS:1999b, Modelling the impact of livestock diseaseon production: Case studies of non-notifiable dis-ease of farm animals in Great Britain. Anim Sci68:681–689.

Bitsch V, Hansen K-EL, Rønsholt L: 2000,Experiences from the Danish programme for eradi-cation of bovine virus diarrhoea (BVD) 1994–1998with special reference to legislation and causes ofinfection. Vet Microbiol 77:137–143.

Bitsch V, Rønsholt L: 1995, Control of bovine viraldiarrhea virus infection without application of vac-cines. Vet Clin North Am Food Anim Pract11:627–640.

Bolin SR, McClurkin AW, Coria MF: 1985, Fre-quency of persistent bovine viral diarrhea virusinfection in selected cattle herds. Am J Vet Res46:2385–2387.

Bolin SR, Ridpath J: 1998, Prevalence of bovine viraldiarrhea virus genotypes and antibody against thoseviral genotypes in fetal bovine serum. J Vet DiagnInvest 10:135–139.

Bolin SR, Ridpath JF: 1996, The clinical significanceof genetic variation among bovine viral diarrheaviruses. Vet Med 91:958–961.

Braun U, Landolt G, Brunner D, Giger T: 1997,Epidemiologishe Untersuchungen über das Vor-kommen von BVD/MD bei 2892 Rindern in 95Milchviehbetriben. Schweizer Archiv fürTierheilkunde 139:172–176.

Braun U, Schönmann M, Ehrensperger F, et al.: 1998,Epidemiology of bovine virus diarrhoea in cattle on

communal alpine pastures in Switzerland. J VetMed A 45:445–452.

Canal CW, Strasser M, Hertig C, et al.: 1998,Detection of antibodies to bovine viral diarrhoeavirus (BVDV) and characterization of genomes ofBVDV from Brazil. Vet Microbiol 63:85–97.

Carman S, van Dreumel T, Ridpath J, et al.: 1998,Severe acute bovine viral diarrhea in Ontario1993–1995. J Vet Diagn Invest 10:27–35.

Chi J, VanLeeuwen JA, Weersink A, Keefe GP: 2002,Direct production losses and treatment costs frombovine viral diarrhoea virus, bovine leukosis virus,Mycobacterium avium subspecies paratuberculosisand Neospora caninum. Prev Vet Med 55:137–153.

Cho HJ, Masri SA, Deregt D, et al.: 1991, Sensitivityand specificity of an enzyme-linked immunosorbentassay for the detection of bovine viral diarrheavirus antibody in cattle. Can J Vet Res 55:56–59.

Couvreur B, Letellier C, Collard A, et al.: 2002,Genetic and antigenic variability in bovine viral di-arrhea virus (BVDV) isolates from Belgium. VirusRes 85:17–28.

David GP, Crawshaw TR, Gunning RF, et al.: 1994,Severe disease in adult dairy cattle in three UKdairy herds associated with BVD virus infection.Vet Rec 134:468–472.

Done JT, Terlecki S, Richardson C, et al.: 1980,Bovine virus diarrhoea-mucosal disease virus:Pathogenicity for the fetal calf following maternalinfection. Vet Rec 106:473–479.

Drew TW, Sandvik T, Wakeley P, et al.: 2002, BVDvirus genotype 2 detected in British cattle. Vet Rec151:551–551.

Drew TW, Yapp F, Paton DJ: 1999, The detection ofbovine viral diarrhoea virus in bulk milk samplesby the use of a single-tube RT-PCR. Vet Microbiol64:145–154.

Duffel SJ, Sharp MW, Bates D: 1986, Financial lossresulting from BVD-MD virus infection in a dairyherd. Vet Rec 118:38–39.

Dufour B, Repiquet D, Touratier A: 1999, Economicstudies in animal health decision-making: The cost-benefit ratio of eradicating bovine virus diarrhoeain France. Rev Sci Tech Off Int Epiz 18:520–532.

Durham PJK, Hassard LE: 1990, Prevalence of anti-bodies to infectious bovine rhinotracheitis, parain-fluenza 3, bovine respiratory syncytial, and bovineviral diarrhea viruses in cattle in Saskatchewan andAlberta. Can Vet J 31:815–820.

Edwards S, Drew TW, Bushnell SE: 1987, Prevalenceof bovine virus diarrhoea virus viraemia. Vet Rec120:71.

Ersbøll AK, Stryhn H: 2000, Epidemiological model-ling of infectious disease in animals: Bovine virusdiarrhoea in Danish dairy herds. In: Proceedings of



the 9th Symposium of the International Society forVeterinary Epidemiology and Economics (ISVEE),Breckenridge, Colorado, USA, August 6–11, 2000,pp. 520–522.

Evermann JF, Ridpath JF: 2002, Clinical and epidemi-ologic observations of bovine viral diarrhea virus inthe northwestern United States. Vet Microbiol89:129–139.

Flores EF, Weiblen R, Gil LHVG, et al.: 2000,Antigenic diversity of Brazilian isolates of bovineviral diarrhea virus (BVDV): Implications for diag-nosis and immunization strategies. Arq Bras MedVet Zootec 52:11–17.

Fredriksen B, Ødegaard SA, Løken T: 1998, The ef-fect of bovine virus diarrhoea virus on reproductionin recently infected Norwegian dairy herds. ActaVet Scand 39:99–108.

Frey HR, Flebbe U, Liess B: 1996, Prävalenz undklinische Symptomatik persistenter BVD-virusinfektionen in Rinderbeständen Nieder-sachsens. Der Praktische Tierarzt 77:49–52.

Fulton RW, Saliki JT, Confer AW, et al.: 2000, Bovineviral diarrhea virus cytopathic and noncytopathicbiotypes and type 1 and 2 genotypes in diagnosticlaboratory accessions: Clinical and necropsy sam-ples from cattle. J Vet Diagn Invest 12:33–38.

Fulton RW, Seger CL: 1982, Infectious bovine rhino-tracheitis, bovine viral diarrhea and parainfluenza-3viral antibodies in Louisiana cattle. Bov Pract17:63–65.

Ganaba R, Bélanger D, Dea S, Bigras-Poulin M:1995, A seroepidemiological study of the impor-tance in cow-calf pairs of respiratory and entericviruses in beef operations from NorthwesternQuebec. Can J Vet Res 59:26–33

Graham DA, German A, McLaren LE, FitzpatrickDA: 2001, Testing of bulk tank milk from NorthernIreland dairy herds for viral RNA and antibody tobovine viral diarrhoea virus. Vet Rec 149:261–265.

Grom J, Barlic-Maganja D: 1999, Bovine viral diar-rhoea (BVD) infections—Control and eradicationprogramme in breeding herds in Slovenia. VetMicrobiol 64:259–264.

Harkness JW, Sands JJ, Richards MS: 1978,Serological studies of mucosal disease virus inEngland and Wales. Res Vet Sci 24:98–103.

Hibberd RC, Turkington A, Brownlie J: 1993, Fatalbovine viral diarrhoea virus infection of adult cat-tle. Vet Rec 132:227–228.

Houe H: 1992a, Age distribution of animals persist-ently infected with bovine virus diarrhea virus intwenty-two Danish dairy herds. Can J Vet Res56:94–98.

Houe H: 1992b, Serological analysis of a small herdsample to predict presence or absence of animals

persistently infected with bovine viral diarrhoeavirus (BVDV) in dairy herds. Res Vet Sci53:320–323.

Houe H: 1993, Survivorship of animals persistentlyinfected with bovine virus diarrhoea virus (BVDV).Prev Vet Med 15:275–283.

Houe H: 1994, Bovine virus diarrhoea virus:Detection of Danish dairy herds with persistentlyinfected animals by means of a screening test of tenyoung stock. Prev Vet Med 19:241–248.

Houe H: 1999, Epidemiological features and econom-ical importance of bovine virus diarrhoea virus(BVDV). Vet Microbiol 64:89–107.

Houe H: 2001, The BVDV eradication effort. SpringMtg Acad Vet Consultants, Omaha, NE, April 5–7,2001, 29: 33–43.

Houe H, Baker JC, Maes RK, et al.: 1995a, Com-parison of the prevalence and incidence of infectionwith bovine virus diarrhoea virus (BVDV) inDenmark and Michigan and association withpossible risk factors. Acta Vet Scand 36:521–531.

Houe H, Baker JC, Maes RK, et al.: 1995b,Prevalence of cattle persistently infected withbovine viral diarrhea virus in 20 dairy herds in twocounties in central Michigan and comparison ofprevalence of antibody-positive cattle among herdswith different infection and vaccination status. J VetDiag Invest 7:321–326.

Houe H, Lloyd JW, Baker JC: 1994, Decision treeanalysis of control strategies in Danish dairy herdswith outbreaks of mucosal disease. Prev Vet Med21:133–146.

Houe H, Meyling A: 1991, Prevalence of bovine virusdiarrhoea (BVD) in 19 Danish dairy herds and esti-mation of incidence of infection in early pregnancy.Prev Vet Med 11:9–16.

Houe H, Palfi V: 1993, Estimation of herd incidenceof infection with bovine virus diarrhoea virus(BVDV) in herds previously without animals per-sistently infected with BVDV. Acta Vet Scand34:133–137.

Houe H, Pedersen KM, Meyling A: 1993a, The effectof bovine virus diarrhoea virus (BVDV) infectionon conception rate. Prev Vet Med 15:117–123.

Houe H, Pedersen KM, Meyling A: 1993b, A comput-erized spread sheet model for calculating total an-nual national losses due to bovine virus diarrhoeavirus (BVDV) infection in dairy herds and sensitiv-ity analysis of selected parameters. In: Proc SecondSymp on Ruminant Pestiviruses, Annecy, France.1–3 October 1992. Ed. Edwards, S, pp.179–184.

Houe H, Uttenthal A, Pedersen KM, et al.: 1997, Riskfactors for a change in bovine virus diarrhoeaseropositivity in 41 Danish dairy herds during twoyears. In: Proc Third ESVV Symp on Pestivirus

Risk Assessment 61


Infections, Lelystad, The Netherlands, 19–20September 1996. Eds. Edwards S, Paton DJ,Wensvoort G, pp. 112–115.

Howard CJ, Brownlie J, Thomas LH: 1986, Preva-lence of bovine virus diarrhoea virus viraemia incattle in the UK. Vet Rec 119:628–629.

Howard TH, Bean B, Hillman R, Monke DR: 1990.Surveillance for persistant bovine viral diarrhea in-fection in four artificial insemination centers. J AmVet Med Assoc 196:1951–1955.

Huirne RBM, Dijkhuizen AA: 1997, Basic methods ofeconomic analysis. In: Animal Health Economics.Principles and Applications. Eds. Dijkhuizen AAand Morris RS. University of Sydney. 1997. pp25–39.

Hyera JMK, Liess B, Frey H-R: 1991, Bovine viraldiarrhoea virus infection in cattle, sheep, and goatsin Northern Tanzania. Trop Anim Hlth Prod23:83–94.

Klingenberg KD, Vagsholm I, Alenius S: 1999,Incidence of diarrhea among calves after strict clo-sure and eradication of bovine viral diarrhea virusinfection in a dairy herd. J Am Vet Med Assoc214:1824–1828.

Kramps JA, van Maanen C, van De Wetering G, et al.:1999, A simple, rapid and reliable enzyme-linkedimmunosorbent assay for the detection of bovinevirus diarrhoea virus (BVDV) specific antibodies incattle serum, plasma and bulk milk. Vet Microbiol64:135–144.

Larsson B, Nikanen R, Alenius S: 1994, Natural in-fection with bovine virus diarrhoea virus in a dairyherd: A spectrum of symptoms including early re-productive failure and retained placenta. AnimRepro Sci 36:37–48.

Liess B, Frey HR, Trautwein G, Peters W: 1987,Häufigkeit des Auftretens persistierender BVD-Virusinfektionen und ihre Auswirkungen in derRinderpopulation. Dtsch tierärztl Wschr 94:583–585.

Liess B, Orban S, Frey HR, et al.: 1984, Studies ontransplacental transmissibility of a bovine virus di-arrhoea (BVD) vaccine virus in cattle. II. Inocu-lation of pregnant cows without detectable neutral-izing antibodies to BVD virus 90–229 days beforeparturition (51st to 190th day of gestation). Zbl VetMed B 31:669–681.

Lindberg ALE, Alenius S: 1999, Principles for eradi-cation of bovine viral diarrhoea virus (BVDV) in-fections in cattle populations. Vet Microbiol64:197–222.

Løken T, Krogsrud J, Larsen IL: 1991, Pestivirus in-fections in Norway. Serological investigations incattle, sheep and pigs. Acta Vet Scand 32:27–34.

Luzzago C, Piccinini R, Zepponi A, Zecconi A: 1999,Study on prevalence of bovine viral diarrhoea virus

(BVDV) antibodies in 29 Italian dairy herds withreproductive problems. Vet Microbiol 64:247–252.

Mainar-Jaime RC, Berzal-Herranz B, Arias P, Rojo-Vazquez, FA: 2001, Epidemiological pattern andrisk factors associated with bovine viral-diarrhoeavirus (BVDV) infection in a non-vaccinated dairy-cattle population from the Asturias region of Spain.Prev Vet Med 52:63–73.

Martin SW, Bohac JG: 1986, The association betweenserological titers in infectious bovine rinotracheitisvirus, bovine virus diarrhea virus, parainfluenza-3virus, respiratory syncytial virus and treatment forrespiratory disease in Ontario feedlot calves. CanVet J 50:351–358.

Martin SW, Shoukri M, Thorburn MA: 1992,Evaluating the health status of herds based on tests applied to individuals. Prev Vet Med14:33–43.


McGowan AC, Murray RD: 1999, Health status ofbulls used for natural breeding on farms in SouthWest Scotland. J Vet Med B 46:311–321.

McGowan MR, Kirkland PD, Richards SG,Littlejohns, IR: 1993, Increased reproductive lossesin cattle infected with bovine pestivirus around thetime of insemination. Vet Rec 133:39–43.

McInerney JP, Howe KS, Schepers JA: 1992, Aframework for the economic analyses of disease infarm livestock. Prev Vet Med 13:137–154.

Moerman A, Straver PJ, de Jong MCM, et al.: 1993,A long term epidemiological study of bovine viraldiarrhoea infections in a large herd of dairy cattle.Vet Rec 132:622–626.

Moerman A, Straver PJ, de Jong MCM, et al.: 1994,Clinical consequences of a bovine virus diarrhoeavirus infection in a dairy herd: A longitudinal study.Vet Q 16:115–119.

Msolla P, Sinclair JA, Nettleton P: 1988, Prevalenceof antibodies to bovine virus diarrhoea-mucosaldisease virus in Tanzanian cattle. Trop Anim HlthProd 20:114–116.

Murray RD: 1990, A field investigation of causes ofabortion in dairy cattle. Vet Rec 127:543–547.

Niskanen R: 1993, Relationship between the levels ofantibodies to bovine viral diarrhoea virus in bulktank milk and the prevalence of cows exposed tothe virus. Vet Rec 133:341–344.

Niskanen R, Alenius S, Larsson B, Jacobsson SO:1991, Determination of level of antibodies tobovine virus diarrhoea virus (BVDV) in bulk tankmilk as a tool in the diagnosis and prophylaxis ofBVDV infections in dairy herds. Arch Virol Suppl3:245–251.



Niskanen R, Emanuelson U, Sundberg J, et al.: 1995,Effects of infection with bovine virus diarrhoeavirus on health and reproductive performance in213 dairy herds in one county in Sweden. Prev VetMed 23:229–237.

Niskanen R, Lindberg A, Larsson B, Alenius S: 2000,Lack of virus transmission from bovine viral diar-rhea virus infected calves to susceptible peers. ActaVet Scand 41:93–99.

Niskanen R, Lindberg A, Traven M: 2002, Failure tospread bovine virus diarrhoea virus infection fromprimarily infected calves despite concurrent infec-tion with bovine coronavirus. Vet J 163:251–259.

Nuotio L, Juvonen M, Neuvonen E, et al.: 1999,Prevalence and geographic distribution of bovineviral diarrhoea (BVD) infection in Finland1993–1997. Vet Microbiol 64:231–235.

Obando C, Baule C, Pedrique C, et al.: 1999,Serological and molecular diagnosis of bovine viraldiarrhoea virus and evidence of other viral infec-tions in dairy calves with respiratory disease inVenezuela. Acta Vet Scand 40:253–262.

Paisley LG, Wells S, Schmitt BJ: 1996, Prevalence ofbovine viral diarrhea antibodies in 256 U.S. cow-calf operations: A survey. Theriogenology46:1313–1326.

Pasman EJ, Dijkhuizen AA, Wentink GH: 1994, Astate-transition model to simulate the economics ofbovine virus diarrhoea control. Prev Vet Med20:269–277.

Paton DJ, Christiansen KH, Alenius S, et al.: 1998,Prevalence of antibodies to bovine virus diarrhoeavirus and other viruses in bulk tank milk inEngland and Wales. Vet Rec 142:385–391.

Pellerin C, van den Hurk J, Lecomte J, Tijssen P:1994, Identification of a new group of bovine viraldiarrhea virus strains associated with severe out-breaks and high mortalities. Virology 203:260–268.

Peréz MJ, Wilks CR, Rice M: 1994, Antibodies tobovine viral diarrhea virus in beef cattle. NewZealand Vet J 42:73.


Polak MP, Zmudzinski JF: 1999, Prevalence of bovineviral diarrhea virus (BVDV) infection in cattle inPoland. Vet Microbiol 64:253–257.

Potgieter LND, McCracken MD, Hopkins FM, WalkerRD: 1984a, Effect of bovine viral diarrhea virus in-fection on the distribution of infectious bovinerhinotracheitis virus in calves. Am J Vet Res45:687–689.

Potgieter LND, McCracken MD, Hopkins FM, WalkerRD, Guy JS: 1984b, Experimental production of

bovine respiratory tract disease with bovine viraldiarrhea virus. Am J Vet Res 45:1582–1585.

Pritchard G, Borland ED, Wood L, et al.: 1989,Severe disease in a dairy herd associated with acuteinfection with bovine virus diarrhoea virus,Leptospira hardjo and Coxiella burnetii. Vet Rec124:625–629.

Reinhardt G, Riedemann S, Ernst S, et al.: 1990,Seroprevalence of bovine viral diarrhea/mucosaldisease in southern Chile. Prev Vet Med 10:73–78.


Ridpath JF, Neill JD, Frey M, Landgraf JG: 2000,Phylogenetic, antigenic and clinical characteriza-tion of type 2 BVDV from North America. VetMicrobiol 77:145–155.

Robinson AJ: 1971. Serological evidence of bovinevirus diarrhoea virus in cattle and sheep in thesouth Island of New Zealand. New Zealand Vet J19:223–224.

Roeder PL, Jeffrey M, Cranwell MP: 1986, Pestivirusfetopathogenicity in cattle: Changing sequelae withfetal maturation. Vet Rec 118:44–48.

Rossmanith W, Deinhofer M: 1998, Untersuchungenüber das Vorkommen von BVD-virusinfektionen inniederösterreichischen Milchviehbetrieben. DtschTierärztl Wschr 105:346–349.

Rønsholt L, Nylin B, Bitsch V: 1997, A BVDV anti-gen- and antibody blocking ELISA (DVIV) systemused in a Danish voluntary eradication program. In:Proceedings of the Third ESVV Symposium onPestivirus Infections, Lelystad, The Netherlands,19–20 September 1996. Eds. Edwards S, Paton DJ,Wensvoort G, pp. 150–153.

Rüfenacht J, Schaller P, Audige L, et al.: 2001, Theeffect of infection with bovine viral diarrhea viruson the fertility of Swiss dairy cattle.Theriogenology 56:199–210.

Schreiber P, Dubois F, Drèze F, et al.: 1999,Prevalence of bovine virus diarrhoea virus infectionin Belgian White Blue cattle in southern Belgium.Vet Q 21:28–32.

Ståhl K, Rivera H, Vågsholm I, Moreno-López J:2002, Bulk milk testing for antibody seroprevalenceto BVDV and BHV-1 in a rural region of Peru.Prev Vet Med 56:193–202.

Stärk KDC, Frei-Stäbeli C, Frei PP, et al.: 1997,Häufigkeit und Kosten von Gesundheitsproblemenbei Schweizer Milchkühen und deren Kälbern(1993–1994). Schweiz Arch Tierheilk 139:343–353.

Sudharshana KJ, Suresh KB, Rajasekhar M: 1999,Prevalence of bovine viral diarrhoea virus antibod-

Risk Assessment 63


ies in India. Revue Scientifique et Technique de LOffice International des Epizooties 18:667–671.

Suzan VM, Onuma M, Aguilar RE, Murakami Y:1983, Prevalence of bovine herpesvirus-1, parain-fluenza-3, bovine rotavirus, bovine viral diarrhoea,bovine adeno virus-7, bovine leukaemia virus andbluetongue virus antibodies in cattle in Mexico. JpnJ Vet Res 31:125–132.

Tajima M, Frey HR, Yamato O, et al.: 2001,Prevalence of genotypes 1 and 2 of bovine viral di-arrhea virus in Lower Saxony, Germany. Virus Res76:31–42.

Taylor LF, Janzen ED, van Donkersgoed J: 1997,Losses over a 2-year period associated with fetal in-fection with the bovine viral diarrhea virus in abeef cow-calf herd in Saskatchewan. Can Vet J38:23–28.

Taylor LF, van Donkersgoed J, Dubovi EJ, et al.:1995, The prevalence of bovine viral diarrhea virusinfection in a population of feedlot calves inWestern Canada. Can J Vet Res 59:87–93.

Toft N, Agger JF, Houe H, Bruun J: 2004, Measuresof disease frequency. In: Introduction to VeterinaryEpidemiology. Biofolia, Frederiksberg, Denmark.Eds. Houe H, Ersbøll AK, Toft N, pp. 77–93.

Valle PS, Martin SW, Skjerve E: 2000a, A hierachicaltrend model for bovine virus diarrhoea virus(BVDV) sero-conversion in Norwegian dairy herdsfrom 1993 through 1997. Prev Vet Med 147:39–52.

Valle PS, Martin SW, Tremblay R, Bateman K: 1999,Factors associated with being a bovine-virus diar-rhoea (BVD) seropositive herd in the Møre andRomsdal county of Norway. Prev Vet Med40:165–177.

Valle PS, Skjerve E, Martin SW, et al.: 2000b, A costbenefit evaluation of the Norwegian bovine virusdiarrhoea control and eradication program. In:Proceedings of the 9th Symposium of theInternational Society for Veterinary Epidemiologyand Economics (ISVEE), Breckenridge, Colorado,USA, August 6–11, 2000, pp.1189–1191.

VanLeeuwen JA, Keefe GP, Tremblay R, et al.: 2001,Seroprevalence of infection with Mycobacteriumavium subspecies paratuberculosis, bovine

leukemia virus, and bovine viral diarrhea virus inMaritime Canada dairy cattle. Can Vet J42:193–198.

Vilcek S, Durkovic B, Bobakova M, et al.: 2002,Identification of bovine viral diarrhoea virus 2 incattle in Slovakia. Vet Rec 151:150–152.

Viltrop A, Alaots J, Parn M, Must K: 2002, Naturalchanges in the spread of bovine viral diarrhoeavirus (BVDV) among Estonian cattle. J Vet Med B49:263–269.

Virakul P, Fahning ML, Joo HS, Zemjanis R.: 1988,Fertility of cows challenged with a cytopathic strainof bovine viral diarrhea virus during an outbreak ofspontaneous infection with a noncytopathic strain.Theriogenology 29:441–449.

Waage S: 2000, Influence of new infection withbovine virus diarrhoea virus on udder health inNorwegian dairy herds. Prev Vet Med 43:123–135.

Waage S, Krogsrud J, Nyberg O, Sandvik T: 1997,Results achieved by a national programme for theeradication of bovine virus diarrhoea. In:Proceedings of the Third ESVV Symposium onPestivirus Infections, Lelystad, The Netherlands,19–20 September 1996S. Eds. Edwards S, PatonDJ, Wensvoort G, pp. 170–172.

Wentink GH, Dijkhuizen AA: 1990, Economischegevolgen van een infectie met het Bovine VirusDiarree virus (BVD-virus) op 14 melkveebedrijven.Tijdschr Diergeneesk 115:1031–1040.

Wentink GH, van Exsel ACA, Goey I, van LieshoutJAH: 1991, Spread of bovine virus diarrhoea virusin a herd of heifer calves. Vet Q 13:233–236.

Wittum TE, Grotelueschen DM, Brock KV, et al.:2001, Persistent bovine viral diarrhoea virus infec-tion in US beef herds. Prev Vet Med 49:83–94.

Wolfmeyer A, Wolf G, Beer M, et al.: 1997, Genomic(5’UTR) and serological differences amongGerman BVDV field isolates. Arch Virol142:2049–2057.

Zimmer G, Schoustra W, Graat EAM: 2002,Predictive values of serum and bulk milk samplingfor the presence of persistently infected BVDV car-riers in dairy herds. Res Vet Sci 72:75–82.



TAXONOMY—DEFININGCHARACTERISTICS OF BOVINEVIRAL DIARRHEA VIRUSES

THE FLAVIVIRUS FAMILY

Bovine viral diarrhea viruses belong to the pes-tivirus genus within the Flavivirus family (Heinz etal., 2000) and, as such, possess many traits charac-teristic of this family. Like all flaviviruses, they havea lipid envelope that is derived from the membranesof the infected host cell. The lipid envelope makesthese viruses susceptible to inactivation by solventsand detergents.

Similar to other flaviviruses, BVDV have a single-stranded positive sense RNA genome. The organiza-tion of the genome, which is conserved within theFlavivirus family, consists of a single open readingframe (ORF) flanked by 5’ and 3’ untranslated re-gions (UTR). The ORF is translated as a singlepolyprotein that is cleaved by viral and cellular pro-teinases into the individual viral proteins. The struc-tural proteins are coded in the 5’ end of the genome,and the nonstructural proteins are coded in the re-maining 3’ end of the genome. The replication of theRNA genome occurs in the host cell cytoplasm. Theviral particles are assembled and enveloped at intra-cellular membranes, transported in cytoplasmicvesicles through the secretory pathway and releasedby exocytosis.

CHARACTERISTICS UNIQUE TO THEPESTIVIRUS GENUS

Viruses from the pestivirus genus have several char-acteristics that differentiate them from other mem-ber viruses of the Flavivirus family. Pestiviruses en-code two proteins that are unique to their genus. Thefirst is the nonstructural Npro, which is the first pro-tein coded by the pestivirus ORF. The Npro is a pro-teinase whose only known function is cleaving itself

from the viral polypeptide. The second unique geneproduct is the Erns. It is an envelope glycoproteinthat possesses an intrinsic RNase activity.

Pestiviruses are also unique among flaviviruses intheir resistance to inactivation by low pH. Most fla-viviruses are inactivated by low pH, but pestivirusesare stable over a broad pH range (Hafez and Liess,1972).

MEMBERS OF THE PESTIVIRUS GENUS

There are currently four recognized species withinthe pestivirus genus: BVDV 1 (type virus BVDV 1a-NADL, accession #M31182); BVDV 2 (type virusBVDV 2a-890, accession #U18059); border dis-ease virus (BDV) (type virus BDV-BD31, accession#U70263); and classical swine fever virus (CSFV),previously known as hog cholera virus (type virusAlfort/187, accession #X87939). The pestiviruseswere originally classified into the species BVDV,CSFV, and BDV based on the animal host they orig-inated from. This classification proved problematicbecause some pestiviruses are not restricted to a sin-gle host. Viruses characterized as BVDV, for exam-ple, were shown to infect cattle, sheep, and swine(Paton, 1995a, b). Subsequent efforts to divide pes-tiviruses into species based on monoclonal antibody(Mab) binding (Moennig et al., 1987; Bolin et al.,1988; Hess et al., 1988; Zhou et al., 1989; Edwardset al., 1991; Edwards and Paton, 1995; Paton et al.,1995b) had limited success due to antigenic crossreactivity between proposed species, antigenic vari-ation within proposed species, and the ease withwhich Mab escape mutants could be generated.Phylogenetic analysis of genomic sequences hasproved very helpful in differentiating pestivirus spe-cies (Hofmann et al., 1994; Harasawa, 1996; Becheret al., 1997; Ridpath and Bolin, 1997; Sullivan et al.,1997; Vilcek et al., 1997; Harasawa et al., 2000;Vilcek et al., 2001b). The comparison of genomic

65

3Classification and Molecular Biology

Julia F. Ridpath



sequence of viruses classified as BVDV led to theconclusion that there were actually two genotypes ofBVDV, BVDV 1, and BVDV 2 (Pellerin et al., 1994;Ridpath et al., 1994a). The degree of sequence iden-tity between pestiviruses in the 5’ UTR region is themost frequently used parameter for differentiationpestivirus species (Ridpath et al., 1994a; Harasawaand Mizusawa, 1995; Harpin et al., 1995; Harasawa,1996; Baule et al., 1997; Becher et al., 1997;Giangaspero et al., 1997; Ridpath and Bolin, 1997;Sandvik et al., 1997; Vilcek et al., 1997; Wolfmeyeret al., 1997; Harasawa and Giangaspero, 1998;Ridpath and Bolin, 1998; Shimazaki et al., 1998;Letellier et al., 1999; Sakoda et al., 1999; Flores etal., 2000; Ridpath et al., 2000; Falcone et al., 2001;Tajima et al., 2001; Tajima et al., 2001; Vilcek et al.,2001b; Beer et al., 2002; Couvreur et al., 2002;Evermann and Ridpath, 2002; Flores et al., 2002).However, differences between BVDV 1 and BVDV2 strains are found throughout the genome (Ridpathand Bolin, 1995b, 1997).

All pestiviruses are antigenically cross-reactive.Although antigenic differences do exist betweenpestivirus species, they are not extensive enough forpestivirus species to be recognized as serotypes.Convelescent sera generated against a virus from aparticular pestivirus species will generally have ahigher titer against other viruses from that species asopposed to viruses from the other pestivirus species(Ridpath, 2003). However, animal-to-animal varia-tion and divergence among viruses from the samepestivirus species can make it difficult to repro-ducibly and reliably differentiate pestivirus speciesbased on serology alone (Muller et al., 1997; Bolinand Ridpath, 1998). Mabs have been developed thatdifferentiate between the pestivirus species (Wens-voort et al., 1989a; Zhou et al., 1989; Dahle et al.,1991; Edwards et al., 1991; Kosmidou et al., 1995).However, cross-reactivity between pestivirus spe-cies and variation within any one species make itdifficult to generate a Mab that simultaneously dif-ferentiates between species and still recognizes allthe viruses within one species. Antigenic variation isparticularly pronounced among BVDV 1 andBVDV 2 strains (Ridpath et al., 1994a) and impactson both detection and control.

In addition to the four recognized pestivirus spe-cies, three putative species have been suggested.These putative species are based on viruses isolatedfrom a giraffe, a reindeer, and a pronghorn antelope(Becher et al., 1997; van Rijn et al., 1997; Harasawaet al., 2000; Vilcek et al., 2001b). Only one virus has been isolated for each of these putative species,

and no correlation with clinical disease has beenreported.

THE PESTIVIRUS VIRION

Pestivirus virions are enveloped, spherical particlesthat are 40 to 60 nm in diameter. The virions aremade up of a central capsid, composed of the vi-rally encoded C protein and the genome RNA, sur-rounded by a lipid bilayer. The capsid appears as anelectron-dense inner core with a diameter of ap-proximately 30 nm (Horzinek et al., 1971). Thestructure and symmetry of the core has not been de-termined. Three virus encoded proteins—Erns, E1,and E2—are associated with the lipid bilayer enve-lope. The envelope surrounding the virion is pleo-morphic, which impedes purification of infectiousparticles by banding in sucrose gradients and iden-tification by electron microscopy. There appear tobe 10–12 nm ring-like subunits on the surface of thevirus envelope (Heinz et al., 2000). The Mr of thevirion is estimated as 6.0 � 107, and the buoyantdensity in sucrose is 1.10–1.15 gm/cm3 (Heinz etal., 2000).

Virions are stable within a pH range of 5.7 to 9.3(Hafez and Liess, 1972). Infectivity is not affectedby freezing but decreases at temperatures above40°C (Heinz et al., 2000). Like other envelopedviruses, BVDV are inactivated by organic solventsand detergents. Other methods of inactivation in-clude Trypsin treatment (0.5 mg/ml, 37°C, 60 min)(Liess, 1990), ethylenimine (reduction of 5 log10

units using 10 mM at 37°C for 2 h) (Preuss et al.,1997), electron beam irradiation (4.9 and 2.5 kGyneeded to reduce virus infectivity 1 log10 unit forfrozen and liquid samples respectively) (Preuss etal., 1997), and gamma irradiation (20–30 kGy)(Miekka et al., 1998).

THE PESTIVIRUS GENOME

The pestivirus genome, in the absence of insertions,is approximately 12.3 Kb in length (Collett et al.,1988a, b; Moormann et al., 1990; Deng and Brock,1992; Ridpath and Bolin, 1995b, 1997). The longopen reading frame (approximately 4000 codons) isbracketed by relatively large 5’ (360–390 bases) and3’ (200–240 bases) untranslated regions (UTR). The5’ terminus does not contain a cap structure (Brock etal., 1992; Deng and Brock, 1993), and no poly(A)tract is present at the 3’ end. All pestivirus genomesterminate at the 3’ end with a short poly(C) tract. Thehighest nucleic acid sequence identity among pes-tiviruses is found in the 5’ UTR (Ridpath and Bolin,


1997). Although sequence identity is high amongpestiviruses in the 5’ UTR, there are two short re-gions that are notable for their variability (Figure3.1A). These are located between nucleotides 208–223 and nucleotides 294–323 (nucleotide positionnumbers based on the sequence of BVDV 1-SD-11).Sequence variations in these regions have been ex-

ploited in PCR-based tests designed to differentiateBVDV genotypes (el-Kholy et al., 1998; Ridpath andBolin, 1998). High conservation of 5’ UTR se-quences is related to formation of tertiary structuresrequired for internal ribosomal entry mediated initia-tion of translation (Deng and Brock, 1993; Pestovaand Hellen, 1999) (Figure 3.1B).

Classification and Molecular Biology 67

Figure 3.1. Conservation and predicted pseudoknots in pestivirus 5’ UTR sequences. Top. The alignment andcomparison of 5’UTR sequences from recognized pestivirus species is shown. Hyperviable regions are boxed.Conserved sequences used in RT-PCR tests to detect pestiviruses are underlined. The first ATG of the open readingframe is in bold type. Bottom. A schematic of predicted tertiary structures in the 5’ UTR is shown. Deng and Brock(1993) domain designations are shown in bold, and Pestova and Hellen (1999) domain designations are shown inparentheses. Note that domain D5 of Deng and Brock is just a portion of the domain IIIe of Pestova and Hellen. Thelocation of hypervariable regions shown in Figure 3.1a are denoted by a thickened line in the schematic.

1Although BVDV 1a-NADL and BVDV 2-890 are the type viruses for genotypes BVDV 1 and BVDV 2, respectively, bothgenomes have inserted sequences. Insertions can cause confusion when indicating genomic location based on nucleotide number.For this reason BVDV 1a-SD-1 is used as the reference for nucleotide position. It was the first noncytopathic BVDV1 sequencedand does not have an insertion. The accession number for BVDV 1a-SD-1 is M96751.


ENTRY OF THE VIRUS INTOCELLS, TRANSLATION, ANDREPLICATIONThe binding and entry of BVDV to susceptible cellshas not been extensively studied. Based on the be-havior of other flaviviruses and pestiviruses, it ishypothisized that binding and entry is a multistepprocess initiated by receptor-mediated endocytosisinvolving cell surface molecules and the viral pro-teins Erns and E2 (Schelp et al., 1995; Xue et al.,1997; Iqbal et al., 2000; Schelp et al., 2000; Iqbaland McCauley, 2002). The genomic RNA is un-coated following endocytosis and serves as themRNA. No subgenomic mRNA molecules havebeen detected. Initiation of translation is mediatedby a cap-independent internal initiation mechanismthat requires an internal ribosome entry site (IRES)located within the 5’ UTR. A secondary structuremodel for the pestivirus 5’ UTR was first proposedby Deng and Brock (1993). Further refinements of aBVDV specific model have been developed (Pooleet al., 1995; Pestova et al., 1998; Pestova andHellen, 1999). Unfortunately a universal terminol-ogy system was not adopted. The different namesused for the same structural domains by different re-search groups can be confusing. The proposed 5’UTR secondary structures shown in Figure 3.1B areidentified by both the domain names given by Dengand Brock and those given by Pestova and Hellen. Itshould be noted that domain D5 of Deng and Brockis just a portion of the IIIe domain of Pestova andHellen. The IIIe domain contains the D5 domainplus the nucleotides between D5 and nucletide 380(numbering based on BVDV 1a-SD-1). Domains C(II) and D (III) contain structures critical for the at-tachment of the initiation complex. A pseudoknotformed just upstream of the start codon is necessaryfor IRES function. The hypervariable regions usedin PCR based tests to differentiate BVDV genotypesare located in domains D2 (IIIb) and D4 (IIId2).

VIRAL PROTEINS

The large open reading frame is translated as apolyprotein. The order of the individual viral pro-teins within the polyprotein is as follows: Npro-C-Erns-E1-E2-p7-NS2/3-NS4A-NS4B-NS5A-NS5B (Figure 3.2A). The polyprotein is processedcotranslationally and posttranslationally by host andviral proteases. The size, function, and characteris-tics of the viral proteins are summarized in Table3.1. Conservation of the polyprotein coding se-quences among pestiviruses is shown in Figure3.2B. The predicted hydrophobicity profile of the

polyprotein is shown in Figure 3.2C. The hydropho-bicity profile of the polyprotein is conserved withinthe pestivirus genus.

Viral structural proteins

The proteins associated with the mature virion areC, Erns, E1, and E2. C is the virion nucleocapsid pro-tein. It is highly basic and relatively conservedamong the pestivirus species. It is postulated that theC terminus of the nucleocapsid protein contains aninternal signal sequence that directs translocation ofstructural glycoproteins to the endoplasmic reticu-lum (Heinz et al., 2000). The Erns protein is glycosy-lated and forms disulfide-linked homodimers (Weil-and et al., 1990). In tissue culture systems Erns maybe found both associated with released virus andfree in the culture medium. This protein has an un-usual ribonuclease activity. Although the function ofthis ribonuclease activity in the viral life cycle is un-known, antibodies that inhibit ribonuclease activityneutralize virus infectivity of classical swine feverviruses (CSFV) (Windisch et al., 1996). The ribonu-clease activity does not require dimer formation orglycosylation (Windisch et al., 1996). The Erns ofCSFV possesses epitopes that induce neutralizingantibodies (hereafter referred to as neutralizing epi-topes) and CSFV Erns subunit vaccines induce neu-tralizing antibodies (Konig et al., 1995). It is notknown whether the Erns of BVDV possesses neutral-izing epitopes that are important in disease control.However, protection afforded by killed BVDV vac-cines does not appear to be dependent upon Erns spe-cific antibodies (Bolin and Ridpath, 1990).

Both the E1 and E2 proteins possess potentialmembrane-spanning domains and are glycosylated.They are predicted to be integral membrane proteinsand interact to form heterodimers (Weiland et al.,1990). The E2 protein is the immunodominant struc-tural protein and possesses neutralizing epitopes(Donis et al., 1988; Bolin and Ridpath, 1989). Mabsproduced against the E2 have been used to differen-tiate pestivirus species (Peters et al., 1986; Wens-voort et al., 1986; Bolin et al., 1988; Hess et al.,1988; Wensvoort et al., 1989a; Corapi et al., 1990a;Edwards and Sands, 1990; Dahle et al., 1991; Deregtet al., 1994; Paton et al., 1994; Ridpath et al., 1994a;Ridpath and Bolin, 1997). The coding region for E2neutralizing epitopes are found in a hypervariable re-gion located in the N-terminal portion of the E2 pro-tein (Paton et al., 1992; Deregt et al., 1998) (refer toFigure 3.2B). E2 antigenic variation is particularlypronounced among BVDV strains and contributes tovaccine failure (Bolin et al., 1988; Bolin and Rid-



path, 1989, 1990; Bolin et al., 1991a; Bolin et al.,1994; Ridpath et al., 1994a; Van Campen et al., 1997;Bolin and Ridpath, 1998; Van Campen et al., 1998;Ridpath et al., 2000; Van Campen et al., 2000).

Viral nonstructural proteins

The first viral protein encoded by the BVDV ORF isthe nonstructural protein, Npro. This protein, whichis unique to the pestivirus genus, is an autoproteasewhose only known function is to cleave itself fromthe polyprotein. The next nonstructural protein, p7,follows the structural protein E2 in the polyprotein(Elbers et al., 1996). It consists of a central chargedregion flanked by hydrophobic termini. The role ofthis cell-associated protein is unknown. It is hypoth-esized that p7 is required for production of infec-tious virus but not for RNA replication (Harada etal., 2000). It is inefficiently cleaved from the E2,leading to two intracellular forms of E2 with differ-ent C termini (E2 and E2-p7) (Elbers et al., 1996).

Neither p7 nor E2-p7 are found associated with in-fectious virus.

Following p7 the next nonstructural protein in thepolyprotein is the serine protease, NS2-3. In BVDVstrains from the cytopathic biotype (see discussionof biotype below), the NS2-3 is cleaved to NS2 andNS3. The serine protease activity of the NS2-3 re-sides in the NS3 portion of the protein. The functionof the NS2 is unknown. It is not required for RNAreplication, and its cleavage from the NS2-3 doesnot affect serine protease activity (Behrens et al.,1998). Sequence analysis reveals regions in the NS2with homology to zinc-finger motifs present in DNAbinding proteins (De Moerlooze et al., 1991).However, a function for the putative DNA bindingactivity of NS2 in viral replication has not beendemonstrated.

Both the uncleaved NS2-3 and the cleaved NS3act as serine proteases (Tautz et al., 1997) that cleavethe remaining nonstructural proteins from the


Figure 3.2. The pestivirus genome. A. Organization of the pestivirus genome is shown. Structural proteins are de-noted by shading. B. Sequence conservation among pestivirus species is shown. These results are a composite ofcomparisons between the four recognized pestivirus species. Sequence similarity is indicated by different levels ofshading. C. A hydrophobicity plot of a pestivirus consensus sequence is shown. A consensus sequence was derivedby comparison of representatives of noncytopathic strains from each of the four recognized pestivirus species.


polyprotein. Purified BVDV NS3 also possessesRNA helicase and RNA-stimulated NTPase activi-ties (Tamura et al., 1993; Warrener and Collett,1995). All three activities (serine protease, RNA he-licase, and RNA-stimulated NTPase) are essential tovirus viability (Grassmann et al., 1999; Gu et al.,2000). Antibodies to the NS2-3 and NS3 do not neu-tralize infectivity. However, the NS2-3 and NS3 (butnot the NS2) are strongly recognized by polyclonalconvalescent sera (Donis et al., 1991). Animals vac-cinated with modified live vaccines also have astrong antibody response to the NS2-3 and/or NS3protein (Bolin and Ridpath, 1989). In contrast, ani-mals vaccinated with inactivated (killed) vaccines

primarily react with structural proteins and not theNS2-3 or NS3 (Bolin and Ridpath, 1990). The dif-ference in recognition of NS2-3 or NS3 may be use-ful in differentiating between immune responses toinactivated vaccines and immune responses to natu-ral infection.

The NS4A and NS4B proteins are similar in size,composition and hydrophobicity to the NS4A andNS4B proteins of other flaviviruses (Lindenbachand Rice, 2001). NS4A acts as a cofactor for theNS2-3 and NS3 serine protease activity. NS4B andNS5A probably are replicase complex components.RNA polymerase activity has been demonstrated forthe NS5B protein (Lai et al., 1999).


Table 3.1. Pestivirus proteins

EstimatedViral Size NeutralizingProtein (K Daltons) Attributes Epitope(s) Function

Npro 020 Nonstructural N AutoproteolysisNot required for RNA replication

C 014 Structural N Forms nucleocapsid of virionConservedHighly basic

Erns 048 Structural Y Envelope-associated glycoprotein7–9 glycosylation sites Ribonuclease activity

E1 025 Structural N Envelope-associated glycoprotein2–3 glycosylation sites Integral membrane protein

E2 053 Structural Y Envelope-associated glycoprotein4–6 glycosylation sites Integral membrane proteinLeast conserved of structural proteins Immunodominant structural protein

p7 007 Nonstructural N Function unknownCentral charged region flanked Required for production of

by hydrophobic termini infectious virus but not RNA replication

NS 2/3 125 Nonstructural N NS2 has a zinc-finger–like domainNS2 054 In cytopathic biotype NS 2/3 cleaved NS2/3 and NS3 contain RNA NS3 080 to NS2 and NS3 helicase and N-terminal serine

Conserved protease domains; cleaves itselfand remaining nonstructural pro-teins from viral polyprotein

Immunodominant nonstructural protein

NS4A 7.2 Nonstructural N Serine protease cofactorHydrophobic

NS4B 38–39 Nonstructural N Replicase componentHydrophobic

NS5A 55–56 Nonstructural N Replicase componentPhosphorylated

NS5B 81–82 Nonstructural N RNA-dependent RNA polymerase


Nomenclature for proteins

The names of the viral proteins have changed overtime and these variations in terminology can be con-fusing as one reads past literature. The names usedabove and in Table 3.1 are those designated in theSeventh Report of the International Committee onTaxonomy of Viruses (van Regenmortel et al.,2000). The first studies of BVDV reported two gly-cosylated structural polypeptides, with molecularweights of 55K and 45K, which were termed VP1and VP2, respectively (Matthaeus, 1979). Later re-views referred to these proteins as E1 and E2(Westaway et al., 1985). When Collett et al. first de-rived the genomic map for BVDV, gene productswere designated by their molecular weights and gly-cosylation status (Collett et al., 1988b). Thus, theviral protein now known as E2, may be referred toas VP1, E1, or gp53 in the literature. Similarly, Erns

may be reported as VP2, E0, E2, or gp48.

TRANSLATION AND PROCESSING OF THEPOLYPROTEIN

As stated above, the first viral protein in the ORF isa nonstructural protein unique to the pestivirusgenus, Npro. The Npro is an autoprotease whose onlyknown function is to cleave itself from the polypro-tein. Cellular signal peptidases are responsible forthe cleavages between C and Erns, E1, and E2, andE2 and p7 (Rumenapf et al., 1993; Stark et al., 1993;Elbers et al., 1996). The mechanism that producesthe cleavage between Erns and E2 is unknown. Asstated above, the cleavage between E2 and p7 maynot be complete. Depending on the biotype of thevirus (see section on biotypes below), the NS2/3protein may be further processed to the NS2 and theNS3 proteins. Both NS2/3 and NS3 have serine pro-tease activity that cleaves the remaining downstreamproteins from the polyprotein. The NS4A viral pro-tein facilitates the protease activity of the NS2/3(and/or NS3) at the cleavage between NS4B andNS5A and between NS5A and NS5B.

VIRAL REPLICATION

Viral proteins, generated by processing the trans-lated polyprotein, participate in viral replication. Ithas been proposed that a secondary structure motifin the 5’ UTR enables the switch of viral RNA froma template for translation to a template for replica-tion (Yu et al., 2000). RNA replication occurs via areplication complex, composed of viral RNA andviral nonstructural proteins, in association with in-tracytoplasmic membranes. The replication ofBVDV RNA is similar to that of other flaviviruses.

After translation, RNA replication begins with thesynthesis of complementary negative strands. Usingthese negative strands as templates, genome-lengthpositive strands are synthesized by a semiconserva-tive mechanism involving replicative intermediatesand replicative forms (Gong et al., 1996; Gong et al.,1998). Because viral proteins are not detected on thesurface of infected cells, it is thought that virionsmature in intracellular vesicles and are released byexocytosis (Heinz et al., 2000). A substantial frac-tion of the infectious virus remains cell associated(Lindenbach and Rice, 2001).

BVDV GENOTYPES ANDPHENOTYPESAs stated above, BVDV strains may belong to oneof two different genotypes, BVDV 1 and BVDV 2.Further, viruses from either genotype may exist asone of two biotypes, cytopathic and noncytopathic.

BVDV GENOTYPES

Genotyping is grouping based on comparison of ge-nomic sequences. There are no hard and fast taxo-nomic rules governing how different two genomeshave to be before they are designated different geno-types. However, for genotypic designations to bemeaningful, some practical considerations should betaken into account. Genotypes should be distinct,discrete, and stable. That is, there should be a cleardefined break between groups (discreet groups asopposed to a continuum), and isolates should not beable to easily switch back and forth between groups.Ideally genotypes would be associated with practi-cal observations such as geographic distribution,antigenic variation, or variations in virulence. Thedifferentiation between the BVDV 1 and the BVDV2 genotypes meets these practical considerations.

The first reported division of BVDV strains intotwo different genotypes was based on comparison ofthe 5’ UTR (Pellerin et al., 1994; Ridpath et al.,1994a). Previous studies of vaccination cross pro-tection and monoclonal antibody binding revealedthat BVDV strains were quite variable (Moennig etal., 1987; Bolin et al., 1988; Corapi et al., 1990a;Bolin et al., 1991a; Donis et al., 1991; Edwards etal., 1991). Although these studies indicated thatthere was considerable variation among BVDVstrains, no standard means of grouping viruses basedon these variations was generated. Meanwhile, hy-bridization analysis and sequence comparison sug-gested that the 5’ UTR was highly conserved com-pared to other portions of the genome (Lewis et al.,1991; Ridpath and Bolin, 1991; Ridpath et al.,



1993). The highly conserved sequences in the 5’UTR were evaluated by several groups as targets forpolymerase chain reaction (PCR) tests designed todetect the wide range of BVDV strains or to differ-entiate BVDV from CSFV or BDV (Hofmann et al.,1994; Vilcek, 1994). Concurrent to this research ahighly virulent form of BVD, termed hemorrhagicsyndrome, was reported in Canada and the UnitedStates (Corapi et al., 1989; Corapi et al., 1990b;Carman, 1995; Carman et al., 1998). Phylogeneticanalysis of the BVDV strains, isolated from animalssuffering from hemorrhagic syndrome, groupedthem separately from the BVDV strains commonlyused, at that time, in vaccine production, diagnostictests, and research (Pellerin et al., 1994; Ridpath etal., 1994a). The newly recognized group of BVDVwas designated BVDV genotype II, and the groupcontaining the strains used in vaccines, detection,and research was termed BVDV genotype I. Thenames of these two genotypes were later modified toBVDV 1 and BVDV 2 in keeping with taxonomicconventions in use with other viruses (Heinz et al.,2000). It was further noted that viruses from theBVDV 2 genotype were also isolated from persist-ently infected calves born to dams that had beenvaccinated with vaccines based on BVDV 1 isolates(Ridpath et al., 1994a). Comparison of monoclonalantibody binding profiles revealed clear differencesbetween BVDV 1 and BVDV 2 strains (Ridpath etal., 1994a). Although initial phylogenetic differenti-ation was based on comparison of 5’ UTR se-quences, comparison of complete genomic se-quences showed that differences between BVDV 1and BVDV 2 strains were found throughout thegenome (Ridpath and Bolin, 1995b).

There are two common misconceptions relating tothe BVDV 2 genotype. One relates to origin of theBVDV 2 genotype and the other relates to the viru-lence of BVDV 2 strains. In the late 1990s there wassome speculation that BVDV 2 strains representednewly emerging viruses that originated in the UnitedStates as a result of use of vaccines by U.S. produc-ers and were then transferred to Europe. AlthoughBVDV 2 strains were first recognized in 1994, ret-rospective characterization of strains collected fromBVDV outbreaks in Ontario that occurred between1981 and 1994 demonstrated that BVDV 2 werepresent in North America at least since the the early1980s. However, the first BVDV 2 strain describedin the literature was isolated in Europe prior to 1979(Wensvoort et al., 1989b). Interestingly, this strainwas isolated from a pig and was referred to as anatypical CSFV. Subsequent comparison of this iso-

late to BVDV 2 strains isolated from North Americaled to the segregation of this virus to the BVDV 2genotype (Paton et al., 1995b). Further, a phyloge-netic survey of BVDV 1 and BVDV 2 strainsshowed similar levels of sequence variation in the 5’UTR (Ridpath et al., 2000). If one assumes that theevolutionary clock ticks at the same rate for bothgenotypes, similar rates of variation in the 5’ UTRsuggests that the two genotypes have been evolvingfor approximately the same time span. This suggeststhat viruses from the BVDV 2 genotype are notnewly emerging viruses, but newly recognized ones.

Initially the BVDV 2 genotype was associatedwith outbreaks of hemorrhagic syndrome (Corapi etal., 1989; Corapi et al., 1990b; Pellerin et al., 1994;Carman et al., 1998), a clinically severe form ofacute BVDV infection. This led to the misconceptionthat all BVDV 2 strains cause severe acute disease.In an initial survey of BVDV 2 strains, only 32 out of76 strains were associated with clinically severe dis-ease (Ridpath et al., 1994a). Since then, animal stud-ies have shown that there is a continuum of virulenceseen in acute infections with type 2 BVDV (Flores etal., 2000; Fulton et al., 2000a; Hamers et al., 2000;Ridpath et al., 2000; Tajima et al., 2001; Evermannand Ridpath, 2002; Kelling et al., 2002; Liebler-Tenorio et al., 2002; Liebler-Tenorio et al., 2003).Thus, it appears that only a minority of BVDV 2strains cause severe acute clinical disease. The ma-jority of BVDV 2 strains present in North Americaappear to be no more virulent than BVDV 1 strains.

Prevalence of BVDV genotypes

Both BVDV genotypes have been reported in NorthAmerica (Pellerin et al., 1994; Ridpath and Neill,1998), Europe (Becher et al., 1995; Edwards andPaton, 1995; Paton, 1995a, b; Sandvik et al., 1997;van Rijn et al., 1997; Giangaspero et al., 2001;Luzzago et al., 2001; Pratelli et al., 2001; Couvreuret al., 2002; Vilcek et al., 2002), and South America(Canal et al., 1998; Jones et al., 2001), although thereported prevalence of BVDV 2 is higher in NorthAmerica than in Europe or South America. Reportsof the prevalence of the BVDV 2 genotype in com-mercial cattle herds in North America range from24–47% (Carman, 1995; Bolin and Ridpath, 1998;Carman et al., 1998; Fulton et al., 1998; Fulton etal., 2000b; Evermann and Ridpath, 2002).

Similarities between viruses from the BVDV 1and the BVDV 2 genotype

Viruses from both genotypes may exist as one oftwo biotypes, cytopathic and noncytopathic (see



below). Noncytopathic viruses from either genotypecan cross the placenta and establish persistent infec-tions in the fetus.

Aside from severe acute BVDV, it is difficult todistinguish BVDV 1 infections from BVDV 2 infec-tions based on clinical signs. There is considerableheterogeneity among strains from both genotypes.Although virulence is strain-dependent, clinical pre-sentation may also be effected by immune status, re-productive status, stress, and the presence of second-ary pathogens.

Heterogeneity, among strains within each geno-type, also affects antigenicity. Although strains fromthe BVDV 1 and BVDV 2 genotypes may be distin-guished by monoclonal antibody binding, there aresmaller, but still significant, differences amongstrains from the same genotype (Ridpath et al.,1994a; Ridpath et al., 2000).

Differences between viruses from the BVDV 1and the BVDV 2 genotype

To date only viruses from the BVDV 2 genotype areassociated with severe acute BVDV (Corapi et al.,1989; Corapi et al., 1990b; Bolin and Ridpath, 1992;Pellerin et al., 1994; Ridpath et al., 1994a; Carmanet al., 1998; Liebler-Tenorio et al., 2002). BVDV 1and BVDV 2 strains are antigenically distinct, asdemonstrated by serum neutralization using poly-clonal sera and monoclonal antibody binding(Ridpath et al., 2000). The practical significance ofantigenic differences is indicated by the birth ofBVDV 2 persistently infected animals to dams thathad been vaccinated against BVDV 1 strains (Bolinet al., 1991a; Ridpath et al., 1994a). Although mod-ified live BVDV 1 vaccines may induce antibodiesagainst BVDV 2 strains, the titers average one logless than titers against heterologous BVDV 1 strains(Cortese et al., 1998).

Subgenotypes of BVDV 1 and BVDV 2

Subgenotypes of both BVDV 1 and BVDV 2 havebeen described, although the biological significanceof these subgroupings is a matter of debate. BVDV1 strains from North America can be segregated totwo subgenotypes, BVDV 1a and BVDV 1b (Pelle-rin et al., 1994). These two subgenotypes can be dis-tinguished by monoclonal antibody binding (Bolinand Ridpath, 1998) and RT-PCR (Ridpath andBolin, 1998). Epidemiological surveys suggest thatBVDV 1b strains may predominate in respiratorycases (Fulton et al., 2002), and BVDV 1a strainsmay predominate in fetal infections occurring late ingestation (>100 days gestation) (Evermann and Rid-

path, 2002). While the reported incidence of BVDV2 strains is lower, European BVDV 1 strains appearto be more variable than North American BVDV 1strains. The European BVDV 1 strains have beenseparated into at least 11 subgroups (Vilcek et al.,2001a). The biological significance of these 11groups has yet to be examined.

Similarly, BVDV 2 strains from North and SouthAmerican have been segregated into two subgroups,BVDV 2a and BVDV 2b (Flores et al., 2002). InNorth America BVDV 2b strains are relatively rare;in South America the prevalence of BVDV 2astrains and BVDV 2b strains are similar.

BVDV BIOTYPES

As stated above, both BVDV 1 strains and BVDV 2strains may exist as one of two biotypes, cytopathicand noncytopathic. The division into biotypes isbased on the activity of the strain when propagatedin cultured epithelial cells (Lee and Gillespie, 1957;Gillespie et al., 1960). Recall that both noncyto-pathic and cytopathic strains code for the nonstruc-tural protein NS2-3. Among cytopathic strains theNS2-3 is processed to an NS2 and an NS3 protein.Processing of the NS2-3 protein in cytopathicstrains may occur by several different strategies de-pending on the viral strain (Meyers et al., 1991a, b;Meyers et al., 1992; Qi et al., 1992; Ridpath et al.,1994b; Kupfermann et al., 1996; Tautz et al., 1996;Becher et al., 1998; Mendez et al., 1998; Meyers etal., 1998; Qi et al., 1998; Becher et al., 1999; Barothet al., 2000; Kummerer and Meyers, 2000; Ridpathand Neill, 2000; Vilcek et al., 2000; Neill andRidpath, 2001; Becher et al., 2002; Ridpath andBolin, 1995a). Most commonly, the generation ofthe NS3 is associated with insertion of sequencesinto the viral genome. Generally the amino acid se-quence flanking the carboxy-terminus of the inser-tion corresponds to either amino acid position 1535or position 1589 (numbering based on the noncyto-pathic BVDV 1a strain SD-1, acc. # M96751, seefootnote a). These two positions are referred to aspositions A and B, respectively (Meyers et al.,1998). Position B corresponds to the cleavage sitethat frees the N-terminus of the NS3 from the NS2-3 precursor. It is thought that insertions at the B po-sition result in the processing of the NS2-3, either byintroducing a new cleavage site at carboxy terminusof the insertion or by introducing sequences with au-tocatalytic activity that act at the carboxy terminusof the insertion. The end result is that cleavage oc-curs at position 1589, making Gly1589 the firstamino acid of NS3. Insertions at position A, al-



though upstream of the cleavage site, also producecleavage at amino acid position 1589. It has beenproposed that insertions upstream of the cleavagesite induce conformational changes that allow cleav-age via a cryptic mechanism at Gly1589 (Meyers andThiel, 1996). The most frequently described inser-tions into cytopathic BVDV 1 strains have been atposition B and consist of duplicated viral genomicsequence and/or ubiquitin coding sequences. Inser-tions in cytopathic BVDV 2 strains are more fre-quently reported in the vicinity of position A andconsist of portions of a gene coding for a DnaJ-likeprotein (Ridpath and Neill, 2000; Neill and Ridpath,2001). Included among the possible explanations forthese differences in relative frequency of positionand source of inserted sequences are (i) differencesin recombination frequencies due to sequence varia-tion; (ii) differences in the stability of recombina-tions; and (iii) differences in the relative amounts ofmRNA coding for ubiquitin and DnaJ-like proteinsin BVDV 1 and BVDV 2 infected cells.

Practical significance of biotypes

Only noncytopathic BVDV have been reported toestablish persistent infections. This has led to somespeculation that cytopathic BVDV are not able tocross the placenta and thus do not infect the fetus.However, the detection of cytopathic BVDV strainsin fetal calf serum and fetal sero conversion in uteroafter exposure to cytopathic BVDV call into ques-tion this assumption (Brownlie et al., 1989; Bolin etal., 1991b; Bolin and Ridpath, 1998). Because cyto-pathic BVDV do not establish persistent infectionsin vivo, they are frequently the biotype of choice formodified live vaccines. Superinfection of an animal,persistently infected with a noncytopathic BVDV,with a cytopathic BVDV can lead to the develop-ment of a highly fatal form of BVDV called mucosaldisease (Brownlie et al., 1984; Bolin et al., 1985).Mucosal disease is a comparatively rare form ofBVDV and does not represent a major source of lossfor producers (Houe, 1995, 1999). The majority oflosses incurred by producers due to BVDV infec-tions are associated with acute infections with non-cytopathic BVDV (Houe, 1995; Fulton et al., 1998;Houe, 1999; Fulton et al., 2000a, b; Evermann andRidpath, 2002; Fulton et al., 2002).

Noncytopathic BVDV strains predominate overcytopathic BVDV in nature. In the field, cytopathicBVDV strains are typically isolated from mucosaldisease outbreaks or post-vaccinal disease out-breaks. Cytopathology in cultured epithelial cellsdoes not correlate with virulence in acute infections

in vivo. All highly virulent viruses studied to datehave been noncytopathic (Corapi et al., 1990b;Bolin and Ridpath, 1992; Carman et al., 1998; Elliset al., 1998; Ridpath et al., 2000; Liebler-Tenorio etal., 2002). In short, from a management standpoint,determining the genotype of a strain is generallymore important than determining the biotype.

Importance of variation among BVDV

Heterogeneity is the defining characteristic ofBVDV viruses. Variation between BVDV genotype1 and BVDV genotype 2 has practical significance.In particular, variation in antigenicity impacts on de-tection and management programs. The impact ofvariation within each genotype is less well defined.Understanding and compensating for variationamong BVDV is essential to developing a success-ful control program.

REFERENCESBaroth M, Orlich M, Thiel HJ, Becher P: 2000,

Insertion of cellular NEDD8 coding sequences in apestivirus. Virology 278:456–466.

Baule C, van Vuuren M, Lowings JP, Belak S: 1997,Genetic heterogeneity of bovine viral diarrhoeaviruses isolated in Southern Africa. Virus Res52:205–220.

Becher P, Konig M, Paton DJ, Thiel HJ: 1995, Furthercharacterization of border disease virus isolates:Evidence for the presence of more than threespecies within the genus pestivirus. Virology209:200–206.

Becher P, Orlich M, Konig M, Thiel HJ: 1999,Nonhomologous RNA recombination in bovineviral diarrhea virus: Molecular characterization of avariety of subgenomic RNAs isolated during anoutbreak of fatal mucosal disease. J Virol73:5646–5653.

Becher P, Orlich M, Shannon AD, et al.: 1997,Phylogenetic analysis of pestiviruses from domesticand wild ruminants. J Gen Virol 78:1357–1366.

Becher P, Orlich M, Thiel HJ: 1998, Ribosomal S27acoding sequences upstream of ubiquitin coding se-quences in the genome of a pestivirus. J Virol72:8697–8704.

Becher P, Thiel HJ, Collins M, et al.: 2002, Cellularsequences in pestivirus genomes encoding gamma-aminobutyric acid (A) receptor-associated proteinand Golgi-associated ATPase enhancer of 16 kilo-daltons. J Virol 76:13069–13076.

Beer M, Wolf G, Kaaden OR: 2002, Phylogeneticanalysis of the 5’-untranslated region of germanBVDV type II isolates. J Vet Med B Infect Dis VetPublic Health 49:43–47.



Behrens SE, Grassmann CW, Thiel HJ, et al.: 1998,Characterization of an autonomous subgenomicpestivirus RNA replicon. J Virol 72:2364–2372.

Bolin S, Moennig V, Kelso Gourley NE, Ridpath J:1988, Monoclonal antibodies with neutralizing ac-tivity segregate isolates of bovine viral diarrheavirus into groups. Brief report. Arch Virol99:117–123.

Bolin SR, Littledike ET, Ridpath JF: 1991a, Serologicdetection and practical consequences of antigenicdiversity among bovine viral diarrhea viruses in avaccinated herd. Am J Vet Res 52:1033–1037.

Bolin SR, Matthews PJ, Ridpath JF: 1991b, Methodsfor detection and frequency of contamination offetal calf serum with bovine viral diarrhea virus andantibodies against bovine viral diarrhea virus. J VetDiagn Invest 3:199–203.

Bolin SR, McClurkin AW, Cutlip RC, Coria MF:1985, Severe clinical disease induced in cattle per-sistently infected with noncytopathic bovine viraldiarrhea virus by superinfection with cytopathicbovine viral diarrhea virus. Am J Vet Res46:573–576.

Bolin SR, Ridpath JF: 1989, Specificity of neutraliz-ing and precipitating antibodies induced in healthycalves by monovalent modified-live bovine viral di-arrhea virus vaccines. Am J Vet Res 50:817–821.

Bolin SR, Ridpath JF: 1990, Range of viral neutraliz-ing activity and molecular specificity of antibodiesinduced in cattle by inactivated bovine viral diar-rhea virus vaccines. Am J Vet Res 51:703–707.

Bolin SR, Ridpath JF: 1992, Differences in virulencebetween two noncytopathic bovine viral diarrheaviruses in calves. Am J Vet Res 53:2157–2163.

Bolin SR, Ridpath JF: 1998, Prevalence of bovineviral diarrhea virus genotypes and antibody againstthose viral genotypes in fetal bovine serum. J VetDiagn Invest 10:135–139.

Bolin SR, Ridpath JF, Black J, et al.: 1994, Survey ofcell lines in the American Type Culture Collectionfor bovine viral diarrhea virus. J Virol Methods48:211–221.

Brock KV, Deng R, Riblet SM: 1992, Nucleotide se-quencing of 5’ and 3’ termini of bovine viral diar-rhea virus by RNA ligation and PCR. J VirolMethods 38:39–46.


Brownlie J, Clarke MC, Howard CJ: 1989,Experimental infection of cattle in early pregnancywith a cytopathic strain of bovine virus diarrhoeavirus. Res Vet Sci 46:307–311.

Canal CW, Strasser M, Hertig C, et al.: 1998,Detection of antibodies to bovine viral diarrhoea

virus (BVDV) and characterization of genomes ofBVDV from Brazil. Vet Microbiol 63:85–97.

Carman S: 1995, Typing of Ontario BVD VirusIsolates from 1981 to 1994. Ontario Ministry ofAgriculture, Food, and Rural Affairs, Guelph.


Collett MS, Anderson DK, Retzel E: 1988a,Comparisons of the pestivirus bovine viral diar-rhoea virus with members of the flaviviridae. J GenVirol 69:2637–2643.

Collett MS, Larson R, Belzer SK, Retzel E: 1988b,Proteins encoded by bovine viral diarrhea virus: thegenomic organization of a pestivirus. Virology165:200–208.

Corapi WV, Donis RO, Dubovi EJ: 1990a, Char-acterization of a panel of monoclonal antibodiesand their use in the study of the antigenic diversityof bovine viral diarrhea virus. Am J Vet Res51:1388–1394.

Corapi WV, Elliott RD, French TW, et al.: 1990b,Thrombocytopenia and hemorrhages in veal calvesinfected with bovine viral diarrhea virus. J Am VetMed Assoc 196:590–596.

Corapi WV, French TW, Dubovi EJ: 1989, Severethrombocytopenia in young calves experimentallyinfected with noncytopathic bovine viral diarrheavirus. J Virol 63:3934–3943.

Cortese VS, Whittaker R, Ellis J, et al.: 1998,Specificity and duration of neutralizing antibodiesinduced in healthy cattle after administration of amodified-live virus vaccine against bovine viral di-arrhea. Am J Vet Res 59:848–850.

Couvreur B, Letellier C, Collard A, et al.: 2002,Genetic and antigenic variability in bovine viral di-arrhea virus (BVDV) isolates from Belgium. VirusRes 85:17–28.

Dahle J, Liess B, Moennig V, Coulibaly CO: 1991,Use of monoclonal antibodies for the differentialdiagnosis of pestivirus infections in swine. TierarztlPrax 19:151–155.

De Moerlooze L, Renard A, Lecomte C, Martial JA:1991, A “zinc finger-like” domain in the 54KDAprotein of several pestiviruses. Arch Virol Suppl3:41–46.

Deng R, Brock KV: 1992, Molecular cloning and nu-cleotide sequence of a pestivirus genome, noncyto-pathic bovine viral diarrhea virus strain SD-1.Virology 191:867–869.

Deng R, Brock KV: 1993, 5’ and 3’ untranslated re-gions of pestivirus genome: primary and secondarystructure analyses. Nucleic Acids Res 21:1949–1957.

Deregt D, Bolin SR, Heckert RA, Loewen KG: 1994,Monoclonal antibodies to bovine viral diarrhea



virus: Cross- reactivities to field isolates and hogcholera virus strains. Can J Vet Res 58:71–74.

Deregt D, Bolin SR, van den Hurk J, et al.: 1998,Mapping of a type 1-specific and a type-commonepitope on the E2 (gp53) protein of bovine viraldiarrhea virus with neutralization escape mutants.Virus Res 53:81–90.

Donis RO, Corapi W, Dubovi EJ: 1988, Neutralizingmonoclonal antibodies to bovine viral diarrhoeavirus bind to the 56K to 58K glycoprotein. J GenVirol 69:77–86.

Donis RO, Corapi WV, Dubovi EJ: 1991, Bovine viraldiarrhea virus proteins and their antigenic analyses.Arch Virol Suppl 3:29–40.

Edwards S, Moennig V, Wensvoort G: 1991, The de-velopment of an international reference panel ofmonoclonal antibodies for the differentiation of hogcholera virus from other pestiviruses. Vet Microbiol29:101–108.

Edwards S, Paton D: 1995, Antigenic differencesamong pestiviruses. Vet Clin North Am Food AnimPract 11:563–577.

Edwards S, Sands JJ: 1990, Antigenic comparisons ofhog cholera virus isolates from Europe, Americaand Asia using monoclonal antibodies. DTW DtschTierarztl Wochenschr 97:79–81.

Elbers K, Tautz N, Becher P, et al.: 1996, Processingin the pestivirus E2-NS2 region: Identification ofproteins p7 and E2p7. J Virol 70:4131–4135.

el-Kholy AA, Bolin SR, Ridpath JF, et al.: 1998, Useof polymerase chain reaction to simultaneously de-tect and type bovine viral diarrhoea viruses isolatedfrom clinical specimens. Rev Sci Tech 17:733–742.

Ellis JA, West KH, Cortese VS, et al.: 1998, Lesionsand distribution of viral antigen following an exper-imental infection of young seronegative calves withvirulent bovine virus diarrhea virus-type II. Can JVet Res 62:161–169.


Falcone E, Cordioli P, Sala G, et al.: 2001, Geno-typing of bovine viral diarrhoea viruses isolatedfrom cattle in northern Italy. Vet Res Commun25:161–167.

Flores EF, Gil LH, Botton SA, et al.: 2000, Clinical,pathological and antigenic aspects of bovine viraldiarrhea virus (BVDV) type 2 isolates identified inBrazil. Vet Microbiol 77:175–183.

Flores EF, Ridpath JF, Weiblen R, et al.: 2002,Phylogenetic analysis of Brazilian bovine viral di-arrhea virus type 2 (BVDV-2) isolates: Evidence fora subgenotype within BVDV-2. Virus Res87:51–60.

Fulton RW, Burge LJ, Saliki JT, et al.: 1998,Prevalence of bovine viral diarrhea virus biotypesand genotypes in diagnostic laboratory accessions:Clinical and necropsy samples from cattle. In: Proc41st Annu Meet Am Assoc Vet Lab Diagn p. 27.

Fulton RW, Purdy CW, Confer AW, et al.: 2000a,Bovine viral diarrhea viral infections in feedercalves with respiratory disease: Interactions withPasteurella spp., parainfluenza-3 virus, and bovinerespiratory syncytial virus. Can J Vet Res64:151–159.

Fulton RW, Ridpath JF, Saliki JT, et al.: 2002, Bovineviral diarrhea virus (BVDV). 1b: PredominantBVDV subtype in calves with respiratory disease.Can J Vet Res 66:181–190.

Fulton RW, Saliki JT, Confer AW, et al.: 2000b,Bovine viral diarrhea virus cytopathic and non-cytopathic biotypes and type 1 and 2 genotypes indiagnostic laboratory accessions: Clinical andnecropsy samples from cattle. J Vet Diagn Invest12:33–38.

Giangaspero M, Harasawa R, Verhulst A: 1997,Genotypic analysis of the 5’-untranslated region ofa pestivirus strain isolated from human leucocytes.Microbiol Immunol 41:829–834.

Giangaspero M, Harasawa R, Zecconi A, Luzzago C:2001, Genotypic characteristics of bovine viral di-arrhea virus 2 strains isolated in northern Italy. JVet Med Sci 63:1045–1049.

Gillespie J, Baker J, McEntee K: 1960, A cytopatho-genic strain of virus diarrhea virus. Cornell Vet50:73–79.

Gong Y, Shannon A, Westaway EG, Gowans EJ: 1998,The replicative intermediate molecule of bovineviral diarrhoea virus contains multiple nascentstrands. Arch Virol 143:399–404.

Gong Y, Trowbridge R, Macnaughton TB, et al.: 1996,Characterization of RNA synthesis during a one-step growth curve and of the replication mechanismof bovine viral diarrhoea virus. J Gen Virol77:2729–2736.

Grassmann CW, Isken O, Behrens SE: 1999, Assign-ment of the multifunctional NS3 protein of bovineviral diarrhea virus during RNA replication: An invivo and in vitro study. J Virol 73:9196–9205.

Gu B, Liu C, Lin-Goerke J, et al.: 2000, The RNA he-licase and nucleotide triphosphatase activities of thebovine viral diarrhea virus NS3 protein are essen-tial for viral replication. J Virol 74:1794–1800.

Hafez SM, Liess B: 1972, Studies on bovine viraldiarrhea-mucosal disease virus. II. Stability andsome physico-chemical properties. Acta Virol16:399–408.

Hamers C, Couvreur B, Dehan P, et al.: 2000,Differences in experimental virulence of bovine



viral diarrhoea viral strains isolated from haemor-rhagic syndromes. Vet J 160:250–258.

Harada T, Tautz N, Thiel HJ: 2000, E2-p7 region ofthe bovine viral diarrhea virus polyprotein:Processing and functional studies. J Virol74:9498–9506.

Harasawa R: 1996, Phylogenetic analysis of pestivirusbased on the 5’-untranslated region. Acta Virol40:49–54.

Harasawa R, Giangaspero M: 1998, A novel methodfor pestivirus genotyping based on palindromic nu-cleotide substitutions in the 5’-untranslated region.J Virol Methods 70:225–230.

Harasawa R, Giangaspero M, Ibata G, Paton DJ:2000, Giraffe strain of pestivirus: Its taxonomic sta-tus based on the 5’-untranslated region. MicrobiolImmunol 44:915–921.

Harasawa R, Mizusawa H: 1995, Demonstration andgenotyping of pestivirus RNA from mammaliancell lines. Microbiol Immunol 39:979–985.


Heinz FX, Collett MS, Purcell RH, et al.: 2000,Genus pestivirus. In: Virus Taxonomy. Eds. vanRegenmortel MHV, Fauquet CM, Bishop DHL, etal., pp. 867–872. Academic Press, New York.

Hess RG, Coulibaly CO, Greiser-Wilke I, et al.: 1988,Identification of hog cholera viral isolates by use ofmonoclonal antibodies to pestiviruses. VetMicrobiol 16:315–321.

Hofmann MA, Brechtbuhl K, Stauber N: 1994, Rapidcharacterization of new pestivirus strains by directsequencing of PCR-amplified cDNA from the 5’noncoding region. Arch Virol 139:217–229.

Horzinek M, Maess J, Laufs R: 1971, Studies on thesubstructure of togaviruses. II. Analysis of equinearteritis, rubella, bovine viral diarrhea, and hogcholera viruses. Arch Gesamte Virusforsch33:306–318.

Houe H: 1995, Epidemiology of bovine viral diarrheavirus. Vet Clin North Am Food Anim Pract11:521–547.


Iqbal M, Flick-Smith H, McCauley JW: 2000,Interactions of bovine viral diarrhoea virus glyco-protein E(rns) with cell surface glycosaminogly-cans. J Gen Virol 81(Pt2):451–459.

Iqbal M, McCauley JW: 2002, Identification of theglycosaminoglycan-binding site on the glycoproteinE(rns) of bovine viral diarrhoea virus by site-directed mutagenesis. J Gen Virol 83:2153–2159.

Jones LR, Zandomeni R, Weber EL: 2001, Genetictyping of bovine viral diarrhea virus isolates fromArgentina. Vet Microbiol 81:367–375.

Kelling CL, Steffen DJ, Topliff CL, et al.: 2002,Comparative virulence of isolates of bovine viraldiarrhea virus type II in experimentally inoculatedsix- to nine-month-old calves. Am J Vet Res63:1379–1384.

Konig M, Lengsfeld T, Pauly T, et al.: 1995, Classicalswine fever virus: Independent induction of protec-tive immunity by two structural glycoproteins. JVirol 69:6479–6486.

Kosmidou A, Ahl R, Thiel HJ, Weiland E: 1995,Differentiation of classical swine fever virus(CSFV) strains using monoclonal antibodies against structural glycoproteins. Vet Microbiol47:111–118.

Kummerer BM, Meyers G: 2000, Correlation betweenpoint mutations in NS2 and the viability and cy-topathogenicity of Bovine viral diarrhea virus strainOregon analyzed with an infectious cDNA clone. JVirol 74:390–400.

Kupfermann H, Thiel HJ, Dubovi EJ, Meyers G:1996, Bovine viral diarrhea virus: Characterizationof a cytopathogenic defective interfering particlewith two internal deletions. J Virol 70:8175–8181.

Lai VC, Kao CC, Ferrari E, et al.: 1999, Mutationalanalysis of bovine viral diarrhea virus RNA-dependent RNA polymerase. J Virol 73:10129–10136.

Lee K, Gillespie J: 1957, Propagation of virus diar-rhea virus of cattle in tissue culture. Am J Vet Res18:952–955.

Letellier C, Kerkhofs P, Wellemans G, VanopdenboschE: 1999, Detection and genotyping of bovine diar-rhea virus by reverse transcription-polymerasechain amplification of the 5’ untranslated region.Vet Microbiol 64:155–167.

Lewis TL, Ridpath JF, Bolin SR, Berry ES: 1991,Detection of BVD viruses using synthetic oligonu-cleotides. Arch Virol 117:269–278.

Liebler-Tenorio EM, Ridpath JF, Neill JD: 2002,Distribution of viral antigen and development of le-sions after experimental infection with highly viru-lent bovine viral diarrhea virus type 2 in calves. AmJ Vet Res 63:1575–1584.

Liebler-Tenorio EM, Ridpath JF, Neill JD: 2003,Distribution of viral antigen and development of le-sions after experimental infection of calves with aBVDV 2 strain of low virulence. J Vet Diagn Invest15:221–232.

Liess B: 1990, Bovine viral diarrhea virus. In: VirusInfections of Ruminants. Eds. Dinter Z, Morein B,pp. 247–266. Elsevier Science Publishers, Uppsala,Sweden.



Lindenbach B, Rice C: 2001, Flaviviridae: Theviruses and their replication. In: Field’s Virology,4th ed. Eds. Knipe DM, Howley PM, Griffin DE, etal., pp. 991–1042. Lippincott Williams and Wilkins,Philadelphia.

Luzzago C, Bandi C, Bronzo V, et al.: 2001, Distri-bution pattern of bovine viral diarrhoea virus strainsin intensive cattle herds in Italy. Vet Microbiol83:265–274.

Matthaeus W: 1979, Detection of three polypeptidesin preparations of bovine viral diarrhoea virus. ArchVirol 59:299–305.

Mendez E, Ruggli N, Collett MS, Rice CM: 1998,Infectious bovine viral diarrhea virus (strainNADL) RNA from stable cDNA clones: A cellularinsert determines NS3 production and viral cy-topathogenicity. J Virol 72:4737–4745.

Meyers G, Rumenapf T, Tautz N, et al.: 1991a,Insertion of cellular sequences in the genome ofbovine viral diarrhea virus. Arch Virol Suppl3:133–142.

Meyers G, Stoll D, Gunn M: 1998, Insertion of asequence encoding light chain 3 of microtubule-associated proteins 1A and 1B in a pestivirusgenome: Connection with virus cytopathogenicityand induction of lethal disease in cattle. J Virol72:4139–4148.

Meyers G, Tautz N, Dubovi EJ, Thiel HJ: 1991b, Viralcytopathogenicity correlated with integration ofubiquitin-coding sequences. Virology 180:602–616.

Meyers G, Tautz N, Stark R, et al.: 1992, Rearrange-ment of viral sequences in cytopathogenic pesti-viruses. Virology 191:368–386.

Meyers G, Thiel HJ: 1996, Molecular characterizationof pestiviruses. Adv Virus Res 47:53–118.

Miekka SI, Busby TF, Reid B, et al.: 1998, Newmethods for inactivation of lipid-enveloped andnon-enveloped viruses. Haemophilia 4:402–408.

Moennig V, Bolin SR, Coulibaly CO, et al.: 1987,Studies of the antigen structure of pestivirusesusing monoclonal antibodies. DTW Dtsch TierarztlWochenschr 94:572–576.

Moormann RJ, Warmerdam PA, van der Meer B, etal.: 1990, Molecular cloning and nucleotide se-quence of hog cholera virus strain Brescia andmapping of the genomic region encoding envelopeprotein E1. Virology 177:184–198.

Muller T, Burow J, Conraths FJ, et al.: 1997, Qualitymanagement in the serodiagnosis of Europeanswine fever: Results of comparative tests. DTWDtsch Tierarztl Wochenschr 104:87–91.

Neill JD, Ridpath JF: 2001, Recombination with acellular mRNA encoding a novel DnaJ protein re-sults in biotype conversion in genotype 2 bovineviral diarrhea viruses. Virus Res 79:59–69.

Paton DJ: 1995, Pestivirus diversity. J Comp Pathol112:215–236.

Paton DJ, Carlsson U, Lowings JP, et al.: 1995a,Identification of herd-specific bovine viral diar-rhoea virus isolates from infected cattle and sheep.Vet Microbiol 43:283–294.

Paton DJ, Lowings JP, Barrett AD: 1992, Epitopemapping of the gp53 envelope protein of bovineviral diarrhea virus [published erratum appears inVirology 1992 Dec;191(2):1013]. Virology190:763–772.

Paton DJ, Sands JJ, Edwards S: 1994, Border diseasevirus: Delineation by monoclonal antibodies. ArchVirol 135:241–252.

Paton DJ, Sands JJ, Lowings JP, et al.: 1995b, A pro-posed division of the pestivirus genus using mono-clonal antibodies, supported by cross-neutralisationassays and genetic sequencing. Vet Res 26:92–109.

Pellerin C, van den Hurk J, Lecomte J, Tussen P:1994, Identification of a new group of bovine viraldiarrhea virus strains associated with severeoutbreaks and high mortalities. Virology203:260–268.

Pestova TV, Hellen CU: 1999, Internal initiation oftranslation of bovine viral diarrhea virus RNA.Virology 258:249–256.

Pestova TV, Shatsky IN, Fletcher SP, et al.: 1998, Aprokaryotic-like mode of cytoplasmic eukaryoticribosome binding to the initiation codon duringinternal translation initiation of hepatitis C andclassical swine fever virus RNAs. Genes Dev12:67–83.

Peters W, Greiser-Wilke I, Moennig V, Liess B: 1986,Preliminary serological characterization of bovineviral diarrhoea virus strains using monoclonal anti-bodies. Vet Microbiol 12:195–200.

Poole TL, Wang C, Popp RA, et al.: 1995, Pestivirustranslation initiation occurs by internal ribosomeentry. Virology 206:750–754.

Pratelli A, Martella V, Cirone F, et al.: 2001, Genomiccharacterization of pestiviruses isolated from lambsand kids in southern Italy. J Virol Methods94:81–85.

Preuss T, Kamstrup S, Kyvsgaard N, et al.: 1997,Comparison of two different methods for inactiva-tion of viruses in serum. Clin Diagn Lab Immunol4:504–508.

Qi F, Ridpath JF, Berry ES: 1998, Insertion of abovine SMT3B gene in NS4B and duplication ofNS3 in a bovine viral diarrhea virus genome corre-late with the cytopathogenicity of the virus. VirusRes 57:1–9.

Qi F, Ridpath JF, Lewis T, et al.: 1992, Analysis of thebovine viral diarrhea virus genome for possible cel-lular insertions. Virology 189:285–292.



Ridpath JF: 2003, BVDV genotypes and biotypes:Practical implications for diagnosis and control.Biologicals 31(2):127–132.

Ridpath JF, Bolin SR: 1991, Hybridization analysis ofgenomic variability among isolates of bovine viraldiarrhoea virus using cDNA probes. Mol CellProbes 5:291–298.

Ridpath JF, Bolin SR: 1995a, Delayed onset postvac-cinal mucosal disease as a result of genetic recom-bination between genotype 1 and genotype 2BVDV. Virology 212:259–262.

Ridpath JF, Bolin SR: 1995b, The genomic sequenceof a virulent bovine viral diarrhea virus (BVDV)from the type 2 genotype: detection of a large ge-nomic insertion in a noncytopathic BVDV. Virology212:39–46.

Ridpath JF, Bolin SR: 1997, Comparison of the com-plete genomic sequence of the border disease virus,BD31, to other pestiviruses. Virus Res 50:237–243.


Ridpath JF, Bolin SR, Dubovi EJ: 1994a, Segregationof bovine viral diarrhea virus into genotypes.Virology 205:66–74.

Ridpath JF, Bolin SR, Katz J: 1993, Comparison ofnucleic acid hybridization and nucleic acid amplifi-cation using conserved sequences from the 5’ non-coding region for detection of bovine viral diarrheavirus. J Clin Microbiol 31:986–989.

Ridpath JF, Neill JD: 1998, Characterization of viru-lent and avirulent BVDV from the type 2 genotype.In: Proc 17th Annu Meet Am Soc Virol, p. 162.

Ridpath JF, Neill JD: 2000, Detection and characteri-zation of genetic recombination in cytopathic type2 bovine viral diarrhea viruses. J Virol74:8771–8774.


Ridpath JF, Qi F, Bolin SR, Berry ES: 1994b, Naturalrecombination in bovine viral diarrhea viruses. ArchVirol Suppl 9:239–244.

Rumenapf T, Unger G, Strauss JH, Thiel HJ: 1993,Processing of the envelope glycoproteins of pes-tiviruses. J Virol 67:3288–3294.

Sakoda Y, Ozawa S, Damrongwatanapokin S, et al.:1999, Genetic heterogeneity of porcine and rumi-nant pestiviruses mainly isolated in Japan. VetMicrobiol 65:75–86.

Sandvik T, Paton DJ, Lowings PJ: 1997, Detectionand identification of ruminant and porcine pes-tiviruses by nested amplification of 5’ untranslatedcDNA regions. J Virol Methods 64:43–56.

Schelp C, Greiser-Wilke I, Moennig V: 2000, Anactin-binding protein is involved in pestivirus entryinto bovine cells. Virus Res 68:1–5.

Schelp C, Greiser-Wilke I, Wolf G, et al.: 1995,Identification of cell membrane proteins linked tosusceptibility to bovine viral diarrhoea virus infec-tion. Arch Virol 140:1997–2009.

Shimazaki T, Sekiguchi H, Nakamura S, et al.: 1998,Segregation of bovine viral diarrhea virus isolatedin Japan into genotypes. J Vet Med Sci 60:579–583.

Stark R, Meyers G, Rumenapf T, Thiel HJ: 1993,Processing of pestivirus polyprotein: Cleavage sitebetween autoprotease and nucleocapsid protein ofclassical swine fever virus. J Virol 67:7088–7095.

Sullivan DG, Chang GJ, Akkina RK: 1997, Geneticcharacterization of ruminant pestiviruses: Sequenceanalysis of viral genotypes isolated from sheep.Virus Res 47:19–29.


Tamura JK, Warrener P, Collett MS: 1993, RNA-stimulated NTPase activity associated with the p80protein of the pestivirus bovine viral diarrhea virus.Virology 193:1–10.

Tautz N, Elbers K, Stoll D, et al.: 1997, Serine pro-tease of pestiviruses: Determination of cleavagesites. J Virol 71:5415–5422.

Tautz N, Meyers G, Stark R, et al.: 1996,Cytopathogenicity of a pestivirus correlates with a27-nucleotide insertion. J Virol 70:7851–7858.

van Campen H, Huzurbazar S, Edwards J, CavenderJL: 1998, Distribution of antibody titers to bovineviral diarrhea virus in infected, exposed, and unin-fected beef cattle. J Vet Diagn Invest 10:183–186.


van Campen H, Williams ES, Edwards J, et al.: 1997,Experimental infection of deer with bovine viraldiarrhea virus. J Wildl Dis 33:567–573.

van Regenmortel MHV, Fauquet CM, Bishop DHL, etal.: 2000, Virus Taxonomy: Seventh Report of theInternational Committee on Taxonomy of Viruses.Academic Press, San Diego, CA.

van Rijn PA, van Gennip HG, Leendertse CH, et al.:1997, Subdivision of the pestivirus genus based onenvelope glycoprotein E2. Virology 237:337–348.

Vilcek S: 1994, Development of PCR tests for the de-tection of bovine herpesvirus-1, bovine respiratorysyncytial viruses and pestiviruses. Vet Med39:687–700.



Vilcek S, Durkovic B, Bobakova M, et al.: 2002,Identification of bovine viral diarrhoea virus 2 incattle in Slovakia. Vet Rec 151:150–152.

Vilcek S, Greiser-Wilke I, Nettleton P, Paton DJ:2000, Cellular insertions in the NS2-3 genome re-gion of cytopathic bovine viral diarrhoea virus(BVDV) isolates. Vet Microbiol 77:129–136.

Vilcek S, Nettleton PF, Paton DJ, Belak S: 1997,Molecular characterization of ovine pestiviruses[published erratum appears in J Gen Virol 1997Aug;78(Pt 8):2129]. J Gen Virol 78:725–735.

Vilcek S, Paton DJ, Durkovic B, et al.: 2001a, Bovineviral diarrhoea virus genotype 1 can be separatedinto at least eleven genetic groups. Arch Virol146:99–115.

Vilcek S, Ridpath JF, Van Campen H, Cevender J:2001b, Identification of a novel pestivirus genotypeisolated from pronghorn antelope. In: Proc 20thAnnu Meet Am SocVirol, p. 196.

Warrener P, Collett MS: 1995, Pestivirus NS3 (p80)protein possesses RNA helicase activity. J Virol69:1720–1726.

Weiland E, Stark R, Haas B, et al.: 1990, Pestivirusglycoprotein which induces neutralizing antibodiesforms part of a disulfide-linked heterodimer. J Virol64:3563–3569.

Wensvoort G, Terpstra C, Boonstra J, et al.: 1986,Production of monoclonal antibodies against swinefever virus and their use in laboratory diagnosis. VetMicrobiol 12:101–108.

Wensvoort G, Terpstra C, de Kluijver EP, et al.:1989a, Antigenic differentiation of pestivirus strainswith monoclonal antibodies against hog choleravirus. Vet Microbiol 21:9–20.

Wensvoort G, Terpstra C, de Kluyver EP: 1989b,Characterization of porcine and some ruminant pes-tiviruses by cross- neutralization. Vet Microbiol20:291–306.

Westaway EG, Brinton MA, Gaidamovich S, et al.:1985, Togaviridae. Intervirology 24:125–139.

Windisch JM, Schneider R, Stark R, et al.: 1996,RNase of classical swine fever virus: biochemicalcharacterization and inhibition by virus-neutralizingmonoclonal antibodies. J Virol 70:352–358.


Xue W, Zhang S, Minocha HC: 1997, Characteri-zation of a putative receptor protein for bovine viraldiarrhea virus. Vet Microbiol 57:105–118.

Yu H, Isken O, Grassmann CW, Behrens SE: 2000, Astem-loop motif formed by the immediate 5’ termi-nus of the bovine viral diarrhea virus genome mod-ulates translation as well as replication of the viralRNA. J Virol 74:5825–5835.

Zhou Y, Moennig V, Coulibaly CO, et al.: 1989, Dif-ferentiation of hog cholera and bovine virus diar-rhoea viruses in pigs using monoclonal antibodies.Zentralbl Veterinarmed [B] 36:76–80.



INTRODUCTIONRNA viruses, including bovine viral diarrhea virus(BVDV), replicate and survive through diverse ge-netic strategies and lifecycles that rely on the biol-ogy and biochemistry of their hosts. Like otherviruses, infection of host cells by BVDV results inmodifications in gene expression that would bebeneficial to virus replication and survival. Up-regulation of genes encoding host proteins involvedin translation and posttranslational processing, andnonlethal down-regulation of genes encoding pro-teins involved in energy production and cell struc-ture have been documented in cells infected withBVDV (Neill and Ridpath, 2003a, 2003b). As seenwith other viruses possessing both cytopathic andnoncytopathic biotypes, cytopathic BVDV amplifiesviral RNA at levels that are several logs higher whencompared to their noncytopathic counterparts (Glewet al., 2003; Kummerer and Meyers, 2000; Mendezet al., 1998; Vassilev and Donis, 2000).

A semiconservative asymmetric model has beenused to describe BVDV RNA replication. Based onfinding of negative-strand template and nascent pos-itive-strand RNA, BVDV replication has been de-tected in cell culture models within 4–6 hours of in-fection, with peak BVDV titer detected at 12–24hours postinfection (Gong et al., 1996; Purchio etal., 1983).

BVDV replication occurs through an RNA-dependent RNA synthesis pathway that utilizes en-zyme activities not typically found in uninfectedhost cells, and therefore, must be encoded in theviral genome and expressed during infection. In eu-karyotic cells, ribosomes require a specific methy-lated cap structure at the 5� end of the genome to sig-nal the initiation of protein synthesis, and in general,each mRNA results in a single polypeptide product.As a family, the RNA viruses have evolved unique

mechanisms to obtain multiple protein productsfrom a single genome by fragmentation at the levelof protein, mRNA, or the gene. Pestiviruses, includ-ing BVDV, utilize their single-stranded positive-sense genome as template for both translation andreplication. BVDV relies on fragmentation by ex-tensive cotranslation and posttranslation proteolyticprocessing using host and viral proteases to obtain atleast 11 viral proteins from a single translatedpolyprotein.

VIRAL GENOME

GENOME ORGANIZATION

The BVDV genome consists of single-strandedpositive-sense RNA approximately 12.5 kb inlength, with some variability in size associated withgenomic deletions, insertions, and duplications. Un-translated regions at the 5� and 3� ends of thegenome (5� UTR and 3� UTR, respectively) flank alarge open reading frame (ORF) that encodes the ap-proximately 4,000 amino acid polyprotein. The highdegree of conservation of nucleotide sequence andstructure for BVDV and the other pestiviruses in the5� and 3� regions of the genome indicates specificgenomic elements required for positive-strand syn-thesis, translation, and possibly packaging of viralgenome (Becher et al., 1999b; Becher et al., 2000;Chon et al., 1998; Deng and Brock, 1993; Yu et al.,1999).

The organization of the BVDV genome is Npro,Capsid, Erns, E1, E2, and p7, which code for struc-tural proteins, followed by NS2-3, NS4A, NS4B,NS5A, and NS5B encoding nonstructural proteins.The genomic organization in cytopathic strains ofBVDV can deviate based on the variety of genomicinsertions, deletions, and replications. The functionand description of proteins encoded by the BVDV

81

4Virus Replication

S. K. Hietala and B. M. Crossley



genome are described in the preceding chapter. Inbrief, the viral-encoded proteins involved in BVDVreplication include an autoprotease encoded by Npro,NS3-encoded serine protease, ATPase, and RNA he-licase, an NS4A-encoded essential cofactor of NS3serine protease, an NS5A-encoded serine phospho-protein, and the NS5B encoded RNA-dependentRNA polymerase. The function of nonstructuralproteins NS4B and NS5A has not been identified,but these proteins have been associated with a mul-tiprotein complex also involving NS3. Blockage ofthe ion channel encoded by the small structural pro-tein p7 has recently been shown to suppress replica-tion of BVDV (Pavlovic et al., 2003).

5�� UTRTranslation among the pestiviruses is cap-independ-ent, relying instead on an internal ribosomal entrysite (IRES), which mediates internal attachment ofribosomes to the translation initiation codon. TheBVDV IRES is encoded in the 5� UTR and containsessential structural or functional components thatmay extend into or include nucleotide sequencewithin the ORF. Although the exact genomic bound-aries of the IRES have not yet been clearly defined,the structural integrity in the region, divided amongthree domains and including a pseudoknot structure,is required for IRES activity (Fields et al., 2001).Structural and biochemical models demonstrate thatthe two stem-loop structures located at the 5� of thegenome and nucleotides of the Npro-coding regionare required for efficient RNA replication. Propersecondary structure in the stem of the 5� terminalstem-loop was found to be critical to efficient trans-lation, and an intact loop and portion of the stem areimportant components for replication (Yu et al.,2000). Based on similarities with other Flaviviridaegenomes, it is postulated that the pseudoknot struc-turally serves to position the ribosomal subunit overthe initiator AUG codon (Lemon and Honda, 1997).The 5� UTR of BVDV, similar to other pestiviruses,is relatively long at approximately 385 nucleotidesand contains multiple initiator codons upstream ofthe actual translation initiation site (Yu et al., 2000).Additionally, four initial nucleotides at the 5� termi-nus of the BVDV genome have been reported to pro-vide an essential signal for replication (Frolov et al.,1998; Yu et al., 2000).

3�� UTRThe 3� end of the BVDV genome also contains crit-ical primary and secondary structures, including aconserved single-stranded region separating two

hairpin loops that functions to direct initiation ofnegative-strand synthesis and is critical for replica-tion (Fields et al., 2001; Yu et al., 1999). The single-stranded region separating the two stem-loop struc-tures is highly conserved among the pestiviruses(Becher et al., 1998b; Yu et al., 1999). It has beensuggested, based on similarities to other single-stranded RNA viruses, that the structurally con-served 3� region may be involved in regulatory func-tions by cross-talk with the 5� structural components(Yu et al., 1999). If analogous to other RNA viruses,the structural interaction of the 5� and 3� regionscould be involved in modulation of RNA-RNA in-teractions, translation, replication, and encapsula-tion steps.

VIRUS BINDING AND RNTRYBased on comparisons to related members of theFlaviviridae, the binding and entry of BVDV in-volves a series of steps, beginning with attachmentor interaction of the virion with specific host cell re-ceptors, followed by internalization and pH depend-ent fusion of the viral envelope and cell membrane.Envelope glycoproteins coded by E2 and Erns havebeen demonstrated to independently bind to cell sur-faces (Hulst and Moorman, 1997; Iqbal et al., 2000).The ability of BVDV to infect a relatively diverserange of cell-types, as well as the tissue and host-species tropisms observed for BVDV have been as-sociated with the E2 envelope glycoprotein (Lianget al., 2003). The E2 protein is translated from ahighly variable region of the genome and may addi-tionally contribute to the ability of BVDV to escapethe host immune response (Ridpath, 2003). Thoughspecific cell receptors for BVDV entry have notbeen well characterized, low-density lipoprotein re-ceptors have been identified (Baranowski et al.,2001). Candidate receptors, as determined by mon-oclonal antibodies capable of blocking infection, in-clude 50kDa, 60kDa, and 93kDa cell surface pro-teins found on bovine cells (Minocha et al.,1997;Schlep et al., 1995; Xue and Minocha, 1993; Xue etal., 1997). After entry into the host cell is complete,viral RNA is released into the host cell cytoplasmand RNA translation begins.

TRANSLATION ANDREPLICATIONInitiation of the translation process is mediated bythe IRES. It has been demonstrated among pes-tiviruses that the IRES binds specifically to the 40Sribosomal subunit in the absence of any additionaltranslation initiation factors. In concert with cellular


components, and functionally linked to the transla-tion process, viral nonstructural proteins assembleinto a functional replicase complex to catalyze tran-scription of positive-sense RNA into full-lengthcomplementary strand negative-sense RNA. Thenegative RNA strands provide template for the repli-case complex to synthesize additional positive-senseRNA molecules, using a semiconservative asym-metric replication model (Warrilow et al., 2000).The model includes three virus-specific RNAs: adouble-stranded replicative form (RF), a partiallysingle-stranded and partially double-stranded repli-cative intermediate (RI), and single-stranded viralRNA.

The replication process begins with a positive-strand replicase complex comprised of viral and cel-lular components formed at the 3’ terminus of thegenome. Progression from initiation to elongationoccurs after the synthesis of nascent RNAs 8–10 nu-cleotides in length (Sun and Kao, 1997). Elongationmediated by viral-polymerase displaces the positivestrand from the RI template, allowing recycling ofthe template while elongation of the prior nascentstrand continues. Approximately six nascent positivestrands per template have been predicted (Gong etal., 1998). The double-stranded replicative form is aproduct of viral RNA used as template for minus-strand synthesis, or may represent the replicative in-termediate during synthesis of the final nascentstrand (no new initiation complexes formed). Duringin vitro experiments, the ratio of positive-to-minus–sense RNA increases from 2:1 at 4 hours to10:1 by 12 hours postinoculation (Gong et al., 1996).

GENETIC RECOMBINATIONThe high rate of genetic insertion and recombinationevents in BVDV as compared to other positive-senseRNA viruses is unique to pestiviruses and is possi-bly related to some unique character of the pes-tivirus polymerase (Fields et al., 2001). Recombina-tion can occur in BVDV when virions havingdifferent genomic sequences coinfect the same hostcell, allowing genetic crossover resulting in a hybridBVDV strain (Moenning et al., 1993; Becher et al.,1999a). Sequence analyses of cytopathic strains ofBVDV, as discussed in more detail below, suggestgenomic hot-spots for viral recombination at theborder of NS2 and NS3, though recent cloning ex-periments using replicons expressing NS3 argue forfunctional selection of mutations or recombinationevents, based on a highly conserved region withinthe NS3 that is critical for effective viral replication(Tautz and Thiel, 2003). Genetic recombinations be-

tween noncytopathic and cytopathic (Ridpath andBolin, 1995b, Fritzemeier et al., 1997, Nagai et al.,2003), BVDV type 1 and type 2 (Ridpath and Bolin,1995b), BVDV persistent infection strains and vac-cine strain (Ridpath and Bolin, 1995b; Becher et al.,2001), and between BVDV and host RNA (Becheret al., 2002; Baroth et al., 2000; Mendez et al., 1998;Meyers et al., 1989; Meyers et al., 1991a; Meyers etal., 1998; Qi et al. 1998; Ridpath and Bolin, 1995a;Ridpath et al., 1994; Ridpath and Neill 2000; Rincket al., 2001; Tautz et al., 1996; Tautz and Thiel.,2003) have been well documented. Cellular inser-tions include duplicated viral sequences, host ubiq-uitin or ubiquitin homologs, ribosomal ubiquitingene fusion protein, host mRNA encoding a DnaJ or J-domain-regulatory proteins, and multiplemicrotubule-associated proteins. The insertion andrecombination events can occur at different sitesalong the viral genome (Desport et al., 1998;Meyers et al., 1992; Meyers and Thiel, 1996); how-ever, those involving the NS2/3 coding sequenceshave been most extensively studied, based on asso-ciation with the generation of cytopathic BVDV(Meyers et al., 1996, Fields et al., 2001). The geneticmodifications associated with NS2-NS3 cleavagetend to be associated with two specific sites locatedwithin 54 base pairs of each other and within BVDVNS2/NS3 (Ridpath and Neill, 2000).

CYTOPATHOLOGYCytopathic strains of BVDV evolve from noncyto-pathic BVDV by mutation, with specific genomicrearrangements varying considerably between dif-ferent cytopathic strains of the virus (Becher at al.,1998a; Kummerer and Meyers, 2000; Meyers et al., 1991a, 1991b; Meyers et al., 1998; Muller et al.,2003; Nakamura et al., 1997; Qi et al.,1992; Qi etal.,1998; Ridpath and Bolin, 1995a; Ridpath andNeill, 2000; Tautz et al., 1993; Tautz et al., 1994;Tautz et al., 1999; Tautz et al., 2003). The genomicrearrangements generating cytopathic BVDV—including duplications of the BVDV genome, inser-tion of cellular sequences at the junction of NS2 andNS3, or in-frame deletions—primarily affect thecoding region for the NS2-3 polypeptide and resultin the production of NS3. Cleavage of NS2/NS3 oc-curs by different theoretical strategies, dependent onthe strain of BVDV and may include introduction ofsequences that form new cleavage sites, introductionof sequences with autocatalytic activity, introduc-tion of conformational changes signaling cellularprotease, or activation of latent protease activity di-rectly encoded by NS2 (Fields et al., 2001; Meyers

Virus Replication 83


and Thiel, 1996; Ridpath and Neill 2000; Rinck etal., 2001). NS3 production, and associated BVDVcytopathogenicity, have also been generated bypoint mutations in the region encoding NS2(Kummerer et al., 1998; Kummerer and Meyers,2000), upstream of the NS2/3 region (Kummerer etal., 1998), or through a specific mutation in the re-gion encoding NS4b (Qu et al., 2001). The majorityof cytopathic BVDV characterized in the literatureappear to be associated with genomic insertions, ofeither host cell sequences or duplications of the viralgenome. A relatively small number of the character-ized cytopathic BVDV sequences appear to arisethrough recombination between BVD viruses hav-ing different genomic sequences. Proteolytic pro-cessing associated with generation of BVDV cy-topathology is further discussed in Chapter 11. Inaddition to BVDV superinfection with a cytopathicand noncytopathic BVDV pair, spontaneous cases ofmucosal disease arise when noncytopathic persistentBVDV mutates or undergoes recombination, as re-viewed in Chapter 8 on reproductive disease andpersistent infections.

DEFECTIVE INTERFERINGPARTICLES ANDCYTOPATHOLOGYViral genomes having deletions or truncations,which can replicate with the support of a helpervirus or helper virus-coreplication, are referred to asdefective interfering (DI) particles (Huang andBaltimore, 1970). BVDV DI particle have been as-sociated with cytopathology in vitro and mucosaldisease in vivo. A BVDV DI particle, approximately4.3 kb smaller than the corresponding noncytopathicBVDV virus and lacking all structural genes plus theamino-terminal region of the nonstructural NS3 pro-tein, was initially described from a persistently in-fected calf in 1994 (Tautz et al., 1994). Other BVDVDI, with slightly different deletions but lacking allstructural genes, have been described (Kupfermannet al., 1996), and at least one (DI9c) has been shownto function as an autonomous replicon (Behrens etal., 1998).

DI particles have been implicated in the onset ofmucosal disease in cattle persistently infected with anoncytopathic BVDV, potentially through recombi-nation events involving structural proteins of thehelper noncytopathic virus (Kupferman et al., 1996).Theoretically, the structural genes of the persistentvirus would not be recognized by the immune sur-veillance of the host, allowing the recombinantBVDV virus to replicate while escaping the host im-

mune response. The existence of DI in naturally oc-curring disease has been confirmed from animalswith mucosal disease; however, DI are difficult todetect due to low virus yields, and they also can getlost easily during virus isolation and plaque purifi-cation (Meyers and Thiel, 1996). DI particles aretypically observed in vitro only during cell culturepassages with a high multiplicity of infection(Meyers and Thiel, 1996).

PROTEOLYTIC PROCESSINGThe single BVDV polyprotein translated from theopen reading frame (ORF) undergoes proteolyticprocessing to derive at least 11 viral proteins(Meyers and Thiel, 1996). BVDV structural proteinsare located within the N-terminus third of thepolyprotein, followed by a single non-virion protein(p7), and nonstructural proteins that comprise the re-mainder of the polyprotein. Proteolytic processingof the polyprotein is mediated by both host and viralenzymes, including a viral autoprotease (Npro)unique to pestiviruses, host signal peptidase, andviral serine protease (Fields et al., 2001). BVDVproteolytic processing and function of the structuraland nonstructural proteins are discussed in detail inChapters 3 and 11.

REGULATION OF TRANSLATIONAND REPLICATIONTight regulation of translation and replication is re-quired during the BVDV lifecycle, since the samepositive-sense RNA is used as template for bothprocesses. Structural and functional studies of the 5’terminal fragment of the BVDV genome, includingthe IRES site, demonstrate secondary structure that could enable viral RNA to switch from a trans-lation to a replication cycle (Yu et al., 2000). Ac-cumulation of viral proteins, such as NS5A andNS5B, which can inhibit IRES-dependent transla-tion may also provide a regulatory mechanism forthe translation-to-replication switch. Kinetic analy-sis of translation and replication using BVDV full-length and replicon genomes indicate that regulationof the translation-to-replication switch is not spe-cific for the BVDV IRES, but may result from regu-latory proteins interacting with viral genome outsidethe IRES or from interaction between viral and hosttranslation protein complexes (Li and McNally,2001). Alternately, competition for replicase andribosome-loading on the same template moleculemay serve to regulate the switch.

The asymmetric replication of the BVDV genomeis consistent with other positive-sense RNA viruses,



where an excess of newly synthesized positive-senseRNA compared to negative-sense RNA is generated.A close functional control appears to exist betweengeneration of the polyprotein and generation ofreplication complexes. The replicase complexes,formed by association of new viral nonstructuralproteins plus cellular components with the 3’ end ofthe genomic RNA, act to catalyze synthesis of a lowcopy number of negative-strand RNA templates forthe transcription of an excess of positive-strandRNA progeny (Behrens et al., 1998; Yu et al., 2000).

VIRION ASSEMBLY ANDEXOCYTOSISLittle information is available on the assembly andrelease of virions from infected host cells. BVDVvirions appear to mature in intracellular vesicles atthe Golgi apparatus or endoplasmic reticulum wherethe lipid envelope is acquired through budding intothe vesicle lumen. Virus maturation, including con-formational stabilization through glycoprotein fold-ing of E1-E2 (Branza-Nichita et al., 2001) and asso-ciated transport to the cell surfaces is mediated byhost cell enzymes and processes (Zitzmann et al.,1999, Branza-Nichita et al., 2001, Durantel et al.,2001). The intact virions are released by buddinginto the cisternae of the endoplasmic reticulum, fol-lowed by exocytosis (Grummer et al., 2001;Bielefeldt-Ohmann and Block, 1982; Gray andNettleton, 1987) with detection reported as early as8 hours postinfection (Nuttall, 1980).

QUASISPECIES AND POPULATIONGENETICSThe RNA-dependent RNA polymerases that catalyzeRNA replication and reverse transcription have mini-mal proofreading activities, resulting in error ratestens of thousands times greater than those encoun-tered during DNA replication (Domingo and Holland,1997; Holland et al., 1992; Malet et al., 2003; Moyaet al., 2000). Because mutation is a frequent event, anRNA virus population does not represent a homoge-neous clone, but is a “cloud of mutants” clusteredaround the most frequent viral sequence (Moya et al.,2000). The related, but nonhomogeneous populationshave been identified as quasispecies (Eigen, 1993).Although the theories of quasispecies and populationgenetics differ in mathematical modeling and the con-tributions of point mutations to viral evolution (Pageand Nowak, 2002; Moya et al., 2000; Domingo,2003), it is agreed that the target of natural selectionis not a single fittest genotype, but a distribution ofgenotypes around a master sequence. The existence

of quasispecies among BVDV strains reflects thehigh replication rate of the virus, as well as the lackof proofreading capacity of the viral RNA-dependentRNA polymerase. BVDV is able to effectively evadeemerging humoral and cellular immune responses(Bolin et al., 1991), which may be a function ofBVDV quasispecies generation in infected hosts.Among BVDV, point mutations occurring approxi-mately once per 10 kb would be equivalent to one ormore mutations per BVDV viral replication cycle andlogically explain the existence of multiple distinct,but closely related, genetic variants of BVDV. BVDVquasispecies have been recovered (Becher et al.,1999a; Jones et al., 2002) from persistently infectedcattle, though other studies have reported apparentstabilization of the genome in persistent infection(Edwards et al., 1991; Edwards and Patton, 1995;Hamers et al., 1998, Hamers et al., 2001). Stabiliza-tion of the antigenic, if not genetic, character of thevirus is consistent with the theory of immunologicelimination of BVDV variants in PI animals. Stabili-zation, or lack of significant genetic change in thegenome of BVDV isolated from PI cattle over time, isalso consistent with observations of herd-specificstrains of BVDV (Paton et al., 1995).

REPLICATION SITESSimilar to most RNA viruses, BVDV replicates inthe cellular cytoplasm. In studies aimed at subcellu-lar localization of BVDV replication, nonstructuralproteins NS2-3 and NS3 were found in associationwith the cytoplasmic face of the endoplasmic retic-ulum, but not with the Golgi apparatus or lysosomalmembrane, suggesting that replication occurs on thecytoplasmic side of the endoplasmic reticulum(Zhang et al., 2003). Both cell and species tropismsare seen with BVDV. Though BVDV can infect awide variety of cell types, both in vitro and in vivo,there is an apparent predilection for cells of the im-mune system, including T cells, B cells, monocytes,macrophages, and dendritic cells (Sopp et al., 1994).The biotypes of BVDV behave differently, withnoncytopathogenic strains having tropism for leuco-cytes, lymphoid tissues, parotic gland, proximalcolon, and respiratory tract; cytopathogenic strainsare generally associated with the gastrointestinaltract (Greiser-Wilke et al. 1993; Liebler et al.,1991)and replicate in the ovaries (Grooms et al., 1998).

SUMMARY AND CONCLUSIONSImportant genetic diversity occurs among BVDVisolates, as evidenced by point mutations, deletions,genetic recombination among virus strains, and by



integration of segments of the host genome. EachBVDV infection represents an encounter betweenthe genetic makeup of the virus and that of its host,allowing BVDV to respond and take advantage of itsability to generate and exist as quasispecies.Knowledge of the replication cycle allows not onlyunderstanding of the epidemiology and a wide rangeof diseases caused by BVDV infection, but also pro-vides opportunities for development of effective de-tection, vaccination, and control strategies based onmediation of receptor recognition, interference withvirus-specific replication, and blockage of viral pro-tein synthesis and exotcytosis.

REFERENCESBaranowski E, Ruiz-Jarabo M, Domingo E: 2001,

Evolution of cell recognition by viruses. Science292:1102–1105.

Baroth M, Orlich M, Thiel H-J, et al.: 2000, Insertionof cellular NEDD8 coding sequences in a pes-tivirus. Virology 278:456–466.

Becher P, Orlich M, Konig M, Thiel H-J: 1999a,Nonhomologous RNA recombination in bovineviral diarrhea virus: Molecular characterization of avariety of subgenomic RNAs isolated during anoutbreak of fatal mucosal disease. J Virol73:5646–5653.

Becher P, Orlich M, Kosmidou M, et al.: 1999b,Genetic diversity of pestiviruses: Identification ofnovel groups and implications for classification.Virology 262:64–71.

Becher P, Orlich M, Thiel H: 2000, Mutations in the5’ nontranslated region of bovine viral diarrheavirus result in altered growth characteristics. J Virol74:7884–7894.

Becher P, Orlich M, Thiel H-J: 1998a, RibosomalS27a coding sequences upstream of ubiquitin cod-ing sequences in the genome of a pestivirus. J Virol72:8697–8704

Becher P, Orlich M, Thiel H-J: 1998b, Complete ge-nomic sequence of border disease virus, a pestivirusfrom sheep. J Virol 72:5165–5173.

Becher P, Orlich M, Thiel H-J: 2001, RNA recombi-nation between persisting pestivirus and a vaccinestrain: generation of cytopathic virus and inductionof lethal disease. J Virol 75:6256–6264.

Becher P, Thiel H-J, Collins M, et al.: 2002, Cellularsequences in pestivirus genomes encoding gamma-aminobutyric acid (A) receptor-associated proteinand Golgi-associated ATPase enhancer of 16 kilo-daltons. J Virol 76:13069–13076.

Behrens SE, Grassmann CW, Thiel H-J, et al.: 1998,Characterization of an autonomous subgenomicpestivirus RNA replicon. J Virol 72:2364–2372.

Bielefeldt-Ohmann H, Bloch B: 1982, Electron micro-scopic studies of bovine viral diarrhea virus in tis-sues of diseased calves and in cell cultures. ArchVirol 71:57–74.

Bolin SR, Littledike ET, Ridpath JF: 1991, Serologicdetection and practical consequences of antigenicdiversity among bovine viral diarrhea viruses in avaccinated herd. Vet Res 52:1033–1037.

Branza-Nichita N, Durantel D, Carrouee-Durantel S,et al.: 2001, Antiviral effect of N-butyldeoxynojir-imycin against bovine viral diarrhea virus correlateswith mis-folding of E2 envelope proteins and im-pairment of their association into E1-E2 het-erodimers. J Virol 75:3527–3536.

Chon SK, Perez DR, Donis RO: 1998, Genetic analy-sis of the internal ribosomal entry segment ofbovine viral diarrhea virus. Virology 251:370–382.

Deng R, Brock KV: 1993, 5� and 3� untranslated re-gions of the pestivirus genome: Primary and sec-ondary structure analyses. Nucleic Acid Res21:1949–1957.

Desport M, Collins ME, Brownlie J: 1998, Genomeinstability in BVDV: An examination of the se-quence and structural influences on RNA recombi-nation. Virology 246: 352–261.

Domingo E, Baranowski E, Ruiz-Jacabo CM, Martin-Hernandez AM, et al.: 1998, Quasispecies structureand persistence of RNA viruses. Emerging InfectDis 4:521–527.

Domingo E, Holland JJ: 1997, RNA mutations andfitness for survival. Ann Rev Microbiol 51:151–178.

Durantel D, Branza-Nichita N, Carrouee-Durantel S,et al.: 2001, Study of the mechanism of antiviral ac-tion of imino sugar derivatives against bovine viraldiarrhea virus. J Virol 75:8987–8998.

Edwards S, Patton D: 1995, Antigenic differencesamong pestiviruses. Vet Clinics N Amer: Food AnimPract 11:563–577.

Edwards S, Wood L, Brockman S, Ibata G: 1991,Clinical and virological observations of a mucosaldisease outbreak with persistently infected seropos-itive survivors. Arch Virol Suppl 3:125–132.

Eigen M: 1993, Viral quasispecies. Sci Amer269:42–49.

Fields BN, Knipe DM, Howley PM, Griffen DE, Eds.:2001, Fields’ Virology, 4th ed., Vol.1. Lippincott,Williams, and Wilkins, Philadelphia, PA.

Fritzemeier J, Haas L, Leibler E, et al.: 1997, The de-velopment of early versus late onset mucosal dis-ease is a consequence of two different pathogenicmechanisms. Arch Virol 142:1335–1350.

Frolov I, McBride MS, Rice CM: 1998, Cis-actingRNA elements required for replication of bovineviral diarrhea virus-hepatitis C virus 5’ nontrans-lated region chimeras. RNA 4:1418–1435.



Glew EJ, Carr BV, Brackenbury LS, et al.: 2003,Differential effects of bovine viral diarrhoea viruson monocytes and dendritic cells. J Gen Virol84:1771–1780.

Gong Y, Shanon A, Westaway EG, et al.: 1998, Thereplicative intermediate molecule of bovine viral di-arrhoea virus contains multiple nascent strands.Arch Virol 143:399–404.

Gong Y, Trowbridge R, MacNaughton TB, et al.:1996, Characterization of RNA synthesis during aone-step growth curve and of the replication mech-anism of bovine viral diarrhea virus. J Gen Virol77:2729–2736.

Gray EW, Nettleton PF: 1987, The ultrastructure ofcell cultures infected with border disease andbovine virus diarrhoea viruses. J Gen Virol68:2339–2346

Greiser-Wilke I, Hass L, Dittmar K, et al.: 1993, RNAInsertions and gene duplications in the nonstruc-tural protein p125 region of pestivirus strains andisolates in vitro and in vivo. Virology 193:977–981.

Grooms DL, Brock KV, Ward LA: 1998, Detection ofcytopathic bovine viral diarrhea in the ovaries ofcattle following immunization with a modified livebovine viral diarrhea virus vaccine. J Vet DiagnInvest 10:130–134.

Grummer B, Beer M, Liebler-Tenorio, et al.: 2001,Localization of viral proteins in cells infected withbovine viral diarrhoea virus. J Gen Virol82:2597–605.

Hamers C, Dehan P, Couvreur B, et al.: 2001, Diver-sity among bovine pestiviruses. Vet J 161:112–122.

Hamers C, Lecomte C, Kulscar G, et al.: 1998,Persistently infected cattle stabilize bovine viral di-arrhea virus leading to herd specific strains. VetMicrobiol 61:177–182.

Holland JJ, DeLaTorre JC, Steinhauser DA: 1992,RNA virus populations as quasispecies. Curr TopMicrobiol Immunol 176:1–20.

Huang AD, Baltimore D: 1970, Defective interferingparticles and viral disease processes. Nature226:325–327.

Hulst MM, Moorman RJ: 1997, Inhibition of pes-tivirus infection in cell culture by envelope proteinsE(rns) and E2 of classical swine fever virus: E(rns)and E2 interact with different receptors. J Gen Virol78:2779–2787.

Iqbal M, Flick-Smith H, McCauley JW: 2000,Interactions of bovine viral diarrhea virus glycopro-tein E(rns) with cell surface glycosaminoglycans. JGen Virol 81:451–459.

Jones LR, Zandomeni R, Weber EL: 2002, Quasi-species in the 5’ untranslated genomic region ofbovine viral diarrhoea virus from a single individ-ual. J Gen Virol 83:2161–2168.

Kummerer BM, Meyers G: 2000, Correlation betweenpoint mutations in NS2 and the viability and cy-topathogenicity of bovine viral diarrhea strainOregon analyzed with an infectious cDNA clone. JVirol 74:390–400.

Kummerer BM, Stoll D, Meyers G: 1998, Bovineviral diarrhea virus strain Oregon: A novel mecha-nism for processing of NS2-3 based on point muta-tions. J Virol 72:4127–4138.

Kummerer BM, Tautz N, Becher P, et al.: 2000, Thegenetic basis for cytopathogenicity of pestiviruses.Vet Microbiol 77:117–128.

Kupferman H, Thiel H-J, Dubovi EJ, et al.: 1996,Bovine viral diarrhea virus: Characterization of acytopathic defective interfering particle with twointernal deletions. J Virol 70:8175–8181.

Lemon SM, Honda M: 1997, Internal ribosome entrysites within the RNA genomes of hepatitis C virusand other flaviviruses. Seminars Virol 8:274–288.

Li Y, McNally J: 2001, Characterization of RNA syn-thesis and translation of bovine viral diarrhea virus(BVDV). Virus Genes 23:149–155.

Liang D, Sainz IF, Ansari IH, et al.: 2003, The enve-lope glycoprotein E2 is a determinant of cell cul-ture tropism in ruminant pestiviruses. J Gen Virol84:1269–1274.

Liebler EM, Waschbusch J, Pohlenz JF, et al.: 1991,Distribution of antigen of noncytopathogenic andcytopathogenic bovine viral diarrhea virus biotypesin the intestinal tract of calves following experi-mental production of mucosal disease. Arch VirolSuppl 3:109–124.

Malet I, Magali B, Argut H, et al.: 2003, From RNAto quasispecies: A DNA polymerase with proof-reading activity is highly recommended for accu-rate assessment of viral diversity. J Virol Methods109:161–170.

Mendez E, Ruggli N, Collett MS, et al.: 1998,Infectious bovine viral diarrhea virus (strainNADL) RNA from stabile cDNA clones: A cellularinsert determines NS3 production and viral cy-topathogenicity. J Virol 72:4737–4745.

Meyers G, Rumenapf T, Tautz N et al.: 1991a, Inser-tion of cellular sequences in the genome of bovineviral diarrhea virus. Arch Virol Suppl 3:133–142.

Meyers G, Rumenapf T, Thiel H-J: 1989, Ubiquitin ina togavirus. Nature 341:491.

Meyers G, Stoll D, Gunn M: 1998, Insertion of a se-quence encoding light chain 3 of microtubule asso-ciated proteins 1A and 1B in a pestivirus genome:Connection with virus cytopathogenicity and induc-tion of lethal disease in cattle. J Virol 72:439–448.

Meyers G, Tautz N, Dubovi EJ, Thiel H: 1991b, Viralcytopathogenicity correlated with integration ofubiquitin-coding sequences. Virology 180:602–616.



Meyers G, Tautz N, Stark R, et al.: 1992,Rearrangement of viral sequences in cytopathicpestiviruses. Virology 191:368–386.

Meyers G, Thiel H-J: 1996, Molecular characteriza-tion of pestiviruses. Advances in Virus Research47:53–118

Minocha HC, Xue W, Reddy JR: 1997, A 50kDamembrane protein from bovine kidney cells is a pu-tative receptor for bovine viral diarrhea virus(BVDV). Adv Exp Med Biol 412:145–148.

Moenning V, Greiser-Wilke I, Frey HR: 1993,Prolonged persistence of cytopathogenic bovineviral diarrhea virus (BVDV) in persistently viremiccattle. Zentralblatt fur Vet Med 40:371–377.

Moya A, Elena SF, Bracho A, et al.: 2000, The evolu-tion of RNA viruses: A population genetics view.Proc Natn Acad Sci 97:6967–6973.

Muller A, Rinck G, Thiel H-J, et al.: 2003, Cell-derived sequences in the N-terminal region of thepolyprotein of a cytopathogenic pestivirus. J Virol77:10663–10669.

Nagai M, Sakoda Y, Mori M, et al.: 2003, Insertion ofcellular sequence and RNA recombination in thestructural protein coding region of cytopathogenicbovine viral diarrhea virus. J Gen Virol84:447–452.

Nakamura S, Sakamoto K, Sakoda Y, et al.: 1997,Variation from cytopathogenic biotype to noncy-topathogenic biotype is correlated with the deletionof cellular sequences from bovine viral diarrheaviruses. J Vet Med Sci 59:361–370.

Neill JD and Ridpath JF: 2003a, Gene expressionchanges in BVDV2-infected MDBK cells.Biologicals 31:97–102.

Neill JD and Ridpath JF: 2003b, Gene expressionchanges in MDBK cells infected with genotype 2of bovine viral diarrhea virus. Vet Microbiol96:301–312.

Nuttall PA: 1980, Growth characteristics of twostrains of bovine virus diarrhea virus. Arch Virol66:365–369.

Page KM, Nowak MA: 2002, Unifying evolutionarydynamics. J Theor Biol 219:93–98.

Paton DJ, Carlsson U, Lowings JP, et al.: 1995,Identification of herd-specific bovine viral diar-rhoea virus isolates from infected cattle and sheep.Vet Microbiol 43:283–94.

Pavlovic D, Neville DC, Argaud O et al.: 2003, Thehepatitis C virus p7 protein forms an ion channelthat is inhibited by long alkyl-chain iminosugar de-rivatives. Proc Natn Acad Sci 100:6104–6108.

Purchio AF, Larson R, Collett MS: 1983,Characterization of virus-specific RNA synthesizedin bovine cells infected with bovine viral diarrheavirus. J Virol 48:320–324.

Qi F, Ridpath JF, Berry ES: 1998, Insertion of bovineSMT3B genes in the NS4A and duplication of NS3in a bovine viral diarrhea virus genome correlatewith the cytopathogenicity of the virus. Virus Res57:1–9.

Qi F, Ridpath JF, Lewis T, Bolin SR, and Berry ES:1992, Analysis of the bovine viral diarrhea virusgenome for possible cellular insertions. Virology189:285–292.

Qu L, McMullan LK, Rice CM: 2001, Isolation andcharacterization of noncytopathic pestivirus mutantsreveals a role for nonstructural protein NS4B inviral cytopathogenicity. J Virol 75:10651–10662.

Ridpath JF: 2003, BVDV genotypes and biotypes:Practical implications for diagnosis and control.Biologicals 31:127–131.

Ridpath JF, Bolin SR: 1995a, The genomic sequenceof virulent bovine viral diarrhea virus (BVDV)from the type 2 genotype: Detection of a large ge-nomic insert in a non-cytopathic BVDV. Virology212:39–46.

Ridpath JF, Bolin SR: 1995b, Delayed onset postvac-cinal mucosal disease as a result of genetic recom-bination between genotype 1 and genotype 2BVDV. Virology 212:259–262.

Ridpath JF, Neill JD: 2000, Detection and characteri-zation of genetic recombination in cytopathic type2 bovine viral diarrhea viruses. J Virol74:8771–8774.

Ridpath JF, Qi F, Bolin SR, et al.: 1994, Natural re-combination in bovine viral diarrhea viruses. ArchVirol Suppl 9:239–244.

Rinck G, Birghan C, Harada T, et al.: 2001, A cellularJ-domain protein modulates polyprotein processingand cytopathogenicity of a pestivirus. J Virol75:9470–9482.

Schelp C, Greiser-Wilke I, Wolf G, et al.: 1995,Identification of cell membrane proteins linked tosusceptibility to bovine viral diarrhea virus infec-tion. Arch Virol 140:1997–2009.

Sopp P, Hooper LB, Clarke MC, et al.: 1994,Detection of bovine viral diarrhoea virus p80 pro-tein in subpopulations of bovine leukocytes. J GenVirol 75:1189–1194.

Sun J-H. Kao CC: 1997, Characterization of RNAproducts associated with or aborted by a viralRNA-dependent RNA polymerase. Virology236:348–353.

Tautz N, Harada T, Kaiser A, et al.: 1999,Establishment and characterization of cytopatho-genic and noncytopathogenic pestivrus replicons. JVirol 73:9422–9432.

Tautz N, Meyers G, Stark R, et al.: 1996,Cytopathogenicity of a pestivirus correlates with a27-nucleotide insertion. J Virol 70:7851–7858.



Tautz N, Meyers G, Thiel H-J: 1993, Processing ofpoly-ubiquitin in the polyprotein of an RNA virus.Virology 197:74–85.

Tautz N, Thiel H-J: 2003, Cytopathogenicity of pes-tiviruses: cleavage of bovine viral diarrhea virusNS2-3 has to occur at a defined position to allowreplication. Arch Virol 148:1405–1412.

Tautz N, Thiel H-J, Dubovi EJ, et al.: 1994,Pathogenesis of mucosal disease: A cytopathogenicpestivirus generated by an internal deletion. J Virol68:3289–3297.

Vassilev VB, Donis RO: 2000, Bovine viral diarrheavirus induced apoptosis correlates with increasedintracellular viral RNA accumulation. Virus Res69:95–107.

Warrilow D, Lott WB, Greive S, et al.: 2000,Properties of the bovine diarrhoea virus replicase inextracts of infected MDBK cells. Arch Virol 145:2163–2171.

Xue W, Minocha HC: 1993, Identification of the cellsurface receptor for bovine viral diarrhea virus byusing anti-idiotype antibodies. J Gen Virol74:73–79.

Xue W, Zhang S, Minocha HC: 1997, Characteriza-tion of a putative receptor protein for bovine viraldiarrhea virus. Vet Microbiol 57:105–118.

Yu H, Grassmann CW, Behrens SE: 1999, Sequenceand structural elements at the 3’ terminus of bovineviral diarrhea virus genomic RNA: Functional roleduring RNA replication. J Virol 73:3638–3648.

Yu H, Isken O, Grassmann CW: 2000, A stem-loopstructure formed by the immediate 5’ terminus ofthe bovine viral diarrhea virus genome modulatestranslation as well as replication of the viral RNA.J Virol 74:5825–5835.

Zhang G, Flick-Smith H, McCauley JW: 2003, Dif-ferences in membrane association and sub-cellulardistribution between NS2-3 and NS3 of bovineviral diarrhea virus. Virus Res 97:89–102.

Zitzmann N, Mehta AS, Carrouee S, et al.: 1999,Imino sugars inhibit the formation and secretion ofbovine viral diarrhea virus, a pestivirus model ofhepatitis C: implications for development of broad-spectrum anti-hepatitis virus agents. Proc. NatnAcad Sci 96:11878–11882.



INTRODUCTIONTransmission of bovine viral diarrhea virus (BVDV)in cattle follows the infectious disease epidemiologycharacteristics of many viral infections, with a no-table exception: Infection with BVDV can manifestin two patterns for the infectious period of the dis-ease, depending on when the animal became in-fected. Infected animals can shed the virus tran-siently, following an acute infection acquiredpostnatally, or they can shed the virus persistently,following congenital infection acquired before120–150 days of gestation. This chapter reviewsconcepts of disease transmission in the context ofBVDV followed by specific discussion of routes oftransmission and factors that may affect virus trans-mission from animals with an acute or a persistentinfection. The chapter concludes with a brief sum-mary of general considerations for minimizingtransmission of the virus in cattle populations.

INFECTIOUS DISEASEEPIDEMIOLOGY IN THECONTEXT OF BVDVTRANSMISSION

FACTORS AFFECTING BVDV TRANSMISSION

Conceptually, four main factors can be expected toinfluence the transmission of infectious agents suchas BVDV (Haloran, 1998). The first factor is the in-fectiousness of the virus strain, given the exposuredosage and route of infection, referred to as the co-efficient of infectiousness (ß). The latter is a meas-ure of the probability that infection will be transmit-ted to a susceptible animal following contact with aninfectious animal. As discussed later, animals per-sistently infected with BVDV generally shed morevirus over a much longer period of time than ani-mals with an acute infection, so the coefficient of in-

fectiousness would be higher for PI animals than foranimals with an acute infection. The second factor isthe number of adequate contacts per time period (k)between infectious and susceptible animals. An ade-quate contact is one that would be sufficient fortransmission to take place, such as direct nose-to-nose contact between an infectious animal and aseronegative, susceptible animal. The third factor isthe duration of the infectious period (d) for the spe-cific host, or, alternatively, the prevalence of infec-tious animals in a herd during a specified period.The fourth factor is the presence of truly susceptibleanimals that lack specific or cross-reacting serumneutralizing (SN) antibodies (humoral immunity)and/or cell-mediated immunity necessary to preventinfection.

These factors can be used to estimate rates ofvirus transmission in a susceptible population. Oneapproach to estimating the rate or force of transmis-sion is by the use of basic reproduction number (R0),referred to as “R-not,” which represents the ex-pected number of new infections resulting fromcontact with an infectious index case animal, and iscalculated as R0 = ßdk. For example, suppose, hypo-thetically, that the probability of a susceptibleweaned calf becoming infected per adequate contactwith an acutely infected animal in a feedlot is 0.02.That is, if an adequate contact was made between anacutely infected animal and 100 susceptible ani-mals, the virus would be transmitted successfully to2 animals. Suppose also that the number of contactsmade, on average, is 50 per day and the duration ofinfectiousness is 4 days, then R0 = ßdk = (0.02/con-tact)(4-day duration)(50 contacts/day) = 4. A valueof 4 would indicate that the acutely infected calf(index case) would infect 4 susceptible calves, andeach of them in turn would infect 4 other calves.

As transmission proceeds through a herd, how-

91

5Virus Transmission

Mark C. Thurmond



ever, infected animals become immune and thenumber of susceptible animals decline. As the num-ber of susceptible animals decline, the likelihoodthat an infectious animal will come in contact with asusceptible animal will decrease, and the R-valuewill decline with each newly infected animal.Eventually, the number of susceptible animals be-comes so small that the transmission rate will de-cline to R = 1, referred to as the endemic state, inwhich each new infection creates only one morenew infection. As the number of susceptible animalsdeclines further, R will become less than 1 and newinfections will stop because transmission is nolonger supported by a sufficient number of suscepti-ble animals. Transmission will not continue unlessadditional susceptible animals are added to the herd,the immunity wanes, the contact rate is increased,the duration of infectiousness increases (i.e., a re-crudescence or extension of shedding), or patho-genicity of the virus changes (i.e., mutation).

Because ß is often unknown, an alternative ap-proximation for R0 is R0 = 1 + L/A, where L is theaverage or median life span of animals in a herd andA is the average or median age of animals at the timeof infection. Suppose, for example, the median life-time of a cow on dry-lot dairies is 40 months andhalf (median) of the animals are infected withBVDV by 8 months of age (Rush et al., 2001), thenR0 = 1 + 40/8 = 6. Thus, one would expect that in afully susceptible population, introduction of 1 in-fected animal would result in transmission of infec-tion to 6 susceptible animals, which in turn wouldtransmit infection to other susceptible animals, andso on. This somewhat hypothetical example does notdistinguish between persistent and acute infections.

Alternatively, ß can be estimated using the for-mula for incidence of BVDV infection, II, alongwith available data and reasonable assumptions.Incidence provides a measure of transmission by es-timating the rate at which new BVDV infectionsoccur, or the proportion of susceptible animals thatbecome infected during a specified time period. Theformula for II is II = ßPk, where ß is the unknownprobability of infection given a contact (coefficientof infectiousness), P is the assumed prevalence ofinfectious animals for a given time period, and k isthe contact rate for the given time period (Haloran,1998). Suppose incidence is 0.5% per day forweaned calves in a feedlot environment (Taylor etal., 1995; Rush et al., 2001), k is about 25 contactsper day, and the prevalence of PI animals is 1% (as-suming that all PI animals are infectious and ignor-ing the prevalence of acutely infectious animals),

then ß = (0.005)/(25)(0.01) = 0.02. After substitutingthis value for ß into the formula for R0, and assum-ing an infectious period of at least 100 days for ani-mals with PI, R0 = (0.02)(25)(100) = 50. Becausethese formulas assume random mixing, constant du-ration of infectiousness, and a constant coefficientof infectiousness, they should be used to obtain ageneral sense for the rate of transmission. The mainbenefit of these equations is that they provide a con-ceptual context for key elements in the transmissionof BVDV.

MEASURING THE EFFECT OF VACCINATIONON TRANSMISSION

In addition to characterizing key forces influencingBVDV transmission (i.e., probability of infection,contact rate, and duration of infectiousness), esti-mates of R0 and II can be used as a benchmark forassessing vaccine efficacy required to control infec-tion. Suppose, for example, that R0 = 4, whereby in-troduction of an acutely infected animal into a com-pletely susceptible herd or group of animals willresult in 4 new infections. If the proportion of theanimals, p, represents those animals that have beeneffectively immunized and that are not susceptible is0.5, 1 new BVDV infection will result in only 2 newinfections rather than 4, because half of the contactswill be immune. In planning a vaccination program,we can use R0 to estimate p, the proportion of theherd that must be immunized to prevent transmis-sion within the herd. If p � R0 animals will not be-come infected because of vaccination, the averagenumber of animals that will become infected by theprimary case is R0 � (p � R0 ).

In order to prevent transmission of infection, thenumber of secondary cases resulting from the pri-mary case will need to be less than 1, or R0 � (p �R0 ) = <1. This is equivalent to R0 � 1 <(p � R0 ) orp >1 � 1/R0, which is the formula for vaccinationprotection (Giesecke, 2002). In other words, if R0 =4, then the proportion of the herd that will need to beeffectively immunized by vaccination (assuming thevaccine is 100% effective in immunizing an animalagainst the particular virus strain) to prevent trans-mission from acutely infected animals is 1 � 1/R0 =1 � 1/4, or 75%. If the R0 for the strain causing theacute infections was 2, rather than 4, only 50% ofthe herd would need to be immunized.

TRANSMISSION UNDER THE INFLUENCE OFHERD IMMUNITY

Herd immunity implies that the overall immunity ofthe herd is greater than the sum of immunity enjoyed


by individual animals in the herd. In other words, anonimmune, completely susceptible animal enjoyssome immunity from infection if it is in a herd withanimals that possess immunity, where immunity mayhave been acquired by natural infection with a fieldstrain, through vaccination, or from colostral anti-bodies. Several factors can contribute to herd immu-nity. Transmission within a herd or group of animalswith a high proportion of immune animals is slower,as shown above using R0 or II, than for herds with alow proportion of immune animals. Thus, a suscepti-ble animal in a herd with a high proportion of im-mune animals will have a lower risk of infection thanif it were in a herd with a low proportion of immuneanimals. Individual animal immunity also can affectthe dynamics of transition from one disease state(i.e., latent, infectious, noninfectious) to the otherand the infectious dose needed to produce infection.For example, animals with little or no immunity canbe expected to have a shorter latent period and to re-quire a lower dose of virus to become infected, com-pared with animals with immunity.

Although specific data are not available forBVDV herd immunity, some generalizations arepossible by applying what is known about the infec-tiousness of the disease. For example, depending onstrain variation, cell-mediated or humoral immunityacquired from exposure to one strain of BVDV mayreduce the amount of virus shed and/or duration ofshedding, and thus reduce ß and d, if the animal isinfected with a different strain of the virus (Howardet al., 1989; Brock et al., 1998: Grooms et al., 2001).Generally, immunity to most infectious agents is notlikely to be absolute; infection may be acquired byanimals with some marginal or incomplete spectrumof immunity if exposed to a sufficiently large doseof the agent. This is particularly true in the case ofBVDV because of the presence of high antigenicdiversity among BVDV strains (Dubovi, 1992;Ridpath, 2003). Low levels of immunity might pro-tect against low doses of a reasonably homologousstrain, but higher levels of immunity would be re-quired to protect against high doses of the virus, par-ticularly if it is a heterologous strain. If an animalwith partial immunity becomes infected, however, itmay not necessarily shed as much virus and not foras long, when compared to a nonimmune animal(Howard et al., 1989; Brock et al., 1998; Grooms etal., 2001). Generally, acute BVDV infection of ani-mals with some immunity can be expected to mani-fest in shorter periods of shedding and lower dosesof the agent being shed, which subsequently wouldreduce the probability of transmission to a fully sus-

ceptible animal. The extent to which viral sheddingis reduced in PI animals with SN antibodies is notknown. However, the transient cessation of viremiain some PI animals after consuming BVDV colostralantibodies or with active SN antibodies could sug-gest that viral shedding might also be reduced, atleast temporarily.

When herd immunity is present, virus transmis-sion is less likely. This is because herd immunity re-sults in a reduction in the number of animals that aresusceptible to viral infection. Thus, the probabilitythat the agent will be transmitted to a susceptible an-imal during an “adequate” contact is reduced, as in-dicated by a diminished value for ß in the equationsabove for R0 and incidence. In addition, the reducedduration of shedding (d), or reduced prevalence ofinfectious animals (P), will further diminish theforce of transmission through a herd, as indicated bythe equations for R0, and the diminished incidencewould not only reduce disease caused by BVDV butwould also reduce virus transmission by limiting theamount of virus shed from infected animals.

HORIZONTAL AND VERTICALTRANSMISSION OF BVDVTransmission of BVDV can occur vertically, result-ing in congenital infection of the fetus, or horizon-tally after birth, sometimes referred to as postnataltransmission. Transmission of BVDV to the fetusbefore 120–150 days in gestation may result in abor-tion, resorbtion, or stillbirth. Surviving fetuses willbe born as BVDV-infected calves. The BVDV infec-tion will persist for the life of the animal and will re-sult, presumably, in the continuous shedding of thevirus into the environment. Thus, PI animals are be-lieved to serve as a reservoir of BVDV. In uterotransmission after about 120–150 days of gestation,when fetus becomes immune-competent, will resultin a congenital infection characterized by abortion,stillbirth, congenital defects, or the birth of a live,normal-appearing calf. Horizontal (postnatal) trans-mission results in an acute infection sometimes re-ferred to as transient infection (TI), varying in sever-ity from inapparent or mild signs to life-threateningdisease. Although animals that have acquired anacute BVDV infection may remain infected forsome time, as indicated by prolonged duration of SNantibodies for a year or more, it is generally believedthat they shed the virus only transiently for a weekor less. Unfortunately, the requisite long-term dataare not available to confirm or deny subsequent re-crudescence of shedding. Several reviews of BVDVhave addressed specific aspects of BVDV infection

Virus Transmission 93


(Duffell and Harkness, 1985; Radostits and Little-johns, 1988; Harkness, 1987; Brownlie, 1990;Houe, 1995; Radostits et al., 1999).

CONGENITAL TRANSMISSION RESULTING IN PI There are two ways in which a fetus can acquire apersistent infection. One is by transmission of thevirus from a PI cow to her fetus, which very likelywill result in PI calf. The other way is by acute infec-tion of the dam in the first 120–150 days of gestationduring which it can transmit infection to the fetus.The rate at which PI animals are created in a herd de-pends on the prevalence of PI among pregnant cowsand on the proportion of cows acquiring an acuteBVDV infection during the first 120–150 days ofgestation. The ability of BVDV to infect embryosprior to attachment may be limited, as suggested bystudies showing only one of six fetuses exposed be-fore 60 days developed into PI calves (Moerman etal., 1993). In contrast the generation rate of PI ani-mals was significantly higher when fetuses were ex-posed after 45–60 days gestation (Liess et al., 1984).

CONGENITAL TRANSMISSION AFTER120–150 DAYS IN GESTATION

Little information exists on the rates of congenitaltransmission in fetuses after they become immune-competent. Two large studies of midwestern slaugh-terhouses found that 20% of the fetuses from culledcows had evidence of BVDV infection (Bolin et al.,1991b; Bolin and Ridpath, 1998), some of which ap-peared to be attributable to cytopathic vaccinestrains. As shown in a recent field study of largedairies, congenital transmission after 120–150 daysgestation may occur in 10% or more of vaccinateddairy cows (Muñoz-Zanzi et al., 2003a). This esti-mate represents only those calves that are born aliveand probably would be considerably higher if con-genital infections in aborted fetuses and stillborncalves are also included. Congenital transmission ofBVDV after 120–150 days in gestation, which is in-dicated by the presence of precolostral BVDV-neu-tralizing antibodies in calves at birth, has recentlytaken on new importance. Calves congenitally in-fected with BVDV were more likely to develop se-vere diseases in the first 4–5 months of life (Muñoz-Zanzi et al., 2003a) and to experience breedingdifficulties as heifers (Muñoz-Zanzi et al., 2003b).

HORIZONTAL TRANSMISSION AND ACUTEINFECTION

Acute infection results from horizontal transmissionof the virus to a susceptible animal following ade-

quate contact either with an acutely infected animalthat is in the infectious (shedding) phase of the dis-ease or with a PI animal that continuously sheds thevirus. Acute infection can be acquired through vari-ous routes of transmission, including direct nose-to-nose contact, fomites contaminated with the virus,and vaccination with a modified live virus (MLV)(Duffell and Harkness, 1985; Harkness, 1987;Brownlie, 1990; Houe, 1995; Radostits and Little-johns, 1988; Radostits et al., 1999).

ROUTES AND MEANS OF BVDVTRANSMISSIONGenerally, one can expect BVDV to be shed in mostexcretions and secretions, including tears, milk,saliva, urine, feces, nasal discharge, and semen ofacutely infected animals during the transient, infec-tious phase of the disease, as well as those of mostif not all PI animals (Houe, 1995). Some key routesof transmission pertaining to current managementconditions are discussed in the following sections.

TRANSMISSION VIA SEMEN

Semen from PI and acutely infected bulls containslarge amounts of BVDV, although semen fromacutely infected bulls typically would have muchless virus than semen from PI bulls (Brock et al.,1991; Voges et al., 1998). Transmission of BVDVvia semen can result in an acute infection or in thedevelopment of PI, or both. The potential for estab-lishment of persistent infection in calves sired by aPI bull was illustrated in a study in which all 12seronegative heifers developed acute BVDV infec-tion following insemination with semen from a PIbull (Meyling and Mikel Jensen, 1988). One of the12 calves (8%) developed persistent infection, indi-cating a significant potential not only for acutetransmission of BVDV from PI bulls but also for theestablishment of PI animals. Even though only 1 of12 calves developed PI, a transmission rate of 8%would suggest that natural breeding by PI bullscould contribute significantly to an accumulation ofPI animals in large herds.

The risk of calves developing persistent infectionmay be high in herds using a large number of bullsin natural breeding, particularly in large dairy herdswhere each bull may service several hundred cowsper year and in herds with a significant proportion ofsusceptible, antibody-negative females. In one case,a single PI bull was found to have transmittedBVDV to 55 cows, resulting in at least 2 PI animals,and was responsible for spreading BVDV to severalclean herds (Kirkland et al., 1994). In contrast, in-



semination of 73 heifers with semen of a bull withacute BVDV infection resulted in virus transmissionto only 3 of the animals, but a second cycle of trans-mission appearing 29 days after insemination re-sulted in 8 of 15 of the heifers that were subse-quently followed becoming infected.

Bulls that acquire an acute BVDV infection alsocan shed the virus in the semen for several days aftercessation of viremia, perhaps up to 14 days after in-fection, even though semen quality may remain nor-mal (Kirkland et al., 1991; Kirkland et al., 1994).The much lower amount of virus found in the semenof acutely infected bulls, as compared with that in PIbulls, and the low rate of shedding of BVDV inacutely infected bulls (Whitmore et al., 1978),would suggest a low risk of BVDV transmission viasemen from acutely infected bulls, compared withthat from PI bulls. In one reported case, however,BVDV was shed persistently for at least 11 monthsin an apparently acutely infected young bull (Vogeset al., 1998), suggesting that some acutely infectedbulls may pose the same threat as PI bulls. Thus, useof semen from bulls with acute BVDV infectioncould pose a substantial risk of transmission to sus-ceptible cows, with possible development of PI intheir fetuses, as previously illustrated (Kirkland, etal., 1997).

The high prevalence of PI (0.8%) found amongwell-screened bulls at artificial insemination centers(Howard et al., 1990) and the high rate of PI in em-bryo transfer (ET) animals (>2%) (Hietala et al.,2000) suggest that bulls selected for breeding, par-ticularly ET bulls, can be expected to have a higherprobability of PI than the general population of cat-tle. Coupled with the high rate of contact with fe-males, therefore, bulls can pose a significant risk ofinfection, either as a PI animal or temporarily as anacutely infected animal. Recommendations for re-ducing bull-related transmission include routinetesting of bulls for PI, testing the semen for virus,and vaccination (Brock et al., 1991; Voges et al.,1998).

TRANSMISSION VIA EMBRYO TRANSFER

Embryo transfer can constitute an important meansof BVDV transmission to valuable genetic stock.The virus can be transmitted to the fetus if the donorhas acute or persistent infection and the embryo isnot adequately washed, if the recipient has acute in-fection or PI, or if the fetal calf serum (FCS) used towash the embryo contains BVDV. The replicationand persistence of BVDV in in vitro embryonic pro-duction systems appears to be strain-specific

(Givens et al., 2000). Large amounts of virus havebeen found in serum (106 TCID50/ml), vaginal mu-cous (106 TCID50/ml), urine (104 TCID50/ml), uter-ine flush medium (102 TCID50/ml), and feces (104

TCID50/ml) of a PI donor heifer (Brock et al., 1991).In addition, several studies have documented BVDVcontamination of commercial FCS. In one study,10% of irradiated lots and 62% of nonirradiated lotsof FCS were contaminated. Heat inactivation ofserum at 56°C for 30 min was not considered to bea reliable method to eliminate BVDV from sera(Rossi et al., 1980). The virus also was found in20.6% of slaughterhouse fetuses, which constitutethe main source of FCS (Bolin et al., 1991b). BVDVwas found in 20.3% of 1000 lots of FCS harvestedby a commercial supplier from an abattoir that killedbeef and dairy cattle from three states (Bolin andRidpath, 1998). Thus, even if donors and recipientsare not infected with BVDV, use of untested FCSmay result in transmission of BVDV to an embryoand establishment of PI. Embryo transfer also canresult in back-transmission of BVDV from an in-fected or contaminated embryo to a susceptible re-cipient animal. It has been proposed that maternalfamilies of animals with PI may be established if fu-ture breeder animals acquire PI as a result of BVDVtransmission through embryo transfer (Brock et al.,1991).

TRANSRECTAL TRANSMISSION

Rectal transmission of BVDV was demonstratedfollowing the palpation of eight seronegativeheifers, all of which seroconverted after a single rec-tal palpation of the uterus and ovaries, using thesame glove that had been used to palpate an animalwith PI. Virus was isolated from five of the heifers,and all eight showed mild signs of clinical BVD.The results suggest that transmission to seronegativeanimals following rectal palpation with the sameglove as used for a PI animal could be an importantmeans of transmitting BVDV (Lang-Ree et al.,1994), and of establishing PI if cows are palpatedbefore 120 days in gestation.

IATROGENIC, FOMITE, ENVIRONMENTAL,AND INSECT TRANSMISSION

The BVD virus can survive in a cool, protected en-vironment for several days or even weeks (Houe,1995). Consequently, fomite or iatrogenic transmis-sion can occur if susceptible animals are exposed tofeed or equipment (e.g., nose tongs, halters, milkbottle nipples, balling guns, etc.) previously used onanimals with PI or acute infection (Radostits and



Littlejohns, 1988; Gunn, 1993; Houe 1995; Rado-stits et al., 1999). Transmission to seronegative ani-mals also is possible by reusing the same 19 gaugeneedle within a few minutes after it has been used ona PI animal, but viable virus may remain in the nee-dle for at least 3 days (Gunn, 1993). Environmentalcontamination of shared calving pens by uterine dis-charges from PI or acutely infected cows may createa common source of exposure for other cows and fornewborn calves that lack colostral protection(Radostits and Littlejohns, 1988).

Environmental conditions that favor crowding andaerosol transmission would increase the likelihoodof BVDV transmission from acutely infected,coughing calves with a respiratory form of BVDVinfection, particularly that caused by BVDV type 1b(Baule et al., 2001). It also is possible for flies tomechanically transmit the virus to susceptible ani-mals. The BVD virus was isolated from face flies(Musca autumnalis) feeding on the nose and face ofa PI animal (Gunn, 1993). In another study, stableflies (Stomoxys calcitrans), horseflies (Haemato-pota pluvialis), and head flies (Hydrotaea irritans)force-fed for 10–15 minutes on a PI animal, wereable to transmit infection to susceptible animals(Tarry et al., 1991).

TRANSMISSION VIA MILK AND COLOSTRUM

Milk from PI cows can be expected to containBVDV and thus will serve as a source of virus forcalves fed unpasteurized waste milk from sick ortreated cows, as is practiced commonly on largecalf-raising operations. Colostrum from cows withpersistent or acute infection may or may not containinfectious BVDV, depending on whether cross-reacting neutralizing antibodies are present.

TRANSMISSION VIA MODIFIED LIVE VIRUSVACCINE

Iatrogenic transmission of BVDV to the fetus, re-sulting in CNS lesions, teratogenic defects, anddeath, can occur if susceptible females are vacci-nated with a MLV BVDV vaccine during pregnancy(Orban et al., 1983; Liess et al., 1984). A survey offetuses harvested from midwestern slaughterhousesfound cytopathic strains of BVDV identical to thoseused in some MLV vaccines (Bolin et al., 1991b),suggesting that fetal infection following BVDVMLV vaccination may not be uncommon. Vaccina-tion with MLV BVDV also can be a highly efficientmeans of transmitting BVDV to the fetus and pro-ducing animals with persistent infection. An experi-mental study of congenital infection following MLV

vaccination found evidence of PI in all live-borncalves of 9 seronegative cows vaccinated before 120days in gestation, while fetuses of 12 other vacci-nated seronegative cows were aborted, had congeni-tal defects, or were stillborn (Liess et al., 1984).

Shedding of virus after vaccination can be mini-mal for some strains. In one study, a transientviremia lasted 3–10 days after vaccination, but noshedding was detected in nasal secretions or by ex-posing susceptible contact animals (Kleiboeker etal., 2003). In contrast, PI calves infected withBVDV type 1b and given a BVDV type 1a MLVvaccine shed the Ia vaccine strain of virus in nasalsecretions for up to 28 days after vaccination (Fultonet al., 2003). These findings suggest that transientshedding of vaccine strains is possible and may rep-resent a means of secondary transmission to preg-nant animals in contact with vaccinated animals.

TRANSMISSION OF BVDV FROMINFECTED ANIMALSContagiousness, or the degree to which BVDV-infected animals can transmit infection followingcontact with a susceptible animal, can depend onmultiple factors, which may include type of infec-tion (PI or acute), the stage of an acute infection,presence of neutralizing antibodies, and the strain ofvirus and its associated virulence.

TRANSMISSION FROM PI ANIMALS

PI animals can be assumed to shed large amounts ofBVDV in secretions and excretions most, if not all,of the time. Pregnant PI cows invariably transmit thevirus to the fetus in the first 120–150 days of gesta-tion, resulting in persistent infection of the fetus.Bulls with persistent infection will shed largeamounts of the virus in their semen, while maintain-ing acceptable semen quality. An understanding thatPI is transmitted from generation to generationalong maternal lines (Radostits and Littlejohn,1988; Brock et al., 1991) can be used in identifica-tion of PI animals by testing the dam, daughters, andsons of PI females.

In contrast to acute infection, PI animals are be-lieved to continuously shed large amounts of virus,thereby posing a constant threat of exposure to sus-ceptible animals. However, considerable data haveaccumulated to indicate that viremia in PI animalswith SN antibodies that cross-react with the PI strainwill cease for some period of time until antibodieswane (Howard et al., 1989; Palfi et al., 1993; Brocket al., 1998; Grooms et al., 2001). Although data onvirus shedding are not readily available, the cessation



of BVDV viremia suggests that the amount of virusshed in secretions and excretions and the duration ofshedding might also be diminished for some period.The existence of some PI animals that could remainundetectable through intermittent or low sheddingwould be a logical explanation for the unexpectedlyprolonged and continued transmission in herdsthought not to have any PI animals, and for whichtransmission was thought to be solely from acutelyinfected animals (Barber et al., 1985; Moerman etal., 1993). In light of these findings and the intermit-tent viremia in PI animals (Brock et al., 1998), itwould be prudent for programs aimed at detecting PIanimals and at reducing BVDV transmission to con-sider the possibility that PI animals may shed thevirus intermittently, rather than continuously.

TRANSMISSION FROM ACUTELY INFECTEDANIMALS

The amount of virus shed and the duration of shed-ding by acutely infected animals very likely dependon the virulence of particular strains of BVDV andtheir propensity to replicate (Bolin and Ridpath,1992), as well as on the presence and repertoire ofSN antibodies (Bolin et al., 1991a). Animals in-fected acutely are believed to shed the virus briefly,perhaps for only a few days or weeks, depending onthe strain of the virus (Duffell and Harkness, 1985;Niskanen et al., 2000). Under field conditions,where the vast majority of acute infections by in-digenous strains in a herd are mild and subclinical,the duration of shedding may be only 1–2 days, orless. Failure to detect viremia by alternate-day PCRtesting in unaffected dairy calves during a 4-weekperiod preceding seroconversion suggests thatviremia and shedding in acute infections can be veryshort-lived, perhaps less than 24 hours in calves ex-periencing no untoward signs of infection (Thur-mond, unpublished observations).

In contrast to endemic BVDV infection and trans-mission of indigenous strains among cattle in a herd,the severity of disease and duration of shedding havebeen observed to be much different following theaddition of new cattle to a herd. Clinical disease hasbeen more severe and the virus has been shed forlonger periods both in newly purchased, out-of-statecattle following week-long exposure to their newherdmates and in some of the herdmates in contactwith the new replacements (Thurmond, unpublishedobservations). A possible explanation is that the an-imals became infected with new strains that weresufficiently different from those that directed theirrepertoire of cell-mediated and SN antibody immu-

nity such that the infecting viruses were able to pro-duce severe disease, with viral shedding extendingfor several days or weeks. Similar observations weremade in an experimental study where acute infec-tion with virulent strains producing severe clinicaldisease resulted in more extensive shedding of virusfor well over a week, compared to infection withstrains that caused only mild disease (Kelling et al.,2002). Thus, the potential for shedding and trans-mission from acutely infected animals should be ex-pected to vary considerably, depending on cross-protection acquired from other exposures, includingvaccines (Bolin et al., 1991a), and the virulence ofthe virus.

As noted elsewhere (Brownlie, 1990), little atten-tion has been given to the presumed latency of acuteBVDV infections. A prevailing assumption has beenthat acute infections result only in transient shed-ding of the virus, perhaps as long as 14–21 days(Duffell and Harkness, 1985), but more typically be-tween 1–2 days or less than a week, with no recrude-scence of virus later on. Unfortunately, there are noreports of the requisite long-term follow-up studiesnecessary to confirm or deny such an assumption.The fact that SN titers persist long after viremiawanes indicates the virus is still replicating andbeing presented to the lymphoid tissue. Althoughseemingly unlikely, given the effect of SN antibod-ies in suppressing viral shedding, exacerbation of anacute infection with renewed shedding, perhapsfrom sequestered, protected sites, could explaintransmission cycles and unusually long time (2–3years) for all animals in some small herds to becomeinfected (Barber et al., 1985; Moerman et al., 1993).

It is not known with confidence whether, undercertain conditions, the latency of acute infectionsthat follows an initial transient shedding can be bro-ken, allowing the virus to be shed again. For exam-ple, in one study, a young bull that had acquired anacute BVDV infection, perhaps at 10–11 months ofage when the blood testicle barrier was being devel-oped, was observed to persistently shed the virus inhis semen for at least 11 months thereafter. Viruscould not be detected in serum, buffy coat, or otherorgans except the testicles, but high SN antibodytiters were maintained to the homologous virus(Voges et al., 1998). The fact that this case resem-bles an acute infection, rather than persistent infec-tion, would suggest the possibility that acutely in-fected bulls could persistently shed virus in thesemen and yet maintain high antibody titers with nodetectable virus in the serum or buffy coat. The au-thors recommend that as part of routine screening of



bulls for BVDV, semen should be tested for the pres-ence of the virus.

The recurrence of shedding in acutely infected an-imals also may depend on the biotype of the virus.Intermittent nasal shedding of the virus was ob-served in animals experimentally infected with acytopathic strain (Fray et al., 1998). Others have de-tected cytopathic virus in tissues and nasal secre-tions, after clearance of the virus from the blood,while SN antibodies were still present (Baule et al.,2001). Authors of this study suggested that seques-tration of cyptopathic BVDV in acutely infected an-imals may present a source of virus for recurrentshedding.

EFFECT OF COLOSTRALANTIBODIES AND VACCINATION-INDUCED IMMUNITY ONTRANSMISSIONPassive or acquired immunity to BVDV can affecttransmission in two ways. One is by potentially re-ducing the amount or duration of shedding in ani-mals with an acute or PI infection, thus diminishingthe amount and duration of infectiousness, as dis-cussed above. Animals that shed lesser virus for ashorter period of time will enhance herd immunityby reducing the likelihood of susceptible animalsacquiring the infection if they contact an infectedanimal. The other way immunity reduces transmis-sion is by reducing the number of animals suscepti-ble to infection where susceptibility exists on a con-tinuous scale, rather than on an all-or-none scale.Animals with considerable immunity (i.e., high SNantibody titers to a diversity of strains) would bemuch less susceptible to a given infectious dose thananimals with lesser immunity (i.e., low SN titers andlimited antibody cross-reactivity). Immunity ofvarying degrees and duration can be acquired fromcolostral antibodies, vaccination, and infection withfield strains.

PREVENTION OF TRANSMISSION BYCOLOSTRAL ANTIBODIES

Colostral antibodies can offer protection fromBVDV infection for the first 2–4 months of life, de-pending on the quality and quantity of colostrumconsumed, after which antibodies decay to levelsthat may no longer protect against infection. A largestudy of calves on two dairies practicing goodcolostrum management estimated that SN colostralantibody titers of half of the calves would decay to<1:16 by about 110 days of age for type 1 and 80days of age for type 2 BVDV (Muñoz-Zanzi et al.,

2002). The rate of antibody decay determines therate at which calves become susceptible to infection,and also to MLV BVDV vaccination, where SN an-tibody titers of 1:16–1:32 have been considered tooffer protection against clinical signs associatedwith BVDV infection (Bolin and Ridpath, 1992;Howard et al., 1989).

Colostral protection depends on the quality of thecolostrum management program, including ingestionof first milking colostrum with a high specific grav-ity from multiparous cows immediately after birth.Cows contributing colostrum should be well immu-nized against BVDV, preferably with both type 1 andtype 2 strains. Because BVDV colostral antibodiescan neutralize the BVDV in a MLV vaccine (Ellis etal., 2001), and thus diminish vaccine efficacy, it isnecessary to know at what age colostral BVDV anti-bodies can be expected to decay to <1:16. At thisage, calves become more susceptible both to infec-tion with BVDV and to an effective immunizationwith a MLV vaccine. Because colostral antibodies ofall calves do not decay at the same rate, the age ofsusceptibility that provides an early window for vac-cination may vary by a month or more. Conse-quently, multiple vaccinations may be needed; anearly vaccination to protect calves with little or nocolostral antibody and a later vaccination for calvesin which antibodies decay more slowly. Coupledwith timing of vaccination to avoid colostral antibod-ies that may negate MLV vaccination is strategic vac-cination before calves would begin to experience arisk of natural infection from herdmates.

The period of maximum transmission will varyfrom herd to herd, depending on the prevalence of PI calves and stocking rates. Transmission of BVDVin dairy calves in large, intensive dry-lot operationscan begin at about 3 months of age and peak at about7 months of age (Rush et al., 2001). Consequently,under conditions that foster such a high risk ofinfection, a narrow window for vaccination existsbetween the age when colostral antibodies decay to <1:16 and when the risk of infection begins toaccelerate.

PREVENTION OF TRANSMISSION BYVACCINATION

Most BVDV vaccination studies have targeted athow well vaccination might prevent or reduce dis-ease caused by BVDV following experimental chal-lenge with large doses of virus (van Oirschot et al.,1999) and not at how well vaccination might preventinfection. Because of the diversity of BVDV strains(Dubovi, 1992; Bolin et al., 1991a), the success of



vaccination aimed at increasing immunity and re-ducing the risk of infection and subsequent trans-mission will likely depend on the homology be-tween the vaccine strain and the field strains presentin the herd. The more homologous the strains, themore cross-protection will be achieved. One ap-proach in using MLV vaccines is to preempt naturalinfections (Harkness, 1987), thereby controlling thetiming of BVDV infection to strategically maximizeimmunity during critical periods of transmission. Asdiscussed above, strategic vaccination of intensivelymanaged calves attempts to induce an active immu-nity before transmission begins to increase at 3–4months of age, but after colostral antibodies havedecayed sufficiently so as not to neutralize the vac-cine virus. Strategic vaccination also may be indi-cated before puberty and before heifers are bred toreduce conception failures associated with infectionwith field strains of BVDV (Muñoz-Zanzi et al.,2003b).

Another example of strategic vaccination is thecommon practice involving prebreeding vaccinationwith MLV vaccines or vaccination of pregnant cowsearly in gestation with killed vaccines. These strate-gies are intended to increase immunity, mainly byincreasing the titer of SN antibodies, of the damsagainst acute infection and transmission of BVDVto the fetus before 120–150 days in gestation. Thestrategy of prebreeding vaccination with a MLVvaccine may reduce the risk of acquiring an acuteinfection in early pregnancy and subsequent devel-opment of PI in the fetus, depending on cross-protection between the strain of virus used in thevaccine and the wild strains circulating in the herd.However, complete protection against transplacentaltransmission to the fetus of cows given an MLV vac-cine prebreeding may not be a realistic expectation.In one study, 2 of 12 (17%) antibody-negativeheifers vaccinated 30 days before breeding with aMLV type 1 strain (NADL) gave birth to a PI calfafter challenge exposure with a type 1 strain be-tween 70 and 75 days in gestation (Cortese et al.,1998). In a similar study, PI was detected in 8 of 19fetuses (42%) from cows vaccinated 45 days beforebreeding with a MLV type 1 strain (NADL) andchallenged with a type 2 strain at 75 days of gesta-tion (Brock and Cortese, 2001). The potential diver-sity and subsequent failure of cross-reactivity withingenotypes also was observed whereby one or moreof six calves vaccinated with either of two commer-cial vaccines, each containing a different BVDVtype 1 strain, produced detectable antibodies against10 cytopathic and 10 noncytopathic BVDV isolates,

but no calf produced antibodies to all 20 viruses(Bolin and Ridpath, 1989).

Further evidence for incomplete protection of thefetus by prebreeding vaccination was presented in astudy in which transmission resulting in PI was notprevented in all cows vaccinated 3 weeks beforebreeding with a MLV type 1 vaccine; 7–15% of thefetuses of vaccinated cows challenged between 55and 100 days in gestation with a selected heterolo-gous type 1 strain developed PI (Dean et al, 2003).Use of a killed vaccine 2–3 weeks before breeding,also failed to prevent transmission of BVDV fromcows to their fetus and establishment of PI (Bolin etal., 1991a). However, three reduced doses of killedvaccine prevented BVDV infection by respiratorychallenge with the same vaccine strain (Howard etal., 1994), suggesting that killed vaccines may havea role in reducing transmission for some strains.

Further evidence for incomplete protectionagainst congenital transmission was found in studiesof newborn dairy calves from cows vaccinated withBVDV type 1 MLV vaccine. The study found that10% of calves born alive had acquired a congenitalBVDV infection, including infection with BVDVtype 1 (Muñoz-Zanzi et al., 2003a), suggesting thatconsiderable transplacental transmission can stilltake place in cows vaccinated once a year prior tobreeding. These studies also illustrate the potentialfor genetic diversity of BVDV (Dubovi, 1992) forvirus strains within a genotype and especially forstrains with different genotypes (Ridpath, 2003).One should expect, therefore, that a vaccine contain-ing only a type 1 strain of BVDV, for example,should not be expected to offer complete protectionof the fetus against infection with other type 1strains or necessarily any protection against infec-tion with type 2 strains of the virus.

In a field study of natural horizontal transmissionof wild strains of BVDV to dairy calves, vaccinationwith a MLV BVDV type 1 vaccine strain was esti-mated to prevent only 48% of infection with type 1strains and effectiveness lasted only about 60 days.It is likely that presence of colostral antibodies wasresponsible for some of the incomplete protection.The vaccine, however, did not offer any detectablereduction in risk of infection with type 2 strains(Thurmond et al., 2001). In a similar study on an-other herd (Thurmond and Hietala, 2002), no differ-ence in infection rates of BVDV type 1 was foundbetween calves given a MLV vaccine containingboth type 1 and type 2 strains and calves given aMLV vaccine containing only a type 1 strain. How-ever, a significantly higher proportion of calves



given a vaccine with only one strain of BVDV type1 became infected with BVDV type 2 than of thosegiven the vaccine with both type 1 and type 2strains. The efficacy in protecting against infectionwith BVDV type 2 was 64% for the vaccine withboth type 1 and type 2 strains, compared with noprotection offered by the vaccine with only a type 1strain.

Field studies following the SN antibody titers ofanimals found considerable evidence for dual geno-type natural infections, in which animals had evi-dence of infection with both type 1 and type 2strains (Rush et al., 2001; Muñoz-Zanzi et al., 2002;Muñoz-Zanzi et al., 2003a), indicating that naturalinfection with one genotype does not necessarilyprotect against natural infection with the other geno-type. Currently, most BVDV type 1 vaccines used inNorth America contain only type 1a strains, with theexception of one killed vaccine (Fulton et al., 2002).No vaccine trials have been reported that indicatethe extent to which vaccines containing BVDV type1a strains will protect against infection with type 1bstrains, which have been found to be a predominantsubtype in calves with respiratory disease (Fulton etal., 2002). These findings indicate that, under typi-cal field conditions, the use of MLV vaccine withonly a type 1 strain should not be expected to offercross-protection against infection with type 2strains, and suggest that vaccines employed to re-duce transmission of both type 1 and type 2 strainsof BVDV should contain both type 1 and type 2strains.

INTRA-HERD TRANSMISSIONThe rate at which BVDV is spread within a herd de-pends on the prevalence of PI animals, the rate ofanimal-to-animal contacts, the virulence of the virusstrain(s), and the susceptibility of the cattle to newand indigenous strains in the herd. In studies ofsmall, susceptible (unvaccinated) dairy herds withlow density management involving seasonal graz-ing, transmission was reported to be slow, requiring2–3 years for all adults to become infected (Barberet al., 1985), presumably in the absence of PI animalin the adult herd. In another similar herd, transmis-sion was cyclical, with temporary increases in infec-tion rates when acutely infected animals were mixedwith susceptible animals, but also showing variationin transmission rates for cattle exposed to animalswith PI (Moerman et al., 1993). Both studies con-cluded that transmission could be sustained for 2 ormore years by shedding from acutely infected ani-mals only, in the absence of PI animals, but that

transmission also can proceed unexpectedly slowlyin a herd with PI animals. The R0 was estimated inthe latter study to be only 3.9.

In contrast, another study of transmission amonganimals in an extensive management system, with astocking density of 67 animals/km2, estimated R0 tobe 2.3 if no PI animals were detected and R0 = 35 ifthe prevalence of PI in the herd was 1.2% (Cherry etal., 1998), suggesting a greater than tenfold increasein the transmission and risk of infection attributableto the presence of PI animals. To achieve eradication(R = 1), the authors concluded that if R0 = 32, PI an-imals would have to be removed from the herd be-fore 11 days of age. In herds without PI, a 2–3-yearcycle was projected for fetal loss attributable to vari-ation in the rate of acute infections. For cattle man-aged under extensive grazing conditions, wherestocking density ranged from 0.2–1 animal/acre, notransmission was observed from animals with acuteinfection, but the incidence of transmission from PIcalves to cows ranged from 0.006–0.04 new casesper day (Radostits and Littlejohns, 1988).

Continual transmission over the course of 2–3years, without exposure to a PI animal, however,would seem to require a much longer period of shed-ding from acutely infected animals than has been re-ported in the literature (Duffell and Harkness, 1985;Niskanen et al., 2000). If transmission in these herdswas in fact due only to shedding by animals with anacute infection, the shedding period for acute infec-tion would have to be considerably longer than the1–3-week period typically quoted (Duffell andHarkness, 1985; Houe, 1995). The slow transmis-sion in the presence of PI animals in one of the herdsalso was incompatible with the prevailing dogmaabout the high, continual risk of transmission fromPI animals. One could reason that if a PI animal con-tinuously shed large amounts of virus, all the ani-mals in the herd, regardless of the animal density,would have come in contact with and been infectedby a PI animal within 2 years. An alternative expla-nation for such a slow transmission would be inter-mittent shedding, either by acutely infected animalsor by PI animals, as discussed above. Intermittentshedding by acutely infected or PI animals could ex-plain both the cyclicity in new cases and the slowrate of spread in such small herds.

Under intensive management conditions, such asfeedlot operations, in which animal density can beas high as 9.7 animals/100 m2 (Rush et al., 2001)and PI animals are invariably present, transmissionof BVDV can be much more rapid. For a dry-lotdairy herd practicing MLV BVDV vaccination, 97%



of the heifers showed evidence of infection by thetime they reached 13–14 months of age (Michel etal., 1993), and calves entering feedlots serocon-verted to BVDV at the rate of 0.4% (Taylor et al.,1995) to 0.7% per day (Martin and Bohac, 1986).The highest incidence of seroconversion on twolarge dry-lot dairy calf operations was 0.5% and1.3% per day for 7-month-old heifers (Rush et al.,2001). In another study, when a PI animal wasyarded with susceptible cows in that same herd, theincidence rate increased from 0.006–0.04 new casesper day to 0.6 new cases in less than 12 hrs (Rado-stits and Littlejohns, 1988).

INTER-HERD TRANSMISSIONConsiderable opportunity exists for transmission ofBVDV from herd to herd, depending on the specificherd biosecurity policies, as reviewed elsewhere(Harkness, 1987). Introduction of new animals intoa herd was found to be the main factor significantlyassociated with high herd seroprevalence of BVDV(Mainar-Jaime et al., 2001). Use of semen from aninfected bull also was observed to spread the virus toa number of closed, BVDV-negative herds (Kirklandet al., 1994). Other means of inter-herd transmissioninclude cross-fence contact with infectious animals,their secretions, or excretions; embryo transfer in-volving introduction of pregnant recipients carryinga fetus with PI, even if the recipient is uninfected;use of MLV BVDV vaccines; and feeding calveswaste or “hospital” milk from dairies with lactatingPI cows. Ruminants other than cattle can becomeinfected with BVDV (Nettleton, 1990); however,the risk of transmission from these animals is notknown.

MINIMIZING TRANSMISSION OFBVDV

REDUCING THE COEFFICIENT OFINFECTIOUSNESS (ß)One strategy in reducing transmission is to reducethe infectiousness of infected animals. The probabil-ity that an infected animal will transmit the infectionto a susceptible animal may be reduced, under someconditions, by maximizing the SN antibody titer, ei-ther through colostral antibodies or vaccination. Asdiscussed above, animals with SN antibodies at thetime they acquire an acute BVDV infection may beexpected to shed less virus, and for a shorter periodof time, than if they had no antibodies. The fact thatviremia in PI animals can cease in the presence ofSN antibodies (Howard et al., 1989; Palfi et al.,

1993; Brock et al., 1998; Grooms et al., 2001) sug-gests that shedding also might be diminished fol-lowing vaccination or ingestion of colostral antibod-ies. Certainly the main means of diminishing theoverall herd infectiousness would be by removal ofPI animals because PI animals shed considerablymore virus than acutely infected animals.

REDUCING THE LIKELIHOOD OF ADEQUATECONTACT

As shown by the equations for R0 and incidence ofBVDV infection, the rate of adequate animal-to-animal contact is very important in fostering trans-mission. In a prospective study (Rush et al., 2001),transmission was related not only to animal densitybut also to the absolute number of animals in a penor corral (stocking rate). A small number of animalsin a high-density corral or pasture would likely ex-perience a lower risk of infection than a large num-ber of animals in a low-density corral because a cor-ral with many animals is more likely to contain a PIanimal, regardless of the density. Strategies for con-trol of BVDV transmission include limiting contactbetween animals by housing calves in individualisolation hutches (Rush et al., 2001), decreasingdensity, increasing feed bunk space and watertroughs, reducing access to cross-fence sharing offeed and water troughs, double fencing to reducecross-fence physical contact, and reducing the num-ber of animals in a pasture or pen.

REDUCING THE DURATION OR PREVALENCEOF INFECTIOUS ANIMALS

Duration of infectiousness may be reduced inacutely and perhaps in persistently infected animalsby providing adequate SN antibody, as discussedabove, for reducing the coefficient of infectiousness.A fundamental strategy in controlling transmissionof BVDV is to identify and remove PI animals,thereby reducing the prevalence of infectious ani-mals in a herd.

REDUCING THE PROPORTION OFSUSCEPTIBLE ANIMALS

Because transmission cannot proceed without sus-ceptible animals, a key concept in herd-basedcontrol of BVDV transmission is to maximize indi-vidual and herd immunity through high-qualitycolostrum management and strategic vaccination tomaximize the repertoire of SN antibody diversity,which are aimed at reducing the probability of ani-mals acquiring a BVDV infection at critical periodswhen BVDV infection could have important nega-



tive consequences. Simulation modeling suggeststhat as herd immunity to BVDV wanes, and moreand more animals become susceptible to infection,the number of PI animals will increase (Innocent etal. 1997).

SUMMARYIn summary, BVDV can be transmitted briefly byanimals following an acute infection and continu-ously by animals with a persistent BVDV infection.Although presumably rare, it is not known the extentto which virus shedding can be reactivated inacutely infected animals some time after an initialinfectious period has waned or by which sheddingcan be intermittent in PI animals. Animals in the in-fectious phase of a BVDV infection shed less virusthan PI animals, and R0 values for transmissionfrom acutely infected animals typically can be ex-pected to be in the order of magnitude of <5,whereas those for transmission from PI animals gen-erally would be >10–20. Consequently, vaccine effi-cacy and/or the proportion of animals immunized byvaccination must be higher to offer protectionagainst infection acquired from PI animals thanfrom acutely infected animals. Vaccine efficacyaimed at preventing infection is correlated with thedegree of homology between the strain of virus usedin the vaccine and the strain of virus being pre-vented. Vaccination may be ineffective in herds withfield strains sufficiently heterologous to the vaccinestrain to prevent development of adequate protectingcross-reacting SN antibodies. Vaccines employingonly one genotype of the virus should not be ex-pected to be efficacious in preventing infection witha strain of the other genotype.

As with other infectious diseases, the forces driv-ing transmission of BVDV in a susceptible popula-tion are the infectiousness of the virus (coefficient ofinfectiousness = ß), the contact rate between infec-tious and susceptible animals, the duration of infec-tiousness (or proportion of the population that is in-fectious), and the proportion of the population that issusceptible to infection. Each of these forces shouldbe addressed in programs to limit or control BVDVinfection. Vaccination is aimed generally at slowingor stopping transmission by promoting herd immu-nity; specifically, this is accomplished by reducingthe proportion of susceptible animals (increasing theproportion that are immune), and by shortening theshedding period (reducing the duration of infec-tiousness) and lowering the amount of virus shed(reducing ß) for animals that do become infected.

REFERENCESBarber DML, Nettleton PF, Herring JA: 1985, Disease

in a dairy herd associated with the introduction andspread of bovine virus diarrhoea virus. Vet Rec117:459–464.

Baule C, Kulcsár G, Belák K, et al.: 2001,Pathogenesis of primary respiratory disease inducedby isolates from a new genetic cluster of bovineviral diarrhea virus type 1. J ClinImmunol39:146–153.

Bolin SR, Littledike ET, Ridpath JF: 1991a, Serologicdetection and practical consequences of antigenicdiversity among bovine viral diarrhea viruses in avaccinate herd. Am J Vet Res 52:1033–1037

Bolin SR, Matthews P, Ridpath J: 1991b, Methods fordetection and frequency of contamination of fetalbovine serum with bovine viral diarrhea virus andantibodies against bovine viral diarrhea virus. J VetDiagn Invest 3:199–203.

Bolin SR, Ridpath JF: 1989, Specificity of neutraliz-ing and precipitating antibodies induced in healthycalves by monovalent modified-live bovine viral di-arrhea virus vaccines. Am J Vet Res 50:817–21.


Bolin S, Ridpath J: 1998, Prevalence of bovine viraldiarrhea virus genotypes and antibody against thoseviral genotypes in fetal bovine serum. J Vet DiagnInvest 10:135–139.

Brock KV, Cortese VS: 2001, Experimental fetal chal-lenge using type 2 bovine viral diarrhea virus incattle vaccinated with modified-live virus vaccine.Vet Therapeutics 2:354–360.


Brock KV, Redman DR, Vickers ML, Irvine NE:1991, Quantitation of bovine viral diarrhea virus inembryo transfer flush fluids collected from a per-sistently infected heifer. J Vet Diagn Invest3:99–100.

Brownlie J: 1990, The pathogenesis of bovine virusdiarrhoea virus infections. Rev Sci Tech Off Int Epiz9:43–59.

Cherry BR, Reeves MJ, Smith G: 1998, Evaluation ofbovine viral diarrhea virus control using a mathe-matical model of infection dynamics. Prev Vet Med33:91–108.

Cortese VS, Grooms DL, Ellis J, et al.: 1998,Protection of pregnant cattle and their fetusesagainst infection with bovine viral diarrhea virustype 1 by use of a modified-live virus vaccine. AmJ Vet Res 59:1409–1413.



Dean HJ, Hunsaker BD, Bailey OD, Wasmoen T:2003, Prevention of persistent infection in calves byvaccination of dams with noncytopathic type-1modified-live bovine viral diarrhea virus prior tobreeding. Am J Vet Res 64:530–537.

Dubovi EJ: 1992, Genetic diversity and BVD virus.Comp Immun Microbiol Inf Dis 15:155–162.

Duffell SJ, Harkness: 1985, Bovine virus diarrhoea-mucosal disease infection in cattle. Vet Rec117:240–245.

Ellis J, West K, Cortese V, et al.: 2001, Effect of ma-ternal antibodies on induction and persistence ofvaccine-induced immune responses against bovineviral diarrhea virus type 2 in young calves. J AmVet Med Assoc 219:351–356.

Fray MD, Clark MC, Thomas LH, et al.: 1998,Prolonged nasal shedding and viraemia of cyto-pathogenic bovine virus diarrhoea virus in experi-mental late-onset mucosal disease. Vet Rec143:608–611.

Fulton RW, Ridpath JF, Saliki JT, et al.: 2002, Bovineviral diarrhea virus (BVDV) 1b: predominantBVDV subtype 1n calves with respiratory disease.Can J Vet Res 66:181–190.

Fulton RW, Step DL, Ridpath JF, et al.: 2003,Response of calves persistently infected with non-cytopathic bovine viral diarrhea virus (BVDV) sub-type 1b after vaccination with heterologous BVDVstrains in modified live virus vaccines andMannheimia haemolytica bacterin-toxoid. Vaccine21:2980–2985.

Giesecke J: 2002, Modern Infectious DiseaseEpidemiology, p. 123. Arnold, New York, NY.

Givens MD, Galik PK, Riddell KP, et al.: 2000,Replication and persistence of different strains ofbovine viral diarrhea virus in an in vitro embryoproduction system. Theriogenology 54:1093–1107.

Grooms DL, Kaiser L, Walz PH, Baker JC: 2001,Study of cattle persistently infected with bovineviral diarrhea virus that lack detectable virus inserum. J Am Vet Med Assoc 219:629–631.

Gunn HM: 1993, Role of fomites and flies in trans-mission of bovine viral diarrhoea virus. Vet Rec132:584–585..

Haloran ME: 1998, Concepts of infectious diseaseepidemiology. In: Modern Epidemiology, 2nd ed.Eds. Rothman R, Greenland S, pp. 524–554.Lippincott-Raven, Philadelphia, PA.

Harkness JW: 1987, The control of bovine viral diar-rhoea virus infection. Ann Rech Vet 18:167–174.

Hietala SK, Muñoz-Zanzi C, Thurmond M: 2000,Herd screening for BVDV PI animals using pooledsample RT-PCR. Virology Committee Report, 43rdAnnual Meeting Am Assoc Vet Lab Diagn,Birmingham, AL.

Houe H: 1995, Epidemiology of bovine viral diarrheavirus. Vet Clin North Am: Food Anim Pract11:521–547.

Howard CJ, Clarke MC, Brownlie J: 1989, Protectionagainst respiratory infection with bovine virus diar-rhoea virus by passively acquired antibody. VetMicrobiol 19:195–203.

Howard CJ, Clarke MC, Sopp P, Brownlie J: 1994,Systemic vaccination with inactivated bovine virusdiarrhoea virus protects against respiratory chal-lenge. Vet Microbiol 42:171–179.

Howard TJ, Bean B, Hilman R, et al.: 1990,Surveillance for persistent bovine viral diarrheavirus viremia in four artificial insemination centers.J Am Vet Med Assoc 196:1951–1955.

Innocent G, Morrison I, Brownlie J, Gettinby G:1997, A computer simulation of the transmissiondynamics and the effects of duration of immunityand survival of persistently infected animals on thespread of bovine viral diarrhoea virus in dairy cat-tle. Epidemiol Infect 119:91–100.

Kelling CL, Steffen DJ, Topliff CL, et al.: 2002,Comparative virulence of isolates of bovine viraldiarrhea virus type 2 in experimentally inoculatedsix- to nine-month-old calves. Am J Vet Res63:1379–1384.

Kirkland PD, Makintosh SG, Moyle A: 1994, Theoutcome of widespread use of semen from a bullpersistently infected with pestivirus. Vet Rec135:527–529.

Kirkland PD, McGowan MR, Mackintosh SG, MoyleA: 1997, Insemintaion of cattle with semen from abull transiently infected with pestivirus. Vet Rec140:124–127.


Kleibocker SB, Lee S-M, Jones CA, Estes DM: 2003,Evaluation of shedding of bovine herpesvirus 1,bovine viral diarrhea virus 1, and bovine viral diar-rhea virus 2 after vaccination of calves with a mul-tivalent modified-live virus vaccine. J Am Vet MedAssoc 222:1399–1403.

Lang-Ree JR, Vatn T, Kommisrud E, Loken T: 1994,Transmission of bovine viral diarrhoea virus by rec-tal examination. Vet Rec 135:412–413

Liess B, Orban S, Frey HR et al: 1984, Studies ontransplacental transmissibility of a bovine virus di-arrhoea (BVD) vaccine virus in cattle. II.Inoculation of pregnant cows without detectableneutralizing antibodies to BVD virus 90–229 daysbefore parturition (51st to 190th day of gestation).Zbl Vet Med B 31:669–681.



Mainar-Jaime RC, Berzal-Herranz B, Arias P, Rojo-Válzquez FA: 2001, Epidemiological pattern andrisk factors associated with bovine viral-diarrhoeavirus (BVDV) infection in a non-vaccinated dairy-cattle population from the Asturias region of Spain.Prev Vet Med 52:63–73.

Martin SW, Bohac JG: 1986, The association betweenserological titers in infectious bovine virus, bovinevirus diarrhea virus, parainfluenza-3 virus, respira-tory syncytial virus and treatment for respiratorydisease in Ontario feedlot calves. Can J Vet Res50:351–358.

Meyling A, Mikel Jensen A: 1988, Transmission ofbovine virus diarrhoea virus (BVDV) by artificialinsemination (AI) with semen from a persistentlyinfected bull. Vet Microbiol 17:97–105.

Michel P, Thurmond M, Hietala S: 1993, Prevacci-nation bovine viral diarrhea virus titers and subse-quent reproductive performance in dairy heifers.Can J Vet Res 57:236–240.


Muñoz-Zanzi C, Hietala S, Thurmond M, Johnson W:2003a, Quantification, risk factors, and health im-pact of natural bovine viral diarrhea virus congeni-tal infection in dairy calves on two large Californiadairies. Am J Vet Res 64:358–365.

Muñoz-Zanzi C, Thurmond M, Hietala S: 2004, Effectof bovine viral diarrhea virus infection on fertility ofdairy heifers. Theriogenology 61:1085–1099.

Muñoz-Zanzi C, Thurmond M, Johnson W, Hietala S:2002, Predicted ages of dairy calves whencolostrum-derived bovine viral diarrhea virus anti-bodies would no longer offer protection against dis-ease or interfere with vaccination. J Am Vet MedAssoc 221:678–685.

Nettleton PF: 1990, Pestivirus infections in ruminantsother than cattle. Rev Sci Tech Off Int Epiz9:131–150.

Niskanen R, Lindberg A, Larsson B, Alenius S: 2000,Lack of virus transmission from bovine viral diar-rhoea virus infected calves to susceptible peers.Acta Vet Scand 41:93–99.

Orban S, Liess B, Hafez SM, et al.: 1983, Studies ontransplacental transmissibility of bovine virus diar-rhea (BVD) vaccine virus. I. Inoculation of preg-nant cows 15 to 90 days before parturition (190th to265th day of gestation). Zbl Vet Med B 30:619–634.

Palfi V, Houe H, Philipsen J: 1993, Studies on the de-cline of bovine virus diarrhoea virus (BVDV) ma-ternal antibodies and detectability of BVDV in per-sistently infected calves. Acta Vet Scand34:105–107.

Radostits OM, Gay CC, Blood DC, et al.: 1999,Veterinary Medicine, pp. 1085–1105. W.B.Saunders, San Francisco, CA.

Radostits OM, Littlejohns IR: 1988, New concepts inthe pathogenesis, diagnosis, and control of diseasescaused by the bovine viral diarrhea virus. Can Vet J29:513–528.

Ridpath JF: 2003, BVDV genotypes and biotypes:practical implications for diagnosis and control.Biologicals 31:127–131.

Rossi CR, Bridgman CR, Kiesel GK: 1980, Viral con-tamination of bovine fetal lung cultures and bovinefetal serum. Am J Vet Res 41:1680–1681.

Rush DM, Thurmond MC, Muñoz-Zanzi C, HietalaSK: 2001, Descriptive epidemiology of postnatalbovine viral diarrhea virus infection in intensivelymanaged dairy heifers. J Am Vet Med Assoc219:1426–1431.

Tarry DW, Bernal L, Edwards S: 1991, Transmissionof bovine virus diarrhoea virus by bloodsuckingflies. Vet Rec 128:82–84.

Taylor LF, Van Donkersgoed J, Dubovi EJ, et al.:1995, The prevalence of bovine viral diarrhea virusinfection in a population of feedlot calves in west-ern Canada. Can J Vet Res 59:87–93.

Thurmond M, Hietala SK: 2002, Field efficacy ofMLV vaccines in preventing BVDV infection. 1stInternational BVDV Conference, Ames, Iowa.

Thurmond MC, Muñoz-Zanzi CA, Hietala SK: 2001,Effect of calfood vaccination on transmission ofbovine viral diarrhea virus under typical drylotdairy conditions. J Am Vet Med Assoc 219:968–975.

van Oirschot JT, Bruschke CJM, van Rijn PA: 1999,Vaccination of cattle against bovine viral diarrhoeavirus. Vet Microbiol 64:169–183.

Voges H, Horner GW, Rowe S, Wellenberg GJ: 1998,Persistent bovine pestivirus infection localized inthe testes of an immuno-competent, non-viraemicbull. Vet Microbiol 61:165–175.

Whitmore HL, Gustafsson BK, Havareshti P, et al.:1978, Inoculation of bulls with bovine virus diar-rhea virus: Excretion of virus in semen and effectson semen quality. Theriogenology 9:153–163.



INTRODUCTIONUnderstanding the clinical manifestations of bovineviral diarrhea virus (BVDV) has been a continualchallenge for the veterinarian (Tremblay, 1996).Grooms et al. (2002) described five forms of clinicalBVDV: acute BVDV infection, severe acute BVDVinfection, hemorrhagic BVDV infection, acuteBVDV infection–bovine respiratory disease (BRD),and acute BVDV infection–immunosuppression. Asignificant percentage (70–90%) of BVDV infec-tions results in subclinical infections. However,whether the BVDV infection is one of the acuteforms or is subclinical in nature, there is a certainperiod of virus shedding. These transiently infected(TI) cattle can be a source of virus to susceptible ru-minants in the population (Cherry et al., 1998). Ifthe susceptible animal is pregnant, there is a risk offetal exposure resulting in early embryonic death,abortion, or congenital infections/defects (Mickel-sen and Evermann, 1994; Muñoz-Zanzi et al.,2003). An important consequence of congenital in-fection results from infection of the fetus between90 and 120 days of gestation. Infection of naivecows during this period results in the birth of calves,which are immunologically vulnerable to BVDV-specific induced tolerance (McClurkin et al., 1984).This tolerance results in a condition commonly re-ferred to as persistent infection (PI). The hallmarkof the BVDV PI is that while the animal may appearclinically normal, it has a prolonged viremia (inmost cases lifelong) during which it sheds largequantities of virus (Figure 6.1; Bolin, 1995).

This chapter focuses primarily on the clinical out-come of BVD viral infection, and in the process,poses key questions on the ecology and epidemiologyof BVDV. These questions will include the following:

• Is the clinical presentation of BVDV infectionchanging as new strains of virus emerge?

• Is the acute (transient) or chronic (persistent)phase of BVDV infection altered due to the useof BVDV vaccination?

• Is fetal infection with BVDV increasing ordecreasing?

We also review the course of initial BVDV infectionin naive immunocompetent cattle, recognition ofclinical symptoms associated with acute and persist-ent BVDV infections, duration and severity of clini-cal symptoms, and recovery from subclinical andclinical symptoms.

An important aspect of BVDV infection is inter-species transmission and the possible spread ofBVDV strains among small and large domestic ru-minants, as well as wildlife species such as deer, elk,and moose (Loken, 1995; Nettleton and Entrican,1995). This topic concludes with the feasibility ofBVDV control in populations, and whether eradi-cation is achievable at different levels of livestockproduction.

COURSE OF INITIAL INFECTIONIN NAIVE IMMUNOCOMPETENTCATTLE POPULATIONInfection of cattle with BVDV can result in a widespectrum of clinical manifestations, from impercep-tible subclinical infections to fulminant signs endingin death (Baker, 1995). The clinical response to in-fection is complex and depends on several host andagent factors (Ames, 1986). Host factors that influ-ence the outcome of clinical disease include im-munocompetance, immunotolerance, pregnancy sta-tus, gestational age of the fetus, the immune status(passive or active immunity from previous infectionor vaccination), and the level of environmentalstress (Baker, 1995). Viral factors influencing clini-cal outcome include genomic and antigenic diversityamong BVDV isolates (Archambault et al., 2000;

105

6Clinical Features

James F. Evermann and George M. Barrington



Bolin and Ridpath, 1996; Ridpath et al., 2000).Currently, two distinct genotypes (BVDV type 1 andBVDV type 2), as well as 11 subgenotypes (BVDVtypes 1a–1k) are recognized (Brownlie et al., 2000;Vilcek et al., 2001). Differences in virulence amongisolates have been recognized for over a decade(Bolin and Ridpath, 1992). Interestingly, recent epi-demiologic studies have revealed that the incidenceof infection with specific genotypes, as well as theclinical outcome after infection, have changed overtime (Evermann and Ridpath, 2002).

Bovine viral diarrhea virus is acquired primarilythrough aerosols that infect nasal mucosa. Theaerosols may become suspended in ambient air andretain infectivity for distances as short as 1.5 m andas far as 10 m (Niskanen and Lindberg, 2003).Direct nose-to-nose contact between an infected an-imal and a susceptible animal is considered to be themost effective route for BVDV transmission. Thereare reports of indirect spread of BVDV by use ofnose tongs and housing animals in recently (within2 hrs) contaminated pens (Niskanen and Lindberg,2003). The virus replicates in draining lymph nodesand from there spreads via circulating lymphoidcells to the blood (Figure 6.2). Viremia can be de-tected as early as 24 hours postinfection (Mills andLuginbuhl, 1968), and this viremic phase is closelyparalleled by the presence of virus in urine 48 hourspostinfection. The course of the viremia and thepresence of virus in the urine are dependent upon

the presence or absence of colostral antibodies.Colostral BVDV antibody is believed to persist forup to 6 months of age, which accounts for the higherincidence of BVDV-associated disease after thistime period (Baker, 1995; Bolin, 1995).

BVDV IN COLOSTRUM-DEPRIVED CALVES

In experimentally infected, colostrum-deprivedcalves there were diphasic pyrexia, leukopenia, ano-rexia, and various degrees of diarrhea (Mills andLuginbuhl, 1968). Infectious BVDV was isolatedfrom the spleen and thymus for at least 25 days post-infection (dpi). Lymph nodes draining the foresto-machs and intestines were positive for virus until 39dpi, while those of the duodenum were positive upto 56 dpi. Virus was also isolated from the trachea,lungs, and bronchial lymph nodes at 56 dpi. On thebasis of these studies, it was concluded that BVDVis carried in the respiratory tract and may constitutethe single most important epizootiologic factor inthe pathogenesis of the disease (Mills and Lugin-buhl, 1968). These prophetic words were supportedrepeatedly in subsequent reports on the clinicaldescriptions of BVDV (Callan and Garry, 2002;Grooms, 1999; Potgieter et al., 1984a).

BVDV IN 6-MONTH-OLD SERONEGATIVECALVES

A later study using 6-month-old calves reported clin-ical symptoms limited to a mild bilateral serous nasal

Figure 6.1. Schematic depicting the clinical and subclinical manifestations of BVDV infections and sequelae to con-genital infection with BVDV. Abbreviations used are bovine respiratory disease (BRD), immunosuppression (IS),transient infection (TI), and persistent infection (PI).


discharge at 12 dpi (Wilhelmsen et al., 1990).Although clinical signs observed with the BVDVstrain used were minimal, there were significant, im-munosuppressive effects of BVDV infection. A 25%or greater decrease in leukocyte count was observedthat lasted up to 12 days. In 6-month-old calves therewas more evidence of lymphoid involvement of thegastrointestinal tract than in the 2-month-oldcolostrum-deprived calves. There was gross swellingand edema of mesenteric lymph nodes and lympho-cytolysis and edema in the gut-associated lymphoidtissue. This makes it clearer that the following com-ments are a continuation of the proceeding discus-sion. Differences noted between BVDV infection of2-month-old and 6-month-old calves may have re-sulted from age differences, BVDV strain variation,or a combination of both (Pellerin et al., 1994).Nonetheless, it is apparent that BVDV is capable ofmultiple systemic effects resulting in different clini-cal manifestations. Grooms et al. (2002), appropri-ately wrote about the “diseases” caused by BVDVand indicated that there were at least six initial formsof BVDV infection in nonpregnant cattle and at leasttwo in pregnant cattle (Table 6.1).

BVDV IN PREGNANT CATTLE

The effects of BVDV on pregnant cattle are listed inTable 6.1. Early embryonic death and abortion maybe seen following BVDV infection of naive heifersand cows (Bolin, 1990b). Usually up to 85% of cat-tle become infected, undergo a transient viremia,and seroconvert prior to breeding age. However, this

leaves approximately 15% of the population at riskat the time of initial breeding. It is estimated thatBVDV causes about 6–10% of infectious abortionsin cattle (Figure 6.3; Dubovi, 1994). Although thisinfection may result in serious reproductive losses ina few herds, the predominant form of infection withBVDV is the congenital infection (Muñoz-Zanzi etal., 2003). The fetus is highly susceptible between45 and 125 days of gestation (Dubovi, 1994). Calvesinfected during this time are at high risk of develop-ing fetal abnormalities that may affect the brain(cerebellar hypoplasia), eyes (retinal atrophy,cataracts), growth retardation with arrested bone de-velopment, and pulmonary hypoplasia. Anotherform of congenital defect is immune tolerance, inwhich the developing fetus becomes infected with anoncytopathic strain (see Chapter 3) of BVDV be-tween 90 and 120 days (Figure 6.4). The immunesystem develops tolerance to the infecting strain ofBVDV. The result is that no immune response to thevirus is mounted by the fetal immune system and thefetus becomes persistently viremic or persistentlyinfected (PI). Figure 6.1 depicts the clinical mani-festations of BVDV, with particular attention on thepropagation of BVDV PI offspring (see Chapter 7).Following birth, the PI calf is vulnerable to develop-ing mucosal disease (MD) by one of several mecha-nisms (Figure 6.5).

The impact of different BVDV strains on repro-ductive performance, in particular age predilectionof the fetus for infection with different BVDV geno-types and subgenotypes, was recently reported

Clinical Features 107

Figure 6.2. Schematic depicting the spread of BVDV in secretions and excretions during a transient infection (TI).Three modes of virus dispersal within the animal’s body are noted. These include lymphatics, viremia, and localizedspread via the alimentary tract.


108

Table 6.1. Subclinical and clinical manifestations of bovine viral diarrhea virus (BVDV)*

Category Manifestations

Acute—Nonpregnant

Acute/asymptomatic Mild fever, leukopenia, seroconversion to BVDV, short-term viremia,short-term shedding.

Acute/symptomatic Fever, leukopenia, depression, anorexia ocular-nasal discharge, oral le-sions, diarrhea, decreased milk production in lactating cattle, short-term viremia (15 days), short-term shedding.

Severe acute Peracute course, fever, pneumonia, sudden death (10–25% mortality).Associated with BVDV type 2.

Hemorrhagic Marked thrombocytopenia, bloody diarrhea, epistaxis, hyphema, bleedingfrom injection and branding sites, pyrexia, leukopenia, and death. As-sociated with BVDV type 2.

Acute/bovine respiratory distress Fever, pneumonia, anorexia, prolonged treatment, and leading cause ofdeath in feedlot cattle. Secondary infections with Pasteurella multo-cida, Mannheimia hemolytica, and Mycoplasma bovis complicate thepathogenesis.

Acute/immunosuppressive Leukocyte function diminished up to 25%, leukopenia, decreased CD+4and CD+8 T lymphocytes, and decreased macrophage and neutrophilfunctions. Secondary or polymicrobial infections with bovine respira-tory syncytial virus, papular stomatitis virus, malignant catarrhal fevervirus have been reported.

Acute—Pregnant

Abortion Naive cattle with no prior exposure to BVDV become infected as in acute/asymptomatic or acute/symptomatic above. Viremia leads to placentalinfection. Abortion, early embryonic death may occur from placentitis.If infection of dam occurs during 90–150 days gestation, congenital in-fection (CI) of fetus occurs. One form of CI between 90–120 days mayresult in fetal tolerance to BVDV.

Congenital defects Infection of a naive dam between 90 and 150 days gestation may result inviremia, which infects fetus resulting in congenital infection (CI). Oneparticular CI occurs between day 90 and 125 and may result in fetaltolerance to BVDV. Calves born with this form of CI are referred to aspersistently infected (PI). Calves that are PI have long-term (lifelong)viremia and are long-term shedders of virus, and are at high risk to de-velop mucosal disease (MD)

Acute/Chronic Sequel to CI

Mucosal disease Calves (6–18 months) that are PI are immunotolerant to noncytopatho-genic (ncp) BVDV (see Ridpath, Chapter 3). These PI calves may ap-pear clinically normal, but upon superinfection with a homologousBVDV, or mutation of the ncp virus to a cytopathic (cp) variant, or vac-cination with a MLV BVDV strain homologous to the ncp PI virus,have a high percentage of developing one of two reported forms ofMD—acute or chronic.


(Evermann and Ridpath, 2002). The results of thisstudy (Table 6.2) indicated that later fetal infectionsleading to congenital defects and PI calves were as-sociated more with BVDV types 1a and 1b than withBVDV type 2.

RECOGNITION OF CLINICALSIGNS

SUBCLINICAL INFECTION

It has been estimated that 70–90% of BVDV infec-tions in immunocompetent, seronegative cattleoccur without manifestation of clinical signs (Ames,1986). In closely monitored animals, infection typi-cally results in mild fever, leukopenia, and the devel-opment of serum-neutralizing antibodies. In dairycattle, a decrease in milk production has been asso-ciated with subclinical infections (Moerman et al.,1994). Viral replication appears to occur in the upper

respiratory tract and adjacent lymphoid tissues(Bolin, 1990a; Meehan et al., 1998). Subclinical in-fections appear to account for high levels of serum-neutralizing antibody titers found in most unvacci-nated cattle. However, it should be noted thatsubclinical disease in the pregnant dam might notreflect the severity of the effects of viral infection tothe exposed fetus.


Category Manifestations

Acute/Chronic Sequel to CI

Mucosal disease (continued) Acute MD is characterized by onset of clinical symptoms within 10–14days postinfection. Symptoms include biphasic fever, anorexia, tachy-cardia, polypnea, decreased milk production, and watery diarrhea.Other clinical signs may include nasal-ocular discharge, corneal opac-ity, excessive salivation, decreased rumination, and bloat. Cattle withacute MD become progressively dehydrated and usually die within3–10 days. Although the mortality usually approaches 100%, a few an-imals may survive the acute MD, but are prone to develop chronic MD

Chronic MD is a sequel of acute MD. Affected cattle are unthrifty,and may have intermittent diarrhea, chronic bloat, decreased appetite,and weight loss. Nasal-ocular discharge is commonly reported. Cattlewith chronic MD rarely survive past 18 months and are usuallyculled due to low performance or die of severe debilitation.

*Modified from Grooms et al., (2002).

Figure 6.3. Schematic depicting the origin of persist-ently infected (PI) calves following viremia and placen-tal infection.

Figure 6.4. Schematic depicting the status of samplesfrom persistently infected calves and the occurrence ofvirus in plasma/white blood cells (buffy coat) and fecesfrom affected animals.


CLINICAL BVDV INFECTION

Acute BVDV infection is the term used to describeclinical disease that occurs in non-persistently in-fected, immunocompetent cattle. The disease hastypically been described in seronegative cattle from6–24 months of age. Conceivably, the disease couldoccur in seropositive cattle that are infected with aviral strain that is heterologous to the one that hascaused seroconversion. Clinical signs include vary-ing levels of fever, anorexia, lethargy, leukopenia,ocular and nasal discharge, oral erosions and ulcers,oral papilla blunting and hemorrhage, diarrhea, anddecreased milk production in lactating cows.Epithelial erosions may be present in the interdigi-tal space, coronet, teats, or vulva. Tachypnea maybe incorrectly correlated with pneumonia, but ismost likely due to fever or other nonpulmonaryfactors.

Acute BVDV infection of neonates may result insigns of enteritis or pneumonia. Such signs arethought to be most commonly associated with calvessuffering from failure of passive transfer, since pas-sively acquired maternal antibodies are thought to beprotective. Clinical disease in the face of adequatepassive transfer may be related to antigenic diversitybetween infecting viral strains and viral strainsagainst which passive immunity was developed.Lastly, inadequate passive immunity combined withimmunosuppressive effects of BVDV infection mayresult in secondary diseases affecting various organsystems.

Acute BVDV infection results in damage to ep-

ithelial surfaces of the gastrointestinal, integumen-tary, and respiratory systems (Blowey and Weaver,2003). Viral antigen has been identified in epithelialsurfaces of the tongue, esophagus, intestine, bronchi,and skin, as well as phagocytic cells in lymph nodes,thymus, Peyer’s patches, tonsils, and spleen (Oh-mann, 1983). Tonsils and respiratory tract tissues areinfected first and from there dissemination occurs toother epithelial surfaces and lymphoid tissues. Virusis retained within mononuclear phagocytic cells ofthe lymphoid tissue (Ohmann, 1983).

Differential diagnoses for diarrheal diseases ofadult cattle include Salmonellosis, winter dysen-tery, Johne’s disease, gastrointestinal parasites, ma-lignant catarrhal fever, arsenic poisoning, Myco-toxicosis, and copper deficiency (Blowey andWeaver, 2003; Kahrs, 2001). Differential diagnosesfor diseases causing oral lesions in cattle includemalignant catarrhal fever, blue tongue, vesicularstomatitis, papular stomatitis, foot and mouth dis-ease, and Rinderpest. Differential diagnoses for dis-eases causing diarrhea in neonatal calves includerota- and coronavirus infections, cryptosporidiosis,E. coli infection, Salmonellosis and coccidiosis.Differential diagnoses for respiratory diseases ofcalves include infection with bovine respiratorysyncytial virus (BRSV), Pasteurella spp., Mann-heimia spp., Hemophilus spp., Salmonella spp., andMycoplasma spp.

SEVERE ACUTE BVDV INFECTION

In the early 1990s, an atypical and significantlymore severe form of BVDV infection was recog-nized in the United States and Canadian cattle pop-ulations (Corapi et al., 1990; Pellerin et al., 1994;


Figure 6.5. Schematic depicting the outcome of apersistently infected calf, which is vulnerable to muco-sal disease upon reinfection with homologous BVDV,mutation of tolerizing noncytopathogenic BVDV tocytopathogenic, or vaccination with a homologousstrain of BVDV.

Table 6.2. Summary of bovine viral diarrheavirus isolates and analysis of genotypecompared with clinical presentation

BVDV Genotype* and Subgenotypes

Clinical Presentation 1a 1b 2

Early fetal infection (abortion) 14% 22% 45%Later fetal infection (includes 86% 78% 55%

animals that are persistently infected, weak calves,chronic poor doers, and mucosal disease)

*From Evermann and Ridpath, 2002.


Carman et al., 1998). A peracute disease course wasdescribed with high morbidity and mortality in allage groups of cattle. In the Quebec outbreak approx-imately 25% of veal calves died, and in Ontario, allage groups of cattle suffered fever, pneumonia, andsudden death (Pellerin et al., 1994; Carman et al.,1998). Severity varied between herds, though someherds experienced 10–20% mortality. Abortionswere also a common occurrence. The gross lesionsfound in these animals were severe, resemblingthose of mucosal disease. The viral isolates fromthese outbreaks were characterized by nucleotide se-quencing and compared to more classical BVDVoutbreak strains (Ridpath et al., 1994). The novelgroup of BVDV was designated BVDV type 2, com-pared to the classical group of viruses, now knownas BVDV type 1.

Importantly, not all BVDV type 2 outbreaks arenecessarily associated with severe disease. There-fore, a severe outbreak of disease should not be as-sumed to have been caused by BVDV type 2.Alternatively, it is likely that some BVDV type 1isolates may be capable of causing severe disease.

Hemorrhagic syndrome is a form of severe acuteBVDV that appears to be associated with a noncyto-pathic, type 2 strain of BVDV. Affected cattle oftensuffer marked thrombocytopenia, petechiation, andecchymoses of mucosal surfaces, epistaxis, bloodydiarrhea, bleeding from injection sites or trauma,fever, leukopenia, and death (Corapi et al., 1990)(Figures 6.6 and 6.7). Importantly, during an out-break variation exists between animals in the ex-pression of clinical signs, and only a minority of an-imals develop fulminant hemorrhagic syndrome.Others may simply suffer marked lymphopenia andthrombocytopenia and succumb without overt clini-cal manifestation of hemorrhage. Therefore, duringan outbreak of severe acute BVDV, practitioners andproducers typically observe a diverse clinical pic-ture. Whereas all cattle with hemorrhagic syndromesuffer from severe acute BVDV, not all cattle withsevere acute BVDV necessarily progress to hemor-rhagic syndrome. The pathophysiologic mechanismof thrombocytopenia and the hemorrhagic syndromeis not clearly understood. Viral antigen has beenshown to be associated with both platelets andmegakaryocytes, and in addition to thrombocytope-nia, platelet function is altered (Walz et al., 1999).Common differential diagnoses for hemorrhagicsyndrome include septicemia with concurrent dis-seminated intravascular coagulation, bracken ferntoxicity, sweet clover poisoning, and anticoagulanttoxicity.

MUCOSAL DISEASE

Typically mucosal disease is a sporadic conditionwherein <5% of herds are affected. In outbreaks, itis common that multiple animals of similar age areaffected because the infection is initiated in the fetus


Figure 6.6. Petechiation of ocular mucous membranesobserved in thrombocytopenic calf.

Figure 6.7. Excessive hemorrhage in a thrombocy-topenic calf several hours after being administered asubcutaneous injection (courtesy of Dr. Sheila McGuirk,University of Wisconsin).


of cows that are at a similar stage of gestation.During such epizootics, up to 25% of a herd may beinfected (refer to Figure 6.7).

An incubation period of 7–14 days has been sug-gested for MD (Grooms et al., 2002). This timeframe is primarily based on experimental infectionswhere a homologous, cp BVDV strain was admin-istered to ncp BVDV persistently infected calves, orsuch calves were administered a modified live vac-cine. Since the majority of cases of MD are thoughtto occur when cp BVDV mutates de novo fromcattle persistently infected with a ncp strain, suchan incubation time frame may not be pertinent(Figure 6.8).

Clinical signs of acute MD include fever, ano-rexia, tachycardia, polypnea, decreased milk pro-duction, and profuse watery diarrhea. Diarrhea isoften characterized by the presence of mucosalshreds, fibrinous casts, blood, and foul odor. Othersigns similar to acute BVDV infection may also bepresent antemortem, but are more pronounced.Erosions and ulcers may be present on the tongue,palate, and gingiva. Oral papilla may be blunted andhemorrhagic. Epithelial erosions may also be pro-nounced in the interdigital regions, coronary bands,teats, vulva, and prepuce. Additional clinical signsoften include ocular and nasal discharge, cornealedema, hypersalivation, decreased ruminal contrac-tions, and bloat. Inflammation of the interdigitalspace and coronary bands may present as lamenessand animals may become laminitic. Clinicopath-ologic findings often include neutropenia (without aleft shift) and thrombocytopenia. Secondary bacter-ial infections may be manifest as pneumonia, masti-tis, and metritis. Cattle suffering acute MD gener-

ally have a case fatality rate approaching 100%.However, a minority of animals may survive theacute phase only to suffer signs of chronic MD (seenext section). Postmortem findings of cattle suffer-ing acute MD include variable necrotizing ulcersand erosions throughout the gastrointestinal tract(esophagus, rumen, abomasum, duodenum, je-junum, ileum, cecum, and colon). Erosions may alsobe present in the nares and upper respiratory tract.Peyer’s patches in the small bowel are often necroticand hemorrhagic. Bowel contents are typically wa-tery and fetid. Differential diagnoses for cattle withacute MD include those listed under acute BVDVinfection (Figures 6.9, 6.10, 6.11). Although theclinical signs of hemorrhagic syndrome and MDmay be similar, animals given supportive therapymay recover from hemorrhagic syndrome. Suppor-tive therapy rarely has any effect in preventing thedeath of animals suffering from MD.


Figure 6.8. Three age-matched beef calves from aherd suffering an outbreak of mucosal disease.

Figure 6.9. Postmortem image of the palate of calfsuffering mucosal disease. Note mucosal ulcerationand blunting of oral papillae. Similar lesions may be ob-served in cattle suffering severe acute BVDV infection.

Figure 6.10. Postmortem image of esophageal ulcersand erosions in calf suffering mucosal disease. Similarlesions may be observed in cattle suffering severeacute BVDV infection.


CHRONIC MUCOSAL DISEASE

A small proportion of cattle with symptoms ofacute MD do not die, but go on to develop a chronicform of the disease. These cattle typically presentwith persistent loose feces or intermittent diarrhea.Additional signs include ill thrift, mild-to-moderateanorexia, chronic recurrent bloat, interdigital ero-sions, and nonhealing erosive lesions of the skin.Ocular and nasal discharge may also be noted. Alopecia and areas of hyperkeratinization may de-velop, especially around the head and neck.Lameness may be manifest as chronic laminitis andabnormal hoof wall growth. Clinicopathologic find-ings often include anemia, neutropenia and throm-bocytopenia.

VENEREAL INFECTIONS

Virus is present in the semen of bulls that are persist-ently infected with BVDV (Revell et al., 1988).Virus is also present in the semen of immunocompe-tent, transiently infected bulls undergoing acute in-fection (Kirkland et al., 1991). Furthermore, recentstudies have demonstrated that transiently infectedbulls can shed BVDV in semen for prolonged peri-ods (months) after they have become non-viremic(Voges et al., 1998; Givens et al., 2003). Therefore,diagnostic testing of semen is recommended due tothe potential for non-viremic bulls to shed BVDV insemen. A recent study demonstrated that the conven-tional testing method of virus isolation (VI) insemen was less sensitive compared to reverse tran-scription-nested polymerase chain reaction (RT-nPCR).

Both acceptable and unacceptable semen qualityhas been reported from bulls that were TI or PI withBVDV. Although results of studies examining theeffect of BVDV infection on conception rates havebeen conflicting, under certain circumstances

BVDV infection is likely associated with transientdecreased conception rates (Baker, 1995).

REPRODUCTIVE CONSEQUENCES

Bovine viral diarrhea virus has been associated withseveral reproductive consequences, including reducedfertility and early embryonic death. All the major or-gans of the female reproductive system are permis-sive to BVDV, and the distribution of virus is similarbetween acutely infected and persistently infectedcows (Fray et al., 2000). Although the precise mech-anism responsible for reduced fertility is unknown,changes in ovarian function have been suggested.Ovarian hypoplasia has been reported in PI cows(Grooms et al., 1996), and diffuse interstitial ovaritishas been reported 61 days postinfection (Ssentongo etal., 1980). Grooms et al. (1998) demonstrated that thegrowth rate and diameter of dominant anovulatoryand ovulatory follicles were significantly reduced fol-lowing acute BVDV infection. Additionally, the num-bers of subordinate follicles associated with theanovulatory and ovulatory follicle were also reducedfollowing infection. Three possible mechanisms ofhow BVDV might compromise ovarian function havebeen suggested (Fray et al., 2000). First, BVDVmight adversely affect pituitary gonadotroph func-tion. Second, BVDV-suppressed plasma estrogen lev-els may adversely affect ovulation and estrus. Last,BVDV-induced general leukopenia may result in de-ficient ovarian leukocyte populations, because thesecells are vital for normal follicular dynamics.

Kafi et al., (2002) demonstrated a detrimental ef-fect of BVDV on in vitro fertilization and early em-bryo development; however, the mechanisms affect-ing these processes were not examined. Clearly,further studies are needed to define the effects ofBVDV infection on the various hormones and cy-tokines that influence early reproductive events.

Infection of immunocompetent pregnant cattlecan result in clinical manifestations in the dam sim-ilar to those described above—i.e., subclinical to se-vere, acute disease, or hemorrhagic syndrome.However, additional clinical outcomes in pregnantcattle are related to the potential transplacentaltransfer of virus to the fetus. Experimental studieshave shown that BVDV crosses the placenta withnear 100% efficiency; therefore, transplacental in-fection of the fetus should be anticipated (Duffelland Harkness, 1985). Importantly, the absence ofclinical signs in the dam does not imply fetal protec-tion or reduce the possibility of transplacental infec-tion. The primary determinant of the outcome offetal infection is clearly the gestational age of the


Figure 6.11. Postmortem image of colon lesions in calfsuffering mucosal disease. Similar lesions may be ob-served in cattle suffering severe acute BVDV infection.


fetus, although other host and viral factors may alsocontribute to the final outcome.

Abortion results from BVDV infection duringpregnancy, and the resultant fetal pathology hasbeen recently reviewed (Fray et al., 2000). Infectionof the dam during the preimplantation phase (<40days) of pregnancy can result in a high incidence ofembryonic or fetal mortality, whereas infection be-tween 40 and 125 days gestation can result in fetaldeath, abortion, mummification, birth of PI calves,and to a lesser degree teratogeneisis. In the case offetal death during the first third of gestation, fetalexpulsion can vary and occur anywhere from days tomonths after infection (Kahrs, 2001). Significantrates of abortion (40%) have been reported whencattle are infected at 100 days gestation under exper-imental conditions (Done et al., 1980); however,lower rates are thought to prevail under field condi-tions. Fetal infections during midgestation (125–180days) typically result in a high incidence of congen-ital abnormalities, which can approach 100% underexperimental conditions (Baker, 1995). Fetal infec-tion during the later stages of gestation does not typ-ically result in abortion but is still possible (Bolin,1990b). Overall, the incidence of abortion is low inimmune herds, but can increase dramatically in non-immune herds (Sanderson and Gnad, 2002).Furthermore, outbreaks of acute severe disease asso-ciated with BVDV type 2 have been associated witha higher incidence of abortion (Grooms et al., 2002).

Congenital defects result from transplacental in-fection of the fetus. It has long been known thatBVDV can act as a teratogen (Mickelsen andEvermann, 1994). Transplacental infection of thefetus from approximately 100–180 days gestationcan be manifest by several congenital defects.During this period, the fetus undergoes the finalstages of organogenesis of the nervous system aswell as immune system development. The mecha-nisms of viral-induced cellular damage are thoughtto include inhibition of cell growth and differentia-tion, or direct cell lysis (Castrucci et al., 1990). Somecommon congenital defects include CNS defects(cerebellar hypoplasia, hydrocephalus, hypomye-linogenesis), ocular defects (retinal atrophy and dys-plasia, cataracts, microphthalmia), thymic hypo-plasia, retarded growth, pulmonary hypoplasia,alopecia, hypotrichosis, brachygnathism, arthrogry-posis, and other skeletal abnormalities (Baker, 1995).

IMMUNOSUPPRESSION

Acute BVDV infection results in the destruction oflymphoid tissue and immunosuppression (Chase et

al., 2004). Recent studies have examined lesions andtissue distribution of virus in calves experimentallyinfected with both high- and low-virulence strains ofBVDV type 2 (Liebler-Tenorio et al., 2002; Liebler-Tenorio et al., 2003a, 2003b). Widespread lymphoiddepletion was noted in calves infected with eitherstrain, and a strong correlation existed between thedistribution of viral antigen and the lesions in lym-phoid tissue. With the high-virulent BVDV strain,apoptosis and lymphocyte depletion was detected inall compartments where viral antigen was observed(Liebler-Tenorio et al., 2002). With the low-virulentstrain, significant lymphoid depletion was rapidlyfollowed by the clearance of viral antigen and sub-sequent recovery of lesions (Liebler-Tenorio et al.,2003a). Interestingly, viral antigen was not associ-ated with lesions in non-lymphoid tissues (lung,liver, kidney, pancreas, testes, or heart), suggestingthat the development of lesions are not solely a func-tion of viral replication but are attributable to thehost’s reaction to infection. Vaccination of cattlewith modified live virus (MLV) vaccines has alsobeen associated with temporary immunosuppression(Potgieter, 1995).

BVDV-induced immunosuppression may increasesusceptibility of the host to other pathogens or exac-erbate the pathogenicity of co-infecting organisms.Synergistic effects of BVDV infection have been de-scribed when animals are concurrently infected withMannheimia haemolytica, bovine herpesvirus type1, or BRSV. BVDV infections have also been asso-ciated with concurrent infection with Salmonellaspp., E. coli, bovine papular stomatitis virus, ro-tavirus, or coronavirus (Grooms, 1999).

The mechanism of BVDV immunosuppressionappears to be multifactorial. Both lymphocytes andmacrophages are targets of BVDV infection(Bruschke et al., 1998). Acute BVDV infection canresult in transient leukopenia and depletion of lym-phoid tissues (Bolin et al., 1985). Decreased B-lymphocytes, CD4+ and CD8+ T-lymphocytes, andneutrophils have been reported (Ellis et al., 1988). Invitro studies revealing potential causes of immuno-suppression have been reviewed and include de-creased mitogen stimulation of infected lympho-cytes, decreased production of interferon, monocyteinterleukin 1, interleukin 2, and tumor necrosisfactor-�, decreased monocyte chemotaxis, and im-paired neutrophil-mediated, antibody-dependant,cell-mediated cytotoxicity (Grooms et al., 2002).Decreased bactericidal activity of neutrophils hasalso been described (Roth et al., 1981). Finally,BVDV-induced immunosuppression may indirectly



result from prostaglandin release by infected cells(Markham and Ramnaraine, 1985).

BOVINE RESPIRATORY DISEASE

Experimental studies have demonstrated that BVDVcan establish infections in the respiratory tract ofcattle and that differences in pneumopathogenicityexist between isolates (Potgieter et al., 1985).Despite this finding, the majority of evidence sug-gests that BVDV infection primarily has a negativeimpact upon the respiratory defenses, most likelythrough immunosuppressive effects (Callan andGarry, 2002). Indeed, synergistic effects have beendocumented between BVDV and M. haemolytica,bovine herpesvirus 1, BRSV, and Mycoplasma bovis(Potgeiter et al., 1984a; Potgeiter et al., 1984b;Pollreisz et al., 1997; Broderson and Kelling, 1998;Shahriar et al., 2002).

DURATION AND SEVERITY OFCLINICAL SIGNSAs stated earlier, infection of immunocompetentnonpregnant cattle can result in a wide spectrum ofdisease syndromes progressing from subclinical tosevere. The incubation period for acute BVD is 5–7days (Grooms et al., 2002). Signs of fever, lethargy,anorexia, oral lesions, oculonasal discharge, diar-rhea, and decreased milk production in lactatingcows follow. Viremia can last up to 15 days. The du-ration of clinical signs is therefore variable and de-pends on the duration of viremia, virulence of the in-fecting virus, the presence of secondary infections,and the normal regenerative capacity of affectedtissues. In general, repair of lesions involving ep-ithelial surfaces requires 1–2 weeks, and repair oflesions involving mucosal surfaces of the gastroin-testinal tract require approximately 3–5 days.Results from experimental studies suggest that re-covery is prolonged with more virulent strains ofBVDV since the virus is more widespread in tissuesand is more slowly cleared (Liebler-Tenorio et al.,2003b). In the absence of secondary bacterial infec-tions, abatement of clinical signs would correlate tothe normal regenerative capacity of the involvedtissues.

Secondary infections influence both the durationand severity of clinical diseases associated withBVDV, especially the acute-transient infections. Thefact that BVDV is immunosuppressive allows forsecondary infections (synonyms include concurrentinfections and polymicrobial infections) to have aprofound effect on animal survival (Bolin, 2002). Insome cases, such as the sheep-associated malignant

catarrhal fever (ovine herpesvirus-2) in bison,BVDV may actually serve as the secondary infec-tion or as a dual infection (Evermann, unpublisheddata). There is laboratory and clinical evidence thatnutritional deficiencies, such as selenium deficiency,predispose cattle to more severe BVDV acute infec-tions and disease and concurrent bacterial infec-tions, such as Hemophilus somnus.

Duration of clinical signs in cattle suffering acuteMD is generally 3–10 days, whereby the eventualoutcome is the death of the animal. Affected cattlethat survive past this expected time frame are con-sidered to develop chronic MD. Cattle with chronicMD typically have an unthrifty appearance with per-sistent loose stool or intermittent bouts of diarrhea,chronic bloat, decreased appetite, weight loss, inter-digital and coronary erosions, and nonhealing skinlesions. Cattle suffering chronic MD rarely survivepast 18 months of age and typically die due to debil-itation and emaciation. Chronic MD is distinguishedfrom chronically poor-doing PI calves because thelatter are typically affected from birth.

RECOVERY FROM CLINICALSIGNSCattle suffering acute BVDV infection without sec-ondary infections generally recover without majorcomplications. Time to recovery is related to the du-ration of viremia as well as the severity of lesions. Ingeneral, recovery can be observed within 2–4 weeksof the onset of signs. Recovery in cattle sufferingsecondary infections may be prolonged and is de-pendent on the severity of secondary infections. Asstated previously, cattle suffering acute MD do notsurvive and death occurs approximately 3–10 daysafter of the onset of signs. Cattle suffering chronicMD typically do not survive beyond 18 months ofage (Bolin, 1995).

OCCURRENCE IN OTHERRUMINANTSEfforts to eradicate BVDV from cattle are hamperedfor several reasons, including short-term carrier cat-tle that have transient infections ranging from 2–8weeks (Cherry et al., 1998), long-term PI carriersthat continue to shed virus to herdmates for life(Bolin, 1995; Wittum et al., 2001), strain variationamong BVDV isolates evading host immune de-fenses (Vilcek et al., 2001; Deregt and Loewen,1995), and the wide host range of BVDV (Evermannet al., 1993; Nettleton and Entrican, 1995). The hostrange for BVDV and related pestiviruses (seeChapter 10) includes all ungulates belonging to the



order Artiodactyla, within which there are 11species of swine, and up to 173 ruminant species(Van Campen, et al., 2001).

Clinical symptoms in wildlife, such as axis deer,roe deer, and moose, include fever, corneal opacity,and depression. The severity of the disease associ-ated with BVDV appears to depend upon the viralstrain, immune competence of the host animal, con-current viral infection, and/or nutritional deficien-cies, such as copper (Van Campen et al., 2001).Domestic animals that can be infected with BVDVinclude sheep, goats, and members of the camelidaefamily (camels, llamas, and alpacas) (Van Campen,et al., 2001; Goyal et al., 2002). Clinical symptomsinclude diarrhea in camels and llamas (Evermann,et al., 1993; Yousif, et al., 2004), abortion, and illthrift in pregnant llamas (Belknap, et al., 2000), andcongenital defects in camel calves (Yousif et al.,2004).

SUMMARY AND CONCLUSIONSAs clinicians and diagnosticians continue to observeand detect the multifaceted effects of BVDV, onemust ask, “Is there such a thing as a true subclinicalBVDV infection?” Recent epidemiological datawould suggest that BVDV poses considerable eco-nomic constraints on a herd (Houe, 1999; Innocentet al., 1997; see Chapter 2). These include

• Immunosuppressive effects of the virus on calvespostnatally

• Effects of transient infections upon pregnantcows and abortion

• Effects of transient infections upon reproductiveage cattle and delayed rebreeding

• Congenital infections leading to calves with con-genital defects and growth retardation, etc.

• Congenital infections resulting in PI calves• Long-term survivability of PI heifers leading to

future PI calves, mortality, and replacement costs

The impact that BVDV has on livestock productionfrom both a subclinical and clinical perspective is ofmajor concern (Houe, 1999; Kelling et al., 2000). Acombination of rigorous testing for BVDV PI ani-mals (see Chapter 12) and effective vaccinationstrategies (see Chapter 13) is imperative for theoverall health of beef and dairy cattle (Smith andGrotelueschen, 2004).

REFERENCESAmes TR: 1986, The causative agent of BVD: Its epi-

demiology and pathogenesis. Vet Med 81:848–869.

Archambault D, Beliveau C, Couture Y, et al.: 2000,Clinical response and immunomodulation followingexperimental challenge of calves with type 2 non-cytopathogenic bovine viral diarrhea virus. Vet Res31:215–227.

Baker JC: 1995, The clinical manifestations of bovineviral diarrhea infection. Vet Clin North Am FoodAnimal Pract 11:425–445.

Belknap EB, Collins JK, Larsen RS, et al.: 2000,Bovine viral diarrhea virus in new world camelids.J Vet Diagn Invest 12:568–570.

Blowey RW, Weaver AD: 2003, Alimentary disorders.In: Color Atlas of Diseases and Disorders of Cattle,2nd Ed. Eds. Blowey RW and Weaver AD, pp.43–46, 123–124. Mosby-Elsevier Science,London.

Bolin SR: 1990a, The current understanding about thepathogenesis and clinical forms of BVD. Vet Med85:1124–1132.

Bolin SR: 1990b, Bovine abortion caused by bovineviral diarrhea virus. In: Laboratory Diagnosis ofLivestock Abortion, 3rd Ed. Ed. Kirkbridge CA,pp. 121–128. Iowa State University Press,Ames, IA.

Bolin SR: 1995, The pathogenesis of mucosal disease.Vet Clin North Am Food Anim Pract 11:489–500.

Bolin SR: 2002, Bovine viral diarrhea virus in mixedinfections. In: Polymicrobial Diseases. Eds.Brogden KM, Guthmiller JM, pp. 33–50. ASMPress, Washington, D.C.

Bolin SR, McClurkin AW, Coria MF: 1985, Effects ofbovine viral diarrhea virus on the percentages andabsolute numbers of circulating B and T lympho-cytes in cattle. Am J Vet Res 46:884–886.


Bolin SR, Ridpath JR: 1996, The clinical significanceof genetic variation among bovine viral diarrheavirus. Vet Med 91: 958–961.

Broderson BW, Kelling C: 1998, Effects of concurrentexperimentally induced bovine respiratory syncytialvirus and bovine viral diarrhea virus infection onrespiratory tract and enteric diseases in calves. AmJ Vet Res 59:1423–1450.

Brownlie J, Thomas I, Curwen A: 2000, Bovine virusdiarrhoea virus—Strategic decisions for diagnosisand control. In Pract 22:176–187.

Bruschke CJM, Weerdmeester K, Van Dirschot JT, etal.: 1998, Distribution of bovine virus diarrhoeavirus in tissues and white blood cells of cattle dur-ing acute infection. Vet Microbiol 64:23–32.

Callan RJ, Garry FB: 2002, Biosecurity and bovinerespiratory disease. Vet Clin North Am Food Pract18:57–77.



Carman S, Van Dreumel T, Ridpath J, et al.: 1998,Severe acute bovine viral diarrhea in Ontario1993–1995. J Vet Diagn Invest 10:27–35.

Castrucci G, Frigeri F, Osburn BI, et al.: 1990, Astudy of some pathogenic aspects of bovine virusdiarrhea virus infection. Comp Immunol MicrobiolInfect Dis 13:41–49.

Chase CC, Elmowalid G, Yousif AA: 2004, The im-mune response to bovine viral diarrhea virus: Aconstantly changing picture. Vet Clin North AmFood Anim Pract 20:95–114.


Corapi WV, Elliot RD, French TW, et al.: 1990,Thrombocytopenia and hemorrhages in veal calvesinfected with bovine viral diarrhea virus. J Am VetMed Assoc 196:590–596.

Deregt D, Loewen KG: 1995, Bovine viral diarrheavirus: Biotypes and disease. Can Vet J 36:37–378.

Done JT, Terlecki S, Richardson C, et al.: 1980,Bovine viral diarrhoea-mucosal disease virus:Pathogenicity for the foetal calf following maternalinfection. Vet Rec 106:473–479.

Dubovi EJ: 1994, Impact of bovine viral diarrheavirus on reproductive performance in cattle. VetClin North Am Food Anim Pract 10:503–514.

Duffell SJ, Harkness JW: 1985, Bovine virus diar-rhoea-mucosal disease infection in cattle. Vet Rec117:240–245.

Ellis JA, Davis WC, Belden EL, et al.: 1988, Flowcytofluorimetric analysis of lymphocyte subset al-teration in cattle infected with bovine viral diarrheavirus. Vet Pathol 25:231–236.

Evermann JF, Berry ES, Baszler TV, et al.: 1993,Diagnostic approaches for detection of bovine viraldiarrhea (BVD) virus and related pestiviruses. J VetDiagn Invest 5:265–269.


Fray MD, Paton DJ, Alenius S: 2000, The effects ofbovine viral diarrhoea virus on cattle reproductionin relation to disease control. Anim Reprod Sci60–61:615–627.

Givens MD, Heath AM, Brock KV, et al.: 2003,Detection of bovine viral diarrhea virus in semenobtained after innoculation of seronegative postpu-bertal bulls. Am J Vet Res 64:428–434.

Goyal SM, Bouljihad M, Haugerud S, et al.: 2002,Isolation of bovine viral diarrhea virus from an al-paca. J Vet Diagn Invest 14:523–525.

Grooms D, Baker JC, Ames TR: 2002, Diseasescaused by bovine virus diarrhea virus. In: LargeAnimal Internal Medicine, 3rd Ed. Ed. Smith BP,pp: 707–714. Mosby, St. Louis, MO.

Grooms DL: 1999, Role of bovine viral diarrhea virusin the bovine respiratory disease complex. BovPract 32:7–12.

Grooms DL, Brock KV, Pate JL, et al.: 1998, Changesin ovarian follicles following acute infection withbovine viral diarrhea virus. Theriogenology49:595–605.

Grooms DL, Ward LA, Brook KV: 1996,Morphologic changes and immunohistochemicaldetection of viral antigen in ovaries from cattle per-sistently infected with bovine viral diarrhea virus.Am J Vet Res 57:830–833.

Houe H: 1999, Epidemiologic features and economi-cal importance of bovine virus diarrhoea virus in-fections. Vet Microbiol 64:89–107.

Innocent G, Morrison I, Brownlie J, et al.: 1997, Theuse of a mass-action model to validate the outputfrom a stochastic simulation model of bovine viraldiarrhoea virus spread in a closed dairy herd. PrevVet Med 31:199–209.

Kafi M, McGowan MR, Kirkland PD. 2002, In vitromaturation and fertilization of bovine oocysts andin vitro culture of presumptive zygotes in the pres-ence of bovine pestivirus. Anim Reprod Sci71:169–179.

Kahrs RF: 2001, Bovine viral diarrhea. In: ViralDiseases of Cattle, 2nd Ed. Ed. Kahrs RF, pp.113–126. Iowa State University Press, Ames, IA.

Kelling CL, Grotelueschen DM, Smith DR, et al.:2000, Testing and management strategies for effec-tive beef and dairy herd BVDV biosecurity pro-grams. Bovine Pract 34:13–22.

Kirkland PD, Richards SG, Rothwell JT, et al.: 1991,Replication of bovine viral diarrhoea virus in thebovine reproductive tract and excretion of virus insemen during acute and chronic infections. Vet Rec128:587–590.

Liebler-Tenorio EM, Ridpath JF, Neill JD: 2002,Distribution of viral antigen and development of le-sions after experimental infection with highly viru-lent bovine viral diarrhea virus type 2 in calves. AmJ Vet Res 63:1575–1584.

Liebler-Tenorio EM, Ridpath JF, Neill JD: 2003a,Distribution of viral antigen and development of le-sions after experimental infection of calves with aBVDV 2 strain of low virulence. J Vet Diagn Invest15:221–232.

Liebler-Tenorio EM, Ridpath JF, Neill JD: 2003b,Lesions and tissue distribution of viral antigen insevere acute versus subclinical acute infection withBVDV2. Biologicals 31:119–122.



Loken T: 1995, Ruminant pestivirus infections in ani-mals other than cattle and sheep. Vet Clin North AmFood Anim Pract 11:597–614.

Markham RJF, Ramnaraine ML: 1985, Release of im-munosuppressive substances from tissue culturecells infected with bovine viral diarrhea virus. Am JVet Res 46:879–881.

McClurkin AW, Littledike ET, Cutlip RC: 1984,Production of cattle immunotolerant to bovine viraldiarrhea virus. Can J Comp Med 48:156–161.

Meehan JT, Lehmkuhl HD, Cutlip RC, et al.: 1998,Acute pulmonary lesions in sheep experimentallyinfected with bovine viral diarrhoea virus. J CompPathol 119:277–292.

Mickelsen WD, Evermann JF: 1994, In utero infec-tions responsible for abortion, stillbirth, and birth ofweak calves in beef cows. Vet Clin North Am FoodAnim Pract 10:1–14.

Mills JHL, Luginbuhl RE: 1968, Distribution and per-sistence of mucosal disease virus in experimentallyexposed calves. Am J Vet Res 29:1367–1375.

Moerman A, Straver PJ, de Jong MCM, et al.: 1994,Clinical consequences of a bovine virus diarrhoeain a dairy herd. Vet Q 16:115–119.

Muñoz-Zanzi CA, Hietala SK, Thurmond MC, et al.:2003, Quantification, risk factors, and health impactof natural congenital infection with bovine viral di-arrhea virus in dairy calves. Am J Vet Res64:358–365.

Nettleton PF, Entrican G: 1995, Ruminant pes-tiviruses. Brit Vet J 151:615–642.

Niskanen R, Lindberg A: 2003, Transmission ofbovine viral diarrhoea virus by unhygienic vaccina-tion procedures, ambient air, and from contami-nated pens. Vet Jour 165:125–130.

Ohmann HB: 1983, Distribution and significance ofBVD antigen in diseased calves. Res Vet Sci34:5–10.

Pellerin C, Van Den Hurk J, Lecomte J, et al.: 1994,Identification of a new group of bovine viral diar-rhea virus strains associated with severe outbreaksand high mortalities. Virology 203:260–268.

Pollreisz JP, Kelling CL, Perino LJ, et al.: 1997, Thepotentiation of bovine respiratory syncytial virusinfection in calves by bovine viral diarrhea virus.Bovine Pract 31:32–38.

Potgieter LN, McCracken MD, Hopkins FM, et al.:1984a, Experimental production of bovine respira-tory tract disease with bovine viral diarrhea virus.Am J Vet Res 45:1582–1585.

Potgieter LN, McCracken MD, Hopkins FM, et al.:1984b, Effect of bovine viral diarrhea virus infec-tion on the distribution of infectious bovine rhino-tracheitis virus in calves. Am J Vet Res 45:686–689.

Potgieter LN, McCracken MD, Hopkins FM, et al.:1985, Comparison of pneumopathogenicity of twostrains of bovine viral diarrhea virus. Am J Vet Res46:151–153.

Potgieter LND: 1995, Immunology of bovine viral di-arrhea virus. Vet Clin North Am Food Anim Pract11:501–520.

Revell SG, Chasey D, Drew TW, et al.: 1988, Someobservations of the semen of bulls persistently in-fected with bovine viral diarrhoea virus. Vet Rec123:122–125.


Ridpath JF, Neill JD, Frey M, et al.: 2000,Phylogenetic, antigenic and clinical characteriza-tion of type 2 BVDV from North America. VetMicrobiol 77:145–155.

Roth JA, Kaekerle ML, Griffith RW: 1981, Effects ofbovine viral diarrhea virus infection on bovinepolymorphonuclear leukocyte function. Am J VetRes 42:244–250.

Sanderson MW, Gnad DP: 2002, Biosecurity for re-productive diseases. Vet Clin North Am Food AnimPract 18:79–98.

Shahriar FM, Clark EG, Janzen E, et al.: 2002,Coinfection with bovine viral diarrhea virus andMycoplasma bovis in feedlot cattle with chronicpneumonia. Can Vet J 43:863–868.

Smith DR, Grotelueschen DM: 2004, Biosecurity andbiocontainment of bovine viral diarrhea virus. VetClin North Am Food Anim Pract 20:131–149.

Ssentongo YK, Johnson RH, Smith JR: 1980,Association of bovine viral diarrhea-mucosal dis-ease virus with ovaritis in cattle. Aust Vet J56:272–273.

Tremblay R: 1996, Transmission of bovine viral diar-rhea virus. Vet Med 91:858–866.

van Campen H, Frolich K, Hofmann M: 2001,Pestivirus Infections. In: Infectious Diseases ofWildlife, 2nd Ed. Eds. Williams ES, Barker IK, pp.232–237. Iowa State University Press, Ames, IA.

Vilcek S, Paton DJ, Durkovic B, et al.: 2001, Bovineviral diarrhea virus genotype 1 can be separatedinto at least eleven genetic groups. Arch Virol146:99–115.

Voges H, Horner GW, Rowe S, et al.: 1998, Persistentbovine pestivirus infection localized in the testes ofan immuno-competent non-viremic bull. VetMicrobiol 61:165–175.

Walz PH, Steficek BA, Baker JC: 1999, Effect of typeII bovine viral diarrhea virus on platelet function inexperimentally infected calves. Am J Vet Res60:1396–1401.



Wilhelmsen CL, Bolin SR, Ridpath JF, et al.: 1990,Experimental primary postnatal bovine viral diar-rhea viral infections in six-month-old calves. VetPathol 27:235–243.

Wittum TE, Grotelueschen DM, Brock KV, et al.:2001, Persistent bovine viral diarrhoea virus infec-tion in US beef herds. Prev Vet Med 49:83–94.

Yousif AA, Braun LJ, Saber MS, et al.: 2004,Cytopathic type 2 bovine viral diarrhea virus indromedary camels. Arab J Biotechnol 7:123–140.



INTRODUCTIONPathogenesis is defined as “development of dis-ease.” After infection with BVDV, clinical signs ofdisease are highly variable (Baker, 1995; Nettletonand Entrican, 1995) because of complex interactionsbetween the etiologic agent and the infected host.Genotype (BVDV 1, BVDV 2—also now recog-nized as two distinct species of virus), biotype (non-cytopathic [ncp], cytopathic [cp]), and virulence ofindividual BVDV strains are determinants for theoutcome of infection on the virus side. Immunecompetence, immune status, and possibly other fac-tors determine the outcome on the host’s side. Bothhost and viral factors vary widely causing the broadrange of clinical signs and lesions.

For discussion on pathogenesis of BVDV infec-tions, the following syndromes are distinguished:acute BVDV infection, transplacental/intrauterineinfection, persistent infection, and mucosal disease(Figures 7.1 and 7.2). Susceptible cattle of all agesmay contract a primary, transient BVDV infectiontermed acute BVDV infection irrespective of theclinical course (subclinical, severe acute, or pro-tracted). The transient acute infection causes severecomplications if the infected animal is pregnant.BVDV will cross the placenta and infect the fetuscausing transplacental/interuterine infection. Thereare several outcomes of fetal infections; the mostimportant is infection of the fetus leading to thebirth of persistently infected (PI) calves. Persistentinfection is the prerequisite for the development offatal mucosal disease later in life. Finally, to closethe circle of infection, animals with persistent infec-tion and mucosal disease are the main sources ofvirus in outbreaks of acute BVDV infections (referto Figure 7.2).

ACUTE INFECTIONAcute infections of cattle with BVDV develop whensusceptible (seronegative), immune-competent cat-tle become infected with BVDV. Seropositive cattle,dependent on the levels of antibody titers, are usu-ally not susceptible. The virus sources are often PIanimals, while horizontal infection from acutely in-fected cattle is less likely (Traven et al., 1991; Nis-kanen et al., 2000). Iatrogenic transmission by con-taminated instruments, vaccines, or semen has alsobeen described (Coria and McClurkin, 1978; Bar-kema et al., 2001; Niskanen and Lindberg, 2003).

ROLE OF VIRAL AND HOST FACTORS INPATHOGENESIS

The most important determinant for the outcome ofacute BVDV infection in susceptible animals is thevirulence of the individual BVDV strain. Histori-cally, acute BVDV infections have received little at-tention, because they were predominantly subclini-cal (Moennig and Plagemann, 1992; Hamers et al.,2001), and the unspecific signs of transient fever andlistlessness did not incite high awareness. Until theearly 1990s, there were only a few reports aboutacute BVDV infections causing severe clinical signsor with special affinity for the respiratory tract (Pot-gieter et al., 1984). Over the last 15 years, however,there has been an increasing interest in acute BVDVinfections because of an increase in incidence offield cases of acute BVDV infection associated withsevere clinical signs and mortality in all age groups(Perdrizet et al., 1987; Pritchard et al., 1989; Hib-berd et al., 1993; David et al., 1994; Carman et al.,1998).

Highly virulent BVDV strains isolated from theoutbreaks that had severe economic impact in North

121

7Pathogenesis

E. M. Liebler-Tenorio


122

Figure 7.1. Outcomes and factors influencing the outcome of BVDV infections.

Figure 7.2. Circulation of BVDV in cattle populations.


America were identified as BVDV 2 (Ridpath et al.,1994; Heinz et al., 2000). Further investigation re-vealed, however, that BVDV 2 encompasses notonly strains of high virulence, but also those of mod-erate to low virulence (Ridpath et al., 2000).Virulence factors associated with BVDV have notyet been defined on an antigenic or genetic level.The occurrence of virulence factors is independentof viral species, although the frequency of virulentstrains appears to be higher in BVDV 2 than inBVDV 1. There are regional differences in the rela-tive frequency of BVDV species. Although a highpercentage of BVDV strains belong to BVDV 2 inNorth America (Bolin and Ridpath, 1998; Carman etal., 1998; Fulton et al., 2000), they are infrequentlyisolated in other countries (Wolfmeyer et al., 1997;Letellier et al., 1999; Sakoda et al., 1999; Tajima etal., 2001; Drew et al., 2002; Vilcek et al., 2002).This is reflected in the different frequency of severeacute BVDV infection in different countries.

Virulence is not correlated with the biotype, sinceboth ncp and cp strains are represented in bothBVDV 1 and BVDV 2 and have a wide spectrum ofvirulence. There are, however, differences betweenthe ncp and cp biotype. ncp BVDV spreads muchwider in the host than the homologous cp virus(Spagnuolo-Weaver et al., 1997). The ncp and cpbiotype also differ in the way they activate the im-mune system (Collen and Morrison, 2000). Al-though ncp BVDV induces a more pronounced hu-moral response, the response to cp BVDV is skewedtoward a cell-mediated immune response. This isparticularly evident upon secondary exposure to cpBVDV (Lambot et al., 1997). One can hypothesizethat this is mediated by the different effects of ncpand cp BVDV on monocytes and dendritic cells.After in vitro infection, monocyte function is alteredby ncp BVDV, but they are killed by cp BVDV.Dendritic cells are not affected by infection with ei-ther ncp or cp BVDV (Glew et al., 2003)

The highly variable clinical signs present difficul-ties for diagnosing BVDV infections in the field.Even if acute BVDV infection is suspected, it maybe difficult to confirm, since virus is only transientlydetectable (Sandvik et al., 1997). Except for im-mune status, the role of other host factors on BVDVinfection has not yet been determined. However, be-cause of differences in clinical signs and speed of re-covery between individual animals after experimen-tal infection, the role of host factors should not beneglected (Liebler-Tenorio et al., 2003a).

VIRUS SPREAD AND DEVELOPMENT OFLESIONS IN ACUTE BVDV INFECTION

The most frequent route of natural infection is byoronasal uptake of BVDV. The development of clin-ical signs and distribution and spread of the virus inthe host and tissue lesions have been documented infield cases and under experimental conditions(Lambert et al., 1969; Corapi et al., 1990; Wilhel-msen et al., 1990; Traven et al., 1991; Bolin andRidpath, 1992; Castrucci et al., 1992; Marshall etal., 1996; Spagnuolo-Weaver et al., 1997; Bruschkeet al., 1998a; Ellis et al., 1998; Odeon et al., 1999;Archambault et al., 2000; Hamers et al., 2000;Stoffregen et al., 2000; Liebler-Tenorio et al., 2002,2003a,b). The data were collected primarily for vir-ulent BVDV strains, but some data do exist forstrains of low virulence (Wilhelmsen et al., 1990;Traven et al., 1991; Bolin and Ridpath, 1992; Mar-shall et al., 1996; Bruschke et al., 1998a; Liebler-Tenorio et al., 2003a). In the following section, thefindings for strains of low and high virulence will bedescribed separately.

Strains of low virulence

After intranasal inoculation of calves with BVDV 1and BVDV 2 strains of low virulence, the animalsdid not develop overt signs of disease (Wilhelmsenet al., 1990; Traven et al., 1991; Bolin and Ridpath,1992; Marshall et al., 1996; Bruschke et al., 1998a;Liebler-Tenorio et al., 2003a). The regular monitor-ing of body temperature showed, however, a mildelevation in temperature for 1–2 days (Liebler-Tenorio et al., 2003a). This elevation was very short-lived, making it necessary to monitor temperaturetwice a day.

Virus isolation and titration data indicate thatBVDV 1 first replicates in the tonsils and nasal mu-cosa (Bruschke et al., 1998a). Viral antigen was ini-tially found in the tonsils, lymph nodes, and Peyer’spatches and then in spleen and thymus (Liebler-Tenorio et al., 2003a) (Figure 7.3). Viral antigen waspredominantly present within lymphoid follicles andthe thymic cortex where it was associated with lym-phocytes and stromal cells. Infection in these sitesaffected the majority of cells present. Multifocal in-fection of the intestinal mucosa was the only siteoutside of lymphoid tissues where viral antigencould be demonstrated (refer to Figure 7.3). Viralantigen was not detectable in the bone marrow atany point during infection. Presence of viral antigenwas not associated with tissue lesions.

Pathogenesis 123


The widest distribution and the largest amount ofviral antigen was found at 6 days postinfection (dpi)followed by a rapid clearance of the virus leading toits disappearance from most sites. The remainingviral antigen was associated with follicular dendriticcells in lymphoid follicles and dendritic cells in thethymus. Viral clearance coincided with the loss ofinfected lymphocytes leading to a marked depletionof lymphoid follicles and thymic cortex of lympho-cytes.

After virus clearance, different degrees of deple-tion and repopulation were observed in the lym-phoid tissues throughout the body. There was amarked difference in the degree of recovery betweenindividual calves indicating the importance of hostfactors. The reversibility of the lesions was mostlikely due to the fact that the stromal elements wereretained in the lymphoid tissues allowing repopula-tion with circulating lymphocytes.

Infection with BVDV strains of high virulence

Clinical signs after infection with strains of highvirulence are severe but are often nonspecific con-sisting of high fever, anorexia, depression, and fre-quently diarrhea (Corapi et al., 1990; Bolin and Rid-

path, 1992; Ellis et al., 1998; Odeon et al., 1999;Archambault et al., 2000; Liebler-Tenorio et al.,2002). Some animals develop severe bleeding(Corapi et al., 1990; Bolin and Ridpath, 1992; Stoff-regen et al., 2000). Bleeding is a clinically very dra-matic and easy-to-recognize alteration and thus theterm hemorrhagic syndrome has been associatedwith severe acute BVDV infections although it doesnot occur regularly. The mortality rate may be higheven in older cattle (Carman et al., 1998). In exper-imental infections, early onset of high fever is a con-sistent finding (Ridpath et al., 2000; Liebler-Tenorioet al., 2002). The other consistent findings are severeprogressive lymphopenia and moderate to severethrombocytopenia (Corapi et al., 1990; Bolin andRidpath, 1992; Marshall et al., 1996; Ellis et al.,1998; Odeon et al., 1999; Archambault et al., 2000;Hamers et al., 2000; Stoffregen et al., 2000; Liebler-Tenorio et al., 2002).

The initial spread of highly virulent strains is sim-ilar to that of low virulence strains (Liebler-Tenorioet al., 2002, 2003b) (refer to Figure 7.3). There is ini-tial infection of tonsils and lymphoid tissues, but theamount of viral antigen in tissues rapidly exceedsthat caused by low virulence strains. The presence of


Figure 7.3. Spread of BVDV of low and high virulence in acute BVDV infection.


highly virulent BVDV is not restricted to follicles inlymphoid tissues but is extended to T-cell–dependentareas. Highly virulent BVDV spreads to the bonemarrow. Its presence in myeloid cells and megakary-ocytes correlates with a decrease in thrombocytenumbers.

In contrast to BVDV of low virulence, which iscleared from infected tissues, virulent BVDV strainskeep spreading (refer to Figure 7.3) resulting in thetissue distribution described for severe acute BVDV(Marshall et al., 1996; Ellis et al., 1998; Odeon etal., 1999; Stoffregen et al., 2000). Antigen of highlyvirulent BVDV is regularly found in lymphoid tis-sues, mucosa of the upper and lower digestive tract,the respiratory tract, and endocrine tissues. Even-tually, there is hardly any organ that does not containviral antigen at least in the interstitium. Antigen isoften initially present in the interstitium or vascularwalls and then in parenchymal cells, indicating ahematogenous spread. Within organs, virus spreadsto additional tissue types; e.g., in the intestine whereinitially the mucosa is antigen-positive, it can laterbe found in the muscular layers.

Despite the wide distribution of viral antigen, thepresence of lesions is restricted. Lesions accompa-nied by the loss of lymphocytes are consistently ob-served in lymphoid tissues (Marshall et al., 1996;Ellis et al., 1998; Odeon et al., 1999; Stoffregen etal., 2000; Liebler-Tenorio et al., 2002). Necrosis ofepithelium in the upper digestive tract and intestinalcrypts may cause erosive to ulcerative lesions alongthe digestive tract (Marshall et al., 1996; Carman etal., 1998; Ellis et al., 1998; Odeon et al., 1999;Stoffregen et al., 2000).

The discrepancy between presence of viral antigenand lesions is particularly evident in the initial phaseof the disease (Liebler-Tenorio et al., 2002). Initially,large numbers of infected cells are present in lym-phoid tissues without corresponding morphologicallesions. Lesions in lymphoid tissues are delayed. Inthe thymus, viral antigen is diffusely present inthymic lobules, but lesions are initially multifocal.The same phenomenon can be found in the intestinalmucosa, where viral antigen has a diffuse distribu-tion, but initial lesions can be observed multifocally.In the lungs, frequently foci of acute purulent bron-chopneumonia are seen, which do not correlate withthe distribution of BVDV antigen in the lung.

Comparison between strains of low and highvirulence

Experimental infection with BVDV strains of lowand high virulence reveals similar initial infection

and spread but differing amounts of virus in tissuesand speed of spreading (Liebler-Tenorio et al.,2003b). This results in a widespread tissue distribu-tion of the virulent strains in later stages of infec-tion, as opposed to an early clearance of the strainsof low virulence from tissues. Strains that cause thehighest degree of viremia result in the most severeclinical signs (Walz et al., 2001). This indicates thatdifferences in replication, and not interactions be-tween virus and host cell receptors, have the highestimpact on the virulence of individual BVDV strains.One can hypothesize that there is a competition be-tween viral spread and reaction of the immune sys-tem. The pathogenesis of tissue lesions is unsolved;the discrepancy between the presence of viral anti-gen and lesions as well as the delayed onset of le-sions seen with low and highly virulent BVDVstrains might indicate that immune-mediated reac-tions contribute to the development of lesions.

IMMUNE SUPPRESSION IN ACUTE BVDVINFECTION

A decrease in the number of both B- and T-lymphocytes in peripheral blood is a consistent find-ing in acute BVDV infection (Bolin et al., 1985a).The severity of lymphopenia helps discriminate in-fections with low or highly virulent BVDV strains(Ridpath et al., 2000; Liebler-Tenorio et al., 2003b).Although lymphocyte numbers decrease to morethan 60% below the baseline levels, sometimesreaching 90% below the baseline levels after infec-tion with highly virulent strains, the decrease doesnot exceed more than 50% below the baseline valuesin strains with low virulence. Lymphopenia corre-lates well with the infection of and lesions in lym-phoid tissue, but it is not resolved yet if lesions aredirectly virus-induced or if the immune responsealso contributes to their development.

Experimental infection with virus of low viru-lence demonstrates clearly that even infections thathave a subclinical course and would go unnoticedunder field conditions will cause a marked, althoughtransient, immune suppression (Liebler-Tenorio etal., 2003a). This explains reports that combined in-fections with BVDV have a potentiating effect onseveral pathogens. Coinfection with BVDV in-creases the severity of rota- and coronavirus infec-tions in young calves (van Opdenbosch et al., 1981;Kelling et al., 2002; Niskanen et al., 2002b) and theseverity of IBRV (infectious bovine rhinotracheitisvirus) and BRSV (bovine respiratory syncytialvirus) infections (Potgieter et al., 1984; Castrucci etal., 1992; Brodersen and Kelling, 1998, 1999). The

Pathogenesis 125


immune response to BRSV is delayed (Elvander etal., 1998). The sequential inoculation of calves withBVDV and Mannheimia haemolytica increased theseverity of lung lesions (Potgieter et al., 1984;Ganheim et al., 2003). After vaccination with modi-fied live vaccines for BVDV, there was a decreasedresponse to Mycobacterium paratuberculosis(Thoen and Waite, 1990). BVDV infection exacer-bated the effects of infection with Salmonella sp.(Wray and Roeder, 1987) and led to impaired clear-ance of bacteremia (Reggiardo and Kaeberle, 1981).

Besides inducing lymphopenia, BVDV has a widerange of effects on the functions of specific as wellas innate immune responses (Peterhans et al., 2003).However, immune suppression is not associatedwith low interferon response or elevated levels ofTGF-ß (Charleston et al., 2002). Peripheral lympho-cytes are hyporesponsive to mitogens (Muscoplat etal., 1973; Pospisil et al., 1975). Different neutrophiland monocyte functions are impaired (Ketelsen etal., 1979; Roth et al., 1981, Roth and Kaeberle,1982, 1983; Jensen and Schultz, 1991). Reductionin nonspecific defense has been suggested as themain contribution of BVDV to respiratory tract in-fections (Potgieter, 1997; Brodersen and Kelling,1998). A multitude of functional defects are seen inalveolar macrophages from infected calves andthose inoculated in vitro with BVDV (Welsh et al.,1995; Liu et al., 1999).

THROMBOCYTOPENIA AND HEMORRHAGESIN SEVERE ACUTE BVDV INFECTIONS

Thrombocytopenia occurs regularly in cases of se-vere acute BVD although the reduction of plateletsdoes not always result in marked hemorrhages(Rebhun et al. 1989; Corapi et al., 1990; Bolin andRidpath, 1992; Ellis et al., 1998; Odeon et al., 1999;Archambault et al., 2000; Hamers et al., 2000;Stoffregen et al., 2000; Liebler-Tenorio et al., 2002).The underlying cause of thrombocytopenia is notcompletely understood. Necrosis of megakary-ocytes, reduced production of thrombocytes bymegakaryocytes, increased consumption of throm-bocytes in the periphery, and functional defects ofthrombocytes have all been suggested as contribut-ing factors (Rebhun et al. 1989; Corapi et al., 1990;Walz et al., 1999, 2001). In some cases of acuteBVDV infection, BVDV antigen has been demon-strated in bone marrow (Marshall et al., 1996; Spag-nuolo et al., 1997). Experimental infection withBVDV strains of different virulence revealed thatnot all BVDV strains spread to the bone marrow(Liebler-Tenorio et al., 2003b). The development of

thrombocytopenia is directly related to the infectionof bone marrow with BVDV. In the bone marrow,BVDV can be detected in all cellular elements in-cluding megakaryocytes (Spagnuolo et al., 1997;Walz et al., 1999; Liebler-Tenorio et al., 2002).Megakaryocytes give rise to thrombocytes by pinch-ing off parts of their cytoplasm. Thus, infection ofmegakaryocytes appears to be crucial in the devel-opment of thrombocytopenia. It is unclear how theinfection of megakaryocytes induces a reduction inthe number of thrombocytes—e.g., whether there isa reduced production of thrombocytes and/orwhether the produced thrombocytes are functionallyaltered. Investigations by Walz et al. (1999) indicatethat both mechanisms are involved.

Bleeding (hemorrhagic diathesis) occurs onlywhen thrombocytes have reached very low numbers.Some infections with highly virulent BVDV maycause mortality early in infection. Since the bonemarrow is infected later than other lympho-hematopoietic tissues, and thrombocytopenia devel-ops after infection of the bone marrow, the death ofthe animal may occur before hemorrhagic diathesisbecomes established. This might explain the variationin frequency of bleeding observed in the field and inexperimental cases of severe acute BVDV infections.

TRANSPLACENTAL/INTRAUTERINE INFECTIONSThe main economic impact of BVDV infections iscaused by intrauterine infections resulting in repro-ductive dysfunctions (Ross et al., 1986; Kirkbride,1992; Dubovi, 1994; McGowan and Kirkland, 1995;Moennig and Liess, 1995; Muñoz et al., 1996;Rufenacht et al., 2001). This is due to the fact thatBVDV is able to infect the female and male genitaltracts, cross the placenta and infect the fetus (refer toFigure 7.2). Transplacental infection may occur inthe course of acute infections shortly before or dur-ing pregnancy, venereal infection by contaminatedsemen, and persistent infection (discussed in thissection in detail). Even acute infections withoutclinical evidence of disease may lead to infection ofthe genital tract and transplacental infection(McGowan et al., 1993). Therefore, any infectionshortly before or during pregnancy has severe con-sequences on reproductive performance.

ROLE OF VIRAL AND HOST FACTORS INTRANSPLACENTAL/INTRAUTERINEINFECTION

The outcome of transplacental/intrauterine infectiondepends primarily on the time of infection during



pregnancy and thus the gestational age of the earlyconceptus or fetus. Infections shortly before or dur-ing the first weeks of pregnancy disturb the hor-monal balance, cause reduced conception rates, andhave direct effect on the early conceptus. In laterpregnancy, the outcome of BVDV infection is influ-enced by the development of target cells, phase oforganogenesis, and development of the fetal im-mune system (Duffel and Harkness, 1985).

The sequelae of infection are also determined byviral properties, especially the biotype of BVDV.Both biotypes are able to infect genital tissues, crossthe placenta, and infect the fetus (Scott et al., 1972;Done et al., 1980; Brownlie et al., 1989; Muñoz etal., 1996; Grooms et al., 1998b, 1998c), but the re-sults of infection differ. Infection with the cp bio-type appears to cause more damage in the early con-ceptus than that with ncp BVDV (Vanroose et al.,1998). Only ncp BVDV is able to induce persistentviremia (Brownlie et al., 1989). It has been sug-gested that the ability to inhibit the induction of type1 interferon enables the ncp biotype to induce viralpersistence (Charleston et al., 2001).

Transplacental infection has been reported forboth BVDV 1 and BVDV 2 (Wittum et al., 2001).Dual infection of the fetus with both BVDV 1 andBVDV 2 has also been demonstrated (Brock andChase, 2000), although differences in the extent ofinfection were observed between BVDV 1 andBVDV 2. Since the BVDV 2 strain was of highervirulence than the BVDV 1 strain, there was widertissue distribution of the BVDV 2 strain suggestingan influence of virulence on the efficiency of virusreplication.

INFECTION DURING THE PREOVULATORYPERIOD

Acute infection of cows with BVDV during the pre-ovulatory period may result in viremia at ovulationleading to a decreased conception rate. If conceptionoccurs, the calves are normal, seronegative, andnon-viremic at parturition (McGowan et al., 1993).Both ncp and cp BVDV have been demonstrated toinfect the ovaries in experimental infections(Grooms et al., 1998b, 1998c). By immunohisto-chemistry, BVDV can be found in granulosa andstromal cells of ovaries between 8 and 30 dayspostinfection (Grooms et al., 1998b, McGowan etal., 2003). Histological examination of the ovariesreveal diffuse nonpurulent interstitial inflammation,necrosis of granulosa cells, and necrosis of follicles(Ssentongo et al., 1980; Grooms, et al. 1998c; Frayet al., 2000a; McGowan et al., 2003). Lesions in the

ovaries occur 2–6 days after estrus during the periodof seroconversion and may persist for extended timeperiods (60 days) postinoculation (Ssentongo et al.,1980; Grooms, et al. 1998c, McGowan et al., 2003).

Ovarian lesions are associated with ovarian dys-functions characterized by decreased secretion ofgonadotrophins and sex steroids, particularly prog-esterone, and absent or reduced preovulatoryluteinizing hormone surge (Fray et al., 2000a, 2002;McGowan et al., 2003). The hormonal imbalanceleads to decreased growth of ovarian follicles, re-duced ovulation rate, reduced numbers of corporalutea, reduced numbers of collected embryos, andincreased numbers of unfertilized ova (Grahn et al.,1984; Grooms et al., 1998a; McGowan et al., 2003).These changes last as long as the lesions persist, atleast the first two estrous cycles after infection. Thisexplains the reduced conception rates and the delaysin conception (Grooms et al., 1998a).

Ovarian dysfunction is one way that BVDV infec-tions decrease conception rates. Infection of oocytesduring infection of ovaries will also impair fertility(Bielanski et al., 1998). The infection of ovaries ismultifocal and does not affect all oocytes.Therefore, cows that conceive despite infection pro-duce normal, seronegative, and non-viremic off-spring (McGowan et al., 1993). Up to 58% of em-bryos collected for in vitro fertilization fromtransiently infected cows were infected with BVDV(Bielanski et al., 1998). Infection of the embryo mayresult in slightly reduced survival rate and retardeddevelopment of the embryo (Archbald et al., 1979;Bielanski and Hare, 1988; Kafi et al., 2002). BVDVcan be isolated from retarded embryos (Kirkland etal., 1993).

EXPOSURE OF THE UTERUS TO BVDV ATINSEMINATION

Exposure of the uterus to BVDV during estrus at in-semination causes a decreased conception rate innaive cattle, and seropositive cows have normal con-ception rates (McClurkin et al., 1979; Whitmore etal., 1981; McGowan et al., 1993; Kirkland et al.,1994). This type of exposure might occur whencows are inseminated with semen contaminatedwith BVDV or contract acute BVDV infection be-cause of animal movement and mixing around thetime of insemination. Experimentally, infection canbe reproduced by direct intrauterine instillation ofBVDV (Whitemore et al., 1981; Grahn et al., 1984).

Semen contaminated with BVDV may originatefrom persistently infected bulls, since not all PI bullsare infertile or produce semen of inferior quality

Pathogenesis 127


(Meyling and Jensen, 1988; Wentink et al., 1989;Kirkland et al., 1994). On the other hand, infectedsemen may also be produced following transient in-fection of postpuberal bulls (Kirkland et al., 1991;Voges et al., 1998; Niskanen et al., 2002a; Givens etal., 2003). Even after transient infection and clear-ance of the virus from other tissues and blood,BVDV can persist in semen for months after expo-sure (Kirkland et al., 1991; Voges et al., 1998; Nis-kanen et al., 2002a; Givens et al., 2003). The mostproductive sites of viral replication in the male gen-ital tract are the seminal vesicles and prostate glandscausing excretion in the seminal fluid (Kirkland etal., 1991). Insemination with infected semen willcause transient infection of naive females (Kirklandet al., 1994).

The effect of BVDV on gametes and the earlyconceptus were investigated. In vitro investigationsrevealed no immediate effect of ncp and cp BVDVon the maturation of oocytes to blastocysts, althoughBVDV replicated well in cells around the embryo(Kirkland et al., 1996; Tsuboi and Imada, 1996; Kafiet al., 2002). In vitro fertilization of oocytes in thepresence of BVDV resulted in reduced fertilizationrates (Kafi et al., 2002). There was no uptake ofBVDV by embryos after in vitro exposure (Potter etal., 1984). It has been suggested that zona pellucidaprotects the embryo against infection with BVDV(Singh et al., 1982). After in vitro removal of zonapellucida, 2- and 4-cell embryos were not suscepti-ble to infection with ncp BVDV (Tsuboi and Imada,1999) and 8-cell embryos were susceptible to cpBVDV (Vanroose et al., 1998). Even extended expo-sure of embryos with removed zona pellucida hadno negative influence on embryonic survival (Bie-lanski and Hare, 1988). With progressive develop-ment, the embryos became increasingly susceptibleto infection (Bak et al., 1992, Vanroose et al., 1998;Tsuboi and Imada, 1999).

FETAL INFECTION

After about 30 days of gestation the fetus is suscep-tible to transplacental infection. There appears to bean increasing efficiency of infection from concep-tion to day 30 postconception (Kirkland et al.,1993). This corresponds with a progressive increasein the loss of pregnancies observed between days 20and 77 after artificial insemination. Development ofplacenta with the formation of placentomes, forma-tion of trophoblasts and development of fetal tissuesmost likely determine the efficiency of infection(Kirkland et al., 1993).

Experimental infections or vaccine trials using

modified live vaccines have shown that fetal infec-tions may result in abortion, persistent infection,stillborn calves, teratogenic effects, and normalcalves born with antibodies to BVDV. Althoughboth biotypes of BVDV can infect the fetus, onlyncp BVDV is able to establish persistent infection(Brownlie et al., 1989). Infection with cp BVDV re-sults either in abortion or healthy seropositive calves(Brownlie et al., 1989). The fetus, before it becomesfully immune-competent, is affected severely by in-terauterine BVDV infection. The intrauterine infec-tion of an immune-competent fetus results in a tran-sient infection of the fetus comparable to acuteBVD. The fetus is able to mount an immune re-sponse and the virus is cleared. Calves are clinicallynormal and non-viremic. The presence of antibodiesto BVDV in serum samples collected before colos-trum uptake indicates intrauterine exposure toBVDV. Neutralizing precolostral antibodies havebeen detected in newborn calves after infectionsfrom day 90 to the end of gestation (Kendrick 1971;Orban et al., 1983; Liess et al., 1987).

Abortions

It was recognized early that BVDV may induceabortions (Kahrs and Ward, 1967; Casaro et al.,1971; Kendrick, 1971; Done et al., 1980; Duffell etal., 1984). Abortions occur most frequently in theearly stages of gestation (less than 125 days of ges-tation), although abortions in the late phase of ges-tation have also been described (Duffel et al., 1984;Roeder et al., 1986). The expulsion of the fetus oftenoccurs not immediately after infection but weeks tomonths later (Duffel et al., 1984). This also explainsthe occurrence of mummified fetuses (Scott et al.,1973; Done et al., 1980).

The pathogenesis of abortion is not clear. Most in-vestigations have reported unremarkable, nonspe-cific lesions of the placenta and placentomes ofaborted calves (Casaro et al., 1971; Done et al.,1980; Murray, 1991) but Baszler et al. (1995) founda necrotizing placentitis associated with the pres-ence of viral antigen. It has been speculated that themild placental lesions might allow opportunistic in-fections to cross the placenta. BVDV antigen waspresent in numerous tissues of aborted fetuses, andinfiltrates of mononuclear cells were found in sev-eral tissues including lung and myocardium (Doneet al., 1980; Murray, 1991; Baszler et al., 1995).Therefore, damage to the fetus is considered as im-portant initiator for abortion.

Tissues from pregnant cows infected with an ncpBVDV at 85–86 day of gestation were examined for



viral antigen by immunohistology (Fredriksen et al.,1999a). BVDV was present in the fetuses at 14 dpiin the lung and liver and at 18–22 dpi in all other tis-sues examined (lung, liver, spleen, intestine, andbrain). The intercotyledonary fetal membranes andthe placentomes contained viral antigen at days 18and 22 postinoculation. BVDV antigen was not de-tected in maternal tissues or in the placenta betweendays 7 and 22 postinoculation. This indicates thatfetal infection can take place without high concen-trations of BVDV in the uterus or placenta ofacutely infected cows and that this is a more likelycause of abortion than replication of the virus in theplacenta.

Persistent infections

Abortion is not the only outcome of infection in thefirst trimester of gestation. Infection of the fetus withan ncp BVDV may result in the birth of a PI animal.Generation of PI animals peaks when infection oc-curs from about 30 until 90 days of gestation and be-comes less frequent as the fetus approaches 125 daysof gestation (Roeder et al., 1986; Radostits andLittlejohns, 1988; Moennig and Liess, 1995). Persis-tent infection has been established with both BVDV1 and BVDV 2. In PI fetuses, BVDV is widely dis-tributed throughout the organs (Fredriksen et al.,1999b). Damage to the fetus is limited and thuspregnancy is maintained. The fetus is immune-incompetent at the time of infection and mounts no re-action to the infecting BVDV. The continued presenceof large amounts of virus throughout fetal tissues to-larizes the fetus, when immune competence sets in.Thus the infecting strain of BVDV is not recognizedas foreign and there is no immune response to it.

The effect of the persisting BVDV on the fetusand newborn varies (McClurkin et al., 1979; Liess etal., 1984). Persistent fetal infection may result in thebirth of normal calves, apparently normal calveswith different functional defects or calves with moreovert defects. Most have been characterized as“poor-doers.” They may be stillborn or may diewithin the first few hours or days of life. Survivingcalves are often stunted and have growth retardation.In the first year of life, up to 50% higher death ratewas found in PI calves when compared to calves notinfected (Duffel and Harkness, 1985). Even appar-ently normal calves have a higher death rate than un-infected calves, because they are predisposed to in-fections and have a higher risk of severe illness(Barber et al., 1985; Werdin et al., 1989; Muñoz-Zanzi et al., 2003). This is most likely due to func-tional defects of the immune system causing im-

mune suppression (Roth et al., 1981, 1986). Persis-tent infection may also cause functional defects ofthe endocrine system: Diabetes mellitus has repeat-edly been reported in PI animals (Taniyama et al.,1995; Buckner, 1997; Murondoti et al., 1999). It hasbeen suggested that different clinical signs associ-ated with persistent infection are predominantly in-fluenced by the time when the fetus becomes in-fected with earlier infections, resulting morefrequently in the birth of normal, healthy virus car-riers (Moennig and Liess, 1995).

Teratogenic effects

Congenital defects have been described after infec-tion between days 80 and 150 of gestation. The fetusappears to be the most susceptible to teratogeniceffects of BVDV when the immune competence be-gins to develop, the ability to mount an inflamma-tory response sets in, and organogenesis is not com-pleted (Duffel and Harkness, 1985).

Intrauterine growth retardation and reduced matu-ration are observed in numerous tissues, includingbrain, thymus, muscles, bone, and lung (Done et al.,1980). Lesions in the brain and eyes are most fre-quent because they are in the final stages of organo-genesis during day 100 and 150 of gestation (Duffeland Harkness, 1985). Alterations affecting both or-gans are referred to as oculocerebellar syndrome(Bielefeldt Ohmann, 1984). The following malfor-mations of the central nervous system have been re-ported: micrencephaly, cerebellar hypoplasia oraplasia, porencephaly, hydranencephaly, hydro-cephalus internus, or dysmyelogenesis (Ward, 1969;Kahrs et al., 1970 a,b; Casaro et al., 1971; Brown etal., 1973, 1974; Scott et al., 1973; Done et al., 1980;Trautwein et al. 1986). Neurological signs of tremor,ataxia, torticollis, or opisthotonus may be seen innewborn calves (Liess et al., 1984). The ocular ab-normalities are characterized by reduced pigmenta-tion of the retina, retinal atrophy, optic neuritis,cataract, microphthalmia, and retinal dysplasia(Kahrs et al., 1970b, Brown et al., 1975). Anotherfrequently affected organ is the thymus, which hasreduced size at parturition (Done et al., 1980). Othercongenital defects are arthrogryposis and alopecia/hypotrichosis (Casaro et al., 1971; Kendrick, 1971;Bielefeldt-Ohmann, 1984). Frequently BVDV can-not be isolated from or detected in non-PI animalswith BVDV-related malformations. In the few caseswhere BVDV is detected, it is often found locally inthe brain or in the cerebrospinal fluid only (Roederet al., 1986; Trautwein et al., 1987; Liess et al.,1987). Some of the newborns with BVDV-related

Pathogenesis 129


malformations are seropositive for BVDV at birth(Liess et al., 1984).

PERSISTENT INFECTIONSThe ability to establish prolonged replication ofvirus in vivo is important both for the epidemiologyand the pathogenesis of BVDV infections. There aretwo forms of prolonged viral replication. The firstform is characterized by viral persistence after in-trauterine infection, which results in animals thatnever clear the virus and thus shed the virus overtheir entire life span. The second form occurs insome, but not all, animals following acute infection.Such animals shed virus for weeks and months fol-lowing infection. However, these animals do mountan immune response to the virus and do clear thevirus from most tissues over time.

VIRUS PERSISTENCE FOLLOWINGINTRAUTERINE EXPOSURE

The more important form of prolonged viral replica-tion results from intrauterine infection of an im-mune-incompetent fetus because

• PI animals are the main source of virus for es-tablishing acute infections.

• PI mothers are important in establishing PI fami-lies by continuous vertical infection.

• Persistent infection is important in the pathogen-esis of mucosal disease.

PI animals are characterized by a wide distribu-tion of BVDV throughout their organs, no virus-associated morphological lesions, and no immune re-sponse to the persisting BVDV strain. The immunetolerance is selectively restricted to the particularstrain of BVDV that has caused the intrauterine in-fection. Infections with other strains of BVDV willinduce an immune response (Fulton et al., 2003).Even single amino acid differences are sufficient forthe PI animal to recognize and mount a response toanother BVDV strain (Collen et al., 2000). PI ani-mals have normal cellular and humoral responses toother antigens (Houe and Heron, 1993). Recent in-vestigations have centered on immunological mech-anisms involved in maintaining viral persistence. Itwas concluded that tolerance is maintained by nonre-active CD4+ T-lymphocytes, and B-lymphocytes andantigen-presenting cells are normally reactive (Frayet al., 2000b; Glew and Howard, 2001).

PI animals as source of virus for acute infection

Some PI calves are born without clinical signs of thedisease and without any macroscopic or micro-

scopic lesions (Done et al., 1980; Binkhorst et al.,1983; Roeder et al., 1986). These clinically incon-spicuous PI animals are an important source for hor-izontal and vertical infections. It was demonstratedexperimentally that transmission of BVDV to naiveanimals was much more efficient from PI animalsthan from those with acute infections (Traven et al.,1991; Niskanen et al., 2000). In PI animals, BVDVis present throughout all organs and tissues (Mey-ling, 1970; Bielfeldt Ohmann, 1988; Liebler et al.,1991), and thus large scale virus shedding occursfrom multiple sites in PI animals. Because BVDV ispresent in the epidermis; salivary glands; nasalglands; lining of the oral and nasal cavity; and mu-cosa and accessory glands of the digestive tract, uro-genital tract, and mammary glands, the virus can befound in sloughed-off skin cells, oronasal fluids,feces, urine, semen, and colostrum/milk of PI ani-mals. Viral antigen is present in the blood althoughlevels of viremia may change (Brock et al., 1998);thus iatrogenic transmission by contaminated nee-dles or surgical instruments is possible.

Vertical infection

PI cows are able to conceive and give birth to livecalves. They have a reduced reproductive perform-ance, which is caused by significant morphologicalchanges in the ovaries leading to reduced follicularmaturation (Grooms et al., 1996). Calves born to PIcows are always persistently viremic; this estab-lishes persistent infection in the next generation,thus forming PI families (Straver et al., 1983).

BVDV transmission may occur at the level ofoocytes, which can be antigen-positive (Fray et al.,1998). On the other hand, since only about 20% offollicles are positive for BVDV in PI cows (Fray etal., 1998), virus-free calves can be obtained afterembryo transfer from PI donors (Wentink et al.,1991; Bak et al., 1992; Smith and Grimmer, 2000).If the oocytes are not infected, infection of the fetuswill occur when it becomes susceptible in the earlyphase of gestation by the continuous presence ofBVDV in the uterus, placentomes, and fetal mem-branes of pregnant PI cows (Fredriksen et al.,1999b). Apparently, pregnancy even enhances theamount of BVDV in the genital tissues (Fredriksenet al., 1999b).

PROLONGED VIRAL SHEDDING FOLLOWINGACUTE INFECTION

A prolonged period of viral shed may develop afteracute infection of immune-competent animals.Immune-suppressive treatment or stress during in-



fection may lead to an extended presence of virus inthe host for up to 80 days postinoculation (Sandviket al., 2000). Sometimes virus is retained in certainimmunologically privileged sites although a sys-temic immune response is mounted and viruscleared from most tissues. This type of prolongedviral replication has been described for the malegenital system where semen contaminated withBVDV was found for up to 7 months after acute in-fection (Kirkland et al., 1991; Voges et al., 1998;Niskanen et al., 2002a; Givens et al., 2003).

MUCOSAL DISEASEMucosal disease (MD) is the most dramatic clinicalevidence of BVDV infection and causes characteris-tic lesions. The name mucosal disease was coinedby Ramsey and Chivers (1953). The authors ob-served a virus infection in cattle that caused severeerosions, ulcerations, and hemorrhages of the mu-cosal surfaces of nuzzle, oral cavity, esophagus,forestomachs, abomasum, and the small and largeintestines. Further examination revealed severe de-pletion of spleen, lymph nodes, and thymus. Thesefindings are in accordance with later descriptions byother authors (Schulz, 1959; Bielefeldt Ohmann,1983; Bolin et al., 1985b; Wilhelmsen et al., 1991).The disease is fatal and the affected animals usuallydie within 2 weeks after onset of clinical symptoms.

PATHOGENESIS OF MUCOSAL DISEASE

Gillespie et al. (1960) recognized that the virus ofmucosal disease was identical with one that causesbovine virus diarrhea (Olafson et al., 1946). Inocula-tion of immune-competent cattle with BVDV-induced acute BVDV infections and not MD. Al-though acute BVDV infections infrequently resultedin severe disease with mucosal lesions, they tended tobe much more limited and less severe than MD. Thepathogenesis of MD remained obscure for severaldecades. It was only when it was recognized that viralpersistence after intrauterine infection of the fetuswith ncp BVDV was a prerequisite for the develop-ment of MD (Coria and McClurkin, 1978; McClurkinet al., 1984) that researchers were able to induce MD.The first two research teams to induce MD experi-mentally used slightly different approaches:

• One group inoculated pregnant naive cows dur-ing the first trimester of pregnancy with the ncpBVDV of an ncp/cp BVDV pair isolated from ananimal with MD. This resulted in the birth of PIcalves (McClurkin et al., 1984). When thesecalves were infected with the cp part of the

BVDV pair, they died from MD (Bolin et al.,1985b).

• The other group identified a herd that containeda number of PI animals. One of these animalsdeveloped spontaneous MD. A cp BVDV wasisolated from the MD animal. When the remain-ing PI animals were inoculated with the cpBVDV isolate from the diseased animal, theydeveloped the characteristic signs of MD anddied within 2–3 weeks (Brownlie et al., 1984).

Several earlier experiments had revealed that notall experimental infections of PI animals with cpBVDV resulted in MD (Liess et al., 1974; Harknesset al., 1984). The comparison of ncp/cp BVDV pairsfrom animals with MD showed a very close anti-genic relationship between ncp and cp of most viruspairs (Howard et al., 1987). Therefore it was hypoth-esized that only cp BVDV with close antigenic re-semblance (homologous strains) can induce MD(Brownlie and Clarke, 1993; Nakajima et al., 1993).In experimental infections with in vitro selected cpBVDV strains, this could not be confirmed, althoughclose homology appears to favor the development ofMD as demonstrated in the following experiments:

• Cytopathic BVDV strains with similar, but notcompletely identical, reactivity patterns to ncpviruses isolated from PI animals were selected.Reactivity patterns were determined based onbinding of monoclonal antibodies producedagainst the immune dominant surface glycopro-tein E2. When the PI animals were inoculatedwith the selected “homologous” cp BVDV, MDwas induced in all animals within 2–3 weeks(Moennig et al., 1990).

• PI animals were inoculated with semihomolo-gous cp BVDV strains. Only some animals suc-cumbed to MD (Bruschke et al., 1998b; Loehr etal., 1998)

• PI animals developed MD after inoculation withan antigenically different cp BVDV (Sentsui etal., 2001).

This led to the hypothesis that in addition to thehomology between ncp and cp BVDV, other factorsinfluence the outcome of infection (Bruschke et al.,1998b). Under experimental conditions, MD was re-produced by nasal or intravenous inoculation withcp BVDV. Under field conditions, horizontal infec-tion is possible, especially from animals in the finalstages of MD, which shed both biotypes of BVDV.Endogenous development of cp BVDV in PI ani-mals has been discussed as an alternative way of in-

Pathogenesis 131


fection (Brownlie and Clarke, 1993; Fricke et al.,2001). When homologous cp BVDV has developedin a PI animal, it will develop MD. During the dis-ease, cp BVDV is excreted and other herd membersare infected. Seropositive animals are immune,seronegative animals will develop acute BVD, andother PI herd members infected with the same ncpBVDV will develop MD.

EARLY AND LATE ONSET MUCOSALDISEASE

In initial experiments, the first signs of MD devel-oped within 2–3 weeks after inoculation with cpBVDV. However, later reports described the devel-opment of MD several months after inoculation withcp BVDV (Brownlie and Clarke, 1993). It was pos-tulated that the course of infection was influencedby the degree of homology between the persistingncp BVDV and the cp BVDV (Brownlie and Clarke,1993). Experimental infections of PI animals usingcp BVDV strains of variable degrees of antigenichomology as determined by monoclonal antibodyreactivity pattern showed that the course of diseaseis not predictable (Loehr et al., 1998). The followingpattern was observed in animals developing MDmonths to years after inoculation with cp BVDV. Aninitial, transient acute BVDV infection with the cp

BVDV was followed by a variably long phase with-out clinical signs and neutralizing antibodies to thecp BVDV. The final phase of infection was denotedby a sudden onset of characteristic signs of MD(Loehr et al., 1998).

Moennig et al. (1993) compared the cp BVDVused for inoculation and the cp BVDV re-isolatedfrom the sick animal from animals that had suc-cumbed to MD months to years after experimentalinfection. It was found that the re-isolated cp BVDVwas a recombination of the cp BVDV used for inoc-ulation with the persisting ncp BVDV. The E2-specific monoclonal antibody reactivity pattern ofthe re-isolated cp BVDV resembled that of the per-sisting ncp BVDV. However, it expressed the ge-netic marker of the cp BVDV used for inoculation.The recombinational events were confirmed on amolecular basis by sequencing the E2 and NS2/3proteins of the viruses involved (Fritzemeier et al.,1995, 1998). The same phenomenon has been de-scribed when modified live vaccines containing cpBVDV are used (Ridpath and Bolin, 1995). Such re-combinational events may even occur betweenBVDV 1 and BVDV 2 (Ridpath and Bolin, 1995).

Based on these findings early onset and late onsetMD can be distinguished as follows (Figure 7.4):Early onset MD occurs within 2–3 weeks after ex-


Figure 7.4. Pathogenesis of mucosal disease.


posure to a cp BVDV strain, and the re-isolated cpBVD is identical to the one the animal was exposedto. Late onset MD occurs months to years after ex-posure and the re-isolated cp BVDV is a recombina-tion between the persisting ncp and the original cpBVDV. This distinction is in general possible onlyunder experimental conditions, because in mostfield cases the identity of the original infecting cpBVDV cannot be determined and the time of expo-sure cannot be pinpointed. Clinical signs in the endphase of early onset and late onset MD are indistin-guishable and there are only subtle differences in tis-sue lesions of animals that have succumbed to thetwo forms of MD (Moennig et al., 1993; Liebler-Tenorio et al., 2000).

PATHOGENESIS OF LESIONS IN MUCOSALDISEASE

Correlation between viral antigen and tissuelesions

In field cases of MD, BVDV antigen can be demon-strated by immunohistochemistry in numerous or-gans and tissues (Meyling, 1970; Bielefeldt Oh-mann, 1983). The distribution of viral antigen doesnot correlate well with tissue lesions, since in gen-eral no distinction can be made between the antigenof the ncp and cp BVDV. The ncp BVDV present inPI animals in a wide distribution is still found in an-imals after they have developed MD, but is not asso-ciated with tissue lesions (Bielfeldt Ohmann, 1988).Under experimental conditions, it is possible to se-lectively demonstrate the distribution of cp BVDVas opposed to the persistent ncp BVDV (Liebler etal., 1991). These conditions require that the cpBVDV used for inoculation differs in binding to oneor more monoclonal antibodies from the persistingBVDV. Thus it can be recognized selectively by spe-cific monoclonal antibodies in tissue sections usingimmunohistochemistry. Further studies indicatedthat the organ distribution patterns for the cp BVDVwere identical whether immunohistochemistry forE2 or whether molecular detection of NS3 proteincharacteristic for the cp BVDV was used (Greiser-Wilke et al., 1993). This suggests that results ob-tained by immunohistochemistry reflect the distri-bution of cp BVDV in the tissues realistically.

Examination of animals with MD revealed that cpBVDV was predominantly detected in sites wheretissue destruction was observed (Liebler et al.,1991). The sites included

• Depleted lymphoid follicles of tonsils, lymphnodes, spleen, mucosa-associated lymphoid

tissues (gut-associated lymphoid tissues,bronchus-associated lymphoid tissues)

• Thymic cortex• Altered mucosa of the upper digestive tract, and

(iv) crypt epithelia of small and large intestine

These observations suggest that cp BVDV is notonly important for the pathogenesis of MD, but alsofor the manifestation of MD at mucosal surfaces andin lymphoid tissues. It appears that the replication ofcp BVDV is associated with marked lesions. Thecomparison of tissue lesions and distribution of viralantigen at consecutive times after inoculation of PIcattle with cp BVDV revealed a spreading pattern forthe cp BVDV in MD that was similar to the one de-scribed for acute BVDV infections either with cp orncp BVDV (Liebler-Tenorio et al., 1997) (Figure 7.5).

After intranasal inoculation, primary infection ofthe tonsillar epithelium and virus replication in thetonsil is observed. The cp BVDV then spreads fromthe tonsils to the regional lymph nodes. With littledelay, cp BVDV can also be detected in Peyer’spatches, mucosa-associated lymphoid follicles inthe large intestine, and lymphoid follicles of thelung and nasal mucosa. In mucosa-associated lym-phoid tissues, the cp viral antigen is primarily foundin the lymphoid tissue and only later in the epithe-lium. This pattern differs from most enteric patho-gens. It suggests a hematogeneous spread of cpBVDV to the gastrointestinal and respiratory tractmucosa. cp BVDV spreads to a lesser degree to theperipheral lymph nodes, thymus, and spleen.

Infection of intestinal mucosa occurs initially inclose association with mucosa-associated lymphoidtissues by local expansion of the infection. In theprogressed phase of MD, cp BVDV is found multi-focally and finally diffusely in the intestinal mucosa.The multifocal infection is most likely caused bymultifocal immigration of infected lymphocytes intothe mucosa. A large number of these cells might berecirculating from Peyer’s patches, since homing isa known phenomenon of the Peyer’s patch lympho-cytes. The preferential infection of the intestinal mu-cosa versus other mucosal surface is most likely in-fluenced by the large amount of mucosa-associatedlymphoid tissue in the intestine and the high num-bers of recirculating lymphocytes. When cp BVDVis present multifocally or diffusely in the intestinalmucosa, the first signs of diarrhea are observed.

Lesions in lymphoid tissues

In the early phase of MD the increased frequency ofcell death by apoptosis and the changes in prolifer-

Pathogenesis 133


ation are key players for the development of tissuelesions both in lymphoid tissues and epithelium. Intonsils, lymph nodes and Peyer´s patches, cpBVDV is predominantly present in the lymphoidfollicles. Initially it is associated with folliculardendritic cells (FDCs) and B-lymphocytes. There isa marked reduction in the number of proliferatingB-lymphocytes and a marked increase in apoptoticcells. The immature B-lymphocytes in the darkzone of the germinal center, which have a very highproliferative activity, are especially affected(Liebler-Tenorio and Pohlenz, 1997). Transmissionelectron microscopy reveals a loss of B-lympho-cytes, whereas FDCs often do not have signs of de-generation (Teichmann et al., 2000). If the loss ofB-lymphocytes is very abrupt, death of FDCs is ob-served leaving empty follicular centers. Histo-logically this presents as cyst formation, a frequentfinding in mucosa-associated lymphoid tissues ofthe intestine, which has led to the descriptive diag-nosis of “colitis cystica.” If depletion of lympho-cytes develops more slowly, follicles consisting ex-clusively of FDCs are found.

Cp BVDV was also detected in dendritic cells(DCs) in the interfollicular areas. DCs can be in-fected with virus and may serve as carriers for viralproteins (Sprecher and Becker, 1993). Since DCsare able to present antigen very effectively to

T-lymphocytes, they may induce a cell-mediated im-mune response. Further investigation of T-cellsubsets in the early phase of MD revealed accu-mulations of CD4+ T-lymphocytes particularly atsites where lesions develop and it was speculatedabout possible cytotoxic effects mediated by CD4+T-lymphocytes (Frink et al., 2002).

Lesions in mucosa-associated lymphoid tissues

Besides changes within the lymphoid follicles, theloss of B-cells also alters the morphology of domes(Liebler et al., 1995). The loss of domes, and conse-quently of the follicle-associated epithelium, im-pedes the selective uptake of antigens from the in-testinal lumen. Thus MD causes severe impairmentof the inductive part of the intestinal mucosal im-mune system.

The changes in the gut-associated lymphoid tissuecause not only local tissue destruction, but also af-fect the whole intestinal mucosa (Liebler et al.,1996). In the mucosa of the small and large intes-tines of animals moribund with MD, the number ofIgA- and IgM-positive plasma cells in the laminapropria is severely reduced. This is most likely dueto the depletion of mucosa-associated lymphoid fol-licles, which are the sites where precursors of theplasma cells in the lamina propria are generated.The reduced number of plasma cells causes reduced


Figure 7.5. Spread of cp BVDV in mucosal disease.


production of secretory immunoglobulins essentialfor the protection of the mucosal surface.

The number of intraepithelial lymphocytes is alsoreduced (Liebler et al., 1996). Since they are alsoderived from Peyer´s patches in part, their reducednumbers might reflect decreased production. On theother hand, the altered microenvironment of the ep-ithelium might also influence their presence be-tween the epithelial cells. In conclusion, all parts ofthe mucosal immune system are severely altered inMD—e.g., FAE, domes, lymphoid follicles, plasmacells in the lamina propria, and T-lymphocytes in theepithelium.

Lesions in mucosa

The intestinal mucosa becomes infected later thanthe Peyer´s patches and mucosa-associated lym-phoid follicles in the large intestine (Liebler-Tenorioet al., 1997). Enterocytes are not infected from theluminal surface as in most enteric viral infections,but from the basolateral surface. Crypt epithelium isparticularly affected. Initially there is a coincidenceof three phenomena in the crypts; focal infection ofa few epithelial cells with cp BVDV, increased num-ber of apoptotic epithelial cells, and increase of ep-ithelial proliferation rate (Liebler-Tenorio andPohlenz, 1997). The induction of apoptosis in thecrypt epithelium appears to play a key role in the de-struction of the intestinal mucosa. In affected crypts,the proliferation rate is severely reduced, and thereis a reactive increase of proliferating cells in adja-cent crypts.

A common finding in moribund animals in thelate phase of MD is the multifocal to diffuse infec-tion of the intestinal mucosa with cp BVDV. In themucosa of these animals, an increasing number ofcrypts is without proliferating epithelial cells. Assoon as the continuous cellular loss of aged cellsfrom the villous tips is not compensated by cryptproliferation, mucosal atrophy and ulceration de-velop. At this stage of infection, clinical symptomsof diarrhea become apparent. The progressive le-sions explain the lack of success in treating appar-ently sick animals.

CONCLUSIONSMany pieces of the complex puzzle of BVDV patho-genesis have been put in place over the last decades,but important pieces are still missing. Observationsfrom experimental infections and naturally occuringdisease continue to create doubt about some of theconcepts of pathogenesis. Currently, advances arebeing made in elucidating the interaction of BVDV

infection with different components of the specificimmune responses (Beer et al., 1997; Rhodes et al.,1999; Collen and Morrison, 2000; Collen et al.,2002), as well as with cells and mediators of the in-nate immune system (Adler et al., 1994, 1996;Baigent et al., 2002; Ganheim et al., 2003; Peterhanset al., 2003). Another approach focuses on the inter-action of BVDV with its target cells—e.g., howBVDV replication influences gene expression in af-fected cells (Neill and Ridpath, 2003), in whichcompartments of the cell BVDV proteins are ex-pressed (Grummer et al., 2001); which receptors areused by BVDV to interact with cells (Minocha et al.,1997; Schelp et al., 2000); and how cp strains areable to exert their cp effect on certain, but not all,cells (Perler et al., 2000; Schweizer and Peterhans,1999, 2001; Grummer et al., 2002; Bendfeldt et al.,2003). Hopefully some of the answers will yield fur-ther insight into the pathogenesis of BVDV andallow better control of this economically importantdisease of ruminants.

REFERENCESAdler H, Frech B, Meier P, et al.: 1994, Noncyto-

pathic strains of bovine viral diarrhea virus primebovine bone marrow-derived macrophages for en-hanced generation of nitric oxide. Biochem BiophysRes Commun 202:1562–1568.

Adler H, Jungi TW, Pfister H, et al.: 1996, Cytokineregulation by virus infection: Bovine viral diarrheavirus, a flavivirus, downregulates production oftumor necrosis factor alpha in macrophages invitro. J Virol 70:2650–2653.

Archambault D, Beliveau C, Couture Y, Carman S:2000, Clinical response and immunomodulationfollowing experimental challenge of calves withtype 2 noncytopathogenic bovine viral diarrheavirus. Vet Res 31:215–227.

Archbald LF, Fulton RW, Seger CL, et al.: 1979,Effect of the bovine viral diarrhea (BVD) virus onpreimplantation embryos. Theriogenology 11:81–89.


Bak A, Callesen H, Meyling A, Greve T: 1992, Calvesborn after embryo transfer from donors persistentlyinfected with BVD virus. Vet Rec 131:37.

Baker JC: 1995, The clinical manifestations of bovineviral diarrhea virus infection. Vet Clin North AmFood Anim Pract 11:425–445.

Barber DM, Nettleton PF, Herring JA: 1985, Diseasein a dairy herd associated with the introduction and

Pathogenesis 135


spread of bovine virus diarrhea virus. Vet Rec117:459–464.

Barkema HW, Bartels CJ, van Wuijckhuise L, et al.:2001, Outbreak of bovine virus diarrhea on Dutchdairy farms induced by a bovine herpesvirus 1marker vaccine contaminated with bovine virus di-arrhea virus type 2. Tijdschr Diergeneeskd126:158–165.

Baszler TV, Evermann JF, Kaylor PS, et al.: 1995,Diagnosis of naturally occurring bovine viral diar-rhea virus infections in ruminants using monoclonalantibody-based immunohistochemistry. Vet Pathol32:609–618.

Beer M, Wolf G, Pichler J, et al.: 1997, Cytotoxic T-lymphocyte responses in cattle infected with bovineviral diarrhea virus. Vet Microbiol 58:9–22.

Bendfeldt S, Grummer B, Greiser-Wilke I: 2003, Nocaspase activation but overexpression of Bcl-2 inbovine cells infected with noncytopathic bovinevirus diarrhea virus. Vet Microbiol 96:313–326.

Bielanski A, Hare WC: 1988, Effect in vitro of bovineviral diarrhea virus on bovine embryos with thezona pellucida intact, damaged and removed. VetRes Commun 12:19–24.

Bielanski A, Sapp T, Lutze-Wallace C: 1998,Association of bovine embryos produced by invitro fertilization with a noncytopathic strain ofbovine viral diarrhea virus type II. Theriogenology49:1231–1238.

Bielefeldt Ohmann H: 1983, Pathogenesis of bovineviral diarrhea-mucosal disease: Distribution andsignificance of BVDV antigen in diseased calves.Res Vet Sci 34:5–10.

Bielefeldt Ohmann H: 1984, An oculo-cerebellar syn-drome caused by congenital bovine viral diarrheavirus-infection. Acta Vet Scand 25:36–49.

Bielefeldt Ohmann H: 1988, BVD virus antigens intissues of persistently viremic, clinically normalcattle: Implications for the pathogenesis of clini-cally fatal disease. Acta Vet Scand 29:77–84.

Binkhorst GJ, Journee DL, Wouda W, Straver PJ, VosJH: 1983, Neurological disorders, virus persistenceand hypomyelination in calves due to intra-uterineinfections with bovine virus diarrhea virus. I.Clinical symptoms and morphological lesions. VetQ 5:145–155.

Bolin SR, McClurkin AW, Coria MF: 1985a, Effectsof bovine viral diarrhea virus on the percentagesand absolute numbers of circulating B and T lym-phocytes in cattle. Am J Vet Res 46:884–886.

Bolin SR, McClurkin AW, Cutlip RC, Coria MF:1985b, Severe clinical disease induced in cattle per-sistently infected with noncytopathic bovine viraldiarrhea virus by superinfection with cytopathic bo-vine viral diarrhea virus. Am J Vet Res 46:573–576.

Bolin SR, Ridpath JF: 1992, Differences in virulencebetween two noncytopathogenic bovine viral diar-rhea viruses in calves. Am J Vet Res 53:2175–2163.

Bolin SR, Ridpath JF: 1998, Prevalence of bovineviral diarrhea virus genotypes and antibodiesagainst these genotypes in fetal bovine serum. J VetDiagn Invest 10:135–139.

Brock KV, Chase CC: 2000, Development of a fetalchallenge method for the evaluation of bovine viraldiarrhea virus vaccines. Vet Microbiol 77:209–214.


Brodersen BW, Kelling CL: 1998, Effect of concur-rent experimentally induced bovine respiratory syn-cytial virus and bovine viral diarrhea virus infectionon respiratory tract and enteric diseases in calves.Am J Vet Res 59:1423–1430.

Brodersen BW, Kelling CL: 1999, Alteration of leuko-cyte populations in calves concurrently infectedwith bovine respiratory syncytial virus and bovineviral diarrhea virus. Viral Immunol 12:323–334.

Brown TT, Bistner SI, de Lahunta A, et al.: 1975,Pathogenetic studies of infection of the bovine fetuswith bovine viral diarrhea virus. II. Ocular lesions.Vet Pathol 12:394–404.

Brown TT, de Lahunta A, Bistner SI, et al.: 1974,Pathogenetic studies of infection of the bovine fetuswith bovine viral diarrhea virus. I. Cerebellar atro-phy. Vet Pathol 11:486–505.

Brown TT, de Lahunta A, Scott FW, et al.: 1973,Virus induced congenital anomalies of the bovinefetus. II. Histopathology of cerebellar degeneration(hypoplasia) induced by the virus of bovine viraldiarrhea-mucosal disease. Cornell Vet 63:561–578.

Brownlie J, Clarke MC: 1993, Experimental andspontaneous mucosal disease of cattle: A validationof Koch’s postulates in the definition of pathogene-sis. Intervirology 35:51–59.


Brownlie J, Clarke MC, Howard CJ: 1989, Experi-mental infection of cattle in early pregnancy with acytopathic strain of bovine virus diarrhea virus. ResVet Sci 46:307–311.

Bruschke CJ, Haghparast A, Hoek A, et al.: 1998b,The immune response of cattle, persistently in-fected with noncytopathic BVDV, after superin-fection with antigenically semi-homologous cyto-pathic BVDV. Vet Immunol Immunopathol62:37–50.

Bruschke CJ, Weerdmeester K, Van Oirschot JT, VanRijn PA: 1998a, Distribution of bovine virus diar-



rhea virus in tissues and white blood cells of cattleduring acute infection. Vet Microbiol 64:23–32.

Buckner R: 1997, Diabetes mellitus in a cow persist-ently infected with bovine viral diarrhea virus.Tierarztl Prax 25:351–352.


Casaro APE, Kendrick JW, Kennedy PC: 1971, Re-sponse of the bovine fetus to bovine viral diarrhea-mucosal disease virus. Am J Vet Res 32:1543–1562.

Castrucci G, Osburn BI, Ferrari M, Traldi V: 1992, Anexperimental contribution to the study of the patho-genesis of bovine viral diarrhea virus infection.Comp Immunol Microbiol Infect Dis 15:163–169.

Charleston B, Brackenbury LS, Carr BV, et al.: 2002,Alpha/beta and gamma interferons are induced byinfection with noncytopathic bovine viral diarrheavirus in vivo. J Virol 76:923–927.

Charleston B, Fray MD, Baigent S, Carr BV,Morrison WI: 2001, Establishment of persistent in-fection with non-cytopathic bovine viral diarrheavirus in cattle is associated with a failure to inducetype I interferon. J Gen Virol 82:1893–1897.

Collen T, Douglas AJ, Paton DJ, et al.: 2000, Singleamino acid differences are sufficient for CD4(+) T-cell recognition of a heterologous virus by cattlepersistently infected with bovine viral diarrheavirus. Virology 276:70–82.

Collen T, Morrison WI: 2000, CD4(+) T-cell re-sponses to bovine viral diarrhea virus in cattle.Virus Res 67:67–80.

Collen T, Carr V, Parsons K, et al.: 2002, Analysis ofthe repertoire of cattle CD4(+) T cells reactive withbovine viral diarrhea virus. Vet ImmunolImmunopathol 87:235–238.

Corapi WV, Elliott RD, French TW, et al.: 1990,Thrombocytopenia and hemorrhages in veal calvesinfected with bovine viral diarrhea virus. J Am VetMed Assoc 196:590–596.


David GP, Crawshaw TR, Gunning RF, et al.: 1994,Severe disease in adult dairy cattle in three UKdairy herds associated with BVD virus infection.Vet Rec 134:468–472.

Done JT, Terlecki S, Richardson C, et al.: 1980,Bovine virus diarrhea-mucosal disease virus:Pathogenicity for the fetal calf following maternalinfection. Vet Rec 106:473–479.

Drew TW, Sandvik T, Wakeley P, et al.: 2002, BVDvirus genotype 2 detected in British cattle. Vet Rec151:551.

Dubovi EJ: 1994, Impact of bovine viral diarrheavirus on reproductive performance in cattle. VetClin North Am Food Anim Pract 10:503–514

Duffell SJ, Harkness JW: 1985, Bovine virus diar-rhea-mucosal disease infection in cattle. Vet Rec117:240–245.

Duffell SJ, Sharp MW, Winkler CE, et al.: 1984,Bovine virus diarrhea-mucosal disease virus-induced fetopathy in cattle: Efficacy of prophylacticmaternal pre-exposure. Vet Rec 114:558–561.

Ellis JA, West KH, Cortese VS, et al.: 1998, Lesionsand distribution of viral antigen following an ex-perimetal infection of young seronegative calveswith virulent bovine virus diarrhea virus-type II.Can J Vet Res 62:161–169.

Elvander M, Baule C, Persson M, et al.: 1998, An ex-perimental study of a concurrent primary infectionwith bovine respiratory syncytial virus (BRSV) andbovine viral diarrhea virus (BVDV) in calves. ActaVet Scand 39:251–264.

Fray MD, Mann GE, Bleach EC, et al.: 2002,Modulation of sex hormone secretion in cows byacute infection with bovine viral diarrhea virus.Reproduction 123:281–289.

Fray MD, Mann GE, Clarke MC, Charleston B:2000a, Bovine viral diarrhea virus: Its effects onovarian function in the cow. Vet Microbiol77:185–194.

Fray MD, Prentice H, Clarke MC, Charleston B:1998, Immunohistochemical evidence for the local-ization of bovine viral diarrhea virus, a single-stranded RNA virus, in ovarian oocytes in the cow.Vet Pathol 35:253–259.

Fray MD, Supple EA, Morrison WI, Charleston B:2000b, Germinal centre localization of bovine viraldiarrhea virus in persistently infected animals. JGen Virol 81:1669–1673.

Fredriksen B, Press CM, Loken T, Odegaard SA:1999b, Distribution of viral antigen in uterus, pla-centa and foetus of cattle persistently infected withbovine virus diarrhea virus. Vet Microbiol64:109–22.

Fredriksen B, Press CM, Sandvik T, et al.: 1999a,Detection of viral antigen in placenta and fetus ofcattle acutely infected with bovine viral diarrheavirus. Vet Pathol 36:267–275.

Fricke J, Gunn M, Meyers G: 2001, A family ofclosely related bovine viral diarrhea virus recombi-nants identified in an animal suffering from mu-cosal disease: New insights into the development ofa lethal disease in cattle. Virology 291:77–90.

Frink S, Grummer B, Pohlenz JF, Liebler-TenorioEM: 2002, Changes in distribution and numbers ofCD4+ and CD8+ T-lymphocytes in lymphoid tis-sues and intestinal mucosa in the early phase of ex-

Pathogenesis 137


perimentally induced early onset mucosal disease incattle. J Vet Med B 49:476–483.

Fritzemeier J, Greiser-Wilke I, Haas L, et al.: 1995,Experimentally induced “late-onset” mucosal dis-ease—characterization of the cytopathogenicviruses isolated. Vet Microbiol 46:285–294.

Fritzemeier J, Haas L, Liebler E, et al.: 1998, The de-velopment of early vs. late onset mucosal disease isa consequence of two different pathogenic mecha-nisms. Arch Virol 142:1335–1350.

Fulton RW, Saliki JT, Confer AW, et al.: 2000, Bovineviral diarrhea virus cytopathic and noncytopathicbiotypes and type 1 and 2 genotypes in diagnosticlaboratory accessions: clinical and necropsy sam-ples from cattle. J Vet Diagn Invest 12:33–38.

Fulton RW, Step DL, Ridpath JF, et al.: 2003,Response of calves persistently infected with non-cytopathic bovine viral diarrhea virus (BVDV) sub-type 1b after vaccination with heterologous BVDVstrains in modified live virus vaccines andMannheimia haemolytica bacterin-toxoid. Vaccine21:2980–2985.

Ganheim C, Hulten C, Carlsson U, et al.: 2003, Theacute phase response in calves experimentally in-fected with bovine viral diarrhea virus and/orMannheimia haemolytica. J Vet Med B 50:183–190.

Gillespie JH, Baker JA, McEntee K: 1960, A cy-topathogenic strain of virus diarrhea virus. CornellVet 50:73–79.

Givens MD, Heath AM, Brock KV, et al.: 2003,Detection of bovine viral diarrhea virus in semenobtained after inoculation of seronegative postpu-bertal bulls. Am J Vet Res 64:428–434.

Glew EJ, Carr BV, Brackenbury LS, et al.: 2003,Differential effects of bovine viral diarrhea virus onmonocytes and dendritic cells. J Gen Virol84:1771–1780.

Glew EJ, Howard CJ: 2001, Antigen-presenting cellsfrom calves persistently infected with bovine viraldiarrhea virus, a member of the Flaviviridae, arenot compromised in their ability to present viralantigen. J Gen Virol 82:1677–1685.

Grahn TC, Fahning ML, Zemjanis R: 1984, Nature ofearly reproductive failure caused by bovine viral di-arrhea virus. J Am Vet Med Assoc 185:429–432.

Greiser-Wilke I, Liebler E, Haas L, et al.: 1993,Distribution of cytopathogenic and noncytopatho-genic bovine virus diarrhea virus in tissues from acalf with experimentally induced mucosal diseaseusing antigenic and genetic markers. Arch VirolSuppl 7:295–302.

Grooms DL, Brock KV, Pate JL, Day ML: 1998a,Changes in ovarian follicles following acute infec-tion with bovine viral diarrhea virus.Theriogenology 49:595–605.

Grooms DL, Brock KV, Ward LA: 1998b, Detectionof cytopathic bovine viral diarrhea virus in theovaries of cattle following immunization with amodified live bovine viral diarrhea virus vaccine. JVet Diagn Invest 10:130–134.

Grooms DL, Brock KV, Ward LA: 1998c, Detectionof bovine viral diarrhea virus in the ovaries of cattleacutely infected with bovine viral diarrhea virus. JVet Diagn Invest 10:125–129.

Grooms DL, Ward LA, Brock KV: 1996, Morphologicchanges and immunohistochemical detection ofviral antigen in ovaries from cattle persistently in-fected with bovine viral diarrhea virus. Am J VetRes 57:830–833.

Grummer B, Beer M, Liebler-Tenorio E, Greiser-Wilke I: 2001, Localization of viral proteins in cellsinfected with bovine viral diarrhea virus. J GenVirol 82:2597–2605.

Grummer B, Bendfeldt S, Wagner B, Greiser-Wilke I,2002: Induction of the intrinsic apoptotic pathwayin cells infected with cytopathic bovine virus diar-rhea virus. Virus Res 90:143–153.

Hamers C, Couvreur B, Dehan P, et al.: 2000,Differences in experimental virulence of bovineviral diarrhea viral strains isolated from haemor-rhagic syndromes. Vet J 160:250–258.

Hamers C, Dehan P, Couvreur B, et al.: 2001, Diver-sity among bovine pestiviruses. Vet J 161:112–122.

Harkness JW, Wood L, Drew T: 1984, Mucosal dis-ease in cattle. Vet Rec 115:283.

Heinz FX, Collett MS, Purcell RH, et al.: 2000,Family Flaviviridae. In: Virus Taxonomy. Eds. vanRegenmortel MHV, Fauquet CM, Bishop DHL, etal., pp. 859–879. Academic Press, San Diego, CA.

Hibberd RC, Turkington A, Brownlie J: 1993, Fatalbovine virud diarrhea infection of adult cattle. VetRec 132:227–228.

Houe H, Heron I: 1993, Immune response to otheragents of calves persistently infected with bovinevirus diarrhea virus (BVDV). Acta Vet Scand34:305–310.

Howard CJ, Brownlie J, Clarke MC: 1987, Com-parison by the neutralisation assay of pairs of non-cytopathogenic and cytopathogenic strains ofbovine virus diarrhea virus isolated from cases ofmucosal disease. Vet Microbiol 13:361–369.

Jensen J, Schultz RD: 1991, Effect of infection bybovine viral diarrhea virus (BVDV) in vitro on in-terleukin-1 activity of bovine monocytes. VetImmunol Immunopathol 29:251–265.

Kafi M, McGowan MR, Kirkland PD: 2002, In vitromaturation and fertilization of bovine oocytes andin vitro culture of presumptive zygotes in the pres-ence of bovine pestivirus. Anim Reprod Sci71:169–179.



Kahrs RF, Scott FW, de Lahunta A: 1970a, Bovineviral diarrhea-mucosal disease, abortion, and con-genital cerebellar hypoplasia in a dairy herd. J AmVet Med Assoc 156:851–857.

Kahrs RF, Scott FW, de Lahunta A: 1970b,Epidemiological observations on bovine viral diar-rhea-mucosal disease virus-induced congenitalcerebellar hypoplasia and ocular defects in calves.Teratology 3:181–184.

Kahrs RF, Ward GM, 1967, Bovine virus diarrheaabortion. Proc Annu Meet U S Anim Health Assoc71:493–499.

Kelling CL, Steffen DJ, Cooper VL, et al.: 2002,Effect of infection with bovine viral diarrhea virusalone, bovine rotavirus alone, or concurrent infec-tion with both on enteric disease in gnotobioticcalves. Am J Vet Res 63:1179–1186.

Kendrick JW: 1971, Bovine viral diarrhea-mucosaldisease virus infection in pregnant cows. Am J VetRes 32:533–544.

Ketelsen AT, Johnson DW, Muscoplat CC: 1979,Depression of bovine monocyte chemotactic re-sponses by bovine viral diarrhea virus. InfectImmun 25:565–568.

Kirkbride CA: 1992, Viral agents and associated le-sions detected in a 10-year study of bovine abor-tions and stillbirths. J Vet Diagn Invest 4:374–379.

Kirkland PD, Mackintosh SG, Mo\yle A: 1994, Theoutcome of widespread use of semen from a bullpersistently infected with pestivirus. Vet Rec135:527–529.

Kirkland PD, McGowan MR, Mackintosh-SG: 1993,Determinants of persistent infection of cattle withpestivirus. Proc 2nd Symp Pestiviruses. Ed.Edwards S. Fondation Marcel Merieux, Lyon, pp.117–121.

Kirkland PD, McGowan MR, Mackintosh SG, MoyleA: 1996, Insemination of cattle with semen from abull transiently infected with pestivirus. Vet Rec140:124–127.

Kirkland PD, Richards SG, Rothwell JT, Stanley DF:1991, Replication of bovine viral diarrhea virus inthe bovine reproductive tract and excretion of virusin semen during acute and chronic infections. VetRec 128:587–590.

Lambert G, Fernelius AL, Cheville NF: 1969,Experimental bovine viral diarrhea in neonatalcalves. J Am Vet Med Assoc 154:181–189.

Lambot M, Douart A, Joris E, et al.: 1997,Characterization of the immune response of cattleagainst non-cytopathic and cytopathic biotypes ofbovine viral diarrhea virus. J Gen Virol78:1041–1047.

Letellier C, Kerkhofs P, Wellemans G, VanopdenboschE: 1999, Detection and genotyping of bovine diar-

rhea virus by reverse transcription-polymerasechain amplification of the 5’ untranslated region.Vet Microbiol 64:155–167.

Liebler EM, Kusters C, Pohlenz JF: 1995,Experimental mucosal disease in cattle: Changes oflymphocyte subpopulations in Peyer’s patches andin lymphoid nodules of large intestine. Vet ImmunolImmunopathol 48:233–248.

Liebler EM, Kusters C, Pohlenz JF: 1996,Experimental mucosal disease in cattle: Changes inthe number of lymphocytes and plasma cells in themucosa of the small and large intestine. VetImmunol Immunopathol 55:93–105.

Liebler EM, Waschbuesch J, Pohlenz JF, et al.: 1991,Distribution of antigen of noncytopathogenic andcytopathogenic bovine virus diarrhea virus biotypesin the intestinal tract of calves following experi-mental production of mucosal disease. Arch VirolSuppl 7:295–301.

Liebler-Tenorio EM, Greiser-Wilke I, Pohlenz JF:1997, Organ and tissue distribution of the antigenof the cytopathogenic bovine virus diarrhea virus inthe early and advanced phase of experimental mu-cosal disease. Arch Virol 142:1613–1634.

Liebler-Tenorio EM, Lanwehr A, Greiser-Wilke I, etal.: 2000, Comparative investigation of tissue alter-ations and distribution of BVD-viral antigen in cat-tle with early onset versus late onset mucosal dis-ease. Vet Microbiol 15:163–174.

Liebler-Tenorio EM, Pohlenz JF: 1997, Experimentalmucosal disease of cattle: Altered cell proliferationin lymphoid tissues and intestinal epithelium. JComp Pathol 117:339–350.

Liebler-Tenorio EM, Ridpath JE, Neill JD: 2002,Distribution of viral antigen and development of le-sions after experimental infection with highly viru-lent bovine viral diarrhea virus type 2 in calves. AmJ Vet Res 63:1575–1584.

Liebler-Tenorio EM, Ridpath JF, Neill JD: 2003a,Distribution of viral antigen and development of le-sions after experimental infection of calves with aBVDV 2 strain of low virulence. J Vet Diagn Invest15:221–232.

Liebler-Tenorio EM, Ridpath JF, Neill JD: 2003b,Lesions and tissue distribution of viral antigen insevere acute versus subclinical acute infection withBVDV2. Biologicals 31:119–122.

Liess B, Frey HR, Kittsteiner H, et al.: 1974, Bovinemucosal disease, an immunobiological explainablelate stage of BVD-MD virus infection with criteriaof a “slow virus infection.” Dtsch TierarztlWochenschr 81:481–487.

Liess B, Orban S, Frey HR, et al.: 1984, Studies ontransplacental transmissibility of a bovine virus di-arrhea (BVD) vaccine virus in cattle. II. Inoculation

Pathogenesis 139


of pregnant cows without detectable neutralizingantibodies to BVD virus 90–229 days before partu-rition (51st to 190th day of gestation). ZentralblVeterinarmed B 31:669–681.

Liess B, Orban S, Frey HR, Trautwein G: 1987,Consequences of the transplacental transmission ofBVD virus to cattle fetuses. Dtsch TierarztlWochenschr 94:585–587.

Liu L, Lehmkuhl HD, Kaeberle ML: 1999, Syner-gistic effects of bovine respiratory syncytial virusand non-cytopathic bovine viral diarrhea virus in-fection on selected bovine alveolar macrophagefunctions. Can J Vet Res 63:41–48.

Loehr BI, Frey HR, Moennig V, Greiser-Wilke I:1998, Experimental induction of mucosal disease:Consequences of superinfection of persistently in-fected cattle with different strains of cytopathogenicbovine viral diarrhea virus. Arch Virol143:667–679.

Marshall DJ, Moxley RA, Kelling CL: 1996, Distribu-tion of virus and viral antigen in specific pathogen-free calves following inoculation with noncyto-pathic bovine viral diarrhea virus. Vet Pathol33:311–318.


McClurkin AW, Littledike ET, Cutlip RC, et al.:1984, Production of cattle immunotolerant tobovine viral diarrhea virus. Can J Comp Med48:156–161.

McGowan MR, Kafi M, Kirkland PD, et al.: 2003,Studies of the pathogenesis of bovine pestivirus-induced ovarian dysfunction in superovulated dairycattle. Theriogenology 59:1051–1066.

McGowan MR, Kirkland PD: 1995, Early reproduc-tive loss due to bovine pestivirus infection. Br Vet J151:263–270.

McGowan MR, Kirkland PD, Richards SG, Little-johns IR: 1993, Increased reproductive losses incattle infected with bovine pestivirus around thetime of insemination. Vet Rec 133:39–43.

Meyling A: 1970, Demonstration of VD-virus by thefluorescent antibody technique in tissues of cattleaffected with bovine viral diarrhea (mucosal dis-ease). Acta Vet Scand 11:59–72.

Meyling A, Jensen AM: 1988, Transmission of bovinevirus diarrhea virus (BVDV) by artificial insemina-tion (AI) with semen from a persistently infectedbull. Vet Microbiol 17:97–105.

Minocha HC, Xue W, Reddy JR: 1997, A 50 kDamembrane protein from bovine kidney cells is aputative receptor for bovine viral diarrhea virus(BVDV). Adv Exp Med Biol 412:145–148.

Moennig V, Frey HR, Liebler E, et al.: 1990, Repro-duction of mucosal disease with cytopathogenicbovine viral diarrhea virus selected in vitro. Vet Rec127:200–203.

Moennig V, Greiser-Wilke I, Frey HR, et al.: 1993,Prolonged persistence of cytopathogenic bovineviral diarrhea virus (BVDV) in a persistentlyviremic cattle. Zentralbl Vet Med B 40:371–377.

Moennig V, Liess B: 1995, Pathogenesis of intra-uterine infections with bovine viral diarrhea virus. Vet Clin North Am Food Anim Pract11:477–487.

Moennig V, Plagemann PG: 1992, The pestiviruses.Adv Virus Res 41:53–98.

Muñoz DP, Lager IA, Mersich S, et al.: 1996, Foetalinfections with bovine viral diarrhea virus inArgentina. Br Vet J 152:175–82.

Muñoz-Zanzi CA, Hietala SK, Thurmond MC, John-son WO: 2003, Quantification, risk factors, andhealth impact of natural congenital infection withbovine viral diarrhea virus in dairy calves. Am J VetRes 64:358–365.

Murondoti A, van der Kolk JH, van der Linde-SipmanJS: 1999, Type 1 diabetes mellitus in a pregnantheifer persistently infected with bovine viral diar-rhea virus. Vet Rec 144:268–269.

Murray RD: 1991, Lesions in aborted bovine fetusesand placenta associated with bovine viral diarrheavirus infection. Arch Virol Suppl 3:217–224.

Muscoplat CC, Johnson DW, Stevens JB: 1973,Abnormalities of in vitro lymphocyte responsesduring bovine viral diarrhea virus infection. Am JVet Res 34:753–755.

Nakajima N, Fukuyama S, Hirahara T, et al.: 1993,Induction of mucosal disease in cattle persistentlyinfected with noncytopathic bovine viral diarrhea-mucosal disease virus by superinfection with cyto-pathic bovine viral diarrhea-mucosal disease virus.J Vet Med Sci 55:67–72.

Neill JD, Ridpath JF: 2003, Gene expression changesin BVDV2-infected MDBK cells. Biologicals31:97–102.

Nettleton PF, Entrican G: 1995, Ruminant pes-tiviruses. Br Vet J 151:615–642.

Niskanen R, Alenius S, Belak K, et al.: 2002a,Insemination of susceptible heifers with semenfrom a non-viraemic bull with persistent bovinevirus diarrhea virus infection localized in the testes.Reprod Domest Anim 37:171–175.

Niskanen R, Lindberg A: 2003, Transmission ofbovine viral diarrhea virus by unhygienic vaccina-tion procedures, ambient air, and from contami-nated pens. Vet J 165:125–130.

Niskanen R, Lindberg A, Larsson B, Alenius S: 2000,Lack of virus transmission from bovine viral diar-



rhea virus infected calves to susceptible peers. ActaVet Scand 41:93–99.

Niskanen R, Lindberg A, Traven M: 2002b, Failure tospread bovine virus diarrhea virus infection fromprimarily infected calves despite concurrent infec-tion with bovine coronavirus. Vet J 163:251–259.

Odeon AC, Kelling CL, Marshall DJ, et al.: 1999,Experimental infection of calves with bovine viraldiarrhea virus genotype II (NY-93). J Vet DiagnInvest 11:221–228.

Olafson P, MacCallum AD, Fox FH: 1946, An appar-ently new transmissible disease of cattle. CornellVet 36:205–213.

Opdenbosch van, Wellemans G, Oudewater J: 1981,Synergistic interaction of bovine diarrhea virus,coronavirus and and rotavirus in diarrhea of new-born calves: Experimental infections in calves.Vlaams Diergeneesk Tijdschr 50:163–173.

Orban S, Liess B, Hafez SM, et al.: 1983, Studies ontransplacental transmissibility of a Bovine VirusDiarrhea (BVD) vaccine virus. I. Inoculation ofpregnant cows 15 to 90 days before parturition(190th to 265th day of gestation). Zentralbl VetMed B 30:619–634.

Perdrizet JA, Rebhun WC, Dubovi EJ, Donis RO:1987, Bovine virus diarrhea—clinical syndromes indairy herds. Cornell Vet 77:46–74.

Perler L, Schweizer M, Jungi TW, Peterhans E: 2000,Bovine viral diarrhea virus and bovine herpesvirus-1 prime uninfected macrophages for lipopolysac-charide-triggered apoptosis by interferon-dependentand -independent pathways. J Gen Virol81:881–887.

Peterhans E, Jungi TW, Schweizer M: 2003, BVDVand innate immunity. Biologicals 31:107–112.

Pospisil Z, Machatkova M, Mensik J, et al.: 1975,Decline in the phytohaemagglutinin responsivenessof lymphocytes from calves infected experimentallywith bovine viral diarrhea-mucosal disease virusand parainfluenza 3 virus. Acta Vet Brno44:369–375.

Potgieter LN: 1997, Bovine respiratory tract diseasecaused by bovine viral diarrhea virus. Vet ClinNorth Am Food Anim Pract 13:471–481.

Potgieter LN, McCracken MD, Hopkins FM, WalkerRD: 1984, Effect of bovine viral diarrhea virus in-fection on the distribution of infectious bovinerhinotracheitis virus in calves. Am J Vet Res45:687–690.

Potter ML, Corstvet RE, Looney CR, et al.: 1984,Evaluation of bovine viral diarrhea virus uptake bypreimplantation embryos. Am J Vet Res45:1778–1780.

Pritchard GC, Borland ED, Wood L, Pritchard DG:1989, Severe acute disease in a dairy herd associ-

ated with acute infection with bovine virus diarrheavirus, Leptospira hardjo and Coxiella burnetti. VetRec 124:625–629.

Radostits OM, Littlejohns IR: 1988, New concepts inthe pathogenesis, diagnosis and control of diseasescaused by the bovine viral diarrhea virus. Can Vet J29:513–528.

Ramsey FK, Chivers WH: 1953, Mucosal disease ofcattle. North Am Vet 34:629–633.

Rebhun WC, French TW, Perdrizet JA, et al.: 1989,Thrombocytopenia associated with acute bovinevirus diarrhea infection in cattle. J Vet Int Med3:42–46.

Reggiardo C, Kaeberle ML: 1981, Detection of bac-teremia in cattle inoculated with bovine viral diar-rhea virus. Am J Vet Res 42:218–221.

Rhodes SG, Cocksedge JM, Collins RA, MorrisonWI: 1999, Differential cytokine responses of CD4+ and CD8+ T cells in response to bovine viral diarrhea virus in cattle. J Gen Virol80:1673–1679.





Ross CE, Dubovi EJ, Donis RO: 1986, Herd problemof abortions and malformed calves attributed tobovine viral diarrhea. J Am Vet Med Assoc188:618–619.

Roth JA, Bolin SR, Frank DE: 1986, Lymphocyteblastogenesis and neutrophil function in cattle per-sistently infected with bovine viral diarrhea virus.Am J Vet Res 47:1139–1141.

Roth JA, Kaeberle ML: 1982, Effects of a modifiedlive BVD vaccine and ACTH on bovine granulo-cyte function. J Reticuloendothelial Soc 32:1.


Roth JA, Kaeberle ML, Griffith RW: 1981, Effects ofbovine viral diarrhea virus infection on bovinepolymorphonuclear leukocyte function. Am J VetRes 42:244–250.

Pathogenesis 141


Rufenacht J, Schaller P, Audige L, et al.: 2001, Theeffect of infection with bovine viral diarrhea viruson the fertility of Swiss dairy cattle.Theriogenology 56:199–210.

Sakoda Y, Ozawa S, Damrongwatanapokin S, et al.:1999, Genetic heterogeneity of porcine and rumi-nant pestiviruses mainly isolated in Japan. VetMicrobiol 65:75–86.

Sandvik T, Drew TW, Bensaude E, et al.: 2000, AcuteBVD in calves—effect of viral dose and immuno-suppression on clinical signs of disease, transmis-sion to susceptible animals and virus persistence intissues. Proc Europ Soc Vet Virol, Brescia, pp.401–402.

Sandvik T, Fredriksen B, Loken T: 1997, Level ofviral antigen in blood leucocytes from cattle acutelyinfected with bovine viral diarrhea virus. ZentralblVet Med B 44:583–590.


Schulz LC: 1959, Pathologisch-anatomische Befundebei der sogenannten “Mucosal-Disease” (Schleim-hautkrankheit des Rindes. Dtsch Tierärztl Wschr21:586–588.

Schweizer M, Peterhans E: 1999, Oxidative stress incells infected with bovine viral diarrhea virus: Acrucial step in the induction of apoptosis. J GenVirol 80:1147–1155.


Scott FW, Kahrs RF, de Lahunte A, et al.: 1973, Virusinduced congenital anomalies of the bovine fetus. I.Cerebellar degeneration (hypoplasia), ocular lesionsand fetal mummification following experimental in-fection with bovine viral diarrhea-mucosal diseasevirus. Cornell Vet 63:536–560.

Scott FW, Kahrs RF, Parsonson IM: 1972, A cytopath-ogenic strain of bovine viral diarrhea-mucosal dis-ease virus isolated from a bovine fetus. Cornell Vet62:74–84.

Sentsui H, Nishimori T, Kirisawa R, Morooka A:2001, Mucosal disease induced in cattle persistentlyinfected with bovine viral diarrhea virus by anti-genically different cytopathic virus. Arch Virol146:993–1006.

Singh EL, Eaglesome MD, Thomas FV, et al.: 1982,Embryo transfer as means of controlling the trans-mission of viral infections. I. The in vitro exposureof preimplanmtatino embryos to Akabane, blue-tongue and bovine viral diarrhea viruses.Theriogenology 17:437–444.

Smith AK, Grimmer SP: 2000, Birth of a BVDV-freecalf from a persistently infected embryo transferdonor. Vet Rec 146:49–50.

Spagnuolo M, Kennedy S, Foster JC, et al.: 1997,Bovine viral diarrhea virus infection in bone mar-row of experimentally infected calves. J CompPathol 116:97–100.

Spagnuolo-Weaver M, Allan GM, Kennedy S, et al.:1997, Distribution of cytopathic and noncytopathicbovine viral diarrhea virus antigens in tissues ofcalves following acute experimental infection. J VetDiagn Invest 9:287–297.

Sprecher E, Becker Y: 1993, Role of Langerhans cellsand other dendritic cells in viral diseases. ArchVirol 132:1–28.

Ssentongo YK, Johnson RH, Smith JR: 1980,Association of bovine viral diarrhea-mucosal dis-ease virus with ovaritis in cattle. Aust Vet J56:272–273.

Stoffregen B, Bolin SR, Ridpath JF, Pohlenz J: 2000,Morphologic lesions in Type 2 BVDV infectionsexperimentally induced by strain BVDV2-1373 re-covered from a field case. Vet Microbiol77:157–162.

Straver PJ, Journee DL, Binkhorst GJ: 1983,Neurological disorders, virus persistence and hy-pomyelination in calves due to intra-uterine infec-tions with bovine virus diarrhea virus. II. Virologyand epizootiology. Vet Q 5:156–164.


Taniyama H, Ushiki T, Tajima M, et al.: 1995,Spontaneous diabetes mellitus associated with per-sistent bovine viral diarrhea (BVD) virus infectionin young cattle. Vet Pathol 32:221–229.

Teichmann U, Liebler-Tenorio EM, Pohlenz JF: 2000,Ultrastructural changes in lymphoid follicles ofsmall intestinal aggregated lymphoid nodules in theearly and advanced phases of experimentally in-duced mucosal disease in cattle. Am J Vet Res61:174–182.

Thoen CO, Waite KJ: 1990, Some immune responsesin cattle exposed to Mycobacterium paratuberculo-sis after injection with modified-live bovine viraldiarrhea virus vaccine. J Vet Diagn Invest2:176–179.

Trautwein G, Hewicker M, Liess B, et al.: 1986,Studies on transplacental transmissibility of abovine virus diarrhea (BVD) vaccine virus in cattle.III. Occurrence of central nervous system malfor-mations in calves born from vaccinated cows.Zentralbl Vet Med B 33:260–268.



Trautwein G, Hewicker M, Liess B, et al.: 1987,Cerebellar hypoplasia and hydranencephaly in cat-tle after transplacental bovine diarrhea virus infec-tion. Dtsch Tierarztl Wochenschr 94:588–590.

Traven M, Alenius S, Fossum C, Larsson B: 1991,Primary bovine viral diarrhea virus infection incalves following direct contact with a persistentlyviremic calf. J Vet Med B 38:453–462.

Tsuboi T, Imada T: 1996, Noncytopathogenic and cy-topathogenic bovine viral diarrhea-mucosal diseaseviruses do not affect in vitro embryonic develop-ment into the blastocyst stage. Vet Microbiol49:127–134.

Tsuboi T, Imada T: 1999, Susceptibility of bovinenaked 2- and 4-cell embryos and hatched blasto-cysts produced in vitro to infection with noncy-topathogenic bovine viral diarrhea virus. J Vet MedSci 61:943–945.

Vanroose G, Nauwynck H, Van Soom A, et al.: 1998,Replication of cytopathic and noncytopathic bovineviral diarrhea virus in zona-free and zona-intact invitro-produced bovine embryos and the effect onembryo quality. Biol Reprod 58:857–566.

Vilcek S, Durkovic B, Bobakova M, et al.: 2002,Identification of bovine viral diarrhea virus 2 in cat-tle in Slovakia. Vet Rec 151:150–152.


Walz PH, Bell TG, Wells JL, et al.: 2001, Relation-ship between degree of viremia and disease mani-festation in calves with experimentally inducedbovine viral diarrhea virus infection. Am J Vet Res62:1095–1103.

Walz PH, Steficek BA, Baker JC, Kaiser L, Bell TG:1999, Effect of experimentally induced bovine viraldiarrhea virus infection on platelet function incalves. Am J Vet Res 60:1396–1401.

Ward GM: 1969, Bovine cerebellar hypoplasia appar-ently caused by BVD-MD virus. A case report.Cornell Vet 59:570–576.

Welsh MD, Adair BM, Foster JC: 1995, Effect ofBVD virus infection on alveolar macrophage func-tions. Vet Immunol Immunopathol 46:195–210.

Wentink GH, Aarts T, Mirck MH, et al.: 1991, Calffrom a persistently infected heifer born afterembryo transfer with normal immunity to (bovine virus diarrhea virus) BVDV. Vet Rec 129:449–450.

Wentink GH, Remmen JL, van Exsel AC: 1989,Pregnancy rate of heifers bred by an immunotoler-ant bull persistently infected with bovine viral diar-rhea virus. Vet Q 11:171–174.

Werdin RE, Ames TR, Goyal SM: 1989, Detectionand elimination of carrier animals in a dairy herdpersistently infected with bovine viral diarrheavirus. J Vet Diagn Invest 1:277–279.

Whitemore HL, Zemjanis R, Olson J: 1981, Effect ofbovine viral diarrhea virus on conception oin cattle.J Am Vet Med Assoc 178:1065–1067.


Wilhelmsen CL, Bolin SR, Ridpath JF, et al.: 1991,Lesions and localization of viral antigen in tissuesof cattle with experimentally induced or naturallyacquired mucosal disease, or with naturally ac-quired chronic bovine viral diarrhea. Am J Vet Res52:269–275.

Wittum TE, Grotelueschen DM, Brock KV, et al.:2001, Persistent bovine viral diarrhea virus infec-tion in US beef herds. Prev Vet Med 49:83–94.


Wray C, Roeder PL: 1987, Effect of bovine virus di-arrhea-mucosal disease virus infection on salmo-nella infection in calves. Res Vet Sci 42:213–218.

Pathogenesis 143


INTRODUCTION AND OVERVIEWBovine viral diarrhea virus (BVDV) is recognizedas one of the most important infectious diseases ofcattle (Baker, 1987; Duffell and Harkness, 1985).The insidious nature of BVDV contributes to sub-stantial economic losses in both the dairy and thebeef cattle industries on a worldwide basis (Duffellet al., 1986; Houe et al., 1993a). The most commonconsequences of BVDV infection in cattle are res-piratory and reproductive disorders. Losses due toBVDV-related reproductive disorders are probablythe most economically important consequence, andthere is evidence to suggest that such losses are in-creasing in the United States (Evermann andRidpath, 2002). Although difficult to measure, theeconomic consequences of reproductive infectionscan be devastating due to negative effects on theindividual animal and on the net return per cow. Inaddition to reduced reproductive efficiency, BVDVhas adapted to the bovine reproductive system,thereby maintaining itself in the cattle populationby inducing immunotolerant fetal infections; thisresults in the birth of calves persistently infectedwith BVDV. The primary consequences of re-productive infection are due to the direct infectionof the fetus. In turn, persistently infected (PI) ani-mals act as a source of BVDV within the cattlepopulations.

Reproductive losses associated with BVDV infec-tion were initially described by Olafson and associ-ates in the first clinical description of bovine viraldiarrhea (Olafson et al., 1946). Pregnant cows thatwere subclinically infected with BVDV, oftenaborted 10–90 days following infection. It has beenobserved that BVDV can cause a wide array of re-productive losses manifested in many different clin-

ical pictures. The outcome of BVDV infection de-pends primarily on the stage of the reproductivecycle or gestation.

IMPACT OF TESTICULARINFECTION ON BULL FERTILITYThe influence of BVDV infection on bull fertilityand reproduction is often overlooked. Both acuteand persistent infections can affect the reproductivesoundness of bulls by reducing semen quality (Coriaand McClurkin, 1978; Barlow et al., 1986; Meylingand Jensen, 1988; Revell et al., 1988; Kirkland et al.,1994; Whitmore and Archbald, 1977; Kirkland etal., 1991; Kommisrud et al., 1996). The primary ef-fect of infection is the subsequent potential for vene-real transmission during breeding and the sheddingof BVDV in semen. Virus has been isolated from thesemen of PI bulls ranging from 104 to 107 CCID50/ml of semen (Kirkland et al., 1991). Following acuteexperimental infection of bulls with noncytopathicBVDV, virus was isolated from the semen between7 and 14 days postinfection at titers ranging from5–75 CCID50/ml of semen (Kirkland et al., 1991).The ability to transmit BVDV via semen is sup-ported by the demonstration that susceptible cowscan become infected following artificial insemina-tion (Kirkland et al., 1994; Paton et al., 1990) or nat-ural service (McClurkin et al., 1979; Wentink et al.,1989).

Persistently infected bulls can successfully sire off-spring but usually with lowered conception rates(Meyling and Jensen, 1988; Kirkland et al., 1994;Paton et al., 1990). Poor reproductive efficiency fol-lowing the use of PI bulls is likely attributable to acombination of several factors including low qualitysemen, ill thrift, and the effects of the virus on the re-

145

8Reproductive Disease and

Persistent Infections

Kenny V. Brock, Daniel L. Grooms, and M. Daniel Givens


productive tract and conceptus of infected females.The quality of semen from PI bulls may range fromacceptable (Barlow et al., 1986; Kirkland et al., 1991)to abnormal with various defects predominantly in-volving the head of the spermatozoa (Barlow et al.,1986; Revell et al., 1988; McClurkin et al., 1979).Prior to the development of standardized BVDV test-ing protocols in semen collection/production facili-ties, semen was unknowingly collected and distrib-uted from PI bulls. A retrospective analysis ofbreeding records in 97 dairy farms in which semenfrom a PI bull was used, indicated a first service con-ception rate of 38% in cows inseminated with thissemen as compared to 66% for cows bred during thesame period on the same farms with semen from adifferent bull (Kirkland et al., 1994).

Following acute infection, bulls can shed BVDVin semen (5–75 CCID50/ml of semen). In contrast tosemen from PI bulls, the semen of acutely infectedanimals (based on the criteria of concentration, mo-tility, and morphology of spermatozoa) will passbreeding soundness examination (Kirkland et al.,1997; Kirkland et al., 1991). It is generally acceptedthat the ability to isolate virus from semen ceaseswhen serum antibodies become detectable (Kirklandet al., 1991; Paton et al., 1989). BVDV isolationfrom raw semen may be less successful than fromextended semen (Revell et al., 1988). This is pre-sumably due to the documented viricidal effects ofsemen. Using semen that was collected prior to se-roconversion (12 days postinoculation), approxi-mately 5% of inseminated heifers became infectedas evidenced by seroconversion (Kirkland et al.,1997). It is interesting to note that in this report theprimary impact of BVDV was the horizontal trans-mission of BVDV to penmates and the subsequentbirth of PI calves (Kirkland et al., 1997).

As is true for most biological systems, there arealways exceptions to the rule. In addition to classi-cal acute and persistent infections, the replication ofBVDV in a “privileged” site has been described.Thus, persistent BVDV infection localized in thetestes of an immunocompetent, seropositive, non-viremic bull has been documented (Voges et al.,1998). The concentration of BVDV in the semen ofthis bull was lower (<2 � 103 CCID50/ml) than thatobserved in PI bulls (104–107 CCID50/ml), buthigher than that observed in acutely infected bulls(5–75 CCID50/ml) (Kirkland et al., 1991; Voges etal., 1998). The insemination of seronegative heiferswith semen collected from this bull resulted inBVDV infection and subsequent seroconversion

(Niskanen et al., 2002). In addition, during the re-ported period of collection, the bull continued tohave consistent, high levels of BVDV neutralizingantibody against the viral strain isolated from thesemen. Bovine viral diarrhea virus could not be iso-lated from white blood cells but was continually iso-lated from semen samples. At postmortem, BVDVwas isolated only from the testicular tissue. It is hy-pothesized that this bull was acutely infected withBVDV near the time of puberty when the blood-testes barrier forms, thus trapping the virus in go-nadal cells away from the animal’s immune re-sponse. These findings suggest that screening bullsfor persistent infection with BVDV using serum orwhite blood cells may not be adequate in assuringBVDV-free semen.

Persistent, localized testicular BVDV infectionsin experimentally infected, post-pubertal, non-viremic bulls have been characterized (Givens et al.,2003a). Following experimental infection, BVDVpersisted within the testicular tissues of some bullsfor at least 7 months (Givens et al., 2003a). Experi-mental results have indicated that an epididymal in-fection may progress to a testicular infection. Due tothe blood-testes barrier, BVDV is protected fromelimination by the immune system and testicular in-fection can persist (Paton et al., 1989; Kirkland etal., 1991). The prevalence of bulls that are non-viremic (based on failure to isolate virus from serumor buffy coat sample) but shed BVDV in semen isprobably extremely low (Givens et al., 2003b; Nis-kanen et al., 2002). Further studies are required todetermine whether persistent-testicular BVDV in-fections contribute to transmission of BVDV bysemen to susceptible cows.

Currently, BVDV contamination of distributedsemen is prevented by practicing standardized test-ing and quarantine procedures in AI semen collec-tion facilities. Certified Semen Services, Inc. (CSS)is a cooperative organization, and membership inCSS ensures that the standardized procedures areused appropriately (CSS guidelines: http://www.naab-css.org/about_css/disease_control-2002.html).Therefore, use of semen from CSS-certified collec-tions is recommended to prevent introduction ofBVDV via semen. Current CSS prevention meas-ures include screening of all bulls 30 days prior toentry by virus isolation or ELISA to detect persist-ent infections. The collection and distribution ofsemen from bulls with persistent testicular infec-tions is prevented by requiring that specimens ofsemen be negative by virus isolation.



IMPACT OF OVARIAN INFECTIONON REPRODUCTIVE CAPACITYBVDV infections prior to breeding often go unno-ticed in production operations, and thus field data togauge its impact are sparse. Limited information isavailable on ovarian function following BVDV in-fection. Reproductive capacity of bulls andcows/heifers can be affected directly by BVDV in-fection. Bovine viral diarrhea virus has been identi-fied by virus isolation and immunohistochemistry inovarian tissues and oviductal cells following acuteinfection. The identification of BVDV and viralantigen has been associated with chronic oophoritisand salpingitis of cattle. Clinical examination ofcattle experimentally infected with BVDV revealedreduced number of corpora lutea and recovered em-bryos. Between 6 and 60 days following experi-mental acute infection, viral antigen (Erns) was de-tected by immunohistochemistry in interstitialstromal cells and macrophage-like cells that wereassociated with primary follicles, secondary folli-cles, antral follicles, corpus luteum, and corpus albi-cans (Grooms et al., 1998a). Viral antigen (Erns) wasdetected by immunohistochemical techniques inovarian sections taken on days 10, 20, and 30 afterimmunization with a modified-live BVDV vaccine(Grooms et al., 1998b). Experimental acute BVDVinfection near estrus has resulted in reduced follicu-lar development following infection (Grooms et al.,1998c).

In a study of cattle being superovulated while un-dergoing experimental challenge with BVDV, thenumber of palpable corpora lutea and recoveredembryos was significantly reduced when comparedto noninfected cows undergoing superovulation(Kafi et al., 1994). BVDV infection resulting inviremia during the preovulatory phase can reducethe rate of follicle growth (Grooms et al., 1998a;Fray et al., 1999). Cows that are persistently in-fected with BVDV have ovaries that are often hy-poplastic with a significantly reduced number ofovarian antral follicles when compared to ovariesfrom cattle not persistently infected with BVDV(Grooms et al., 1996). Alteration of ovarian hor-mone secretions has been demonstrated followingacute BVDV infection and has been postulated tocontribute to BVDV-induced infertility (Fray et al.,1999; Fray et al., 2000a; 2000b; 2002). Thechanges in follicular development and ovarian hor-monal dynamics associated with BVDV infectionmay subsequently lead to a transient and/or long-term reduction in fertility.

IMPACT OF STAGE OFGESTATION ON THE OUTCOMEOF REPRODUCTIVE DISEASEAs stated previously, the outcome of reproductive dis-ease depends on the stage of gestation at which fetalinfection occurs. Abortions associated with BVDVinfection were first described in 1946 (Olafson et al.,1946). Initial reports linked abortions to epizootics ofdisease described as bovine viral diarrhea althoughdefinitive causes of abortions were not identified(Dow et al., 1956; Nielson et al., 1955; Swope andLuedke, 1956). Early studies involving experimentalinfection with BVDV resulted in abortion althoughvirus was not isolated from the fetus (Baker et al.,1954; Huck, 1957). Subsequently noncytopathic(Gillespie et al., 1967) and cytopathic BVDV (Shope,1968; Scott et al., 1972) were isolated from abortedfetuses. Experimental, transplacental infection offetus with BVDV was first demonstrated in 1969(Ward et al., 1969). Under experimental conditions,both cytopathic (Brownlie et al., 1989) and noncyto-pathic BVDV (Done et al., 1980; Duffell et al., 1984;Liess et al., 1984) can cause fetal death following in-fection of seronegative dams. An abortion rate of 21%in a 6-month period has been documented to have oc-curred in a susceptible herd following the introduc-tion of BVDV in the herd (Roeder et al., 1986). In en-demically infected herds without BVDV preventionand control programs (vaccination, biosecurity, test,and removal), it has been estimated that 7% of fetaldeaths may be attributable to infection with BVDV(Rufenacht et al., 2001).

Fetal death following BVDV infection of suscep-tible dams can occur at any stage during gestation,although abortions are most common during the firsttrimester (Duffell and Harkness, 1985; Done et al.,1980; Roeder et al., 1986; Casaro et al., 1971;Kahrs, 1968; Kendrick, 1971; Sprecher et al., 1991;McGowan et al., 1993). However, BVDV can alsobe associated with late-term abortions. In a field in-vestigation of an abortion outbreak in a large dairyoperation, BVDV was isolated from 18 fetuses, 14of which were aborted during the last trimester ofgestation (Grooms, unpublished data). Dependingon the age of the fetus, fetal reabsorption, mummifi-cation, or expulsion can occur following infectionwith BVDV (Done et al., 1980; Casaro et al., 1971).

CONSEQUENCES OF BVDV INFECTIONPRIOR TO IMPLANTATION (30–45 DAYS)Understanding the effects of BVDV infection duringthe early stages of embryo development has proven

Reproductive Disease and Persistent Infections 147


to be difficult. Characterization of the complexevents that influence conception is difficult and con-tributes to the lack of understanding of the effects ofBVDV. From epidemiological and clinical observa-tions it is apparent that BVDV infection reduces notonly the conception rates but also the viability of theearly stage conceptus. Conception rates were re-duced in cattle exposed to BVDV during breedingwhen compared to cattle seropositive to BVDV(Houe et al., 1993a, 1993b; McGowan et al., 1993;McGowan et al. 1995; Virakul et al., 1988). Themechanism(s) for reduced conception is not directlyknown. However, it has been demonstrated thatBVDV infection results in the replication of BVDVin ovarian tissues (Grooms et al., 1998a, 1998b).The most common consequence of BVDV infectionduring the early stages of gestation is infertility. Invitro experimental studies have demonstrated thatthe zona pellucida provides a protective effect to theearly developing embryo.

When a group of seronegative cattle was acciden-tally exposed to a persistently infected cow duringbreeding, the conception rates in cattle that sero-converted to BVDV before, during, or after breed-ing were 78.6%, 44.4%, and 22.2%, respectively(Virakul et al., 1988). Cattle that seroconverted toBVDV at breeding or soon after breeding were lesslikely to conceive than cattle that had seroconvertedprior to breeding (Virakul et al., 1988). McGowan(1993) compared BVDV seropositive heifers toheifers that seroconverted between breeding andpregnancy diagnosis at 51 days postinseminationand found that the pregnancy rate was significantlylower in heifers seroconverting following breeding.Houe et al. (1993b) identified and defined a specificrisk period for BVDV infection in dairy herds inwhich cattle persistently infected with BVDV werepresent. The risk period was defined as the periodof time previous to when the oldest PI animal was 6months old. In all herds studied, conception rateswere significantly lower during the defined risk pe-riod than during the post-risk period (Houe et al.,1993b). In an experimental study examining BVDVinfection around breeding, conception rates inheifers infected intranasally 9 days before insemi-nation was 44% compared to 79% for the controlgroup (McGowan et al., 1993). The reduction inconception rates was attributed to either failure offertilization or early embryonic death. In the samereport, the conception rate in heifers exposed to apersistently infected cow and calf 4 days followinginsemination was 60% (McGowan et al., 1993).However, significant embryo loss was experienced

in this group, resulting in a 77-day pregnancy rateof only 33% as compared to 79% for the controlgroup.

The mechanism for decreased conception rates isnot clear but may depend on the time of infectionwith respect to the stage of early reproductiveevents. Virus has been localized in ovarian tissue forprolonged periods of time following acute infectionwith cytopathic (Ssentongo et al., 1980; Grooms etal., 1998b) and noncytopathic BVDV (Grooms etal., 1998a). BVDV has also been isolated from fol-licular fluid collected from slaughterhouse ovaries(Bielanski et al., 1993). Exposure of developingoocytes to BVDV could result in reduced survivabil-ity either through direct cell damage or indirectlythrough changes in the oocyte at the cellular level.Following acute infection with cytopathic BVDV,interstitial oophoritis has been described with le-sions lasting up to 60 days (Ssentongo et al., 1980;Grooms et al., 1998a). Significant long-termoophoritis could result in ovarian malfunction withsubsequent poor conception rates.

Because of its essential role in fertilization,changes in the oviductal environment could have adetrimental effect on the conception rate. BVDV hasbeen detected in oviductal cells (Bielanski et al.,1993; Booth et al., 1995). Archbald et al. (1973) iso-lated BVDV from oviductal tissue and detected evi-dence of salpingitis for up to 21 days following ex-perimental intrauterine infusion with cytopathicBVDV. Similar findings have not been reported withnoncytopathic BVDV.

Studies have suggested that the interruption ofnormal fertilization or embryonic death may be themechanism for a reduction in conception rates asso-ciated with acute BVDV infection. This conclusionwas drawn from the observation that infusion of cy-topathic BVDV into the uterus at insemination ofsuperovulated cows resulted in a significant reduc-tion in the number of fertilized ova found at slaugh-ter 3 and 13 days later (Grahn et al., 1984). Archbaldet al. (1979) provided evidence that BVDV may in-terfere with early embryonic development. In super-ovulated cattle in which BVDV had been infusedinto one uterine horn, the quality of the embryos col-lected from the infected horn was significantly re-duced compared to those collected from the nonin-fected horn (Archbald et al., 1979). In a similarstudy, the conception rate in seronegative heifers in-fused with BVDV 2 hours following breeding was27% and was significantly reduced (67%) as com-pared to sham-inoculated cows (Whitmore et al.,1981). However, in the same study, conception rates



of seropositive cows inoculated in utero with BVDVor seronegative cows inoculated orally and in-tranasally were not significantly different than thosein control cows.

Although it is thought that BVDV has a direct ef-fect on the developing embryo, inflammatorychanges in the uterus following BVDV infectionmay result in an incompatible environment for em-bryo development. In a study of uterine pathologyfollowing intrauterine infusion of cytopathic BVDV,histological changes in both the uterus and oviductwere evident from 6–21 days postinfection (Arch-bald et al., 1973). Several in vitro studies have beenundertaken to elucidate the effects of BVDV onearly reproductive function. Ova exposed to BVDVin vitro can have virus particles attached to the zonapellucida (Gillespie et al., 1990). However, in vitrostudies have shown that the intact zona pellucidaprotects the developing embryonic cells fromBVDV infection, allowing normal development tocontinue (Singh et al., 1982; Potter et al., 1984). Inmorula and blastocyst stage bovine embryos withthe zona pellucida intact or damaged, no cytopathiceffects were seen for 48 hours following exposure tocytopathic BVDV (Bielanski and Hare, 1988).Similarly, zona pellucida intact embryos exposed tononcytopathic BVDV infected bovine oviductal ep-ithelial cells for 7 days showed no adverse effects intheir rates of development (Zurovac et al., 1994). Incontrast, blastocysts hatched from the zona pellu-cida (day 8 of gestation) have been shown to havedecreased viability when exposed to cytopathicBVDV in vitro. In the same study, noncytopathicBVDV did not decrease blastocyst survivability(Brock and Stringfellow, 1993). These studies sug-gest that the zona pellucida protects the developingembryo from direct effects of BVDV. However, fol-lowing removal of the zona pellucida, cytopathicBVDV may have detrimental effects on the surviv-ability of blastocysts. Noncytopathic BVDV has notbeen shown to have the same effects. Because non-cytopathic BVDV is the most common virus iso-lated in acute outbreaks of BVDV, and has been thebiotype associated with reported decreases in con-ception rates, further characterization of the effect ofnoncytopathic BVDV on the early stages of the de-veloping embryo is necessary. Interestingly, at day14 post-hatching, BVDV antigen has been detectedin embryos inoculated with noncytopathic BVDV athatching (Brock and Stringfellow, 1993). In contrastto cytopathic BVDV, it is possible that the effect ofexposure of embryos to noncytopathic BVDV maybe delayed.

CONSEQUENCES OF BVDV INFECTION FROM30–125 DAYS OF GESTATION

Following implantation, transplacental infection ofthe developing fetus can occur in susceptible cowswith either noncytopathic or cytopathic strains ofBVDV. The outcome of infection is largely depend-ent on the timing of infection, the immunocompe-tence of the developing fetus, the virus biotype in-volved and the virulence of the virus. Although themechanism of fetal infection is not clear, evidencesuggests that BVDV may cross the placenta by caus-ing vasculitis on the maternal side of placentation al-lowing access to fetal circulation (Fredriksen et al.,1999). Following infection of pregnant animals,BVDV crosses the placenta, and replication can bedemonstrated in the fetus within 7 days postinfection.

The phenomenon of persistent infection was firstdescribed in an apparently healthy bull (Coria andMcClurkin, 1978). Experimental production of per-sistent infections was reported in 1984 (Brownlie etal., 1984; Liess et al., 1984; McClurkin et al., 1984).During gestation, fetuses from 18–125 days of agethat survive noncytopathic BVDV infection developimmunotolerance between 125–150 days of gesta-tion and subsequently harbor lifelong persistentinfection. Although there are several hypotheses re-garding the mechanism of induction of immunolog-ical tolerance, it is clear that the circulation ofBVDV during the period of gestation when immu-nocompetence is developing (90–120 days) is a pre-requisite for immunotolerance and persistent infec-tion. Persistent BVDV infection in cattle appears toarise from specific B- and T-lymphocyte immuno-tolerance (Coria and McClurkin, 1978; McClurkinet al., 1984). In persistently infected animals, BVDviral proteins are accepted as self-antigens with a re-sulting negative selection or down-regulation ofBVDV specific B- and T-lymphocytes during on-togeny. This negative selection results in an absenceof neutralizing and non-neutralizing antibodies andof cell-mediated immunity to the persistent virus(Donis and Dubovi, 1987). It appears that the win-dow of opportunity during fetal development forBVDV to establish an immunotolerant infection isapproximately 100 days. Under experimental condi-tions, persistent infection occurred in 86% and100% of calves derived from cows infected withBVDV at days 18 and 30 of gestation, respectively(Kirkland et al., 1993).

Persistent infections have been obtained in 100%of fetuses in dams challenged at 75 days of gestation(Cortese et al., 1998; Brock and Cortese, 2001;



Brock and Chase, 2000). Although persistent infec-tions have been reported to occur following infec-tion up to 125 days of gestation, they are only rarelyestablished in fetuses older than 100 days of gesta-tion (Baker, 1995). Noncytopathic BVDV is theonly biotype that has been observed clinically or ex-perimentally to produce persistent BVDV infections(Brownlie et al., 1984; Brownlie et al., 1989; Bolinet al., 1985). Experimental infection of pregnantdams in the first trimester of gestation with cyto-pathic BVDV failed to produce persistently infectedcalves (Casaro et al., 1971; Brownlie et al., 1989;McClurkin et al., 1984). The simultaneous experi-mental infection of type 1 and type 2 genotypes dur-ing gestation has resulted in dual persistent infec-tions with the two BVDV genotypes (Brock andChase, 2000).

CONSEQUENCES OF BVDV INFECTION FROM125–175 DAYS OF GESTATION

Fetal infection during fetal development andorganogenesis in the middle trimester can result innumerous types of congenital anomalies. The com-bination of direct cellular damage by virus and theresultant inflammatory response to the foreign viralantigens have been proposed as pathogenic mecha-nisms for congenital anomalies (Castrucci et al.,1990). Congenital defects of the central nervous sys-tem are the most common and may include cerebel-lar hypoplasia (Kahrs et al., 1970; Scott et al., 1973),microencephaly, hydrocephalus, hydranencephaly(Badman et al., 1981), porencephaly (Hewicker-Trautwein and Trautwein, 1994), and hypomyelina-tion (Binkhorst et al., 1983). Cerebellar hypoplasiais the most commonly recognized congenital defectassociated with BVDV infection (Ward et al., 1969,Kahrs et al., 1970; Scott et al., 1973; Wilson et al.,1983). Congenital infection has been correlated withataxia, tremors, wide-based stance, stumbling with-out compensation or resolution, “dummy calves,”and failure or inability to nurse. The defects fre-quently are severe enough that compensation doesnot occur and the calves may either die or requireeuthanasia (Baker, 1987). Histologically, the cere-bellar lesions have been described as containing re-duced numbers of molecular layer cells and granularlayer cells (Done et al., 1980; Brown et al., 1973;Brown et al., 1974; Bielefeldt-Ohmann, 1984) andreduced numbers and displaced Purkinje cells(Brown et al., 1973). Fetal cerebellar effects havebeen seen following infection as early as 79 daysand as late as 150 days of gestation (Brown et al.,1973). The severity of cerebellar lesions apparently

increases with the age of the fetus at the time of in-fection up to 150 days of gestation (Brown et al.,1973). Numerous other congenital defects have beenassociated with BVDV infection and include hyenadisease (Espinhasse et al., 1986) (osteochondrosis),growth retardation, optic neuritis (Bielefeldt-Ohmann, 1984), retinal degeneration (Scott et al.,1973), thymic hypoplasia (Done et al., 1980),hypotrichosis/alopecia (Baker, 1987; Kendrick,1971), curly haircoat (Larsson et al., 1991), de-ranged osteogenesis (Constable et al., 1993), mi-croopthalmia (Kahrs et al., 1970; Scottt et al., 1973;Brown et al., 1975), cataracts (Bielefeldt-Ohmann,1984; Wohrmann et al., 1992), mandibular brachyg-nathism (Scott et al., 1972), and growth retardation(Baker, 1987; Done et al., 1980, Constable et al.,1993). From a diagnostic aspect, it is important tonote that it is generally difficult to isolate BVDV ordemonstrate viral antigen in calves that exhibit signsassociated with the congenital defects as described.

CONSEQUENCES OF BVDV INFECTION FROM175 DAYS OF GESTATION TO TERM

In the later stages of gestation, immunocompetence,and organogenesis are generally complete. Althoughabortions and the birth of weak calves have been at-tributed to infection with BVDV late in gestation(Ward et al., 1969), fetuses infected during this timeperiod are normally able to mount an effectiveimmune response and clear the virus. These calvesare usually normal at birth and have precolostral-neutralizing antibodies to BVDV (Casaro et al.,1971; Kendrick, 1971; Braun et al., 1973; Ohmann etal., 1982; Orban et al., 1983). However, calves con-genitally infected with BVDV may be at more risk ofexperiencing a serious postnatal health event. It hasrecently been described that congenital infectionsduring the last trimester of gestation may negativelyaffect neonatal performance and survivability(Muñoz-Zanzi et al., 2003). It was reported thatcalves born with BVDV-neutralizing antibodies weretwice as likely to experience a severe illness duringtheir first 10 months of life as compared to calvesborn free of neutralizing antibodies. Further studiesare needed to determine whether detrimental effectsmay continue long-term (Muñoz-Zanzi et al., 2003).

IMPACT OF PERSISTENTINFECTION AND THERELATIONSHIP TO BVDV-ASSOCIATED DISEASEAs discussed previously, there are several outcomesof BVDV infection during gestation. However, fetal



infection resulting in persistent infection has severeimpact on BVDV-associated losses. It is universallyaccepted that persistent infections are the primarymethod for the spread of BVDV and the maintenanceof the virus in the cattle population. The birth of PIanimals leads to increased calf mortality and respira-tory tract disease. The PI animal is most commonlythe source of BVDV spread when different groups ofcattle are mixed. The birth of PI calves commonlyoccurs following the purchase of pregnant animals.Therefore, the ability of BVDV to cause persistentfetal infection has a significant impact on the contin-ued ability of BVDV to cause disease. Persistent in-fections occur in spite of using vaccines that effec-tively modulate and prevent acute disease.

IMPACT OF VIRAL FACTORS ONTHE OUTCOME OFREPRODUCTIVE DISEASEThe genotypic classification (1a, 1b, and 2) ofBVDV strains is based primarily on antigenic differ-ences in the E2 glycoprotein and comparative nu-cleotide sequencing. All BVDV genotypes (1a, 1b,and 2) are capable of causing persistent infections.In addition, persistent infection can be caused simul-taneously by a mixture of type 1 and type 2 BVDV(Brock and Chase, 2000). Certain BVDV outbreaksinvolving type 2 genotypes have been associatedwith increased rates of abortion, which would resultin a reduced number of PI calves being born follow-ing an outbreak. It has been demonstrated that thedistribution of type 1 and type 2 persistent infectionsin field studies is approximately equal (Wittum etal., 2001).

Differences between cytopathic and noncyto-pathic BVDV affect the outcome of reproductive in-fections. Persistent infection is the result only ofnoncytopathic BVDV infection. Cytopathic BVDVreplicates in the reproductive tract, and vaccine ori-gin virus has been isolated in ovarian tissues follow-ing vaccination. Transplacental transmission of cy-topathic BVDV to the fetus is rare as opposed torapid transplacental transmission of noncytopathicstrains. The mechanism for transplacental transmis-sion of BVDV is unknown but is thought to berelated to viremia in the pregnant dam. The mecha-nism influencing the differential ability of noncyto-pathic and cytopathic BVDV to infect the fetus isunknown. It is unclear why cytopathic BVDV doesnot establish persistent infection in the fetus.

Although several published studies are availableon the mechanisms of persistence and the role of in-terferon (IFN) during pregnancy, their findings are

conflicting. Infection of the fetus with cytopathicBVDV leads to IFN production in the fetus that mayprevent the establishment of persistent infection. Inexperimental studies, ncp BVDV fetal infection didnot induce IFN-� production. However, ncp BVDVfetal infection did result in IFN-� production in thecow (Charleston et al., 2001). It has been hypothe-sized that the protective occurrence of apoptosis andthe IFN response seen during the earliest stages ofpregnancy are important elements of the innate im-mune system protecting the fetus. Evasion of thesekey elements may be crucial not only for transmis-sion to the fetus, but also for the maintenance of im-munotolerance (Charleston et al., 2002)

Experimental data indicate that immunosuppres-sion caused by ncp BVDV infection may not be as-sociated with low interferon responses or elevatedlevels of TGF-ß (Schweizer and Peterhans, 2001).However, it does not exclude an influence of IFN in-hibition by ncp BVDV early in infection on the out-come of the immune response. More studies areneeded to better understand the mechanism of fetaltransplacental infection as well as the mechanism ofimmunotolerance.

IMPACT OF HOST FACTORS ONTHE OUTCOME OFREPRODUCTIVE DISEASEFrom experimental studies attempting to infect otherspecies, such as sheep and pigs, it is clear thatBVDV is highly adapted for replication in thebovine species. Attempts to induce persistent BVDVinfections by infecting pregnant sows at differentstages of gestation have not produced consistent re-sults. Although BVDV has been shown to inducepersistent infection in lambs following in utero inoc-ulation, congenital defects are more common fol-lowing fetal infection than persistent infection.Lambs persistently infected with BVDV generallydo not live for more than a few months. There havebeen no recognized differences between breeds ofcattle when considering susceptibility to and sever-ity of BVDV infection. Environmental factors suchas cattle density, type of cattle operation (dairy orbeef), type of housing, and biosecurity practices in-fluence the severity and prevalence of disease in cer-tain populations of cattle.

IMPACT OF VACCINATION ONTHE OUTCOME OFREPRODUCTIVE DISEASEVaccination against BVDV has a positive impact onreproductive health. Both experimental and field



studies have proven that vaccination reduces the po-tential impact of fetal infections with BVDV. Under-standing the strengths and limitations of vaccines hasbecome important with the recognition of the signif-icance of antigenic and genetic diversity in BVDV.Early vaccines were developed with little knowledgeof their ability to provide fetal protection. Accordingto 9CFR guidelines, for approval of BVDV vaccinesin the United States, the vaccines were required todemonstrate reduced clinical disease in vaccinatedanimals but did not address the ability of the vaccineto prevent fetal infection. Recently, the USDA hasdeveloped guidelines for conducting fetal challengeexperiments to obtain vaccine labels that can claimprevention of persistent infection and abortion due toBVDV infection. Vaccine efficacy in preventing thedevelopment of persistent infections is generally de-termined by experimental intranasal challenge ofpregnant animals at 75 days of gestation with ap-proximately 5 � 105 CCID50/ml of BVDV (Corteseet al., 1998; Brock and Cortese, 2001). The status ofthe fetus and persistent infection are determined bydetection of virus at least 75 days postchallenge (150days of gestation).

The influence of antigenic diversity on fetal pro-tection is evident by different rates of protection af-forded by monovalent (type 1) vaccines when exper-imental challenge is done with type 1 or type 2BVDV. Several field studies suggest that immuno-logical protection against heterologous BVDV chal-lenge may be incomplete with respect to fetal pro-tection (Bolin et al., 1991; Kelling et al., 1990; VanCampen et al., 2000). The efficacy of killed vaccinesin experimental studies has varied from 25–100%.In studies on evaluating the fetal protection efficacyof a modified-live vaccine, 88% and 58% levels offetal protection were demonstrated in heifers immu-nized once with a commercial BVDV-1 vaccine andchallenged at 75 days of gestation with type 1 ortype 2 BVDV, respectively (Cortese et al., 1998;Brock and Cortese, 2001). Therefore, it was con-cluded that BVDV-1 vaccine was less likely to stim-ulate fetal protective immunity against BVDV type2. To reduce the prevalence of persistent infections,it is important to promote routine vaccination aswell as strict biosecurity practices to prevent the in-troduction and continued maintenance of cyclicBVDV infections.

For controlling BVDV, it is important to preventfetal infections and the birth of calves that are per-sistently infected with BVDV. Eliminating fetal in-fections will correspondingly eliminate other repro-ductive losses. Prevention of fetal infections, and the

subsequent economic consequences, depend on re-ducing or eliminating exposure to BVDV and en-hancing specific anti-BVDV immunity by vaccina-tion (Sockett et al., 1997). Cattle persistentlyinfected with BVDV are the primary source of virusspread within and between farms. Identifying andeliminating persistent infections should be a majorfocus when attempting to control and preventBVDV. By eliminating persistently infected ani-mals, the major source of virus capable of causingtransient infections in pregnant dams and subse-quent fetal infections is removed.

CONCLUSIONSThe reproductive consequences of BVDV infectionsare economically important because of the impact ofboth persistent and acute infections and of losses as-sociated with producing animals persistently in-fected with BVDV. The primary consequence of re-productive infections is the occurrence of persistentinfections. In addition, losses are due to a decreasein fertility and reproductive efficiency. Although themechanism of persistent infections is known, themechanism of reproductive pathogenesis is notknown. Due to the capability of BVDV of crossingthe placenta and infecting the fetus, vaccination isnot considered to be an effective method for BVDVcontrol.

REFERENCESArchbald LF, Fulton RW, Seger CL, et al.: 1979,

Effect of the bovine viral diarrhea virus on preim-plantation bovine embryos: A preliminary study.Theriogenology 11:81–89.

Archbald LF, Gibson CD, Schultz RH, et al.: 1973,Effects of intrauterine inoculation of bovine viraldiarrhea-mucosal disease virus on uterine tubes anduterus of nonpregnant cows. Am J Vet Res34:1133–1137.

Badman RT, Mitchell G, Jones RT, Westbury HA:1981, Association of bovine viral diarrhoea virusinfection to hydrancephaly and other central nerv-ous system lesions in perinatal calves. Aust Vet J57:306–307.

Baker JA, York CJ, Gillespie JH, Mitchell GB: 1954,Virus diarrhea in cattle. Am J Vet Res 15:525–531.

Baker JC: 1987, Bovine viral diarrhea virus: a review.J Am Vet Med Assn 190:1449–1458.


Barlow RM, Nettleton PF, Gardiner AC, et al.: 1986,Persistent bovine virus diarrhoea virus infection ina bull. Vet Rec 118:321–324.



Bielanski A, Hare WC: 1988, Effect in vitro of bovineviral diarrhea virus on bovine embryos with thezona pellucida intact, damaged and removed. VetRes Commun 12:19–24.

Bielanski A, Loewen KG, DelCampo M, et al.: 1993,Isolation of bovine herpesvirus-1 (BHV-1) andbovine viral diarrhea (BVDV) in association withthe in vitro production of bovine embryos.Theriogenology 40:531–538.

Bielefeldt-Ohmann H: 1984, An ocular-cerebellar syn-drome caused by congenital viral diarrhea virus-in-fection. Acta Vet Scand 25:36–49.

Binkhorst GJ, Journee DL, Wouda W, et al.: 1983,Neurological disorders, virus persistence and hy-pomyelination in calves due to intrauterine infec-tion with bovine virus diarrhoea virus. Vet Q5:145–155.

Bolin SR, Littledike ET, Ridpath JF: 1991, Serologicdetection and practical consequences of antigenicdiversity among bovine viral diarrhea viruses in avaccinated herd. Am J Vet Res 52:1033–1037.

Bolin SR, McClurkin AW, Cutlip RC, Coria MF:1985, Severe clinical disease induced in cattle per-sistently infected with noncytopathic bovine viraldiarrhea virus by superinfection with cytopathicbovine viral diarrhea virus. Am J Vet Res46:573–576.

Booth PJ, Stevens DA, Collins ME, et al.: 1995,Detection of bovine viral diarrhoea virus antigenand RNA in oviduct and granulosa cells of persist-ently infected cattle. J Reprod Fertil 105:17–24.

Braun RK, Osburn BI, Kendrick JW: 1973,Immunologic response of bovine fetus to bovineviral diarrhea virus. Am J Vet Res 34:1127–1132.

Brock KV, Chase CC: 2000, Development of a fetalchallenge method for the evaluation of bovine viraldiarrhea virus vaccines. Vet Microbiol 77:209–214.

Brock KV, Cortese V: 2001, Experimental fetal chal-lenge using type II bovine viral diarrhea virus incattle vaccinated with modified-live virus vaccine.Vet Therapeutics 2:354–360.

Brock KV, Stringfellow DA: 1993, Comparative ef-fects of cytopathic and noncytopathic bovine viraldiarrhea on bovine blastocysts. Theriogenology39:196.

Brown TT, Bistner SI, de Lahunta A, et al.: 1975,Pathogenetic studies of infection of the bovine fetuswith bovine viral diarrhea virus. II. Ocular lesions.Vet Pathol 12:394–404.

Brown TT, de Lahunta A, Bistner SI, et al.: 1974,Pathogenetic studies of infection of the bovine fetuswith bovine viral diarrhea virus. I. Cerebellar atro-phy. Vet Pathol 11:486–505.

Brown TT, de Lahunta A, Scott FW, et al.: 1973,Virus induced congenital anomalies of the bovine

fetus. II. Histopathology of cerebellar degeneration(hypoplasia) induced by the virus of bovine viraldiarrhea-mucosal disease. Cornell Vet 63:561–578.

Brownlie J, Clarke MC, Howard CJ: 1984,Experimental production of fatal mucosal disease incattle. Vet Rec 114:535–536.

Brownlie J, Clarke MC, Howard CJ: 1989,Experimental infection of cattle in early pregnancywith a cytopathic strain of bovine virus diarrhoeavirus. Res Vet Sci 46:307–311.

Casaro AP, Kendrick JW, Kennedy PC: 1971,Response of the bovine fetus to bovine viral diar-rhea-mucosal disease virus. Am J Vet Res32:1543–1562.

Castrucci G, Frigeri F, Osburn BI, et al.: 1990, Astudy of some pathogenetic aspects of bovine viraldiarrhea virus infection. Comp Immunol MicrobiolInfect Dis 13:41–49.

Charleston B, Brackenbury LS, Carr BV, et al.: 2002,Alpha/beta and gamma interferons are induced byinfection with noncytopathic viral diarrhea virus invivo. J Virol 76:923–927.

Charleston B, Fray MD, Baigent S, et al.: 2001,Establishment of persistent infection with non-cytopathic bovine viral diarrhea virus in cattle isassociated with a failure to induce type I interferon.J Gen Virol 82:1893–1897.

Constable PD, Hull BL, Wicks JR, Myer W: 1993,Femoral and tibial fractures in a newborn calf aftertransplacental infection with bovine viral diarrheavirus. Vet Rec 132:383–385.


Cortese VS, Grooms DL, Ellis J: 1998, Protection ofpregnant cattle and their fetuses against infectionwith bovine viral diarrhea virus type 1 by use of amodified-live virus vaccine. Am J Vet Res59:1409–1413.


Donis RO, Dubovi EJ: 1987, Molecular specificity ofthe antibody responses of cattle naturally and ex-perimentally infected with cytopathic and noncyto-pathic bovine viral diarrhea virus biotypes. Am JVet Res 48:1549–1554.

Dow C, Jarret W, McIntyre W: 1956, A disease of cat-tle in Britain resembling the virus diarrhea-mucosaldisease complex. Vet Rec 68.




Duffell SJ, Sharp MW, Bates D: 1986, Financial lossresulting from BVD-MD virus infection in a dairyherd. Vet Rec 118:38–39.

Duffell SJ, Sharp MW, Winkler CE, et al.: 1984, Bo-vine viral diarrhea-mucosal disease virus-inducedfetopathology in cattle: Efficacy of prophylacticmaternal pre-exposure. Vet Rec 119:558–561.

Espinasse J, Parodi AL, Constantin A, et al.: 1986,Hyena disease in cattle: A review. Vet Rec118:328–330.


Fray MD, Mann GE, Bleach EC, et al.: 2002,Modulation of sex hormone secretion in cows byacute infection with bovine viral diarrhoea virus.Reproduction 123:281–289.

Fray MD, Mann GE, Clarke MC, Charleston B: 1999,Bovine viral diarrhea virus: Its effects on estradiol,progesterone and prostaglandin secretion in thecow. Theriogenology 51:1533–1546.

Fray MD, Mann GE, Clarke MC, Charleston B:2000a, Bovine viral diarrhoea virus: Its effects onovarian function in the cow. Vet Microbiol77:185–194.

Fray MD, Paton DJ, Alenius S: 2000b, The effects ofbovine viral diarrhoea virus on cattle reproductionin relation to disease control. Anim Reprod Sci60–61:615–627.

Fredriksen B, Press CM, Loken T, Odegaard SA:1999, Distribution of viral antigen in uterus, pla-centa and foetus of cattle persistently infected withbovine virus diarrhoea virus. Vet Microbiol64:109–122.

Gillespie JH, Bartholomew PR, Thompson RG,McEntee K: 1967, The isolation of noncytopathicvirus diarrhea virus from two aborted fetuses.Cornell Vet 57:564–571.

Gillespie JH, Schlafer DH, Foote RH, et al.: 1990,Comparison of persistence of seven bovine viruseson bovine embryos following in vitro exposure.Dtsch Tierarztl Wochenschr 97:65–68.

Givens MD, Heath AM, Brock KV, et al.: 2003a,Detection of bovine viral diarrhea virus in semenafter infection of seronegative, post-pubertal bulls.Am J Vet Res 64:428–434.

Givens MD, Heath AM, Carson RL, et al.: 2003b,Analytical sensitivity of assays used for detectionof bovine viral diarrhea virus in semen samplesfrom the Southeastern United States. Vet Microbiol96:145–155.

Grahn TC, Fahning ML, Zemjanis R: 1984, Nature ofearly reproductive failure caused by bovine viral di-arrhea virus. J Am Vet Med Assoc 185:429–432.

Grooms DL, Brock KV, Pate JL, et al.: 1998c.Changes in ovarian follicles following acute infec-tion with bovine viral diarrhea virus.Theriogenology 49:595–605.

Grooms DL, Brock KV, Ward LA: 1998a, Detectionof bovine viral diarrhea virus in the ovaries of cattleacutely infected with bovine viral diarrhea virus. JVet Diagn Invest 10:125–129.

Grooms DL, Brock KV, Ward LA: 1998b, Detectionof cytopathic bovine viral diarrhea virus in theovaries of cattle following immunization with amodified live bovine viral diarrhea virus vaccine. JVet Diagn Invest 10:130–134.

Grooms DL, Ward LA, Brock KV: 1996, Morphologicchanges and immunohistochemical detection ofviral antigen in ovaries from cattle persistently in-fected with bovine viral diarrhea virus. Am J VetRes 57:830–833.

Hewicker-Trautwein M, Trautwein G: 1994, Changesin porencephaly, hydranencephaly and leukoen-cephalopathy in ovine fetuses following trans-placental infection with bovine virus diarrhoeavirus: Distribution of viral antigen and characteriza-tion of cellular response. Acta Neuropathol87:385–397.

Houe H, Myrup-Pedersen K, Meyling A: 1993a, Theeffect of bovine virus diarrhea virus infection onconception rate. Prev Vet Med 15:117–123.

Houe H, Pedersen KM, Meyling A: 1993b, A comput-erized spread sheet model for calculating total an-nual national losses due to bovine viral diarrhoeavirus infection in dairy herds and sensitivity analy-sis of selected parameters. Second Symposium onPestiviruses, Lyon, France. pp. 179–184.

Huck R: 1957, A mucosal disease of cattle. Vet Rec69:1207–1215.

Kafi MR, McGowan M, Jillella D: 1994, The effect ofbovine viral diarrhoea virus (BVDV) during follicu-lar development on the superovulatory response ofcattle. Theriogenology 41:223.

Kahrs RF: 1968, The relationship of bovine viral diar-rhea-mucosal disease to abortion in cattle. J Am VetMed Assoc 153:1652–1655.

Kahrs RF, Scott FW, de Lahunta A: 1970, Congenitalcerebella hypoplasia and ocular defects in calvesfollowing bovine viral diarrhea-mucosal disease in-fection in pregnant cattle. J Am Vet Med Assoc156:1443–1450.

Kelling CL, Stine LC, Rump KK, et al.: 1990,Investigation of bovine viral diarrhea virus infec-tions in a range beef cattle herd. J Am Vet MedAssoc 197:589–593.

Kendrick JW: 1991, Bovine viral diarrhea-mucosaldisease virus infection in pregnant cows. Am J VetRes 32:533–544.



Kirkland PD, Richards SG, Rothwell JT, Stanley DF,et al.: 1991, Replication of bovine viral diarrhoeavirus in the bovine reproductive tract and excretionof virus in semen during acute and chronic infec-tions. Vet Rec 128:587–590.

Kirkland PD, McGowan MR, Mackintosh SG: 1993,Factors influencing the development of persistentinfection of cattle with pestivirus. Second Sym-posium on Pestiviruses, Lyon, France. pp. 117–121.

Kirkland PD, Mackintosh SG, Moyle A: 1994, Theoutcome of widespread use of semen from a bullpersistently infected with pestivirus. Vet Rec135:527–529.

Kirkland PD, McGowan MR, Mackintosh SG, et al.1997, Insemination of cattle with semen from a bulltransiently infected with pestivirus. Vet Rec140:124–127.

Kommisrud E, Vatn T, Lang-Ree JR, Loken T: 1996,Bovine virus diarrhoea virus in semen from acutelyinfected bulls. Acta Vet Scand 37:41–47.

Larsson B, Jacobsson SO, Bengtsson B, et al.: 1991,Alenius Summer Congenital curly haircoat as asymptom of persistent infection with bovine virusdiarrhoea virus in calves. Arch Virol Suppl3:143–148.

Liess B, Orban S, Frey H-R, et al.: 1984, Studies ontransplacental transmissibility of a bovine virusdiarrhoea (BVD) vaccine virus in cattle. II.Inoculation of pregnant cows without detectableneutralizing antibodies to BVD virus 90–229 daysbefore parturition (51st to 190th day of gestation).Zentbl Vet Med B 31:669–681.


McClurkin AW, Littledike ET, Cutlip RC: 1984,Production of cattle immunotolerant to bovine viraldiarrhea virus. Can J Comp Med 48:156–161.

McGowan MR, Kirkland PD: 1995, Early reproduc-tive loss due to bovine pestivirus infection. Br vet J151:263–270.

McGowan MR, Kirkland PD, Richards SG,Littlejohns IR.: 1993, Increased reproductive lossesin cattle infected with bovine pestivirus around thetime of insemination. Vet Rec 133:39–43.

Meyling A, Jensen AM: 1988, Transmission of bovinevirus diarrhoea virus (BVDV) by artificial insemi-nation (AI) with semen from a persistently infectedbull. Vet Microbiol 17:97–105.

Muñoz-Zanzi CA, Hietala SK, Thurmond MC, et al.:2003, Quantification, risk factors, and health impactof natural congenital infection with bovine viral di-arrhea virus in dairy calves. Am J Vet Res64:358–365.

Nielson S, Horney F, Hulland T, et al.: 1955, Mucosaldisease of cattle in Ontario. Can J Comp Med19:318–324.

Niskanen R, Alenius S, Belak K, et al.: 2002,Insemination of susceptible heifers with semenfrom a non-viraemic bull with persistent bovinevirus diarrhoea virus infection localized in thetestes. Reprod Domest Anim 37:171–175.

Ohmann HB, Jensen MH, Sorensen KJ, Dalsgaard K:1982, Experimental fetal infection with bovine viraldiarrhea virus. I. Virological and serological stud-ies. Can J Comp Med 46:357–362.

Olafson P, MaCallum A, Fox F: 1946, An apparentlynew transmissable disease of cattle. Cornell Vet36:205–213.

Orban S, Liess B, Hafez SM, et al.: 1983, Studies ontransplacental transmissibility of bovine virus diar-rhoea (BVD) vaccine virus. Zenblt Vet Med B30:619–634.


Paton DJ, Goodey R, Brockman S, et al.: 1989,Evaluation of the quality and virological status ofsemen from bulls acutely infected with BVDV. VetRec 124:63–64.

Potter ML, Corstvet RE, Looney CR, et al.: 1984, Eval-uation of bovine viral diarrhea virus uptake by pre-implantation embryos. Am J Vet Res 45:1778–1780.

Revell SG, Chasey D, Drew TW, Edwards S: 1988,Some observations on the semen of bulls persist-ently infected with bovine virus diarrhoea virus. VetRec 123:122–125.

Roeder PL, Jeffrey M, Cranwell MP: 1986, Pestivirusfetopathogenicity in cattle: Changing sequella withfetal maturation. Vet Rec 118:44–48.

Rufenacht J, Schaller P, Audige L, et al.: 2001, Theeffect of infection with bovine viral diarrhea viruson the fertility of Swiss dairy cattle. Theriogen-ology 56:199–210.


Scott FW, Kahrs RF, de Lahunta A, et al.: 1973,Virus-induced congenital anomalies of the bovinefetus. I. Cerebellar degeneration (hypoplasia),ocular lesions and fetal mummification followingexperimental infection with bovine viral diarrhea-mucosal disease virus. Cornell Vet 63:536–560.

Scott FW, Kahrs RF, Parsonson IM: 1972, A cytopath-ogenic strain of bovine viral diarrhea-mucosal dis-ease virus isolated from a bovine fetus. Cornell Vet62:74–84.



Shope R: 1968, Comments on bovine viral diarrhea-mucosal disease. J Am Vet Med Assoc 152:769.

Singh E, Eaglesome M, Thomas F, et al.: 1982,Embryo transfer as a means of controlling thetransmission of viral infections I. The in vitro expo-sure of preimplantation embryos to Akabane virus,blue tongue virus, and bovine viral diarrhea virus.Theriogenology 17:437–444.

Sockett D, Modderman E: 1997, Fetal exposure tobovine viral diarrhea virus despite vaccination in aWisconsin diary herd. First International VirtualConference on Infectious Diseases of Animals.National Animal Disease Center, Ames, IA, USA.http://www.nadc.ars.usda.gov/virtconf/submabs/ab-stracts/B00029.html.

Sprecher D, Baker J, Holland R, et al.: 1991, An out-break of fetal and neonatal losses associated withthe diagnosis of bovine viral diarrhea virus in adairy herd. Theriogenology 36:567–606.

Ssentongo YK, Johnson RH, Smith JR: 1980, Asso-ciation of bovine viral diarrhea-mucosal diseasevirus with ovaritis in cattle. Aust Vet J 56:272–273.

Swope RE, Luedke AJ: 1956, A mucosal disease incattle in Pennsylvania. J Am Vet Med Assoc129:111–115.


Virakul P, Fahning ML, Joo HS, Zemjanis R: 1988,Fertility of cows challenged with a cytopathic strainof bovine viral diarrhea virus during an outbreak ofspontaneous infection with a noncytopathic strain.Theriogenology 29:441–449.

Voges H, Horner GW, Rowe S, Wellenberg GJ: 1998,Persistent bovine pestivirus infection localized in

the testes of an immuno-competent, non-viraemicbull. Vet Microbiol 61:165–175.

Ward GM, Roberts SJ, McEntee K, Gillespie JH:1969, A study of experimentally induced bovineviral diarrhea-mucosal disease in pregnant cowsand their progeny. Cornell Vet 59:525–538.

Wentink GH, Remmen JL, van Exsel AC: 1989,Pregnancy rate of heifers bred by an immunotoler-ant bull persistently infected with bovine viral diar-rhoea virus. Vet Q 11:171–174.

Whitmore HL, Archbald LF: 1977, Demonstrationand quantitation of immunoglobulins in bovineserum, follicular fluid, and uterine and vaginal se-cretions with reference to bovine viral diarrhea andinfectious bovine rhinotracheitis. Am J Vet Res38:455–457.

Whitmore HL, Zemjanis R, Olson J: 1981, Effect ofbovine viral diarrhea virus on conception in cattle.J Am Vet Med Assoc 178:1065–1067.

Wilson TM, de Lahunta A, Confer L: 1983, Cerebellardegeneration in dairy calves: clinical, pathologic,and serologic features of an epizootic caused bybovine viral diarrhea virus. J Am Vet Med Assoc183:544–547.

Wittum TE, Grotelueschen DM, Brock KV, et al.:2001, Persistent bovine viral diarrhea virus infec-tion in U.S. beef herds. Vet Prev Med49:83–94.103.

Wohrmann T, Hewicker-Trautwein M, Fernandez A,et al.: 1992, Distribution of bovine virus diarrhoeaviral antigens in the central nervous system of cat-tle with various congenital manifestations.Zentralbl Vet Med B 39:599–609.

Zurovac O, Stringfellow DA, Brock KV, et al.: 1994,Noncytopathic bovine viral diarrhea virus in a sys-tem for in vitro production of bovine embryos.Theriogenology 41:841–853.



INTRODUCTIONClinical manifestations of BVDV may include inap-parent, acute, or persistent subclinical infections;fetal death and congenital abnormalities; wastingdisease; severe acute disease (which may progress tohemorrhagic syndrome); or mucosal disease. Theclinical outcome of infection depends on the im-mune status of the animal and time of infection. Acomplete understanding of the immune processesand immunogens of BVDV, as given in the next sec-tion, should be helpful in designing better preventivestrategies for BVDV infections.

IMMUNE RESPONSES TO BVDVBVDV strains vary in their tropism for bovine tis-sues. Hence, the effect of different strains of BVDVon the immune system also differs and may affectthe type of disease exhibited. The natural route oftransmission of BVDV is oronasal by contact withsuspended droplets or mucus, although genitaltransmission may also occur. The virus first repli-cates in the nasal mucosa and tonsils from wherewhite blood cells help spread the virus throughoutthe body by binding to surface receptors (Bruschkeet al., 1998a, 1998b). Although the biochemical na-ture of BVDV receptors is not well understood, a 50kDa protein is believed to be the viral receptor (Xueet al., 1997).

Pestivirus infections are associated with leukope-nia, immunosuppression, and in some cases, hem-orrhages. In general, the immunosuppressive prop-erties of BVDV lead to a reduction in local defensemechanisms, thereby predisposing calves to otherrespiratory pathogens. Immune responses to BVDVmay develop following vaccination, infection, ex-posure to cross-reactive pestiviruses, or by pas-sively acquiring BVDV-specific antibodies fromcolostrum.

INNATE IMMUNE RESPONSE

The innate/natural (non–antigen-specific) immuneresponse can influence the outcome of BVDV infec-tion. BVDV can infect cells of the innate immunesystem (e.g., neutrophils, monocytes, macrophages,and dendritic cells) and affect their function (Pot-gieter, 1995; Glew et al., 2003; Lambot et al., 1998;Peterhans et al., 2002). Infection with BVDV mayresult in impairment of microbicidal, chemotatic,and antibody-dependent cell-mediated cytotoxicityof neutrophils (Potgieter, 1995). In monocytes, in-fection with cp BVDV may lead to apoptosis (Glewet al., 2003; Lambot et al., 1998). A 30–70% de-crease in monocyte numbers may occur followinginfection of calves with virulent BVDV (Archam-bault et al., 2000). In vitro or in vivo infection ofalveolar macrophages (AM) with BVDV may leadto decreases in phagocytosis, expression of Fc (FcR)and complement receptors (C3R), microbicidal ac-tivity, and chemotatic factors (Welsh et al., 1995;Liu et al., 1999; Peterhans et al., 2002). Infection ofAM also causes increased LPS-induced (lipopoly-saccharide) procoagulant activity, which can lead tobacterial colonization and may adversely affect thenormal defense mechanism of the lung (Olchowy etal., 1997).

Cytokines can mediate the effects of both innateand specific immunity (Nobiron et al., 2001). Cyto-kines are small soluble proteins secreted by certaincells. They alter not only the function of cells produc-ing them but also of other cells on which they mightact. Tumor necrosis factor -� (TNF-�) is a cytokineproduced mainly by macrophages. It plays an impor-tant role in the activation of the immune response bymodulating the production and activity of many othercytokines (Chase, 2004). Infection of AM withBVDV leads to a decrease in superoxide anion andTNF, enhanced nitric oxide (NO) synthesis in re-

157

9Immunity and Immunosuppression

Sanjay Kapil, Paul Walz, Melinda Wilkerson, and Harish Minocha


sponse to LPS (Potgieter, 1995; Adler et al., 1994;Adler et al., 1997), stimulation of prostaglandin E2synthesis (Van Reeth and Adair, 1997; Welsh andAdair, 1995), induction of IL-1 inhibitors (Jensen andSchultz, 1991), and a decrease in cytokine-inducedchemotaxis (Ketelsen et al., 1979). Certain solublefactors released by infected monocytes and macro-phages induce apoptosis when added to uninfectedcells (Lambot et al., 1998; Adler et al., 1997). Apop-tosis is the process that commits cells to programmedcell death to eliminate an infected cell and is believedto be the cause of lymphoid tissue pathology seen inmucosal disease and in disease caused by highly vir-ulent BVDV (Liebler-Tenorio et al., 2003; Liebler-Tenorio and Ridpath, 2002; Stroffregen et al., 2000).Cytokines such as IL-2 and granulocyte-macrophagecolony-stimulating factor enhance both humoral andcellular immune responses against BVDV and play acritical role in fetal-maternal interface by ensuringthat pregnancy proceeds successfully (Graham et al.,1992). IL-1 is one of the endogenous pyrogens thatact upon the hypothalamus to alter the regulation ofbody temperature.

Interferon (IFN) is the most important innate de-fense antiviral cytokine. Type I interferons are twoserologically distinct proteins including IFN -� pro-duced by phagocytes and IFN-ß produced by fibro-blasts. Viral infections, including BVDV infections,strongly signal the induction of type I IFN, whichincreases the cytotoxic potential of natural killer(NK) cells. Treatment of cells with high doses (104

units/ml) of human IFN -� prevented the replicationof both ncp and cp BVDV in vitro while that withhuman INF-�, TNF-� or TNF-ß did not (Sentsui etal., 1998). Infection of the fetus with cp BVDVleads to IFN production, which probably preventsthe establishment of persistent infection. Infectionof the fetus with ncp BVDV does not induce IFN-�production although it did induce abundant amountsof all IFN (�, ß, and �) in gnotobiotic calves (Char-leston et al., 2001a, 2002). These responses were as-sociated with depressed levels of transforminggrowth factor beta (TGF-ß) in serum. These resultsindicate that the immunosuppression caused by ncpBVDV may not be associated with low interferonresponses or elevated levels of TGF-ß (Charleston etal., 2001a).

Antigen-presenting cells (APC; dendritic cells,macrophages, and monocytes) internalize the viralantigen and present it to T-helper cells with assis-tance from IFN-� and IL-12. However, infection ofAPC with BVDV causes a reduction in Fc and C3receptor expression, receptors that are required for

phagocytic activity (Welsh et al., 1995; Adler et al.,1996). It also reduces the ability of monocytes topresent antigen to T-helper cells (Glew et al., 2003).Dendritic cells, the most important APC in thelymph node, on the other hand, were not affected intheir ability to present antigen to T-helper cells or insurface marker expression (Glew et al., 2003).

HUMORAL IMMUNE RESPONSE

Although both active and passive humoral immuneresponses are protective, they differ in longevity andtheir ability to potentiate an immune response fol-lowing a subsequent exposure. Antibody response toBVDV is detectable 2–3 weeks postinfection andmay plateau in approximately 10–12 weeks postin-fection (Howard et al., 1992). Humoral immunity, asdetected by serum antibodies against BVDV, mayresult from an active immune response following anexposure to BVDV antigen (active immunity) or theingestion of antibody present in colostrum (passiveimmunity). Three glycoproteins of BVDV (gp53/E2, gp 48/E0, and gp 25/E1) induce neutralizingantibodies, with E2 protein being immunodominant(Bolin and Ridpath, 1990). Antibodies against sev-eral other viral proteins (115, 90, 48, and 25 kDa)have also been detected in some cattle (Bolin andRidpath, 1990; Boulanger et al., 1991).

Colostral antibodies

Antibodies do not cross the placenta of cattle as theydo in humans. Thus, calves receive passive immu-nity by absorption, through the gastrointestinal tract,of immunoglobulins contained in the ingested colos-trum. However, calves can absorb colostral antibod-ies only during the first 24–48 hrs of life when theirgastrointestinal tract is permissive to the transfer ofthese molecules across the mucosal epithelium, aphenomenon called “the open gut.” The highest con-centration of BVDV antibodies in colostrum occursonly in the first few days of lactation after parturi-tion. After that, the colostrum changes to normalmilk and the amount of antibodies decreases rapidly.

Passive antibodies play an important role in pro-tection from BVDV infection in the neonatal calf.However, the presence of high concentrations ofmaternal antibody in animals may block the induc-tion of active B-cell immune response to BVDVvaccination. This has led to the suggestion that vac-cination should be administered when passively ac-quired antibodies are declining (Ellis et al., 2001).However, T-cell immune responses have been ob-served when calves were infected intranasally withBVDV in the presence of maternal antibodies



(Endsley et al., 2003; Ridpath et al., 2003), suggest-ing that vaccination in the face of maternal antibod-ies may be effective.

Passive immunity is not fail safe, however.Factors such as dose of challenge virus, poor nutri-tion, exposure to harsh weather, poor ventilation,transportation stress, and increased stocking densitymay affect the outcome of infection in the presenceof colostral antibodies. In addition, antigenic varia-tion among BVDV may result in incomplete protec-tion, and colostrum may vary in quality and quantityof available antibodies. Nutrition and vaccinationstatus of the dam will also affect the immune qual-ity of colostrum. For example, vaccination of preg-nant cows in the last month of pregnancy with killedvaccines will improve the amount of antiviral anti-bodies in colostrum. Vaccination during pregnancywith live, modified vaccines will also increase theamount of antibody in colostrum, but this type ofvaccination is advised only when the dam has preex-isting antibodies. Further, the benefits of colostrumare not limited to the action of passive antibodies;immune lymphocytes and macrophages present incolostrum are also of help. In this regard, freshcolostrum is better than frozen colostrums.

CELLULAR IMMUNE RESPONSE

Infection with BVDV results in mild (10–20% de-crease) or severe lymphopenia (50–60% decrease)depending upon the strain of the virus (Brodersenand Kelling, 1999; Archambault et al., 2000).Cytotoxic T-lymphocytes (CD8+) are affected morethan helper T-lymphocytes (CD4+) with little or noaffect on circulating �/� T-cells (Brodersen andKelling, 1999; Ellis et al. 1988). Depletion of CD4+cells increases the period of virus shedding; that ofCD8+ and �/� T-cells does not, indicating that CD4+helper cells play a pivotal role in coordinating a cell-mediated response early in infection. These CD4+responses are directed primarily at NS3 and E2 pro-teins (Lambot et al., 1997; Collen and Morrison,2000; Collen et al., 2000; Collen et al., 2002) andalso against the capsid protein (C), glycoprotein Erns,amino-terminal proteinase (Npro), and nonstructuralprotein 2-3 (NS2-3) (Collen et al., 2002).

Proliferation assays are an indirect measure ofCD4+ responses to viral antigens. The cells of Th1phenotype produce IL-2 and IFN-�, but not IL-4 orB-cell stimulatory activity. In contrast, cells of Th2phenotype express high levels of B-cell growth fac-tor and IL-4 activity with relatively low levels of IL-2 and IFN-� (Rhodes et al., 1999). IL-4 increasesthe expression of MHC II on B-cells and is also im-

portant for the growth and survival of Th2 re-sponses. In cattle, unlike human and mouse systems,pregnancy has no observable shift in IL-4 pattern(Waldvogel et al., 2000).

The response generated by cp and ncp BVDV isdifferent. For example, ncp viruses tend to shift theimmune response toward the Th2 response andavoid the production of high levels of cell-mediatedimmunity (Lambot et al., 1997; Rhodes et al., 1999).cp BVDV, on the other hand, produces higher CMIresponse (Th1), along with up-regulation of IL-2 re-ceptor (IL-2R) in response to increased levels of IL-2. The down-regulation of IFN-� by acute ncpBVDV infection inhibits the cell-mediated responseto Mycobacterium bovis that could result in the fail-ure to identify cattle with tuberculosis (Charleston etal., 2001b). Delayed-type hypersensitivity is a cellu-lar immune response that occurs approximately 18hours after exposure to an antigen and is used as anassay for cell-mediated immunity. This response ischaracterized by induration and erythema at the siteof antigen injection such as with the Mycobacteriumantigen. After exposure to BVDV, the delayed typehypersensitivity to Mycobacterium is inhibited.Thus, BVDV causes general and nonspecific inhibi-tion of cellular immune responses in cattle (Thoenand Waite, 1990) and may interfere with the diagno-sis of bovine tuberculosis (Charleston et al., 2001b).

Proliferating CD8+ cytotoxic T-lymphocytes(CTL) produce IL-2 and IFN-�, indicating a type 1memory response in BVDV-seropositive cattle(Howard et al., 1992; Rhodes et al., 1999). Althoughfine mapping of BVDV CTL epitopes has not beendone, computer predictions based on MHC I bind-ing domains indicate that regions in the C, Erns, E2,and NS2-3 are likely BVDV CTL epitopes (Hegdeand Srikumaran, 1997).

IMMUNE RESPONSE OF PERSISTENTLYINFECTED ANIMALS

Persistent BVDV infections can arise in animals in-fected as fetuses. Persistently infected cattle are im-munotolerant to the infecting BVDV isolate but maymount an immune response to heterologous BVDV.It has been demonstrated that heifers carrying PIcalves developed BVDV antibody titers 5–10 timeshigher than those carrying non-PI calves (Brownlieet al., 1998). A number of studies have been done tounderstand the immunological defects in PI animals(Chase, 2004). The inability of ncp BVDV to induceIFN-� in the fetus is certainly one of the major im-mune evasion mechanisms that allow BVDV to es-tablish persistence (Charleston et al., 2002).

Immunity and Immunosuppression 159


ASSESSMENT OF IMMUNITY

HERD IMMUNITY

When assessing herd immunity, it is important tosample several animals or use a pooling strategy.The exact parameters of cellular and humoral im-mune responses that indicate protection againstBVDV infection have not been determined.Veterinary practitioners routinely screen cattle herdsfor humoral immune response against BVDV to as-sess the effectiveness of vaccination programs.However, it is difficult to interpret serological pro-files when only a single serum sample is tested.Because antibody titers resulting from natural infec-tions usually exceed those generated from vaccina-tion, uniformally high antibody titers within a herdor jointly housed group may indicate a recentBVDV outbreak. In addition, the presence of anti-bodies in fetal thoracic fluid or of precolostral anti-bodies in neonates is suggestive of fetal BVDV in-fection. Cattle that are not immunotolerent toBVDV antigens and are not sufficiently immunizedcan then be immunized with a BVDV vaccine con-taining a slightly different antigen. Enough individ-ual variation exists among outbred cattle that someanimals may need booster injections to maintainhigh levels of herd immunity.

The serum neutralization (SN) test has been ex-tensively used as a correlate of protective immunity.Since the SN test measures the total neutralizing ac-tivity of a serum sample or a body fluid, it representsa composite of neutralizing activity due to all classesof antibodies that have activity against surface pro-teins of BVDV. Some regions of viral proteins arehighly conserved between members of pestiviruses,and others are variable between different membersof the family. Most of the highly conserved regionsare found in the inner part of the virion. Type-specific variable regions are present on the outersurface of the BVDV protein. Members of pes-tiviruses, such as BVDV, classical swine fever virus,and border disease virus often cross-react in SNtests (de Smit et al., 1999).

There are two major problems with BVDV serol-ogy as it is currently used: a lack of test standardiza-tion and the existence of antigenic variation amongBVDV isolates. In an interlaboratory study of 14U.S. diagnostic laboratories, SN titers for a singleserum sample varied from 1:8 to 1:3642 (Vaugn,1997). This variability may be due to technical ex-pertise of the person performing the test and to virusstrain, cell type and passage, and amount of virusused in the test. In one study, the SN titers varied by

5–10-fold with the use of different BVDV strains(Fort Dodge, 1999b) while in another study theyvaried from 2–250 fold (Fulton et al., 1997). Al-though ELISA tests hold the promise of negatingmany of the above variables, they have not beenwidely used because they suffer from a lack of cor-relation with SN titers, which appear to correlate rel-atively well with protective immunity againstBVDV (Bolin and Saliki, 1996). At present, thereare two recognized genotypes of BVDV (1 and 2)and several proposed subgenotypes. Although thereis cross-reactivity between BVDV 1 and BVDV 2,cross-reactivity is significantly higher within agenotype than between genotypes (Jones et al.,2001). Most of the veterinary diagnostic laboratoriesin the U.S. now perform both type 1 and type 2 SNtests.

PROTECTIVE IMMUNITY

A large number of BVDV vaccines are available inthe U.S. They are usually approved on the basis oftheir safety, their ability to induce serological re-sponse against the virus, and protection achieved inchallenge protection experiments. In view of anti-genic diversity of BVDV, the panel of challengeviruses should include both homologous and het-erologous BVDV isolates from geographical areaswhere vaccines are to be administered. Inclusion ofnovel, antigenically different variants of BVDV inthe challenge panel would also be of value.

BVDV has at least two genotypes (genotype 1 and2) and two biotypes (cytopathic and noncytopathic).Despite widespread vaccination and the availabilityof a wide range of vaccines (Van Oirschot et al.,1999), BVDV remains a problem in most areas ofthe United States, raising the concern that evolutionof the virus may occur under immunological pres-sure leading to the continued emergence of anti-genic variation among BVDV isolates. With the re-cent recognition of genotype 2 BVDV, most vaccinemanufacturers have now started to include bothBVDV 1 and BVDV 2 in their vaccines. Althoughtype 1 BVDV vaccines provide some protectionagainst type 2 BVDV (Cortese et al., 1998; Fair-banks et al., 2003; Makoschey et al., 2001), a num-ber of challenge studies have indicated that the bestprotection rates are obtained with the use of homol-ogous BVDV vaccines (Potgieter, 1995; Fulton etal., 2003).

IMMUNOSUPPRESSIONDiverse clinical manifestations are associated withBVDV infection in cattle. Although a majority of



BVDV infections in immunocompetent cattle aretransient and self-limiting, it is apparent that wheninfection occurs in the presence of other microor-ganisms, BVDV can contribute to a disease that be-comes clinically evident (Baker, 1995). Both naturaland experimental studies have demonstrated the re-lationship between BVDV and other infectiousagents, suggesting that BVDV has the ability to in-duce immunosuppression in immunocompetent ani-mals. Impairment of lymphocyte and neutrophilfunctions decreases in the number of circulating andtissue immune cells, and environmental and/or man-agement stressors have been cited as contributors toimmunosuppression (Potgieter, 1995).

BVDV AND SECONDARY INFECTIONS

Increased susceptibility to secondary infections is aconsequence of BVDV-induced immunosuppres-sion. BVDV increases susceptibility to secondaryinfections because lymphocytes from BVDV-infected animals have impaired memory responsesto other antigens and BVDV (Lamontage et al.,1989). Some pathogens may induce disease alone,but the disease is enhanced in the presence ofBVDV. In some cases, BVDV-induced disease canbe enhanced by opportunistic organisms. Substantialdata indicate that BVDV infection is important inmultiple-etiology diseases. Bovine respiratory dis-ease complex in feedlot animals and in intensivelyhoused calves is an example of this type of multiple-etiology disease process.

It is debatable whether BVDV-induced immuno-suppression or primary infection of the respiratorytract plays a major role in the bovine respiratorydisease complex. Little experimental evidence ex-ists establishing a primary role of BVDV in thebovine respiratory disease complex (Potgieter,1997). To support a primary role, BVDV is fre-quently isolated from pneumonic lungs of cattle(Greig et al., 1981). In a study on respiratory dis-ease outbreaks with multiple virus infections,BVDV was isolated from pneumonic cattle morefrequently than any other virus (Richer et al.,1988). Suggestive evidence exists for certainBVDV isolates to be pneumotropic (Jewett et al.1990; Potgieter et al., 1985). In an experimental in-fection study, endobronchial inoculation of two dif-ferent isolates of BVDV resulted in interstitialpneumonia. However, when calves were challengedwith Mannheimia hemolytica, more severe diseaseresulted with only one of the two isolates (Potgieteret al., 1985). In another study, certain BVDV iso-lates were shown to be able to induce severe respi-

ratory disease in gnotobiotic lambs (Jewett et al.,1990).

Although BVDV is frequently isolated from cattlewith pneumonia, it is often present with other infec-tious agents, including bovine herpes virus-1 (BHV-1) (Biuk-Rudan et al., 1999; Greig et al., 1981),parainfluenza-3 virus (PI-3) (Dinter and Bakos,1961; Fulton et al., 2000), bovine respiratory coron-avirus (BCV), bovine respiratory syncytial virus(BRSV) (Brodersen and Kelling, 1998, 1999),Mannheimia hemolytica (Fulton et al., 2002), Pas-teurella multocida (Fulton et al., 2002), Myco-plasma bovis (Martin et al., 1990; Shahriar et al.,2002; Haines et al., 2001), and Hemophilus somnus.

In addition to causing immunosuppression,BVDV may interact directly with pathogens to makethem more virulent. Combined infection of bovinealveolar macrophages with BRSV and BVDV pro-duces synergistic depression of alveolar macro-phage functions (Liu et al., 1999). Concurrent infec-tion with BVDV and BRSV in cattle causes moresevere enteric and respiratory disease, and these an-imals shed higher concentrations of BVDV in theirnasal secretions (Brodersen and Kelling, 1998).Reports from disease outbreaks and experimentalstudies have supported a role for synergism withBHV-1 also (Greig et al., 1981; Potgieter et al.,1984). An experimental infection study was per-formed for evaluating this synergism. Calves inocu-lated with BVDV 7 days prior to inoculation withthe Cooper strain of BHV-1 developed severe clini-cal disease, with dissemination of BHV-1 into non-respiratory tissues, including intestinal and oculartissues, as compared to calves inoculated with BHV-1 alone (Greig et al., 1981; Potgieter et al., 1984).These observations suggested that initial BVDV in-fection impaired the ability of calves to clear BHV-1 from the lungs and to contain BHV-1 at the localinfection site.

The ability of BVDV to synergistically interactwith pathogens does not appear to be confined to therespiratory tract either, since synergism betweenBVDV and enteric pathogens has also been reported(Kelling et al., 2002a; Woods et al., 1999). From ex-perimental infection studies in calves, it has beenobserved that BVDV and rotavirus may interact in asynergistic manner to induce more severe clinicaldisease (Kelling et al., 2002a). Synergism betweenBVDV and transmissible gastroenteritis virus wasdemonstrated by generalized lymphocyte depletionthroughout the lymphatic system and villous atro-phy in the intestinal tract of experimentally infectedpigs (Woods et al., 1999).



CLINICOPATHOLOGICAL ASSESSMENT OFIMMUNOSUPPRESSION

Although a majority of BVDV infections in im-munocompetent, seronegative cattle are subclinical(Baker, 1995), pyrexia, mild inappetance, and a de-crease in milk production can be detected on closeexamination. In clinical infections in cattle, hemato-logic abnormalities such as thrombocytopenia andleukopenia are frequently reported (Archambault etal., 2000; Bolin et al., 1985; Ellis et al., 1998; Kell-ing et al., 2002a, 2002b: Walz et al., 2001). De-creases in white blood cells in the peripheral circu-lation (quantitative disorder or leukopenia) as wellas alterations in function of these cells (qualitativedisorder) in the peripheral circulation or in tissuesare the basis for BVDV-induced immunosuppres-sion. Although both qualitative and quantitativewhite blood cell disorders may be observed in acuteinfections, only qualitative disorders are observed inPI cattle (Brown et al., 1991; Muscoplat et al., 1973;Roth et al., 1986).

A transient leukopenia occurs in most cattleacutely infected with BVDV. The very first descrip-tion of BVDV by Olafson et al. (1946) demon-strated leukopenia as a finding in both naturally andexperimentally infected cattle. White blood cellconcentrations decrease to various degrees duringBVDV infection. White blood cell count forhealthy cattle is 4,000–12,000 leukocytes/µl with amean of 8,000 leukocytes/µl (Kramer, 2000), andthe leukocyte concentration is higher in neonatalcalves (Tennant et al., 1974). By definition,leukopenia is a decrease in white blood cell con-centration below the normal level. Early reportshave identified mild reductions in white blood cellconcentration in calves with minimal evidence ofclinical disease. There seems to be no differencebetween cytopathic and noncytopathic biotypes asfar as production of leukopenia is concerned. Afterintravenous challenge with a cytopathic BVDV, themean white blood cell concentration decreasedfrom 7,850 leukocytes/µl to 5,050 leukocytes/µl at4 days after infection (Bolin et al., 1985). In an ex-perimental infection with a noncytopathic isolate ofBVDV, a mild leukopenia was observed on days 3,5, and 7 after infection (Ellis et al., 1988) withleukocyte concentrations of <5,000 leukocytes/µl.The typical time frame for depression in whiteblood cell concentrations is between 3 and 12 daysafter infection (Bolin and Ridpath, 1992; Ellis etal., 1998; Kelling et al., 2002a, 2002b; Walz et al.,1999).

Differences in the degree of leukopenia are asso-ciated with the virulence of the virus. Although se-vere acute disease has been associated only withtype 2 BVDV, genotype is not a determinant for vir-ulence. Though some type 2 strains cause severeacute disease, infection with other BVDV 2 strainsresults in subclinical to mild disease under naturaland experimental conditions (Marshall et al., 1996;Walz et al., 2001; Liebler-Tenorio et al., 2003).Kelling et al. (2002b) evaluated five different iso-lates of BVDV 2 and found that although all five in-duced leukopenia, the highly virulent isolates in-duced a significantly more severe lymphopenia thanthe less virulent isolates. Further study revealed thatvirulence correlated with depression in lymphocytecounts. In an experimental infection, a highly viru-lent strain of BVDV 2–induced depression in lym-phocyte counts of greater than 80% from preinocu-lation levels; a less virulent BVDV strain induceddepressions of less than 50% (Liebler-Tenorio et al.,2003).

Not only are absolute counts affected by virus vir-ulence, but duration of leukocyte depressions alsocorrelates with virulence (Liebler-Tenorio et al.,2003; Walz et al., 2001). In a study involving twotype 2 isolates and one type 1 isolate, the more vir-ulent type 2 isolate induced a more severe leukope-nia of a longer duration than the less virulent type 2isolate and the type 1 isolate (Walz et al., 2001).Differences have also been reported in the type ofaffected leukocytes following experimental chal-lenge. For example, neutropenia was observed as themajor hematologic abnormality with some BVDVisolates (Walz et al., 2001; Hamers et al., 2000;Archambault et al., 2000), and lymphopenia was ob-served with some others (Bolin et al., 1985; Kellinget al., 2002b). In yet other studies, reduction in bothlymphocytes and neutrophils has been documented(Ellis et al., 1988, 1998).

Unlike acutely infected animals, the PI calveshave normal numbers of leukocytes and lympho-cytes (Bolin et al., 1985; Larsson et al., 1988) butthe proportion of lymphocyte subpopulation mightchange. For example, PI calves have an increasedproportion of B-cells and diminished numbers oflymphocytes not identified as B-cells or T-cells (nullcells) (Larsson et al., 1988). The mechanism ofBVDV-induced leukopenia is currently unknown.Several possibilities exist, including immune systemremoval of BVDV-infected immune cells, destruc-tion of immune cells by BVDV, and increased traf-ficking of immune cells into tissue sites of viralreplication.



BVDV-INDUCED IMMUNE ORGANDYSFUNCTION

Specific changes in immune system function occurwith acute BVDV infection. As described above, alltypes of immune cells are infected by BVDV andtheir functions affected.

Effect of BVDV on bone marrow

Viral antigen is present in megakaryocytes andmyeloid cells of BVDV-infected cattle (Spagnuoloet al., 1997). Bone marrow–derived macrophagescan be infected in vitro with BVDV, and after infec-tion, they release type I interferon (Adler et al.,1997). It has been suggested that type I interferonmight prime the oral cavity and gastrointestinal tractto decrease NO production and to induce apoptosisin response to LPS (Alder et al., 1997). This bio-chemical pathway may be the basis of mucosal dis-ease caused by BVDV. In acute BVDV infection, thedevelopment of lesions in lymphoid tissue is a func-tion of BVDV replication and host reaction to infec-tion (Liebler-Tenorio et al., 2002).

Effect of BVDV on thymus

Thymus plays an important role in the maturation ofCD4+ and CD8+ lymphocytes. Thymus is a centrallymphoid organ and all lymphocytes that enter thy-mus get selected based on molecules expressed ontheir surface. Lymphocytes that strongly recognizeself-antigens are negatively selected in the thymus.Only 1–2% of lymphocytes that enter the thymusmature. The remaining lymphocytes undergo clonaldeletion during the selection process. It is from theseselected cells that BVDV-specific T-cells are gener-ated. Infection with BVDV significantly decreasesthymocyte function (Marshall et al., 1994). Like alllymphoid tissues, there are lesions (such as deple-tion of lymphocytes) but viral antigen is not detectedexcept in vascular walls only (Liebler-Tenorio et al.,2002).

Effect of BVDV on Peyer’s patches

BVDV significantly alters T-cells in Peyer’s patches.After BVDV infection, there is significant depletionof lymphocytes in Peyer’s patches, and cattle in ter-minal stages of mucosal disease have extensive lossof lymphocytes in the gut-associated lymphoid tis-sues (Figure 9.1). The tips of the domes are depletedof lymphocytes and the epithelia of the follicles areinvaginated. The number of CD4+ cells decreases inthe follicles and the number of lymphocytes in inter-follicular areas is also reduced (Liebler et al., 1995).B-lymphocytes are also depleted in lymphoid folli-

cles leading to decreased size of the follicles(Liebler et al., 1995).

Effect of BVDV on spleen

Small arterioles in the spleen are surrounded by pe-riarteriolar lymphoid sheaths that contain CD4+ andCD8+ T-cells. Masses of B-cells called lymphoidfollicles are attached to T-cell sheaths. Both cyto-pathic and non cytopathic biotypes of BVDV infectspleen cells and the virus has been isolated fromspleen of cattle that die from BVDV (McClurkin etal., 1985 and Bolin et al., 1987). In the initial phaseof BVDV infection, virus is present but no lesionsoccur. However, lesions develop later in infectionand antigen disappears in about 2 weeks.

IMMUNOLOGICALLYPRIVILEGED SITESIn immunologically privileged sites, the immune re-sponse is inaccessible or suppressed. In these or-gans, the immune response is contained to preventinflammation that could be destructive to the virus-infected tissues.

TESTES

BVDV infection of breeding bulls leads to a de-crease in semen quality, although the virus may ormay not be detectable in the seminal ejaculate (Frayet al., 2000). In one study, BVDV was isolated fromraw, unprocessed, and diluted extended semen(Kirkland et al., 1991) with titers ranging from 5–75TCID50/ml. Testes can be a site of persistent BVDVinfection in non-viremic bulls (Niskanen et al.,2002; Givens et al., 2002). In calves infected withBVDV, seminiferous tubules are lined with sertolicells and intratubular giant cells (Binkhorst et al.,1983). Studies done in other species (such as hu-mans) indicate that testicular and epidymal lympho-cytes express T-cell markers (Yakirevich et al.,2002). Studies to investigate immune responses toBVDV in testes of cattle are sorely needed.

OVARIES

Cattle persistently infected with BVDV contain viralRNA and antigen in ovarian tissues (Booth et al.,1995). In cattle acutely infected with ncp BVDV, thelevels of gonadotrophins and sex strands are modu-lated (Fray et al., 2002). Bovine follicular cells andoocytes are permissive to BVDV infection, and acuteinfection decreases estradiol secretion (Fray et al.,2000). The replication of BVDV in ovaries causesovarian dysfunction and reduced fertility (Shin andAclaud, 2001).



CNSCalves born to BVD-infected cows develop CNSdisorders at various times after birth, including nys-tagmus, eye disorders, gait disorders, and tremors.Congenital malformations occur during the 3rd to5th months of gestation. Cerebellar hypoplasia is themost common lesion and is also associated with hy-drocephalus and hydraencephaly. BVDV infectionis accompanied by lens opacity and retinal atrophy.Blindness in calves is associated with no blinkingreflex. BVDV-infected animals develop demyelina-tion of the spinal cord, which is consistently presentin the white matter (Hewicker-Trautwein et al.,1995).

SUMMARY AND CONCLUSIONSBovine viral diarrhea virus continues to be a majorproblem for the cattle industry worldwide. The virusis intimately associated with immunological dis-arrangement because it infects the immune cells ofthe body, including the antigen-presenting cells/

macrophages, and B- and T-cells. The virus has animpressive ability to evade immune recognition andelimination, which is best exemplified by its abilityto infect the developing fetus, induce immunotoler-ance and persistent infection, and thus perpetuate it-self in the herd for future generations of calves. Inaddition, BVDV may persist in immunologicallyprivileged sites, such as ovaries and testicles, andthus is present in the germ plasm. Antigenic and ge-netic variations are also hallmarks of BVDV due tothe extreme plasticity of the virus, and this variationaffects immune system recognition of the virus.

Vaccines developed for BVDV have existed forseveral decades and the use of BVDV vaccines isfairly common. In spite of their widespread usage,BVDV vaccines appear to offer incomplete fetalprotection due to the inability to achieve total pro-tection against viremia. Understanding the methodsby which BVDV interacts with the bovine immunesystem should provide insight into the pathogenesisof the virus, which may ultimately lead to the devel-opment of more effective vaccines. Thus, the key


Figure 9.1. Left. Histologic section of ileum from an uninfected calf. Right. Histologic section of ileum from a calf12 days after infection with BVDV 890 (BVDV type 2). Note the depletion of lymphoid follicles.


areas of future research with respect to BVDV andthe immune system are

• Understanding the roles of different componentsof the immune system in protection as they re-late to vaccine development and evolution ofBVDV genetic and antigenic diversity

• Determining the mechanisms of BVDV-inducedimmunosuppression and immunotolerance


phages infected with cytopathic bovine viral diar-rhea virus release a factor(s) capable of priming un-infected macrophages for activation-inducedapoptosis. J Virol 71:3255–3258.

Adler H, Frech B, Meier P, et al.: 1994,Noncytopathic strains of bovine viral diarrhea virusprime bovine bone marrow-derived macrophages.Biochem Biophy Res Commun 202:1562–1568.

Adler H, Jungi TW, Pfister H, et al.: 1996, Cytokineregulation by virus infection: Bovine viral diarrheavirus, a flavivirus, downregulates production oftumor necrosis factor alpha in macrophages invitro. J Virol 70:2650–2653.

Archambault D, Beliveau C, Couture Y, Carman S:2000, Clinical response and immunomodulationfollowing experimental challenge of calves withtype 2 noncytopathogenic bovine viral diarrheavirus. Vet Res 31:215–227.

Atluru D, Xue W, Polam S, et al.: 1990, In vitrointeractions of cytokines and bovine viral diarrheavirus in phytohemagglutinin-stimulated bovinemononuclear cells. Vet Immunol Immunopathol25:47–59.


Bielefeldt Ohmann H: 1987, Double-immunolabelingsystems for phenotyping of immune cells harboringbovine viral diarrhea virus. J Histochem Cytochem35:627–33.

Bielefeldt Ohmann H, Ronsholt L, Bloch B: 1987,Demonstration of bovine viral diarrhea virus in pe-ripheral blood mononuclear cells of persistently in-fected, clinically normal cattle. J Gen Virol68:1971–1982.

Biuk-Rudan N, Cvetnic S, Madic J, Rudan D: 1999.Prevalence of antibodies to IBR and BVD virusesin dairy cows with reproductive disorders.Theriogenology 51:875–881.

Bolin SR, McClurkin AW, Coria MF: 1985, Effects ofbovine viral diarrhea virus on the percentages andabsolute numbers of circulating B and T lympho-cytes in cattle. Am J Vet Res 46:884–886.

Bolin S, Moennig V, Kelso-Gourley NE, Ridpath JF:1988, Monoclonal antibodies with neutralizing ac-tivity segregate isolates of bovine viral diarrheavirus into groups. Arch Virol 99:117–123.

Bolin SR, Ridpath JF: 1990, Range of neutralizing ac-tivity and molecular specificity of antibodies in-duced in cattle by inactivated bovine viral diarrheavirus vaccines. Am J Vet Res 51:703–707.


Bolin S, Ridpath JF: 1996. The clinical significance ofgenetic variation among bovine viral diarrheaviruses. Vet Med 10:958–961.

Bolin SR, Roth JA, Uhlenhopp EK, Pohlenz JF: 1987,Immunologic and virologic findings in a bull chron-ically infected with noncytopathic bovine viral diar-rhea virus. J Am Vet Med Assoc 190:1015–1017.

Bolin SR, Saliki JT: 1996, Serology interpretation:Bovine viral diarrhea virus (BVDV). Am Assoc VetLab Diagn, http://www.aavld.org/MainMenu2/InterSerology/bovine_viral_diarrhea.html

Booth PJ, Stevens DA, Collins ME, Brownlie J: 1995,Detection of bovine viral diarrhea virus antigen andRNA in oviduct and granulosa cells of persistentlyinfected cattle. J Reprod Fertil 105:17–24.

Boulanger D, Waxweiler S, Karelle L, et al.: 1991,Characterization of monoclonal antibodies tobovine viral diarrhea virus: Evidence of a neutraliz-ing activity against gp48 in the presence of goatanti-mouse immunoglobulin serum. J Gen Virol72:1195–1198.

Brodersen BW, Kelling CL: 1998, Effect of concur-rent experimentally induced bovine respiratory syn-cytial virus and bovine viral diarrhea virus infectionon respiratory tract and enteric diseases in calves.Am J Vet Res 59:1423–1430.

Brodersen BW, Kelling CL: 1999, Alteration of leuko-cyte populations in calves concurrently infectedwith bovine respiratory syncytial virus and bovineviral diarrhea virus. Viral Immunol 12:323–334.

Brown GB, Bolin SR, Frank DE, Roth JA: 1991,Defective function of leukocytes from cattle persist-ently infected with bovine viral diarrhea virus, andthe influence of recombinant cytokines. Am J VetRes 52:381–387.

Brownlie J, Hooper LB, Thompson I, Collins ME:1998, Maternal recognition of foetal infection withbovine virus diarrhoea virus (BVDV)-the bovinePestivirus. Clin Diag Virol 10:141–150.

Bruschke CJ, Haghparast A, Hoek A, et al.: 1998b,The immune response of cattle, persistently in-fected with noncytopathic BVDV, after superinfec-tion with antigenically semi-homologous cytopathicBVDV. Vet Immunol Immunopathol 62:37–50.



Bruschke CJ, Weerdmeester K, van Oirschot JT, vanRijn PA: 1998a Distribution of bovine virusdiarrhoea virus in tissues and white blood cells ofcattle during acute infection. Vet Microbiol64:23–32.

Castrucci G, Frigeri F, Osburn BI, et al.: 1990, Astudy of some pathogenetic aspects of bovine viraldiarrhea virus infection. Comp Immunol MicrobiolInfect Dis 13:41–49.

Charleston B, Brackenbury LS, Carr BV, et al: 2002,Alpha/beta and gamma interferons are induced byinfection with noncytopathic bovine viral diarrheavirus in vivo. J Virol 76:923–927.

Charleston B, Fray MD, Baigent S, et al.: 2001,Establishment of persistent infection with non-cytopathic bovine viral diarrhea virus in cattle isassociated with a failure to induce type I interferon.J Gen Virol 82:1893–1897.

Charleston B, Hope JC, Carr BV, Howard CJ:2001,Masking of two in vitro immunological assays forMycobacterium bovis (BCG) in calves acutely in-fected with non-cytopathic bovine viral diarrhoeavirus. Vet Rec 149:481–484.

Chase CC, Elmowalid G, Yousif AA: 2004, The im-mune response to bovine viral diarrhea virus: Aconstantly changing picture. Vet Clin North AmFood Anim Pract 20:95–114.

Collen T, Carr V, Parsons K, et al: 2002, Analysis ofthe repertoire of cattle CD4(+) T cells reactive withbovine viral diarrhoea. Vet Immunol Immunopathol87:235–238.

Collen T, Douglas AJ, Paton DJ, et al: 2000, Singleamino acid differences are sufficient for CD4(+) T-cell recognition of a heterologous virus by cattlepersistently infected with bovine viral diarrheavirus. Virology 276:70–82.

Collen T, Morrison WI: 2000, CD4(+) T-cell re-sponses to bovine viral diarrhea virus in cattle.Virus Res 67:67–80.

Cortese VS, West KH, Hassard LE, et al.:1998,Clinical and immunologic responses of vaccinatedand unvaccinated calves to infection with a virulenttype-II isolate of bovine viral diarrhea virus. J AmVet Med Assoc 213:1312–1319.

de Smit AJ, de Smit AJ, Eble PL, de Kluijver EP,Bloemraad M, Bouma A: 1999, Laboratory deci-sion-making during the classical swine fever epi-demic of 1997–1998 in the Netherlands. Prev VetMed 42:185–199.

Dinter Z, Bakos K: 1961, Viruses associated withacute respiratory and enteric disease in Swedishcattle. Bull Off Internat Epizoot 56:29–34.

Donis RO, Dubovi EJ: 1987, Glycoproteins of bovineviral diarrhea-mucosal disease virus in infectedbovine cells. J Gen Virol 68:1607–1616.

Ellis JA: 2001b, The immunology of the bovine respi-ratory disease complex. Vet Clin North Am FoodAnim Pract 17:535–550.

Ellis JA, Davis WC, Belden EL, Pratt DL: 1988, Flowcytofluorimetric analysis of lymphocyte subset al-terations in cattle infected with bovine viral diar-rhea virus. Vet Pathol 25:231–236.

Ellis JA, West K, Cortese V, et al.: 2001a, Effect ofmaternal antibodies on induction and persistence ofvaccine-induced immune responses against bovineviral diarrhea virus type II in young calves. J AmVet Med Assoc 219:3351–3356.

Ellis JA, West KH, Cortese VS, Myers SL et al.: 1998,Lesions and distribution of viral antigen followingan experimental infection of young seronegativecalves with virulent bovine virus diarrhea virus-type II. Can J Vet Res 62:161–169.

Endsley JK, Quade MJ, Terhaar B, Roth JA: 2002,Bovine viral diarrhea virus type 1-and type 2-specific bovine T lymphocyte-subset responsesfollowing modified live virus vaccination. Vet Ther3:364–372.

Endsley JJ, Roth JA, Ridpath J, Neill J:2003,Maternal antibody blocks humoral but not T cell re-sponses to BVDV. Biologicals 31:123–125.

Fairbanks K, Schnackel J, Chase CC: 2003,Evaluation of a modified live virus type-1a bovineviral diarrhea virus vaccine (Singer strain) against atype-2 (Strain 890) challenge. Vet Ther 4:24–34.

Fort Dodge Animal Health: 1999, Challenge studyone year post vaccination with a Canadian type IIstrain. Technical Information, September 16, 1999,Fort Dodge Animal Health, Overland Park, KS.

Fray MD, Mann GE, Bleach EC, et al.: 2002,Modulation of sex hormone secretion in cows byacute infections with bovine viral diarrhea virus.Reproduction 123:281–289.

Fray MD, Paton DJ, Alenius S: 2000, The effects ofbovine viral diarrhea virus on cattle reproduction inrelation to disease control. Anim Reprod Sci60–61:615–627.

Fulton RW, Purdy CW, Confer AW et al.: 2000,Bovine viral diarrhea viral infections in feedercalves with respiratory disease: Interactions withPasteurella spp., parainfluenza-3 virus, and bovinerespiratory syncytial virus. Can J Vet Res64:151–159.

Fulton RW, Ridpath JF, Confer AW et al.: 2003,Bovine viral diarrhea virus antigenic diversity:Impact on disease and vaccination programmes.Biologicals 31:89–95.

Fulton RW, Ridpath JF, Saliki JT et al.: 2002, Bovineviral diarrhea virus (BVDV) 1b: PredominantBVDV subtype in calves with respiratory disease.Can J Vet Res 66:181–190.



Fulton RW, Saliki JT, Burge LJ, et al.: 1997, Neutra-lizing antibodies to type 1 and 2 bovine viral diar-rhea viruses: Detection by inhibition of viral cy-topathology and infectivity by immunoperoxidaseassay. Clin Diag Lab Immunol 4:380–383.

Givens MD, Riddell KP, Galik PK, et al.: 2002,Diagnostic dilemma encountered when detectingbovine viral diarrhea virus in IVF embryo produc-tion. Theriogenology 58:1399–1407.

Glew EJ, Carr BV, Brackenbury LS, et al.: 2003,Differential effects of bovine viral diarrhoea viruson monocytes and dendritic cells. J Gen Virol84:1771–1780.

Glew EV, Howard CV: 2001, Antigen-presenting cellsfrom calves persistently infected with bovine viraldiarrhea virus, a member of the Flaviviridae, arenot compromised in their ability to present viralantigen. J Gen Virol 82:1677–1685.

Graham CH, Lysiak JJ, McCrae KR, LaLa PK: 1992,Localization of transforming growth factor-beta atthe human fetal maternal interface: Role in tro-phoblast growth and differentiation. Biol Reprod46:561–572.

Greig A, Gibson IR, Nettleton PF, Herring JA: 1981Disease outbreak in calves caused by a mixedinfection with infectious bovine rhinotracheitisvirus and bovine virus diarrhea virus. Vet Rec108:480.

Haines DM, Martin KM, Clark EG et al.: 2001, Theimmunohistochemical detection of Mycoplasmabovis and bovine viral diarrhea virus in tissues offeedlot cattle with chronic, unresponsive respiratorydisease and/or arthritis. Can Vet J 42:857–860.

Hamers C, Couvreur B, Dehan P, et al.: 2000,Differences in experimental virulence of bovineviral diarrhoea viral strains isolated from haemor-rhagic syndromes. Vet J 160:250–258.

Hamers C, Lambot M, Onclin M, et al.: 1998,Antigenic diversity of bovine viral diarrhea viralisolates contradicts the concept of herd specificstrain. Epidemiol Infect 121:451–457.

Harpin S, Hurley DJ, Mbikay M, et al.: 1999, Vacci-nation of cattle with a DNA plasmid encoding thebovine viral diarrhea virus major glycoprotein E2. JGen Virol 80:3137–3144.

Hegde N, Srikumaran S: 1997, The use of bovineMHC class I allele-specific peptide motifs and pro-teolytic cleavage specificities for the prediction ofpotential cytotoxic T lymphocyte epitopes ofbovine viral diarrhea virus. Virus Genes14:111–121.

Hewicker-Trautwein M, Liess B, and Trautwein G:1995, Brain lesions in calves following transplacen-tal infection with bovine-viral diarrhea virus.Zentralbl Vet Med B 42:65–77.

Houe H: 1995, Epidemiology of bovine viral diarrheavirus. Vet Clin North Am Food Anim Pract11:521–547.

Howard CJ, Howard CJ, Clarke MC, Sopp P,Brownlie J: 1992, Immunity to bovine virus diar-rhea virus in calves: The role of different T-cellsubpopulations analyzed by specific depletion invivo with monoclonal antibodies. Vet ImmunolImmunopathol 32:303–314.

Jensen J, Schultz, RD: 1991, Effect of infection bybovine viral diarrhea virus (BVDV) in vitro on in-terleukin-1 activity of bovine monocytes. VetImmunol Immunopathol 29:251–265.

Jewett CJ, Kelling CL, Frey ML, Doster AR: 1990,Comparative pathogenicity of selected bovine viraldiarrhea virus isolates in gnotobiotic lambs. Am JVet Res 51:1640–1644.

Johnson DW, Muscoplat CC: 1973, Immunologic ab-normalities in calves with chronic bovine viral diar-rhea. Am J Vet Res 34:1139–1141.

Jones L, Van Campen H, Xu Z-C, Schnackel JA:2001, Comparison of neutralizing antibody titers totype 1a, 1b and 2 bovine viral diarrhea virus fromexperimentally infected and vaccinated cattle.Bovine Practitioner 35:137–140.

Kelling CL, Steffen DJ, Cooper VL, et al.: 2002a,Effect of infection with bovine viral diarrhea virusalone, bovine rotavirus alone, or concurrent infec-tion with both on enteric disease in gnotobioticneonatal calves. Am J Vet Res 63:1179–1186.

Kelling CL, Steffen DJ, Topliff CL, et al.: 2002b,Comparative virulence of isolates of bovine viraldiarrhea virus type II in experimentally inoculatedsix- to nine-month-old calves. Am J Vet Res63:1379–1384.

Ketelsen AT, Johnson DW, Muscoplat CC: 1979,Depression of bovine monocyte chemotactic re-sponses by bovine viral diarrhea virus. InfectImmun 25:225–230.

Kirkland PD, Richards SG, Rothwell JT, et al.: 1991,Replication of bovine viral diarrhea virus in thebovine reproductive tract and excretion of virus insemen during acute and chronic infections. Vet Rec128:587–590.

Knoetig SM, Summerfield A, Spagnuolo-Weaver M,McCullough KC: 1999, Immunopathogenesis ofclassical swine fever: Role of monocytic cells.Immunology 97:359–366.

Kramer, J: 2000, Normal hematology of cattle, sheep,and goats. In: Schalm’s Veterinary Hematology.Eds. Feldman B, Zinkl J, Jain N, pp. 1075–1084.Lippincott, Philadelphia, PA.

Lambot M, Douart A, Joris E, et al.: 1997,Characterization of the immune response of cattleagainst non-cytopathic and cytopathic biotypes of



bovine viral diarrhoea virus. J Gen Virol78:1041–1047.

Lambot M, Hanon E, Lecomte C, et al.: 1998, Bovineviral diarrhoea virus induces apoptosis in bloodmononuclear cells by a mechanism largely depend-ent on monocytes. J Gen Virol 79:1745–1749.

Lambot M, Joris E, Douart A, et al.: 1998, Evidencefor biotype-specific effects of bovine viral diar-rhoea virus on biological reponses in acutely in-fected calves. J Gen Virol 79:27–30.

Lamontagne L, Lafortune P, Fournel M: 1989,Modulation of the cellular immune responses to T-cell–dependent and T-cell–independent antigens inlambs with induced bovine viral diarrhea virus in-fection. Am J Vet Res 50:1604–1608.

Larsson B, Fossum C, Alenius S, et al.: 1988, A cellu-lar analysis of immunosuppression in cattle withmucosal disease. Res Vet Sci 44:71–75.

Liebler EM, Kusters C, Pohlenz JF: 1995,Experimental mucosal disease in cattle: Changes oflymphocyte subpopulations in Peyer’s patches andin lymphoid nodules of large intestine. Vet ImmunolImmunopathol 48:233–248.

Liebler-Tenorio EM, Ridpath JF: 2002, Distribution ofviral antigen and development of lesions after ex-perimental infection with highly virulent bovineviral diarrhea virus type 2 in calves. Am J Vet Res63:1575–1584.

Liebler-Tenorio EM, Ridpath JF, Neill JD, et al.:2002, Distribution of viral antigen and developmentof lesions after experimental infection with highlyvirulent bovine viral diarrhea virus type 2 in calves.Am J Vet Res 63:1575–1584.

Liebler-Tenorio EM, Ridpath JF, Neill JD: 2003,Lesions and tissue distribution of viral antigen insevere acute versus subclinical acute infection withBVDV2. Biologicals 31:119–122.

Liu L, Lehmkuhl HD, Kaeberle ML: 1999,Synergistic effects of bovine respiratory syncytialvirus and noncytopathogenic bovine viral diarrheavirus infection on selected bovine alveolarmacrophage functions. Can J Vet Res 63:41–48.

Makoschey B, Janssen MG, Vrijenhoek MP, et al.:2001, An inactivated bovine virus diarrhea virus(BVDV) type 1 vaccine affords clinical protectionagainst BVDV type 2. Vaccine 19:3261–3268.

Marshall DJ, Moxley RA, Kelling CL, et al.: 1996,Distribution of virus and viral antigen in specificpathogen-free calves following inoculation withnoncytopathic bovine viral diarrhea virus. VetPathol 33:311–318.

Marshall DJ, Perry GA, Kuszynski CA, et al.: 1994,Flow cytometric analyses of lymphocyte subsets inperipheral blood and lymphoid tissues of gnotobi-otic calves during primary acute postnatal infec-

tions of bovine viral diarrhea virus. Viral Immunol7:141–149.

Martin SW, Bateman KG, Shewen PE, et al.: 1990, Agroup level analysis of the associations between an-tibodies to seven putative pathogens and respiratorydisease and weight gain in Ontario feedlot calves.Can J Vet Res 54:337–342.

McClurkin AW, Bolin SR, Coria MF: 1985, Isolationof cytopathic and noncytopathic bovine viral diar-rhea virus from the spleen of cattle acutely andchronically affected with bovine viral diarrhea. JAm Vet Med Assoc 186:568–569.

Muscoplat CC, Johnson DW, Teuscher E, et al.: 1973,Surface immunoglobulin of circulating lymphocytesin chronic bovine diarrhea: Abnormalities in cellpopulations and cell function. Am J Vet Res34:1101–1104.

Niskanen R, Alenius S, Belak K, et al.: 2002,Insemination of susceptible heifers with semenfrom a non-viraemic bull with persistent bovinevirus diarrhoea virus infection localized in thetestes. Reprod Domest Anim 37:171–175.

Nobiron I, Thompson I, Brownlie J, Collins ME:2001, Cytokine adjuvancy of BVDV DNA vaccineenhances both humoral and cellular immune re-sponses in mice. Vaccine 19:4226–4235.

Olafson P, MacCallum AD, Fox FH: 1946, An appar-ently new transmissible disease of cattle. CornellVet 36:205–213.

Olchowy TWJ, Slauson DO, Boschsler PN: 1997,Induction of procoagulant activity in virus infectedbovine alveolar macrophages and the effect of lipo-polysaccharide. Vet Immunol Immunopath 58:27–37.

Pastoret PP, Lambot M: 1995, Biology of bovine pes-tivirus infection responsible for mucosal disease.Bull Mem Acad R Med Belg 150:244–250.

Pellerin C, van den Hurk J, Lecomte J, Tussen P:1994, Identification of a new group of bovine viraldiarrhea virus strains associated with severe out-breaks and high mortalities. Virology 203:260–268.

Peterhans E, Jungi TW, Schweizer M: 2002, BVDvirus and innate immunity. Biologicals 31:107–112.

Potgieter LND: 1995, Immunology of Bovine ViralDiarrhea Virus. Vet Clin North Am Food Anim Pract11:501–520.

Potgieter LND: 1997, Bovine respiratory tract diseasecaused by bovine viral diarrhea virus. Vet ClinNorth Am Food Anim Pract 13:471–481.

Potgieter LN, McCracken MD, Hopkins FM, Guy JS:1985, Comparison of the pneumopathogenicity oftwo strains of bovine viral diarrhea virus. Am J VetRes 46:151–153.

Potgieter LN, McCracken MD, Hopkins FM, WalkerRD: 1984, Effect of bovine viral diarrhea virus in-fection on the distribution of infectious bovine



rhinotracheitis virus in calves. Am J Vet Res45:687–690.

Reggiardo C, Kaeberle ML: 1981, Detection of bac-teremia in cattle inoculated with bovine viral diar-rhea virus. Am J Vet Res 42:218–221.

Rhodes SG, Cocksedge JM, Collins RA, MorrisonWI: 1999, Differential cytokine responses of CD4+and CD8+ T-cells in response to bovine viral diar-rhea virus in cattle. J Gen Virol 80:1673–1679.

Richer L, Marois P, et al.: 1988, Association of bovineviral diarrhea virus with multiple viral infections inbovine respiratory disease outbreaks. Can Vet J29:713–717.


Ridpath JF, Neill JD, Endsley J, Roth JA; 2003, Effectof passive immunity on the development of a pro-tective immune response against bovine viral diar-rhea in calves. Am J Vet Res 64:65–69.

Roth JA, Bolin SR, Frank DE: 1986, Lymphocyteblastogenesis and neutrophil function in cattle per-sistently infected with bovine viral diarrhea virus.Am J Vet Res 47:1139–1141.

Roth JA, Kaeberle ML, Griffith RW: 1981, Effects ofbovine viral diarrhea virus infection on bovinepolymorphonuclear leukocyte function. Am J VetRes 42:244–250.

Sandvik T, Larsen IL, Nyberg O: 2001, Influence ofmilk from cows persistently infected with BVDVon bulk milk antibody levels. Vet Rec 148:82–84.

Schelp C, Greiser-Wilke I, Wolf G, et al.: 1995,Identification of cell membrane proteins linked tosusceptibility to bovine viral diarrhoea virus infec-tion. Arch Virol 140:1997–2009.

Sentsui H, Takami R, Nishimori T, et al.: 1998, Anti-viral effect of interferon-� on bovine viral diarrheavirus. J Vet Med Sci 60:1329–1333.

Shahriar FM, Clark EG, Janzen E, et al.: 2002,Coinfection with bovine viral diarrhea virus andMycoplasma bovis in feedlot cattle with chronicpneumonia. Can Vet J 43:863–868.

Shin T, Aclaud H 2001: Tissue distribution of bovineviral diarrhea virus antigens in persistently infectedcattle. J Vet Sci 2:81–84.

Spagnuolo M, Kennedy S, Foster JC, et al: 1997,Bovine viral diarrhea virus infection in bone mar-row of experimentally infected calves. J CompPathol 116:97–100.

Sopp P, Hooper LB, Clarke MC, et al.: 1994,Detection of bovine viral diarrhoea virus p80 pro-tein in subpopulations of bovine leukocytes. J GenVirol 75:1189–1194.

Stoffregen B, Bolin SR, Ridpath JF, Pohlenz J: 2000,Morphologic lesions in type 2 BVDV infections

experimentally induced by strain BVDV2-1373 re-covered from a field case. Vet Microbiol77:157–162.

Swasdipan S, Bielefeldt-Ohmann H, Phillips N, et al.:2001, Rapid transplacental infection with bovinepestivirus following intranasal inoculation of ewesin early pregnancy. Vet Pathol 38:275–280.

Tennant B, Harrold D, Reina-Guerra M, et al.: 1974,Hematology of the neonatal calf: Erythrocyte andleukocyte values of normal calves. Cornell Vet64:516–532.

Thoen CO, Waite KJ: 1990, Some immune responsesin cattle exposed to Mycobacterium paratuberculo-sis after injection with modified-live bovine viraldiarrhea virus vaccine. J Vet Diagn Invest2:176–179.

Van Oirschot JT, Bruschke CJM, Van Rijn PA: 1999,Vaccination of cattle against bovine viral diarrhoea.Vet Microbiol 64:169–183.

Van Reeth K, Adair B: 1997, Macrophages and respi-ratory viruses. Pathol Biol 45:184–192.

Vaughn M: 1997, SN titer response comparisonsamong fourteen diagnostic laboratories for antibod-ies protecting against IBR, BVD and BRSV virusfollowing the administration of modified live vac-cine to weaning aged calves. Bovine Proceedings30:141–143.

Waldvogel AS, Hediger-Weithaler BM, Eicher R, etal.: 2000, Interferon-gamma and interleukin-4mRNA expression by peripheral blood mononu-clear cells from pregnant and non-pregnant cattleseropositive for bovine viral diarrhea virus. VetImmunol Immunopathol 77:201–212.

Walz PH, Bell TG, Steficek BA, et al.: 1999,Experimental model of type II bovine viral diarrheavirus-induced thrombocytopenia in neonatal calves.J Vet Diagn Invest 11:505–514.

Walz PH, Bell TG, Wells JL, Grooms DL: 2001,Relationship between degree of viremia and diseasemanifestation in calves with experimentally in-duced bovine viral diarrhea virus infection. Am JVet Res 62:1095–1103.

Welsh, MD, Adair BM:1995, Prostaglandin E2 pro-duction by bovine virus diarrhoea virus infectedcell culture and monocytes. In: Immunobiology ofviral infections: Proc. 3rd Congress of theEuropean Society of Veterinary Virology, Interlaken,Switzerland, pp. 63–67.

Welsh MD, Adair BM, Foster JC: 1995, Effect ofBVDV infection on alveolar macrophage function.Vet Immunol Immunopathol 46:195–210.




Woldehiwet Z, Hussain AA: 1994, Border diseasevirus antigens in lymphocyte subpopulations in theperipheral flood of persistently infected sheep. VetImmunol Immunopathol 42:127–135.

Woods RD, Kunkle RA, Ridpath JF, Bolin SR: 1999,Bovine viral diarrhea virus isolated from fetal calfserum enhances pathogenicity of attenuated trans-missible gastroenteritis virus in neonatal pigs. J VetDiag Invest 11:400–407.


Yakirevich, Yanai O, Sova Y, et al.: 2002, Cytotoxicphenotype of intraepithelial lymphocytes in normaland cryptorchid human testicular excurrent ducts.Human Reprod 275–283.



INTRODUCTIONPestiviruses of domestic animals include bovineviral diarrhea virus (BVDV) of cattle, border diseasevirus (BDV) of sheep, and classic swine fever virus(CSFV) of pigs. The latter virus is also known ashog cholera virus (HCV). These viruses were ini-tially found to be associated only with their speciesof origin but are now recognized to infect numerousother animal species indicating they are not species-specific. This chapter focuses on BVDV and hostsother than cattle in which infection results in dis-ease, virus isolation, or seroconversion. Addition-ally, cell types supporting the growth of BVDV invitro are also discussed.

Concern exists over non-bovine hosts of BVDVfor a number of reasons. In many agricultural situa-tions, cattle and other wild and domestic ruminantsshare the same pasture, range, and water source.Transmission of virus from non-bovine hosts to cat-tle may affect control programs and/or productivityin cattle herds. Similarly, spread of BVDV from cat-tle to sheep, goats, and wild ruminants may affectthe productivity of domestic animals or the health ofwildlife populations. In addition, infection ofspecies such as swine with BVDV may interferewith HCV control programs through false positivetest results. In Europe, formal monitoring of wildlifefor domestic animal pathogens (such as pestivi-ruses) is carried out in some countries (Frolich et al.,2002). This type of surveillance would be neededshould control programs for pestiviruses be consid-ered in North America.

An additional concern worldwide is the possibleBVDV contamination of biologics (particularlymodified live virus vaccines) used in non-bovinehosts. Biologic manufacturers may use BVDV-contaminated fetal bovine serum in vaccine produc-tion as a result of inadequate testing or improper

quality control standards. Pandemics may start whena pathogen “jumps” from one species to another.The parvovirus jump from feline to canine hostsmay be an example of such a pandemic occurringdue to contaminated biologics.

DISEASE SYNDROMES IN NON-BOVINE HOSTS DUE TONATURAL BVDV INFECTIONReports of disease resulting from natural infectionwith BVDV in species other than cattle are infre-quent. BVDV was initially isolated from litters ofdying piglets as well as from stillborn pigs in 1988(Terpstra and Wensvoort, 1988). In a subsequent re-port, these authors reported a series of 19 naturallyoccurring and 8 experimentally induced incidencesof congenital BVDV infections in pigs (Terpstra andWensvoort, 1991). The disease was restricted to theinvolved litter without any horizontal transmission.The signs resembled those of CSF with pigs showingsigns of infection at birth and dying by 4 months ofage. Postpartum infections, in contrast, usually re-sulted in subclinical disease. Rarely, congenitally in-fected pigs may survive and become lifelong carriersof the virus, shedding the virus in urine, semen, andpharyngeal fluid (Terpstra and Wensvoort, 1997).

Outbreaks of a disease resembling border diseasein sheep have been reported as a result of BVDVbeing spread from persistently infected calves(Carlsson, 1991). BVDV has also been isolated fromcongenitally infected kids and lambs showing entericsymptoms resulting in death in the first week of life(Nettleton et al., 1980). BVDV was also detected in astillborn alpaca by reverse transcription-polymerasechain reaction (RT-PCR) (Goyal et al., 2002). Thevirus was subsequently isolated and typed as BVDVtype 1b. BVDV may have been the cause of death forthis alpaca although no gross or histologic lesions

171

10Hosts

Trevor R. Ames


were found. Mucosal disease-like lesions have beendescribed in a free-ranging white-tail deer fromwhich a cytopathic BVDV was isolated (Ludwig andMcClurkin, 1981). BVDV type 1a was isolated froma free-ranging female mule deer in Wyoming (VanCampen et al., 2001). The deer was emaciated, weak,salivating, and had lung abscesses from whichArcanobacter pyogenes was cultured.

DISEASE SYNDROMES IN NON-BOVINE HOSTS DUE TOEXPERIMENTAL BVDVINFECTIONBVDV has also been shown to produce disease in arange of domestic animals following experimentalchallenge. Oral-intranasal infection with fieldstrains of BVDV in pregnant gilts produced in-trauterine infection in 1 of 20 gilts, with an addi-tional 3 gilts having reduced fetal numbers com-pared to corpora lutea and 2 gilts being barren(Stewart et al., 1980). Intranasal and subcutaneousadministration of a vaccine strain of BVDV (OregonC24V) in pregnant sows in the second trimester ofgestation produced asymptomatic seroconversionduring second and third trimesters of pregnancy(Kulscar et al., 2001). Some of the piglets born atterm had clinically apparent disease consistent withthat caused by CSFV, indicating transplacental in-fection. Fifty-seven percent of the piglets died by 60days of age, but neither the sows nor their progenyshed the virus. Similarly, a BVDV or BDV that waspresent as a contaminant in a CSF vaccine inducedcongenital pestivirus infection in piglets born oneight swine farms utilizing the vaccine (Wensvoortand Terpstra, 1988).

Experimental, intranasal challenge of 2-month-old pigs with various doses of type 1 and type 2BVDV was performed to evaluate the pathogenicityof the two BVDV isolates in a swine model (Paul etal., 1999). Neither virus induced clinical signs al-though both were reisolated antemortem and post-mortem. Type 2 BVDV isolate that was experimen-tally more pathogenic in calves failed to producedisease in pigs. indicating the virulence was species-specific. These reports support the arguments thatcongenital infection of pigs with BVDV may resultin disease but postnatal infection rarely producesclinical signs (Terpstra and Wensvoort, 1988;Terpstra and Wensvoort, 1991).

Transplacental infection of ovine fetuses withBVDV has been studied extensively in sheep.BVDV can cross the placenta and produce fetal le-sions within 10 days postinfection (Hewicker-Traut-

wein, 1994). Type 2 BVDV has also been shown toproduce congenital lesions in sheep fetuses (Schereret al., 2001). BVDV infection of pregnant ewes, likeBDV infection, results in abortions, stillbirths, unvi-able lambs, and viable lambs that are both virus-pos-itive and virus-negative (Scherer et al., 2001).Because of the similarity of BVDV to BDV, and thebehavior of the BVDV in the pregnant ewe, experi-mental challenge of vaccinated ewes with BVDVhas been used to evaluate protective immunity, in-cluding fetal protection, afforded by various BVDVvaccines (Brushke et al., 1996).

BVDV can produce pulmonary lesions in 6–8-month-old lambs following intranasal and intra-bronchial challenge (Meefhan et al., 1998). New-born lambs and kids challenged with BVDV isolatesmost consistently have CNS lesions (Jewett et al.,1990; Loken et al., 1990). Deer and elk have alsobeen experimentally challenged with BVDV to de-termine whether these wild ruminants are capable ofbecoming infected and shedding the virus (Tessaroet al., 1999; van Campen et al., 1997). Both deer andelk can be infected, shed the virus, and seroconvertto the virus (Terpstra and Wensvoort, 1997; vanCampen and Williams, 1996). In addition, lateraltransmission to nonchallenged in-contact animalshas been demonstrated with elk (Tessaro et al.,1999). Experimental infection of pregnant does witha deer isolate of BVDV resulted in mummified fe-tuses, stillborn fawns, and live fawns (Ludwig andMcClurkin, 1981).

VIRUS ISOLATION ANDSEROCONVERSION IN NON-BOVINE HOSTSAs stated in the previous section, BVDV has beenisolated from sheep, goats, and pigs undergoing nat-ural infections (Carlsson, 1991; Pratelli et al., 1999;Terpstra and Wensvoort, 1988). BVDV has alsobeen isolated from German roe deer (Tessaro et al.,1999), Scottish deer (Fischer et al., 1998), and ex-otic captive ruminants (Nattleton, 1990). Isolation ofpestiviruses from a variety of ruminants other thancattle indicates that many species not yet studiedmay harbor them (Doyle and Heuschele, 1983). Re-cently, persistent infection has been documented ina family of captive mousedeer indicating that thisphenomenon also occurs in cervids (Grondahl et al.,2003). This finding has significant implications foreradication campaigns.

Serological surveys also provide useful informa-tion on which species may become infected withBVDV and the seroprevalence within those species.



A survey of 1,133 dairy cows, 3,712 ewes, and1,317 adult pigs in Norway found that 18.5% of cat-tle, 4.5% of sheep, and 2.2% of pigs were seroposi-tive for BVDV (Loken et al., 1991). Of the swinethat tested positive, all had higher antibody titersagainst BVDV than against CSFV indicating thatexposure was likely to BVDV rather than to CSFV.No such comparisons were made between BVDVantibody titers in the seropositive sheep, so it is un-clear whether these titers are a true reflection ofBVDV seroprevalence. In a study of ruminants inNamibia, neutralizing antibodies to BVDV werefound in 58% of cattle, 14% of sheep, and 4.6% ofgoats (Depner et al., 1991). Again no attempt wasmade to differentiate the pestivirus species causingthese high serologic titers. In a similar study inNorthern Ireland, 5.3% of ewes and 30.4% of flockswere found positive to BDV (Graham et al., 2001).Upon further testing it was found that all sheep hadhigher titers to BVDV type 1 than to BDV. In addi-tion, the only pig found seropositive of 680 testedfor HCV actually had higher titers to BVDV type 1than to CSFV. These studies point out the impor-tance of determining which pestivirus a group of an-imals is infected with or is responding to serologi-cally, especially when control programs are beingconsidered. Serologic evidence also exists for pes-tivirus infection of goats (Nattleton, 1990).

Serologic surveys of approximately 50 captiveand free living ruminant species distributed withinthe families of Camelidae, Cervidae, Anteloca-pridae, and Bovidae have shown presence of pes-tivirus antibodies (Loken, 1995; Nattleton et al.,1980). Seventeen different species of African wild-life have tested positive for antibodies to BVDV(Hambli and Hedger, 1979). In North America, 2%of pronghorn antelope and 31% of American Bisonhave antibodies to BVDV (Stauber et al., 1980;Taylor et al., 1997). Seropositivity in caribou herdsmay range from 41–100% ((Dieterich, 1987; Elaz-hary et al., 1981; Stuen et al., 1993; van Campen andWilliams, 1996) but with moose the seroprevalenceranges from 12-18% (Kocan et al., 1986; Thorsenand Henderson, 1971). This suggests that wild rumi-nants that are part of large herds, such as caribouand bison herds, may have higher seroprevalencethan more solitary wild ruminants such as the mooseor those that travel in smaller groups such as the an-telope (van Campen and Williams, 1996).

Two percent of llamas in Argentina have antibod-ies to BVDV indicating that imported llamas mayalso harbor this virus and may transmit to cattleherds if in close proximity (Puntel et al., 1999).

However, the seroprevalence of BVDV antibodies indeer has not been shown to be affected by cattlepopulation densities in deer habitats (Frolich, 1995).Sequence analysis of BVDV isolated from roe deerfound the isolate to be genetically unique comparedto other BVDV isolates (Frolich and Hofmann,1995). It is likely that BVDV-strains circulate inde-pendently within deer populations, with transmis-sion rarely occurring between deer and domesticlivestock (Fischer et al., 1998). Serologic evidencefor BVDV infection in rabbits has been shown with40% of free ranging rabbits in Germany having an-tibody titers to BVDV in one report (Frolich andStreich, 1998). None of the rabbit spleens examinedin this study were positive for BVDV.

Some evidence exists for BVDV infection in hu-mans (Giangaspero et al., 1993). The seroprevalenceof BVDV antibodies in one study of human subjectsin Zambia ranged from 14.75% in healthy non–HIV-infected humans to 19.2% in HIV-infected individu-als with chronic diarrhea. The increased seropreva-lence in this latter group was statistically significantleading the authors to speculate that some interac-tion between HIV and BVDV may exist. There are anumber of concerns, however, that arise from thisreport:

• No other reports have documented seroconver-sion to BVDV in humans who work with BVDVor BVDV-infected animals.

• BVDV has not been successfully grown in pri-mate cells, raising doubts about the potential forinfection and seroconversion in primates.

• Antibody titers against other pestiviruses werenot determined, making it difficult to concludewhether the antibodies detected were specificallyagainst BVDV

• The researchers failed to provide evidence thatno reagents used in their investigation were con-taminated with BVDV or BVDV antibodies.

CELL LINES SUPPORTING THEGROWTH OF BVDV AS POSSIBLEINDICATORS OF HOST RANGEIn a large survey of 41 common cell lines, cells ofcattle, sheep, goat, deer, bison, rabbit, domestic cat,and swine origin were found capable of supportingBVDV growth. A subpopulation of these cell linesincluding those of cattle, sheep, goat, deer, bison,rabbit, and domestic cat origin were found to becontaminated by BVDV using immunohistochemi-cal and polymerase chain reaction procedures (Bolinet al., 1994). This study not only points out the abil-

Hosts 173


ity of BVDV to grow in a variety of cell lines butalso the potential for it to contaminate cell lines in-advertently. Interestingly, although the virus is capa-ble of infecting many cell lines, cell tropisms doexist, and the viral glycoprotein E2 has been shownto play a role in this cell tropism (Liang et al., 2003).As previously discussed, researchers and diagnosti-cians must control for possible cell line contamina-tion with BVDV or the results of investigative stud-ies will be compromised.

CONCLUSIONSThe reports of naturally occurring disease indicatethat pigs and a variety of domestic and wild rumi-nants can be infected with BVDV. The virus hasbeen shown to produce congenital infection in pigs,sheep, and goats. In some cases, congenital infectionresults in persistently infected animals that can be asignificant source of viral transmission. Wild rumi-nants are known to be susceptible to acute infectionswith BVDV, and limited evidence suggests thatcervids may also undergo congenital infectionsleading to persistent infection (Grondahl et al.,2003). This suggests a new potential for this virus tobe shed in large amounts from wild ruminants thatcould then infect cattle. Current evidence suggeststhat wild ruminants may serve as a transient sourceof virus while undergoing acute infections and maypossibly be a more prolonged source of virus frompersistently infected animals. This would be ofgreatest concern where these animals are in contactfor prolonged periods of time, as opposed to tran-sient fence line contact.

The role of species from which the virus has neverbeen isolated but in which seroconversion has beenobserved is not clear. The significance of in vitrogrowth of BVDV in certain cell lines without infec-tion in whole animals is also not known. It is un-likely, however, that these species would be viremicfor significant periods of time and thus are unlikelyto shed BVDV in the environment. Based on theprevious discussion, it seems that non-bovine do-mestic ruminants and swine remain the greater con-cern for disrupting management attempts to controlthe diseases caused by BVDV in cattle.

REFERENCESBolin SR, Ridpath JF, Black J, et al.: 1994, Survey of

cell lines in the American Type Culture Collectionfor bovine viral diarrhea virus. J Virol Methods48:211–221.

Brushke CJ, von Rijn PA, Moormann RJ, vanOerschot JT: 1996, Antigentically different pes-

tivirus strains induce congenital infection in sheep:a model for bovine virus diarrhea virus vaccine effi-cacy studies. Vet Microbiol 50:33–43.

Carlsson M: 1991, Border disease in sheep caused bytransmission of virus from cattle persistently in-fected with bovine virus diarrhea virus. Vet Rec128: 145–147.

Depner K, Hubschle OJ, Liess B: 1991, Prevalence ofruminant pestivirus infection in Namibia. Onder-stepoort J Vet Res, 58:107109.

Dieterich RA: 1981, Respiratory viruses. In: AlaskanWildlife Diseases. Ed. Deiterich RA. University ofAlaskan Press, Fairbanks, Alaska, pp.28–29.

Doyle LF, Heushchele WP: 1983, Bovine viral diar-rhea virus infection in captive exotic ruminants. JAm Vet Med Assoc 183:1257–1259.

Elazhary MA, Frechette JL, Silim A, Roy RS: 1981,Serologic evidence of some bovine viruses in thecaribou (Rangifer tarandus caribou) in Quebec. JWildl Dis 17:609–612.

Fischer S, Weiland E, Frolich K: 1998, Characteri-zation of a bovine viral diarrhea virus isolated from roe deer in Germany. J Wildl Dis 34:47–555.

Frolich K: 1995, Bovine viral diarrhea and mucosaldisease in free–ranging and captive deer (Cervidae)in Germany. J Wildl Dis 31:247–250.

Frolich K, Hofmann M: 1995, Isolation of bovineviral diarrhea virus-like particles from roe deer(Capreolus capreolus). J Wildl Dis 31:243–246.

Frolich K, Streich WJ: 1998, Serologic evidence ofbovine viral diarrhea virus in free-ranging rabbitsfrom Germany. J Wildl Dis 34:173–178.

Frolich K, Thiede S, Kozikowski T, et al: 2002, A re-view of mutual transmission of important infectiousdiseases between livestock and wildlife in Europe.Ann New York Acad Sci 969:4–13.

Giangaspero M, Vacirca G, Morgan D, et al: 1993,Anti-bovine viral diarrhoea virus antibodies in adultZambian patients infected with the human immun-odeficiency virus. Internat J Sexually TransmittedDis Acquired Immune Deficiency Syndrome4:300–302.

Goyal SM, Bouljihad M, Haugerud S, Ridpath JF:2002, Isolation of bovine viral diarrhea virus froman alpaca. J Vet Diagn Invest 14:523–525

Graham DA, Calvert V, German A, McCullough SJ:2001, Pestivirus infections in sheep and pigs inNorthern Ireland. Vet Rec 148:69–72.

Grondahl C, Uttenthal A, Houe H, et al: 2003, Char-acterization of pestivirus isolated from a persist-ently infected mousedeer (Tragulus javanicus).Arch Virol 148:1455–1463.

Hambli C, Hedger RS: 1979, The prevalence of anti-bodies to bovine viral diarrhea/mucosal disease



virus in African Wildlife. Comp Immunol MicrobiolInfect Dis 2:295–303.

Hewicker-Trautwein M, Trautwein G: 1994, Pore-cephaly, hydroncephaly, and leukoencephalopathyin ovine fetus following transplacental infectionwith bovine virus diarrhea virus: Distribution ofviral antigen and characterization of cellular re-sponses. Acta Neuropathol (Berl) 87:385–397.

Jewett CJ, Kelling CL, Frey ML, Doster AR: 1990,Comparative pathogenicity of selected bovine viraldiarrhea virus isolates in gnotobiotic lamb. Am J VetRes 51: 1640–1744.

Kocan AA, Franzmann AW, Waldrup KA, Kubat GJ:1986, Serologic studies of select infectious diseaseof moose (Alces alces L.) from Alaska. J Wildl Dis22:418–420.

Kulscar G, Soos P, Kucsera L, et al: 2001, Patho-genicity of a bovine viral diarrhea virus strain inpregnant sows: Short communication. Acta VetHung 49:117–1200.

Liang D, Sainz FS, Ansari IH, et al: 2003, The enve-lope glycoprotein E2 is a determinant of cell cul-ture tropism in ruminant pestiviruses. J Gen Vir84:1269–1274.

Loken T: 1995, Ruminant pestivirus infection in ani-mals other than cattle and sheep. Vet Clin NorthAm: Food Anim Pract 11:597–614.

Loken T, Bjerkas I, Larsen HJ: 1990, Experimentalpestivirus infection in newborn goat kids. J CompPathol 103:277–288.

Loken T, Krogrud J, Larsen IL: 1991, Pestivirus infec-tions in Norway, serologic investigation in cattle,sheep, and pigs. Acta Vet Scand 32:27–34.

Ludwig J, McClurkin A: 1981, BVD in a Minnesotawhite-tailed deer. Wildl Dis Assoc Mtg Abstracts p. 38.

Meefhan JT, Lehmkuhl HD, Cutlip RC, Bolin SR:1998, Acute pulmonary lesions in sheep experimen-tally infected with bovine viral diarrhea virus. JComp Pathol 119:277–292.

Nettleton PF: 1990, Pestivirus infection in ruminantsother than cattle. Rev Sci Tech 9:131–150.

Nettleton PF, Herring JA, Corigall W: 1980, Isolationof bovine virus diarrhea virus from a Scottish reddeer. Vet Rec 107:425.

Pratelli A, Ballo E, Martella V, et al: 1999, Pestivirusinfection in small ruminants: virological and his-pathological findings. New Microbiol 22:351–356.

Puntel M, Fondvila NA, Blanco VJ, et al: 1999,Serologic survey of viral antibodies in llamas(Lama glama) in Argentina. Zentrabl VeterinarmedB 46:157–161.

Scherer CF, Flores, Weiblen R, et al: 2001, Experi-mental infection of pregnant ewes with bovine viral

diarrhea virus type-2 (BVDV-2): Effects on thepregnancy and fetus. Vet Microbiol 79:285–299.

Stauber EH, Autenreith R, Markham OD, Whitveck V:1980, A seroepidemiologic survey of three prong-horn (Antelocapra Americana) populations insoutheast Idaho, 1975–1977. J Wildl Dis16:109–115.

Stewart WC, Miller LD, Kresse JI, Synder ML: 1980,Bovine viral diarrhea infection in pregnant swine.Am J Vet Res 41:459–462.

Stuen S, Krogsrud J, Hyllseth B, Tyler NJ: 1993,Serosurvey of three virus infections in reindeer innorthern Norway and Svalbard. Rangifer13:215–219.

Taylor SK, Lane VM, Hunter DL, et al: 1997,Serologic survey for infectious pathogens in free-ranging American bison. J Wildl Dis 33:308–311.

Terpstra C, Wensvoort G: 1988, Natural infection ofpigs with bovine viral diarrhea virus associatedwith signs resembling swine fever. Res Vet Sci45:137–142.

Terpstra C, Wensvoort G: 1991, Bovine virus diarrheavirus infections in swine. Tijdschr Dergeneeskd116:943–948.

Terpstra C, Wensvoort G: 1997, A congenital infectionof bovine virus diarrhea virus in pigs: Clinical, vi-rological, and immunological observations. Vet Q19:97–101.

Tessaro SV, Carman PS, Deregt D: 1999, Viremia andvirus shedding in elk with type1 and virulent typebovine viral diarrhea virus. J Wildl Dis 35:671–677.

Thorsen J, Henderson JP: 1971, Survey for antibodyto infectious bovine rhinotracheitis (IBR), bovineviral diarrhea (BVD) and parainfluenza 3 (PI-3) inmoose sera. J Wildl Dis 7:93–95.

van Campen H, Ridpath J, Williams E, et al: 2001,Isolation of bovine viral diarrhea virus from a free-ranging mule deer in Wyoming. J Wildl Dis37:306–331.

van Campen H, Williams ES: 1996, Wildlife andBovine Virus Diarrhea Virus: InternationalSymposium Bovine Viral Diarrhea Virus—A 50 YearReview. Cornell University Press, Ithaca, NY, pp167–175.

van Campen H, Williams ES, Edwards J, et al: 1997,Experimental infection of deer with bovine viral di-arrhea virus. J Wildl Dis 33:567–573.

Walz PH, Baker JC, Mullaney TP, et al: 1999,Comparison of type I and type II bovine diarrheavirus infection in swine. Can J Vet Res 63:119–123.

Wensvoort G, Terpstra C: 1988, Bovine viral diarrheavirus infection in piglets born to sows vaccinatedagainst swine fever with contaminated vaccine. ResVet Sci 45:143–148.

Hosts 175


INTRODUCTIONBovine viral diarrhea virus (BVDV) is a ubiquitouspathogen of ruminants that is often associated withsevere economic losses. Understanding theseviruses, particularly at the cellular and molecular lev-els, is important to develop new vaccination andtreatment strategies. The events that transpire follow-ing infection of the host animal are only now begin-ning to become clear. Upon entry into a susceptiblehost cell, the virus replicates utilizing viral as well ascellular proteins and machinery. Our understandingof BVDV replication comes from studies of BVDVdirectly or is extrapolated from studies of classicalswine fever virus (CSFV, known previously as hogcholera virus) and other members of the Flaviviridae,particularly hepatitis C virus (HCV). Most proteinsencoded by BVDV, CSFV, and HCV, as well as RNAstructural motifs, are considered functionally equiva-lent (Rice, 1996; Branza-Nichita et al., 2001). Thetwo exceptions are the Npro and Erns proteins that areunique to pestiviruses. This chapter describes func-tions provided by the host cell that are absolutely re-quired for virus replication, including protein trans-lation, protein modification, and transport andrelease of progeny virus. Results from studies ofother pestiviruses and flaviviruses are discussed incontext of BVDV and will be considered common toall viruses unless stated otherwise.

RECEPTOR AND VIRUSATTACHMENTArguably, the single most important event in the lifecycle of a virus, the “make or break point,” is the at-tachment of the infectious virus particle to the sur-face of a susceptible cell. If this does not occur,downstream events such as uptake, release ofgenome, replication of viral RNA(s) and translationof viral proteins cannot take place. Binding of the

virus to the cell surface has been shown in a numberof systems to involve the recognition of specific cel-lular molecules that are embedded in, or intimatelyassociated with, the cell’s plasma membrane. Thesespecific cellular molecules are usually proteins, al-though many examples of viruses binding to carbo-hydrate moieties are known (Jackson et al., 1996;Hilgard and Stockert, 2000; Martínez and Melero,2000; Dechecchi et al., 2001; Terry-Allison et al.,2001). Virus receptors are normal components of theplasma membrane and serve diverse functions suchas internalization of ligands, cell signaling, and cell-to-cell interactions (Mendelsohn et al., 1989; Wykeset al., 1993; Colston and Racaniello, 1994; Roi-vainen et al., 1994; Agnello et al., 1999). The typesand nature of these receptor molecules affects boththe host and tissue tropism of the virus. If the recep-tor molecule of an organism is sufficiently differentfrom that of the normal host, infection will notoccur. Additionally, the receptor molecule(s) may beexpressed in only a subset of cell types in the host,only at specific points in the cells cycle, or on spe-cific surfaces of polarized cells (Compans, 1995).Any of these factors would limit the infection to spe-cific cells or tissues. This in turn gives rise to thecharacteristic disease symptoms, lesions, andpathology of the viral infection.

CELLULAR FACTORS

Several groups have studied the interaction ofBVDV and the cell surface. Initial studies involvedthe production of monoclonal antibodies (Mab) di-rected against cell surface epitopes. Teyssedou et al.(1987) reported the production of a Mab that was re-active against a protein on the surface of Madin-Darby bovine kidney (MDBK) cells. When at-tached, this Mab completely prevented infection bybovine enterovirus-3 (BEV-3) strain 240A and par-

177

11Interactions of Virus and Host

John D. Neill


tially inhibited infection by BEV-2 and BVDV. Themechanism behind this incomplete inhibition wasunclear and probably did not represent an antibodyreactive against the true receptor molecule. This wasfollowed shortly by a report by Moennig et al.(1988) of a bovine-specific Mab that prevents infec-tion of bovine cells with a number of cytopathicstrains of BVDV (cp BVDV) while not preventinginfection with CSFV, bovine herpesvirus-1, parain-fluenza-3, and pseudorabies virus. Based on thesefindings, the authors suggested a single receptormolecule for BVDV. Further analysis of the protec-tion afforded by this Mab, using the more sensitiveimmunoperoxidase method (Collett et al., 1989), re-vealed small numbers of infected foci in Mab-treated MDBK cultures, indicating the presence ofmultiple cell surface receptors for BVDV.

Another type of Mab, anti-idiotypic antibodies,were produced to study specific virus-receptor inter-actions (Xue and Minocha, 1993; Xue and Minocha,1996, Xue et al., 1997). Anti-idiotypic Mabs wereraised against neutralizing Mabs that bound the viralprotein anti-gp53 (E2). These anti-idiotypic Mabsthus mimicked the viral epitopes presumed to be as-sociated with binding of the virus to the cellular re-ceptor. The anti-idiotypic Mabs bound to the surfaceof MDBK cells in a manner similar to that of thevirus and inhibited infection by BVDV. A 50 kDaprotein was identified in immunoprecipitation ex-periments as the target of these Mabs (Xue andMinocha, 1993; Minocha et al., 1997). Xue et al.(1997) determined that this receptor molecule wasprotein in nature because protease treatment of cellsresulted in concurrent loss of virus binding to thecell surface. Glycosidase treatment of cells prior toinfection was used to demonstrate that glycosylationof the receptor protein was not necessary for bindingof virus. The receptor molecule was found on anumber of different cell types, both susceptible andnonsusceptible to infection with BVDV (Xue andMinocha, 1993; Xue and Minocha, 1996), indicat-ing a protein of conserved function. The receptorprotein was recognized by BVDV on the surface ofporcine cells, based on the resulting productive in-fection. However, the replication of the virus wasslower, with progeny virus levels eventually reach-ing those produced by bovine cells. The differencewas probably not a result of cell surface binding anduptake but rather in functions related to replication.

Three additional Mabs (Moennig et al., 1988) thatbound to the surface of cells susceptible to infectionwith BVDV were characterized by Schelp et al.(1995). These Mabs bound to the surface of cultured

cells as well as leukocytes freshly isolated from theblood of cattle and blocked infection with BVDV tovarying degrees. When used in immunoprecipitationexperiments, all precipitated cellular proteins of 60and 93 kDa. In follow-up experiments (Schelp et al.,2000), one of these Mabs, BVD/CA 26, recognized aprotein consisting of 28 and 56 kDa subunits that,under nonreducing conditions, formed multimers ofapproximately 200 kDa. The 56 kDa subunit wasshown to bind to F-actin, giving some insight into itspossible biological function. It is unknown why therewas a discrepancy in protein sizes between the tworeports. It is also unknown if the 56 kDa protein de-scribed here and the 50 kDa protein described by Xueand Minocha (1993) are the same or related proteins.

A different approach examining BVDV bindingand uptake was taken by Flores and Donis (1995).They generated an MDBK cell line that was resist-ant to infection by BVDV. This mutant cell line,CRIB, was developed by infection of a susceptiblemonolayer of cells with a cytopathic strain ofBVDV (Singer strain) and culturing of the survivors.The resulting culture contained no virus (infectiousor defective) as determined by cocultivation, animalinoculation, immunofluoresence, western and north-ern blots, and reverse-transcription polymerasechain reaction (RT-PCR). Infection of CRIB cellswith 24 additional strains of BVDV did not result ina productive infection as determined by expressionof viral proteins. Transfection of viral RNA or viri-ons in the presence of polyethylene glycol (PEG)did result in a productive infection as measured byimmunofluoresence, indicating CRIB cells were de-fective at virus entry and not at a postuptake func-tion. However, CRIB cells were not resistant to in-fection by 10 other bovine viruses, indicating thatthe block of BVDV replication was specific and wasnot a general antiviral activity.

In additional studies, Flores et al. (1996) reportedthat resistance of CRIB cells to infection withBVDV was blocked at virus entry and suggestedthat a cell membrane function that was important invirus uptake following viral attachment was mutatedor missing. Using PEG-mediated virus uptake inthese cells, the authors concluded that while CRIBcells bound saturating levels of virus, entry wasblocked by a defect in endocytosis. This was furtherdemonstrated by blockage of PEG-mediated uptakeof virus using inhibitors of endocytosis and endoso-mal acidification.

Characterization of the cell surface moleculebound by the E2 envelope provided evidence thatthe low density lipoprotein receptor (LDLR) was the



E2 receptor molecule as well as the mediator of en-docytosis of the attached virus particle protein(Agnello et al., 1999). The LDLR is a cell surfaceendocytic receptor that mediates the uptake of extra-cellular ligands into the cell (May et al., 2003).Three lines of evidence support the LDLR as aBVDV receptor: complete inhibition of endocytosisby anti-LDLR antibodies, inhibition of endocytosisby phenylarsine oxide (an inhibitor of endocytosis),and inhibition of uptake by chemical methods thatprevent lipoprotein/LDLR interactions. In addition,CRIB cells (Flores and Donis, 1995; Flores et al.,1996) lack LDLR based on failure to bind anti-LDLR antibodies. The loss of infection by inhibitionof endocytosis with anti-LDLR antibody indicatesthat LDLR-mediated endocytosis may be the mainmechanism of virus entry (Agnello et al., 1999).However, the observation that there was a low back-ground of infection associated with high levels of in-fecting virus indicated the presence of other, low-affinity receptor molecules.

The possibility of the CD46 molecule as a cellsurface receptor for BVDV has been proposed(Rümenapf et al., 2000; Maurer et al., 2002). TheCD46 protein is a known receptor for measles virusand herpesvirus 6 (Greenstone et al., 2002). Anti-bodies against the CD46 molecule, a complementregulatory protein, inhibited infection with BVDV.Expression of the bovine CD46 protein in porcinecells increased plaquing efficiency of cytopathicBVDV forty- to hundredfold. Expression of CD46in nonpermissive cells did not confer susceptibility.

VIRAL FACTORS

The outer envelope of the BVDV virion contains thestructural proteins Erns, E1, and E2. These proteinsare highly glycosylated and possess the major anti-genic determinants of the virus. For a more com-plete discussion concerning the biology and struc-ture of these proteins, refer to Chapter 3. Recentwork has demonstrated that two of these proteins,Erns and E2, play important roles in the attachmentof the virus particle to the cell surface. Hulst andMoormann (1997), working with purified CSFVErns and E2 synthesized in insect cells, showed thatErns added to susceptible cells prior to infection (100µg/ml) could irreversibly bind to cells and preventinfection. Purified E2 at 10 µg/ml also provided100% inhibition of infection but this inhibition wasreversible and required additional supplementationof E2 to maintain inhibition. Following removal ofE2, infection still occurred, presumably by virusparticles bound to the surface of the cells. Treatment

with 100 µg/ml Erns released these particles. The dif-ference between concentration of proteins requiredto reach complete inhibition and differences in bind-ing properties indicated that these proteins bound todifferent molecules on the cell surface. These differ-ences also indicate that the E2 binding site is oflower prevalence.

Experiments have demonstrated that dengue virus(Hilgard and Stockert, 2000) and BVDV (Iqbal etal., 2000) Erns molecules first bind to cell surface he-parin sulfate proteoglycans and sulfated heparin-likeglycosaminoglycans, respectively. This type of ini-tial virus/cell interaction has been reported for anumber of viruses that utilize two or more cell sur-face receptors (Mettenleiter et al., 1990; Okazaki etal., 1991; Jackson et al., 1996; Chen et al., 1997;Jusa et al., 1997; Krusat and Streckert, 1997; Asagoeet al., 1997). Erns binds to both permissive and non-permissive cells because of the near ubiquitous pres-ence of glycosaminoglycans (Iqbal et al., 2000).Soluble Erns added to medium blocked replication ofBVDV. As supporting evidence, Erns binding wasnot observed when cells were treated with hepari-nase I or III, when soluble glycosaminoglycans werepresent, or to Chinese hamster ovary (CHO) cellsthat did not produce glycosaminoglycans or heparinsulfate. These lines or evidence led to the hypotho-sis that binding of Erns to cell surface glycosamino-glycans is the initial event in viral attachment.Subsequently, Iqbal and McCauley (2002) showedthat the conserved KKLENKSK motif near the C-terminus of the Erns protein mediated binding of theglycosaminoglycan molecule.

The mechanism of release of genomic RNA intothe cytoplasm of the cell is unclear but probably in-volves acidification of endocytic vesicles (Floresand Donis, 1995; Flores et al., 1996). Uptake ofvirus appears to occur by endocytosis and not by amembrane fusion mechanism. Treatment of BVDV-infected MDBK cells with phenylarsine oxide, aninhibitor of endocytosis and inhibitors of endosomeacidification (chloroquine and ammonium chloride)prevented uptake of virus in the presence of PEG(Flores and Donis, 1995). Dengue virus, a memberof the Flaviviridae, has been shown to traffic themajor clathrin-dependent endocytic pathway duringinfection (Hilgard and Stockert, 2000). Work withWest Nile virus demonstrated that acidification(<pH 6.6) caused rapid loss of viral infectivity(Gollins and Porterfield, 1986). Ammonium chlo-ride also inhibited uncoating of virus and infection,demonstrating the dependence on acid pH in the in-fection process.

Interactions of Virus and Host 179


Taken together, these results suggest that attach-ment and entry of BVDV into a cell is a multistepprocess. First, the virus attaches to the cell surfacethrough interaction of Erns envelope protein and adocking glycosaminoglycan receptor molecule. Thisstep can occur in both susceptible and nonsuscepti-ble cells because these molecules are present onmost cells. It appears that binding to the docking re-ceptor is insufficient by itself to bring about uptakeof the virus (Flores and Donis, 1995; Flores et al.,1996). Entry into the cell is further mediated by at-tachment of the E2 envelope protein to the LDLRreceptor and internalization via endocytosis. This isprobably the rate-limiting step due to the relativelylow abundance of the LDLR on most cell types(Agnello et al., 1999). As was stated earlier, thepresence of other, minor receptor molecules cannotbe ruled out at this time.

INTERACTIONS WITHCELLULAR FACTORS DURINGREPLICATIONUpon entry into the cell, the genomic RNA must actas mRNA, directing the translation of viral proteins.Viral proteins participate in the necessary functionsfor RNA replication, protein processing (proteasecleavages), and protein trafficking, but are insuffi-cient by themselves to bring about all of theseevents. Recruitment of cellular factors is necessaryfor the successful completion of the replicationprocess. Many of these factors are known, but theidentity and function of others remain elusive.

Probably the most important of the cellular func-tions utilized by BVDV for replication is proteinsynthesis. This process is much too complex to beencoded by a simple virus. Thus, BVDV must uti-lize the existing translational factors employed bythe cell. In noninfected cells, eukaryotic cap-dependent translation initiation begins by the bind-ing of eIF4E (the cap binding protein), as part of theheterotrimeric eIF4F complex (composed of eIF4A,eIF4E, and eIF4G), to the 5’ m7G cap of the mRNAmolecule (Figure 11.1A). The 40S small ribosomesubunit, associated with the ternary complex eIF2/GTP/Met-tRNAi, eIF2 and eIF3, forms the 43Spreinitiation complex. The binding of the 43S sub-unit to the mRNA is directed by the eIF4F complexand results in the formation of the 48S complex. TheATP-dependent helicase activity of the eIF4A sub-unit is thought to unwind the RNA, allowing thebinding of the 43S complex to the mRNA.

Both eIF1 and eIF1a are necessary for proper as-sembly of the 48S preinitiation complex as well as

making the 48S preinitiation complex competent forscanning to the initiation codon. The scanningprocess to the initiation codon also requires eIF4Band eIF4F. At the initiation codon, eIF5, through in-teraction with both eIF2 and eIF3, causes the hy-drolysis of the GTP moiety of eIF2, resulting in the


Figure 11.1. Translation initiation of cellular transcriptsas compared to that of the pestivirus genomic RNA.A. Eukaryotic translation initiation takes place by bind-ing of the cap structure (M7G) as directed by the eIF4Fcomplex. The attachment of the 40S small ribosomalsubunit and other factors necessary to form the 48Spreinitiation complex are illustrated. The scanning com-petent complex moves down the transcript until en-countering the initiation codon, at which point eIF5 andeIF5B direct formation of the 80S ribosomal complexand initiation of translation. B. Formation of the 48Spreinitiation complex at the BVDV IRES requires onlythe 40S small ribosomal subunit, eIF3 and eIF2/GTP/Met-tRNAi. This results in the assembly of the transla-tion-competent 80S ribosomal complex at the initiatorAUG and translation of the BVDV polyprotein.


release of eIF2-GDP. Initiation factor eIF5B, againwith hydrolysis of GTP, directs the attachment of the60S ribosomal subunit in the formation of the trans-lationally competent 80S ribosome. For a more de-tailed discussion of the eukaryotic protein transla-tion initiation process, see Pestova et al. (1998) andPestova et al. (2001).

POLYPROTEIN TRANSLATION

Following the uptake and the release of the viralgenome, the next critical event in the replicationcycle is the production of viral proteins, both thosenecessary for replication and those that interact withhost factors to generate a cellular environment opti-mized for production of virus progeny. The BVDVgenomic RNA functions as mRNA for the immedi-ate synthesis of a large viral polyprotein that is en-coded in the single open reading frame (ORF) of thevirus. The ORF encodes a large protein of approxi-mately 4000 amino acids that is co- and posttransla-tionally cleaved to produce the individual, matureproteins. The proteins, in order of placement withinthe genome from N- to C-terminus are NH2-Npro-capsid-Erns-E1-E2-p7-NS2/3-NS4a-NS4b-NS5a-NS5b-COOH. The envelope glycoproteins—Erns,E1, and E2—are translocated into the lumen of theendoplasmic reticulum for further posttranslationalmodification (Rümenapf et al., 1993). The E1 andE2 are inserted into the membrane of the endoplas-mic reticulum during translocation while Erns re-mains soluble.

A major difference between the BVDV genomicRNA and most cellular mRNA molecules is thepresence of a long 5’ nontranslated region (5’ UTR)containing a high degree of secondary structure(Figure 11.2) as well as multiple upstream AUGcodons. This roughly 390 base region, termed the in-ternal ribosome entry site (IRES), functions to directthe attachment and assembly of the ribosome at theinitiation codon of the ORF, insuring that initiationof translation begins at the proper AUG codon.

Computational analysis of the 5’ untranslated re-gion of pestivirus genomes revealed the presence ofhigher order folding and secondary structure(Brown et al., 1992; Deng and Brock, 1993; Le etal., 1995). The presence of multiple AUG codonswas noted, several with stronger matches to theKozak consensus sequence (Kozak, 1987) than theauthentic initiation codon (Deng and Brock, 1993).These analyses resulted in the development of amodel of secondary structure that revealed a seriesof stem-loops that were highly conserved struc-turally between BVDV and CSFV. These tertiarystructures were conserved despite the sequence di-vergence between the two viruses. Brown et al.(1992) used double- and single-strand specificRNases to confirm the structure of the stem-loopstructures and further found them to be highly simi-lar to that described for the IRES of HCV(Tsukiyama-Kohara et al., 1992). Modeling studiesby Le et al. (1995) predicted stem-loop structures aswell as a psuedoknot in the BVDV and CSFV 5’


Figure 11.2. The secondary structure of the 5’ UTR/IRES of the BVDV genomic RNA. The conserved stem-loopstructures of this region are illustrated, along with the location of the binding sites of eIF3 and the 40S ribosomalsubunits (Pestova et al., 2001). The nucleotide sequence is not shown because of variation between BVDV strains.


UTRs that were consistent with prior RNase sensi-tivity studies (Brown et al., 1992). The pestivirusIRES also contained a structural motif that was pro-posed to base pair with the 3’ end of the 18S rRNA,in a manner similar to that described for the picor-naviruses (Deng and Brock, 1993; Le et al., 1995).This base pairing would occur at sequences immedi-ately 5’ of the AUG initiation codon. It was proposedthat this base pairing between the viral and 18srRNA functioned in initiation of cap-independenttranslation by positioning of the ribosome.

The first description of the 5’ UTR as a functionalIRES was by Poole et al. (1995). In vitro transcrip-tion of the 5’ UTR along with the p20 (Npro) codingsequences of the BVDV ORF yielded expression ofthe p20 protein in an in vitro cell-free expressionsystem. Inclusion of the 5’ UTR as a spacer betweenthe CAT and luciferase genes in a bicistronic con-struct yielded expression of both genes. Expressionstudies using constructs with both genes without aspacer resulted in only CAT expression. Deletion ofnucleotides 173 through 236 of the 5’ UTR reducedexpression of luciferase by about two-thirds. Theseresults suggested that the BVDV 5’ UTR possessesIRES activity similar to that of HCV. Rijnbrand etal. (1997), in a separate study, demonstrated that theCSFV IRES sequences functioned similarly whenused as a spacer in a bicistronic construct. Muta-tional analysis of the nucleotide sequences thatformed the pseudoknot revealed that RNAs unableto form base pairs in stem II of the pseudoknot weretranslationally inactive. However, translational ac-tivity could be restored by introduction of compen-satory base changes that restored the stem loopstructure (Rijnbrand et al., 1997; Fletcher and Jack-son, 2002). Neither an AUG codon 7 base upstreamof the authentic initiation AUG codon nor an AUGdownstream could initiate translation. This indicatesthat scanning by the ribosome on the IRES is limitedto a very small region.

A genetic approach to dissection of BVDV IRESfunction (Chon et al., 1998) using mutational analy-sis through a number of in vitro–introduced dele-tions or insertions, showed that stem loops Ia and Ibwere dispensable for efficient translation, and IIIband IIIe were partially required (Figure 11.3).Deletions or insertions in II, IIIa, IIIc or IIId couldcause a tenfold or greater reduction in translation.Confirmation of these data was provided by chemi-cal and enzymatic footprinting (Sizova et al., 1998),which demonstrated that the translation initiationfactor eIF3 bound to and protected distinct regionsof domain III, particularly IIIb and IIIc (refer to

Figure 11.2). Further, deletions of these sequencesabrogated translation. They suggested that eIF3bound the IRES in a sequence- or structure-specificmanner. In UV cross-linking experiments to deter-mine which subunits of the 10 protein subunit eIF3complex actually contacts the IRES, 4 subunits werefound to be labeled by the radiolabeled RNA (p170,p116, p66, and p47). The p116 and p66 subunitsboth contain RNA recognition motifs and thus arelikely to be determinants of the interaction of eIF3with the IRES (Sizova et al., 1998).

Pestova and Hellen (1999) provided evidence thatthe IRES was bound independently of eIF3 and the40S ribosomal subunit. The IRES contains complexstructure determinants (refer to Figure 11.2) thatmediate attachment of the 48S ribosomal subunitcomplexes to the initiation codon (Pestova et al.,1998). Determinants of domain III mediate eIF3binding. Pestivirus translation initiation is depend-ent on eIF3, eIF2, GTP, and Met-tRNAmet (48Ssubunit assembly) but not on any constituent of theeIF4G subunit (refer to Figure 11.1B). eIF3 en-hances 48S complex formation, probably by stabi-lizing ribosomal complexes by its interaction withthe IRES, and is absolutely required for subsequentsubunit joining to form active 80S ribosomes(Sizova et al., 1998). The probable function of eIF3is to recruit factors eIF1 and eIF5 to the 48S com-plex (Pestova and Hellen, 1999). Using cryoelectronmicroscopy, Spahn et al. (2001) demonstrated thatbinding of the HCV IRES caused a conformationalchange in the 40S ribosomal subunit that resulted inclosing of the mRNA binding cleft, and in IRES-mediated positioning of the initiation codon in the Psite of the ribosome. This was the first description ofa viral RNA that actively manipulates the structureof the transcriptional machinery, resulting in thepromotion of translation initiation without the assis-tance of otherwise necessary initiation factors(Spahn et al., 2001).

Several groups have reported that initiation factorseIF4A, eIF4B, and eIF4F were not required for trans-lation initiation from the pestivirus IRES (Rijnbrandet al., 1997; Chon et al., 1998; Pestova and Hellen,1999; Sizova et al., 1998; Fletcher et al., 2002).These findings were derived from different types ofexperiments, including reconstituted in vitro transla-tion initiation systems (Pestova and Hellen, 1999)and coexpression of picornavirus 2A protease(Rijnbrand et al., 1997; Chon et al., 1998). In addi-tion, this IRES-driven translation was shown to be independent of the translation initiation factoreIF-4F by coexpression of the poliovirus 2A protease



that cleaves eIF4F. This protease-inhibited cap-dependent translation in picornavirus-infected cells(Rijnbrand et al., 1997; Chon et al., 1998; Pestova etal., 1998; Pestova et al., 2001; Fletcher et al., 2002).The report by Pestova et al. (1998) was the first de-scribing the lack of reliance by the pestivirus IRESon canonical translation initiation factors that arenecessary for normal cellular cap-dependent transla-tion initiation. It was concluded that pestivirus pro-tein translation initiation is simpler than that em-ployed by the picornaviruses and more closelyresembles prokaryotic translation initiation. Trans-lation initiation from the flavivirus or pestivirusIRES requires so few translation initiation factorsthat a simpler, equally efficient means of translationinitiation is unlikely (Pestova et al., 1998). CellularRNAs that contain an IRES are generally those thatare expressed under stress conditions (i.e., heatshock, reactive oxygen species) when the activitiesof eIF4B, E, and F are reduced (Pestova et al., 1998).

The 5’ end of the IRES was mapped to betweennucleotides 28 and 75 (Rijnbrand et al., 1997; Chonet al., 1998). The 3’ end of IRES extends into Nprocoding sequences of the ORF (Chon et al., 1998) asfar as 51 bases (Fletcher et al., 2002) and possessessingle-strandedness that appears important for ribo-some binding and translation initiation. Studiesmapping the termini of the IRES were done utilizingin vitro systems where the IRES was directly fusedto the ORF encoding the expression reporter.Additional work showed that efficient IRES func-tion, in the context of translation from a functionalreplicon, required the 5’ Ia stem-loop structure (Yuet al., 2000). Utilizing the subgenomic RNA repli-con DI9c (Behrens et al., 1998), translation wasshown to be inefficient when the Ia stem-loop struc-ture was mutated or missing. In addition, it wasfound that the Ia structure was important in RNAreplication as well. The authors postulated that thisstructural motif may be involved in RNA:RNA in-teractions that may be necessary for committing theRNA molecule to function as template for eithertranslation or RNA replication.

PROCESSING OF ENVELOPE GLYCOPROTEINSBY CELLULAR ENZYMES

The BVDV polyprotein, encoded by the single ORFof the genomic RNA, is co-and posttranslationallyprocessed by both host and viral proteases to yieldthe mature viral proteins. The protease cleavages ofthe nonstructural proteins are carried out by theNS2/3 serine protease (Wiskerchen and Collett,1991; Tautz et al., 1997; Xu et al., 1997). However,

the processing of the virus structural proteins, en-coded in the N-terminal one-third of the polyprotein,is carried out by host signal peptidases (Figure11.3A) following translocation into the lumen of theendoplasmic reticulum (Figure 11.3; Rümenapf etal., 1993; Lin et al., 1994). A model of this translo-cation and proteolytic processing has been proposed(Rümenapf et al., 1993). Cotranslationally, the enve-lope proteins are translocated into the ER utilizing asignal sequence located between the capsid and Erns

proteins. In doing so, the ribosome attaches to theSec61p complex, a protein translocation channel inthe ER membrane. The signal peptide is insertedinto the membrane and the nascent protein chain istranslocated into the lumen of the ER in a loop struc-ture (Matlack et al., 1998). A rapid signalase cleav-age occurs at the N-terminus of the Erns protein, pos-sibly leaving the capsid protein attached to thecytoplasmic side of the membrane.

The core (capsid) protein of HCV is released fromthe endoplasmic reticulum membrane by the cellularprotease SPP (McLauchlan et al., 2002), and it islikely that process occurs in a similar manner inBVDV-infected cells. Transfer of the remainingErns/E1/E2 envelope protein precursor continueswith interruption by two hydrophobic domains atthe end of the E1 protein. The first hydrophobic do-main acts as a stop transfer for the nascent protein,and the second acts as a signal sequence to begintranslocation of the E2 protein. Both may functionto anchor the E1 protein into the membrane. Thereare similar stop/start transfer signals at the C-terminus of the E2 region. The internal signal se-quences of the Erns/E1/E2 precursor are the onlydeterminants for compartmentalization of the ma-ture envelope proteins (Wu, 2001). Cleavage of theErns/E1/E2 precursor from the downstream transla-tion product occurs rapidly. Processing of the pre-cursor begins with the signalase cleavage at theE1/E2 border. This is followed by the separation ofthe Erns/E1 proteins and is probably dependent on aconformational change taking place following theinitial E2 cleavage reaction. The Erns protein formshomodimers by intermolecular disulfide bond for-mation and, having no hydrophobic transmembranedomain, remains soluble and associates with the E1-E2 heterodimers by an unknown mechanism (Hulstand Moorman, 2001). Unassociated Erns is also se-creted into the extracellular domain and can befound free in the medium of infected cells culturedin vitro (Rümenapf et al., 1993). As discussed previ-ously, Erns is essential for virus attachment and up-take (Hulst and Moormann, 1997).



The E1 and E2 proteins form disulfide-linked het-erodimers after proper folding of the proteins in as-sociation with the endoplasmic reticulum proteincalnexin (Branza-Nichita et al., 2001). The foldingof these proteins is dependent on intramoleculardisulfide bridge formation and varies in the rate thisoccurs between the two envelope proteins. The fold-ing of the E2 protein is rapid, showing completionwithin 2.5 minutes; the folding of the E1 proteinstakes approximately 30 minutes. Thus the rate offolding of the E1 protein is the rate-limiting step inthe formation of E1-E2 dimers. The interaction ofenvelope proteins with the lectin-like calnexin mol-ecule is dependent on the removal of glucose moi-eties on the N-linked glycans by endoplasmic retic-ulum �-glucosidases (Branza-Nichita et al., 2001).If the removal of the glucose moieties does notoccur, interaction with calnexin is inhibited and theimproperly folded proteins are targeted for degrada-tion by the proteosome (Branza-Nichita et al.,2002). Decrease in size of cleaved proteins that in-dicated trimming of sugar moieties on the side-chains has been reported (Rümenapf et al., 1993).Jordan et al. (2002a) demonstrated that inhibition of�-glucosidase removal of terminal glucose moieties

from the nascent E2 glycoprotein prevented golgiprocessing of the E2 glycoprotein with the loss ofproduction of infectious progeny virus.

RNA REPLICATION

Replication of the BVDV genomic RNA occursdownstream of polyprotein translation, being de-pendent on the presence of viral proteins that func-tion specifically in RNA replication. Replication ofthe genomic RNA involves the copying of the in-fecting plus-sense RNA into minus-sense RNAcopies, forming the replicative form (RF). Thenegative-strand of the RF is copied by strand-displacement, giving rise to the RNA replicative in-termediate (RI), and releasing the displaced plus-sense RNA strand when nascent strand transcriptionis complete. The RF is reutilized, primarily by recy-cling of the negative-sense. The kinetics of synthesisshowed that the rate of synthesis of the positive-sense strand at 12 hours postinfection was 10 timesthat of the negative strand (Gong et al., 1996). Atany given time, there were approximately 6–7 nas-cent plus-stranded RNA strands on the template RI(Gong et al., 1998).

Viral proteins known to participate in RNA repli-


Figure 11.3. Proposed mechanism for transporting and processing of pestivirus envelope proteins. A. Genomicorganization of the BVDV genome showing cleavage sites of the mature proteins with those cleaved by hostsignalases denoted by asterisks. The sizes of the proteins are not drawn to scale. B. Import into the lumen of theendoplasmic reticulum and cleavage of the mature proteins from the polyproteins result in the release of free Erns

into the lumen of the endoplasmic reticulum, while cleavage of the E1 and E2 proteins results in retention of themembrane anchors in the membrane, probably facilitating downstream processing, heterodimerization and transport.


cation include NS2/3, NS4B, NS5A, and NS5B.NS2/3 possesses RNA helicase and nucleotidetriphosphatase activities essential for RNA replica-tion (Warrener and Collett, 1995; Grassmannn et al.,1999; Gu et al., 2000). NS5B contains the GDDamino acid motif characteristic of RNA-dependentRNA polymerases and has been demonstrated topossess RNA polymerase activity and be essentialfor RNA replication (Zhong et al., 1998; Lai et al.,1999). NS4B is an integral endoplasmic reticulumprotein that, besides being part of the multiproteincomplex composed of NS3, NS4B, and NS5A, canalso play a role in suppression of the cytopathic phe-notype by mutation of the highly conserved tyrosineresidue at position 2441 (Qu et al., 2001). This ap-parent tripartite protein complex may be associatedwith the endoplasmic reticulum membrane via thetransmembrane domain of NS4B (Qu et al., 2001).The function of NS5A remains elusive and is dis-cussed later with its known interactions with cellu-lar proteins.

Currently, little is known concerning the identityand the role of host factors in the replication of theBVDV genomic RNA. Highly conserved RNA sec-ondary structures at both the 5’ and 3’ ends of theBVDV genomic RNA indicate functional roles(Becher et al., 2000; Yu et al., 2000). As illustratedby the function of the IRES, these secondary struc-tures probably bind specific cellular factors, as wellas viral proteins, that participate in RNA replication(Deng and Brock, 1993; Becher et al., 2000). The re-quirement for the secondary structure contained inthe 5’ nontranslated region of the BVDV genomicRNA in viral replication has been analyzed (Yu etal., 2000). Stem-loop Ia, found outside the IRES se-quences at the extreme 5’ end of the genomic RNA(Rijnbrand et al., 1997; Chon et al., 1998), appearsto regulate the switching of the genomic RNA ofBVDV as template for either translation or RNAreplication (Behrens et al., 1998; Yu et al., 2000; Liand McNally, 2001). This was supported by findingsthat regulation of RNA replication and translationwas outside the IRES region and was probably dueto interaction with regulatory proteins or viral pro-tein(s) associated with specific host protein transla-tion complexes (Li and McNally, 2001).

Several reports have demonstrated interactions ofcellular proteins with 3’ nontranslated region sec-ondary structures (both plus and minus strands) offlaviviruses. Several of these proteins have beenidentified (Blackwell and Brinton, 1995; Blackwelland Brinton, 1997; Ito and Lai, 1997; Li et al.,2002). Two separate reports may provide evidence

for one possible function of NS5A. The first studydemonstrated that the cellular protein elongationfactor-1� (EF-1�) recognized and bound to highlyconserved secondary structure found on the plus-strand of the 3’ nontranslated region of West Nilevirus (Blackwell and Brinton, 1997). The seconddemonstrated that NS5A protein of BVDV interactsspecifically with EF-1� (Johnson et al., 2001).Taken together, these data indicate that binding ofEF-1� to both the secondary structure of the 3’ non-translated region and to NS5A may act to bring thegenomic RNA template and the NS5A containingendoplasmic reticulum membrane-bound replica-tion complex into the correct position or orientationfor RNA replication to take place. EF1-� plays nu-merous roles in RNA sorting and regulation of ex-pression of mRNA in eukaryotic cells and may playa role in RNA replication of BVDV.

The RNA-binding proteins TIA-1 and TIAR havebeen identified as cellular proteins that interact withthe 3’ nontranslated region of West Nile virus (Li etal., 2002). In contrast to EF-1�, TIAR and TIA-1bind to the 3’ nontranslated region of the minusstrand of the West Nile virus genomic RNA. WestNile virus grew to significantly lower titers inmurine TIAR-/- knockout cells; however, growthwas not eliminated. TIA-1-/- knockout cells showedno decrease in final virus titer but these levels werereached 6 hours later than those grown in normalcontrol cells, indicating distinct requirements for thetwo proteins. The identification of proteins that dif-ferentially bind to the plus and minus strands of thegenomic RNA may indicate a mechanism to distin-guish between the two strands during replication.These data demonstrate that host proteins that nor-mally interact with cellular RNA are essential to theflavivirus RNA replication process. It is expected tobe similar in BVDV RNA replication.

The function of some BVDV proteins may be reg-ulated through phosphorylation by cellular proteinkinases. The NS5A protein of BVDV was shown tobe phosphorylated (Reed et al., 1998). The ability ofNS5A to interact with both cellular and viral pro-teins and perhaps regulate their function may be de-pendent on its phosphorylation state (Kapoor et al.,1995; Reed et al., 1998). The phosphorylation of theBVDV NS5A protein indicates that it is associatedwith serine/threonine protein kinases (Reed et al.,1998). Kinase inhibitor studies indicated that the as-sociated kinases belong to the CMGC family of pro-tein kinases (Reed et al., 1997). This interaction hasbeen postulated to regulate viral replication and pos-sibly host gene expression, whether through trans-



port of NS5A into the nucleus or through interac-tions with cellular signaling pathways. An exampleof the latter includes the interaction with the hostdouble-stranded RNA-activated protein kinase(PKR) by the HCV NS5A protein. This interactioninhibits the host interferon antiviral response, in-cluding inhibition of the phosphorylation of transla-tion initiation factor eIF-2� and subsequent de-crease in protein translation (Gale et al., 1997).

CELL DEATHApoptosis, or programmed cell death, is the geneti-cally and biochemically defined mechanism of cel-lular suicide that may be induced for a number ofphysiological reasons. From a virological stand-point, this is one of the first lines of defense againstviral infection. Its primary purpose is to kill thevirus-infected cell before the virus replicates andspreads to neighboring cells. Apoptosis is also ameans to kill the cell without inducing an inflamma-tory response that may damage surrounding tissue(Steller, 1995). Apoptosis is characterized by manycellular changes, including cellular swelling, loss ofplasma membrane integrity, loss of mitochondrialmembrane potential, release of compartmentalizedmolecules into the cytoplasm, proteolytic cleavageof proteins (both activation and inactivation of func-tion), and degradation of nuclear DNA. Apoptosis isinduced in cells by two major pathways, the extrin-sic and intrinsic pathways. The extrinsic pathway isinduced when an extracellular signal is received andtransduced into the cytoplasm. This results in the ac-tivation of caspase 8, initiating the caspase-mediateddestruction of the cell. The intrinsic pathway is acti-vated by an intracellular stimulus that results in lossof mitochondrial membrane potential (��m) and re-lease of cytochrome c. Cytochrome c, free in the cy-toplasm, is bound by APAF-1 and, along with ATP,activates caspase 9, thus beginning the caspase-mediated cellular destruction.

Many viruses encode proteins that interact withcellular defense mechanisms, resulting in the inhibi-tion of apoptosis until viral replication steps havebeen conducted, allowing production of maximallevels of progeny virus. At this point, no inhibitor ofapoptosis has been specifically identified in theBVDV genome. The hallmark of the cp BVDVstrains is the induction of cell death in cultured ep-ithelial cells soon after infection, generally within24–48 hours. The inability of the cp BVDV strainsto infect susceptible cells without inducing celldeath is considered loss of function, because themore prevalent noncytopathic BVDV (ncp BVDV)

strains possess this ability. The loss of the ability toprevent cell death by cp BVDV strains appears to berelated to the genetic changes that occur within theNS2/3 protein coding sequences that give rise to theformation of NS3 (Hoff and Donis, 1997; Lambot etal., 1998), although other mechanisms cannot beruled out (Bruschke et al., 1997; Vassilev and Donis,2000; Qu et al., 2001).

Late in the infection process, cells infected withcp BVDV strains show many of the classic signs ofapoptosis. These include rounding of cells, cleavageof nuclear DNA to oligonucleosomal fragments, andcleavage and inactivation of poly (ADP-ribose)polymerase, a nuclear enzyme important in DNA re-pair (Zhang et al., 1996; Hoff and Donis, 1997). Inan investigation of the mechanism of induction ofapoptosis in cp BVDV-infected cells, Grummer etal. (2002a) demonstrated that cp BVDV-infectedcells induced the intrinsic apoptotic pathway. Thiswas shown by translocation of cytochrome c into thecytoplasm, increased expression of APAF-1, and in-creased caspase 9 activity that is indicative ofAPAF-1 activation. Inhibition of loss of mitochondr-ial ��m and subsequent loss of cytochrome c intothe cytoplasm delayed onset of apoptosis. Addi-tionally, treatment of cells with apoptosis inhibitorsdelayed induction of apoptosis (Grummer et al.,2002b). Disruption of the ��m disrupts normal cel-lular oxidation/reduction. Schweizer and Peterhans(1999) provided evidence that cp BVDV-infectedcells show an increase in reactive oxygen species,indicative of oxidative stress that preceded caspaseactivation. Antioxidants that protected the cell fromoxidative stress, prevented apoptosis. Interestingly,this had no effect on virus replication and virustiters, indicating that the loss of ��m and inducedoxidative stress plays no role in cp BVDV replica-tion (Schwiezer and Peterhans, 1999; Grummer etal., 2002a).

The envelope glycoprotein Erns has been reportedto possess RNase activity (Schneider et al., 1993;Hulst et al., 1994), the function of which is un-known. The effect of Erns on cells infected withCSFV was investigated by Bruschke et al. (1997).This was done because of the immunomodulatoryeffects of RNases (Tamburrini et al., 1990;D’Alessio, 1993) and the known immunosuppres-sion associated with pestiviral infections. Treatmentof lymphocytes from porcine, bovine, ovine, andhuman sources with Erns completely inhibited con-canavalin A activation in vitro. In addition, proteinsynthesis was inhibited in the Erns-treated cells with-out disruption of the plasma membrane. Examina-



tion of cells following treatment revealed that apop-tosis was induced in the treated cells. These resultssuggest that Erns may, at least in part, be responsiblefor the leukopenia observed in pestivirus infectedanimals. Earlier work had shown that Erns was se-creted in the medium of cultured cells (Rümenapf etal., 1993). If this is true in vivo, circulating 0.0 maybe able to effect cell death in lymphocytes withoutdirect viral infection and replication. The mode ofentry into the susceptible cell remains a mystery.Langedijk (2002) demonstrated that short peptidescorresponding to the C-terminus of Erns were trans-ported across the plasma membrane and targeted tothe nucleolus. The region involved, as short as 13amino acid residues in length, appeared to be asso-ciated with conserved basic residues. These C-terminal sequences, when attached to unrelated pro-teins, would still function to mediate transport intothe cell. The translocation was rapid, occurring in 1minute or less, and was not dependent on energy orcell surface receptors. Delivery of the peptide to thenucleolus may indicate that Erns has a role in geneexpression or protein synthesis modulation.

BVDV, like all members of the Flaviviridae, uti-lize the endoplasmic reticulum in their replicationprocess with the envelope proteins being insertedinto the membrane of the endoplasmic reticulum forfurther modification. When the endoplasmic reticu-lum experiences loss of Ca2+ or the accumulation ofmisfolded or unassembled proteins, the endoplasmicreticulum stress response is triggered. This acts toslow translation of proteins and is mediated throughphosphorylation of eIF-2� by the ER kinase PERK(Ma et al., 2002). This is maintained until the prob-lem of the accumulated proteins is resolved(Kaufman, 1999). PERK phosphorylation of eIF-2�also leads to transcriptional induction of endoplas-mic reticulum chaperone proteins as well as tran-scriptional repression of bcl-2 (anti-apoptotic pro-tein). If the PERK-mediated protein translationinterruption is of sufficient duration, apoptosis maybe induced by decreased levels of bcl-2 (Friedman,1996). Jordan et al. (2002b) demonstrated thatMDBK cells infected with cp BVDV induce the ERstress response. This response was characterized byactivation of PERK, hyperphosphorylation of eIF-2�, and decreased transcription of bcl-2. In addition,increased caspase 12 activity and decreased glu-tathione levels were noted. The latter would be aconsequence of decreased bcl-2 protein levels andcould potentially contribute to increased susceptibil-ity to reactive oxygen species (Schweizer andPeterhans, 1999; McCullough et al., 2000). Overex-

pression of bcl-2 in cells infected with dengue virusand Japanese encephalitis virus was demonstrated tochange the cellular response to the infection fromapoptotic to chronic (Su et al., 2001; Su et al., 2002).

INHIBITION OF SIGNALING

CYTOKINES

The innate immune response represents the first lineof defense against an invading virus and occurs atthe cellular level. Interferon production and induc-tion of apoptosis represent two mechanisms a cellmay employ to attempt to limit the infection.Synthesis and release of interferon act not only atthe infected cell but also to notify other cells of thedanger posed by the virus. Many viruses encodeproteins that target components of the innate im-mune system to limit the response, giving the virustime to replicate to maximal levels. BVDV appearsto be no exception; however, the mechanisms uti-lized to bring this about remain unclear. Adler et al.(1994), working in vitro with bovine bone mar-row–derived macrophages (BBMM), demonstratedthat only ncp BVDV strains primed BBMM for en-hanced nitric oxide production in the presence ofSalmonella. Infection of cells by cp BVDV resultedin cytopathic effect (apoptosis), demonstrating thatthe two biotypes acted differently in affecting spe-cific functions of the host. In a later study, Adler etal. (1996), again working with BBMM, showed thatinfection with both biotypes resulted in decreasedproduction of tumor necrosis factor � (TNF�) uponstimulation with Salmonella, while other specificmacrophage functions were not affected. This sug-gested that decline in the ability to produce TNF�may contribute to immunosuppression often ob-served in BVDV infections. In addition, the infectedBBMM also produced a substance that primed unin-fected cells for decreased nitric oxide productionand for apoptosis following treatment with lipopoly-saccharide (Adler et al., 1997; Jungi et al., 1999).This substance was believed to be interferon, basedon its physicochemical properties (Adler et al.,1997; Perler et al., 2000).

INTRACELLULAR SIGNALING INHIBITION

The production of interferons by virus-infected cellsis an important function of the innate immune re-sponse. The mechanism of inhibition of interferonsynthesis by ncp BVDV in infected cells remainsunclear. ncp BVDV strains, but not cp BVDVstrains, possess a function that inhibits interferonproduction in response to infection or treatment with



dsRNA, but not the response to exogenous interfer-ons (Schweizer and Peterhans, 2001). dsRNA is amolecule produced during the infection process bymost viruses and has been shown conclusively to bea specific trigger of apoptosis and interferon produc-tion (Jacobs and Langland, 1996; Kibler et al.,1997). This inhibition was observed only in cellsthat had been infected with the ncp BVDV strain forat lease 12 hours, indicating that inhibition wasvirus-induced. ncp BVDV inhibited induction ofapoptosis by dsRNA but not by staurosporine or ac-tinomycin D. In addition, this inhibition wasthrough an unknown intracellular mechanism andnot by inhibition of uptake of the dsRNA. This sug-gested that some component encoded by ncp BVDVinterfered specifically with intracellular signaling.By using different cell lines and different strains ofncp BVDV, Schweizer and Peterhans (2001)demonstrated that interference with apoptosis andinterferon synthesis was a general and importantfunction of ncp strains.

Baigent et al. (2002) extended the above studiesto further characterize the lack of intracellular re-sponse in cells infected with ncp BVDV. As de-scribed by Schweitzer and Peterhans (2001), priorinfection with ncp BVDV did not block the responseto exogenous interferons but could block interferonand MxA gene transcription in response to dsRNAor infection by the heterologous virus SemlikiForest virus (SFV). Prior infection with ncp BVDVdid not block induction of apoptosis by cp BVDV orSFV. The nature of the interferon response requirestranscriptional up-regulation or posttranslationalmodification of specific transcription factors, in-cluding NF-�B, ATF2, and c-jun. Examination ofthese proteins revealed that NF-�B was not acti-vated, and ncp BVDV infection did not block activa-tion of NF-�B by SFV or by tumor necrosis factor-�. The stress-activated kinases JNK1 and JNK2were not activated nor were the transcription factorsATF2 and c-Jun phosphorylated. These events werenot inhibited following superinfection by SFV fol-lowing ncp BVDV-infection. Interferon regulatoryfactor-3 (IRF-3), a transcriptional activator responsi-ble for the increased transcription of interferongenes, was translocated to the nucleus in ncpBVDV-infected cells but was shown to lack DNAbinding activity in nuclear extracts (Baigent, et al.,2002). In addition, IRF-3 DNA binding activity waspresent in SFV-infected cells but could be blockedby prior infection with ncp BVDV.

HCV was recently shown to also inhibit IRF-3transcriptional activation of interferon genes (Foy et

al., 2003) but by an apparently mechanistically dif-ferent means. The HCV NS3/4A serine proteaseprevented the phosphorylation of IRF-3, blocking itstranslocation to the nucleus. This is in contrast to theIRF-3 in ncp BVDV-infected cells where IRF-3 istranslocated to the nuclease but does not bind DNA.The two viruses appear to utilize two differentmechanisms to achieve the same goal.

A recent report by Ruggli, et al. (2003) suggestedthat the pestivirus protein Npro may play a role in in-terference with the induction of the interferon re-sponse. Porcine SK-6 cells infected with wild-typeCSFV possessed the ability to resist induction ofapoptosis when treated with dsRNA, and macro-phages were inhibited from producing an �/ß inter-feron response. When infected with CSFV mutantslacking the Npro coding sequences, SK-6 cells werenot protected from dsRNA-induced apoptosis andmacrophages produced a type I interferon responsein the absence of additional stimuli. In addition, re-duced replication was noted for the Npro-deficientmutant. These data imply that Npro plays a role inthe inhibition of the interferon response, the mecha-nism of which still remains to be determined.

CELLULAR REMODELINGRecent advances in functional genomics have madeit possible to examine changes in gene expression incells under a variety of conditions that allow an un-derstanding of the basic mechanism(s) that bringabout distinct cellular changes. These changes, indi-cated by increased or decreased transcription of spe-cific genes, allow the dissection of the host responsebased on the function of the genes with altered ex-pression levels. One functional genomics technol-ogy, serial analysis of gene expression (SAGE), wasapplied to examining gene expression changes thattake place in BVDV-infected cells in response to theviral infection (Neill and Ridpath, 2003a, b). Thisanalysis revealed that a number of gene expressionchanges occur following BVDV infection and thatsome of the changes act to increase efficiency offunctions that are beneficial to viral replication, mat-uration, and release. Although expression level ofgenes involved in energy production and metabo-lism were relatively unchanged, genes encodingproteins involved in protein translation were alteredin a pro-virus manner. Protein translation plays suchan integral role in the replication of a virus that al-terations in the translational machinery that en-hances viral protein synthesis are advantageous.These changes included ribosomal proteins, elonga-tion factors, and tRNA synthetases. In addition,



transcription of genes involved in transport of nas-cent polypeptides into the lumen of the endoplasmicreticulum was increased, indicating possible in-creased capacity for ER translocation and process-ing. One protein in particular, TRAM, functions inregulating the Sec61p pore complex through whichthe nascent polypeptide passes into the lumen of theendoplasmic reticulum (Hegde et al., 1998a).TRAM also regulates exposure of the nascent pro-tein to the cytoplasm in response to pause transfersignals (Hegde et al., 1998b). This may be importantfor BVDV replication in allowing the correct pro-cessing of the structural proteins. Gene expressionchanges were also noted with unknown effect orbenefit to the replicating virus. Several tubulin iso-types showed sharp declines in transcript numbers,indicating possible disruption of microtubule func-tion. This may suggest a possible mechanism for thedisruption of platelet production by BVDV-infectedmegakaryocytes. Proper microtubule function is es-sential for pro-platelet release, and disruption of themicrotubule network would have a detrimental ef-fect on this process.

VIRUS RELEASELittle is currently known concerning release of in-fectious BVDV virions from the infected cell. Thisis one of the more confusing aspects of flavivirus/pestivirus biology because of apparent differences inthe mechanisms of viral egress used by differentviruses. Egress by budding from the cell surface wasreported for West Nile (Sarafend) virus and wasdemonstrated by both transmission and scanningelectron microscopy (Ng et al., 1994). Examinationof release of virus in polarized epithelial cells byChu and Ng (2002) showed that West Nile (Sara-fend) was released from the apical surface, andKunjin virus was released from both apical and ba-solateral surfaces. In addition, microtubules wereshown to play a role in the sorting and transport ofviral proteins to the cell surface by West Nile(Sarafend) virus, and disruption of the microtubulenetwork had no effect on Kunjin virus maturationand egress (MacKenzie and Westaway, 2001; Chuand Ng, 2002). Transport of Kunjin virus to the cellsurface from the golgi apparatus appears to be bymovement of virus-containing vesicles and fusionwith the plasma membrane.

The release of mature pestivirus particles appearsat this point to more closely resemble that of Kunjinvirus. A study examining location of viral proteinsin cells infected with CSFV (Weiland et al., 1999)showed that the viral proteins were associated only

with released viral particles. Lack of detection ofviral proteins in the plasma membrane of BVDV-infected cells was confirmed by Grummer et al.(2001) using indirect fluorescence microscopy, con-focal microscopy, or FACS analysis. Using subcel-lular fractionation techniques, the envelope glyco-proteins Erns and E2 were found associatedexclusively with intracellular membranes. In addi-tion, with the possible disruption of microtubule net-works in BVDV-infected cells (Neill and Ridpath,2003), the microtubule-associated sorting of viralproteins required for surface budding would be af-fected. This provides further evidence for the use ofan egress mechanism more closely resembling thatof Kunjin virus.

BVDV REPLICATION CYCLEOVERVIEWThe mechanisms utilized by BVDV to enter a sus-ceptible cell and replicate itself are becomingclearer. It is now possible to put together the pictureof what takes place from attachment of the infectingvirus particle to release of infectious progeny. As de-tailed in Figure 11.4, the infection process begins byattachment of the BVDV particle to the plasmamembrane, probably first by attachment of Erns to adocking surface glycosaminoglycan(s) followed bybinding of the LDLR E2 (Figure 11.4A). The inter-nalization of the virion is probably mediated by theLDLR receptor through endocytosis (Figure11.4B,C), and release of the genomic RNA occursfollowing acidification of the endosomal vesicle(Figure 11.4D,E). The genomic RNA, now in the cy-toplasm of the cell first acts as mRNA for translationof the polyprotein at least initially taking place onthe endoplasmic reticulum (Figure 11.4F). At somepoint, a switch occurs that causes the genomic RNAto be used as template for RNA replication ratherthan for protein translation (Figure 11.4G). It is pre-sumed that some of the newly synthesized daughterRNAs are then used for protein translation whileothers participate in RNA replication. The genomicRNAs also interact with the capsid protein followedby recognition of the cytoplasmic domains of theenvelope proteins and budding into the lumen of the endoplasmic reticulum (Figure 11.4H,I,J). Thepassage of the immature particles through the endo-plasmic reticulum and the golgi body result in thematuration of the particles by processing and glyco-sylation of the envelope proteins (Figure 11.4K).The virus-laden vesicles (Figure 11.4L) movethrough the cytoplasm to the cell surface (Figure11.4M) where the vesicle fuses with the plasma



membrane, releasing the virus particle into the ex-tracellular environment. The budding of the capsid-bound genomic RNA and the immature virus parti-cle into the lumen of the endoplasmic reticulum,followed by the release of mature particles directlyfrom a vesicle fused with the plasma membrane,would explain why viral antigens are not found as-sociated with plasma membrane of the infected cell.

FINAL REMARKSReplication of an RNA virus in a susceptible cell isa complicated affair, requiring essentially all of thecell’s resources. This is an intricate combination ofcellular machinery and virus-encoded proteinsworking in concert, most often to the detriment ofthe cell, to bring about production of progeny virus.Some of these effects are obvious, but others arevery subtle in their effect. Increasingly more exper-imental evidence is available showing the depend-ence on host cell proteins in carrying out many as-pects of virus replication. BVDV is no different,although much of what we know is supported or in-ferred from work with other pestiviruses or fla-viviruses. Regardless, much is left to do. This is par-

ticularly true in two areas. The first is how the virusinteracts with the host cell in establishing a chronicinfection in utero that results in the birth of a persist-ently infected calf. The second is the subtle interac-tions with the host immune system and how it ismanipulated to delay both the innate and acquiredimmune responses. Continued work will givegreater insight into replication strategy of pestivi-ruses and possible means to intervene, yielding bet-ter vaccination strategies and perhaps pharmaco-logical treatments to prevent, inhibit, or moderateinfection.


phages infected with cytopathic bovine viral diar-rhea virus release a factor(s) capable of priming un-infected macrophages for activation-inducedapoptosis. J Virol 71:3255–3258.

Adler H, Frech B, Meier P, et al.: 1994, Noncyto-pathic strains of bovine viral diarrhea virus primebovine bone marrow derived macrophages for en-hanced generation of nitric oxide. Biochem BiophysRes Comm 202:1562–1568.

Adler H, Jungi TW, Pfister H, et al.: 1996, Cytokineregulation by virus infection: bovine viral diarrheavirus, a flavivirus, downregulates production oftumor necrosis factor alpha in macrophages invitro. J Virol 70:2650–2653.

Agnello V, Ábel G, Elfahal M, et al.: 1999, HepatitisC virus and flaviviridae viruses enter cells via lowdensity lipoprotein receptor. Proc Natl Acad SciUSA 96:12766–12771.

Asagoe T, Inaba Y, Jusa E.R, et al.: 1997, Effect ofheparin on infection of cells by equine arteritisvirus. J Vet Med Sci 59:727–728.


Becher P, Orlich M., Thiel H-J: 2000, Mutations inthe 5’ nontranslated region of bovine viral diarrheavirus result in altered growth characteristics. J Virol74:7884–7894.

Behrens S-E, Grassmannn CW, Thiel H-J, et al.:1998, Characterization of an autonomous subge-nomic pestivirus RNA replicon. J Virol72:2364–2372.

Blackwell JL, Brinton MA: 1995, BHK cell proteinsthat bind to the 3’ stem-loop structure of the WestNile virus genome RNA. J Virol 69:5650–5658.

Blackwell JL, Brinton M.A: 1997, Translation elonga-tion factor-1 alpha interacts with the stem-loop


Figure 11.4. Overall view of BVDV infection, replica-tion, and release of progeny virus derived usingcurrently available information. Please see text foradditional details.


region of West Nile virus genomic RNA. J Virol71:6433–6444.

Branza-Nichita N, Durantel D, Carrouée-Durantel S,et al.: 2001, Antiviral effect of N-butyldeoxynojir-imycin against bovine viral diarrhea virus correlateswith misfolding of E2 envelope proteins and im-pairment of their association into E1-E2 het-erodimers. J Virol 75:3527–3536.

Branza-Nichita N, Lazar C, Durantel D, et al.: 2002,Role of disulfide bond formation in the folding and assembly of the envelope glycoproteins of apestivirus. Biochem Biophys Res Comm296:470–476.

Brinton MA: 2001, Host factors involved in West Nilevirus replication. Ann NY Acad Sci 951:207–219.

Brown EA, Zhang H, Ping L-H, Lemon SM: 1992,Secondary structure of the 5’ nontranslated regionsof hepatitis C virus and pestivirus genomic RNAs.Nucl Acids Res 20:5041–5045.

Bruschke CJM, Hulst MM, Moormann RJM, et al.:1997, Glycoprotein Erns of pestiviruses inducesapoptosis in lymphocytes of several species. J Virol71:6692–6696.

Chen Y, Maguire T, Hileman RE, et al.: 1997, Denguevirus infectivity depends on envelope protein bind-ing to target cell heparan sulfate. Nature Med3:866–871.

Chon SK, Perez DR, Donis RO: 1998, Genetic analy-sis of the internal ribosome entry segment of bovineviral diarrhea virus. Virology 251:370–382.

Chu JJH, Ng ML: 2002, Infection of polarized epithe-lial cells with flavivirus West Nile: Polarized entryand egress of virus occur through the apical sur-face. J Gen Virol 83:2427–2435.

Collett MS, Moenning V, Horzinek, MC: 1989,Recent advances in pestivirus research. J Gen Virol70:253–266.

Colston E, Racaniello, VR: 1994, Soluble receptor-resistant poliovirus mutants identify surface and in-ternal capsid residues that control interaction withthe cell receptor. EMBO J 13:5855–5862.

Compans RW: 1995, Virus entry and release in polar-ized epithelial cells. Curr Top Microbiol Immunol202:209–219.

D’Alessio G: 1993, New and cryptic biological mes-sages from RNases. Trends Cell Biol 3:106–109.

Dechecchi MC, Melotti P, Bonizzato A, et al.: 2001,Heparan sulfate glycosaminoglycans are receptorssufficient to mediate the initial binding of aden-ovirus types 2 and 5. J Virol 75:8772–8780.

Deng R, Brock KV: 1993, 5’ and 3’ untranslated re-gions of pestivirus genome: Primary and secondarystructure analyses. Nucl Acids Res 21:1949–1957.

Fletcher SP, Ali IK, Kaminski A, et al.: 2002, The in-fluence of viral coding sequences on pestivirus

IRES activity reveals further parallels with transla-tion initiation in prokaryotes. RNA 8:1558–1571.

Fletcher SP, Jackson RJ: 2002, Pestivirus internal ri-bosome entry site (IRES) structure and function:Elements in the 5’ untranslated region important forIRES function. J Virol 76:5024–5033.

Flores EF, Donis RO: 1995, Isolation of a mutantMDBK cell line resistant to bovine viral diarrheavirus infection due to a block in viral entry.Virology 208:565–575.

Flores EF, Kreutz LC, Donis RO: 1996, Swine and ru-minant pestiviruses require the same cellular factorto enter bovine cells. J Gen Virol 77:1295–1303.

Foy E, Li K, Wang C, et al.: 2003, Regulation of in-terferon-3 by the hepatitis C virus serine protease.Science 300:1145–1148.

Friedman AD: 1996, GADD153/CHOP, a DNA dam-age-inducible protein, reduced CAAT/enhancerbinding protein activities and increased apoptosis in32D c13 myeloid cells. Cancer Res 56:3250–3256.

Gale MJ Jr, Korth MJ, Tang NM, et al.: 1997,Evidence that hepatitis C virus resistance to inter-feron is mediated through repression of the PKRprotein kinase by the nonstructural 5A protein.Virology 230:217–227.

Gollins SW, Porterfield JS: 1986, The uncoating andinfectivity of the flavivirus West Nile on interactionwith cells: Effects of pH and ammonium chloride. JGen Virol 67:1941–1950.

Gong Y, Shannon A, Westaway EG, Gowans EJ: 1998,The replicative intermediate molecule of bovineviral diarrhea virus contains multiple nascentstrands. Arch Virol 143:399–404.

Gong Y, Trowbridge R, Macnaughton TB, et al.: 1996,Characterization of RNA synthesis during a one-step growth curve and of the replication mechanismof bovine viral diarrhoea virus. J Gen Virol77:2729–2736.

Grassmann CW, Isken O, Behrens S-E: 1999,Assignment of the multifunctional NS3 protein ofbovine viral diarrhea virus during RNA replication:An in vivo and in vitro study. J Virol73:9196–9205.

Greenstone HL, Santoro F, Lusso P, Berger EA: 2002,Human herpesvirus 6 and measles virus employdistinct CD46 domains for receptor functions. JBiol Chem 277:39112–39118.

Grummer B, Beer M, Liebler-Tenorio E, Greiser-Wilke I: 2001, Localization of viral proteins in cellsinfected with bovine viral diarrhoea virus. J GenVirol 82:2597–2605.

Grummer B, Bendfeldt S, Greiser-Wilke I: 2002b,Apoptosis inhibitors delay the cytopathic effect ofbovine viral diarrhoea virus (BVDV). J Vet Med B49:298–303.



Grummer B, Bendfeldt S, Wagner B, Greiser-Wilke I:2002a, Induction of the intrinsic apoptotic pathwayin cells infected with cytopathic bovine viral diar-rhoea virus. Virus Res 90:143–153.

Gu B, Liu C, Lin-Goerke J, et al.: 2000, The RNA he-licase and nucleotide triphosphatase activities of thebovine viral diarrhea virus NS3 protein are essen-tial for viral replication. J Virol 74:1794–1800.

Hegde RS, Voigt S, Lingappa VR: 1998a, Regulationof protein topology by trans-acting factors at theendoplasmic reticulum. Mol Cell 2:85–91.

Hegde RS, Voigt S, Rapoport TA, Lingappa VR:1998b, TRAM regulates the exposure of nascent se-cretory proteins to the cytosol during translocationinto the endoplasmic reticulum. Cell 92:621–631.

Hilgard P, Stockert R: 2000, Heparan sulfate proteo-glycans initiate dengue virus infection of hepato-cytes. Hepatology 35:1069–1077.

Hoff HS, Donis RO: 1997, Induction of apoptosis andcleavage of poly (ADP-ribose) polymerase by cyto-pathic bovine viral diarrhea virus infection. VirusRes 49:101–113.

Hulst MM, Himes G, Newbigin E, Moormann RJM:1994, Glycoprotein E2 of classical swine fevervirus: Expression in insect cells and identificationas a ribonuclease. Virology 200:558–565.

Hulst MM, Moorman RJM: 1997, Inhibition of pes-tivirus infection in cell culture by envelope proteinsErns and E2 of classical swine fever virus: Erns andE2 interact with different receptors. J Gen Virol78:2779–2787.

Hulst MM, Moorman RJM: 2001, Erns protein of pes-tiviruses. Meth Enzymol 342:431–440.

Iqbal M, Flick-Smith H, McCauley JW: 2000,Interactions of bovine viral diarrhea virus glycopro-tein Erns with cell surface glycosaminoglycans. JGen Virol 81:451–459.

Ito T, Lai MC: 1997, Determination of the secondarystructure of and cellular protein binding to the 3’-untranslated region of the hepatitis C virus RNAgenome. J Virol 71:8698–8706.

Jackson T, Ellard FM, Ghazaleh RA, et al.: 1996,Efficient infection of cells in culture by type Ofoot-and-mouth disease virus requires binding tocell surface heparan sulfate. J Virol 70:5282–5287.

Jacobs BL, Langland JO: 1996, When two strands arebetter than one: The mediators and modulators ofthe cellular responses to double-stranded RNA.Virology 219:339–49.

Johnson CM, Perez DR, French R, et al.: 2001, TheNS5a protein of bovine viral diarrhea virus inter-acts with the � subunit of elongation factor-1. J Gen Virol 82:2935–2943.

Jordan R, Nikolaeva OV, Wang L, et al.: 2002a,Inhibition of host ER glucosidase activity prevents

golgi processing of virion-associated bovine viraldiarrhea virus E2 glycoproteins and reduces infec-tivity of secreted virions. Virology 295:10–19.

Jordan R, Wang L, Graczyk TM, et al.: 2002b,Replication of a cytopathic strain of bovine viral di-arrhea virus activates PERK and induces endoplas-mic reticulum stress-mediated apoptosis of MDBKcells. J Virol 76:9588–9599.

Jungi TW, Schweizer M, Perler L, Peterhans E: 1999,Supernatants of virus-infected macrophages primeuninfected macrophages for lipopolysaccharide-induced apoptosis by both an interferon-dependentand an independent mechanism. Pathobiology67:294–297.

Jusa ER, Inaba Y, Kouno M, Hirose O: 1997, Effect ofheparin on infection of cells by porcine reproduc-tive and respiratory syndrome virus. Am J Vet Res58:488–491.

Kapoor M, Ramachandra M, Kusukawa J, et al.:1995, Association between NS3 and NS5 proteinsof dengue virus type 2 in the putative RNA repli-case is linked to differential phosphorylation ofNS5. J Biol Chem 270:19100–19106.

Kaufman RJ: 1999, Stress signaling from the lumenof the endoplasmic reticulum: coordination of genetranscriptional and translational controls. GenesDev 13:1211–1233.

Kibler KV, Shors T, Perkins KB, et al.: 1997, Double-stranded RNA is a trigger for apoptosis in vacciniavirus-infected cells. J Virol 71:1992–2003.

Kozak M: 1987, At least six nucleotides preceding theAUG initiator codon enhance translation in mam-malian cells. J Mol Biol 196:947–950.

Krusat T, Streckert JJ: 1997, Heparin-dependent at-tachment of respiratory syncytial virus (RSV) tohost cells. Arch Virol 142:1247–1254.

Lai VCH, Kao CC, Ferrari E, et al.: 1999, Mutationalanalysis of bovine viral diarrhea virus RNA-dependent RNA polymerase. J Virol73:10129–10136.

Lambot M, Hanon E, Lecomte C, et al.: 1998, Bovineviral diarrhoea virus induces apoptosis in bloodmononuclear cells by a mechanism largely depend-ent on monocytes. J Gen Virol 79:1745–1749.

Langedijk JPM: 2002, Translocation activity of C-terminal domain of pestivirus Erns and ribotoxinL3 loop. J Biol Chem 277:5308–5314.

Le S-Y, Sonenberg N, Maizel JV: 1995, Unusual fold-ing regions and ribosome landing pad within hepati-tis C virus and pestivirus RNAs. Gene 154:137–143.

Li W, Li Y, Kedersha N, et al.: 2002, Cell proteinsTIA-1 and TIAR interact with the 3’ stem-loop ofthe West Nile virus complementary minus-strandRNA and facilitate virus replication. J Virol76:11989–12000.



Li Y, McNally J: 2001, Characterization of RNA syn-thesis and translation of bovine viral diarrhea virus(BVDV). Virus Genes 23:149–155.

Lin C, Lindenbach BD, Pragai BM, et al.: 1994,Processing in the hepatitis C virus E2-NS2 region:Identification of p7 and two distinct E2-specificproducts with different C termini. J Virol68:5063–5073.

Ma K, Vattern KM, Wek RC: 2002. Dimerization andrelease of molecular chaperone inhibition facilitateactivation of eukaryotic initiation factor 2 (eIF2) ki-nase in response to endoplasmic reticulum stress. JBiol Chem 277:18728–18735.

MacKenzie JM, Westaway EG: 2001, Assembly andmaturation of the flavivirus Kunjin virus appear tooccur in the rough endoplasmic reticulum andalong the secretory pathway, respectively. J Virol75:10787–10799.

Martínez I, Melero JA: 2000, Binding of human respi-ratory syncytial virus to cells: implication of sul-fated cell surface proteoglycans. J Gen Virol81:2715–2722.

Matlack KES, Mothes W, Rapoport TA: 1998, Proteintranslocation: tunnel vision. Cell 92:381–390.

Maurer K, Himmelreich, Krey T, et al.: 2002, CD46serves as receptor for bovine viral diarrhea virus.Abstract 5th ESVV Pestivirus Symposium.

May P, Bock HH, Herz J: 2003, Integration of endo-cytosis and signal transduction by lipoprotein re-ceptors. Sci. STKE 2003 12.

McCullough KD, Martindale JL, Klotz L, et al.: 2000,Gadd153 sensitizes cells to endoplasmic reticulumstress by downregulating bcl2 and perturbing thecellular redox state. Mol Cell Biol 21:1249–1259.

McLauchlan J, Lemberg MK, Hope G, Martoglio B:2002, Intramembrane proteolysis promotes traffick-ing of hepatitis C virus core protein to lipiddroplets. EMBO J 15:3980–3988.

Mendelsohn CL, Wimmer E, Racaniello VR: 1989,Cellular receptor for poliovirus: molecular cloning,nucleotide sequence, and expression of a new mem-ber of the immunoglobulin superfamily. Cell56:855–865.

Mettenleiter TC, Zsak L, Zuckermann F, et al.: 1990,Interaction of glycoprotein gIII with cellular he-parin-like substance mediates adsorption ofpseudorabies virus. J Virol 64:278–286.

Minocha HC, Xue W, Reddy JR: 1997, A 50 kDamembrane protein from bovine kidney cells is a pu-tative receptor for bovine viral diarrhea virus(BVDV). Adv Exp Med Biol 412:145–148.

Moennig V, Bolin SR, Mateo A, et al.: 1988,Identification of a single receptor for bovine viraldiarrhea virus (BVDV) on the surface of culturedbovine cells. Abstract 88th Annual Meeting ASM.

Neill JD, Ridpath JF: 2003a, Gene expression changesin BVDV2-infected MDBK cells. Biologicals31:97–102.

Neill JD, Ridpath JF: 2003b, Gene expressionchanges in MDBK cells infected with genotype 2bovine viral diarrhoea virus. Vet Microbiol96:301–312.

Ng ML, Howe J, Sreenivasan V, Mulders JJL: 1994,Flavivirus West Nile (Sarafend) egress at theplasma membrane. Arch Virol 137:3030–313.

Okazaki K, Matsuzaki T, Sugahara Y, et al.: 1991,BHV-1 adsorption is mediated by the interaction ofglycoprotein gIII with heparin-like moiety on thecell surface. Virology 181:666–670.

Perler L, Schweizer M, Jungi TW, Peterhans E: 2000,Bovine viral diarrhoea virus and bovine her-pesvirus-1 prime uninfected macrophages forlipopolysaccharide-triggered apoptosis by inter-feron-dependent and –independent pathways. J GenVirol 81:881–887.

Pestova TV, Hellen CUT: 1999, Internal initiation oftranslation of bovine viral diarrhea virus RNA.Virology 258:249–256.

Pestova TV, Kolupaeva VG, Lomakin IB, et al.: 2001.Molecular mechanisms of translation initiation ineukaryotes. Proc Natl Acad Sci USA 98:7029–7036.

Pestova TV, Shatsky IN, Fletcher SP, et al.: 1998, Aprokaryotic-like mode of cytoplasmic eukaryotic ri-bosome binding to the initiation codon during inter-nal translation initiation of hepatitis C and classicalswine fever virus RNAs. Genes Develop 12:67–83.

Poole TL, Wang C, Popp RA, et al.: 1995, Pestivirustranslation initiation occurs by internal ribosomeentry. Virology 206:750–754.

Qu L, McMullan LK, Rice CM: 2001, Isolation andcharacterization of noncytopathic pestivirus mutantsreveals a role for nonstructural proteins NS4B inviral cytopathogenicity. J Virol 75:10651–10662.

Reed KE, Xu J, Rice CM: 1997, Phosphorylation ofthe hepatitis C virus NS5A protein in vitro and invivo: properties of the NS5A-associated kinase. JVirol 71:7187–7197.

Reed KE, Gorbalenya AE, CM Rice: 1998, TheNS5a/NS5 proteins from three genera of the familyFlaviviridae are phosphorylated by associated ser-ine/threonine kinases. J Virol 72:6199–6206.

Rice CM: 1996, Flaviviridae: The viruses and theirreplication, pp. 931–959. In: Fields Virology, 3rdedition, vol. 1. Eds. Fields BN, Knipe DM, HowleyPM, et al. Lippincott-Raven Publishers,Philadelphia, PA.

Rijnbrand R, van der Straaten T, van Rign PA, et al.:1997, Internal entry of ribosomes is directed by the5’ noncoding region of classical swine fever virusand is dependent onthe presence of an RNA



pseudoknot upstream of the initiation codon. J Virol71:451–457.

Roivainen M, Piirainen L, Hovi T, et al.: 1994, Entryof coxsackievirus A9 into host cells: Specific inter-actions with alpha v beta 3 integrin, the vitronectinreceptor. Virology 203:357–365.

Ruggli N, Tratschin J-D, Schweizer M, et al.: 2003,Classical swine fever virus interferes with cellularantiviral defense: Evidence for a novel function ofNpro. J Virol 77:7645–7654.

Rümenapf T, Thiel H-J, Moennig V, Müller K: 2000,CD46 is a receptor for the pestivirus BVDV onbovine cells. Am Soc Virol Mfg.

Rümenapf T, Unger G, Strauss JH, Thiel H-J: 1993,Processing of the envelope glycoproteins of pes-tiviruses. J Virol 67:3288–3294.


Schelp C, Greiser-Wilker I, Wolf G, et al.: 1995,Identification of cell membrane proteins linked tosusceptibility to bovine viral diarrhea virus infec-tion. Arch Virol 140:1997–2009.

Schneider R, Unger G, Stark R, et al.: 1993,Identification of a structural glycoprotein of anRNA virus as a ribonuclease. Science261:1169–1171.

Schweizer M, Peterhans E: 1999, Oxidative stress incells infected with bovine viral diarrhoea virus: Acrucial step in the induction of apoptosis. J GenVirol 80:1147–1155.


Sizova DV, Kolupaeva VG, Pestova TV, et al.: 1998,Specific interaction of eukaryotic translation factor3 with the 5’ nontranslated regions of hepatitis Cvirus and classical swine fever virus RNAs. Virol72:4775–4782.

Spahn CMT, Kieft JS, Grassucci RA, et al.: 2001,Hepatitis C virus IRES RNA-induced change in theconformation of the 40S ribosomal subunit. Science291:1959–1962.

Steller H: 1995, Mechanisms and genes of cellularsuicide. Science 267:1445–1449.

Su H, Liao C, Li Y: 2002, Japanese encephalitis virusinfection initiates endoplasmic reticulum stress andan unfolded protein response. J Virol76:4162–4171.

Su H, Lin Y, Yu H, et al.: 2001, The effect of hu-man bcl-2 and bcl-X genes on dengue virus-induced apoptosis of cultured cells. Virology282:141–153.

Tamburrini M, Scala G, Verde C, et al.: 1990,Immunosuppressive activity of bovine seminalRNase on T-cell proliferation. Eur J Biochem190:145–148.

Tautz N, Kaiser A, Thiel H-J: 1997, NS3 serineprotease of bovine viral diarrhea virus:Characterization of active site residues, NS4A co-factor domain, and protease-cofactor interactions.Virology 273:351–363.

Terry-Allison T, Montgomery RI, Warner MS, et al.:2001, Contributions of gD receptors and gly-cosaminoglycan sulfation to cell fusion mediatedby herpes simplex virus 1. Virus Res 74:39–45.

Teyssedou E, Magar R, Justewicz D M, Lecomte J:1987, Cell-protective monoclonal antibodies tobovine enterovirus-3 and partial or no activityagainst other serotypes. J Virol 61:2050–2053.

Tsukiyama-Kohara K, Iizuka N, Kohara M, NomotoA: 1992, Internal ribosome entry site within hepati-tis C virus RNA. J Virol 66:1476–1483.

Vassilev VB, Donis RO: 2000, Bovine viral diarrheavirus induced apoptosis correlates with increasedintracellular viral RNA accumulation. Virus Res69:95–107.

Warrener P, Collett MS: 1995, Pestivirus NS3 (p80)protein possesses RNA helicase activity. J Virol69:1720–1726.

Weiland F, Weiland E, Unger G, et al.: 1999,Localization of pestiviral envelope proteins Ernsand E2 at the cell surface and on isolated particles.J Gen Virol 80:1157–1165.

Wiskerchen M, Collett MS: 1991, Pestivirus gene ex-pression: Protein p80 of bovine viral diarrhea virusis a proteinase involved in polyprotein processing.Virology 184:341–350.

Wu, JZ: 2001, Internally located signal peptides directhepatitis C virus polyprotein processing in the ERmembrane. IUBMB Life 51:19–23.

Wykes MN, Shellam GR, McCluskey J, et al.: 1993,Murine cytomegalovirus interacts with major histo-compatibility complex class I molecules to estab-lish cellular infection. J Virol 67:4182–4189.

Xu J, Mendez E, Caron PR, et al.: 1997, Bovine viraldiarrhea virus NS3 serine protease: polyproteincleavage sites, cofactor requirements, and molecu-lar model of an enzyme essential for pestivirusreplication. J Virol 71:5312–5322.

Xue W, Minocha HC: 1993, Identification of the cellsurface receptor for bovine viral diarrhoea virus byusing anti-idiotypic antibodies. J Gen Virol74:73–79.

Xue W, Minocha HC: 1996, Identification of bovineviral diarrhea virus receptor in different cell types.Vet Microbiol 49:67–79.




Yu H, Isken O, Grassmann CW, Behrens S-E: 2000, Astem-loop motif formed by the immediate 5’ termi-nus of the bovine viral diarrhea virus genome mod-ulates translation as well as replication of the viralRNA. J Virol 74:5825–5835.

Zhang G, Aldridge S, Clarke MC, McCauley JW:1996, Cell death induced by cytopathic bovine viral

diarrhoea virus is mediated by apoptosis. J GenVirol 77:1677–1681.

Zhong W, Gutshall L, Del Vecchio AM: 1998,Identification and characterization of an RNA-dependent RNA polymerase activity within thenonstructural protein 5B of bovine viral diarrheavirus. J Virol 72:9365–9369.



INTRODUCTIONInfections with bovine viral diarrhea virus (BVDV)are endemic in many countries, leading to heavyeconomic losses for the cattle industry. Sweden wasone of the first countries to introduce a nationalBVDV control program in 1993, which now formsthe basis for control programs in many other coun-tries (Moennig and Greiser-Wilke, 2003). The pri-mary aim of BVDV control programs in Scandina-vian countries is the identification of BVDV-freeherds and prevention of reinfection of these herds sothat there is a gradual decrease in the number of in-fected herds. A crucial requirement for the successof these programs is the availability of rapid, eco-nomical, and simple diagnostic methods that arehighly sensitive and specific. In the Scandinavianmodel, different diagnostic tests are used for the de-tection of infection at various levels. Initially, bulktank milk is screened for BVDV or anti-BVDV an-tibody using a reverse transcription-polymerasechain assay (RT-PCR) or an enzyme-linked im-munosorbent assay (ELISA), respectively. If posi-tive, identification of individual, persistently in-fected (PI) animals is undertaken. For this purpose,an ELISA for the demonstration of virus in bloodhas proved reliable (Bitsch and Ronsholt, 1995).

The Scandinavian model may work well in coun-tries where cattle density is low and vaccination isnot allowed. It may not work, however, in countrieswhere both virus prevalence and cattle densities arehigh and where vaccination is permitted. Economiclosses in these countries can be minimized by low-ering the infection pressure (Moennig and Greiser-Wilke, 2003). No matter what control program isused, the identification and elimination of PI ani-mals is of the utmost importance (Schelp andGreiser-Wilke, 2003), which is possible only if thediagnostic tests are highly efficient and reliable.

The diagnosis of BVDV infection can sometimesbe made on the basis of history and clinical signs.However, clinical signs following BVDV infectionare highly variable depending on viral strain, age,and immune status of the animal; reproductive sta-tus of the animal; and the presence of other patho-gens. Thus, BVDV infection may result in subclini-cal acute infections; severe acute infectionscharacterized by fever, leukopenia, and thrombocy-topenia; persistent infections; reproductive diseasepresenting as congenital defects, repeat breeding,abortion, or mummification; enteric disease; respira-tory disease; and immunosuppression. Because somany different types of clinical presentation are as-sociated with BVDV infection, a diagnosis on thebasis of history, clinical signs, and postmortem ex-amination of dead animals can only be consideredpresumptive. Accurate and definitive detection ofBVDV infection depends on laboratory diagnosis.

The availability of accurate and rapid diagnostictests is necessary not only for control programs butalso for prognosis, monitoring, and epidemiology ofBVDV infection. Another area in which accurateBVDV diagnosis is important is the presence ofBVDV or its antibody in biologics of bovine origin,such as fetal bovine serum, cell cultures grown inBVDV-contaminated fetal bovine serum, and evenstocks of viruses prepared in BVDV-contaminatedcell cultures. For example, Nakamura et al. (1993)isolated noncytopathic (ncp) BVDV from three dif-ferent stocks of cytopathic (cp) BVDV using a re-verse plaque formation method based on intrinsicinterference phenomenon.

As stated above, fetal bovine serum (FBS) andcalf serum are extensively used in cell cultures asnonspecific nutrients and are often contaminatedwith BVDV (Bolin et al., 1991; Potts et al., 1989).The use of contaminated sera can result in contami-

197

12Diagnosis

Sagar M. Goyal


nation of cell cultures affecting the production of bi-ological reagents and the results of diagnosis. Bolinet al. (1994) examined 41 cell lines in the ATCC col-lection and found viral antigen or RNA in 13 ofthem by immunohistochemistry (IHC) and RT-PCR.The use of contaminated cells may result in contam-inated vaccines, which may lead to seroconversionor disease in the vaccinated animals or humans(Levings and Wessman, 1991; Wessman and Lev-ings, 1999). Hypervirulent outbreaks of BVDV inthe Netherlands and Italy were attributed to BVDVcontamination of bovine herpesvirus type 1 markervaccine (Falcone et al., 1999). The use of contami-nated serum or cell cultures may also interfere withthe diagnosis of viral infections by interfering withthe growth of other viruses.

Genital tracts are often obtained from abattoirs toharvest cumulus-oocyte complexes and coculture offeeder cells (oviduct epithelial cells and granulosecells) for use in vitro bovine embryo production sys-tems. A certain number of these tissues are likely tobe contaminated with BVDV and their use repre-sents a risk of BVDV transmission. Hence, all mate-rials of animal origin used in the production ofbovine embryos for in vitro fertilization should bescreened for BVDV (Givens et al., 2002).

It is important, therefore, to continually monitorFBS, cell cultures, seed viruses, and live vaccinesprepared in cell cultures for BVDV. Both cp and ncpstrains of BVDV should be looked for althoughmost of the contamination is associated with ncpstrains, which infect cells without morphological al-terations, inducing problems that arise after severalcell generations. Gamma irradiation, exposure to ul-traviolet light, and inactivation with beta propiolac-tone have been used to cure BVDV from FBS (Zabalet al., 2000).

A number of different tests are available for the de-tection of antigen, antibody, and viral components(antigen and nucleic acid) of BVDV. Each methodhas its advantages, disadvantages, and applicability.Factors that can affect the efficiency of a particulardiagnostic method include antigenic and/or geneticdiversity of the virus, variation in virus load, and in-terference from maternal antibodies obtained throughcolostrum. Methods using monoclonal antibodies(Mabs) can be used to differentiate pestiviruses(BVDV, border disease virus of sheep, and classicalswine fever virus of pigs). Monoclonal antibodiesprepared against NS2-3 protein are panpestivirus be-cause they recognize highly conserved epitopes(Mignon et al., 1992). Methods are also available toclassify BVDV into subtypes 1a, 1b, and 2.

DIRECT ANTIGEN DETECTIONMethods for direct antigen detection in clinical sam-ples are rapid and are often as sensitive as some ofthe other methods. However, the presence of viralantigen in tissues is often not associated with le-sions, particularly in subclinical and persistent in-fections. When lesions appear, they are seen prima-rily in lymphoid tissues, where the presence of viralantigen is associated with lymphoid depletion(Liebler-Tenorio et al., 2002). In persistent infec-tions and mucosal disease, virus can be isolatedfrom almost all tissues. Similarly, infection with avirulent BVDV 2 usually results in a widespreaddissemination of viral antigen in the host tissues.The tests that can be used for direct antigen detec-tion in fresh, frozen, or fixed tissues include ELISA,IHC (including immunoperoxidase staining of pe-ripheral blood leukocytes and skin biopsies), andimmunofluorescence.

IMMUNOFLUORESCENCE

In this procedure, cryostat sections of fresh tissuesor smears of buffy coat cells are stained with fluo-rescein-conjugated anti-BVDV antibody and thenexamined under a fluorescent microscope. The pres-ence of apple green fluorescence indicates a positivetest. This method is known as direct fluorescent an-tibody (DFA) or direct immunofluorescence test.Although DFA on buffy coat cells has been advo-cated for the detection of PI animals (Bezek et al.,1988), in our hands, the technique was not very suc-cessful (Werdin et al., 1989a).

IMMUNOHISTOCHEMISTRY OF PERIPHERALBLOOD LEUKOCYTES

An indirect immunoperoxidase test has been used todetect BVDV in smears of buffy coat cells (Saino etal., 1994). Deregt and Prins (1998) developed a Mab-based immunoperoxidase monolayer assay (IPMA)for the detection of BVDV and compared it with abovine polyclonal antibody (Pab)-based IPMA. A poolof five Mabs (four Mabs against BVDV 1 and oneMab against BVDV 2) was employed. These Mabswere chosen because of their broad cross-reactivity,antigenic avidity, reactivity to different BVDV pro-teins, and lack of competition for binding sites orbinding to unusual BVDV isolates. The Mab-IPMAwas found to outperform the Pab-IPMA in staining,ease of reading test results, and relative sensitivity.

IMMUNOHISTOCHEMISTRY OF SKIN BIOPSIES

Immunohistological testing of skin biopsies (earnotch samples) has been used to detect PI animals.



The technique can also be applied to dead animalsusing thyroid gland, skin, oral mucosa, esophagus,and abomasum as samples for IHC (Thur et al.,1996). The IHC staining of formalin-fixed, paraffin-embedded tissues is an efficient method for the de-tection of BVDV and is often considered to be bet-ter than histopathology (Haines and Ellis, 1994;Hewicker-Trautwein et al., 1995). In a study of 41cell lines for the presence of BVDV antigen orRNA, Bolin et al. (1994) found an excellent correla-tion between IHC and RT-PCR. The presence ofcolostrum-derived antibodies did not interfere withIHC of skin biopsy (Grooms and Keilen, 2002).

To compare IHC with virus isolation (VI),Grooms and Keilen (2002) screened samples from332 calves. Formalin-fixed skin biopsy sampleswere stained by IHC and virus was isolated frombuffy coat cells. Six calves were positive by bothtechniques. One was VI-positive and IHC-negativedue probably to acute infection since IHC does notdetect acute infections but VI does (Ridpath et al.,2002). In another study, skin from 41 of 42 calves,known to be PI by repeated virus isolation, werefound to be positive by IHC (Njaa et al., 2000).

The Mab used for IHC should be chosen carefullybecause only one of 32 Mabs against BVDV pro-teins and glycoproteins was able to detect BVDV informalin-fixed tissues by IHC (Haines et al., 1992).This Mab (designated 15C5) is widely employed todetect viral antigen by IHC (Baszler et al., 1995;Ellis et al., 1995).

ENZYME-LINKED IMMUNOSORBENT ASSAY

Many antigen-capture ELISAs have been devel-oped for the direct detection of BVDV antigen inbuffy coat cells, serum, and ear notch samples. Thebasic principle consists of the use of monoclonalantibodies to capture viral antigen followed by de-tection of antigen-antibody complex with enzyme-conjugated antibody (Bottcher et al., 1993;Ludeman and Katz, 1994). A test that can detect allcurrently circulating strains of BVDV is the mostdesirable. The most commonly used antigen cap-ture ELISA (AC-ELISA) uses Mab directed againsta conserved antigenic domain of a nonstructuralprotein (NS2/3) of pestiviruses. The captured anti-gen is then detected with a pestivirus-specific poly-clonal peroxidase conjugate (Gottschalk et al.,1992). Serum is a good sample for the detection ofPI animals by antigen-capture ELISA. Acute ani-mals are rarely detected because the virus is presentin the blood of an acutely infected animal only fora short time. Monoclonal antibodies against NS2-3

have been used in antigen capture ELISA tests.These tests have been found to yield results compa-rable to those of virus isolation (Sandvik andKrogsrud, 1995; Greiser-Wilke et al., 1992).Entrican et al. (1995) developed a double MabELISA for the detection of viral antigen in bloodsamples. Two Mabs against p125/p80 were used tocapture viral antigen from blood and another twoMabs were used to detect the captured antigen. TheMab ELISA was found to be more sensitive thanPab ELISA.

VIRUS ISOLATIONThe most reliable method for the detection ofBVDV infection has been the isolation of BVDV incell cultures followed by identification of the viralisolate by immunofluorescence or immunoperoxi-dase monolayer assay (IPMA; Meyling, 1984;Werdin et al., 1989a) or RT-PCR (Ridpath et al.,2002). During viremia, virus can be isolated fromnasal discharge, PBL, lungs, and feces. Semen,blood, serum, fetus, and feces can be used for virusisolation. However, the presence of anti-BVDV an-tibody may interfere with virus isolation from serumand buffy coat samples.

Many different cells of bovine origin support thegrowth of BVDV but bovine turbinate (BT) cells arethe most widely used for virus isolation becausethey are more sensitive to BVDV-induced cytopathiceffects, which makes it easier to differentiate cpfrom ncp strains.. A comparative study was carriedout to determine the susceptibility of five differentcell types to BVDV. The cell systems used wereswine testicle (ST), mink lung (ML), bovineturbinate (BT), porcine kidney (PK15), and equinedermal (ED) cells. The titers obtained on day 8postinfection were 101.13, 103.25, 104.13, 100.00, and100.00, in ST, ML, BT, PK15, and ED cells, respec-tively, indicating that BT and ML cells are optimalfor the propagation of BVDV (Onyekaba et al.,1987). In another study, primary bovine embryo kid-ney (pBEK) cells and two cell lines originating frombovine embryonic trachea (EBTr) and buffalo lung(IMR-31) were found to be equally susceptible toBVDV (Ferrari, 1985).

Isolated virus can be confirmed by DFA (directfluorescent antibody assay), immunoperoxidase, an-tigen capture ELISA, or RT-PCR. For DFA, infectedcells are rinsed in PBS and fixed in anhydrous ace-tone for 10 minutes. Fluorescein (FITC)-conjugatedanti-BVDV conjugate is allowed to react at 37°C for25 minutes. After rinsing lightly, the infected cellsand soaked in pH 9 carbonate-bicarbonate buffer

Diagnosis 199


with 0.05% Tween-20 for 10 minutes. The stainedcells are examined under a fluorescent microscope.Positive cells are characterized by the appearance ofapple green fluorescence. In immunoperoxidasetests, a peroxidase-labeled conjugate is used insteadof an FITC-labeled one and the stained cells are ex-amined by light microscopy.

Saliki et al. (1997) compared two techniques forthe identification of BVDV isolated in cell cultures.Serum samples were inoculated in indicator cellscontained in 96-well microtiter plates followed byimmunostaining of infected cells with a pool ofMabs using either immunoperoxidase monolayerassay (IPMA) or the monolayer enzyme-linked im-munosorbent assay (M-ELISA). In IPMA, positivesamples developed a red intracellular precipitate; ayellow color appears in solution in m-ELISA. Theoptimal time for staining was determined to be 4days postinoculation. Although both tests were sen-sitive and specific, the authors preferred M-ELISAbecause of its rapidity and greater objectivity. TheIPMA virus isolation-immunoperoxidase test (IPX)was also found to be sensitive for the detection ofBVDV in another study (Castro et al., 1997). The in-creased sensitivity of these tests is due to virus iso-lation component associated with them (Saliki et al.,1997). Pooled Mabs are often used in IPMA andELISA to detect BVDV after amplification in cellcultures because all Mabs do not give equivalent re-sults. For example, some of the E2-specific mono-clonal antibodies generated with BVDV 1 are cross-reactive with BVDV 2 and others are not (Deregtand Prins, 1998; Ridpath et al., 1994).

ANTIBODY DETECTIONAn indirect measure of virus infection is the detec-tion of virus-specific antibodies in the sera of ani-mals. Unfortunately, it is often difficult to differenti-ate among antibodies produced in response to acuteinfection, vaccination, or transfer of maternal anti-bodies from dam to offspring. In cattle, calves areusually born without antibody but seroconvert aftercolostrum consumption. These passive antibodieswane after 3–8 months. Hence, the presence of anti-body in colostrum-deprived calves can be due onlyto active infection (either in utero or postnatal) orvaccination. Seroconversion of sentinel animals canbe used as an evidence for possible exposure to PIanimals. Many tests are available for the detection ofanti-BVDV antibodies—namely, virus-neutralization(VN), indirect immunofluorescence (IIF) assay, indi-rect immunoperoxidase (IIP), and ELISA tests(Muvavarirwa et al., 1995).

VIRUS NEUTRALIZATION TEST (VN)The virus neutralization (VN), also known as serum-neutralization (SN), is considered to be the goldstandard test for the detection of anti-BVDV anti-bodies and is used worldwide (Rossi and Kiesel,1971). In this test, twofold serial dilutions of serumsample are incubated with a constant amount(200–500 TCID50) of the virus for 1 hour followedby the addition of indicator cells. The test is readafter 4–5 days of incubation at 37°C. The highest di-lution of the serum that inhibits virally induced cy-topathic effects in approximately 50% of inoculatedcells is considered to be the antibody titer of theserum. The test can be used for the detection of an-tibodies against BVDV 1 or BVDV 2 dependingupon the virus used in the test. In most situations, cpstrains of BVDV are used in the test so that the pres-ence of neutralizing antibodies can be detected byinhibition of viral infectivity as detected by the ab-sence of viral cytopathology. However, the test canalso be used with ncp strains, in which case the in-hibition of viral infectivity is measured by im-munoperoxidase staining of infected cells (Fulton etal., 1997).

Cross-neutralization tests can be used to charac-terize antigenic differences among pestiviruses(Dekker et al., 1995), and titers due to active infec-tion can be differentiated from vaccination titers bydemonstrating a fourfold rise in antibody titers usingpaired serum samples. Virus neutralizing antibodiesusually appear 3–4 weeks after infection and persistfor years. Titers induced by vaccination may alsopersist for a long time (Oguzoglu et al., 2003).Passive antibodies decline at 105–230 days (but maypersist for more than a year).

ENZYME-LINKED IMMUNOSORBENTASSAY

Various ELISA tests have been developed for the de-tection of anti-BVDV antibodies in serum samples.The antigens used in ELISA tests include wholevirus antigen, nonstructural protein, monoclonal an-tibodies, and peptides. Several factors can influencethe results of an ELISA test—e.g., antigen, conju-gated antibody, test sample, etc. (Schrijver andKramps, 1998). The procedure used to preparewhole virus antigen can also affect the specificityand sensitivity of the ELISA test. For example,Pilinkiene et al. (1999) found that antigens preparedby mild treatment showed the most specificity andactivity. Cho et al. (1991) prepared antigen fromMDBK-grown BVDV. The antigen was solubilized



with MEGA-10 (decanonyl-N-methylglucamide)followed by the removal of hydrophobic proteinswith Triton X-100 treatment. Compared to VN, thistest was 100% specific and 97% sensitive. Moenniget al. (1991) described the development of anELISA using the nonstructural protein p125/80 ofBVDV as antigen. The results were comparable tothose obtained by the VN test.

The specificity of serodiagnosis has been en-hanced greatly by the use of monoclonal antibody incompetitive ELISA systems, and further improve-ments are possible with the use of defined antigensderived by recombinant DNA techniques (Hainesand Ellis, 1994). Lecomte et al. (1990) prepared apanel of monoclonal antibodies that recognized 80kDa (NS3) antigen of cytopathic BVDV strains. Acompetitive ELISA was developed using mono-clonal antibodies and recombinant protein expressedin Escherichia coli as a fusion protein with ß-galac-tosidase. Beaudeau et al. (2001) developed a block-ing ELISA using monoclonal antibody against NS2-3 and used this test for mass screening of milk andserum samples with sensitivity and specificity of ap-proximately 97% as compared to VN.

Langedijk et al. (2001) developed a solid phaseantibody ELISA using peptides deduced from the C-terminal end (residues 191–227) of pestivirus enve-lope protein Erns. This ELISA was cross-reactive forseveral types of pestiviruses and could be used forgeneral detection of pestivirus antibodies. To detecttype-specific antibody, a liquid phase ELISA usinga labeled specific CSFV peptide, and an unlabeledBVDV peptide (to block cross-reactivity) was used.This test can potentially be used for the differentia-tion of vaccinated animals from infected ones if vac-cination is based on another envelope protein (E2).

A single serum dilution can be used in an ELISAtest to quantitate antibodies. Graham et al. (1997)standardized a commercial ELISA test for detectionof serum antibodies to BVDV so that a single serumdilution could be tested and the results expressedquantitatively using a standard curve. Various dilu-tions of known sera were tested and their endpointtiters calculated by an algebraic method directlyfrom a plot of each titration series and also from aregression line fitted to this plot.

IMMUNOPEROXIDASE ANDIMMUNOFLUORESCENCE

Indirect immunoperoxidase and indirect immuno-fluorescence tests have also been used for the detec-tion of anti-BVDV antibodies. However, ELISA andVN tests appear to have replaced them.

REVERSE TRANSCRIPTION-POLYMERASE CHAIN REACTION(RT-PCR) ASSAYSSeveral molecular diagnostic tests have been de-scribed for the detection of BVDV infection. Lewiset al. (1991) used 32P end-labeled, syntheticoligonucleotide probes as a tool for the detection ofBVDV RNA. A probe originating nearest the 5’ endof the viral RNA detected 86% of the 22 viral iso-lates tested. Brock (1991) compared DNA dot blothybridization and RT-PCR for the detection ofBVDV in serum and blood samples. RT-PCR wasfound to provide clearer identification of PI animalsthan DNA hybridization.

Due to its high sensitivity, RT-PCR is consideredas an alternative to current standard methods for de-tecting BVDV especially in pooled samples such asbulk tank milk. Because the nucleotide sequence ofBVDV is highly variable, it is important to carefullyselect and test primers (Ridpath et al., 1993). Thedetection of virus by RT-PCR has been found to bemore sensitive and rapid than virus isolation. In ad-dition, contrary to virus isolation, RT-PCR is not af-fected by the presence of antibodies in serum sam-ples. Weinstock et al. (2001) were able to detect thepresence of viral RNA when a single viremic sam-ple was pooled with 100 BVDV-negative sera.Horner et al. (1995) compared antigen captureELISA, RT-PCR, and virus isolation-IPMA. RT-PCR was the most sensitive and ELISA was theleast sensitive. The use of RT-PCR is not withoutdisadvantages, however. For instance, the test doesnot differentiate between nucleic acid from live orinactivated virus and may yield false positive re-sults. In a study by Givens et al. (2002), RT-PCRwas positive for BVDV for fetal bovine serum andprimary uterine tubal cells used for in vitro embryofertilization. However, no virus was detectable byvirus isolation, and the recipients of washed em-bryos did not seroconvert.

The prolonged stability of viral nucleic acid ascompared to the virus itself has led to a simplemethod for the collection, storage, transport, andtesting of blood samples. In this procedure, 10 µl ofblood or serum is applied to a Whatman No. 1 paper,the sample is air dried and then tested. Using thisprocedure, Vilcek et al. (2001) were able to detectviral RNA for up to 6 months by RT-PCR whilevirus isolation was positive for a few days only.

A prerequisite for the performance of RT-PCR isan efficient and simple method for sample prepara-tion. A study was conducted to compare the effi-ciency of three commercial kits for RNA extraction.

Diagnosis 201


The most sensitive RT-PCRs were obtained whensamples were prepared by acidic guanidinium-isothiocyanate-phenol-chloroform extraction withthe Trizol (Gibco) reagent. A kit based on the bind-ing of RNA to silica membrane in a spin column hadsomewhat lower sensitivity, and the kit using saltprecipitation of DNA and proteins was unsuitablefor the isolation of viral RNA (Scheibner et al.,2000).

Several single- and two-tube RT-PCR assays havebeen described for the detection of BVDV RNA inserum, buffy coat cells, and fresh and formalin-fixedtissues (including ear notches). Most of these testsare based on the detection of 5’ untranslated region(5’ UTR) and E2 gene of BVDV. Many different for-mats of RT-PCR have been described includingnested PCR (RT-nPCR) and multiplex PCR. Bothvirus isolation and ELISA tests are unreliable in thepresence of high levels of antibody in the sample butRT-PCR test is not affected (Zimmer et al., 2004).

For retrospective studies, it may be necessary toisolate and amplify viral RNA from formalin-fixed,paraffin-embedded tissues. Gruber et al. (1993) usedproteinase K digestion of deparaffinized tissue sec-tions followed by nPCR. This procedure consistentlydetected an 803 bp fragment of the gene coding forthe nonstructural protein p125 of BVDV and wasalso able to detect BVDV RNA from fresh brain tis-sue after 10 days of autolysis. Bhudevi and Wein-stock (2003) described a TaqMan test for the detec-tion of BVDV in formalin-fixed tissues. The viralnucleic acid was detectable in freshly fixed tissues aswell as tissues that were 7 years old. To determinethe effect of temperature on RNA degradation,freshly collected tissues from a PI animal werestored at 4°C or at room temperature prior to forma-lin fixation. Although there was a mild drop in signalstrength, tissues stored at 4°C were positive for up to1 week prior to formalin fixation, but those stored atroom temperature were positive until 74 hours only.

NESTED PCRGruber et al. (1994) recommended nested RT-PCRover single-step RT-PCR for archival studies becausethe former was found to be 100-fold more sensitivethan the latter. Deregt et al. (2002) described rPCR(conventional RT-PCR after RNA extraction) anddPCR (direct RT-PCR without RNA extraction) andfound that both were comparable to virus isolation.

TAQMAN

Open-tube RT-PCRs, in which PCR product is iden-tified by agarose gel electrophoresis, are subject to

contamination with amplicons and are not suitablefor high throughput of samples. Several one-tube,real-time formats of PCRs have been developed tominimize sample manipulation. These assays utilizefluorescent signals from oligonucleotide probes todetect amplification of nucleic acid. Hamel et al.(1995) described a simple, one-tube RT-PCR for therapid detection of BVDV. Total RNA was extracteddirectly from whole blood and tissue samples. Drewet al. (1999) also described a single step, single-tubeRT-PCR test to detect the presence of BVDV in so-matic cells from bulk milk samples. The test washighly specific and sensitive, detecting one PI ani-mal in a herd of 162 lactating animals.

Bhudevi and Weinstock (2001) developed a sin-gle-tube fluorigenic RT-PCR assay that was found tobe 10–100fold more sensitive than the two-tubeTaqMan and the standard single-tube RT-PCR as-says. They were also able to differentiate betweenBVDV 1 and BVDV 2 by using two different gene-specific labeled fluorigenic probes for the 5’ UTRregion. A single-tube RT-PCR, using panpestivirus324/326 primers targeting the 5’ UTR region wasdeveloped by Rossmanith et al. (2001).

Mahlum et al. (2002) reported a closed-tube for-mat of nucleic acid amplification and detection. Forthe development of TaqMan RT-PCR, the forwardand reverse primers to 5’ UTR (El-Kholy et al.,1998) were lengthened, and degeneracy at specificnucleotide sites was included to enable them to an-neal to a wider range of BVDV strains. Theseprimers (forward: 5’-GGGNAGTCGTCARTG-GTTCG-3’; reverse: 5’GTGCCATGTACAGCA-GAGWTTTT-3’) amplify a fragment of approxi-mately 190 bases in BVDV types 1 and 2. TheTaqMan probe (5’-6-FAM-CCAYGTGGAC-GAGGGCAYGC-TAMRA-3’) included a 6-FAM(6-carboxy-fluorescein) fluorescent reporter mole-cule at the 5’ end and a TAMRA (6-carboxy-tetramethyl-rhodamine) quencher molecule at the 3’end. This test was found to be more sensitive thanvirus isolation and IPMA in detecting BVDV in seraand more sensitive than virus isolation and IHC indetecting BVDV in tissue samples. This test is nowroutinely used at the Minnesota Veterinary Diagnos-tic Laboratory and is considered to be rapid, sensi-tive, and specific.

STRAIN DIFFERENTIATION

The antigenic and genetic variability of BVDV iso-lates has direct implications on sensitivity of BVDVdiagnosis. The diagnostic tests (such as ELISA andRT-PCR) should detect all prevailing strains of



BVDV in the area. A number of investigators havedeveloped RT-PCRs for the identification and differ-entiation of BVDV and classical swine fever virus.Sullivan and Akkina (1995) developed an nPCR forthe identification and differentiation of BVDV 1,BVDV 2, and BDV. Ridpath et al. (1994) designed aPCR to differentiate between types 1 and 2 ofBVDV. Later, Ridpath and Bolin (1998) described adifferential PCR (based on phylogenetic analysis of5’ UTR) for the segregation of types 1a, 1b, and 2 ofBVDV. Harpin et al. (1995) used PCR followed byrestriction endonuclease analysis of PCR product totype BVDV. Single-strand conformation polymor-phism (SSCP) analysis of polymerase chain reaction(PCR) products, a genetic screening technique forrapid detection of nucleotide substitutions in PCR-amplified genomic DNA or cDNA, has also beenemployed to identify and differentiate amongBVDV isolates (Jones and Weber, 2001).

Fulton et al. (1999) developed a 2-step RT-PCRassay for typing ruminant pestiviruses—e.g., BVDV1, BVDV 2, and border disease virus (BDV). Thefirst PCR (consensus PCR) detected all threeviruses. The amplification of the consensus PCRproduct by a second (nested) PCR, which uses type-specific primers, resulted in the differentiation of allthree viruses. Gilbert et al. (1999) described a nestedmultiplex PCR for genotyping of BVDV with orwithout RNA extraction.

Letellier and Kerkhofs (2003) developed a quan-titative real-time PCR for the detection and genotyp-ing of BVDV in genotypes 1 and 2. They used aprimer pair that was specific for highly conservedregions of the 5’ UTR and two TaqMan probes la-beled with FAM and VIC. The probe sequences dif-fered by three nucleotides, allowing differentiationbetween genotypes 1 and 2. Using this procedurethey detected 1000 and 100 copies of BVDV 1 andBVDV 2, respectively.

MILK AS A DIAGNOSTIC SAMPLEScreening of bulk milk samples for antibody or anti-gen is a promising tool for the detection of PI ani-mals. The milk is centrifuged to remove fat, andundiluted skim milk is tested most commonly in anELISA test. Because of the ease of obtaining sam-ple, milk sampling is preferable in large scale epi-demiological studies and for herd screening.According to the scheme of Alenius, milk samplescan be divided into four classes on the basis of anti-body levels. Samples belonging to class 0–1 arerarely virus-positive and those in class 2–3 are likelyto be positive (Meier et al., 2003). Commercial

ELISAs detecting antibodies against BVDV havebeen used as a screening test (Bottcher et al., 2003;Meier et al., 2003; Sandvik, 1999). Niskanen et al.(1991) found an excellent correlation between thelevel of antibodies in the bulk tank milk as detectedby an indirect ELISA and the prevalence of BVDVantibody-positive cows. Analysis of bulk milk sam-ples for BVDV antibodies is now routinely used inScandinavian countries as a tool in the diagnosis ofBVDV infections in dairy herds.

To minimize the number of animals tested, Houe(1994) conducted a study to determine whether 10young stock (8–18 months of age) from a herd can beused as a reliable screening test. The herds were di-vided into “slightly” to “heavily” infected if less than3 or more than 8 samples, respectively, were positivefor BVDV antibody. No PI animal was found in 24slightly infected herds and 5 of 18 heavily infectedherds were positive for PI animals in a follow-upwhole-herd test. In addition, bulk tank milk titerswere also generally higher in heavily infected herds.

RT-PCR assay to screen bulk milk samples forBVDV has proven to be a sensitive and economicmethod for the detection of a single PI animal withina group of several hundred cows (Renshaw et al.,2000) because virus titers are usually higher in milkthan in serum samples (Radwan et al., 1995). In thisprocedure, viral RNA is extracted from somaticcells purified from whole milk using a guanidiniumisothiocyanate and phenol/chloroform extractionmethod. Oligonucleotide primers are selected fromthe 5’ UTR and NS3 region of BVDV genome. RT-PCR of somatic cells was fifteenfold more sensitivethan virus isolation, and the presence of antibodiesdid not interfere with the test. Using RT-PCR onmilk samples, Drew et al. (1999) were able to detectone of 162 PI animals. In a comparative study,BVDV could be detected by both virus isolation andRT-PCR when milk from a single PI animal was di-luted 1:600 with milk from a herd of BVDV-negative animals. The correlation between the twoassays was 95.9% (Renshaw et al., 2000).

SCREENING FOR PERSISTENTLYINFECTED (PI) ANIMALSThe PI animals are considered “virus factories” fortheir herdmates. They shed large amounts of virusthroughout their lifetime and hence should be iden-tified and removed from the herd as soon as possi-ble. After removal of PI animals and cleaning of theherd, all “brought-in” stock should be isolated un-less found to be BVDV-free. Although PI animalsare an important source of virus on a farm (Duffel

Diagnosis 203


and Harkness, 1985), immunocompetent animalsthat undergo acute infections can also be a signifi-cant source of the virus and should not be ignored.Most of the diagnostic tests are designed and vali-dated based on the detection of PI animals. Sinceacute infections may also lead to positive reactionsfrom samples of serum, buffy coat, semen, and skinbiopsy, it is important to distinguish between acuteand persistent infections (Njaa et al., 2000; Ridpathet al., 2002). Confirmation of persistent infectioncan be done by obtaining positive results from twodifferent serum samples taken 3–4 weeks apart.

Immunohistochemical staining of ear notch sam-ples from calves is considered an effective methodto detect PI animals. The ear punch samples are easyto obtain and are relatively stable, and the test is notaffected by the presence of maternal antibodies.However, sample pooling is not possible. In addi-tion, tests based on the detection of viral antigen andnucleic acid (IHC and RT-PCR) do not yield a viralisolate that can be further characterized and studied.In a landmark study, Njaa et al. (2000) used IHCstaining of formalin-fixed, paraffin-embedded skinbiopsy specimens to detect the presence of PI ani-mals. Of the 42 samples that were shown to be per-sistently infected by repeated virus isolation, 41were positive by IHC of skin biopsy. Based on thesite of immunohistochemical staining, these authorswere able to distinguish between persistent andacute infections. Staining in PI animals was mostpronounced in keratinocytes, hair follicle epithe-lium, matrix cells of the hair bulb, and the dermalpapilla; in acute infection, small foci of stainingwere confined to the nonfollicular epidermis andfollicular ostia (Njaa et al., 2000).

Buffy coats (PBL) have been used for the isola-tion of BVDV for PI detection. In conventionalmethod of buffy coat isolation, heparanized blood issedimented, the plasma layer containing mononu-clear and polymorphonuclear leukocytes is aspi-rated, and then centrifuged at 160 xg for 5–10 min-utes. The cell pellet is suspended in 0.5–1.0 mL ofEagle’s MEM followed by inoculation in cell cul-tures for virus isolation. To maximize leukocyteyield, Ficoll-Paque/Macrodex (F-P/M) is often used.In this procedure, heparanized blood is diluted 1:1with saline solution and layered on FP solution in aratio of 3:1 and then centrifuged at 400 xg for 30minutes at room temperature. The mononuclear cellfraction lays on top of the FP gradient. This layer isremoved, washed twice in MEM followed by resus-pension in 2 mL of MEM containing 5% fetalbovine serum (Howell et al., 1979).

Serum is sometimes preferred as a specimen forvirus isolation because the virus survives longer inserum and because serum can be collected and storedeasily as compared to buffy coat cells (Saliki et al.,2000). However, the presence of antibodies in serummay interfere with virus isolation. In some instances,pooled samples of serum may be the specimen ofchoice. Serum pools are used in the voluntary controlprogram initiated in the German federal stateSaxony-Anhalt in 2002 (Gaede et al., 2003). Initially,pools of sera were tested with real-time RT-PCR in aLight-Cycler system. Individual PI animals in posi-tive pools were detected by using commercial Erns-antigen-ELISA. The prevalence of PI animals wasfound to be 0.2% during a 3-year period.

A prenatal test that can detect PI animals is notavailable although fetal fluids obtained percuta-neously from late stages of gestation have beentested by virus isolation (Callan et al., 2002). Unfor-tunately, the procedure is labor-intensive and is notvery safe. Lindberg et al. (2001) used an indirectELISA to identify dams pregnant with PI calves andclaimed good sensitivity and specificity after 7months of gestation.

Some practitioners use paired serology to screenfor PI animals. The procedure is to bleed a herd,identify animals with no or low VN titers, revacci-nate, and rebleed the whole herd. The animals thatdo not seroconvert in response to vaccination or thatdie from mucosal disease are considered to be PI an-imals. It must be emphasized, however, that this isnot a reliable method to screen for PI animals; if thevaccine strain is sufficiently different from the per-sisting strain, the PI animal may seroconvert to thevaccine strain and be declared non-PI although itstill is persistently infected (Werdin et al., 1989a,1989b; Ferreira et al., 2000). Such animal will notbe removed from the herd and will continue to be asource of the virus for its herdmates. Other factorscan also result in failure to detect PI animals—e.g.,maternal antibodies may interfere with virus isola-tion, virus load may not be enough for detection,and detection by PCR and ELISA may not be suc-cessful if the test is not universal.

SUMMARY AND CONCLUSIONSA tentative diagnosis of BVDV infection can oftenbe made on the basis of history, clinical signs, andpostmortem examination. For definite diagnosis,however, laboratory investigations using rapid andreliable tests are necessary. The methods for the de-tection of BVDV infections have greatly improvedrecently; several methods are available for the detec-



tion of virus, antibody, or viral components (antigenand nucleic acid). High throughput methods are nowavailable for whole herd screening to identify andremove PI animals. Bulk milk screening for BVDVor anti-BVDV antibody is used to identify infectedherds in Sweden. However, this procedure may notbe successful in countries with high cattle densitiesand in which vaccination is permitted.

Bulk milk and serology of young animals byELISA test is good. Herds with no or low antibodytiters in milk and negative serology in young ani-mals are considered free of PI animals. On the otherhand, high titers in milk with or without positivetiters in young animals is taken as the evidence thatthe herd is positive and that further investigationshould be undertaken to detect and remove PI ani-mals. The use of serum samples for RT-PCR and ofskin biopsies for IHC or RT-PCR has greatly im-proved BVDV diagnosis. Using these new tools, itshould now be possible to consider national controlprograms for BVDV.

REFERENCESBaszler TV, Evermann JF, Kaylor PS, et al.: 1995,

Diagnosis of naturally occurring bovine viral diar-rhea virus infections in ruminants using monoclonalantibody-based immunohistochemistry. Vet Pathol32:609–618.

Beaudeau F, Belloc C, Seegers H, et al.: 2001,Informative value of an indirect enzyme-linked im-munosorbent assay (ELISA) for the detection ofbovine viral diarrhoea virus (BVDV) antibodies inmilk. J Vet Med B Infect Dis Vet Public Health48:705–712.

Bezek DM, Baker JC, Kaneene JB: 1988, Immuno-fluorescence of bovine virus diarrhea viral antigenin white blood cells from experimentally infectedimmunocompetent calves. Can J Vet Res52:288–290.

Bhudevi B, Weinstock D: 2001, Fluorogenic RT-PCRassay (TaqMan) for detection and classification ofbovine viral diarrhea virus. Vet Microbiol 83:1–10.

Bhudevi B, Weinstock D: 2003, Detection of bovineviral diarrhea virus in formalin fixed paraffin em-bedded tissue sections by real time RT-PCR(Taqman). J Virol Methods 109:25–30.

Bitsch V, Ronsholt L: 1995, Control of bovine viraldiarrhea virus infection without vaccines. Vet ClinNorth Am Food Anim Pract 11:627–640.

Bolin SR, Matthews PJ, Ridpath JF: 1991, Methodsfor detection and frequency of contamination offetal calf serum with bovine viral diarrhea virus andantibodies against bovine viral diarrhea virus. J VetDiagn Invest 3:199–203.

Bolin SR, Ridpath JF, Black J, et al.: 1994, Survey ofcell lines in the American Type Culture Collectionfor bovine viral diarrhea virus. J Virol Methods48:211–221.

Bottcher J, Gottschalk E, Greiser-Wilke I, et al.: 1993,Diagnosis of bovine virus diarrhoea by two en-zyme-linked immunosorbent assays. Rev Sci Tech12:461–469.

Bottcher J, Meier N, Stengel KH, et al.: 2003,Feasibility of serological bulk milk testing as amethod for BVD surveillance. Berl Munch TierarztlWochenschr 116:244–251.

Brock KV: 1991, Detection of persistent bovine viraldiarrhea virus infections by DNA hybridization andpolymerase chain reaction assay. Arch Virol Suppl3:199–208.

Callan RJ, Schnackel JA, Van Campen H, et al.: 2002,Percutaneous collection of fetal fluids for the detec-tion of bovine viral diarrhea virus infection in cat-tle. J Am Vet Med Assoc 220:1348–1352.

Castro MD, Stoffregen WC, Brigman GP, Hillard KA:1997, A method to detect bovine viral diarrheavirus contamination in cell cultures using im-munoperoxidase staining. J Vet Diagn Invest9:427–431.

Cho HJ, Masri SA, Deregt D, et al. : 1991, Sensi-tivity and specificity of an enzyme-linked im-munosorbent assay for the detection of bovine viraldiarrhea virus antibody in cattle. Can J Vet Res55:56–59.

Dekker A, Wensvoort G, Terpstra C: 1995, Six anti-genic groups within the genus pestivirus as identi-fied by cross neutralization assays. Vet Microbiol47:317–329.

Deregt D, Carman PS, Clark RM, et al.: 2002, Acomparison of polymerase chain reaction with andwithout RNA extraction and virus isolation fordetection of bovine viral diarrhea virus in youngcalves. J Vet Diagn Invest 14:433–437.

Deregt D, Prins S: 1998, A monoclonal antibody-based immunoperoxidase monolayer (micro-isolation) assay for detection of type 1 and type 2bovine viral diarrhea viruses. Can J Vet Res62:152–155.

Drew TW, Yapp F, Paton D: 1999, The detection ofbovine viral diarrhoea virus in bulk milk samplesby the use of a single-tube RT-PCR. Vet Microbiol64:145–154.


Ellis JA, Martin K, Norman GR, Haines DM: 1995,Comparison of detection methods for bovine viraldiarrhea virus in bovine abortions and neonataldeath. J Vet Diagn Invest 7:433–436.

Diagnosis 205


El-Kholy AA, Bolin SR, Ridpath JF, et al.: 1998,Use of polymerase chain reaction to simultaneouslydetect and type bovine viral diarrhoea viruses iso-lated from clinical specimens. Rev Sci Tech17:733–742.

Entrican G, Dand A, Nettleton PF: 1995, A doublemonoclonal antibody ELISA for detecting pes-tivirus antigen in the blood of viraemic cattle andsheep. Vet Microbiol 43:65–74.

Falcone E, Tollis M, Conti G: 1999, Bovine viral diar-rhea disease associated with a contaminated vac-cine. Vaccine 18:387–388.

Ferrari M: 1985, Susceptibility of various cell culturesystems to bovine viral diarrhea virus. Microbio-logica 8:17–22.

Ferreira GM, Lourens DC, van Vuuren M: 2000, Theprevalence of bovine viral diarrhoea antibodies inselected South African dairy herds, and control ofthe disease. J S Afr Vet Assoc 71:10–13.

Fulton RW, d’Offay JM, Saliki JT, et al.: 1999, Nestedreverse transcriptase-polymerase chain reaction(RT-PCR) for typing ruminant pestiviruses: Bovineviral diarrhea viruses and border disease virus. CanJ Vet Res 63:276–281.

Fulton RW, Saliki JT, Burge LJ, et al.: 1997,Neutralizing antibodies to type 1 and 2 bovine viraldiarrhea viruses: Detection by inhibition of viralcytopathology and infectivity by immunoperoxidaseassay. Clin Diagn Lab Immunol 4:380–383.

Gaede W, Gehrmann B, Korber R: 2003, Eliminationof persistently BVDV infected animals: efficientherd screening using RT-PCR and antigen ELISAin milk and serum samples. Berl Munch TierarztlWochenschr 116:234–239.

Gilbert SA, Burton KM, Prins SE, Deregt D: 1999,Typing of bovine viral diarrhea viruses directlyfrom blood of persistently infected cattle by multi-plex PCR. J Clin Microbiol 37:2020–2023.

Givens MD, Riddell KP, Galik PK, et al.: 2002,Diagnostic dilemma encountered when detectingbovine viral diarrhea virus in IVF embryo produc-tion. Theriogenology 58:1399–1407.

Gottschalk EE, Greiser-Wilke I, Frey HR, et al.: 1992,An antigen capture test for the detection of cattleviremic with bovine viral diarrhoea virus—A com-parison with BVD virus isolation from buffy coatcells in bovine kidney cells. Zentralbl VeterinarmedB 39:467–472.

Graham DA, Mawhinney KA, McShane J, et al.:1997, Standardization of enzyme-linked im-munosorbent assays (ELISAs) for quantitative esti-mation of antibodies specific for infectious bovinerhinotracheitis virus, respiratory syncytial virus,parainfluenza-3 virus, and bovine viral diarrheavirus. J Vet Diagn Invest 9:24–31.

Greiser-Wilke I, Dittmar KE, Liess B, Moennig V:1992, Heterogeneous expression of the non-structural protein p80/p125 in cells infected withdifferent pestiviruses. J Gen Virol 73:47–52.

Grooms DL, Keilen ED: 2002, Screening of neonatalcalves for persistent infection with bovine viral di-arrhea virus by immunohistochemistry on skinbiopsy samples. Clin Diagn Lab Immunol9:898–900.

Gruber AD, Greiser-Wilke IM, Haas L, et al.: 1993,Detection of bovine viral diarrhea virus RNA informalin-fixed, paraffin-embedded brain tissue bynested polymerase chain reaction. J Virol Methods43:309–319.

Gruber AD, Moennig V, Hewicker-Trautwein M,Trautwein G: 1994, Effect of formalin fixation andlong-term storage on the detectability of bovineviral-diarrhoea-virus (BVDV) RNA in archivalbrain tissue using polymerase chain reaction.Zentralbl Veterinarmed B 41:654–661.

Haines DM, Clark EG, Dubovi EJ: 1992, Mono-clonal antibody-based immunohistochemicaldetection of bovine viral diarrhea virus in formalin-fixed, paraffin-embedded tissues. Vet Pathol29:27–32.

Haines DM, Ellis JA: 1994, Special tests for the diag-nosis of infectious causes of reproductive failure inruminants. Vet Clin North Am Food Anim Pract10:561–585.

Hamel AL, Wasylyshen MD, Nayar GP: 1995, Rapiddetection of bovine viral diarrhea virus by usingRNA extracted directly from assorted specimensand a one-tube reverse transcription PCR assay. JClin Microbiol 33:287–291.


Hewicker-Trautwein M, Liess B, Trautwein G: 1995,Brain lesions in calves following transplacental in-fection with bovine virus diarrhea virus. ZentralblVeterinarmed B 42:65–77.

Horner GW, Tham KM, Orr D, et al.: 1995,Comparison of an antigen capture enzyme-linkedassay with reverse transcription—polymerase chainreaction and cell culture immunoperoxidase testsfor the diagnosis of ruminant pestivirus infections.Vet Microbiol 43:75–84.

Houe H: 1994, Bovine viral diarrhoea virus: Detectionof Danish dairy herds with persistently infected ani-mals by means of a screening test of ten youngstock. Prev Vet Med 19:241–248.

Howell CL, Miller MJ, Martin WJ: 1979, Comparisonof rates of virus isolation from leukocyte popula-tions separated from blood by conventional and



Ficoll Paque/Macrodex methods. J Clin Microbiol10:533–537.

Jones LR, Weber EL: 2001, Application of single-strand conformation polymorphism to the study ofbovine viral diarrhea virus isolates. J Vet DiagnInvest 13:50–56.

Langedijk JP, Middel WG, Meloen RH, et al.: 2001,Enzyme-linked immunosorbent assay using a virustype-specific peptide based on a subdomain of en-velope protein E(rns) for serologic diagnosis ofpestivirus infections in swine. J Clin Microbiol39:906–912.

Lecomte C, Pin JJ, De Moerlooze L, et al.: 1990,ELISA detection of bovine viral diarrhoea virusspecific antibodies using recombinant antigen andmonoclonal antibodies. Vet Microbiol 23:193–201.

Letellier C, Kerkhofs P: 2003, Real-time PCR for si-multaneous detection and genotyping of bovineviral diarrhea virus. J Virol Methods 114:21–27.

Levings RL, Wessman SJ: 1991, Bovine viral diarrheavirus contamination of nutrient serum, cell culturesand viral vaccines. Dev Biol Stand 75:177–181.

Lewis TL, Ridpath JF, Bolin SR, Berry ES: 1991,Detection of BVD viruses using synthetic oligonu-cleotides. Arch Virol 117:269–278.

Liebler-Tenorio EM, Ridpath JE, Neill JD: 2002,Distribution of viral antigen and development of le-sions after experimental infection with highly viru-lent bovine viral diarrhea virus type 2 in calves. AmJ Vet Res 63:1575–1584.

Lindberg A, Groenendaal H, Alenius S, EmanuelsonU: 2001, Validation of a test for dams carryingfoetuses persistently infected with bovine viral-diarrhoea virus based on determination of antibodylevels in late pregnancy. Prev Vet Med 51:199–214.

Ludemann LR, Katz JB: 1994, Enzyme-linked im-munosorbent assay assessment of bovine viral diar-rhea virus antigen in inactivated vaccines usingpolyclonal or monoclonal antibodies. Biologicals22:2l–27.

Mahlum CE, Haugerud S, Shivers JL, et al.: 2002,Detection of bovine viral diarrhea virus by TaqManreverse transcription-polymerase chain reaction. JVet Diag Invest 14:120–125.

Meier N, Meier B, Banzhaf K, et al.: 2003, Antibodytiter against bovine viral diarrhea virus (BVDV) inbulk milk samples from Bavaria. Berl MunchTierarztl Wochenschr 116:240–243.

Meyling A: 1984, Detection of BVD virus in viremiccattle by an indirect immunoperoxidase technique.In: Recent Advances in Virus Diagnosis. Ed.McNulty MS, MacFerran JB, pp. 37–46. MartinusNijhoff, Boston, MA.

Mignon B, Waxweiler S, Thiry E, et al.: 1992,Epidemiological evaluation of a monoclonal ELISA

detecting bovine viral diarrhoea pestivirus antigensin field blood samples of persistently infected cat-tle. J Virol Methods 40:85–93.

Moennig V, Greiser-Wilke I: 2003, Perspectives onBVD eradication in Germany. Berl Munch TierarztlWochenschr 116:222–226.

Moennig V, Leder L, Greiser-Wilke I, et al.: 1991, Anew enzyme immunoassay for the detection of anti-bodies against the bovine viral diarrhea virus.Tierarztl Prax 19:35–38.

Muvavarirwa P, Mudenge D, Moyo D, Javangwe S:1995, Detection of bovine-virus-diarrhoea-virusantibodies in cattle with an enzyme-linked im-munosorbent assay. Onderstepoort J Vet Res62:241–244.

Nakamura S, Fukusho A, Inoue Y, et al.: 1993,Isolation of different non-cytopathogenic bovineviral diarrhoea (BVD) viruses from cytopathogenicBVD virus stocks using reverse plaque formationmethod. Vet Microbiol 38:173–179.

Niskanen R, Alenius S, Larsson B, Jacobsson SO:1991, Determination of level of antibodies tobovine virus diarrhoea virus (BVDV) in bulk tankmilk as a tool in the diagnosis and prophylaxis ofBVDV infections in dairy herds. Arch Virol Suppl3:245–251.


Oguzoglu TC, Frey HR, Eicken K, et al.: 2003,Kinetics and persistence of neutralizing anti-bodies against bovine viral diarrhea virus-1 and -2 and border disease virus after two step vaccina-tion of cattle. Dtsch Tierarztl Wochenschr110:14–17.

Onyekaba C, Fahrmann J, Bueon L, et al.: 1987,Comparison of five cell lines for the propagation ofbovine viral diarrhea and infectious bovine rhino-tracheitis viruses. Microbiologica 10:311–315.

Pilinkiene A, Pieskus J, Salomskas A: 1999,Obtaining immunologically active bovine viraldiarrhea virus (BVDV) proteins for enzyme linkedimmunosorbent assay (ELISA). Ukr Biokhim Zh71:59–64.

Potts BJ, Sawyer M, Shekarchi IC, et al.: 1989,Peroxidase-labeled primary antibody method fordetection of pestivirus contamination in cell cul-tures. J Virol Methods 26:119–124.

Radwan GS, Brock KV, Hogan JS, Smith KL: 1995,Development of a PCR amplification assay as ascreening test using bulk milk samples for identify-ing dairy herds infected with bovine viral diarrheavirus. Vet Microbiol 44:77–91.

Diagnosis 207





Ridpath JF, Bolin SR, Katz J: 1993, Comparison ofnucleic acid hybridization and nucleic acid amplifi-cation using conserved sequences from the 5’ non-coding region for detection of bovine viral diarrheavirus. J Clin Microbiol 31:986–989.

Ridpath JF, Hietala SK, Sorden S, Neill JD: 2002,Evaluation of the reverse transcription-polymerasechain reaction/probe test of serum samples and im-munohistochemistry of skin sections for detectionof acute bovine viral diarrhea infections. J VetDiagn Invest 14:303–307.

Rossi CR, Kiesel GK: 1971, Microtiter tests for de-tecting antibody in bovine serum to parainfluenza 3virus, infectious bovine rhinotracheitis virus, andbovine virus diarrhea virus. Appl Microbiol22:32–36.

Rossmanith W, Vilcek S, Wenzl H, et al.: 2001,Improved antigen and nucleic acid detection in abovine virus diarrhoea eradication program. VetMicrobiol 81:207–218.

Saino H, Watanabe H, Ikehata T: 1994, Immuno-peroxidase procedures for rapid detection of bovineviral diarrhea-mucosal disease virus antigen. J VetMed Sci 56:805–807.

Saliki JT, Fulton RW, Hull SR, Dubovi EJ: 1997,Microtiter virus isolation and enzyme immunoas-says for detection of bovine viral diarrhea virus incattle serum. J Clin Microbiol 35:803–807.

Saliki JT, Huchzermeier R, Dubovi EJ: 2000,Evaluation of a new sandwich ELISA kit that usesserum for detection of cattle persistently infectedwith BVD virus. Ann NY Acad Sci 916:358–363.

Sandvik T: 1999, Laboratory diagnostic investigationsfor bovine viral diarrhoea virus infections in cattle.Vet Microbiol 64:123–134.

Sandvik T, Krogsrud J: 1995, Evaluation of an antigencapture ELISA for detection of bovine viral diar-rhea virus in cattle blood samples. J Vet DiagnInvest 7:65–71.

Scheibner H, Frey HR, Eicken K, et al.: 2000,Detection of the genome of bovine viral diarrheavirus (BVDV) using the polymerase chain reactionafter reverse transcription (RT-PCR): Comparisonof methods for the isolation of ribonucleic acid(RNA) from clinical samples. Dtsch TierarztlWochenschr 107:431–437.

Schelp C, Greiser-Wilke I: 2003, BVD diagnosis: Anoverview. Berl Munch Tierarztl Wochenschr116:227–233.

Schrijver RS, Kramps JA: 1998, Critical factors af-fecting the diagnostic reliability of enzyme-linkedimmunosorbent assay formats. Rev Sci Tech17:550–561.

Sullivan DG, Akkina RK: 1995, A nested polymerasechain reaction assay to differentiate pestiviruses.Virus Res 38:231–239.

Thur B, Zlinszky K, Ehrensperger F: 1996, Immuno-histology as a reliable and efficient method for thediagnosis of BVDV infections. Schweiz ArchTierheilkd 138:476–482.

Vilcek S, Strojny L, Durkovic B, et al. 2001. Storageof bovine viral diarrhoea virus samples on filterpaper and detection of viral RNA by a RT-PCRmethod. J Virol Methods 92:19–22.

Weinstock D, Bhudevi B, Castro AE: 2001, Single-tube single-enzyme reverse transcriptase PCR assayfor detection of bovine viral diarrhea virus inpooled bovine serum. J Clin Microbiol 39:343–346.

Werdin RE, Ames TR, Goyal SM: 1989b. Detectionand elimination of carrier animals in a dairy herdpersistently infected with bovine viral diarrhea. JVet Diag Invest 1:277–279.

Werdin RE, Ames TR, Goyal SM, DeVries GP:1989a. Diagnostic investigation of bovine viral di-arrhea infection in a Minnesota dairy herd. J VetDiag Invest 1:57–61.

Wessman SJ, Levings RL: 1999, Benefits and risksdue to animal serum used in cell culture production.Dev Biol Stand 99:3–8.

Zabal O, Kobrak AL, Lager IA, et al.: 2000,Contamination of bovine fetal serum with bovineviral diarrhea virus. Rev Argent Microbiol32:27–32.

Zimmer GM, Van Maanen C, De Goey I, et al.: 2004,The effect of maternal antibodies on the detectionof bovine virus diarrhea virus in peripheral bloodsamples. Vet Microbiol 100:145–149.



INTRODUCTIONBovine viral diarrhea virus (BVDV) infects cattle ofdifferent age groups, from fetuses infected duringpregnancy through adult cattle. The virus is preva-lent in cattle populations worldwide and is responsi-ble for considerable economic losses—e.g., reduc-tion in feed conversion efficiency for meat and milkproduction, sick cattle, fatal diseases, and costs as-sociated with treatment and prevention includingvaccination.

Control programs for BVDV focus on three areas:

• Identification and removal of persistently in-fected (PI) cattle believed to be the major reser-voirs of infection

• Biosecurity measures including prevention ofBVDV exposure to herds believed to be BVDV-free through testing of new additions

• Vaccination of postnatal calves to protect themagainst disease after maternal antibody protec-tion is lost and of heifers/cows to prevent fetalinfections causing fetal losses and PI calves.

Vaccination to prevent fetal infections is likely themost important measures to control the occurrenceof PI calves.

Since the 1960s, considerable research has beendevoted to BVDV vaccines with numerous pub-lished studies. Two previous reviews offer extensivecoverage, and readers are referred to them for dis-cussion of earlier studies (van Oirschot et al., 1999;Bolin, 1995). In many BVDV vaccine studies, sheephave been used as a model because of their suscep-tibility to BVDV. The review by van Oirschot et al.(1999) covers numerous BVDV vaccine trials usingsheep. This chapter focuses on BVDV vaccines incattle, especially the efficacy of the vaccines andtheir impact on BVDV diversity.

Effective vaccines against infectious agents must

account for both antigenic and genetic diversity.There are two BVDV biotypes, cytopathic (cp) andnoncytopathic (ncp) based on the presence or ab-sence of visible cytopathic effects in infected cellcultures (Baker, 1995). There are two genotypes ofBVDV (BVDV 1 and BVDV 2) based on genomicnucleotide differences, which are detectable by ge-nomic sequence comparisons (Pellerin et al., 1984;Ridpath et al., 1984). Regions of the genome used indetecting genomic differences include the 5’ un-translated region (5’ UTR), NS (nonstructural pro-tein) region, and E2 (envelope glycoprotein (gp53)region (Collett, 1996). The BVDV 1 genotype canbe further separated into subgenotypes BVDV 1aand BVDV 1b (Ridpath and Bolin, 1998). Antigenicdifferences among these genotypes and subgeno-types are detected by differential virus neutraliza-tion tests (Fulton et al., 2003a; Jones et al., 2001).

Antigenic diversity in BVDV is a reflection of se-quence diversity in the regions of the genome cod-ing for the envelope protein E2, a glycoprotein (gp53) E1, and Erns (ribonuclease) protein (Collett,1996; Potgeiter, 1995). The differences in the 5’UTR are reflected in the E2 region (Pellerin et al.,1994; Van Rijn et al. 1997). Bruschke et al. (1999)reported that the E2 immunogens of BVDV 1a andBVDV 1b may not provide immunity to a heterolo-gous challenge. However, natural infections and theuse of killed or modified live virus (MLV) vaccinesdoes induce antibodies to several strains of BVDVincluding BVDV 1a, BVDV 1b, and BVDV 2 (Ful-ton et al., 1995; Cortese et al., 1998a; Jones et al.,2001; Fulton et al., 1997; Fulton and Burge, 2000;Fulton et al., 2000a; Fulton et al., 2002a; Groomsand Coe, 2002; Fulton et al., 2003a). However, vac-cines containing BVDV 1a were found to inducehigher antibody titers to the homologous BVDV 1astrains than to the heterologous BVDV 1b and

209

13Vaccines

Robert W. Fulton


BVDV 2 strains. Although BVDV genotype is im-portant in antigenic diversity, this does not appear tobe true of biotypes. Studies have shown no differ-ences in antibody titers when comparing cp and ncpstrains of BVDV (Fulton et al., 1997; Fulton andBurge, 2000).

The impact of BVDV diversity on clinical formsof the disease is evident when examining BVDV iso-lates from cases submitted by veterinarians to the di-agnostic laboratories (Fulton et al., 2000b). Of the105 BVDV positive samples (Fulton et al., 2000b),26 were BVDV 1 cp strains (24.8%), 38 were BVDV1 ncp strains (36.2%), 10 were BVDV 2 cp strains(9.5%), and 31 were BVDV 2 ncp strains (29.5%).The ncp biotype was isolated more frequently(65.7%) than the cp biotypes (34.3%), and BVDV 1genotype was more frequently isolated (61%) thanBVDV 2 genotype (39.0%). Cattle with respiratorydisease had more ncp biotypes than cp and moreBVDV 1 genotypes than BVDV 2. In cases of fibri-nous pneumonia, more BVDV 1 were isolated thanBVDV 2 and of the 41 BVDV 1 isolates, 68.3% wereBVDV 1b strains and 31.7% were BVDV 1a (Fultonet al., 2003a). In two studies of over 300 cattle com-mingled and observed for approximately 5 weeks,BVDV 1b was the most prevalent isolate (Fulton etal., 2002a). The results of these studies indicate thatdiverse BVDV strains (BVDV 1b, BVDV 1a, andBVDV 2) are found in various BVDV-associated dis-eases in the cattle population. BVDV 1a, BVDV 1b,and BVDV 2 were isolated from tissues of cattle re-ceiving BVDV 1a vaccines weeks to months prior(Fulton et al., 2000b; Van Campen et al., 2000; andFulton et al., 2003a). These BVDV strains may havebeen present due to two possible scenarios: theBVDV vaccinal strains did not induce protective im-munity to these diverse strains, these BVDV strainsmay be PI strains.

VACCINESOver 160 BVDV vaccines are listed in the Compen-dium of Veterinary Products (2003). These USDA li-censed vaccines are based either on BVDV alone orBVDV in combination with bovine herpesvirus-1(BHV-1), parainfluenza virus type 3 (PI-3), bovinerespiratory syncytial virus (BRSV), Leptospira spp.,Campylobacter spp., Haemophilus somnus, Mann-heimia haemolytica and/or Pasteurella multocidaimmunogens. These vaccines meet the requirementsfor safety, purity, potency, and efficacy by theUSDA’s Center for Veterinary Biologics (CVB)Code of Federal Regulations Sections 113.311(“Live virus vaccines,” “Bovine virus diarrhea vac-

cine”) and 113.215 (“Killed virus vaccines;”“Bovine viral diarrhea vaccine, killed virus”). Thisinformation is available at the USDA APHIS CVBwebsite (www. Aphis.usda.gov/vs/cvb/index.htm).

The requirements for efficacy of the killed orMLV vaccines include protection against clinical ill-ness after challenge with virulent BVDV. For MLVvaccines, the challenge occurs at 14–28 days post-vaccination with an observation period of 14 days.For killed vaccines, the challenge occurs 14 days ormore after the last dose of the vaccine is given.There is no requirement for the use of a specificBVDV subtype as a challenge in either of these effi-cacy requirements. In September 2002, the USDAAPHIS Center for Biologics Notice No. 02-19 pub-lished “Vaccine Claims for Protection of the FetusAgainst Bovine Viral Diarrhea Virus.” Three cate-gories for label claims were identified: “Aids in theprevention of abortion,” “Aids in prevention of per-sistently infected calves,” or “Aids in the preventionof fetal infection” or “Aids in the prevention of fetalinfection including persistently infected calves.” Aproduct label claim must be supported by researchdata filed by the vaccine manufacturing companyduring the licensing process.

BVDV vaccines have been available for use incattle for over 40 years in the U.S. These vaccines,either MLV or killed, have been an integral part ofbovine vaccination programs for beef breeding herd,stocker-feeder cattle for pasture forage or feedlots,dairy cattle, and veal operations. The merits of killedversus MLV vaccines have been debated over time.There are three important issues that have directedattention on the appropriate use of BVDV vaccines.First, in 1993 ncp BVDV strains with enhanced vir-ulence were isolated in Canada from cases of severeacute BVD disease in cattle (Carman et al., 1998).Analysis of the isolates indicated that they wereBVDV 2 strains. Cattle not properly vaccinated ac-cording to manufacturer’s instructions died with se-vere acute disease and hence efforts were made toincrease the usage of BVDV vaccines. Also, therecognition of antigenic differences between BVDV1 and BVDV 2 indicated the need for protection ofcattle against BVDV 2 either by BVDV 1 vaccine orby developing BVDV 2-specific vaccines. The sec-ond issue is the recognition that PI cattle, becausethey shed virus throughout their lifetime, are likelyreservoirs of infection for susceptible cattle. This fo-cused attention on the need to control fetal BVDVinfections by vaccination prior to exposure. Prior tothis, fetal protection studies were not a part of effi-cacy requirements. The third issue is that the ability



of the vaccines in use from the 1960s through the1990s to confer protection against BVDV 2 has notbeen established. The vaccines in use since the1960s had been, in retrospect, BVDV 1a cp strains,such as the Singer, NADL, and C24V (Oregon)strains. Questions then arose as to whether to acceptor prove that BVDV 1 vaccines protected againstBVDV 2 when used according to the label orwhether BVDV 2 strains should be added to theBVDV 1a vaccines.

VACCINE STRAINSVirus strains being used in various BVDV vaccinesare shown in Table 13.1. In 2002, the genotypeBVDV 2 was further divided into subgenotypes,BVDV 2a and BVDV 2b (Flores et al., 2002).Selected BVDV strains have been subtyped (J.F.Ridpath, personal communication, 2003) and thoseare reflected in the Table 13.1. The vaccines in theU.S. primarily contain BVDV 1a cp strains, al-though some vaccines do contain BVDV 2 cp strains(refer to Table 13.1). All but four vaccines containcp biotypes; only two killed and two MLV vaccinescontain ncp strains. Of the latter vaccines, one con-tains a ncp strain, WRL, referred to as BVDV 1without any reference to the subtype, and anothercontains a BVDV 1 ncp strain (GL 760) and a sub-type is not described. One killed vaccine contains

strain 6309, a BVDV 1 ncp strain whose subtype isnot specified. Only one killed vaccine contains ncpBVDV 1b. A dendrogram representing genetic relat-edness for reference BVDV strains, including manyvaccinal strains, is shown in Figure 13.1. There areseveral companies in the U.S. that manufacture vac-cines or single/multiple immunogens that are mar-keted by other companies. Thus, vaccines sold underdifferent names may have identical immunogens.

Isolation of BVDV 1b and BVDV 2 from cliniccases cited above demonstrates the need for BVDVvaccines that confer a broad spectrum of completeimmunity to these multiple BVDV subtypes. Theaddition of BVDV 2 immunogen to either MLV orkilled BVDV 1a vaccines has been the subject ofmarketing efforts by certain companies; some othercompanies, rather than including type 2 immuno-gens in their vaccine, have submitted data that satis-fied licensing requirements allowing a label claimfor cross protection against BVDV 2.

MLV VACCINESThe advantages and disadvantages of MLV BVDVvaccines are similar to those of other MLV vaccines.The MLV vaccines require lesser amounts of virusthan do killed vaccines because the vaccine repli-cates in the host to build immunogenic mass. In gen-eral, MLV vaccines require only one dose for initial

Vaccines 211

Figure 13.1. Dendogram representing the relatedness of nucleotide sequences from reference BVDV 1a, BVDV 1b,and BVDV 2 strains.


immunization but do require more rigid handlingprocedures because the vaccine virus is susceptibleto inactivation by chemicals and/or exposure tohigher temperatures. Upon administration of MLVvaccine, the BVDV vaccinal strains replicate in thesusceptible bovine, resulting in viremia (Cortese etal., 1997; Fulton et al., 2003c; Grooms et al., 1998).The duration of viremia is between 3 and 7 daysafter which the virus is cleared as the calves developantibodies (Fulton et al., 2003c). Thus vaccinalstrains must be differentiated from field strains ifBVDV is isolated from calves within 2 weeks afterMLV vaccination. Live virus in MLV vaccine mayalso cause immunosuppressive effects on leukocytefunction of vaccinated cattle (Roth and Kaberle,1983), potentially rendering them more susceptibleto other infections. Another concern is the effect oflive vaccine virus on the reproductive tract, becauseBVDV virus or antigen was detected in the ovariesof heifers receiving an MLV BVDV vaccine up to 30days prior to testing (Grooms et al., 1998).

A major concern for MLV BVDV use was the ob-servation of a postvaccinal disease following MLVvaccination. Within 1–4 weeks after MLV vaccina-tion, a mucosal disease (MD)–like syndrome oc-curred in cattle (Bittle, 1968; McKercher et al.,1968; Peter et al., 1967; Rosner, 1968; Lambert,1968). Postvaccinal MD may result when an animalpersistently infected with a ncp virus is exposed tothe cp BVDV included in the vaccine. However,vaccination of PI animals with an MLV vaccine willnot invariably cause MD. There have been multiplestudies whereby PI calves have been given MLVBVDV vaccines with cp strains and no MD resulted(Bolin et al., 1985b; Bolin et al., 1988; Fulton et al.,2003b). Those studies used cp strains that were anti-genically different from the PI strains, and thecalves developed antibodies to the cp strain in thevaccines. Thus PI animals may not be cleared fromherds by vaccination, and failure to develop anti-body titers following vaccinations is not a goodmethod to screen for PI animals.

In addition, postvaccinal disease can result frominfection with adventitious ncp field strains ofBVDV introduced via contaminated fetal bovineserum or cell lines used in vaccine manufacturing.The ncp BVDV is a frequent contaminant of fetalbovine serum (Bolin and Ridpath, 1998; Studer etal., 2002). An example of BVDV contamination ofan MLV vaccine occurred when vaccination with aBHV-1 marker vaccine with BVDV 2 contaminationcaused severe disease beginning 6 days after vacci-nation with high morbidity on 11 of 12 farms

(Barkema et al., 2001). Signs included nasal dis-charge, fever, and diarrhea. Necropsy revealed ero-sions and ulcers of mucosa of the digestive tract.

The ability of BVDV to infect the fetus resultingin abortions, stillbirths, and development defects hascaused MLV BVDV vaccines to be contraindicatedin pregnant cattle (Orban et al., 1983; Liess et al.,1984). In one study, MLV vaccine containing aBVDV 1a cp strain (C24V) was used in pregnantheifers resulting in fetal infection and disease. In ad-dition there were ncp strains isolated from affectedcalves diagnosed as PI. The source of the ncp strainswas not identified; however, a cp vaccine containingncp strains is possible. Another study indicated thatcp challenge inoculums may give rise to ncp strains(Done et al., 1980). Pregnant heifers were inocu-lated with a pool of BVDV cp strains (10), andtransplacental infections occurred with all ncpstrains being isolated, but no cp isolates.

Potentially a vaccine containing a ncp BVDVcontaminant could induce a PI calf if the fetus wereexposed between days 42–125 of gestation(McClurkin et al., 1984). Fetuses exposed in the lasttrimester of pregnancy could survive, but may haveantibody titers in the precolostral serums after calv-ing. Data suggest that MD was induced after recom-bination between a ncp BVDV 2 strain with aBVDV 1a cp NADL vaccine strain (Ridpath andBolin, 1995). The cp strains can cause fetal infec-tions; however, experimental inoculation of preg-nant heifers with a cp strain resulted in no PI calves(Brownlie et al., 1989). The potential for a replicat-ing vaccine virus or a contaminating ncp to be trans-mitted to susceptible contacts must be addressed forany MLV vaccine. Although an MLV BVDV mayreplicate in a susceptible animal causing viremia,virus may or not be shed. More importantly, theremust be sufficient virus shed to infect contacts as in-dicated by viremia and/or seroconversion. In a re-cent study (Fulton et al., 2003c) calves vaccinatedwith each of three BVDV 1a MLV vaccines devel-oped a transient viremia that was cleared after anti-bodies were induced; however, there was no trans-mission of the virus to susceptible contact calves inthe same pen. The use of ear notch immunohisto-chemistry (IHC) for BVDV antigen did not detectIHC positives in a preliminary study testing calvesafter they received MLV BVDV vaccine (Dubois etal., 2000).

KILLED VACCINESKilled vaccines also have advantages and disadvan-tages. From a production cost standpoint, killed vac-



cines are expensive because larger amounts of virusare required to prepare each dose of the vaccine ascompared to MLV vaccines and there is the addedcost of adjuvants. The process of virus inactivationfor the production of killed vaccine is likely to alsoinactivate possible contaminants if any; however,this is not guaranteed unless the final product istested for replicating virus. Killed BVDV vaccinesare generally safer in the pregnant cow, and somevaccination programs advocate killed BVDV vac-cines during pregnancy. One disadvantage might bethat two doses are generally required for the initialimmunization. Both MLV and killed BVDV vac-cines have induced antibodies to a wide range ofBVDV subtypes, usually resulting in higher anti-bodies to the specific BVDV subtype(s) in the re-spective vaccine (Fulton et al., 1997; Fulton andBurge, 2000).

The duration of antibody titers to various BVDVstrains was found to vary among different studies.Cortese et al. (1998a) reported that cattle receivingan MLV BVDV 1a (NADL) vaccine induced anti-bodies to numerous BVDV 1 and BVDV 2 strainsdetectable through 18 months after vaccination. Inother studies, there was a decline in BVDV antibod-ies by 140 days after initial vaccination (Fulton etal., 1995; Fulton and Burge, 2000). Revaccination atday 140 with either killed or MLV vaccine did in-duce increased antibodies in calves, especially thosewith low antibody titers (Fulton et al., 1995 andFulton and Burge, 2000). This rapid anamnestic re-sponse points out that, while antibody titers may de-cline or disappear, an improved immune responseremains in effect. From this standpoint, the true du-ration of immunity cannot be determined by justmeasuring serum antibody levels but should ratherbe determined by challenge with virulent virus.

PROTECTION BASED ONDISEASE FORMThe efficacy of a BVDV vaccine should be meas-ured by three different methods (Van Oirschot et al.,1999). The first is experimental vaccination undercontrolled conditions followed by direct experi-mental viral challenge and observation of clinicalsigns. Secondly, there should be transmission stud-ies to determine whether the vaccine prevents or re-duces the transmission of the challenge virus.Finally, field trials of the vaccine are needed to de-termine protection/cost benefits under productionconditions and natural exposure. To truly help dairyand beef cattle clinicians and producers make ra-tional decisions on appropriate vaccine use, the

studies outlined above should go beyond the mini-mal licensure requirements.

Recently the USDA APHIS CVB began requiringdata on file to support label claims. In an ideal set-ting, methods should be available for veterinariansand the public to analyze the documentation for ap-proved and marketed vaccines. All too frequently,the public’s only access to vaccine efficacy informa-tion is from marketing materials and advertisementsfor vaccines. Use of peer-reviewed publicationswould be an appropriate means to that end for pub-lic information. Intellectual property issues shouldbe honored, but the experimental design, results, sta-tistical analysis, and interpretation (discussion)should be available to allow producers and veteri-narians to make informed decisions.

MECHANISM OF PROTECTIONFOR BVDVCattle are capable of mounting both humoral (anti-body) and cell-mediated (T-cell) immune responsesto BVDV. These responses occur after vaccinationor field/natural infection of susceptible, seronegativecattle. As cited above, the vaccines induce antibod-ies to multiple BVDV subtypes, but antibody titersare generally higher to the vaccine strain and tostrains belonging to the same genotype or subgeno-type. Numerous vaccine efficacy/immunogenicitystudies in which antibody titers have been measuredare available in published literature. Unfortunately,in many cases, the authors do not identify the BVDVsubtype in either the vaccine, natural infection, orchallenge strain. The challenge strain used in thelaboratory for a virus neutralization test is also notalways known. Our current understanding of anti-genic diversity among genotypes and subgenotypescalls for more attention to subtypes in future immu-nity studies.

The fact that antibodies provide protection is clearfrom studies that demonstrate passive protectionagainst BVDV challenge in calves fed colostrumcontaining BVDV antibodies (Howard et. al., 1989;Bolin and Ridpath, 1995). Resistance either to dis-ease severity or viral infection/shedding was depend-ent upon BVDV antibody titers in the sera. Corteseet al. (1998) showed that calves fed colostrum withBVDV antibodies were protected against experimen-tal BVDV. However, high concentrations of mater-nally derived BVDV antibodies have been shown toblock a protective response to MLV BVDV 1aNADL vaccine (Ellis et al., 2001). Kirkpatrick et al.(2001) reported that dairy calves fed colostrum withBVDV antibodies had low BVDV antibody titers of

Vaccines 213


1:4–1:16 and did not seroconvert to either BVDV 1aor BVDV 2 after vaccination with an MLV BVDV 1aand BVDV 2 vaccine.

Postnatal calves develop T-cell immune responsesafter vaccination or infection. It is assumed thatthese calves can develop concurrent humoral anti-body and T-cell responses. Trying to determinewhich arm of the acquired immunity (humoral orcell-mediated) is more important in protection is dif-ficult because in the intact, healthy susceptible ani-mal, it is hard to immobilize one arm of the immunesystem to determine its role in protection. Beer et al.(1997) demonstrated that immunized and experi-mentally exposed calves responded with positive cy-totoxic T-cell responses, along with high neutraliz-ing antibody titers. More recently, Endsley et al.(2002) reported that calves developed antigen-specific T-cell responses when given either a BVDV1 MLV or BVDV 1 and BVDV 2 MLV vaccine. In-formation on humoral antibody responses to BVDV1 and BVDV 2 was not provided. Using a variety oftests for T-cell immunity and antibodies, the protec-tive immune mechanism was not apparent in calvesreceiving an MLV BVDV 1 vaccine and subse-quently challenged with heterologous BVDV 2(Cortese et al., 1998b).

Passively acquired BVDV antibodies preventedclinical disease in colostrum-fed calves exposed tovirulent BVDV (Ridpath et al., 2003). The serumantibody titers decayed at the same rate as in un-challenged colostrum-fed calves. These samecolostrums-fed calves previously challenged werestill protected from clinical disease after the serumantibodies had decayed to undetectable levels. Aprotective response was thus mounted in calveswith passive immunity but which was not reflectedby antibody titers. According to the authors, cell-mediated immunity may have a role in preventingpostnatal disease.

EFFICACY STUDIESIn a review of published studies for both postnatalcalves and fetal infection/disease, van Oirschot et al.(1999) cited numerous trials in both sheep and cat-tle for the prevention of fetal infections and trials incattle for preventing postnatal infections. In that re-view, numerous reports were cited indicating the useof sheep as a model for BVDV vaccine developmentand evaluation. Although sheep are an active, inex-pensive BVDV model, this chapter focuses on stud-ies of current North American vaccines in cattle.Selected other studies will be included that addressvaccine technology and experimental design, which

have worldwide relevance. Until recently, the impactof BVDV antigenic diversity on vaccines and chal-lenge models had not been addressed. In addition,only recently efforts have been made to study vac-cine protection against fetal infections. One of thefew early studies was done by Duffel et al. (1984)who reported that immunity to BVDV was inducedin heifers prior to breeding, and that their fetuseswere protected after challenge at day 100 of gesta-tion (Duffel et al., 1984).

This chapter updates previous published studieswith the current designation of subtypes. However, itshould be realized that a large database of all knownBVDV strains with respect to specific subtype desig-nations does not exist. Experimental studies areavailable that have evaluated both killed and MLVvaccines. The biotype/genotype in the vaccines untilrecently was confined to the BVDV 1a cp strains.Likewise, challenge strains have varied, as have ex-perimental challenge methods and parameters forevaluation of clinical illness. In general, the experi-mental challenge has been relatively soon after thelast vaccine dose, usually 14–28 days, and the dura-tion of immunity for the postnatal protection andfetal protection studies has not always and not uni-formly been performed. The general consensus forpostnatal protection is that there is incomplete pro-tection against clinical signs/disease. Brock andChase (2000) described an experimental protocol toevaluate fetal protection in cattle by BVDV vaccines.

PROTECTION AGAINST POSTNATALINFECTION/DISEASE

A combination killed BVDV vaccine containing aBVDV 1a cp strain and a BVDV 1 ncp strain (sub-type not designated in Table 13.1) given in twodoses induced BVDV antibodies, and vaccinateswere protected against clinical illness when com-pared to controls. The challenge virus, BVDV 1bncp New York (NY-1) strain, was given 14 days afterthe second dose. The antibody titers of vaccinates atthe time of challenge ranged between 1:16 and1:128 (mean = 1:42.7). The strain used in neutraliza-tion assay was not identified. The protection was notcomplete because some vaccinates had illness(Talens et al., 1989).

Calves given a MLV vaccine containing theBVDV 1a NADL strain were challenged with BVDV1b ncp NY-1 strain on postvaccination day 27. Noneof the vaccinated calves had clinical signs in the 14-day postchallenge period nor did they have rectaltemperatures above 39.7°C (Phillips et al., 1975).With the evident need for protection against BVDV



2 strains, there have been published studies whereBVDV 1-vaccinated cattle have been protected whenchallenged by BVDV 2. Cortese et al. (1998b) re-ported a study in which a BVDV 1a NADL vaccinalstrain was given to seronegative 10–14-day-old dairycalves and the calves then challenged intranasallywith a virulent BVDV 2 24515 ncp strain 21 dayslater. The antibody titers at the time of challenge

ranged from negative to 1:66 (mean = 1:15) toBVDV 1a NADL and negative to 1:6 (mean = 1:1) toBVDV 2 125C. Vaccinated calves were protectedagainst clinical signs and viremia in the 14-daypostchallenge observation period.

A related study by Ellis et al. (2001) indicated thatthe MLV BVDV 1a NADL vaccinal strain was givento 10–14-day-old or 4-month-old seronegative

Vaccines 215

Table 13.1. Strains and types of bovine viral diarrhea virus (BVDV) in modified-live virus (MLV)and killed vaccines.

Vaccine Type and Trade Name Name of Strain Genotype/Biotype Company

MLVArsenal 4.1 GL 760 1 ncp1 NovartisExpress 5 Singer 1a cp Boehringer Ingelheim Vet Medica

296 2a cp2

BoviShield 4 NADL 1a cp Pfizer Animal HealthPyramid 4 Singer 1a cp Fort Dodge Animal HealthReliant 4 NADL 1a cp MerialFrontier 4 Plus C24V 1a cp Intervet

296 2a cp2

Titanium 5 C24V 1a cp Agrilabs296 2a cp2

Jencine 4 WRL 1 ncp Schering-Plough Animal HealthHerd Vac 3 Singer 1a cp Biocor Animal Health

Bovine ViralDiarrhea Vaccine C24V 1a cp Colorado Serum Co.

KilledElite 4 Singer 1a cp Boehringer Ingelheim Vet MedicaHorizon 4 Plus C24 V 1a cp Intervet

125C 2a cpMaster Guard 5 C24 V 1a cp Agrilabs

125C 2a cpRespishield 4 Singer 1a cp MerialTriangle 4 + type II Singer 1a cp Fort Dodge Animal Health

5912 2a cpCattleMaster 4 5960 1a cp Pfizer Animal Health

6309 1 ncp

ViraShield 5 KY22 1a cp Grand LaboratoriesTN 131 2 ncp

Surround 4 Singer 1a cp Biocor Animal HealthNY 1b ncp

cp = cytopathic; ncp = noncytopathic 1Information provided by the company.2Based on sequencing information by Dr. J. F. Ridpath, USDA, ARS, NADC.


calves that were later challenged intranasally withthe virulent BVDV 2 24515 ncp strain at 4.5 monthsof age (3 weeks after vaccination of the 4-month-oldcalves and 4 months after vaccination of the 10–14-day-old calves). At time of challenge, the neutraliz-ing antibody titers for the 10–14-day-old seronega-tive calves vaccinated calves were the following:mean of 162 to BVDV 1a NADL (range of 108 to>162) and mean of 108 to BVDV 2 125C (range of18 to >162). The neutralizing antibody titers at chal-lenge for the 4-month-old calves vaccinated calveswere the following: mean of 36 (range of 18–54) toBVDV 1a NADL and mean of 18 (range of 12–108)to BVDV 2 125C. As in the prior study, the calveswere protected against clinical illness, and all butone vaccinated calf were protected against viremia.

An MLV vaccine containing the BVDV 1 ncpstrain (subtype not identified, Table 13.1) was givento seronegative calves, and the cattle were held forapproximately 7 months and then challenged in-tranasally with the BVDV 2 ncp 890 strain (Deanand Ley, 1999). The neutralizing antibody titers toBVDV 2 890 strain at time of challenge were thefollowing: intramuscular vaccinates, mean of 78(range 16–362) and subcutaneous vaccinates, meanof 73 (range of 23–181). Clinical signs were reducedin the vaccinated animals as compared to controls inthe 21-day postchallenge period. Vaccination elimi-nated nasal shedding in 87% of the cattle and com-pletely prevented viremia and leukopenia, whereasall unvaccinated cattle shed virus nasally and be-came viremic. This study was unique in that itdemonstrated there was a long duration of immunity(7 months) and the calves were protected againstBVDV viremia.

In a study by Howard et al. (1994), two inacti-vated vaccines were prepared using 11249 NC andKy 1203 NC using ß-propriolactone inactivation.Three doses of each vaccine were given on day 0,week 3, and week 6. The subtypes for these twostrains were not identified in the study. Subse-quently, studies refer to 11249 NC and Ky 1203 NCBVDV 1a ncp strains (Nobiron et al., 2003; J. Patel,personal communication). Three weeks later the11249 NC vaccinates were challenged with homol-ogous virus intranasally. The neutralizing antibodytiters in vaccinates to 1249 NC was a mean of 1318(3.12 log10). Nonvaccinated controls became vire-mic 4–6 days postchallenge and shed virus in na-sopharyngeal samples 4–8 days postchallenge; how-ever, the vaccinated animals did not have positivenasopharyngeal swabs or viremia. In the secondgroup, the Ky 1203 NC strain was given on days 0,

weeks 3 and 6, and cattle challenged on week 8 withthe heterologous 11249 NC strain. Again, vaccinateswere protected against viremia and nasopharyngealshedding. These studies support the point that killedvaccines, although requiring multiple doses, canprotect against viremia and nasopharyngeal shed-ding.

In another report, Makoschey et al. (2001) foundprotection by an inactivated BVDV 1 vaccine(Bovilis BVD) against clinical signs includingthrombocytopenia after challenge with a BVDV 2strain. This vaccine contained BVDV 1a cp strainC86 (Patel et al., 2002; J. Patel, personal communi-cation). Calves received 2 doses of the vaccine 4weeks apart and were challenged 4 weeks later withthe BVDV 2 Giessen-1 challenge strain(s), whichcontained both BVDV cp and ncp biotypes. The an-tibody titers in vaccinates at time of challenge usingthe BVDV 2 challenge virus in the serologic testranged from 5–10 log2 (1:32–1:1024). Calves weregiven the challenge virus both intranasally and intra-venously, and the observation period was 14 daysfor clinical signs. After challenge, unvaccinated cat-tle developed signs of respiratory disease, diarrheawith erosions and hemorrhage of the digestive tract,and depletion of lymphocytes in lymphatic organs.These signs were absent or markedly less severe invaccinates. The unvaccinated and vaccinated ani-mals were evaluated for thrombocytopenia andleukopenia for 22 days postchallenge. Beneficial ef-fects of vaccination indicated protection againstleukopenia and thrombocytopenia compared to un-vaccinated cattle. Vaccinated calves did not shedvirus after challenge and had reduced numbers ofBVDV isolates from plasma and blood cells. Thusthe inactivated BVDV 1 vaccine gave protectionagainst heterologous BVDV 2 infection and diseaseand reduced viral shedding and viremia.

In another study, two doses of an inactivated vac-cine containing BVDV 1 PT 810 strain and BVDV2 890 strain were given to cattle 28 days apart (Beeret al., 2000). The BVDV PT 810 strain is a BVDV1c strain (Giangaspero and Harasawa, 1999). Vacci-nated and nonvaccinated animals were challengedintranasally with the BVDV 1 PT810 strain 38 daysafter vaccination. The neutralizing antibody titers invaccinates at time of challenge ranged from1:10–1:8192 using the homologous virus BVDV 1cPT 810 in the serotest. Vaccinated cattle had reducedviral shedding in nasal samples, reduced leukopenia,and reduced viremia as compared to controls. This isanother example where an inactivated BVDV vac-cine given in multiple doses provided protection



after homologous challenge as measured by leuko-penia, viremia, and viral shedding.

A Japanese study recently demonstrated protec-tion by a Japanese MLV BVDV vaccine containingBVDV 1 ncp strain 12-43 in calves against a U.S.strain of BVDV 2 (Shimazaki et al., 2003). Fourweeks after vaccination, the vaccinated (n = 3) andunvaccinated calves (n = 2) were challenged intra-venously with the BVDV 2 890 ncp strain. The vac-cinated calves were seronegative to the BVDV 2 890strain at the time of challenge; however, the neutral-izing titers to BVDV 1 ncp were >1:64. The vacci-nated calves did not develop clinical signs or fevernor have hematological changes (decreased WBC)after challenge. The vaccinated animals were pro-tected against blood leukocyte infection. The au-thors suggested that other studies be performed toconfirm effectiveness of the vaccine against BVDV2 strains isolated in Japan.

PROTECTION AGAINST FETALINFECTION/DISEASE

In the 1990s attention was directed towards develop-ing new vaccines or evaluating existing vaccines toprotect the fetus against BVDV infection/disease.Until then vaccine licensure relied on efficacy ofprotection against acute disease or infection forpostnatal exposure of calves. Also label claims bymanufacturers required extensive, long-term studiesto generate supporting data for the label claim. Aninactivated vaccine, Bovidec (C-Vet) containing aCompton prototype virus, was given to heifers as ei-ther two or three doses each, 3 weeks apart near thetime of breeding (Brownlie, et al., 1995). The vac-cine strain in Bovidec vaccine is a BVDV 1a ncpstrain (J. Patel, personal communication). Pregnantheifers were exposed intranasally between 25 and80 days gestation with BVDV Pe 515 NC C1.Vaccinated heifers were protected against viremiapostchallenge. There was no evidence of infection inthe live calves born to vaccinated dams or in the 2aborted fetuses from the vaccinates. Thus vaccina-tion gave 100% protection to the 15 fetuses of 15vaccinated dams. There was, however, fetal infec-tion in 14/15 fetuses in the unvaccinated group as in-dicated by PI status of newborn calves, infectedaborted fetuses, or active immune response prior tobirth.

The MLV BVDV 1a NADL vaccine strain wasevaluated for protection of fetuses against BVDV 1strain (BJ ncp strain) (Cortese et al., 1998c). Twelveheifers were vaccinated once, and there were 6 non-vaccinates. Thirty days after vaccination the heifers

were synchronized with a prostaglandin product andthen housed with 10 BVDV-free bulls. They re-mained with bulls for 7 days. Approximately 110days after vaccination (approximately 75 days ges-tation) all 18 heifers were challenged intranasallywith the BJ ncp strain, a BVDV 1 a strain (K.V.Brock, personal communication). Mean neutralizingantibody titers to the BVDV 1a BJ strain and BVDV1a NADL strain at time of challenge was 1:40 (BJ)and 1:20 (NADL) for 10 vaccinates and 1:10 (BJ)and <1:5 (NADL) for 2 vaccinates. The vaccinatedanimals did not become viremic postchallenge, but4 of 6 nonvaccinated animals did 6–8 days afterchallenge. All 6 unvaccinated heifers carried theirfetuses to term and gave birth to PI calves as deter-mined by multiple virus isolations. Of the 12 vacci-nated heifers, there were 2 with PI calves. Thus fetalprotection was demonstrated against homologousBVDV 1 challenge in 10 of 12 (83.3%) of the vac-cinated heifers.

The ability of the BVDV 1a MLV NADL vaccineto provide protection against heterologous BVDV 2was evaluated (Brock and Cortese, 2001) among 19vaccinated and 6 unvaccinated heifers. Forty-fivedays after vaccination, the cattle were synchronizedwith prostaglandin and exposed to 10 BVDV-freebulls for 1 week. Thirty days later, the heifers wereexamined by ultrasound. All open animals weregiven a second dose of prostaglandin and were ex-posed to bulls again for 7 days. Thus challenge ap-pears to have been approximately 4–5 months aftervaccination. Mean neutralizing antibody titers invaccinated animals at the time of challenge was 1:36to both BVDV 1a (NADL) and BVDV 2 (PA 131).At 75 days gestation, the pregnant heifers were ex-posed intranasally to PA 131 strain, a ncp BVDV 2strain. After challenge, 5 of 6 nonvaccinates wereviremic (between days 5 and 8), but none of the 19vaccinates became viremic. The experiment was ter-minated at 150–180 days gestation, and spleen, thy-mus, and small intestine of fetuses were collectedfor virus isolation. All 6 fetuses from the nonvacci-nates were PI while 6 of 19 vaccinated dams had PIfetuses. Thus 11 of 19 (57.9%) pregnant heifersgiven a BVDV 1a NADL vaccine strain were pro-tected against PI. While protection was demon-strated against BVDV 2, it was not complete andwas less than protective against homologous BVDV1 (83.3%) demonstrated previously (Cortese et al.,1998c).

A two-step vaccination protocol, first dose withan inactivated BVDV vaccine (Mucobovin) fol-lowed by an MLV BVDV vaccine (Vacaviron), was

Vaccines 217


evaluated for fetal protection (Frey et al., 2002). Theinactivated BVDV vaccine contained BVDV 1b ncpNY-1 strain and a Border Disease virus strain,Aveyronite. The MLV vaccine contained the BVDV1a cp C24V strain. Heifers were vaccinated with thetwo-step vaccination program starting with the inac-tivated BVDV vaccine followed by MLV vaccine 4weeks later. Neutralizing antibody titers in vacci-nates 16 weeks after first vaccination indicatedBVDV 1 titers of 1:100–1:600, and BVDV 2 titers of1:10–1:240. Five months after the first vaccination(at 30–120 days of gestation), the heifers were chal-lenged with a mixture of BVDV 1 ncp 22146/Han 81 and BVDV 2 ncp (strain not identified). Atransient viremia was seen 5–9 days after challengein the nonvaccinated heifers, and BVDV 1 was iso-lated from 1 heifer and BVDV 2 from the remaining5. One vaccinate was positive for BVDV 2 on onecollection day, but the other vaccinates remainednonviremic. All vaccinated heifers (n = 9) gave birthto calves that were not viremic, did not have evi-dence of congenital fetal damage, and all pre-colostral sera were BVDV seronegative. The 6 calvesborn to nonvaccinates had evidence of fetal infec-tion; 1 was stillborn, 1 died 2 days after calving, and4 were underdeveloped and had signs of ocular de-fects. Precolostrum serum samples were positive forBVDV 2 by virus isolation. This study demonstratedthat a combination of killed BVDV 1b and BorderDisease virus vaccine followed by an MLV BVDV1a vaccine conferred protection against challengewith both BVDV 1 and BVDV 2, and also againsttransplacental infection with BVDV 2.

A unique study was performed to evaluate theability of an inactivated BVDV 1a vaccine to protectheifers exposed to PI animals 6 months after initialvaccination (Patel et al., 2002). The use of PI ani-mals to introduce challenge virus mimics the mostlikely route of exposure under field conditions. Twodoses of an inactivated vaccine (Bovilis BVDV)containing BVDV 1a cp strain C 86 were given to 20heifers 4 weeks apart. Approximately 100 days later,the heifers were synchronized followed by two serv-ices. There were 11 vaccinated and 7 unvaccinatedanimals. Six months after the second vaccination (atabout 87 days of gestation) the heifers were chal-lenged. The neutralizing antibody titers in vacci-nates at the time of challenge to BVDV 1a rangedfrom 1:64–1:256. Three PI heifers were introducedinto the pen for 2 weeks to initiate challenge. The PIcattle were infected with a BVDV 1a ncp strain (J.Patel, personal communication). All 7 nonvacci-nated heifers became viremic, as detected by virus

isolation from serum and/or leukocytes. Only 5 of11 vaccinated calves were found to be viremic as de-termined by viral isolation from leukocytes; all werenegative for virus in serum. Virus was detected innasal samples of 5 of 7 nonvaccinates and 2/11 vac-cinates. The vaccinated heifers had two abortions(not related to BVDV) and 9 normal calves. These 9calves were negative for BVDV and BVDV anti-bodes in precolostral serum. All 7 nonvaccinatedheifers delivered BVDV-infected calves. This studydemonstrated that two doses of an inactivatedBVDV 1a vaccine gave complete protection againsthomologous viral challenge of pregnant heifers 6months after vaccination.

The results of these trials evaluating efficacy ofBVDV vaccines under challenge conditions indicateprotection for both postnatal acute infections/dis-ease and fetal infections. Challenge in many trialswas relatively soon after vaccination, 2–4 weeksafter last dose of initial immunization, which is pre-sumably at a peak of immunity. In other trials, thecattle were challenged approximately 6 months aftervaccination. Interestingly, all of the above studiesindicated protection against postnatal acute disease.In some, but not all studies, vaccination eliminatedviremia and nasal shedding in postchallenge vacci-nated animals. Thus, it may be difficult to totallyeliminate nasal shedding from the respiratory tractor viremia (systemic infection) in even highly im-mune cattle after challenge. It is important that anyBVDV vaccine protects against viremia, becauseviremia indicates the potential for fetal infections.Likewise, elimination of nasal shedding in exposedvaccinated cattle is important to prevent furthertransmission of BVDV. There appears to be in-creased interest in protection against differentBVDV subtypes, particularly against BVDV 2.

The protection against fetal infection/disease var-ied from approximately 60–100%. Fetal protectionis extremely important to prevent or minimize PIcalves. Elimination of PI cattle will likely reduceBVDV transmission. In the two reports using theMLV BVDV 1a vaccinal strain there was protection(83%) against homologous BVDV 1a challenge;however, there was reduced protection against het-erologous BVDV 2 challenge, (58%). Although pro-tection was demonstrated, the MLV BVDV 1a vac-cine did not provide 100% protection. Thus eventhough the vaccines may have efficacy demonstratedfor licensure, vigilance for PI cattle in vaccinatedcattle is important because there may be potentialfor PI calves born to vaccinated cows/heifers.

Up to this point, this review has focused on killed



and modified live whole virus vaccines produced byclassic methods. However, several experimentalvaccines for delivering BVDV immunogens havebeen reported in the literature. Predominantly thesevaccines focus on the delivery of the E2 glycopro-tein by various methods, including recombinant pro-tein from baculovirus constructs (Bolin and Ridpath,1996; Bruschke et al., 1997), expression as con-stituent of recombinant adenovirus (Elahi et al.,1999) and as a DNA construct (Harpin et al., 1999;Norbiron et al. 2003).

GOALS FOR VACCINEDEVELOPMENTThere are significant goals for BVDV vaccines that,if accomplished, would greatly provide better con-trol measures for BVDV. The first involves fetal pro-tection. In order to completely eliminate the birth ofPI calves, vaccines need to provide 100% lifetimeprotection of pregnant heifers/cows against viremia.The second goal involves the prevention of acute in-fection in nonpregnant animals. Acute BVDV dis-ease remains a problem for both calf and dairy pro-duction. Stocker/feeder beef production operationsare faced with acute disease caused by BVDV(Fulton et al., 2000a; Fulton et al., 2002a). The goalof vaccination in these animals would be to elimi-nate clinical signs and virus shedding. In addition,current management practices require a rapid onsetof protective immunity for vaccination to be effec-tive. The marketing of beef cattle in North Americatypically involves assembly of calves from diversesources, resulting in comingling of calves with un-known vaccination status. Likewise, many cattle ofunknown vaccination status are delivered directly tofeedlots. Cattle are usually vaccinated againstBVDV along with other viral immunogens shortlyafter purchase at auctions and/or delivery to thefeedlot. The period immediately after comingling iscritical for exposure to pathogens.

Current BVDV vaccine onset of immunity has notbeen thoroughly investigated, although efficacy tri-als have been successful when challenge occurred14 days after the last dose of initial immunization.The duration of immunity should be established foreach vaccine, particularly those with a fetal protec-tion label claim. For example, beef breeding cattle,particularly under range conditions may not alwaysbe able to be gathered for vaccinations due to largepastures; hence the need for vaccines inducingdurable immunity.

There is an ongoing debate regarding the impor-tance of BVDV vaccines in controlling acute respi-

ratory disease in postnatal calves. A review of fieldstudies on vaccine efficacy in bovine respiratory dis-ease did not demonstrate benefits by BVDV vac-cines (Perino and Hunsaker, 1997). In contrast, alater study demonstrated that BVDV immunity wasa predictor of illness and performance parameters infeedlot calves (Fulton et al., 2002b). In that study,there was a correlation between higher levels ofBVDV 1a antibodies and lower morbidity rate.Calves with low antibody levels to BVDV 1a andBVDV 2 had decreased net value to owners (carcassvalue minus total feedlot costs). Calves treated twiceor more had lower levels of antibody to BVDV 1athan those treated once or not at all. Thus, BVDVimmunity appears to demonstrate benefit againstdisease and for increased profitability. Herds withhigh morbidity and treatment costs are often shownto follow incomplete vaccination programs.

Finally, there are safety issues that should be ad-dressed. These include assurances that MLV vac-cines do not cross the placenta and infect the fetus.The bovine fetus is very susceptible to the field orvaccinal strains of BVDV. In addition, there alsoneeds to be better efficacy studies to heterologousBVDV subtypes.

REFERENCESBaker JC: 1987, Bovine viral diarrhea virus: a review.

J Am Vet Med Assoc 190:1449–1458.Barkema HW, Bartels CJ, van Wuijckhuise L, et al.:

2001, Outbreak of bovine virus diarrhea on Dutchdairy farms induced by a bovine herpesvirus 1marker vaccine contaminated with bovine virusdiarrhea virus type 2. Tijdschr Diergeneeskd126:158–165.

Baxi MK, Deregt D, Robertson J, et al.: 2000, Re-combinant bovine adenovirus type 3 expressingbovine viral diarrhea virus glycoprotein E2 inducesan immune response in cotton rats. Virology278:234–243.

Beer M, Hehnen HR, Wolfmeyer A, et al.: 2000, Anew inactivated BVDV genotype I and II vaccine.An immunization and challenge study with BVDVgenotype I. Vet Microbiol 77:195–208.

Beer M, Wolf G, Pickler J, et al.: 1997, Cytotoxic T-lymphocyte responses in cattle infected with bovineviral diarrhea virus. Vet Microbiol 58:9–22.

Bittle, JL.: 1968, Vaccination for bovine viral diarrhea-mucosal disease. J Am Vet Med Assoc 152:861.

Bolin SR: 1988, Viral and viral protein specificity ofantibodies induced in cows persistently infectedwith noncytopathic bovine viral diarrhea virus aftervaccination with cytopathic bovine viral diarrheavirus. Am J Vet Res 49:1040–1044.

Vaccines 219


Bolin SR: 1995, Control of bovine viral diarrhea in-fection by use of vaccination. Vet Clin North Am:Food Anim Pract 11:615–625.

Bolin SR, McClurkin AW, Cutlip RC, Coria MF:1985a, Severe clinical disease induced in cattlepersistently infected with noncytopathic bovineviral diarrhea virus by superinfection with cyto-pathic bovine viral diarrhea virus. Am J Vet Res46:573–576.

Bolin SR, McClurkin AW, Cutlip RC et al: 1985b,Response of cattle persistently infected with noncy-topathic bovine viral diarrhea virus to vaccinationfor bovine viral diarrhea and to subsequent chal-lenge exposure with cytopathic bovine viral diar-rhea virus. Am J Vet Res 46:2467–2470.

Bolin SR, Ridpath JF: 1995, Assessment of protectionfrom systemic infection or disease afforded by lowto intermediate titers of passively acquired neutral-izing antibody against bovine viral diarrhea virus incalves. Am J Vet Res 56:755–759.

Bolin SR, Ridpath JF: 1996, Glycoprotein E2 ofbovine viral diarrhea virus expressed in insect cells provides calves limited protection from sys-temic infection and disease. Arch Virol141:1463–1477.

Bolin SR, Ridpath JF: 1998, Prevalence of bovineviral diarrhea virus genotypes and antibody againstthose viral genotypes in fetal bovine serum. J VetDiagn Invest 10:135–139.

Brock KV, Chase CCL: 2000, Development of a fetalchallenge method for the evaluation of bovinediarrhea virus vaccines. Vet Microbiol 77:209–214.

Brock KV, Cortese VS: 2001, Experimental fetal chal-lenge using type II bovine viral diarrhea virus incattle vaccinated with modified-live virus vaccine.Vet Therapeutics 2:354–360.

Brownlie J, Clarke MC, Hooper, LB et al.: 1995,Protection of the bovine fetus from bovine viral di-arrhoea virus by means of a new inactivated vac-cine. Vet Rec 137:58–62.


Brownlie J, Clarke MC, Howard CJ: 1989, Experi-mental infection of cattle in early pregnancy with acytopathic strain of bovine viral diarrhoea virus.Res Vet Sci 46:307–311.

Bruschke, CJM, Moormann RJM, van Oirschot JT,van Rijn PA: 1997, A subunit vaccine based on gly-coprotein E2 of bovine viral diarrhea virus inducesfetal protection in sheep against homologous chal-lenge. Vaccine 15:1940–1945.

Bruschke CJM, van Oirschot JT, van Rijn PA: 1999,An experimental multivalent bovine virus diarrheavirus E2 subunit vaccine and two experimental con-

ventionally inactivated vaccines induce partial fetalprotection in sheep. Vaccine 17:1983–1991.


Collett MS: 1996, Genomic structure of BVDV. In:Bovine Viral Diarrhea Virus: A 50 year review.Cornell University Press, Ithaca, NY.

Compendium of Veterinary Products: 2003, SeventhEdition. North American Compendiums, PortHuron, MI, pp. 302–309.

Cortese VS, Ellis J, Whittaker R: 1997, BVDV virustransmission following attenuated vaccines toBVDV seronegative cattle. Large Animal Pract,September/October pp. 18–24.

Cortese VS, Grooms DL, Ellis J, et al.: 1998c, Protec-tion of pregnant cattle and their fetuses against in-fection with bovine viral diarrhea virus type 1 byuse of a modified-live virus vaccine. Am J Vet Res59:1409–1413.

Cortese VS, West KH, Hassard LE, et al.: 1998b,Clinical and immunological responses of vacci-nated and unvaccinated calves to infection with avirulent type II isolate of bovine viral diarrheavirus. J Am Vet Med Assoc 213:1312–1319.

Cortese VS, Whittaker R, Ellis J, et al: 1998a,Specificity and duration of neutralizing antibodiesinduced in healthy cattle after administration of amodified-live virus vaccine against bovine viral di-arrhea. Am J Vet Res 59:848–850.

Dean HJ, Leyh R: 1999, Cross-protective efficacy of abovine viral diarrhea virus (BVDV) type 1 vaccineagainst BVDV type 2 challenge. Vaccine17:1117–1124.

Done JT, Terlecki S, Richardson C, et al.: 1980,Bovine virus diarrhoea-mucosal disease virus: path-ogenicity for the fetal calf following maternal infec-tion. Vet Rec 106:473–479.

Dubois WR, Cooper VL, Duffy JC, et al.: 2000, Apreliminary evaluation of the effect of vaccinationwith modified live viral diarrhea virus (BVDV) ondetection of BVDV antigen in skin biopsies usingimmunohistochemical methods. Bovine Pract34:98–100.

Duffel SJ, Sharp MW, Winkler CE, et al.: 1984,Bovine virus diarrhoea-mucosal disease virus-induced fetopathy in cattle: efficacy of prophylacticmaternal pre-exposure. Vet Rec 114:558–561.

Elahi SM, Shen SH, Talbot BG, et al.: 1999, Recom-binant adenoviruses expressing the E2 protein ofbovine viral diarrhea virus induces humoral andcellular immune responses. FEMS MicrobiolLetters 177:159–166.

Ellis J, West K, Cortese V, et al.: 2001, Effect of ma-ternal antibodies on induction and persistence of



vaccine-induced immune responses against bovineviral diarrhea virus type II in young calves. J AmVet Med Assoc 219:351–356.

Endsley JJ, Quade MJ, Terhaar B, Roth JA: 2002,Bovine viral diarrhea virus type 1-and type 2-specific bovine T lymphocyte-subset responsesfollowing modified-live virus vaccination. VetTherapeutics 3:364–372.

Flores EF, Ridpath JF, Weiblen R, et al.: 2002,Phylogenetic analysis of Brazilian bovine viral di-arrhea virus type 2 (BVDV-2) isolates: eveidencefor a subgenotype within BVDV-2. Virus Research87:51–60.

Frey HR, Eicken K, Grummer B, Kenklies S, et al.:2002, Foetal protection against bovine viral diar-rhoea virus after two-step vaccination. J Vet Med B49:489–493.

Fulton RW, Burge LJ: 2000, Bovine viral diarrheavirus types 1 and 2 antibody response in calves re-ceiving modified live virus or inactivated vaccines.Vaccine 19:267–274.

Fulton RW, Confer AWW, Burge LJ, et al.: 1995,Antibody responses by cattle after vaccination withcommercial viral vaccines containing bovine her-pesvirus-1, bovine viral diarrhea virus, parain-fluenza-3 virus, and bovine respiratory syncytialvirus immunogens and subsequent revaccination atday 140. Vaccine 13:725–733.

Fulton RW, Cook BJ, Step DL, et al.: 2002b,Evaluation of health status of calves and the impacton feedlot performance: assessment of a retainedownership program for postweaning calves. Can JVet Res 66:173–180.

Fulton RW, Purdy CW, Confer AW, et al: 2000a,Bovine viral diarrhea type 1 and 2 infections infeeder calves with respiratory disease: interactionswith Pasteurella spp., parainfluenza-3 virus andbovine respiratory syncytial virus. Can J Vet Res64:151–159.

Fulton RW, Ridpath JF, Confer AW, et al.: 2003a,Bovine viral diarrhoea virus antigenic diversity: im-pact on disease and vaccination programs.Biologicals. In press.

Fulton RW, Ridpath JF, Saliki JT, et al.: 2002a,Bovine viral diarrhea virus (BVDV) 1b: Predomi-nant subtype in calves in calves with respiratorydisease. Can J Vet Res 66:181–190.

Fulton RW, Saliki JT, Burge LJ, et al: 1997, Neutrali-zing antibodies to type 1 and 2 bovine viral diar-rhea viruses: detection by inhibition of viral cy-topathology and infectivity by immunoperoxidaseassay. Clin Diagn Immunol 4:380–383.

Fulton RW, Saliki JT, Burge LJ, Payton ME: 2003c,Humoral immune response and assessment of vac-cine virus shedding in calves receiving modified

live virus vaccines containing bovine herpesvirus-1and bovine viral diarrhoea virus 1a. J Vet Med B50:31.

Fulton RW, Saliki JT, Confer AW, et al: 2000b,Bovine viral diarrhea virus cytopathic and noncyto-pathic biotypes and type 1 and 2 genotypes in diag-nostic laboratory accessions: clinical and necropsysamples from cattle. J Vet Diagn Invest 12:33–38.

Fulton RW, Step DL, Ridpath JF, et al.: 2003b,Response of calves persistently infected with non-cytopathic bovine viral diarrhea virus (BVDV) sub-type 1b after vaccination with heterologous BVDVstrains in modified live virus vaccines andMannheimia haemolytica bacterin-toxoid. Vaccine21:2980–2985.

Giangaspero M, Harasawa, R: 1999, Ovine pestivi-ruses: Their taxonomic status revealed by palin-dromic nucleotide substitutions. Vet Microbiol70:33–39.

Grooms DL, Brock KV, Ward LA: 1998, Detection ofcytopathic bovine viral diarrhea virus in the ovariesof cattle following immunization with a modifiedlive bovine viral diarrhea virus vaccine. J Vet DiagnInvest 10:130–134.

Grooms DL, Coe P: 2002, Neutralizing antibody re-sponses in preconditioned calves following vaccina-tion for respiratory viruses. Vet Therapeutics3:119–127.

Harpin S, Hurley DJ, Mbikay M, et al.: 1999,Vaccination of cattle with a DNA plasmid encodingthe bovine viral diarrhoea virus major glycoproteinE2. J Gen Virol 80:3137–3144.

Howard CJ, Clarke MC, Brownlie J: 1989, Protectionagainst respiratory infection with bovine virus diar-rhoea virus by passively acquired antibody. VetMicrobiol 19:195–203.

Howard CJ, Clarke MC, Sopp P, Brownlie J: 1994,Systemic vaccination with inactivated bovine virusdiarrhoea virus protects against respiratory chal-lenge. Vet Microbiol 42:171–179.

Jones L, Van Campen H, Xu ZC, et al.: 2001,Comparison of neutralizing antibodies to type 1a,1b, and 2 bovine viral diarrhea virus from experi-mentally infected and vaccinated cattle. BovinePract 35:137–140.

Kirkpatrick J, Fulton RW, Burge LJ, et al.: 2001,Passively transferred immunity in newborn calves,rate of antibody decay, and effect on subsequentvaccination with modified live virus vaccine.Bovine Pract 35:47–55.

Lambert G: 1973, Bovine viral diarrhea: Prophylaxisand postvaccinal reactions. J Am Vet Med Assoc163:874–876.

Liess B, Orban S, Frey HR, et al.: 1984, Studies ontransplacental transmissability of a bovine virus

Vaccines 221


diarrhoea (BVD) vaccine virus in cattle II.Inoculation of pregnant cows without detectableneutralizing antibodies to BVD virus 90–229 daysbefore parturition (51st to 190th day of gestation).Zbl Vet Med B 31:669–681.

Makoschey B, Janssen MGJ, Vrijenhoek MP, et al.:2001, An inactivated bovine virus diarrhoea virus(BVDV) type 1 vaccine affords protection againstBVDV type 2. Vaccine 19:3261–3268.


McKercher DG, Saito JK, Crenshaw GL, et al.: 1968,Complications in cattle following vaccination witha combined bovine viral diarrhea-infectious bovinerhinotracheitis vaccine. J Am Vet Med Assoc152:1621–1624.

Nobiron I, Thompson I, Brownlie, J, Collins ME:2003, DNA vaccination against bovine viral diar-rhoea virus induces humoral and cellular responsesin cattle with evidence for protection against viralchallenge. Vaccine. In press.

Orban S, Liess B, Hafez SM, et al.: 1983, Studies ontransplacental transmissibility of a bovine virus di-arrhoea (BVDV) vaccine virus in cattle. I.Inoculation of pregnant cows 15 to 90 days beforeparturition (190th to 265th day of gestation). ZblVet Med B 30:619–634.

Patel JR, Shilleto RW, Williams J, Alexander DCS:2002, Prevention of transplacental infection ofbovine foetus by bovine viral diarrhoea virusthrough vaccination. Arch Virol 147:2453–2463.

Pellerin C, vanden Hurk J, Lecomte J, Tijssen P:1994, Identification of a new group of bovine viraldiarrhea virus strains associated with severe out-breaks and high mortalities. Virology 203:260–268.

Perino LJ, Hunsaker BD: 1997, A review of bovinerespiratory disease vaccine field efficacy. BovinePract 31:59–66.

Peter CP, Tyler DE, Ramsey FK: 1967,Characterization of a condition following vaccina-tion with bovine virus diarrhea vaccine. J Am VetMed Assoc 150:46–52.

Phillips RM, Heuschele WP, Todd JD: 1975,Evaluation of a bovine viral diarrhea vaccine pro-duced in a porcine kidney cell line. J Am Vet MedAssoc 36:135–140.

Potgeiter LND: 1995, Immunology of bovine viraldiarrhea virus. Vet Clin North Am: Food AnimPract 11:501–520.


Ridpath JF, Bolin SR: 1998, Differentiation of types1a, 1b and 2 bovine viral diarrhea virus (BVDV) byPCR. Mol Cell Prob 12:101–106.


Ridpath JF, Neill JD, Endsley J, Roth JA: 2003, Effectof passive immunity on the development of a pro-tective immune response against bovine viral diar-rhea virus in calves. Am J Vet Res 64:65–69.

Rosner SF: 1968, Complications following vaccina-tion of cattle against infectious bovine rhinotra-cheitis, bovine viral diarrhea-mucosal disease, andparainfluenza type 3. J Am Vet Med Assoc152:898–902.


Shimazaki T, Nakamura S, Taguchi K, Inoue Y, et al.:2003, Efficacy of bovine viral diarrhea vaccineused in Japan against bovine viral diarrhea virustype 2 strain 890. J Vet Med Sci 65: 263–266.

Studer E, Bertoni G, Candrian U: 2002, Detection and characterization of pestivirus contaminations in human live viral vaccines. Biologicals30:289–296.

Talens LT, Beckenhauer WH, Thurber ET, et al.:1989, Efficacy of viral components of a nonaborti-genic vaccine for prevention of respiratory and re-productive system diseases in cattle. J Am Vet MedAssoc 194:1273–1280.

van Campen H, Vorpahl P, Huzurbazar S, et al.: 2000,Evidence for bovine viral diarrhea virus type 2(BVDV)-associated disease in beef herds vacci-nated with a modified-live type 1 BVDV vaccine. JVet Diagn Invest 12:263–265.

van Oirschot JT, Bruschke CJM, van Rijn PA: 1999,Vaccination of cattle against bovine viral diarrhoea.Vet Microbiol 64:169–183.

van Rijn PA, van Gennip HGP, Leenderste CH, et al.:1997, Subdivision of the Pestivirus genus based onenvelope glycoprotein E2. Virology 237:337–348.



INTRODUCTIONThe management and control of bovine viral diar-rhea virus (BVDV) infection in cattle herds musttake into consideration two methods of virus trans-mission: postnatal horizontal transmission and ges-tational vertical transmission from a viremic dam toher fetus (Meyling et al., 1990). Postnatal horizontaltransmission results in a transient infection that isusually mild or subclinical, but can result in severedisease if susceptible cattle are exposed to a virulentstrain of the virus (Kelling et al., 2002; Hamers etal., 2000a). In addition, postnatal horizontal trans-mission can lead to vertical transmission of BVDV,because postnatal horizontal infection is the primarymethod by which a pregnant dam becomes viremicand subsequently infects her fetus.

PRINCIPAL RESERVOIRS OF BVDVThe primary source of BVDV in postnatal transmis-sion is cattle persistently infected (PI) with BVDV.Persistently infected animals are much more efficienttransmitters of BVDV than transiently infected ani-mals because they shed large amounts of virus for along period of time. Transiently infected animals ex-perience a short period of viremia and shed virus intheir body secretions and excretions from day 4–15postinfection (Brownlie et al., 1987; Duffell andHarkness, 1985). In contrast, PI animals usually havea very high and persistent viremia, and BVDV isshed throughout life from virtually all secretions andexcretions, including nasal discharge, saliva, semen,urine, tears, milk, and to a lesser extent, feces (Rae etal., 1987; Brock et al., 1991; Bezek et al., 1995;Brock et al., 1998). Horizontal transmission ofBVDV to seronegative cattle has been shown tooccur after only 1 hour of direct contact with a PI an-

imal (Traven et al., 1991). Over-the-fence contactwith a PI animal from a neighboring herd can also in-troduce BVDV into a susceptible herd (Roeder andHarkness, 1986; Miller et al., 2002).

Although transiently infected cattle are far lessefficient at transmitting the virus to susceptible in-contact animals (Meyling and Jensen, 1988; Nis-kanen et al., 2000; Niskanen et al., 2002), the occur-rence of seroconversion among assembled cattlewithout the presence of PI animals indicates thattransmission from transiently infected animals doesoccur (Meyling et al., 1990). In a field study byMoerman et al. (1993) it was found that BVDV in-fections circulated within a herd over a period of 30months in the absence of PI cattle and BVDV vacci-nation, but in the presence of transiently viremic cat-tle. Horizontal transmission of the virus from eitherpersistently or transiently infected animals to sus-ceptible cattle may be via inhalation or ingestion ofvirus-containing body fluids (Duffell and Harkness,1985). Although aerosol transmission over short dis-tances seems likely; the spread of infection is slowor absent when cattle are housed at long distancesfrom PI animals (Wentink et al., 1991).

REPRODUCTIVE EFFECTS OF BVDVEven mild or subclinical infections of pregnant andsusceptible seronegative breeding females may resultin conception failure, abortion, or vertical fetal infec-tion. The immune status of the dam, the stage of ges-tation, and the viral biotype are important factors indetermining the result of vertical infection. Trans-placental infection occurs with high efficiency inpregnant, seronegative dams (Done, et al., 1980;McClurkin et al., 1984). However, naturally acquiredimmunity can prevent later fetal infection (Orban et

223

14Management Systems and Control Programs

Robert L. Larson


al., 1983). Infection of fetuses can lead to early em-bryonic death, abortion, congenital defects, stunting,and the birth of either PI or normal calves (Figure14.1) (Baker, 1987). Fetal infection of susceptibledams with a cytopathic biotype of BVDV early ingestation will usually result in abortion. However,fetal infection with a noncytopathic biotype of BVDVwill result either in abortion or, in a certain proportionof animals, the birth of a calf that is immunotolerantto and persistently infected with that particular non-cytopathic strain of BVDV. Thus, PI cattle are the re-sult of in utero exposure to the noncytopathic biotypeof BVDV prior to the development of a competentfetal immune system (Casaro et al., 1971; McClurkinet al., 1984). Age at immune competence in the faceof BVDV exposure is variable and has been reportedto range from 90–125 days (Casaro et al., 1971;McClurkin et al., 1984; Roeder et al., 1986). If PI fe-tuses survive to term, they are continually viremic,but immunotolerant to the homologous BVDV(Duffell and Harkness, 1985; Roeder et al., 1986).

In addition to BVDV causing conception failureand abortion, reproductive efficiency can be de-creased due to fatal congenital defects followingfetal infection between 100 and 150 days of gesta-tion (Duffell and Harkness, 1985). The teratogeniclesions associated with fetal infection with BVDVinclude microencephaly, cerebellar hypoplasia, hy-dranencephaly, hydrocephalus, hypomyelination ofthe spinal cord, cataracts, retinal degeneration, opticneuritis, microphthalmia, thymic aplasia, hypotri-chosis, alopecia, brachygnathism, growth retarda-tion, and pulmonary hypoplasia (Baker, 1987).

EXPOSURE OF HERDS TO PI CATTLE

Suckling calves are commonly in contact with thebreeding herd and thus come in contact with dams inearly stages of pregnancy. As a result, PI sucklingcalves may be a source of BVDV infection in breed-

ing herds, causing decreased pregnancy percentage,pregnancy loss, preweaning mortality and the induc-tion of PI calves in the next generation (Wittum etal., 2001; Duffell and Harkness, 1985; McClurkin etal., 1984).

Although mortality of PI calves prior to weaninghas been reported to be very high due to fatal con-genital defects and secondary infections that causeenteritis, pneumonia, and arthritis (McClurkin et al.,1979; McClurkin et al., 1984), in some situations17–50% of PI calves may reach breeding age (Bar-ber et al., 1985; Binkhorst et al., 1983; Houe, 1993).Persistently infected females of breeding age are notonly a source of horizontal transfer of BVDV, butalso will produce a PI calf themselves, which if itsurvives, creates a familial clustering of PI animals(Houe et al., 1995; McLurkin et al., 1979).

The consequence of introducing a PI animal intoa beef herd (confined breeding and calving seasons)depends on the timing of the introduction relative tothe breeding season and the subsequent immuno-logic status of the herd during early gestation (Table14.1). A likely scenario for a BVDV-exposed herd isto experience an initial peak of disease followed bylow-level chronic reproductive losses in subsequentmonths and years. If a PI animal enters the herd ei-ther by birth or by purchase near the start of thebreeding season, a high percentage of the herd maynot be immunologically protected to the degree nec-essary to prevent viremia, conception failure, abor-tion, or fetal infection. If the PI animal is in contactwith the breeding herd for a long enough period oftime, the majority of the herd should become in-fected and seroconvert. Seropositive animals are lesslikely to have conception failures, abortions, or in-fected fetuses compared to seronegative animals. Ifno intervention is applied to the herd, the number ofsusceptible females the following year should begreatly decreased and the number of abortions and


Figure 14.1. The effect of stage of gestation at the time of BVDV infection of susceptible pregnant cows on clinicaloutcome.


infected fetuses (both PI and immunocompetent)should decrease. Thus, even in the absence of vacci-nation and culling, the number of PI animals andBVDV infections in a closed herd may be self-limiting over time (Houe, 1995). A model developedby Cherry et. al. (1998) indicates that in continuouscalving situations, the proportion of PI animals in adairy herd will reach an equilibrium of about

0.9–1.2% in closed herds with no BVDV controlprocedures (Cherry et al., 1998). This model doesnot hold true if the herd is not closed (replacementfemales and bulls are added). Prevalence of PI ani-mals and the resulting problems would not be ex-pected to reach equilibrium in BVDV-infected herdswith no control measures that are purchasingheifers, particularly bred heifers.

Management Systems and Control Programs 225

Table 14.1. Various outcomes of bovine viral diarrhea virus (BVDV) infection depending ontransmission and host.

Timing and Route of Transmission Infected Animal Outcome

Postnatal Seropositive nonpregnant animal • Mild or subclinical diseasehorizontal Susceptible nonpregnant animal • Mild or subclinical diseasetransmission • Moderate to severe clinical disease with

virulent strains

Susceptible breeding animal • Conception failure• Conception success

Susceptible pregnant animal • Fetal infection

Gestational Fetus of persistently infected dam • Fetal infectionvertical •• Embryo loss/abortiontransmission •• Immunotolerant, persistently infected

fetus/calf

Fetus of susceptible dam (d 0 to d • Fetal infection90–125 gestation) — cytopathic biotype •• Embryo loss/abortion

Fetus of susceptible dam (d 0 to d • Fetal infection90–125 gestation) — noncytopathic •• Embryo loss/abortionbiotype •• Immunotolerant, persistently infected

fetus/calf•• Fetal malformations (d 100–150):

cerebellar hypoplasia, hypotrichosis,brachygnathism, depleted lymphnodes/tissue, etc.

Fetus of susceptible pregnant animal with • Fetal infectionimmune-competent fetus (>90–125 d) •• Abortion

•• Fetal malformations•• Normal fetus/calf

Fetus of seropositive pregnant dam • Fetal protection from infection(seropositive following natural infection) • Fetal infection ?

Fetus of seropositive pregnant dam • Fetal protection from infection(seropositive following vaccination) • Fetal infection

•• Embryo loss/abortion ?•• Immunotolerant, persistently infected

fetus/calf•• Normal fetus/calf


Estimates of the prevalence of PI animals in thegeneral cattle population range between 0.13% and2.0% (Bolin et al., 1985; Houe et al., 1995; Howardet al., 1986; Wittum et al., 2001). Differences in re-ported prevalence may be due to the populationtested, the country/continent where the populationwas located, management system in effect and/orthe diagnostic tests utilized. Persistent infection hasa clustered distribution, which means a few herdsmay contain several cattle but most herds containonly non-PI cattle (Bolin, 1990). Clustering of mul-tiple PI animals in a herd is primarily due to expo-sure of numerous susceptible dams to a PI or tran-siently infected source of noncytopathic BVDVprior to day 125 of gestation, but can also be due tosurviving PI offspring of a PI dam.

DIAGNOSTIC TESTS FOR THEDETECTION OF PI ANIMALSBecause the PI animal is an important reservoir andtransmitter of BVDV, control programs must firstidentify and remove these animals from the breedingherd. Because of vertical transmission of the virusfrom viremic dams to their fetuses, PI animalsshould be removed prior to the start of the breedingseason in beef herds with a controlled breeding sea-son. In dairy herds, the PI animals must be removedas soon as possible from direct or indirect contactwith the breeding herd. In order to find and removePI beef cattle prior to the start of the breeding sea-son, all calves, all replacement heifers, all bulls, andall nonpregnant dams without calves must be testedfor PI status (Kelling et al., 2000). Nonpregnantcows may lack calves prior to the start of the breed-ing season due to not becoming pregnant, aborting,or calf mortality. Any female that is still pregnant atthe time the herd is tested should be isolated fromthe breeding herd and kept isolated until her calf istested and found to be negative. Calves persistentlyinfected with BVDV can be identified by testingstrategies that utilize virus isolation from wholeblood (buffy coats) or serum, immunohistochem-istry (IHC) staining of viral antigen in skin biopsies,antigen-capture enzyme-linked immunosorbentassay (ELISA) from skin biopsies, and polymerasechain reaction (PCR) methods from whole blood orserum (Dubovi, 1996)

VIRUS ISOLATION

Persistently infected animals produce an exception-ally large number of BVDV particles that can be iso-lated from virtually any tissue sample, includingblood, serum, spleen, liver, and lymphoid tissue

(Ellis et al., 1995; Straver et al., 1983). Virus isola-tion is considered to be very specific for BVDV in-fection; however, colostral antibodies may tem-porarily reduce the amount of free virus in the serumof young calves, thus making it less sensitive to iso-late virus from the blood or serum of young calves(Kelling, et al., 1990; Palfi et al., 1993; Brock et al.,1998). However, when the maternal antibodies havedisappeared, BVDV can be isolated easily from PIcalves. BVDV could be isolated by 6 weeks of agein colostrum-fed PI calves in one study (Brock et al.,1998), and by 8 weeks of age in colostrum-fed PIcalves in another study (Palfi et al., 1993). Althoughit is rare, some PI calves develop neutralizing anti-body and clear the persistent BVDV strain fromtheir serum. However, in these animals it is still pos-sible to isolate virus from leukocytes (Brock et al.,1998). Virus isolation methods are labor-intensiveand take several days to complete. An additionalshortcoming is that virus isolation may not differen-tiate between transiently infected animals and PI an-imals, unless positive cattle are re tested and remainpositive at a later date (i.e., 3 weeks later).

IMMUNOHISTOCHEMISTRY

An IHC test for BVDV infection using skin biopsysamples, such as ear notches, is available that differ-entiates between PI animals and transient BVDV in-fections (Njaa et al., 2000). Transiently infected an-imals may have internal organ tissue samples thatare IHC-positive. However, when skin samples wereevaluated, transiently infected animals either had nostaining, or staining was confined to the epidermalkeratinocytes and follicular ostia, in contrast to PIcattle with antigen-positive staining cells in all lay-ers of the epidermis, all levels of hair follicles, andthe hair bulb (Njaa et al., 2000). This test is suitablefor herd screening because samples can be takenfrom cattle of any age, sample collection is simple,the samples are stable for transport and handling, itis not affected by the presence of passive antibodiesand the test is both sensitive and specific for PI cat-tle (Ellis et al., 1995; Baszler et al., 1995; Njaa et al.,2000). In addition, the use of modified live BVDVvaccine does not appear to produce false positiveIHC reactions on skin biopsies when testing for PIanimals (DuBoise et al., 2000).

POLYMERASE CHAIN REACTION

Polymerase chain reaction testing for BVDV infec-tion is more rapid than virus isolation and can alsodetect virus in antigen-antibody complexes (Brocket al., 1998). Polymerase chain reaction tests are



sensitive and have been shown to differentiate be-tween BVDV genotypes. However, a single BVDV-positive blood sample tested by PCR does not allowthe diagnostician to differentiate between viremiafrom a postnatal acquired infection and that frombeing a PI animal. Because PCR tests can identifyminute amounts of virus, this test can be used inpooled samples of blood or milk in surveillanceprograms.

SEROLOGY

Although PI cattle are usually seronegative toBVDV, an immune response can be elicited to a het-erologous strain (Brock et a., 1998). Presumably,this response can follow either natural or vaccine ex-posure. In addition, some cattle in both vaccinatedand unvaccinated herds are seronegative, makingserology alone unsuitable for identification of PI an-imals (Bolin et al., 1985; Houe et al., 1995; Wittumet al., 2001).

DIAGNOSTIC TESTINGSTRATEGIES TO IDENTIFY PICALVESIf a herd has had confirmed PI calves, or if the his-tory strongly suggests the presence of PI calves, thea priori assessment of PI prevalence is fairly high,making the predictive value of a positive test highenough so one can conclude that the animal is per-sistently infected and that PI animals are present inthe herd. Consequently, a second confirmatory testmay not be needed. In contrast, if there is no a priorievidence of PI prevalence greater than 0.3–0.5%, thepredictive value of a positive test is low and a differ-ent confirmatory test may be advisable before mak-ing conclusions about the individual animal or theherd (Table 14.2). When a calf is identified as PI, it

should be euthanized or removed for slaughter andthe dam should be tested. Most dams of PI calves arenot PI themselves, and if confirmed as non-PI, canreenter the breeding herd because naturally acquiredimmunity is considered to prevent future fetal infec-tions (Orban et al., 1983). Dams identified as PI,however, should be sold to slaughter immediately.

In most whole-herd testing situations, IHC testingof skin samples is currently the test of choice be-cause it can be accurately performed on animals ofany age and a single sample is all that is usually suf-ficient. The use of virus isolation or PCR to identifyBVDV-infected cattle requires a second test 3 weeksfollowing any positive results to differentiate be-tween transient viremia and PI with BVDV.

IDENTIFICATION OF DAMSCARRYING A PI FETUSPurchase of non-PI dams pregnant with PI fetuses(PI-carriers) is a potential source of BVDV spreadbetween herds. Any pregnant heifer or cow that ispurchased should not come in contact with otherpregnant cattle unless they have been tested andconfirmed not to be PI. In addition, because the sta-tus of the fetus is unknown, the calf must be testedsoon after birth and prior to the breeding season toprevent contact with pregnant cattle in the first 210days of gestation. Because it would be advantageousto identify PI fetuses prior to purchase or entry ofthe dam into a herd, other strategies to determine thePI status of the fetus are being investigated.

Unvaccinated non-PI dams carrying PI fetuses areseropositive and virus-negative, but they havemarkedly higher titers of antibodies to BVDV com-pared to dams carrying healthy fetuses (Lindberg etal., 2001). Optical density (OD) of an indirectELISA technique has a strong positive correlation to


Table 14.2. Possible tests to identify persistently infected (PI) cattle.

Initial Test Confirmatory Test

Immunohistochemistry (IHC) of skin biopsy Virus isolation (VI) from serum or whole bloodor

Polymerase chain reaction (PCR) from serum or whole blood

VI-positive from two whole blood samples taken IHC of skin biopsy3 weeks apart

PCR-positive from two whole blood samples taken IHC of skin biopsy3 weeks apart


virus neutralization titer, and for unvaccinatedseropositive dams carrying normal calves, the aver-age OD value stays the same irrespective of when inpregnancy the sample is taken (Lindberg et al.,2001). In contrast, unvaccinated dams that carry PIfetuses have an increasing antibody titer from earlypregnancy (when they become infected) through thelast month of gestation, resulting in significantlyhigher OD by the 7th month of gestation comparedto dams carrying normal calves (Lindberg, 2001).The sensitivity of this method to identify dams car-rying PI calves ranged from 0.79–0.96 and thespecificity ranged from 0.37–0.70, depending on the test cutoff (Lindberg et al., 2001). Because of therelatively high sensitivity, relatively low specificity,and low prevalence of dams carrying PI fetuses, thenegative predictive value will be high and the posi-tive predictive value will be poor. Therefore, a neg-ative test should be considered good evidence forthe absence of a PI fetus, but a positive test wouldnot be a particularly strong indication for the pres-ence of a PI fetus in tested unvaccinated dams inherds with endemic BVDV.

Another method that has been investigated in aneffort to identify dams carrying a PI fetus is the col-lection of fetal fluids and their evaluation for thepresence of BVDV (Callan et al., 2002). Callan et al.(2002) were able to use abdominal ultrasound toguide a spinal needle introduced through a surgi-cally prepared area of the right flank into the preg-nant uterus to collect a sample of fetal fluid. Virusisolation performed on the fetal fluid samples cor-rectly identified 1/1 BVDV-infected and 168/168noninfected fetuses. In addition to the technical ex-pertise required, this method may have limited ap-plication in most herds because 8% (14/169) oftested dams aborted or delivered premature calveswithin 3 weeks of fetal fluid collection that mayhave been related to the sample collection (Callan etal., 2002).

MONITORING HERDS FOR BVDVPI RISKThe cost of initiating a BVDV PI whole-herdscreening protocol on a farm or ranch is significant.Because of the low prevalence of herds with at leastone PI animal, veterinary practitioners may not beeconomically justified to initiate whole-herd screen-ing protocols to find PI BVDV beef cattle for herdsat low risk for the presence of PI cattle or herds thatcannot gain significant market price advantage forselling groups of cattle that have been tested and de-termined to be free of PI individuals (Wittum et al.,

2001; Larson et al., 2002). However, if ranch historyraises a suspicion of BVDV PI cattle being presentin the herd, or if significant marketing advantagesexist, a protocol to screen the herd can be defendedbased on its likelihood to improve or protect eco-nomic return (Larson et al., 2002). Several strategiescan be employed to monitor herds for their risk ofhaving PI cattle present. The interpretation of resultsfrom these strategies would be different if the goalwere to monitor for the presence of BVDV ratherthan PI animals. If complete eradication of BVDV isdesired, the effort and cost of monitoring is muchgreater than for monitoring for the presence of PIcattle.

USE OF PRODUCTION RECORDS ANDLABORATORY EVALUATION OF MORIBUNDAND DEAD CALVES

The minimal level of surveillance for every herdshould include monitoring of herd fertility (earlybreeding season pregnancy proportion, pregnancyper insemination proportion, and total pregnancyproportion), neonatal calf morbidity and mortalityproportions, and weaning proportions. Because ofthe negative effect of the presence of PI calves in abreeding herd on measures of reproductive effi-ciency, the presence of physical abnormalities atbirth, and calf survivability to weaning, an unaccept-able level of these symptoms increases the risk thatBVDV is a problem in the herd and increases thelikelihood that whole-herd screening for PI cattlewill be economically rewarding (Houe and Meyling,1991). Although, in many situations, pregnancy ratedrops significantly at the time of conception of theoldest PI animal, and about 6 months later calf mor-tality increases; using production records alonelacks sensitivity for identifying herds with PI ani-mals because the clinical indications of PI presencemay be less noticeable in some outbreaks (Houe andMeyling, 1991). The clinical signs and time se-quence following introduction of BVDV infectioninto different herds varies considerably due to thedifferent proportions of seronegative animals in thecritical period of pregnancy and different virulenceamong BVDV strains (Houe, 1995).

In addition to monitoring production records,minimal surveillance should include the necropsyexamination of as many aborted fetuses, stillborncalves, and calves that die preweaning as possible,with whole blood submitted for determination ofBVD viremia and serum submitted for serologic ev-idence of infection. In addition, moribund calvesfrom clusters of pneumonia, neonatal scours, or sep-



ticemia outbreaks that are not easily explained bysanitation or other problems should also be testedfor BVDV exposure and PI status. If most perinataland preweaning mortalities are examined for BVDVantigen via IHC and found to be negative, it is notlikely that PI animals are present in the herd. Thepresence of PI animals in the herd will be estab-lished by a single confirmed IHC-positive skin sam-ple. The presence of PI animals is not ruled out andmay be considered likely if few moribund or deadcattle are tested and found to be IHC-negative, butother tests indicate the presence of viremia or serol-ogy indicates recent BVDV infection and the possi-bility of PI animals being in contact with the mori-bund or dead sample animals (Table 14.3).

The advantage of utilizing production measuresand necropsies to determine whether herds have ei-ther a high or low risk for the presence of BVDV PIanimals is that minimal expense is involved andthese management tactics are inclusive for the mon-itoring of other disease and production problems.This level of monitoring is probably appropriate inherds with no evidence for the presence of PI ani-mals and that are at low risk of PI introduction(Larson et al., 2002). The disadvantage is that atleast one PI animal is allowed into the herd beforeproduction losses are identified, and productionlosses will continue for at least 1 year after interven-tion is initiated.

USE OF POOLED SAMPLES OF WHOLEBLOOD FOR PCR TESTING

Herd monitoring for the introduction of PI animalscan also be accomplished with pooled whole bloodsamples for PCR testing. By pooling samples, theexpense of screening herds with a low prevalence ofPI animals is minimized. Polymerase chain reactionis well suited to pooled-sample testing for the pres-ence of BVDV PI animals because it is sensitiveenough to detect minute amounts of virus. A singlePI animal was detectable in pools of 200–250 nega-tive samples (Muñoz-Zanzi et al., 2000). Animalscontributing to negative samples are all assumed tobe non-PI, whereas positive pools may contain sam-ples from PI animals or transiently viremic animals.If the initial pool is PCR-positive, it must be splitand retested to differentiate viremic and non-viremic animals. After the viremic animals are iden-tified, they must be classified as transiently infectedor PI with either a subsequent PCR or virus isolationtest in 3 weeks or via IHC of a skin sample.

The best size of the initial pool is determined bythe balance between the cost savings of having large

numbers of individuals represented in negative poolsand few individuals represented in positive pools thatrequire further diagnostics. If pool size is too large,there is an increased chance that any single pool willtest positive, requiring additional testing to identifythe few truly viremic individuals in the pool. If thesamples are grouped in unnecessarily small pools,the cost benefit of pooling samples is lost to the largenumber of negative pools tested for each positivepool identified (Muñoz-Zanzi et al., 2000). Muñoz-Zanzi et. al. (2000) developed a simulation model todetermine that the economically optimum samplesize depends on prevalence of true positives in thepopulation. For a PI prevalence of 0.5–1.0%, the op-timum number of samples in an initial pool is 20–30,using a described repooling strategy for test-positiveinitial pools (Muñoz-Zanzi et al., 2000). As preva-lence increases, the least-cost initial pool size de-creases (Muñoz-Zanzi et al., 2000).

If whole blood samples are collected for pooledPCR from all suckling calves prior to the start of thebreeding season, PI cattle can be identified and re-moved before contact with pregnant females,thereby eliminating the opportunity for a PI animalwithin the herd to cause reproductive failure and tocreate more PI animals in the next calf crop.Screening for PI animals at a later time, such asweaning, is discouraged. If samples are taken atweaning, although PI cattle can be removed fromthe herd, those continuously viremic animals werein contact with pregnant females throughout muchof gestation and can cause reproduction and produc-tion losses, including the creation of PI cattle in thenext calf crop.

USE OF SEROLOGIC EVALUATION OFSENTINEL ANIMALS

Herd surveillance of dairy herds has been describedusing sentinel animals. Pillars and Grooms (2002)


Table 14.3. Diagnostic test results of clinicallyill animals with BVDV infection.

BVDV infection Viremiastatus IHCa Serology (PCRb/VIc)

Transiently infected – – +Convalescent – + –Persistently

infected (PI) + +/– +

aImmunohistochemistry.bPolymerase chain reaction.cVirus isolation.


found that serologic evaluation of unvaccinated 6- to12-month–old heifers for the presence of a highserum neutralizing titer had a sensitivity of 66% andspecificity of 100% for detecting herds that have PIcattle present when reference strains from bothgenotypes 1 and 2 were used. Herds that were iden-tified as containing PI animals could then utilizeother diagnostic tests to identify individual PI ani-mals for removal. In countries that have BVDV con-trol programs that do not allow the use of vaccina-tion, a similar strategy to identify herds with PIcattle has been investigated. This strategy is basedon the fact that in unvaccinated herds, there are sig-nificantly more antibody-positive animals, espe-cially in young stock, in herds with PI animals thanin herds without PI animals (Houe, 1992).

However, because of the variable percentage ofantibody-positive animals in herds without PI ani-mals, it was not possible to predict the presence ofPI animals in dairy herds in The Netherlands usingthis method (Zimmer et al., 2002). Because thismonitoring occurs when potential PI animals are atleast 6 months old, a positive herd test would indi-cate that the herd has been exposed to at least one PIanimal for at least 6 months, reproductive losses andthe presence of PI calves in the next calf crop wouldbe expected. When utilizing serology strategies forBVDV monitoring or diagnostics, it is important touse reference strains from both genotypes 1 and 2because it is not unusual for BVDV-exposed cattleto have a low titer against one genotype and a highertiter against the other.

USE OF ANNUAL WHOLE-HERD TESTING

Certain high biosecurity herds, such as herds sellingor developing replacement breeding animals, mayelect to undergo a high level of surveillance even inthe absence of evidence that PI animals are present.This high level of biosecurity may be important totheir marketing plan or may indicate a high valueplaced on avoiding the small, but real, risk of intro-ducing BVDV virus into the herd with subsequentnegative reproductive, health, and marketing conse-quences. The first year that a beef herd adopts thisstrategy, all suckling calves, all females that werebred that failed to present a calf on test day, all re-placement heifers, and all bulls should be tested. Ifany calf is confirmed as a PI animal, his dam shouldbe tested as well. In subsequent years, only sucklingcalves and any purchased animals need to be tested.If pregnant animals are purchased, the dam shouldbe tested prior to or at arrival and the calf should betested immediately after birth. In beef herds with a

confined breeding season, this testing must occurbefore the start of the breeding season to ensure thatno PI animals are in contact with pregnant femalesduring gestation. Heifer development operationsshould test every heifer prior to or at arrival at the fa-cility. Dairy operations should test every calf thatstays on the same farm as the breeding herd or iscared for by breeding herd employees, or if anyequipment used on the calf-rearing farm is used onthe breeding herd farm.

Following the identification and removal of PI an-imals from a herd, testing of all suckling calvesshould be done for one or more breeding seasons toensure the complete accounting for PI animals.Because no test or testing strategy is perfectly sensi-tive, and because risk factors involved in the initialintroduction may still be present, a vigilant monitor-ing system is wise until a high confidence for PI-freestatus is achieved.

OTHER POTENTIAL SOURCES OFBVDVMale PI calves will occasionally be selected for useas breeding bulls. The amount of BVDV excreted inthe semen of persistently infected bulls is very high(104–106 TCID50/ml) (Revell et al., 1988). WhenBVDV is infused into the uterus at the time ofbreeding, seronegative cattle exhibit a significant re-duction in conception rate, but seropositive animalsmay not be adversely affected (Whitmore et al.,1981). BVDV-contaminated semen is an efficienthorizontal transmitter of disease from bull to cow(Paton, et al., 1989). If PI bulls are used for naturalservice, the cows may conceive when immunity hasdeveloped, resulting in the birth of normal (non-PI)calves (Barber et al., 1985; McClurkin et al., 1979).If PI bulls are used for AI, all or most seronegativefemales bred with the semen will become infectedwith BVDV although most will not produce a PI calf(Meyling and Jensen, 1988).

Although some evidence exists for BVDV to causelatent infections, particularly in gonads and acces-sory sex glands (Kirkland, et al., 1991; Voges et al.,1998), recrudescence and excretion in immune-competent animals has not been shown to date to beinvolved in the epidemiology of the disease(Brownlie, 1990). And although PI bulls will shedBVDV in semen for prolonged periods of time, virusexcretion in semen from transiently infected bullswas confined to days 10–14 post–experimental-infection, and the virus titer in semen of transientlyinfected bulls was much lower than for PI bulls(Paton et al., 1989).



EMBRYO TRANSFER

Embryo transfer is a potential route of transmissionof BVDV. If the embryo recipient is PI, verticaltransmission to the transferred embryo will occurcausing embryonic loss or the creation of a PI fetus.Although there is no evidence to suggest that BVDVis present inside the embryos of viremic females, thevirus can be present on the intact zona pellucida ofPI and transiently infected females, and the virus ispresent at high levels in the uterine environment ofPI donors (Singh et al., 1982). Established washingprocedures will remove contaminating virus, but ifthese procedures are not followed, BVDV from thecollection fluids or virus present on the zona pellu-cida can be horizontally transferred to a susceptiblerecipient cow (Singh, 1982; Singh et al., 1985).Vertical transmission from the recipient cow to thefetus can occur, resulting in embryonic/fetal deathor the birth of a PI calf (Brock et al., 1991). BVDVinfection of the recipient cow and fetus can alsooccur when both the donor and recipient are free ofBVDV if BVDV-contaminated fetal serum is used inthe embryo transfer process or if contaminated liq-uid nitrogen is in direct contact with embryos (Singhet al., 1985; Bielanski et al., 2000).

OTHER UNGULATE SPECIES (DOMESTIC ANDWILDLIFE)Other ungulate species may be potential sources ofBVDV to susceptible cattle herds. Transmission ofBVDV between sheep and cattle has been demon-strated, but the importance of this transmission hasnot been established (Carlsson and Belak, 1987).BVDV has also been isolated from pigs, but again,the importance of pigs as a source of the virus to sus-ceptible herds is not established (Liess andMoenning, 1990; Terpstra and Wensvoort, 1988).Deer seropositive to BVDV have been identified inNorth America and Europe (Davidson and Crow,1983; Nielsen et al., 2000; Frolich et al., 2002).However, the existence of PI deer has not beendemonstrated, and cattle are assumed to be the sourceof BVDV infection for free-ranging ruminants.

FOMITES

Fomites may serve in the transmission of BVDVfrom PI cattle to susceptible animals. A 19-gaugeneedle was able to infect susceptible cattle withBVDV when used IV within 3 minutes of drawingblood from a PI animal (Gunn, 1993). Nose tongswere able to infect susceptible cattle with BVDVwhen used for 90 seconds within 3 minutes of beingused in a PI animal (Gunn, 1993).

No evidence has been presented that insects are asource of BVDV transmission in field outbreaks(Table 14.4). However, a role is possible: BVDVwas isolated from nonbiting flies (Musca autum-nalis) collected from the face of a PI animal, and ex-perimental BVDV transmission between a PI animaland susceptible animals occurred when 50 bitingflies fed on the PI animal for 5 minutes and 15 min-utes later fed on susceptible animals (Gunn, 1993;Tarry et al., 1991).

VACCINATION TO CONTROLBVDV-INDUCED DISEASE ANDPRODUCTION LOSSESIn addition to removal of PI reservoirs, BVDV trans-mission to and within the herd can be reduced withan appropriate vaccination program. Informationfrom serological data and limited field trials can beused to make empirical recommendations regardingwhat constitutes an effective vaccination program tolimit postnatal and gestational BVDV transmission.

IN VITRO EVIDENCE OF VACCINE EFFICACY

Although there were large variations in the vaccine-induced virus neutralizing antibody titers of individ-ual colostrum-deprived calves vaccinated with twodoses (21-day separation between doses) of an inac-tivated BVDV vaccine (Hamers et al., 2002) or witha modified live, temperature-sensitive BVDV vac-cine (Hamers et al. 2000b), sera from all animalswere capable of neutralizing a wide range of anti-genically diverse European and American isolates ofBVDV, including genotypes 1 and 2. In anotherstudy, administration of a single dose of a modifiedlive BVDV vaccine in seronegative cows inducedantibodies that were able to cross-neutralize 12 anti-genically diverse strains of BVDV (Cortese et al.,1998c) and were detectable for at least 18 months.

COLOSTRAL IMMUNITY AND VACCINATIONOF YOUNG CALVES

Adequate intake of colostrum from BVDV seropos-itive dams provides protection from clinical diseasein young calves (Cortese et al., 1998b; Ridpath etal., 2003). Vaccination of young calves has also beendemonstrated to reduce clinical disease and mortal-ity following experimental challenge (Cortese et al.,1998b). Calves that did or did not receive colostralantibodies and were vaccinated at 10–14 days of agewith a single dose of modified live vaccine (MLV)containing type-1 BVDV were protected from clini-cal disease when experimentally challenged 21 dayspostvaccination with a virulent type-2 BVDV. In



contrast, calves that did not receive colostral anti-bodies to BVDV and did not receive the MLV vac-cine suffered severe clinical disease and required eu-thanasia (Cortese et al., 1998b). Clinical scores,indicating severity of illness, were not significantlydifferent between seropositive-vaccinated andseropositive-unvaccinated calves after experimentalviral challenge in this trial (Cortese et al., 1998b).Most of the vaccinated calves that were seronegativeprior to vaccination did not have measurable serumantibody response 21 days following vaccination atthe time of experimental challenge, even thoughthese calves were protected from clinical disease(Cortese et al., 1998b). Similarly, Ridpath et. al.(2003) demonstrated that an active protective re-sponse, not correlated to serum-neutralizing anti-bodies, was mounted in young calves in the pres-ence of colostral-derived passive immunity. Thusserum antibody titers may represent an inadequate

measure of protection against disease (Cortese et al.,1998b; Ridpath et al., 2003).

ABILITY OF VACCINES TO PROVIDE FETALPROTECTION

Using vaccination strategies to prevent or reduceclinical disease caused by BVDV is important in themanagement of bovine respiratory disease in feedlotand other confined cattle situations. However, whencontrolling BVD in cowherds, controlling clinicaldisease is minimally important compared to prevent-ing fetal infection that results in embryonic/fetalloss or creation of PI animals. Fetal protection is im-munologically more difficult than protection fromclinical disease; however, the majority of vaccinesare licensed based on their ability to reduce clinicalsigns in acute infection, not to reduce reproductiveloss or fetal infection. Fetal infection can be pre-vented if the immune system of an exposed herd is


Table 14.4. Sources of bovine viral diarrhea virus and mode of transmission to susceptible cattle.

Sources of BVDV for Transmission to Susceptible Cattle with Convincing Evidence to Support Their Role inthe Epidemiology of BVD

Source of BVDV Target Animal Mode of Transmission

Cattle persistently infected with and Postnatal cattle Direct horizontal oronasal contact with immunotolerant to BVDV viral-contaminated secretions and

excretionsTransiently infected cattle Postnatal cattle Direct horizontal oronasal contact with

viral-contaminated secretions andexcretions

Dam persistently infected with and Fetus Vertical transmission across placenta fromimmunotolerant to BVDV maternal viremia

Transiently infected dam Fetus Vertical transmission across placenta frommaternal viremia

Sources of BVDV for Transmission to Susceptible Cattle with Experimental or Serologic Evidence toSupport a Potential Role in the Epidemiology of BVD

Source of BVDV Target Animal Mode of Transmission

Domestic farm animals: sheep, pigs, Postnatal cattle Direct horizontal oronasal contact with goats viral-contaminated secretions and

excretionsWildlife: deer, elk Postnatal cattle Direct horizontal oronasal contact with

viral-contaminated secretions andexcretions

Fomites: palpation sleeves, nose Postnatal cattle Indirect horizontal oronasal contact with tongs, injection needles viral-contaminated secretions and

excretionsFlies Postnatal cattle Blood meals


primed so that it can effectively neutralize circulat-ing virus before it has a chance to cross the placentaand cause fetal infection. Evidence from earlier aswell as recently reported trials indicates that mater-nal vaccination provides some protection of thefetus, although protection does not extend to 100%of fetuses (Kelling et al., 2000). In fact, efficacy ofmaternal vaccination to provide fetal protection hasbeen reported to range from 25–100% for inacti-vated vaccines (Meyling et al., 1987; Harkness etal., 1987; Brownlie et al., 1995) and from 58 to 88%for modified live vaccines (Cortese et al., 1998a;Brock and Cortese, 2001). The presence of measur-able levels of BVDV antibody in dams followingvaccination appears to provide fetal protection andis important in a planned vaccination program forBVDV control. It should be realized, however, thata sufficient amount of virus is able to escape neutral-ization inactivation by circulating antibodies insome dams to cause transplacental infection, abor-tion, and the development of persistent fetal infec-tion. Thus, vaccination programs by themselves areinadequate in controlling BVDV (Cortese et al.,1998a; Brock and Cortese, 2001; Brownlie et al.,1995).

CONTROL PROGRAMS TO LIMITLOSSES DUE TO BVDV

BEEF CATTLE

The primary goals of BVDV control in breedingherds are to prevent fetal infection in order to elimi-nate BVDV-associated reproductive losses (therebypreventing the birth of PI calves) and to reducelosses from transient BVDV infections (Harkness,1987). Cattle that have been infected with BVDVafter birth and have recovered are considered to beprotected from clinical disease following subse-quent exposure to the virus even if they are seroneg-ative (Ridpath et al., 2003). Animals that are sero-positive in response to natural exposure are alsoconsidered to be protected from future transmissionof the virus to a fetus. An immunocompetent animalthat gives birth to a persistently infected animal willdevelop high antibody titers and is likely to elimi-nate future BVDV infections rather than pass themto the fetus. Thus, an immunocompetent dam (non-PI) could have, at most, one PI calf.

While vaccination does provide some protectionfrom fetal infection, the herd level protection is notequal to that generated by natural exposure. As a re-sult, BVDV control is generally achieved by a com-bination of removal of PI cattle, vaccination, and a

biosecurity system that prevents the introduction ofPI animals into the herd and minimizes the contactwith potentially viremic cattle or wildlife (Kelling,1996).

Removal of PI animals

Herds should be routinely monitored to determinethe presence of PI animals. If the presence of PI cat-tle is confirmed or strongly suspected, a whole-herdscreening protocol, most likely utilizing IHC of skinsamples or PCR as described earlier, should be un-dertaken to identify and remove PI individuals. Asecond whole-herd screening the following yearmay be advisable in some herds where risk ofcontinued fence-line or other exposure to PI animalsis high.

Biosecurity to prevent herd exposure to PIanimals

Biosecurity to prevent herd exposure to PI or tran-siently infected animals is important, especiallyafter the removal of PI cattle, because with the re-moval of PI animal (BVDV) shedders, the numbersof naturally exposed seropositive animals in a herddecreases (Kelling, 1996). All replacement heifersand bulls that enter the breeding herd, whetherraised or purchased, should be tested prior to thestart of breeding to ensure that they are not PI ani-mals. If a pregnant animal is purchased, it should besegregated from the breeding herd until both thedam and the calf are confirmed to be non-PI. Fence-line contact with neighboring cattle should be man-aged so that stocker cattle are not adjacent to thebreeding herd during early gestation, and othercowherds are not adjacent unless they also have astrict biosecurity and vaccination program in place.

Vaccination as a component of biosecurity

Biosecurity also involves application of a vaccina-tion protocol to reduce the risk of fetal infection inthe event of cowherd exposure to a viremic andvirus-shedding animal. Modified live vaccines arebelieved to stimulate more complete protectionagainst transplacental infection (Kelling, 1996). Forthat reason, one recommendation is to vaccinate un-stressed, healthy heifers with MLV vaccines. Vac-cine administration should be timed so that a protec-tive immune response coincides with the first 4months of gestation. This is done to maximize thepotential for adequate immunity to protect againstfetal infection and reproductive failure or the birthof PI calves. If heifers have not been previously vac-cinated, the primary vaccination should be done



twice. The first dose should be given when theheifers are 6 months of age or older, and the seconddose should be given 2 months before breeding.Beef cows should be revaccinated annually beforebreeding according to label directions (Kelling,1996).

DAIRY CATTLE

The management of U.S. dairies is vastly differentthan that of beef cows; dairy calves are removedfrom their dams soon after birth and are not in con-tact with the breeding herd during gestation.Hence, calf transmission of BVDV to pregnant fe-males is eliminated or greatly reduced in dairyherds as compared to beef herds. However, becausemany pregnant replacement females are purchasedor raised off-site on dairies, the risk of introductionof a PI replacement animal or pregnant non-PI an-

imal with a PI fetus is greater in dairy operationsthan in beef herds. Biosecurity for dairy herdsshould include screening of potential replacementheifers prior to the start of their first breeding sea-son, biosecurity of heifer and adult cows duringgestation to prevent exposure to PI or transientlyinfected animals, and a vaccination program toprovide some level of protection in the face of ex-posure to viremic animals.

STOCKER/FEEDLOT OPERATIONS

Because pregnancy is not a common or desirablecomponent of stocker and feedlot operations, verti-cal transmission and reproductive losses due toBVDV are not a concern. However, BVDV viremiaor seroconversion has been associated with respira-tory disease outbreaks in feedlot situations (Martinet al., 1989; Fulton et al., 2000; Fulton et al., 2002).


Table 14.5. Control program for BVDV in beef cow herds.

I. Monitor herd for risk of PI presencea. If PI presence is confirmed or strongly sus-

pected, a whole-herd screening to identifyand remove PI individuals is undertaken

b. Monitoring options:i. Monitor production (reproduction effi-

ciency, neonatal and postnatal deathloss), monitor ill neonates for BVDVviremia, and necropsy (with laboratorysubmissions) as many abortions, still-births, and neonatal deaths as possible

ii. Use pooled samples of whole bloodtaken from suckling calves prior tobreeding for PCR testing

iii. Use serologic evaluation of sentinelanimals

iv. Use annual whole-herd screening priorto breeding season (all suckling calvesand purchased replacements)

II. Biosecurity for herd against BVDV entrancea. Isolate herd during breeding/gestation from

cattle of unknown BVD PI status(avoid fence-line contact with stocker cattle,neighbors’ cattle, etc.)b. Vaccinate to reduce virus circulation and to

reduce production of PI calves in face ofvirus circulation

i. Use MLV vaccine two or more timesbetween weaning and two months priorto first breeding as heifers

ii. Use vaccines according to label tobooster cowherd annually

c. Biosecurity to prevent introduction of virusinto herd

i. Test purchased bulls and heifersii. Test raised replacement heifers and

bullsiii. Monitor population for evidence of

introduction of PI animalsd. Purchasing bred heifers

i. If heifers are tested prior to purchase,still need to test calf after birth.

ii. If heifers are not tested prior to pur-chase, can test heifer prior to entry intoherd and then test calf after birth, orisolate heifer until calf is born and testthe calf (only need to test heifer if calfis positive).

iii. Percutaneous collection of fetal fluidsfor detection of PI calf has beendescribed – not recommended at thistime

iv. Serology in late gestation to identify PI-carrying unvaccinated dams

(Good negative predictive value, poor posi-tive predictive value)

e. Fomites i. Maintain sanitation to prevent viral

spread via nose tongs, injection needles,palpation sleeves, etc.


Persistently infected cattle are a primary source ofBVDV transmission to in-contact susceptible cattleduring marketing, trucking, and while in feedingpens and pastures. Vaccination is currently the pri-mary control intervention for BVDV in stocker andfeedlot operations. Screening cattle for the presenceof PI individuals prior to purchase or at arrival hasnot been adequately evaluated for economic return.The economic return will depend on the prevalenceof PI cattle, the sensitivity and specificity of the testused, and the economic cost of the disease to theoperation.

REFERENCESBaker JC: 1987, Bovine viral diarrhea virus: A review.

J Am Vet Med Assoc 190:1449–1458.Barber DML, Nettleton PF, Herring JA: 1985, Disease

in a dairy herd associated with the introduction andspread of bovine virus diarrhoea virus. Vet Rec117:459–464.

Baszler TV, Evermann JF, Kaylor PS, et al.: 1995,Diagnosis of naturally occurring bovine viral diar-rhea virus infections in ruminants using monoclonalantibody-based immunohistochemistry. Vet Pathol32:609–618.


Table 14.6. Control program for BVDV in dairyherds.

I. Monitor herd for risk of PI presencea. If PI presence is confirmed or strongly sus-

pected, a whole-herd screening to identifyand remove PI individuals is undertaken

b. Monitoring options:i. Monitor production (reproduction effi-

ciency, neonatal and postnatal deathloss), monitor ill neonates for BVDVviremia, and necropsy (with laboratorysubmissions) as many abortions, still-births, and neonatal deaths as possible

ii. Use serologic evaluation of sentinelanimals

iii. Test all heifer calves that may be savedas replacements (IHC of skin samples or PCR of pooledwhole blood samples)

II. Biosecurity to prevent introduction of BVDVinto herda. Test all heifer calves that are potential re-

placements if raising own replacementsb. Test purchased replacement heifers and

bulls prior to or at deliveryc. Obtain replacement heifers from source that

has a strict biosecurity program in placed. Use vaccines according to label to booster

cowherd e. Monitor population for evidence of intro-

duction of PI animalsIII. Biosecurity for heifer raisers

a. Test all heifers prior to or at arrival at devel-opment facility/farm

b. Isolate heifers during breeding/gestationfrom cattle of unknown BVD PI status

c. Vaccinate with MLV product two or moretimes between six months of age and startof breeding

Table 14.7. Control program for BVDV instocker/feedlot operations.

I. Vaccination of incoming cattle with MLVproduct

II. Screening incoming cattle for BVDV PIanimalsa. Low prevalence causes even a test with high

specificity to have more false positive thantrue positives

b. Testing should be done prior to cattle beingcommingled and trucked. Transmission andinfection likely to occur during stress andcrowding of commingling and trucking.

III. Purchasing PI-free certified cattlea. All cattle in a group are test negative

(IHC of skin sample or pooled PCR ofwhole blood)

b. Economic benefit is determined by multi-plying the cost of having a PI feeder calfpresent (increased pen morbidity, mortality,treatment failure, and performance) by theexpected prevalence for similar cattle.

IV. Purchasing PI-low risk cattlea. All cattle in the group originate from

farm(s) with a complete vaccination pro-gram and BVDV PI screening protocol

V. Purchasing cattle of unknown PI riska. Cost of unknown status is determined by

multiplying the cost of having a PI calfpresent by the expected prevalence forsimilar cattle

b. Cost of unknown PI risk is added to othercosts for break-even calculation


Bezek DM, Stofregen D, Posso M: 1995, Effect of cy-topathic bovine viral diarrhea virus (BVDV) super-infection on viral antigen association with platelets,viremia, and specific antibody levels in two heiferspersistently infected with BVDV. J Vet Diagn Invest7:395–397.

Bielanski A, Nadin-Davis S, Sapp T, Lutze-WallaceC: 2000, Viral contamination of embryos cryopre-served in liquid nitrogen. Cryobiology 40:110–116.

Binkhorst GJ, Journee DLH, Wouda W, et al.: 1983,Neurological disorders, virus persistence and hy-pomyelination in calves due to intrauterine infec-tion with bovine virus diarrhoea virus. I. Clinicalsymptoms and morphological lesions. Vet Q5:145–155.

Bolin SR: 1990, The current understanding about thepathogenesis and clinical forms of BVD. Vet Med85:1124–1132.

Bolin SR, McLurkin AW, Cutlip RC, Coria MF: 1985,Frequency of persistent bovine viral diarrhea virusinfection in selected cattle herds. Am J Vet Res46:2385–2387.

Brock KV, Cortese VS: 2001, Experimental fetal chal-lenge using type II bovine viral diarrhea virus incattle vaccinated with modified-live vaccine. VetTherapeutics 2:354–360.

Brock KV, Grooms DL, Ridpath, J, Bolin SR: 1998,Changes in levels of viremia in cattle persistentlyinfected with bovine viral diarrhea virus. J VetDiagn Invest 10:22–26.

Brock KV, Redman DR, Vickers ML, Irvine NE: 1991,Quantification of bovine viral diarrhea virus in em-bryo transfer flush fluids collected from a persis-tently infected heifer. J Vet Diagn Invest 3:99–100.

Brownlie J: 1990, The pathogenesis of bovine virusdiarrhoea virus infections. Rev Sci Tech Off Int Epiz9:43–59.

Brownlie J, Clarke MC, Hooper LB, Bell GD: 1995,Protection of the bovine fetus from bovine viral di-arrhea virus by means of a new inactivated vaccine.Vet Rec 137:58–62.

Brownlie J, Clarke MC, Howard CJ, Pocock DH:1987, Pathogenesis and epidemiology of bovinevirus diarrhoea virus infection of cattle. Annls RechVet 18:157–166.

Callan RJ, Schnackel JA, van Campen H, et al.: 2002,Percutaneous collection of fetal fluids for detectionof bovine viral diarrhea virus infection in cattle. JAm Vet Med Assoc 220:1348–1352.

Carlsson U, Belak K: 1987, Border disease virustransmitted to sheep and cattle by a persistently in-fected ewe: Epidemiology and control. Acta VetScand 35:79–88.

Casaro APE, Kendrick JW, Kennedy PC: 1971,Response of the bovine fetus to bovine viral diar-

rhea-mucosal disease virus. Am J Vet Res32:1543–1562.


Cortese VS, Grooms DL, Ellis J, et al.: 1998a,Protection of pregnant cattle and their fetusesagainst infection with bovine viral diarrhea virustype 1 by use of a modified-live virus vaccine. AmJ Vet Res 59:1409–1413.

Cortese VS, West KH, Hassard LE, et al.: 1998b,Clinical and immunologic responses of vaccinatedand unvaccinated calves to infection with a virulenttype-II isolate of bovine viral diarrhea virus. J AmVet Med Assoc 213:1312–1319.

Cortese VS, Whittaker R, Ellis J, et al.: 1998c,Specificity and duration of neutralizing antibodiesinduced in healthy cattle after administration of amodified-live virus vaccine against bovine viraldiarrhea. Am J Vet Res 59:848–850.

Davidson WR, Crow CB: 1983, Parasites, diseases,and health status of sympatric populations of sikadeer and white-tailed deer in Maryland andVirginia. J Wildlife Dis 19:345–348.


DuBois WR, Cooper VL, Duffy JC, et al.: 2000, Apreliminary evaluation of the effect of vaccinationwith modified live bovine viral diarrhea virus(BVDV) on detection of BVDV antigen in skinbiopsies using immunohistochemical methods. BovPract 34:98–100.

Dubovi EJ: 1996, Laboratory diagnosis of bovineviral diarrhea virus infections. Vet Med 91:867–872.

Duffell SJ, Harkness JW: 1985, Bovine virus diar-rhea—mucosal disease infection in cattle. Vet Rec117:240–245.

Ellis JA, Martin K, Norman GR, Haines DM: 1995,Comparison of detection methods for bovine viraldiarrhea virus in bovine abortions and neonataldeath. J Vet Diagn Invest 7:433–436.

Frolich K, Thiede S, Kozikowski T, Jakob W: 2002, Areview of mutual transmission of important infec-tious diseases between livestock and wildlife inEurope. Ann New York Acad Sci 969:4–13.

Fulton RW, Purdy CW, Confer AW, et al.: 2000,Bovine viral diarrhea viral infections in feedercalves with respiratory disease: Interactions withPasteurella spp., parainfluenza virus 3, and bovinerespiratory syncytial virus. Can J Vet Res64:151–159.




Gunn HM: 1993, Role of fomites and flies in thetransmission of bovine viral diarrhoea virus. VetRec 132:584–585.

Hamers C, Couvreur B, Dehan P, et al.: 2000a,Differences in experimental virulence of bovineviral diarrhoea viral strains isolated from haemor-rhagic syndromes. Vet J 160:250–258.

Hamers C, di Valentin E, Lecomte C, et al.: 2000b,Virus neutralizing antibodies against a panel of 18BVDV isolates in calves vaccinated with RispovalRS-BVD. J Vet Med B 47:721–726.

Hamers C, di Valentin E, Lecomte C, et al.: 2002,Virus neutralizing antibodies against 22 bovineviral diarrhoea virus isolates in vaccinated calves.Vet J 163:61–67.

Harkness JW: 1987, The control of bovine viraldiarrhea virus infection. Ann Rech Vet 18:167–174.

Harkness JW, Roeder PL, Drew T, et al.: 1987, Theefficacy of an experimental inactivated BVD-BDvaccine. In: Pestivirus Infection of Ruminants. Ed.J.W. Harkness. Brussels, Commission of theEuropean Communities. pp. 233–251.

Houe H: 1992, Serological analysis of a small herdsample to predict presence or absence of animalspersistently infected with bovine viral diarrhoeavirus (BVDV) in dairy herds. Res Vet Sci53:320–323.

Houe H: 1993, Survivorship of animals persistentlyinfected with bovine virus diarrhoea virus (BVDV).Prev Vet Med 15:275–283.

Houe H: 1995, Epidemiology of bovine viral diarrheavirus. Vet Clin N Amer: Food Anim Pract11:521–547.


Houe H, Baker JC, Maes RK, et al.: 1995, Prevalenceof cattle persistently infected with bovine viraldiarrhea virus in 20 dairy herds in two counties incentral Michigan and comparison of prevalence ofantibody-positive cattle among herds with differentinfection and vaccination status. J Vet Diagn Invest7:321–326.

Houe H, Meyling A: 1991, Surveillance of cattleherds for bovine virus diarrhoea virus (BVDV) in-fection using data on reproduction and calf mortal-ity. Arch Virol Suppl 3:157–164.

Howard CJ, Brownlie J, Thomas LH: 1986, Pre-valence of bovine virus diarrhoea virus viremia incattle in the UK. Vet Rec 119:628–629.

Kelling CL: 1996, Planning bovine viral diarrheavirus vaccination programs. Vet Med 91:873–877.

Kelling CL, Grotelueschen DM, Smith DR, BrodersenBW: 2000a, Testing and management strategies foreffective beef and dairy herd BVDV biosecurityprograms. Bov Pract 34:13–22.

Kelling CL, Steffen DJ, Topliff CL, et al.: 2002,Comparative virulence of isolates of bovine viraldiarrhea virus type II in experimentally inoculatedsix- to nine-month-old calves. Am J Vet Res63:1379–1384.

Kelling CL, Stine LC, Rump KK, et al.: 2000b,Investigation of bovine viral diarrhea virus infec-tions in a range beef cattle herd. J Am Vet MedAssoc 197:589–593.


Larson RL, Pierce VL, Grotelueschen DM, WittumTE: 2002, Economic evaluation of beef cowherdscreening for cattle persistently infected withbovine viral diarrhea virus. Bov Pract 36:106–112.

Liess B, Moenning V: 1990, Ruminant pestivirus in-fection in pigs. Rev Sci Tech Off Int Epiz9:151–161.

Lindberg A, Groenendaal H, Alenius S, EmanuelsonU: 2001, Validation of a test for dams carrying fe-tuses persistently infected with bovine viral diar-rhoea virus based on determination of antibody lev-els in late pregnancy. Prev Vet Med 51:199–214.

Martin SW, Bateman KG, Shewen PE, et al.: 1989,The frequency, distribution and effects of antibodiesto seven putative respiratory pathogens on respira-tory disease and weight gain in feedlot calves inOntario. Can J Vet Res 53:355–362.



Meyling A., Houe H, Jensen AM: 1990,Epidemiology of bovine virus diarrhoea virus. RevSci Tech Off Int Epiz 9:75–93.

Meyling A, Jensen AM: 1988, Transmission of bovinevirus diarrhoea virus (BVDV) by artificial insemi-nation (AI) with semen from a persistently infectedbull. Vet Microbiol 17:97–105.

Meyling A, Ronsholt L, Dalsgaard K, Jensen AM:1987, Experimental exposure of vaccinated andnon-vaccinated pregnant cattle to isolates of bovineviral diarrhoea virus (BVDV). In: Pestivirus



Infection of Ruminants. Ed. J.W. Harkness.Brussels, Commission of the EuropeanCommunities. pp. 225–232.

Miller MA, Ramos-Vara JA, Kleiboeker SB, et al.:2002, Elimination of persistent bovine viral diar-rhea virus infection in a beef cattle herd. 45th AnnuMtg Amer Assoc Vet Lab Diagn, St. Louis, MO, Oct19–21, 2002. pp. 69.


Muñoz-Zanzi CA, Johnson WO, Thurmon MC,Hietala SK: 2000, Pooled-sample testing as a herd-screening tool for detection of bovine viral diarrheavirus persistently infected cattle. J Vet Diagn Invest12:195–203.

Nielsen SS, Roensholt L, Bitsch V: 2000, Bovinevirus diarrhea virus in free-living deer fromDenmark. J Wildl Dis 36:584–587.

Niskanen, R, Lindberg A, Larsson B, Alenius S: 2000,Lack of virus transmission from bovine viral diar-rhoea virus infected calves to susceptible peers.Acta Vet Scand 41:93–99.

Niskanen R, Lindberg A, Traven M: 2002, Failure tospread bovine virus diarrhoea virus infection fromprimarily infected calves despite concurrent infec-tion with bovine coronavirus. Vet J 163:251–259.


Orban S, Liess B, Hafez SM, et al.: 1983, Studies ontransplacental transmissibility of a bovine virus di-arrhoea (BVD) vaccine virus. I. Inoculation ofpregnant cows 15 to 90 days before parturition(190th to 265th day of gestation). Zbl Vet Med B30:619–634.

Palfi V, Houe H, Philipsen J: 1993, Studies on the de-cline of bovine virus diarrhoea virus (BVDV) ma-ternal antibodies and detectability of BVDV in per-sistently infected calves. Acta Vet Scand34:105–107.

Paton DJ, Goodey R, Brockman S, Wood L: 1989,Evaluation of the quality and virological status ofsemen from bulls acutely infected with BVDV. VetRec 124:63.


Rae AG, Sinclair JA, Nettleton PF: 1987, Survival ofbovine virus diarrhoea virus in blood from persist-ently infected cattle. Vet Rec 120:504.

Revell SG, Chasey D, Drew TW, Edwards S: 1988,Some observations on the semen of bulls persist-ently infected with bovine virus diarrhoea virus. VetRec 123:122–125.

Ridpath JF, Neill JD, Endsley J, Roth JA: 2003, Effectof passive immunity on the development of a pro-tective immune response against bovine viral diar-rhea virus in calves. Am J Vet Res 64:65–69.

Roeder PL, Harkness JW: 1986, BVD virus infection:Prospects for control. Vet Rec 118:143–147.


Singh EL: 1985, Disease control procedures for han-dling embryos. Rev Sci Tech Off Int Epiz4:867–872.

Singh EL, Eaglesome MD, Thomas FC, et al.: 1982,Embryo transfer as a means of controlling thetransmission of viral infections. I. The in vitro ex-posure of preimplantantion embryos to akabane,bluetongue and bovine viral diarrhea viruses.Theriogenology 17:437–444.

Straver PJ, Journee DLH, Binkhorst GJ: 1983,Neurological disorders, virus persistence and hy-pomyelination in calves due to intrauterine infec-tion with bovine virus diarrhoea virus. II. Virologyand epizootiology. Vet Q 5:156–164.

Tarry DW, Bernal L, Edwards S: 1991, Transmissionof bovine virus diarrhoea virus by biting flies. VetRec 128:82–84.

Terpstra C, Wensvoort G: 1988, Natural infections ofpigs with bovine viral diarrhea virus associatedwith signs resembling swine fever. Res Vet Sci45:137–142.

Traven M, Alenius S, Fossum C, Larsson B.: 1991,Primary bovine viral diarrhoea virus infection incalves following direct contact with a persistentlyviraemic calf. J Vet Med B 38:453–462.


Wentink GH, Van Exsel ACA, Goey I, van LieshoutJA: 1991, Spread of bovine virus diarrhea virus in aherd of heifer calves. Vet Q 13:233–236.

Whitmore HL, Zemjanis R, Olson J: 1981, Effect ofbovine viral diarrhea virus on conception in cattle.J Am Vet Med Assoc 178:1065–1067.

Wittum TE, Grotelueschen DM, Brock KV, et al.:2001, Persistent bovine viral diarrhoea virus infec-tion in U.S. beef herds. Prev Vet Med 49:83–94.

Zimmer G, Schoustra W, Graat EAM: 2002,Predictive values of serum and bulk milk samplingfor the presence of persistently infected BVDV car-riers in dairy herds. Res Vet Sci 72:75–82.



PROGRESS IN BVDV RESEARCHSince the first publication on bovine viral diarrheavirus (BVDV) 60 years ago (Olafson et al., 1946),significant progress has been made in elucidatingthe pathogenesis, transmission, diagnosis, and mo-lecular virology of this virus. We have come to un-derstand that the pathogenesis of BVDV is compli-cated and that it produces a multifaceted disease(Bolin et al., 1985; Brownlie et al., 1984; Corapi etal., 1989). Monoclonal antibodies (Mabs) have beendeveloped (Moennig et al., 1987) and the genome ofBVDV has been completely sequenced (Collett etal., 1988). Antigenic and genomic diversity has beenfound to be a hallmark of BVDV, making the controlof this virus difficult. In addition to different bio-types (Gillespie et al., 1960), different genotypesand subgenotypes have now been identified (Pelle-rin et al., 1994; Ridpath et al., 1994). Biotype differ-ences in cytopathology in culture are associatedwith the production of a nonstructural protein (NS3or P80) in the cytopathic strains (Meyers et al.,1991). Cytopathic effect observed in vitro (i.e., cy-topathic biotype) does not correlate with virulencein vivo, because all highly virulent viruses belong tothe noncytopathic biotype (Ridpath et al., 2000).Although hemorrhagic syndrome has been associ-ated only with viruses from the BVDV 2 genotype,variation in virulence exists in both the BVDV 1 andBVDV 2 genotype (Evermann and Ridpath, 2002;Fulton et al., 2000; Fulton et al., 2002; Liebler-Tenorio et al., 2003).

Numerous studies have been undertaken to eluci-date the mechanism of development of persistentlyinfected (PI) animals and how to control them(Harding et al., 2002; Ohmann, 1982; Ohmann etal., 1982; Van Campen and Woodard, 1997). It iswidely accepted that PI animals are a major sourceof virus on the farm and in feedlots. Although viral

shed is limited and normally confined to a period ofless than 7 days, transiently infected animals canalso be a source of virus to herdmates. Both acutelyand persistently infected bulls can shed virus in theirsemen with a minimal effect on the traditional meas-ures of semen quality (Givens et al., 2003; Kirklandet al., 1991; Kommisrud et al., 1996; Revell et al.,1988). The presence of BVDV in semen leads tovirus transmission by the reproductive route, result-ing in abortions, early embryonic death, congenitalinfections and PI animals (Givens et al., 2003; Kirk-land et al., 1991; Kommisrud et al., 1996; Kirklandet al., 1997; Meyling and Jensen, 1988; Paton et al.,1990; Schlafer et al., 1990).

Major advances have been made in detectingBVDV infection, including the use of monoclonalantibody– and polymerase chain reaction–basedtests. Although BVDV may be spread by animalsthat are either persistently or acutely infected, themain emphasis until now has been on the detectionof PI animals. This is because the removal of PI an-imals is considered to be integral to an effective con-trol strategy. Eradication programs in Sweden,Finland, and Denmark rely on nonvaccination ofcattle so that positive animals can be identified andremoved easily (Greiser-Wilke et al., 2003). Theseprograms are reported to be successful in reducingBVDV-positive herds. In Germany, vaccination withkilled vaccine followed by a booster dose of modi-fied live vaccine (MLV), in addition to testing andremoving PI animals, is used to reduce virus circu-lation in the herd (Moennig and Greiser-Wilke,2003). In the U.S., reliance is placed on detectionand removal of PI animals and vaccination of breed-ing animals before conception. Recent research sug-gests that, to be effective, BVDV vaccines shouldcontain at least a BVDV 1 and a BVDV 2 strain(Fulton et al., 2003).

239

15Conclusions and Future Research

Julia F. Ridpath and Sagar M. Goyal


BVDV research as summarized above has bene-fited from the efforts of a cadre of outstanding sci-entists and diagnosticians, as reported and illus-trated by the authors of the preceding chapters.However, BVDV has proven to be a difficult andchallenging subject and despite these efforts muchremains to be done. As this book goes to press thereis gathering momentum for the development and im-plementation of a national program for BVDV con-trol in the United States. Although eradication ofBVDV in the U.S. may be a difficult goal to achieve,the reduction of the losses currently caused byBVDV is a readily attainable goal and is well worthpursuing. Ideally, the drive to reduce BVDV losseswould be multifaceted and widely implemented. Itwould require the cooperation of producers, practi-tioners, diagnosticians, and regulators, and a refo-cusing of research and control efforts.

REFOCUSING BASIC RESEARCH

NONCYTOPATHIC BVDVTo date, heavy emphasis has been placed on exam-ining differences between cytopathic and noncyto-pathic viruses in vitro. While intriguing from a sci-entific standpoint, such research does little towardsreducing the incidence of BVDV in the field.Cytopathic viruses are rare and isolated only in as-sociation with mucosal disease or postvaccinal dis-ease resulting from inoculation with a cytopathicBVDV vaccine. Mucosal disease, although an inter-esting phenomenon, is not a source of major eco-nomic loss for producers. The major economic im-pact of BVDV infection is the result of lossesassociated with reproductive or respiratory disease,which are almost always the result of infection witha noncytopathic virus. In contrast, no clinical out-breaks of acute disease have been traced to infectionwith a cytopathic BVDV.

Immunosuppression and acerbation of secondaryinfections, associated with infection with noncyto-pathic BVDV, also contribute to economic losses.There is no evidence to indicate that acute, uncom-plicated infections with cytopathic BVDV are moreclinically severe than infection with noncytopathicBVDV. In fact, the most clinically severe forms ofacute BVDV infections are associated with noncyto-pathic BVDV. Thus, comparing cytopathic and non-cytopathic viruses in vitro yields little informationthat would help limit BVDV infections in vivo.Efforts to limit the damage caused by BVDV infec-tions would be better served by research that ad-dresses the real significance of noncytopathic

BVDV strains and the host’s response to them. Inaddition, the nature of immunosuppression and ofprotective immune response engendered by BVDVshould be studied. Because noncytopathic virus mayestablish persistent infections, all available BVDVvaccines contain only cytopathic BVDV. It is notknown, however, if cytopathic BVDV contained invaccines are any safer than the noncytopathicviruses, especially for fetuses.

T-CELL RESPONSES

Considerable amount of information is available onB-cell immune response to BVDV vaccines, but lit-tle information is available regarding the T-cell re-sponses. This is partly due to the availability of sim-ple and reliable technology to examine and compareB-cell responses, as expressed by neutralizing anti-bodies present in serum. Unfortunately, such tech-nology is not yet available for comparing T-cell re-sponses, although it appears that the T-cell responseis as important, if not more important, as B-cell re-sponse in the development of acquired immunityagainst BVDV. Research that focuses on methods toimprove T-cell response to vaccination is sorelyneeded. This is dependent upon developing simple,robust, and reliable methods for measuring andcomparing T-cell responses in infected, vaccinated,and nonvaccinated animals.

ACUTE, PROLONGED INFECTIONS

Historically, BVDV infections have been catego-rized as acute or persistent. Persistent infections aredefined as lifelong infections resulting from in uteroexposure of the animal to a noncytopathic BVDV.Lifelong infections result because the animal devel-ops a specific immune tolerance for the virus towhich it was exposed in utero. Acute infections, onthe other hand, are defined as infections in which theimmune system clears the virus within 14 days.However, the real picture is somewhat more compli-cated. For example, some animals exposed afterbirth (acute infection) may mount an immune re-sponse but fail to clear the virus within 14 days,some animals may clear the virus after a prolongedviral shed, and others may allow the virus to repli-cate in immunologically privileged sites such astestes and ovaries (see Chapter 9).

The damage resulting from acute infections maynot be over after the virus has been cleared. Thus,there may be lingering problems that are not de-tected until several months after viremia has passed.Prolonged infections and persistent infectionswithin privileged sites also contribute to the spread



of BVDV in cattle populations. Current surveillanceprograms, designed to detect classic persistently in-fected animals, will invariably miss these types ofinfections. In addition, the true cost of BVDV mustfactor in the problems in production that occur afterthe virus has been shed. This is difficult to track be-cause onset of acute infection and its repercussionson animal production may be months apart.

LOW-VIRULENCE STRAINS OF BVDVIt is easy to determine the impact of infection withhighly virulent strains of BVDV. The impact of low-virulence strains, on the other hand, is difficult to as-certain. However, infection with these strains shouldnot be considered a benign event even though theireffects may not be immediately apparent. Infectionwith low-virulence strains of animals that are ac-tively laying down bone and muscle tissues mayhave long-term effects on their growth potential. Inaddition, low-virulence strains may lead to immuno-suppression, setting up the animal for more severesecondary infections.

The impact of low-virulence strains on vulnerablepopulations of cattle should also be studied in detail.Examples of vulnerable animals include the fetus,weaned animals from which passive antibodies havedisappeared or are waning, animals newly intro-duced into the herd, and stressed animals. Biosecur-ity to prevent the introduction of BVDV in animalherds and adequate vaccination programs are partic-ularly important to limit infection in vulnerable ani-mals. The significance of BVDV isolates from gi-raffe, antelope, and reindeer also need to be studied.

REFOCUSING CONTROLEFFORTSHistorically, BVDV control efforts have been basedon vaccination. However, 40+ years of vaccinationhave not resulted in a significant decrease in BVDVlosses nationwide. Although vaccination can be aneffective component of a BVDV control plan, it isnot a stand-alone answer to the problem. For a con-trol effort to be effective, proactive managementplans must be in place, and the development of suchplans requires a complete understanding of BVDVtransmission within populations, surveillance strate-gies, biosecurity on the farm, and the role of vacci-nation.

TRUE IMPACT OF BVDVThe compelling reason for a BVDV control programis that reducing BVDV infections is cost-effective.This premise assumes that the expenses incurred in

preventing BVDV infections are offset by savingsrealized by more efficient animal production. Al-though most producers agree that BVDV infectionsare detrimental, the full extent of BVDV’s impact onproduction is frequently underestimated. In a lay-man’s mind, BVDV-associated reproductive diseasepresents as the birth of PI animals. However, thebirth of PI animals following the introduction ofBVDV into a group of dams may represent just thetip of the iceberg. Clinical presentation of BVDV-associated reproductive disease also includes failureto conceive (repeat breeding), abortion, mummifica-tion, congenital abnormalities, and the birth ofneonates that appear normal but fail to thrive. Inorder to determine the true impact of BVDV repro-ductive disease, animal producers and cliniciansneed to be alert to the many different forms ofBVDV-induced reproductive problems.

BIOSECURITY ON THE FARM

Vaccination is a component of the control planrather than an end unto itself and vaccines shouldnot be considered “silver bullets” in the eradicationof BVDV. In the absence of a good managementprogram, which by definition would include a biose-curity plan, protection via vaccination will eventu-ally fail. Preventing long- and short-term exposuresto BVDV by limiting the introduction of BVDV intoa herd is as important as protecting the animals byvaccination. Thus, PI animals must be removed andnew additions to the herd should be isolated untiltheir BVDV-free status has been established. In thecase of bred heifers, BVDV-free status cannot be as-sured until the calf has been tested. Care should betaken to assure that a good BVDV management pro-gram is in place for all parties when facilities andequipment are shared between production units. Aproducer’s biosecurity program is only as good asthat of the neighbor he shares a fence with or the fel-low exhibitor with whom he shares a show ring.

DISEASE SURVEILLANCE

The present testing for BVDV is sporadic and isusually done in response to an outbreak of the dis-ease and focuses on the detection of PI animals.However, exposure to BVDV may result in subclin-ical disease that may impact profitability but is notimmediately apparent from casual observation ofthe herd. Routine BVDV surveillance programswould be more effective at detecting and controllingBVDV outbreaks. Further, not all outbreaks ofBVDV-related disease can be traced to contact witha PI animal. PI animals are certainly a major factor

Conclusions and Future Research 241


in keeping BVDV in circulation, but acute BVDVinfections cannot be ignored for surveillance pro-grams to be effective. As discussed above, non-PIanimals may experience prolonged viremia or mayhave localized infections that are not cleared.Further, aborted or stillborn fetuses may serve assources of BVDV infection for a herd. Effectivecontrol programs must include the use of reliabletests for the detection of both acute and persistentinfections and examination of aborted and stillbornfetuses.

VACCINE DESIGN

When developing vaccines, one of the questions thatshould be asked is whether the strains used in thevaccine reflect and protect against viruses the animalwill be exposed to in the field. The first BVDV vac-cines were produced in the 1960s. Viral strains usedin these vaccines—e.g., Singer, NADC, NY-1, andC24V (Oregon strain), are still found in vaccines 40years later. Retrospective analysis of these strains hasshown that all of them belong to a single branch ofthe pestivirus family tree—e.g., BVDV genotype 1.In the early 1990s, a newly recognized group ofBVDV field isolates, eventually termed BVDV geno-type 2, was found to break through BVDV vaccina-tion programs. A second generation of BVDV vac-cines that contain the old vaccine strains plus strainsfrom the BVDV 2 genotype are now becoming avail-able. Time will tell whether these vaccines are moreefficacious than the old ones.

As more BVDV strains are characterized, sub-genotypes of BVDV are being recognized withinboth BVDV 1 and BVDV 2 genotypes. Thus, BVDV1 has been subdivided into BVDV 1a and BVDV 1b.Recent surveys in the U.S. have shown that BVDV1b is found most frequently in association with clin-ical disease and that most BVDV 1 strains isolatedfrom the field belong to the BVDV 1b subgenotype.However, when BVDV 1 strains used in vaccineswere characterized, it was found that they predomi-nantly belonged to the BVDV 1a subgenotype. Thepractical significance of this finding is not clear andit is also not known whether the inclusion of BVDV1b in vaccines would offer better protection. Furtherresearch is needed to answer these questions.

The observation of incomplete fetal protectionfollowing prebreeding vaccination should also beexamined. Currently used vaccines are evaluated bydetermining whether they protect against BVDV-induced respiratory disease. Although some vaccinemanufacturers have started to evaluate their vaccinesfor fetal protection, more needs to be done in this

area. Current vaccines with fetal protection labelclaims were licensed based on a one-time challengeof BVDV. Animals in contact with PI animals willbe challenged by virus shed by the PI on a dailybasis. It is not known how well vaccination standsup to such repeated challenges. Further, fetal protec-tion appears to be dependent upon the prevention ofviremia, not reduction of viremia or clinical disease.Thus the “perfect” BVDV vaccine would prevent in-fection rather than disease and induce both cellularand humoral immune responses.

EFFECTS OF STRESS ON VACCINE EFFICACY

Vaccines are commonly administered when animalsare gathered for weaning, sorting, branding, andshipping. Unfortunately, the animal is under stressfrom a number of different factors at these times andits immune response is not functioning at its peak.Vaccination at these times may not result in an opti-mum immune response and may even cause tran-sient losses in production. Thus, management prac-tices may impact on the efficacy of the vaccine andshould be taken into account when implementing acontrol program with vaccination.

OTHER QUESTIONSThe recognition of severe acute BVD in the early1990s changed our perception of BVDV-induceddisease severity. It is important to continue to moni-tor the role of newly emerging strains of BVDV ondisease severity and on acute and persistent infec-tions. The role of vaccination in the emergence ofnew BVDV strains should be monitored, and thecurrent methods for the detection of BVDV insemen samples need improvement. The develop-ment of a prenatal test to determine the existence ofPI animals would also be useful.

REFERENCESBolin SR, McClurkin AW, Cutlip RC, et al.: 1985,

Severe clinical disease induced in cattle persistentlyinfected with noncytopathic bovine viral diarrheavirus by superinfection with cytopathic bovine viraldiarrhea virus. Am J Vet Res 46:573–576.

Brownlie J, Clarke MC, Howard CJ. 1984:Experimental production of fatal mucosal disease incattle. Vet Rec 114:535–536.

Collett MS, Larson R, Belzer SK, et al.: 1988,Proteins encoded by bovine viral diarrhea virus:The genomic organization of a pestivirus. Virology165:200–208.

Corapi WV, French TW, Dubovi EJ: 1989, Severethrombocytopenia in young calves experimentally



infected with noncytopathic bovine viral diarrheavirus. J Virol 63:3934–3943.


Fulton RW, Ridpath JF, Confer AW, et al.: 2003,Bovine viral diarrhoea virus antigenic diversity:Impact on disease and vaccination programmes.Biologicals 31:89–95.


Fulton RW, Saliki JT, Confer AW, et al.: 2000, Bovineviral diarrhea virus cytopathic and noncytopathicbiotypes and type 1 and 2 genotypes in diagnosticlaboratory accessions: Clinical and necropsy sam-ples from cattle. J Vet Diagn Invest 12:33–38.

Gillespie J, Baker J, McEntee K: 1960, A cytopatho-genic strain of virus diarrhea virus. Cornell Vet50:73–79.

Givens MD, Heath AM, Brock KV, et al.: 2003,Detection of bovine viral diarrhea virus in semenobtained after inoculation of seronegative postpu-bertal bulls. Am J Vet Res 64:428–434.

Greiser-Wilke I, Grummer B, Moennig V: 2003,Bovine viral diarrhoea eradication and control pro-grammes in Europe. Biologicals 31:113–118.

Harding MJ, Cao X, Shams H, et al.: 2002, Role ofbovine viral diarrhea virus biotype in the establish-ment of fetal infections. Am J Vet Res63:1455–1463.

Kirkland PD, McGowan MR, Mackintosh SG, et al.:1997, Insemination of cattle with semen from a bulltransiently infected with pestivirus. Vet Rec140:124–127.

Kirkland PD, Richards SG, Rothwell JT, et al.: 1991,Replication of bovine viral diarrhoea virus in thebovine reproductive tract and excretion of virus insemen during acute and chronic infections. Vet Rec128:587–590.

Kommisrud E, Vatn T, Lang-Ree JR, et al.: 1996,Bovine virus diarrhoea virus in semen from acutelyinfected bulls. Acta Vet Scand 37:41–47.

Liebler-Tenorio EM, Ridpath JF, Neill JD: 2003,Lesions and tissue distribution of viral antigen insevere acute versus subclinical acute infection withBVDV 2. Biologicals 31:119–122.

Meyers G, Rumenapf T, Tautz N, et al.: 1991,Insertion of cellular sequences in the genome of

bovine viral diarrhea virus. Arch Virol Suppl3:133–142.

Meyling A, Jensen AM: 1988, Transmission of bovinevirus diarrhoea virus (BVDV) by artificial insemi-nation (AI) with semen from a persistently-infectedbull. Vet Microbiol 17:97–105.

Moennig V, Bolin SR, Coulibaly CO, et al.: 1987,Studies of the antigen structure of pestivirusesusing monoclonal antibodies. DTW Dtsch TierarztlWochenschr 94:572–576.

Moennig V, Greiser–Wilke I: 2003, Perspectives onBVD eradication in Germany. Berl Munch TierarztlWochenschr 116:222–226.

Ohmann HB: 1982, Experimental fetal infection withbovine viral diarrhea virus. II. Morphological reac-tions and distribution of viral antigen. Can J CompMed 46:363–369.

Ohmann HB, Jensen MH, Sorensen KJ, et al.: 1982,Experimental fetal infection with bovine viral diar-rhea virus. I. Virological and serological studies.Can J Comp Med 46:357–362.

Olafson P, MacCallum AD, Fox FH. 1946: An appar-ently new transmissible disease of cattle. CornellVet 36:205–213.


Pellerin C, van den Hurk J, Lecomte J, et al.: 1994,Identification of a new group of bovine viral diar-rhea virus strains associated with severe outbreaksand high mortalities. Virology 203:260–268.

Revell SG, Chasey D, Drew TW, et al.: 1988, Someobservations on the semen of bulls persistently in-fected with bovine virus diarrhoea virus. Vet Rec123:122–125.


Ridpath JF, Neill JD, Frey M, et al.: 2000, Phylo-genetic, antigenic and clinical characterization oftype 2 BVDV from North America. Vet Microbiol77:145–155.

Schlafer DH, Gillespie JH, Foote RH, et al.: 1990,Experimental transmission of bovine viral diseasesby insemination with contaminated semen or duringembryo transfer. DTW Dtsch Tierarztl Wochenschr97:68–72.

van Campen H, Woodard L: 1997, Fetal infection maynot be preventable with BVDV vaccines. J Am VetMed Assoc 210:480.

Conclusions and Future Research 243


Note: Italicized page numbers indicate table or figure.

Abattoirsgenital tracts obtained from, 198sampling from, 38

Abortions, 5, 8, 53, 57, 59, 93, 107causes of, 114, 223, 224, 241fetal infection and, 128–129initial reports on, 147

Academy of Veterinary Consultants, BVDV positionstatement by, 25–26

AC-ELISA. See Antigen capture ELISAActive humoral immune response, 158Active immunity, 105Acute, prolonged infections, 240–241Acute BVDV infections, 121, 123–126

bovine respiratory disease and, 105esophageal ulcers in cattle suffering from, 112horizontal transmission of BVDV and, 94immune suppression in, 105, 125–126role of viral and host factors in pathogenesis of, 121,

123spread of BVD of low and high virulence in, 124viral shedding and, 130–131virus spread and development of lesions in, 123–125

comparing strains of high and low virulence, 125strains of high virulence, 124–125strains of low virulence, 123–124

Acutely infected animals, BVDV transmission from,97–98

Acute mucosal disease, clinical signs of, 112Adequate contact, herd immunity and, 93Adequate contacts per time period, number of, 91Aerosols, role of, in acquiring BVDV, 106Agar gel-diffusion precipitin tests, 6Agarose gel electrophoresis, PCR product identified by,

202Akkina, R.K., 20, 203Alopecia, 150, 224Alpacas, 116Alveolar macrophages, 157, 161American Bison, 173

American Type Culture Collection, 23Antelocapridae family, serologic surveys of ruminant

species within, 173Antibodies against BVDV, epidemiological studies for

estimation of prevalence of animals with, 39–42Antibody carriers, prevalence of, 51Antibody decay rate, calves susceptibility of infection

and, 98Antibody detection, 200Antibody status, of individual animals, 50Anticoagulant toxicity, 111Antigen capture ELISA, 199Antigenic diversity, 209, 210, 239Antigenic variation, among BVDV 1 and BVDV 2

strains, 66Antigen nucleic acid, detection of, 198Antigen-presenting cells, 158, 164Anti-idiopathic Mabs, 178AP. See Apparent prevalenceAPAF-1, 186APC. See Antigen-presenting cellsApoptosis, 22, 157, 158, 186, 187Apparent prevalence, 36AR. See Attributable riskArcanobacter pyogenes, 172Archbald, L.F., 148Arsenic poisoning, 110Artificial insemination (AI) centers

prevalence of PI at, 95testing requirements for bulls in, 17

Artiodactyla order, occurrence in ungulates belonging to,115–116

Ataxia, 129, 150ATPase, 82Attributable risk, 50Atypical persistent infection, 16–17Austria, prevalence in, 46AVC. See Academy of Veterinary ConsultantsAxis deer, 116

Baigent, S.J., 188Baker, J.C., 13, 14

245

Index

2550_Goyal_Bovine Viral Diarrhea Virus_Index 3/18/05 11:51 AM

Barlow, R.M., 10Baule, C., 14, 22BBMM. See Bovine bone marrow-derived macrophagesB-cells, 164, 240BCV. See Bovine respiratory coronavirusBDV. See Border disease virusBeaudeau, F., 201Becher, P., 15Beef calves, age-matched, from herd suffering from

mucosal disease, 112Beef cattle, control programs and, 233Beef cow herds, control program for BVDV in, 234Beer, M., 214Belgium

BVDV genotypes identified in, 46prevalence in, 52

Bennett, R.M., 58BEV-3. See Bovine enterovirus-3Bhudevi, B., 202BHV-1. See Bovine herpes virus-1Bielefeldt-Ohmann, H., 12Biologics, possible BVDV contamination of, 171Biosecurity

control programs and, 209for dairy herds, 234, 235on farm, 241importance of, 152inter-herd transmission and, 101to prevent herd exposure to PI animals, 233whole-herd testing and, 230

BiotypesBVDV, 73–74practical significance of, 74recurrent shedding and, 98

Bison, cell lines survey and, 173Bitsch, V., 21Bittle, J.L., 6Blindness, in calves, 164Blue tongue, 110B-lymphocytes

decrease of, in acute BVDV infection, 125lymphoid tissue lesions and, 134

Bolin, S.R., 9, 13, 15, 16, 203Bone marrow, effect of BVDV on, 163Border disease virus, 5, 65, 171, 198, 218Bovidec (C-vet), 217Bovine bone marrow-derived macrophages, 187Bovine enterovirus-3, 177Bovine herpes virus-1, 161, 210Bovine papular stomatitis, 114Bovine polyclonal antibody (Pab)-based IPMA, 198Bovine respiratory coronavirus, 161Bovine respiratory disease, 105, 106, 115, 161Bovine respiratory syncytial virus, 110, 125, 161, 210Bovine species, BVDV adaptation for replication in,

151Bovine turbinate cells, 199Bovine viral diarrhea-mucosal disease, 4

Bovine viral diarrhea virus, xi, 3acute and persistent infections in bull, 10atypical persistent infection, 16–17binding and entry of, 82biotypes, 73–74

practical significance of, 74characterizations of, 4–5circulation of, in cattle populations, 122clinical/subclinical manifestations of and sequelae to

congenital infection with, 106control by vaccination, 20–21, 25control strategies for, 12–14control without vaccination, 21detection of, 12diagnosis and control

1960s, 5–61970s, 7–8

diagnosis by virus isolation, 18economic impact of, 56–58, 59effects/consequences of, on disease and production,

53–56entry of into cells, translation, and replication, 68–71epidemiological framework for description of occur-

rence of, 36eradication programs, 25–26experimental production of persistent infection and

mucosal disease, 8–10factors affecting transmission of, 91–92forms of, 105future research on, 165genetic recombination and spontaneous and post-

vaccinal mucosal disease, 15genotypes, 71–73

differences between BVDV 1 and BVDV 2, 73prevalence of, 72similarities between viruses from BVDV 1 and 2

genotype, 72–73subgenotypes of BVDV 1 and BVDV 2, 73

heterogeneity of, 74immune responses to, 157–159isolates/analysis of genotype compared with clinical

presentation, 110late-onset mucosal disease, 15–16measurable/quantifiable epidemiological variables,

35–38molecular actions of cytopathic and noncytopathic

BVDV, 22–23molecular biology advances and, 11–12, 17–18monoclonal antibody-based tests for, 18–19normal calves from persistently infected cows and, 16old cytopathic BVDV strains, 23–24outcomes and factors influencing outcome of infec-

tions from, 122overall view of infection from, 190persistent infections in sick and apparently healthy

cattle, 7phylogenetic studies, 14–15polymerase chain reaction, 19–20

246 Index


in pregnant cattle, 107, 109prevalence of infections with, 36–38, 46principal reservoirs of, 223progress in research on, 239–240refocusing research on, 240–241reproductive effects of, 223–224respiratory disease, 10–11respiratory disease, 2000-present, 21–22risk factors for occurrence of infections, 49–52skin biopsies and, 24subclinical and clinical manifestation of, 108–109taxonomy — defining characteristics of, 65–67thrombocytopenia, 11transient infection by, in combination with other

pathogens, 54–55true impact of, 241type 2, emergence of, 3, 14various outcomes of, depending on transmission and

host, 225Bovine viral diarrhea virus 1 (BVDV 1), 65, 209

differences between viruses from BVDV 2 and, 73similarities between viruses from BVDV 2 and, 72–73subgenotypes of, 210transplacental infection with, 127vaccine design and, 242

Bovine viral diarrhea virus 1a (BVDV 1a), 209dendogram representing relatedness of nucleotide

sequences from BVDV 1b, BVDV 2 and, 211Bovine viral diarrhea virus 1b (BVDV 1b), 209Bovine viral diarrhea virus 2 (BVDV 2), xi, 65, 209

differences between viruses from BVDV 1 and, 73incorporating strains of, in vaccines, 21, 25misconceptions related to, 72monoclonal antibodies and, 18similarities between viruses from BVDV 1 and, 72–73subgenotypes of, 211transplacental infection with, 127vaccine design and, 242virulence factors and, 121, 123

Bovine viral diarrhea virus 2a (BVDV 2a), 211Bovine viral diarrhea virus 2b (BVDV 2b), 211Brachygnathism, 224Bracken fern toxicity, 111BRD. See Bovine respiratory diseaseBrinkhof, J., 19Brock, K.V., 16, 18, 67, 68, 201, 214Bronchopneumonia, colostrum source and, 54Brownlie, J., 8, 9, 15BRSV. See Bovine respiratory syncytial virusBT cells. See Bovine turbinate cellsBuffy coat cells, 198, 199Buffy coat isolation, 204Bulk milk testing, in Scandinavia, 21Bulk tank milk

antibody levels in Sweden, 48samples, 37

Bullsacute and persistent infections in, 10

persistently infected, BVDV transmission and, 230reducing BVDV transmission related to, 95testicular infection and fertility of, 145–146

BVD. See Bovine viral diarrheaBVD-MD. See Bovine viral diarrhea-mucosal diseaseBVDV. See Bovine viral diarrhea virusBVDV genomic RNA, secondary structure of 5’

UTR/IRES of, 181BVDV 1. See Bovine viral diarrhea virus 1BVDV 1a. See Bovine viral diarrhea virus 1aBVDV 1b. See Bovine viral diarrhea virus 1bBVDV-specific nucleic acids, 25BVDV 2. See Bovine viral diarrhea virus 2BVDV 2a. See Bovine viral diarrhea virus 2a BVDV 2b. See Bovine viral diarrhea virus 2b

Calvesbeef, from herd suffering from mucosal disease, 112birth of, from persistently infected cow, 16blindness in, 164colon lesions in, from mucosal disease, 113colostral immunity and vaccination of, 231–232colostral protection and, 98colostrum-deprived, BVDV in, 106coronavirus infections in, 125esophageal ulcers in, from mucosal disease, 112histologic sections from ileum of both uninfected and

infected calves, 164occurrence of other diseases and transient infection in,

54postmortem image of palate of, from mucosal disease,

112rota-virus infections in, 1256-month-old seronegative, BVDV in, 106–107thrombocytopenic, petechiation of ocular mucous

membranes in, 111transient infection in, 59

Calving pens, environmental contamination of, 96Camelidae family, serologic surveys of ruminant species

distributed within, 173Camels, 116Campylobacter spp., 210Canada

BVDV type 2 emergence in, 14mucosal disease in, 4severe acute BVDV infection in cattle population in, 110total annual costs in, 57

Caribou herds, seropositivity in, 173Case-control studies, sampling in, 50Caspase-mediated cellular destruction, 186Cataracts, 107, 224Catarrhal fever, 110Cats (domestic), cell lines survey and, 173Cattle. See also Persistently infected cattle

acute infections of, 121BVDV transmission between sheep and, 231cell lines survey and, 173concurrent infection with BVDV and BRSV, 161

Index 247


Cattle (continued)pregnant, BVDV in, 107, 109recovery in, 115susceptible, sources of BVDV and mode of transmis-

sion to, 232Cattle density, 52, 59, 197, 205Cattle herds, risk factors for presence of PI animals in, 51Cattle populations, circulation of BVDV in, 122CD4+ T-lymphocytes, lymphoid tissue lesions and, 134CD46 protein, 179CD-87 isolate, 11cDNA clone, 17, 18, 25 Cell death, 186–187Cell lines, and support of growth of BVDV as possible

indicators of host range, 173–174Cell-mediated immune responses, to BVDV, 213, 214Cell tropism, 174Cellular enzymes, processing of envelope glycoproteins

by, 183–184Cellular immune response, 159, 242Cellular remodeling, 188–189Center for Veterinary Biologics (USDA), 210Central nervous system

congenital defects of, 150as immunological privileged site, 164malformations in, 129

Cerebellar hypoplasia, 5, 107, 129, 150, 164, 224Certified Semen Services, Inc., 146Cervidae family, serologic surveys of ruminant species

distributed within, 173Cervids, persistent infections in, 172Chase, C.C.L., 214Cherry, B.R., 225Chinese hamster ovary cells, 179Chivers, W.H., 4, 131Cho, H.J., 200CHO cells. See Chinese hamster ovary cellsChronic mucosal disease, 113Chu, J.J.H., 189Classic swine fever virus, 5, 65, 173, 189, 198

BVDV replication and, 177cellular factors and, 178hosts and, 171identification/differentiation of BVDV and, 203

Classification, molecular biology and, 65–74Clinical BVDV infection, 110Clinical features, 105–116

course of initial infection in immunocompetent cattlepopulation, 105–107, 109

duration and severity of clinical signs, 115occurrence in other ruminants, 115–116recognition of, 109–115

bovine respiratory disease, 115chronic mucosal disease, 113clinical BVDV infection, 110immunosuppression, 114–115mucosal disease, 111–112reproductive consequences, 113–114

severe acute BVDV infection, 110–111subclinical infection, 109venereal infections, 113

recovery from clinical signs, 115Coefficient of infectiousness (â), 91, 92, 102“Colitis cystica,” 134Collett, M.S., 71Colon lesions, in calves suffering from mucosal disease,

113Colostral antibodies, 18

age of dairy calves and, 25humoral immune response and, 158–159prevention of transmission by, 98role of, in acquiring BVDV, 106

Colostral immunity, vaccination of young calves and,231–232

ColostrumBVDV transmission via, 96as mechanism of protection from BVDV, 213–214

Colostrum-deprived calves, BVDV in, 106Colostrum management program, 98Compendium of Veterinary Products, 210Complement receptors (C3R), 157Compton prototype virus, 217Conception failure, 223, 224, 241Conception rates, 55, 59

BVDV infection and, 148–149ovarian dysfunction and, 127

Concurrent infections, 115Confocal microscopy, 189Congenital defects, 53, 55, 93, 241

of central nervous system, 150types of, 114

Congenital infection, consequences of, 105Congenital transmission

after 120–150 days in gestation, 94persistent infection and, 94

Contact rate, 92Contagiousness factors, 96Control

by vaccination, 20–21, 25without vaccination, 21

Control efforts, refocusing, 241–242Control programs

for BVDV in beef cow herds, 234for BVDV in dairy herds, 235for BVDV in stocker/feedlot operations, 235focus of, 209to limit losses due to BVDV, 233–235

beef cattle, 233–234dairy cattle, 234stocker/feedlot operations, 234–235

vaccination to control BVDV-induced disease andproduction losses, 231–233

Cooper strain, of BHV-1, 161Copper deficiency, 110Corapi, W.V., 9, 11, 12Coria, M.F., 7, 10

248 Index


Cornell University, College of Veterinary Medicine at, 3,11

Coronavirus, 114Cortese, V.S., 213, 215Cows

occurrence of other diseases and transient infection in,53–54

transient infection in, 48Coxiella burnetii, 55, 57cp BVDV (cytopathic BVDV), 12, 121, 123, 160, 197 ,

240cell death and, 186 cellular immune response and, 159early and late onset mucosal disease and, 132, 133intracellular signaling inhibition and, 188lymphoid tissue lesions and, 134molecular actions of, 22–23pathogenesis of mucosal disease and, 131, 132reproductive infections and, 151 spread of, in mucosal disease, 134tissue lesions and, 133vaccines and, 209, 210

CRIB cells, 178, 179Cross-fence contact, inter-herd transmission and, 101Cross-sectional studies, sampling in, 50Crowding, BVDV transmission and, 96Crypt epithelium, mucosal disease and infection of, 135CSFV. See Classic swine fever virusCSS. See Certified Semen Services, Inc.C3 receptor expression, BVDV and reduction in, 158C24V strain, 242CTL. See Cytotoxic T-lymphocytesCurly haircoat, 150CVB. See Center for Veterinary Biologics (USDA)Cytokines, 157, 158, 187Cytopathic biotypes, 71, 72, 81Cytopathic BVDV. See cp BVDVCytopathic strains, discovery of, 4Cytopathology, 83–84Cytotoxic T-lymphocytes (CD8+), 159

Dairy calves, transmission of BVDV in, 98Dairy cattle, control programs and management of, 234Dairy herds, PCR assay for testing of, 20Dairy operations, whole-herd testing by, 230Dams

BVDV and clinical manifestations in, 113identification of those carrying a PI fetus, 227–228

Danish Cattle Database, 52DCs. See Dendritic cellsDead calves, use of production records and laboratory

evaluations of, 228–229Decision tree analyses, 58Deer, 105, 172, 173Defective interfering (DI) particles, cytopathology and,

84Delayed-onset mucosal disease. See Late-onset mucosal

disease

Dendritic cells, 134, 157, 158Deng, R., 67, 68Dengue virus, 179Denmark

cross-sectional study in, 51economic evaluation of eradication program in, 58economic losses in, before eradication campaign in, 57eradication programs in, 21, 25, 239incidence risk in, 48persistently infected cattle in, 12prevalence in, 52

Department of Agriculture, 5Deranged osteogenesis, 150Deregt, D., 18Dexamethasone, 10DFA. See Direct fluorescent antibodyDiabetes mellitus, in PI animals, 129Diagnosis, 197–205

antibody detection, 199–200virus neutralization test, 200

direct antigen detection, 198–199enzyme-linked immunosorbent assay, 199immunofluorescence, 198immunohistochemistry of peripheral blood leuko-

cytes, 198immunohistochemistry of skin biopsies, 198–199

enzyme-linked immunosorbent assay, 200–201immunoperoxidase and immunofluorescence, 201milk as a diagnostic sample, 2031946-1969, 4–61970s, 7–81980s, 12reverse transcription-polymerase chain reaction,

201–203nested PCR, 202strain differentiation, 202–203TaqMan, 202

screening for persistently infected animals, 203–204skin biopsies and, 24virus isolation and, 18, 199–200

Diagnostic testsof clinically ill animals with BVDV infection, 229for detection of PI animals, 226–227

virus isolation, 226results, of clinically ill animals with BVDV infection,

229Diagnostic tests for detection of PI animals

for identifying PI calves, 227immunohistochemistry, 226

polymerase chain reaction, 226–227serology, 227

DI particles. See Defective interfering (DI) particlesDirect antigen detection, 198–199

immunofluorescence, 198immunohistochemistry of peripheral blood leukocytes,

198immunohistochemistry of skin biopsies, 198–199

enzyme-linked immunosorbent assay, 199

Index 249


Direct economic cost, 56Direct fluorescent antibody, 198Direct RT-PCR, 202Disease form, vaccine protection based on, 213Disease surveillance, 241–242DNA, 25Domes, loss of, 134Domestic animals, BVDV transmission and, 231Done, J.T., 8Donis, R.O., 22, 178Double fencing, minimizing BVDV transmission and,

101Double-stranded replicative form, of RNA, 83dPCR. See Direct RT-PCR Drew, T.W., 20, 202, 203dsRNA, 188 Dual genotype natural infections, 100Dual persistent infections, 150Dubovi, E.J., 18Duffel, S.J., 214“Dummy calves,” 150Duration of infectiousness, 91, 92, 101, 102Dysmyelogenesis, 129

E. coli, 110, 114, 201Eagle’s MEM, 204Earlier values, 56Early mucosal disease, 132–133Ear notch immunohistochemistry, 212Ear notch samples, 198, 204EBTr. See Embryonic trachea cell lineEconomic analyses, functions in performance of, 56Economic impact of BVDV, 56–58, 59

economic evaluation of control strategies at herd level, 58of national eradication programs, 58

at herd level, 57losses in larger populations and at national level,

57–58optimizing decisions based on economic calculations,

58Economic losses, 145, 177, 197ED cell system. See Equine dermal cell systemEdwards, S., 12Electron beam irradiation, pestivirus virion and, 66ELISA. See Enzyme-linked immunosorbent assayElk, 105, 172Ellis, J., 215Embryonic trachea cell line, 199Embryo transfer

BVDV transmission and, 95, 231inter-herd transmission and, 101

Endemic state, 92END method, 23Endsley, J.J., 214Envelope glycoproteins

processing of, by cellular enzymes, 183–184virus binding and, 82

Environment BVDV transmission and, 95–96outcome of reproductive disease and, 151

Environmental stress, 105Enzyme-linked immunosorbent assay, 19, 37, 197,

198–199, 200–201, 202, 226skin biopsies and, 24testicular infection testing and, 146

EO protein, 71Epidemiological studies

for estimation of herd level prevalence based on screening

samples/bulk milk samples, 45of incidence, 47–49of prevalence of animals with antibodies against

BVDV, 39–42of prevalence of virus positive and persistently

infected animals, 43–44on occurrence of different genotypes, 46

Epithelial surfaces, BVDV and damage to, 110Equine dermal cell system, 199Eradication programs, 25–26

national, economic evaluation of, 58in Scandinavia, 46virus isolation, seroconversion and, 172

Esophageal ulcers, mucosal disease and postmortemimage of, 112

Estonia, persistently infected animals in, 46Ethylenimine, pestivirus virion and, 66Etiology, 49E2 (envelope glycoprotein) region, 209E2 monoclonal antibodies, 12E2 protein, 68, 71, 82E2-specific monoclonal antibodies, 18Europe

formal monitoring of wildlife in, 171prevalence in, 46

European Osloss strain, 11, 12EV. See Earlier valuesExchange rate between currencies, 56Exocytosis, virion assembly and, 85Experimental BVDV infection, disease syndromes in

non-bovine hosts due to, 172

Face flies, BVDV transmission and, 96Failure to thrive, 241FAT. See Fluorescent antibody techniqueFBS. See Fetal bovine serumFcR, 157, 158FCS. See Fetal calf serumFDCs. See Follicular dendritic cellsFeed bunk space, minimizing BVDV transmission and, 101Feedlot cattle

respiratory disease in, 21–22vaccines for, 6

Feedlot operationscontrol programs and management of, 234–235intra-herd transmission and, 100

250 Index


Feedlots, “shipping fever” pneumonia and, 10Fence breakout, risk factors and, 51, 59Fence-line contact, managing, 233Fertility, reduced, 113Fetal abnormalities, 107Fetal bovine serum, 5, 7, 8, 197Fetal calf serum, 95Fetal death, 157Fetal fluid collection, 228Fetal infection, 114, 128–130

abortions and, 128–129incidence risk of, 48during middle trimester, 150

noncytopathic BVDV and, 23, 24persistent infections and, 129preventing, 152teratogenic effects of, 129–130vaccines and protection against, 25, 217–219

Fetal protectionagainst BVDV, 21vaccines and, 232–233

Fetus, BVDV transmission to, 93Ficoll-Paque/Macrodex, 204Finland

cattle density in, 52eradication programs in, 21, 239prevalence in, 46

5� UTR (5� untranslated region), 66, 81, 182, 202, 209degree of sequence identity between pestiviruses in,

66virus replication and, 71, 82

Flativirus family, 65Flaviviridae, 12, 187

genomes, 82Flavivirus IRES, translation initiation from, 183Flies, BVDV transmission and, 96Flores, E.F., 15, 178Flow cytometry, 19Fluorescent antibody technique, 6, 8Fluorescent probes, 20Folding of proteins, 184Follicular dendritic cells, 134Fomites

acute infection and, 94BVDV transmission and, 95–96, 231

Foot and mouth disease, 110Formulas

herd sensitivity/herd specificity, 37incidence of BVDV infection, 92incidence risk over animal’s life, 49true prevalence, 36vaccination protection, 92

Fox, Francis, 3, 4F-P/M. See Ficoll-Paque/MacrodexFray, M.D., 16Freisen bulls, 10Fritzemeier, J., 16

Fuller, D.A., 6Fulton, R.W., 22, 203

Gait disorders, 164Gamma irradiation, pestivirus virion and, 66Gastroenteritis virus, synergism between BVDV and, 161Gastrointestinal parasites, 110Genetic recombination

in BVDV, 83spontaneous and postvaccinal disease and, 15

Genomic diversity, 239Genotypes, 209, 239

BVDV, 71–73, 160occurrence of, and epidemiological studies, 46

Genotyping, 71German roe deer, 172Germany

BVDV genotypes identified in, 46eradication programs in, 25prevalence in, 52

Gestational age of fetus, 105Gestational vertical transmission, management and

control of BVDV and, 223Gestation stage

consequences of BVDV infection from 30–125 days of gestation, 149–150from 125–175 days of gestation, 150from 175 days of gestation to term, 150prior to implantation (30–45 days), 147–149

effect of, at time of BVDV infection of susceptiblepregnant cows, 224

impact of, on outcome of reproductive disease,147–150

Gilbert, S.A., 203Gillespie, J.H., 3, 131Giraffe, viruses isolated from, 66Givens, M.D., 201Gnotobiotic calves, 158Gnotobiotic lambs, severe respiratory disease in, 161Goats, 116, 172

cell lines survey and, 173congenital infection in, 174

gp48, 71 Graham, D.A., 54, 201Granulocyte-macrophage colony-stimulating factor, 158Great Britain, economic losses in, 57Grooms, D.L., 24, 107, 229Growth retardation, 150, 224Gruber, A.D., 202Grummer, B., 189Gutekunst, D.E., 6

Haematopota pluvialis, 96Haines, D.M., 19, 22Hamel, A.L., 202Hannover Veterinary School, 3Harpin, S., 19HCV. See Hepatitis C virus; Hog cholera virus

Index 251


HCV Ns3/4A serine protease, 188Head flies, BVDV transmission and, 96Heifer development operations, whole-herd testing by,

230Hellen, C.U., 67, 68, 182Helper T-lymphocytes (CD4+), 159Hemophilus somnus, 115, 161, 210Hemophilus spp., 110Hemorrhages, in severe acute BVDV infections, 126Hemorrhagic BVDV infection, 105Hemorrhagic diathesis, 126Hemorrhagic syndrome, 11, 14, 54, 72, 124, 157

symptoms of, 111Hepatitis C virus, 177Herd diagnosis, 37Herd immunity, 160

transmission under influence of, 92–93Herd infection, defined, 35Herd level

economic evaluation of control strategies at, 58economic losses at, 57

Herd level prevalence, epidemiological studies forestimation of, based on screening samples/milksamples, 45

Herdsannual testing of, 230BVDV and economic constraints on, 116

Herd screening, factors in, 14Herd sensitivity, 37Herd specificity, 37Herd test, 37Highly infected antibody carriers, 51High virulence strains

development of lesions and, 124–125low virulence strains compared with, 125role of, in acute infection, 121, 123

HIV, speculation on interaction between BVDV and, 173Hog cholera virus, 5, 65, 171, 177Homing, Peyer’s patch lymphocytes and, 133Horizontal transmission

acute infection and, 94postnatal, management and control of BVDV and, 223

“Hospital” milk, inter-herd transmission and, 101Host factors

clinical disease and, 105impact of, on outcome of reproductive disease, 151role of, in transplacental/intrauterine infection,

126–128Host range, cell lines supporting growth of BVDV as

possible indicators of, 173–174Hosts, 171–174. See also Interactions of virus and host

non-bovinedisease syndromes in, due to experimental BVDV

infection, 172disease syndromes in, due to natural BVDV

infection, 171–172virus isolation and seroconversion in, 172–173

overview of, 171

Houe, H., 24, 49, 203Howard, C.J., 216HSe. See Herd sensitivityHSp. See Herd specificityHulst, M.M., 179Humoral immune response, 242

to BVDV, 213, 214colostral antibodies and, 158–159

Humoral immunity, 91Hydranencephaly, 150, 164, 224Hydrocephalus, 150, 164, 224Hydrocephalus internus, 129Hydrophobicity plot, of pestivirus consensus sequence,

68, 69Hydrotaea irritans, 96Hyena disease, 150Hypomyelination, 150Hypotrichosis, 224

Iatrogenic transmission of BVDV, 95–96IBR. See Infectious bovine rhinotracheitisIETS. See International Embryo Transfer SocietyIHC. See ImmunohistochemistryIIF. See Indirect immunofluorescence assayIIP. See Indirect immunoperoxidaseIL-2, 158Ileum

histologic section of, from calf 12 days after infection,164

histologic section of, from uninfected calf, 164Immune responses to BVDV, 157–159

cellular, 159humoral, 158–159innate, 157–158of persistently infected animals, 159

Immune suppression, in acute BVDV infection,125–126

Immunityassessment of, 160herd, 92–93

Immunocompetence, 105Immunofluorescence, 198, 201Immunohistochemistry, 24, 198, 226

of peripheral blood leukocytes, 198of skin biopsies, 198–199, 205

Immunologically privileged sites, 163–164central nervous system, 164ovaries, 163, 164testes, 163, 164

Immunoperoxidase, 201Immunoperoxidase monolayer assay, 12, 198, 200Immunosuppression, 105, 106, 114–115, 157, 160–163

in acute BVDV infection, 125–126BVDV and secondary infections, 161BVDV-induced immune organ dysfunction, 163clinicopathological assessment of, 162multifactorial, 114

Immunotolerance, 105, 164

252 Index


Incidence, 35, 58of BVDV infections, 47–49epidemiological studies for estimation of, 47–49prevalence and, 49

Incidence rate, 47Incidence risk, 47, 48Incubation periods, for acute BVD, 115Indirect fluorescence microscopy, 189Indirect immunofluorescence assay, 200Indirect immunoperoxidase, 200INF. See InterferonInfected animals, BVDV transmission from, 96–98 Infection status, 49Infectious bovine rhinotracheitis, 6, 10, 54, 125Infectious cDNA clones, 17 Infectiousness of infected animals, reducing coefficient

of, 101Infertility, 148Innate immune response, to BVDV, 157–158Insect transmission of BVDV, 95–96Insemination

transmission via semen and, 94–95uterine exposure to BVDV at, 127–128

Interference test, 8Interferon, 187

cytopathic BVDV and production of, 23innate immune response and, 158role of, during pregnancy, 151

Interferon regulatory factor-3, 188Inter-herd transmission, 101Internal ribosome entry site, 68, 82, 181, 182International Committee on Taxonomy of Viruses,

Seventh Report of, 71International Embryo Transfer Society, 16Inter-species transmission, 105Intestinal mucosa, mucosal disease and infection of, 135Intracellular signaling inhibition, 187–188Intra-herd transmission, 100–101Intrauterine exposure, virus persistence after, 130Intrauterine infections, 5. See also Transplacental/in-

trauterine infectionsIn utero transmission, after 120–150 days of gestation,

93, 94In vitro fertilization, BVDV and, 113IPMA. See Immunoperoxidase monolayer assayIqbal, M., 179IRES. See Internal ribosome entry siteIRF-3. See Interferon regulatory factor-3Isolation hutches, minimizing BVDV transmission and,

101i-VVNADL, 23

Japanese MLV BVDV vaccine, protection by, 217Jensen, A.M., 10JNK1, 188JNK2, 188Johne’s disease, 110Jordan, R., 184

Kafi, M., 113Kahrs, R.F., 6Kerkhofs, P., 203Killed vaccines, 13, 209, 212–213

efficacy of, in experimental studies, 152requirements for efficacy of, 210strains and types of bovine viral diarrhea virus in, 215

Kirkland, P.D., 10Kirkpatrick, J., 213Kozak consensus sequence, 181Kümmerer, B.M., 17Kunjin virus, 189

Label claims, for vaccine products, 210, 211Laboratory evaluation, use of, in evaluating moribund

and dead calves, 228–229Lambert, G., 8Lambs, 151Lamina propria, lesions and, 134, 135Laminitis, 5Langedijk, J.P.M., 187, 201Latency, of acute infections, 97Late-onset mucosal disease, 9, 132–133LDLR. See Low-density lipoprotein receptorLecomte, C., 201Leptospira hardjo, 55, 57Leptospira spp., 210Lesions

in brain and eyes, 129of BVDV, 4correlation between viral antigen and, 133development of, in acute BVDV infection, 123–125fetal infections and, 224in lymphoid tissues, 133–134in mucosa, 135in mucosa-associated lymphoid tissues, 134–135ovarian, 127pathogenesis of, in mucosal disease, 133–135

Letellier, C., 203Leukopenia, 4, 17, 113, 157, 162Lewis, T.L., 201Licensing of vaccines, 25Liess, Bernd, 3, 13Lifelong infections, 240Light-Cycler system, 204Lightly infected antibody carriers, 51Lindberg, A., 204Lipid envelope, in flativiruses, 65Littlejohns, I.R., 13, 20Livestock production, impact of BVDV on, 116Llamas, 116, 173Low-density lipoprotein receptor, 82, 178, 179, 189Low virulence strains, 241

development of lesions and, 123–124high virulence strains compared with, 125

Lung lesions, 126Lymph node involvement, with BVDV, 106, 107Lymphoid depletion, 114

Index 253


Lymphoid follicles, 163Lymphoid tissues

lesions in, 133–134mucosa-associated, lesions in, 134–135

Lymphopenia, 124, 125, 126, 159, 162“Lytic” viruses, 22

Mabs. See Monoclonal antibodiesMacCallum, A.D., 4Macrophages, 157Madin-Darby bovine kidney cells, 6, 177, 178Mahlum, C.E., 202Makoschey, B., 216Malmquist, W.A., 5Management systems

diagnostic tests for detection of PI animals, 226–227for identifying PI calves, 227

identification of dams carrying PI fetus, 227–228monitoring herds for BVDV PI risk, 228–230other potential sources of BVDV, 230–231

Mandibular brachygnathism, 150Mannheimia haemolytica, 10, 11, 114, 115, 126, 161,

210Mannheimia spp., 110Markov Chain model, 58Martin, S.W., 21Mastitis, 53, 54McCauley, J.W., 179McClurkin, A.W., 7, 8, 9, 10, 13, 24McGoldrick, A., 20MD. See Mucosal diseaseMDBK cells. See Madin-Darby bovine kidney cellsMEGA-10, 201M-ELISA. See Monolayer enzyme-linked immunosor-

bent assayMendez, E., 17Meyers, G., 12, 15, 17Meyling, A., 10, 12, 49Michigan, prevalence in, 52Microencephaly, 129, 150, 224Microopthalmia, 150, 224Milk

bulk testing of, in Scandinavia, 21BVDV transmission via, 96yield reductions, 53

Milk samplesclasses of, 203diagnostic, 197, 203epidemiological studies for estimation of herd level

prevalence based on, 45PCR assay for testing of, 20

Mink lung cell system, 199Minnesota Veterinary Diagnostic Laboratory, 202ML cell system. See Mink lung cell systemMLV type 1 strain (NADL), strategic vaccination and, 99Modified live virus (MLV) vaccines, 13, 114, 171, 209,

211–212, 239

BVDV transmission via, 96biosecurity and, 233requirements for efficacy of, 210strains and types of BVDV in, 215

Moennig, V., 9, 16, 201Moerman, A., 48Molecular biology

advances in, 11–12, 17–18classification and, 65–74

Monitoring herds for BVDV PI risk, 228–230annual whole-herd testing, 230pooling samples and whole blood for PCR testing, 229production records and laboratory evaluation of mori-

bund and dead calves, 228–229serologic evaluation of sentinel animals, 229–230

Monoclonal antibodies (Mabs), xi, 3, 11, 239tests for, 18–19

Monocytes, 157Monolayer enzyme-linked immunosorbent assay, 200Moormann, R.J.M., 179Moose, 105, 116, 173Moribund calves, use of production records and labora-

tory evaluations of, 228–229Mucosa, lesions in, 135Mucosa-associated lymphoid tissues, lesions in, 134–135Mucosal disease, 3, 4, 5, 57, 74, 111–112, 121, 131–135,

157acute/chronic sequel to CI, 108–109age-matched beef calves from herd suffering from, 112chronic, 113defective interfering particles and, 84early-onset, 132–133experimental production of, 8–10late-onset, 9, 15–16, 132–133pathogenesis of, 131–132postmortem images

of colon lesions in calves suffering from, 113of esophageal ulcers in calf suffering from, 112of palate of calf suffering from, 112

spontaneous and postvaccinal, 15spread of cp BVDV in, 134

Multiplex assays, 19, 20Mummification, 147, 241Mummified fetuses, 128Muñoz-Zanzi, C.A., 24, 25, 229Musca autumnalis, 96, 231Mycobacterium bovis, 159Mycoplasma bovis, 22, 115, 161Mycoplasma spp., 110Mycotoxicosis, 110

NADC strain, 242NADL, 14Naive immunocompetent cattle population

course of initial infection in, 105–107, 109BVDV in colostrum-deprived calves, 106BVDV in 6-month-old seronegative calves, 106–107BVDV in pregnant cattle, 107, 109

254 Index


Nakamura, S., 197National Animal Disease Laboratory, 6National eradication programs, economic evaluation of,

58Natural BVDV infection, disease syndromes in non-

bovine hosts due to, 171–172ncp BVDV, 12, 121, 123, 160, 197

cellular immune response and, 159intracellular signaling inhibition and, 188molecular actions of, 22–23pathogenesis of mucosal disease and, 131, 132persistent infections and, 129 reproductive infections and, 151 research on, 240vaccines and, 209, 210

NDV. See Newcastle disease virusNeonatal calves

BVDV infections in, 5preventing BVD in, 8

Neonates, acute BVDV infection of, 110Nested polymerase chain reaction, 202Neutralizing antibodies, neonates and, 150Neutropenia, 162Neutrophils, 157Newcastle disease virus, 23New York, original outbreaks in, 6New York 93 strain, 17Ng, M.L., 189Nitric oxide, 157Njaa, B.L., 204NO. See Nitric oxideNoncytopathic biotypes, 71, 72, 81Noncytopathic BVDV. See ncp BVDVNoncytopathic viruses, 9, 10Nonstructural protein region, 209North America

NADL strain in, 11subgroups of BVDV 2 in, 73

Northern Ireland, incidence risk in, 48Norway

cattle density in, 52cost-benefit evaluation of eradication program in, 58economic losses in, 57eradication programs in, 21herd study in, 51prevalence in, 46

NS2, 17, 69genetic recombination and, 83

NS2-3, 12, 15, 16, 19, 69, 73NS3

genetic recombination and, 83Position A and Position B and, 73

NS3-encoded serine protease, 82NS3 monoclonal antibodies, 12, 19NS4A protein, 70NS4B protein, 70NS5A protein, 185, 186NS region. See Nonstructural protein region

Nucleic acids, BVDV-specific, 25Nuttall, P.A., 8NY-1 strain, 14, 214, 218, 242Nystagmus, 164

Ocular defects, 5Oculocerebellar syndrome, 129OD. See Optical densityOdds ratio, 50Olafson, P., 4, 145, 162Ontario

BVDV outbreaks in, 55BVDV type 2 emergence in, 14severe acute BVDV infection in cattle population in,

111Oophoritis, 147, 148“Open gut” phenomenon, 158Open reading frame, 11, 65, 66, 68, 82

polyprotein translation and, 181proteolytic processing and, 84

Open-tube RT-PCRs, 202Opisthotonus, 129Optical density, of indirect ELISA technique, 227–228Optic neuritis, 150, 224OR. See Odds ratioOra-nasal route of BVDV transmission, 157Oregon (C24V) strain, 4, 6, 14, 17, 242ORF. See Open reading frameOro-nasal uptake, virus spread and, 123, 124Osloss strains, 14Ovarian hypoplasia, 113Ovaries

as immunologically privileged site, 163, 164lesions in, 127reproductive capacity and infection of, 147

Over-the-fence contact, with PI animals, 223Oviductal cells, BVDV infection and, 148

P. haemolytica, 54Palate of calf, mucosal disease and postmortem image of,

112Palfi, V., 18Passive humoral immune response, 158Passive immunity, 105, 159Pasteurella multocida, 161, 210Pasteurella spp., 110Pasturing, risk factors and, 51, 59Pathogenesis, 121–135

acute infection, 121, 123–126immune suppression in, 125–126thrombocytopenia and hemorrhages in, 126viral and host factors in, 121, 122virus spread and lesions in, 123–125

defined, 121of lesions in mucosal disease, 133–135

correlation between viral antigen and tissue lesions,133

lesions in lymphoid tissues, 133–134

Index 255


Pathogenesis (continued)lesions in mucosa-associated lymphoid tissues,

134–135in mucosa, 135

mucosal disease, 131–135early- and late-onset, 132–133

overview of, 121persistent infections, 130–131

prolonged viral shedding following acute infection,130–131

virus persistence following intrauterine exposure,130

transplacental/intrauterine infections, 126–130exposure of uterus to BVDV at insemination,

127–128fetal infection, 128–130infection during preovulatory period, 127viral and host factors in, 126–127

Paton, D.J., 10pBEK. See Primary bovine embryo kidney cells PCR. See Polymerase chain reactionPCR-probe test, 24PCR tests. See Polymerase chain reaction testsPEG. See Polyethylene glycolPellerin, C., 14Peripheral blood leukocytes, immunohistochemistry of,

198PERK, 187Persistent infection, 3, 105, 106, 121, 130, 152

atypical, 16–17congenital transmission resulting in, 94consequences of, during 30–125 days of gestation,

149–150defined, 240experimental production of, 8–10fetal infection and, 128–129impact of, and relationship to BVDV-associated

disease, 150–151reproductive disease and, 145–152

Persistently infected animals, 197BVDV transmission from, 96–97coefficient of infectiousness for, 91death among, 53epidemiological studies for estimation of prevalence of

virus positive and, 43–44immune response of, 159incidence risk and, 48intra-herd transmission and, 100losses among, 55–56prevalence and, 36, 46as primary source of BVDV, 223removal of, 233, 239screening for, 203–204as source of virus for acute infection, 130

Persistently infected calvesorigin of, after viremia and placental infection, 109outcome of, and vulnerability to mucosal disease,

110

samples from, and occurrence of BVDV in plasma andfeces, 109

resistantly infected carriers, 227Persistently infected cattle, 12, 13

control programs and, 209cytopathic BVDV vaccines and, 9exposure of herds to, 224–226possible tests for identification of, 227

Persistently infected cows, normal calves from, 16Pestivirus consensus sequence, hydrophobicity plot of,

69Pestivirus envelope proteins, proposed mechanisms for

transporting/processing of, 184Pestiviruses, 14, 17

conservation of polyprotein coding sequences among,68, 69

genetic recombination and, 83replication and, 81

Pestivirus 5� UTR, sequences, conservation and predictedpseudoknots in, 67

Pestivirus genome, 66–67organization of, 69

Pestivirus genomic RNA, translation initiation of cellular transcripts compared to, 180

Pestivirus genuscharacteristics unique to, 65members of, 65–66

Pestivirus infections, 157Pestivirus IRES, translation initiation from, 183Pestivirus proteins, 70Pestivirus virion, 66Pestova, T.V., 67, 68, 182Petechiation, of ocular mucous membranes in thrombo-

cytopenic calf, 111Peterhans, E., 23, 186, 188Peyer’s patches, 110, 134, 135

acute mucosal disease and, 112effect of BVDV on, 163homing and, 133

Phylogenetic studies, 14–15Pigs, 172

classic swine fever virus in, 171, 198congenital infection in, 174

Pilinkiene, A., 200Pillars, R.B., 24, 229PK15 cell system. See Porcine kidney cell systemPlacenta, BVDV and, 113Plaque neutralization assays, 4Pneumonia, 4, 11, 161

antibiotic-resistant, 22Polyethylene glycol, 178poly IC, 23Polymerase chain reaction, 19–20, 203, 226–227Polymerase chain reaction tests, 72

pooled samples of whole blood for, 229Polymicrobial infections, 115Polyprotein translation, 71, 181–183P125 protein, 12

256 Index


Pooled blood samples, 37Pooled monoclonal antibodies, 19Pooled samples, for diagnostic testing, 24Pooling strategies, herd immunity and, 160“Poor-doers,” 129Population genetics, quasispecies and, 85Porcine kidney cell system, 199Porencephaly, 150Postnatal horizontal transmission, management and

control of BVDV and, 223Postnatal infection/disease, vaccines and protection

against, 214–217Postnatal transmission, 93Postvaccinal mucosal disease, 6, 8, 9, 13, 15Potgieter, L.N.D., 10, 11Prebreeding vaccination, 99Precolostral BVDV-neutralizing antibodies, 94Preconditioning programs, 13Pregnancy status, 105Pregnant cattle, BVDV in, 107, 109Pregnant PI cows, and transmission to fetus, 96Prenatal testing, 204, 242Preovulatory period, infection during, 127Present values, 56Prevalence, 35, 58, 59

of BVDV genotypes, 72of BVDV infections, 36–38, 46definition of, 36epidemiological studies for estimation of, 38epidemiological studies for estimation of, in animals

with antibodies against BVDV, 39–42incidence and, 49intra-herd transmission and, 100reducing, 101

Primary bovine embryo kidney (pBEK) cells, 199Prins, S., 18Pritchard, W.R., 4, 5Probability of infection, 91, 92Production records, use of, in evaluating moribund and

dead calves, 228–229Prolonged infections, acute, 240–241Pronghorn antelope, 66, 173Protective immunity, 160Protein kinase, 186Protein translation, virus replication and, 188Proteolytic processing, 84Pseudoknot structures, 82Pulmonary hypoplasia, 107, 224Pulmonary lesions, in lambs, 172Putative species, pestiviruses and, 66PV. See Present valuespvMD. See Postvaccinal mucosal disease Pyrexia, 162

Quantitative disorder, 162Quasispecies, population genetics and, 85Quebec

BVDV type 2 emergence in, 14

severe acute BVDV infection in cattle population in,111

Qvist, P., 19

Rabbits, cell lines survey and, 173Radostits, O.M., 13, 20Radwan, G.S., 20Rae, A.G., 18Ramsey, F.K., 4, 131Real rate of interest, 56Rebhun, W.C., 11Recovery, from acute BVDV infections, 115Rectal transmission of BVDV, 95Reduced milk yield, 53Reindeer, viruses isolated from, 66Relative risk, 50Renshaw, R.W., 20Repeat breeding, 53Replication, 177

interactions with cellular factors during, 180–186overall view of, 190regulation of, 84–85RNA, 184–186viral, 71

Replicative form, of RNA, 83Reproductive consequences, from BVDV, 113–114Reproductive disease

impact of gestation stage on outcome of, 147–150host factors on outcome of, 151vaccination on outcome of, 151–152viral factors on outcome of, 151

ovarian infection, 147overview of, 145persistent infections and, 145–152testicular infection, 145–146

Reproductive disorders, 55, 59Reproductive effects, of BVDV, 223–224Reproductive infections, 152Reproductive losses, initial descriptions of, 145Reproductive performance, impact of different BVDV

strains on, 107, 108–109Research

BVDV, progress in, 239–240refocusing of, 240–241

acute, prolonged infections, 240–241low-virulence strains of BVDV, 241noncytopathic BVDV, 240T-cell responses, 240

Respiratory disease, 10–11, 53, 1452000 to the present, 21–22

Respiratory tract, BVDV and damage to, 110Retained placenta, 54Retinal atrophy, 107, 164Retinal degeneration, 150, 224Revell, S.G., 10Reverse transcription-nested polymerase chain reaction,

113

Index 257


Reverse transcription-polymerase chain reaction, 19, 171,178, 199, 205

assays, 197, 201–202RF. See Replicative formRidpath, J.F., 14, 15, 16, 18, 20, 24, 203Rinderpest, 4, 110Risk assessment, 35–59

economic impact of BVDV, 56–58economic evaluation of control strategies at herd

level, 58economic losses at herd level, 57economic losses in larger populations and at

national level, 57–58evaluation of eradication programs and, 58optimizing decisions based on economic calcula-

tions, 58incidence of BVDV infections, 47–49

epidemiological studies for estimating incidence,47–49

prevalence and incidence, 49measurable and quantifiable epidemiological variables,

35overview of, 35prevalence of BVDV infections, 36–38, 46

epidemiological studies for estimation of preva-lence, 38, 46

epidemiological studies on occurrence of differentgenotypes and, 46

quantification of effects/consequences of BVDVinfections, 53–56

losses among PI animals, 55–56occurrence of other diseases, 53–54reduced milk yield, 53reproductive disorders, 55subclinical infections, 53virulent strains of BVDV with other pathogens, 54

risk factors for occurrence of BVDV infections, 49–52Risk factors, 51

for BVDV infections, 59defined, 49epidemiological framework for description of, 36for occurrence of BVDV infections, 49–52

RNA, 25virus-specific forms, 83

RNA helicase, 70, 82RNA replication, 183, 184–186RNA replicative intermediate (RI), 184RNases, immunomodulatory effects of, 186RNA-stimulated NTPase, 70RNA viruses, 81R-not (reproduction number), 91Roe deer, 116Rotavirus, 114RR. See Relative riskRT-nPCR. See Reverse transcription-nested polymerase

chain reactionRT-PCR. See Reverse transcription-polymerase chain

reaction

Ruggli, N., 188Ruminants

infection of, with BVDV, 101occurrence of BVDV in, 115–116spread of BVDV among, 105

SAGE. See Serial analysis of gene expressionSaliki, J.T., 19, 200Salmonella spp., 114, 126Salmonellosis, 110Sampling, prevalence measures and, 38Savan, M., 5Saxony-Anhalt (Germany), serum pools used in, 204Scandinavia

BVDV control programs in, 197eradication programs in, 25, 46

Schelp, C., 178Scheme of Alenius, 203Schweizer, M., 23, 186, 188Screening samples, epidemiological studies for estima-

tion of herd level prevalence based on, 45Se. See SensitivitySecondary infections, BVDV and, 161Semen. See also Testicular infection

BVDV transmission via, 94–95, 230, 239inter-herd transmission and, 101quality of, in persistently infected bulls, 10testing of, before bull’s entry into AI center, 17venereal infections and, 113

Semliki Forest virus, 188Sensitivity, prevalence and, 36, 37Sentinel animals

serologic evaluation of, 229–230use and testing of, 25

Septicemia, 111Serial analysis of gene expression, 188Serine protease, 70Seroconversion

incidence risk and, 48in non-bovine hosts, 172–173

Serologic evaluation, of sentinel animals, 229–230Serology, 227

problems with, 160Seronegative animals, BVDV transmission to, 96Serum neutralization assays, 4, 8, 15Serum neutralization test, 5, 160, 200Serum neutralizing (SN) antibodies, 91, 96, 97Serum PCR-probe test, 24Severe acute BVDV infection, 105, 110–111, 242

lesions in cattle suffering from, 112, 113thrombocytopenia and hemorrhages in, 126

Severe acute disease, 157SFV. See Semliki Forest virusShahriar, F.M., 22Shedding, 203, 223

acute infection and, 94clinical features of BVDV and, 105herd immunity and, 93

258 Index


intra-herd transmission and, 100prolonged, following acute infection, 130–131in semen, 146transmission

from acutely infected animals and, 97–98from persistently infected animals and, 96–97via modified live virus vaccine, 96

Sheep, 116, 151border disease virus of, 171, 198BVDV transmission between cattle and, 231cell lines survey and, 173congenital infection in, 174studies of transplacental infection of ovine fetuses with

BVDV in, 172in vaccine studies, 209

Shimizu, M., 9“Shipping fever” pneumonia, 10Signaling inhibition, 187–188

cytokines, 187intracellular, 187–188

Singer strain, 8, 11, 14, 17, 242vaccine, 13

Single-strand conformation polymorphism, 203Single-stranded viral RNA, 83SIR (susceptible/infectious/removed) model, 47Skin biopsies, 24, 205

immunohistochemistry of, 198–199staining of viral antigen in, 226

Slaughterhouse fetuses, BVDV found in, 95Slovakia, BVDV genotypes identified in, 46SN antibodies. See Serum neutralizing (SN) antibodiesSN test. See Serum neutralization testSouth America

BVDV genotypes identified in, 46subgroups of BVDV 2 in, 73

Southern Africa, subtypes of BVDV 1 in, 22Sp. See SpecificitySpahn, C.M.T., 182Specific immunity, 157Specificity, prevalence and, 36, 37Sperm heads, in infected bulls, 10Spinal cord, hypomyelination of, 224Spleen, BVDV and damage to, 110, 163Spontaneous mucosal disease, 3, 15SSCP. See Single-strand conformation polymorphismStable flies, BVDV transmission and, 96ST cell system. See Swine testicle cell systemStillbirths, 8, 55, 129Stocker operations, control programs and management

of, 234–235Stocking density, intra-herd transmission and, 100Stocking rate, minimizing BVDV transmission and, 101Stomoxys calcitrans, 96Strain differentiation, 202–203Strategic vaccination, examples of, 99Stress, vaccine efficacy and, 242Subcellular fractionation techniques, 189Subclinical acute infections, 197

Subclinical infections, 53, 59, 109Subgenotypes, 239

of BVDV 1 and BVDV 2, 73Suckling calves, testing of, 230Sullivan, D.G., 20, 203Superinfections, 9Superovulatory response, in persistently infected cows,

16Surveillance programs, 59, 241–242Susceptible animals

decline in, 92reducing proportion of, 101–102

Swedenantibody levels from bulk tank milk in, 48bulk milk screening for BVDV in, 205bulk tank milk samples from dairy herds in, 53BVDV control program in, 197eradication programs in, 21, 25, 239self-clearance in, 46

Sweet clover poisoning, 111Swine testicle cell system, 199Switzerland, incidence risk in, 48Synergism, between BVDV and enteric pathogens, 161

Tamoglia, T.W., 7TaqMan, 20, 202T-cell immune responses

development of, 213, 214research on, 240

T-cells, 164Temperature monitoring, 123Temperature-sensitive mutant virus vaccine, 13Teratogenic effects, of BVDV, 129Teratogenic lesions, fetal infections and, 224Testes, as immunologically privileged site, 163, 164Testicular infection, bull fertility and, 145–146TGAC virus, 16Thomson, R.G., 53� UTR (3� untranslated region), 81, 82Thrombocytopenia, 11, 14, 111, 124, 126, 162, 197Thrombocytopenic calf

excessive hemorrhage in, 111petechiation of ocular mucous membranes in, 111

Thür, B., 24Thymic aplasia, 224Thymic hypoplasia, 150Thymus, effect of BVDV on, 163TI. See Transient infectionTIA-1, 185TIAR, 185Tissue lesions, correlation between viral antigen and,

133T-lymphocytes, decrease of, in acute BVDV infection,

125Tonsils, BVDV and damage to, 110Torticollis, 129TP. See True prevalenceTRAM, 189

Index 259


Transient infection, 93, 106, 223acute, 121BVDV spread in secretions/excretions during, 107in calves, 59in calves, occurrence of other diseases and, 54in cows, 48, 53–54in other ruminants, 115by virulent strains or BVDV in combination with other

pathogens, 54–55Translation

of polyprotein, 71regulation of, 84–85replication and, 82–83

Transmission. See Virus transmissionTransplacental infection, 114Transplacental infections/intrauterine infections, 121,

126–130fetal infection, 128–130

abortions, 128–129persistent infections, 129teratogenic effects, 129–130

infection during preovulatory period, 127role of viral and host factors in, 126–127uterine exposure to BVDV at insemination, 127–128

Transrectal transmission of BVDV, 95Tremors, 164Triton-X-100 treatment, 201Trizol (Gibco) reagent, 202tRNA syntheses, 188 True prevalence, 36Trypsin treatment, pestivirus virion and, 66Tuberculosis, 159Tumor necrosis factor, 157Type 1c virus, 22Type 1d virus, 22

Udder health, 53Ungulate species, BVDV transmission and, 231United Kingdom, BVDV genotypes identified in, 46United States

BVDV genotypes identified in, 46BVDV type 2 emergence in, 14severe acute BVDV infection in cattle population in,

110Untranslated regions, 65. See also 3’ UTR; 5’ UTRUSDA, BVDV guidelines, 152Uterus, exposure of, to BVDV at insemination,

127–128UTR. See Untranslated regions

Vaccination, 102, 197, 205biosecurity and, 233–234to control BVDV-induced disease and production

losses, 231–233control by, 20–21, 25control without, 21eradication without, 3in Germany, 25

impact of, on outcome of reproductive disease,151–152

measuring effect of, on transmission, 92prevention of transmission by, 98–100in vitro evidence of efficacy of, 231of young calves, colostral immunity and, 231–232

Vaccination programs, control programs and, 209“Vaccine Claims for Protection of the Fetus Against

Bovine Viral Diarrhea Virus,” 210Vaccine-induced mucosal disease, in persistently infected

cattle, 3Vaccines, 3, 209–219

appropriate use of, 210design of, 242efficacy studies, 214

protection against fetal infection/disease, 217–219protection against postnatal infection/disease,

214–217fetal protection and, 232–233goals for development of, 219killed, 212–213mechanism of protection for BVDV, 213–214modified live virus, 114, 211–212overview of, 209–210protection based on disease form and, 213recent research on BVDV strains in, 239serological results and different uses of, 38strains, 211stress and efficacy of, 242

Van Oirschot, J.T., 21, 209, 214Vassilev, V.B., 22, 25Veal calves, BVDV-related hemorrhages in, 11Venereal infections, 113, 130Vertical transmission, 93–94

embryo transfer and, 231management and control of BVDV and, 223

VI. See Virus isolationVilcek, S., 201Viral factors

impact of, on outcome of reproductive disease, 151role of, in transplacental/intrauterine infection,

126–128Viral genome, genome organization and, 81–82Viral nonstructural proteins, 69–70Viral proteins, 68

nomenclature for, 71nonstructural, 69–70RNA replication and, 184–185structural, 68–69

Viral replication, 71Viral shedding. See also Shedding

in persistently infected bulls, 10prolonged, following acute infection, 130–131

Viral structural proteins, 68–69Viremia, 5, 17

detection of, 106testing for, 17

Virion assembly, exocytosis and, 85

260 Index


Virions, pestivirus, 66Virulence factors, role of, in acute infection, 121, 123Virulent strains, transient infection by, in combination

with other pathogens, 54–55Virus and host interactions, 177–190

BVDV replication cycle overview, 189–190cell death, 186–187with cellular factors during replication, 180–186

polyprotein translation, 181–183processing of envelope glycoproteins and, 183–184RNA replication, 184–186

cellular remodeling, 188–189inhibition of signaling, 187–188

cytokines, 187intracellular signaling inhibition, 187–188

receptor and virus attachment, 177–180cellular factors, 177–179viral factors, 179–180

virus release, 189Virus isolation, 199–200, 226

diagnosis by, 17, 18in non-bovine hosts, 172–173in semen, 113

Virus-neutralization test, 200Virus pair, 9Virus release, 189Virus replication, 81–86

cytopathology, 83–84defective interfering particles and cytopathology, 845� UTR, 82genetic recombination, 83overview of, 81proteolytic processing, 84quasispecies and population genetics, 85regulation of translation and replication, 84–85sites, 853� UTR, 82translation and, 82–83viral genome, 81–82

genome organization, 81–82virion assembly and exocytosis, 85virus binding and reentry, 82

Virus sheddingin persistently infected bulls, 10

Virus transmission, 91–102of BVDV from infected animals, 96–98

from acutely infected animals, 97–98from persistently infected animals, 96–97

horizontal and vertical, 93–94

infectious disease epidemiology, 91–93BVDV transmission factors, 91–92effect of vaccination on transmission, 92transmission under influence of herd immunity,

92–93inter-herd, 101intra-herd, 100–101minimizing BVDV transmission, 101–102

reducing coefficient of infectiousness, 101reducing duration of prevalence of infectious

animals, 101reducing likelihood of adequate contact, 101reducing proportion of susceptible animals,

101–102overview of, 91prevention of

by colostral antibodies, 98by vaccination, 98–100

routes and means of BVDV transmission, 94–96via embryo transfer, 95 iatrogenic, fomite, environmental, and insect, 95–96via milk and colostrum, 96via modified live virus vaccine, 96via semen, 94–95transrectal transmission, 95

VN. See Virus-neutralization testVoges, H., 16, 17VP1, 71VP2, 71

Washing procedures, embryo transfer and, 231Wasting disease, 157Weak-born calves, 55, 59Weaning, mortality of PI calves prior to, 224Weinstock, D., 24, 201, 202Wentink, G.H., 16Westenbrink, F., 9, 15West Nile virus, 189White blood cell disorders, 162Whitmore, H.L., 10Whole-herd testing, 227

annual, 230Wildlife

BVDV transmission and, 231pestiviruses and, 14

Winter dysentery, 110World Organization for Animal Health, 17

Zona pellucida, BVDV infection and, 149

Index 261


bovine viral diarrhea virus

Documents