laboratory techniques in rabies

Laboratory techniques

in rabies

F.-X. Meslin Chief

Veterinary Public Health Division of Communicable

Diseases World Health Organization

Geneva, Switzerland

FOURTH EDITION

Edited by

M. M. Kaplan H. Koprowski Former Director Director

Research Promotion Center for Neurovirology and Development Jefferson Cancer Institute

World Health Organization Thomas Jefferson University Geneva, Switzerland Philadelphia

PA, USA

World Health Organization

Geneva 1996

WHO Library Cataloguing ip P b b i c a t o ~ i Data

Laboratory tec?r?ques in rabies edited by F - X Mes in V h; Kapdr i H Koprnwsk - 4th ed

1 Rabies diagnosis laboratorv marLals 2 Haoies vaccine I Mesin F.-X l1 Kaplan. M.M. lli.tioprowsk!, H

ISBN 92 4 154,479 1 (NLM Classification WC 550)

The W o r d Health O ~ g a i i z a t ~ o n welcomes requests for permission to reproduce or translate its poblica- tions in part or in full Applicaliiins ar?ri enquiries should be addressed !o the Office of Publ~cat ions World Health Oryari i iation Geneva Switzerland whicti will be glad to provide the latest ir~forrnation o n any

changes made to the text p a l l s for new editions, and reprints arid lranslatlons already available

@World Health Organization 1996

Publicat~ons of the World Healtil Organiratioi l enjoy copyright protection in accordaricc with the provisions of Protocol 2 of the Universal Copyrigtit Conbention Al l rights reserved

The oesgnations ernp:oyed a r d the prese.itaton of t t ~ ? material in this publication do not irnply the express10:i ui ariy op'lior: wiiatsoeve~ on the part of thr Sccretaria: of the World l ieal lh Organization c o n c e r ~ i n g !?e legal s:a!us o: any country terrilory citv or area or of its authorities, or concerning the

delimitation o i i s frontiers or DO~rlddrleS

The -:lention of specitic companies or of certain rnan~fac:~i rers ' products ooes no: imply that they are

endorsed or recommended by :he Worla Heaitb Orgar~7at ion ~ r , preference to otners of a sirnilar natdre

that are no! ment io led Errors and owissior3s exceo:ed, tqe names of propfietary pronucts are d is t ingu is t i~~ i by iiiitial capi!al leiters

The authors aloqe are responsible fo i the vleivs expressed in this uublicatiori Where the designation

' ,court ry of area ' appears in the headings of tables. I ? covers countrles, te:rltories cities or areas

Typeset i l India

Printed in Finiarid

93~9814-Macm1l lars!Vam~ala8500

Contents

Preface xiii

List of acronyms and abbreviations used in this book xv

Part I. General considerations 1

Chapter 1 Safety precautions in handling rabies virus (M M Kaplan) 3

Properties of the virus 3

Pathogenesis 3

Laboratory precautions 4

Treatment of wounds 5

Pre-exposure immunization

References

Chapter 2 An overview of laboratory techniques in the diagnosis and

prevention of rabies and in rabies research (F.-X Mes/in &

6

7

MM�� 9

Introduction 9

Diagnostic procedures for antigen detection

Tests for the determination of rabies antibody

Potency tests

Research techniques

Conclusion

References

Chapter 3 Characteristics and molecular biology of the rabies virus

10

13

14

15

16

16

( N. Torda) 28

Introduction

Morphology and structure

Functional analysis of the infection

Molecular biology of the rabies virus

Evolution of the rabies virus

References

Part II. Routine laboratory procedures

Chapter 4 Rapid microscopic examination for Negri bodies and

preparation of specimens for biological tests

(E. S. Tierkel & P. Atanasiu)

Dissection of the brain

iii

28

28

32

36

43

45

53

55

55

LABORATORY TECHNIQUES IN RABIES

Preparation of slides Tlie Negri body: differential diagnosis The mouse inoculation test Annex Pre~arat ion of Sellers' stain

Chapter 5 Histopathological diagnosis (P Lepine & P. Atanasiu) 66 Removal of the b r a ~ n and preparation of tissue samples for examination 68 Ernbedciirig, shining and examination for Negri bodies 77

Chapter 6 The lnouse inoculation test (H. Koprowsk~) Choice of mice Preparation of suspect material for rnoculation Inoculation of mice Observation of inoculated mice Further passages of infected material Removal of tile brain Complications

Ctiapter 7 The fluorescent antibody test (D J Dean M K Abelseth & P Atanas~u) 88 Principle 88 Materials and methods 89 Discussion 93

Chapter 8 Virus isolation in neuroblastorna cell culture A. Webster & G. A. Casey) 96 Rabies tissue-culture infection test (RTCIT) 97 References 101 Annex l Media 102 Annex 2 Avidin-biotin staining method 103

Chapter 9 Rapid rabies enzyme irnmunodiagnosis (RREID) for rab~es ant~gen detection (H Eourhy & P Perrin) 105 Introduction 105 Method 105 Evaluation of the technique 11 1 References 11 1 Annex Preparation of reagents 112

Chapter 10 Cell culture of rabies virus ( A A King) Susceptible cells, cell lhnes and strains Methods of virus propagation Cytopathology Persistent infection Virus in intected cells Application of cell-culture methods

References

CONTENTS

Part Ill. Special diagnostic and research techniques

Chapter 11 Techniques for the production, screering and characterization of rnonoclonal antibodies (M, Lafon) Immunization of animals Myelomas Fetal calf sera

Fusion Screening of hybrdoma supernatants

Cloning of hybridomas by l im~l ing di l i~t ion Prodi~ction of large n~nounts of rnonnclonal antibodies Freezing and thawing of hybridomas Characterization of rnonoclonal antibodies Use of rnonoclonal antibodies References Aniiex Dulbeccos modified Eagle's medium (DMEM)

Chapter 12 Monoclonal antibodies for the identification of rabies and non-rabies lyssaviruses (J. S. SmitA & A. A. King) Materials and methods lmmunofliiorescence tests using MAb-RNPs Antigenic analysis using MAb-Gs Applications Discussion References

Chapter 13 The polynlerase chain reaction (PCR) technique for diagnosis. typing and epidemiological studies of rabies (N. Tordo, D. Sacrament0 & H. Bourhy) Introduction Amplification of the rabies transcripts Diagnosis Typing and molecular epidemiological studies References Annex Preparation of buffers and reagents

Chapter 14 Techniques for the purification of rabies virus, its subunits and recornbinaril products (B. Dietzschold) Introduction Pcrrificatiori of rabies virus particles Purification of rabies virus siibunits and structural proteins under non-denaturing conditions Purification of rabies virus proteins under denaturing coriditions References Annex Preparation of reagents

Chapter 15 A rapid fluorescent focus inhibition test (RFFIT) for determining rabies virus-~ieiitralizing antibody (J. S. Smith. P. A. Yager & G. M. Baer)


Standard procedure Calcu!at~on of virus-neutralizing antibody titres Alternative test procedures Interpretation of results References Annex 1 Growth media for MNA and BHK-21 S13 cells Annex 2 Calculation of titres

Chapter 16 An in vltro virus neutralization test for rabies antibody (C. V Triinarchi. R. D. Rudd & M. Saffoid. Jr) Method Interpretation of results References

Chapter 17 Competitve ELISA for the detection of rabies virus-neutralizing antibodes (L. D. Eimgren & A. l. Wandeler) Method Interpretation of results References Annex 1 Preparaion of buffers and reagents Annex 2 Conjugaton of monoclona! antibodies Annex 3 ELISA software

Chapter 18 Electron microscopy (K. Hummeler & P. Atanasiu) Structural studes Studies of morphogenesis Annex Medium for agarose: Eagle's basal medium (EBM)

Part IV. Methods of vaccine production

Section A. Brain-tissue vaccines 221 Chapter 19 General consideratons in the production and use of brain-

tissue and purified chicken-embryo rabes vaccines for human use (F.-X. Meslin & M. M. Kaplan) 223 Introduction 223 Adverse efTects of brain-tissue vacclnes 224 Recent developments in brain-tssue vaccine production 226 References 228

Chapter 20 P-Propolactone-inactivated sheep brain vaccine (H. Sir7gii) 234

Cornpositon 234 Preparation of the seed vrus 234 Preparation of the vaccine 235 Quality control tests 237 Biochemical tests 238 Preparation of standard vaccine 240 Dosage schedule 240 References 241 Annex 1 Preparation of 0.5 mol/l sodium-potassium phosphate buffer, pH 7.6 24 1

CONTENTS

Annex 2 Preparation of 0.05 rnol/l phosphate-buffered saline (PBS), pH 7 0 Annex 3 Preparation of stabilizer for rabies vaccine, pH 7.2

Chapter 21 Suckl~ng-mouse brain vaccine ( A M Dlaz) Formilla

Preparat~on of the inoculum Inoculation and harvest

Preparaton of the vaccine Control tests

Expiry date References Annex 1 Ultraviolet irradiation for inactivation of vaccines Annex 2 Pre~arat ion of stabilizer solutions

Section B. Ernbryonating egg vaccines Chapter 22 Purified duck-embryo vaccine for Run~ans (R. Gluck)

Preparation of the vaccine Control tests Expiry date References Annex Preparation of stabilizing medium

Chapter 23 Chicken-embryo vaccine for dogs (H. Koprowski) =ccine Preparation of the v-

Control tests References Annex Preparation of stabilizing solution, pH 7.6

Section C. Cell-culture vaccines Chapter 24 Cell-culture vaccines for Iiunian use general considerations

(K G Nicholson) Human diploid cell vaccine Other cell-culture vaccines Safety Efficacy Economical post-exposure treatment regimens Pre-exposure ~mrnun~zation References

Chapter 25 Vaccine for humans prepared in human diploid cells (R. Brarictie) Preparation of the vaccine Contl-ol tests Conclusion References Annex Flow chart for the production of HDC vaccine using

the MRC5 cell strain


Chapter 26 Purified Vero cell vaccine for humans (B. Montagnon & B. Fanget) Cell cultures Preparation of the vaccine Control tests Expiry date References

Chapter 27 Purified chick-ernbryo cell vaccine for humans [R. Bartti & K Franke) History Preparatiori of the vaccine Control tests Administratioii of the vaccine Expiry date Laboratory tests References

Chapter 28 Fetal rhesus monkey lung diploid cell vaccine for humans (R. Barth & V Franke) History Preparat~on of the vaccine Control tests Expiry date Adinnistration of the vaccine Laboratory tests References

Chapter 29 Dog kidney cell vaccine for humans (R. Barih, V Franke & G. van Steenis) Preparation of tile vaccine Control tests Administration of ihe vaccine References Annex Preparation of medium 199

Chapter 30 Primary hamster kidney cell vaccine for humans (R. Barth, I/ Franke & F. T. L I ~ ) Preparaton of the vaccne Control tests

Expiry date Administratior? of ihe vacclne References Annex Hanks' balanced salt solution

Chapter 31 Vnukovo-32 primary hamster kidney cell vaccines for humans [R. Barfii, V Franke & M. A. Selimov) Preparation of the vaccines Control tests

Adrn~nistration of the vaccines References

CONTENTS

Chapter 32 Cell-culture vaccines for veternary use (P. Reculard) Substrates for the production of seed virus and vaccne Preparation of the vaccines Manufacturing requirements for cell-culture rabies vaccines and recommendations for their use Planning a facility for the production of rabies vaccine for veterinary use References

Chapter 33 Modified live-virus rabies vaccines for oral immunization of carnivores (J Riancou & F-X. Meslin)

Modified live-virus vaccines Guidelines for assessing the safety and efficacy of MLV vaccines References

Section D. Genetically engineered vaccines Chapter 34 General considerations in the use of recombinant rabies

vaccines for oral iminunzation of wildlife (C. E. Rupprecht, C. A Hanlon & H. Koprowsk~) References

Chapter 35 Expressron of rabies proteins using prokaryotc and eukaryotic expression systems (B. Diefzschold) Prokaryotic expression systems Eukaryotic expression systenis References

Part V. Vaccine safety and tests for potency and antigen quantification

Chapter 36 General considerations in testing the safety and potency of rabies vaccines (P. Sizaret) References

Chapter 37 The NIH test for potency (L. A. Wilbur & M. F. A. Aubert) Standard test Modified NIH test References Annex Preparation of Challenge V~rus Standard (CVS) diluent, pH 7.6

Chapter 38 Habel test for potency (K. Habel) Standard test Modifled Habel test Annex Calculating 50% end-point dilutions by the method of Reed & Muench

Chapter 39 Guinea-pig potency test for chicken-embryo vaccine (H. Kopro wski) Immunization orocedure


Preparation of challenge material Challenge of the guinea-pigs Interpretation of results

Chapter 40 Single radial irnrnunodiffusiorl test for the determination of the glycoproiein content of inactivated rabies vaccines (M. Ferguson) Principle Materials Method lnterpretation of results References Annex Preparation of Dulbecco's phosphate-buffered saline solution A (PBSA), pH 7.2

Chapter 41 Enzyme-linked immunosorbenl assay (ELISA) Tor the determination of the glycoprotein content of rabies vaccines (P. Perrin, M. Lafon & P. Sureau) Pr~nciple Preparation of antibodies Precautions Sensitization of microtitration plates Assay procedure lnterpretation of results Evaluation of in v~iro potency References

Chapter 42 The Essen-ELISA for the determ~nation of the glycoprotein content of inactivated cell-culture rabies vaccines (0. Thraenhart) Principle Method Evaluation of results References Annex Preparation of buffers and reagents

Chapter 43 The modified antibody-binding test for in vitro quantification of rabies virus antigen in inactivated rabies vaccines (R. Barth) Principle Method Interpretation of results Evaluation of the test

References

Part V!. Antirabies serum and immunoglobulin

Chapter 44 Production of antirabies serum of equine origin ( T Luekrajang, J. Wangsa~ & P. Phanuphak) Introduction Method

CONTENTS

Factors affecting the production of ERlG Safety References

Chapter 45 Purification techniques for heterologous rabies antserum (R G l m k & D Laberti

Preservation and storage of serilm or plasma Purification by enzyme tieatmept and heat denaturation Purflcation by precipitation using ethacridine lactate and cthanol Stabilization and preservation of purrfled HRIG Standardizatoi? of the final product References Annex Prewarat~on of buffers

Chapter 46 Production of human rabies immunoglobulin (P. Fournier & R K Sikes) Introduction Formula Source and shipment of blood Reagents Technique Disadvantages References Annex Preparation of reagents

Chapter 47 Potency test for antrabies serum and lmmi~nogiobulln (E A Fitigerald) Principle Preparation and titration of challenge virus Serum virus neutralization Interpretation of results Currently used potency tests References

Appendices -

Appendix 1 Simple technique for the collection and shipment of brain specimens for rabies diagnosis (J Barrat)

Appendix 2 Techniques for the preparat~on of rabies conjugates (P Perrin)

Appendix 3 Methods for the calculat~on of titres (h1 F A. Aubert)

Append~x 4 Addresses of international institutions for technical cooperation in rabies control

Index

Preface

During the 20 years that have elapsed since the publication of the third edition of Laboratory techniques in rabies, enormous progress has been made in improving methods of rabies vaccine and antisera production, and in developing new diagnostic and assay procedures. Major advances in molecular biology techniques have been extensively applied to the study of the rabies virus during recent years, and a fourth edition of the monograph has therefore become necessary. This edition includes some 30 new chapters, which describe new diagnostic. research and vaccine production techniques. Although some of these methods are currently restricted to relatively advanced laboratories (e.9, monoclonal antibody techniques, the polymerase chain reaction and virus expression systems), they are expected to become routine procedures in the future. Nevertheless, many laboratories will not have the facilities or equipment to use these methods. therefore the basic classical techniques described in the previous edition have been retained and, where necessary, brought up to date.

The production of rabies vaccines for animal and human use is extensively reviewed. The production of modified live-virus vaccines and recombinant vaccines is also briefly covered. It should be rioted that there has been a dramatic increase in the number of cell-culture vaccines available for human use and that production is no longer restricted to developed countries. Many of these vaccines have now replaced those derived from nerve tissue. Accordingly, only two chapters deal with the production of the latter, which are still used in some developing countries.

It should be stressed that clalms for the efficacy of particular vaccines are entirely the responsibility of the authors, and that their inclusion ~n this book does not imply offrcial recogn~tion by WHO Vaccine manufacturers ~ntending to use the production techniques described here should refer to the requirements for rabies vaccines for human and veterrnary use, as defined by the WHO Expert Committee on Biological Standardization.' - 3

An early draft manuscript of this fourth edition of Laboratory techniques I n

rabieswas examined by the WHO Expert Committee on Rabies in September 1991 ,4

and a number of suggestions were inade for changes to the text and for the

' W H O Expert Committee on Bfoiogicai Stariddrdiratron. Thfrty-f!rst report Geneva, World Health Organl~atlon, 1981 (WHO Technical Report Serles No 658). Annex 2; Annex 3

' W H O Expert Corrirnittee on B~ologfral Slandard~zat~on Thfriy-seventh report Geneva. Worid Health Organization. 1987 (WHO Technlca Rcport Series, No 760). Annex 9

3WH0 Expert Coinrn!ttee on Bioiogical Standardrzat~on Forty-Nxrd report Geneva, World Health Organization 1994 (WHO Technical Rcport Selies, No. 840). Annex 4: Annex 5; Arinex 6

"WHO Expert CommiJtee on Rab!es Eighili feport. Geneva, World Health Orcjanlzat~on, 1992 (WHO Technical Report Series. No 824).


inclusion of some add~tional material. In preparing the final manuscript, the editors have ensured that the information IS as up to date as possible. Where new material could not be incorporated in the existing text, it has been added in the form of appendices at the end of the book.

The World Health Organization gratefully acknowledges the collaboration of the many eminent specialists who have contributed to this volume. The ed~tors thank Miss C. Allsopp, Office of Publications, WHO, for her assistance in the preparation of this book.

List of acronyms and abbreviations used in this book

ABT ATCC BHK BPL BSA CDC cDNA C-ELISA CER

Cl95 CVS D I DIA DMEM DNA E AE EBL EBM

ED50 EIA ELlSA EMEM ERA ERIG FA FCS FDA

FFD50

FFU FlTC FRhMDC FWR G protein H DC HEP HN HRIG

ant~body-bindlng test American Type Cultilre Collection baby hamster kidney cells beta-propiolactone bovine serum albumin Centers for Disease Control and Prevention (USA) complementary deoxyribonucleic acid competitive enzyme-linked immunosorbent assay chick embryo-related cells 95% confidence interval Challenge Virus Standard defective interfering (particles) dot-immunobinding assay Dulbecco's modified Eagle's medium deoxyribonucleic a c d experimental allergic encephalomyelitis European bat lyssavirus Eagle's basal medium median effective dose, 50% end-point dilution enzyme immunoassay enzyme-linked irnmunosorbent assay Eagle's minimum essential medium Evelyn Rokitniki Abelseth strain of rabies virus equine rabies immunoglobulin fluorescent antibody fetal calf serum Food and Drug Administration (USA) dilution at which 50% of the observed m~croscopic fields contain one or more f o c of infected cells focus-forming units fluorescein isothocyanate fetal rhesus monkey diploid cell French wild rabies isolates rab~es glycoprotein human d ip lo~d cell Flury high egg passage strain of rabies v~rus haemagglutinin-neuraminidase protein human rabies immunoglobulin


l g INHV INPPAZ

LD50 LEP L protein M1 protein M2 protein MAb MAb-G MAb-N MAb-RNP MEM MICLD,, MIT M LV MNA MNT MO1 Mok mRNA N A NIH N protein OD PAb PAb-G PAHO PAS PBS PCEC PCR PDE PDL PFU PHKC P M

PSRV PV PVRV RIG RNA RNP proteln RTCIT RFFIT RREID

SAD

immunoglobulin ~nfectious haematopoietic necrosis virus PAHOiWHO Pan American Institute for Food Protection and Zoonoses (Argentina) International Unit median lethal dose Flury low egg passage strain of rabies vrriis rabies RNA-dependent RNA polyrnerase rabies phosphoprotein rabies matrix or ineinbrane protein monoclonal antibody anti-glycoprotein monoclonal antibody anti-nucleoprotein monoclonal antibody anti-ribonucleoprotein monoclonal antibody minimum essential medium median lethal dose for mice inoculated by the ~ntracerebral route mouse inoculatiori test modified live-virus mouse neuroblastoma cells mouse neutralization test mult~plicity of infec'rlon Mokola virus messenger ribonucleic acid neuroblastoma cells National Institutes of Health (USA) rabies nucleoprotein optical density polyclonal antibody anti-glycoprotein polyclonal ant~body Pan American Health Organization Louis Pasteirr strain of rabies virus phosphate-buffered saline purified chick embryo cell polymerase chain reaction purified duck embryo population doubling level plaque-iorming units primary Syrian hamster kidney cell Pitman-Moore strairi of rabies virus

product-specific reference vaccine Pasteur strain of rabies virus purified Vero cell rabies vaccine rabies immunoglobulin ribonucleic acid rabies riboniicleoprotein rabies tissue-culture infection test rapid fluorescent focus inhibition test rapid rabies enzyme imnlunodiagnosis

Street-Alabama-Duiferin strain of rabies virus

ACRONYMS AND ABBREVIATIONS

SClD severe combined immunodeficient SDS PAGE sodiurn dodecyl sulfate-polyacrylarnide gel electrophoresis SMB sucklng mouse b r a n TCID,, median tissue-culture infective dose VRG reconibinant vaccinia virus expressing the G proten gene of rabies

virus

VSV vescular stomatits virus

General considerations

CHAPTER 1

Safety precautions in handling rabies virus M. M. Kaplan '

Various publications are available describing the precautions to be taken in laboratories working witti infective agents. The reterer~ices i~sted at the end of this chapter include four manuals (1-4) that deal with problems associated w ~ t h handling such agents, and Chapter 4 and Appendix 1 of this monograph are concerned with the shipment and preparation of specimens from animals suspected of being rabid. The present chapter discusses some properties of the rabies virus and the precautions needed to protect personnel. It is noteworthy that apparently only one hurnan death has occurred from a laboratory infection with fixed virus, despite accidental needle punctures and other exposures to rabies virus during many decades of handling this virus in research, diagnostic and vaccine production laboratories. This death was reported in 1972 in a laboratory worker in Texas, USA. who was engaged in the preparation of vaccine and was exposed to fixed virus in an aerosol from a blender (5). Another laboratory infection from an aerosol subsequently occurred in New York slate, USA, and the victim was still alive some 10 years later, although with severe residual mental impairment (W.G. Winkler, personal communication). Nevertheless. the fear surrounding work involving rabies virus has resulted in relatively few laboratories conducting needed research in this field. It is hoped that the following discussion will dispel undue concern, and will provide guidance on rational steps to be taken to avoid accidents and on how to deal with them when thev do occur.

Properties of the virus

The rabies virus belorigs to the genus Lyssavirus, family Rhabdoviridae: enveloped and bullet-shaped viruses that contain lipid and single-stranded RNA (see Chapter 3). It is sensitive to lipid solvents (soap solution, ether, chloroform, acetone), 4 5 7 0 % ethanol, iodine preparations, and quaternary ammonium compounds (6). Other relevant properties are resistance to drying, to repeated freezing and thawing, relative stability at pH 5-10, and sensitivity to pasteurization temperatures and ultraviolet light The virus is readily inactivated by p-propiolactone, but is more resistant to 0.25-0 594 phenol used in Sempie-type vaccines, where several days are required to obtain complete inactivation (see Chapter 20).

Pathogenesis

The virus is usually introduced by a bite wound, although penetration can occur through intact mucous membranes and the digestive tract ( i), but not through

' Foriner Director Research Promotion and Deveioprnel! World Health Organ7ation Geneva Switrer land


intact skin. In tissue-culture systems, the virus penetrates into the cell within 15 minutes, after which it can no longer be neutralized by specific antiserum (8). Airborne natural infection is possible in exceptional circumstances, for example in caves harbouring large numbers of bats carrying the virus (9).

There are marked differences between different strains of virus in their ability to infect, spread within the body, and produce disease Laboratory strains of "fixed" virus used to produce vaccines, or employed in diagnostic and research procedures, have low pathogenicity when inoculated peripherally in low doses. The Fermi-type vaccine, as well as the original Pasteur vaccine contained residual, live, fixed virus. Such vaccines were used for many decades, but are no longer recommended by the WHO Expert Committee on Rabies. Accidents occurred in humans when such fixed virus vaccines were not properly incubated to reduce their virus content, and large amounts of material of a high titre were injected into humans ( T O ) , Laboratory accidents involving small puncture wounds are not very dangerous when fixed or modified attenuated virus is handled. They need not cause undue worry if appropriate wound treatment procedures and a booster vaccine dose are employed (see below). This includes all fixed or modified strains such as SAD (Street-Alabama-Dufferin), ERA (Evelyn Rokitniki Abeseth), CVS (Challenge Virus Standard), LEP (Low Egg Passage) and HEP (High Egg Passage) strains.

Street virus, however, must always be handled with respect, especially in the presence of hyaluronidase-containing saliva, despite apparent differences in the ability of different strains of street virus to infect through peripheral inoculation. This ability to infect and cause disease in animals is mainly a function of virus dosage, i.e. there appears to be a threshold below which disease is not produced. The susceptibility of humans to small amounts of street rabies virus is apparently not as great as that of foxes and cattle, but since human infection has been known to occur even after relat~vely small puncture wounds (on the fingers for example), it is wise to consider all wounds contaminated with street virus as potentially very dangerous. In any event, all laboratory personnel should receive pie-exposure immunization (see page 6).

Laboratory precautions

All laboratories have their individual arrangements and consequent rules of discipline for handling infective materials (1-4). No attempt, therefore, is made here to cover the many possible variations of measures that could be used. Instead, general recommendations are glver? below for procedures that could be adapted to meet the major laboratory operations employed with rabies virus.

Protective clothing

Gross operations requiring the opening of skulls and spinal column, or those in which splintered material is encountered (broken glassware, bone), should be performed with thick protective gloves, sleeved gowns, and goggles or a plastic face shield. Close-fitting goggles and a face mask are always wise precautions, especially when street virus IS handled. Rubber or strong plastic aprons that can easily be disinfected or discarded should also be worn. Close-fitting plastic or

SAFETY PRECAUTIONS

rubber gloves should be worn when animals are being inoculated with street virus; gloves are not necessary, and are often a hindrance, when fixed or other attenuated strains of rabies virus are used in tissue culture and when titrations are erf forme d.

Aerosols

Since airborne rabies infection has been demonstrated (see above), high-speed mixing and centrifugation procedures should be carried out in tightly closed conta~ners and ~ ~ n d e r a negatlve draught hood Other operations that might cause aerosols (e g pipetling) should also be carried out under a negative draught hood Hoods and cubicles should be provided with ultraviolet lamps for disinfection when not in use Pipetting by mouth should be prohibited

Disinfectants

Quaternary ammonium disinfectants in 1 :500 dilution, 45-70% alcohol, l % soap solution, and 5-7% iodine solutions kill the rabies virus within one minute (6) and are indicated for the treatment of wounds (see below). For pipette receptacles a 1 : 1000 dilution of a quaternary ammoniurn compound, any iodine disinfectant with residual available iodine of at least 1: 10000, or a 1 % concentration of soapy water or detergent can be used. The solution should be autoclaved and discarded after each use. Hot soapy water or detergent can be used for swabbing floors and tables.

Glassware, plasticware and instruments

These should be drscarded into plastic or glass receptacles containing one of the disinfectants merit~oned above They should be autoclaved before reuse or disposal

Carcasses and animal tissues

These are best disposed of in plastic bags and incinerated. Oral transmission of rabies in laboratory animals may occur when brain tissue contain~ng large amounts of virus is fed (7). In countries at an early stage of economic development the carcasses of domesticated animals (sheep and goats) that have been used to prepare Semple-type vaccines are sometimes employed for human consumption- after the head, viscera and vertebral column are removed. Although the risk is thought to be negligible once the meat has been thoroughly cooked, this practice 1s not recommended,

Treatment of wounds

All wounds should be washed immediately and thoroughly for several minutes with soap and water. T h ~ s is perhaps the single most important procedure in preventing infection, The washing should be gentle in order to avoid further traumatization of the tissues. All soap should be removed before one of the chemical disinfectants


mentioned under "Laboratory precautions" is applied. Suturing of gaping wounds should be delayed for as long as possible (from several hours up to 3 days). Puncture wounds should be probed gently with an appropriate chemical disinfectant, taking care to minimize further trauma, Local infiltration of wounds with human or animal antrabies irnniunoglobulin (or serum) should also be employed. if the wounded area so permits; this procedure is used particularly in puncture wounds where adequate cleansing and disinfection are not feasible. Additional measures, such as administration of antimicrobials or antitetanus procedures. when indicated, should follow the local treatment.


Various schedules have been tried for pie-exposure immunization of laboratory personnel. The course of action recommended in the eighth report of the WHO Expert Committee on Rabies (1992) should be followed ( 1 1 ) . The report states:

Pre-exposure immunization should be offered to persons at high risk of exposure, such as laboratory staff working w ~ t h rabies virus veterinarans animal handlers and wildlife officers, and other individuals who are living in or travelling to areas where rabies is endemic

Such immunization should preferably consist of three full intramuscular doses of tissue-culture rabies vaccine of potency at least 2 5 IU per dose given on days 0 7 and 28 (Afew days variation is not important ) The presence of virus- neutralizing antibodies in vaccinated individuals should be ascertained where feasible using serum samples collected 1-3 weeks after the last dose For adults the vaccine should always be administered in the deltoid area of the arm For young children the anterolateral area of the thigh is also acceptable The guteal area sliould never be used for vaccine injections since administra- t ~ o n in this area results in lower neutralizing antibody titres

Tissue-culture or purified duck-embryo rabies vaccines of potency at least 2.5 IU per dose have been shown to induce adequate antibody titres when carefully administered intradermally in 0.1 ml volumes on days 0, 7 and 28. After reconstitution of the vaccine, the entire volume should be used as soon as possible. Separate syringes and needles must be used for each dose. lntradermal application of the vaccine is of special interest in areas where economic constraints limit vaccine availability. However, pre-exposure vaccination with human diploid cell (HDC) vaccine administered intradermally should, whenever possible, be performed before starting antimalarial prophylaxis, since virus-neutralizing antibody hires have been shown to be lower in patients receiving chloroquine phosphate. When this is not feasible, HDC vaccine should be administered intramuscularly.

Periodic booster injections are recommended for persons at continuing risk of exposure to rabies. The following guidelines are recommended for determining when boosters should be administered:

All persons who work with live rabies virus in a diagnostic, research or vaccine production laboratory should have a serum sample tested for rabies virus- neutralizing antibodies every 6 months and a booster administered when the

SAFETY PRECAUTIONS

titre falls below 0.5 IU/ml. Responsible authorities should ensure that all staff are properly immunized.

All other persons at continuing risk of exposure to rabies should have a serum sample tested for I-abies virus-neutralizing antibodies every year; a booster should be administered when the titre falls below 0.5 U/ml.

Satisfactory antibody responses have been obtained with as little as two or three doses of vaccine given 3 days apart, and this may be tried if there is a shortage of time. A booster dose should be given one to several montiis after the last inoculation. Any potent tissue-cultiire vaccine can be employed either intradermally (0.2 m1 total dose in two sites in the deltoid reglon of the upper arm, 0.1 ml in each) or subcutaneously or intramuscularly (in three doses of at least 2.5 IU per dose). Alternatively, 2 ml of 5% nerve-tissue emulsion or its equivalent may be given subcutaneously or intramuscularly if cell-culture vaccine is not available. It is important to determine, as stated above, whether virus-neutralizing antibody has in fact resulted from the procedure. With pre-existing antibody and the known rapid recall response elicited by booster doses, i f re-exposure occurs a single inoculation of vaccine can be employed with reasonable confidence. The development of new, concentrated, and purified vaccines of cell-culture origin, with greatly increased antgenic content, has considerably improved the antibody levels now achieved with most conventional antirabies vaccines (12).

Exposed individuals who have not received pre-exposure immunization should be treated according to the recommendations made in the eighth report of the WHO Expert Committee on Rabies ( 1 1 ) . i.e. local treatment of the wound followed by a complete post-exposure course of vaccine, including rabies immunoglobulin if indicated.

References

1 . Pal SB et a l , eds. Handbook of laboratory health and safety measures Lancaster, MTP Press. 1985.

2. Miller BM et al , eds. Laboratory safety: principles and practices. Washington, DC, American Society for Microbiology, 1986.

3, Laboratory biosafety manual. 2nd ed. Geneva, World Health Organization, 1993.

4. Schmidt NH, Emmons RW. Diagnostic procedures for viral, rickettsia1 and chlamydial infections. Washington. DC, American Public Health Association. 1989.

5. Winkler WG et al. Airborne rabies transinisson in a laboratory worker Journal of the American Medical Association, 1973, 226: 1219-1 221

6 Kaplan MM et al. An intracerebral assay procedure in mice for chemical inactivation of rabies virus. Bulletin of the World Health Organization, 1966, 34: 293- 297.


7 Fischman H Ward FE Oral transmission of rabies virus in experimental animals. Americari journal of epidemiblogy. 1968, 88: 132-1 38.

8. Kaplan M M et al. Effect of polyions on the infectivity of rabies virus in tissue culture: construct~on of a single-cycle growth curve. Journal of viroioyy, 1967, 1: 145-151.

9. Constantine DG. Rabies transm~ssion by non-b~te route. Pubiic health reports, 1962, 77. 287 289.

10. Para M An outbreak of post-vaccnal rabies (rage de laboratore) in Forta- leza, Brazil. Residual fixed virus as the etiological agent. Bulletin of the World Health Oryannation, 1965, 33: 177-1 82.

11. WHO Expert Committee on Rabies. Eighth report. Geneva, World Health Organization, 1992 (WHO Technical Report Series, No 824).

12. Nicholson KG. Modern vaccines. Rabies Lancet. 1990, 335: 1201-1205

CHAPTER 2

An overview of laboratory techniques in the diagnosis and prevention of rabies and in rabies research F.-X. Meslin ' & M. M. Kaplan2

Introduction

The laboratory occupies a central place in efforts to meet the threat of rabies. Laboratory results influence both the decision whether or not to proceed with a course of treatment, and the decison on the need to institute elaborate measures for controlling an epizootic n a community. The laboratory must also provide the necessary assurance that the biological products used for treatment and pre- venton in humans and animals are efficient and safe.

The succeeding chapters in t h s manual describe selected methods for arriving at a diagnosis in the laboratory, for determining the acceptability of biological products in rabies prophylaxs, and for conducting rabies research. Most laboratory workers can decde for themselves whether one or other of the techniques given here IS within their competence, but often they are not aware of certan pitfalls and limitations of particular methods. In addition, a choice of procedures can ease the work and provide a decisive answer more quickly. These considerations are partally covered in the relevant sections of this manual; here the various techniques are revlewed and evaluated comparatively to serve as a possible guide for selecting procedures and for interpreting the results obtained. Many of these techniques can be used by laboratories with limited resources.

The Institution of treatment measures In exposed ndividuals, as recommended in the eighth report of the WHO Expert Committee on Rabies ( I ) , should never await the results of laboratory diagnosis. A laboratory diagnosis may be delayed for a variety of reasons and early treatment, both local and systemic, can be a critcal factor in saving the life of the patient.

A laboratory report should be as clear and unequivocal as possible, and should stipulate exactly the procedures used. A positive test by any one of several recognized procedures overrides negative reactions in the others. Where a doubtful result IS obtained in any single test, recourse to the other tests available I S

essential in order to arrive at a definitive conclusion. Until this conclusion is reached, treatment should be continued. Even with a negative laboratory report, circumstances may occasionally justify the in~tiation or continuation of treatment by the physican, e.g. suspicious cllnical signs in the animal, or an attack in an area

' Chief Veterinary Public Health Dlv~sion of Communlcable Dlseases Worid Health Organ~zation Geneva Switzerland Former Director Research Promotion and Development World Health Organization Geneva Swbtzer land


where rabies is enzootic by an animal that could not be caught or killed. Considerable experience has demonstrated, however, that a complete set of negative results obtained in a reliable diagnostic laboratory can usually be accepted with confidence, and that the treatment can be terminated or modified at that point.

Diagnostic procedures for antigen detection

Fluorescent antibody (FA) test

A quick and easy procedure for the diagnosis of rabies is the use of a suitable dye for the detection of Negri bodies (see Chapter 5) Histopathological techniques have however been replaced in most laboratories by the fluorescent antibody (FA) test whicti was first developed in 1958 by Goldwasser & Kissling (2) and subsequeritly modified in 1973 by Dean & Abelseth (3) and in 1975 by Kss lng (4) The FA test is now the most widely used method for diagnosing rabies infection in animals and humans It is based on microscopic examination under ultraviolet light of impressions smears or frozen sections of brain or nervous tissue after treatment with antirabies serum or globulin conjugated with fluorescein isothio cyanate The test is accurate and results can often be obtained within 30 minutes of receipt of the specimen although for routine purposes a period of 2-4 hours is desirable for the fixation in cold acetone

Apart from an appropriate microscope the two main requirements for success in using this technique are well trained personnel and conjugated antiserum or globulin of good quality After one year S experience most laboratories flnd over 99% agreement between the FA test and the mouse inoculation (MI) test In the first year however some laboratories may miss up to 10% or even 20% of the positives with the FA test and for this reason both tests (M1 and FA) should be run in parallel during this period (5) Stringent control of the labelled antrabies antibodies should be carried out to determine the specficity of the fluorescence and to minimize the number of false-positives ( 3 6) Appropriate tlssue sampling is also important Examination of impressions or smears of tissue samples from Amrnon s horn and brain stem are recommended (see Chapter 4) I abelled rabies antibodies can be prepared against the whole rabies virus Details of the preparation of the immunizrig antigen and inoculation schedule in laboratory animals and the separation and abellirig of rdbies aritibodies are given i r i Appendix 2 More highly potent antisera can be prepared using purified and concentrated rabies virus ( 7 8) or virion components such as ribonucleoprotein (9) Conjugated monoclonal rabies antibodes are being increasingly used in routine diagnosis (10) The

specficity of these latter conjugates IS greater than those prepared against the whole vlrlon or virion components A conjugate composed of two labelled monoclonal antibodies (502 2 and 1037) is now w d e y used and is ava~lable commercially Panels of monoclonal antibodies are also used in studying the epidemiology of rabies (see page 12 and Chapters 11 and 12)

The FA technique is a tighly sensitive method for detect~ng rabies antigen in fresh specimens However it may also be periormed on fixed specimens ( 11-73) The specimen should be treated with one or more proteolytc enzymes such as Lrypsirl or pepsiri before stairling to unmask the antigenic sites The sensitivity of the test using fixed specimens has been reported to be 90-10090 of that obtained

OVERVIEW

usng fresh specmens (13). However, it is reco~nniended that fresh tissue be examined where possible.

When specinlens are received in 509'0 glycerol-saline, it is imperative that the tissue be washed several times in saline before stairing

Cell-culture isolation techniques

Fixed rabies viruses car1 grow in a wide variety of cells ( 14, 15). Successful in vitro cult ivaior~ of rabies virus was first reported in 1936 ( 14) This property has been used extensively in research on rabies (page 15). However, it IS only recently that techniques tor the isolation of street rabies from suspect material in cell cultures have been developed. Tests for the isolation of street rabies n cell culture were first carried out in the mid-1970s using baby hamster kidney cells, line 21 (BHK-21), and chick embryo-related (CER) and neuroblastoma cells. These studies demonstrated that rabies infection could be detected by immunofluorescence from as early as 4-5 hours up to 5 days following inoculation (5, 16-19), Furthermore. it was found that BHK-21 cells were comparable in sensitivity to mice, whereas neuroblastoma cells were more sensitive than mice to infection by street rabies virus. The difference in sensitivity between neuroblastoma cells and BHK-21 cells and other cell lines was reported to be associated with the neural origin of the former (20.21). However. Webster & Casey (22) suggested that the difference may also be related to viral strain differences, as well as to cell type.

In routine rabies diagnosis, positive specimens contain amounts of antigen that can easily be detected by the FA test, the MI test or virus isolation in neuroblastonia cells, In the latter case, the result is obtained within 18-24 hours, although rabies antigen may be detected in these cells by FA as early as 4-5 hours after inoculation. Furthermore, virus isolation in cell culture has been shown to be as efficient as the FA test and the MI test for demonstrating small amounts of rabies virus (20). However, specimens containing a small amount of rabies virus and which are negative by FA and subsequently positive by virus isolation in cell culture require an incubat~on period of 4 days after inoculation of the cells.

In view of the usually short delay in obtair~ing the result, isolation of rabies virus in cell culture should replace intracerebral mouse inoculation whenever possible, It should. however, be borne in mind that only laboratories where cell-ci~lture techniques are currently used car1 successfully maintain neuroblastoma cells for diagnosis.

Enzyme-linked immunosorbent assay (ELISA)

In the rabies field the enzyme-linked immunosorbent assay (ELISA) was initially developed for the titration of rabies virus-neutralizing antibodies (23). The technique was applied to the quantification of rabies antigen by Atanasiu et al. using fluorescein-labelled IgG to the purified nucleocapsid (24-26). Subsequently, Perrin et al. (27) developed an ELISA called rapid rabies enzyme immunodiagnosis (RREID), which was based upon the detection of rabies virus nucleocapsid antigen in brain t~ssue. In this lest, microplates are coated with purified IgG and an IgG-peroxdase conjugate is used to react with immunocaptured antigen. This technique was compared with the FA test in a collaborative study involving six


laboratories in Europe and North America The study showed a good correlation between the FA test and RREID although the latter lest was less sensit iv~ (28) A further study was organized to evaluate the RREID under conditions prevailing in rabies laboratories in developing countries The study found over 96% agreement between the FA test and RREID ( 2 9 30) Similar results were obtained in a study on more than 3000 specimens (31) It should be noted however that in view of its lower sensitivity RREID should not replace FA in laboratories where FA is already performed

RREID is a simple and relatively cheap technique which can be especially useful for epidemiological surveys It may be used to examine partially decom posed tissue specimens for evidence of rabies infection but it cannot be used with specimens that have beer? fixed in forrnalin Since the antigen can be visualized with the naked eye the test can be carried out in laboratories that do not have the necessary equipment for FA tests (32)

Virus identification using monoclonal antibodies

Monoclonal antibodies are produced by hybridomas of fused mouse myeloma cells and splenocytes from mice immunized with either the rabies virus or rabies- related viruses They were first produced by Wiktor et at in 1978 ( l0 33) These hybridomas secreted monoclonal antibodies directed against the glycoprotein ( G protein) or nucleocapsid of rabies virus Detailed information on the production and use of monoclonal antbod~es is given in Chapters 11 and 12 The monoclonal antibodies displayed specific reactivity patterns which were used to characterize and classify rabies and rabies-related viruses into groups corresponding to antigenic determinants Since then other hybridomas have been produced and different panels of monoclonal antibodies have been established to allow differen- tration of rabies virus isolates from terrestrial and bat host specles in the USA western Europe and to a lesser extent Africa Asia eastern Europe and Latin America (34-42) Between 1982 and 1990 WHO coordinated collaborative studies on the use of monoclonal antibodies in rabies didgnosis and research (43-48) These studies led to the establishn~ent of two panels of rnonoclonal antibodies allowing identification of the various yssavirus serotypes arid the dfferent~ation of major virus strains used for vaccine production from field virus isolates An additional panel of monoclonal antibodies was also selected to differentiate rabies viruses isolated from terrestrial animal species from t h o s ~ isolated from European bat species

Although monoclonal antibodies are mainly used for epidemiological inves tigatioris they were found to be very useful for rabies diagnosis in certair-i circumstances such as ~mported cases of human rabies and rabes associated with uncertain exposure ( 4 9 50) and also routinely in countries where large-scale programmes for oral vaccination of foxes are under way to establish that no infections are caused by the vaccine straiii ( 4 1 51 52)

lntra vitam diagnosis

In addition to the b r a n and spinal cord rabiesvrus antigen can be detected by FA

In the peripheral nerves salivary glands saliva and also In the cornea a n d s k n

OVERVIEW

during the final stages of the disease (53 54) Intra v~tamdiagnosis of rabies by FA in corneal impressions was f i~st described by Schneder in animals (55) and by Cifuentes et a1 in humans (56) However a study of the reliability of corneal impressions for rabies diagnosis showed that especially when sampling is done under field conditions a negative resiilt could not rule out the diagnosis of rabies

(57) Examination of skin biopsy niaterial was shown to be a valuable technique for tntra vitamdiagnosis of rabies in animals and humans (58-60) Anderson et a1

(61) showed that rabies antigen may be detected in skin biopsies from humans at the onset of clinical signs In contrast Blenden et al (59 60) found that only some

patients (25-500/0) showed positive results during the early phase of clinical illness and that the proportion of positive results increased as the disease progressed

Specimens for intra v~tam d~agnos~s should be of a good quality They should be refrigerated immediately after collection and until the test is carried out This is important since partially autolysed specimens will reduce the percentage of positive results and contamination of the material may lead to false-positive results (62) Examination of skin biopsies may also be used for post-mortem diagnosis in countries where opening of the skull of the dead person is not accepted by relatives on cultural or religious grounds

Tests for the determination of rabies antibody

Reference tests

Serum neutralization assays are used to determine the potency of rabies serum and immunoglobulins for post exposure treatment and to evaluate the immuno- gencity of human and to a lesser deglee animal rabies vaccnes The standard procedures recommended at the seventh meeting of the WHO Expert Committee on Rabies (51) were the mouse ne~riralization test (MNT) (63) and the plaque reduction assay (64) Since then plaque reduction methods have been superseded by fluorescent focus inhibition tests which are more convenient Although the MNT is still widely used as a reference test the rapid fluorescent focus inhibition test (RFFIT) (65) has become the test of choice in most modern laboratories (see Chapter 15) The RFFIT has been shown to be at least as sensitive as the MNT in measuring virus-neutralizing antibodies (66 67) and results have also been shown to correlate well with other lesis such as the soluble antigen fluorescent antibody test (68) passive haemaggluiinalion (69) and radiommuno assays (70) Reported differences (71 73) in the potencies of rabies antibody preparations as measured by the MNT and RFFIT were not confirmed in a collaborative study carried out under the aegis of WHO (page 15)

A modified version of the RFFIT called the fluorescent inhibition microtest (FIMT) which uses microtitration plates instead of tissue culture chamber-slides has recenlly been described (74 75) Results of the test have been shown to correlate well with the RFFIT and MNT in measuring virus-n~utralizing antibodies

Techniques under development

Enzyme immunosorbent assays Enzyme im~nunosorbent assays for rabies antibody determination were first used during the late 1970s (23). Various test systems have since been developed


(76-80) for antibody titration in body fliiids and for screening supernalarits of hybridoma cultures (77, 80).

Various adsorbed antigens are used in these systems including whole virion, purified G protein and purified ribonucleoprotein (RNP) (81). An ELISA using purified G prote~n has been used to determine virus-neutralizing antibody levels in the serum of several species, includirlg humans. The test appears to correlate well with the MNT. In laboratories where rnouse colonies are scarce and tissue-culture techniques are unavailable, this test may be considered a suitable alternative to the MNT.

-Rapid seini-quant~tative assays Dot-immunobinding assays (DIA) have recently beer? developed for the diagnosis of rabies based upon the detection of viral antibodies in the serum (82). A DIA has been used to determine rabies antibody levels in human sera and the results compared with the RFFIT (83). However. the test will need to be slightly modified to enable rabies virus-neutralizing antibody levels to be determined. A DIA using nitrocellulose for antigen support has been used to test for the presence of rabies antibodies in a drop of blood (without centrifugation) (84) The test can be carried out in the field and provides results within about 30-40 minutes. It also appears to correlate well with the RFFIT.

A rapid agglutination test (RAT) based upon the ability of specific antibodies to agglutinate sensitized latex beads has recently been developed (85). The test can detect post-vaccination antibody levels equal to or higher than 2.5 IU/ml. Above this value. the agreenierit between this test and the MNT is over 97%. The test is currently being adapted to use whole blood and a mixture with the latex beads on paper instead of a glass slide.

Potency tests

Rabies vaccines

The potency of every vaccine batch must be checked before its release The appropriate potency tests described in Part V should be used In spite of d s a d vantages such as lack of accuracy, variability and poor reproducibility the NIH test remains the test of choice for inactivated rabies vaccines for both manufacturing and control laboratories (86 88) (see Chapter 37)

During the past decade WHO has coordinated research on potency tests for rabies vaccines (89-92) The possibility of replacing the standard NIH test by a modified i f? vivo test was investigated Varioi~s modifications were u s e d n the challenge strain number of vaccinations route of vaccination, and route of challenge (93) Some of these modified tests (especallv the peripheral challenge test developed by the Centers for Disease Control and Prevent~on) were shown to correlate better wlth the degree of protect~on conferred by the vaccine However the results of the latter test still showed a high degree of variability and required further modificdtlons for routine use A number of in vitrotests for the measurement of vaccine G protein content were also assessed (94-102) A protocol for the replacement of In vlvo tests by I ~ J v~tro tests was suggested by a WHO Consultation on Rabies in July 1988 (92) and was further elaborated during a WHO Cons~l ta t ion on fables vaccine potency testing ~n May 1991 (103) However stud~es showed the

effect of strain differences and substrates for the cult~vation of the rabies antigen on the results of the vattous tests Sorne of the in vitro tests failed to differentiate between free and vrion-bound G protein although it is now well accepted that only the latter is inimunogenic

In addition recent studies (104 105) have shown the importance of RNP in the immune response to rabies infection thereby underlining the need for further studies on the content of RNP Iri rabies vacclnes in relation to that of the G protein and the level of prote~l ior i

Rabies immunoglobulins

Until 1984, antirabies sera were caiibrated against the lnternatioriai Standard for Antirabies Serum, which was established by the WHO Expert Committee on Biological Star?dardizaton in 1955 (706). It was a crude hyperimmune horse serum In 1984. the International Standard for Antirabies Serum, Equine, was replaced by the International Standard for Rabies immunoglobulin, with a potency of 59 IU per ampoule. The calibration of the human rabies immunoglobulin was made by the RFFIT relative to the equine antirabies serum (107).

In 1986. it was reported that some laboratories had obtained significantly different estimates of relative potencies of rabies aritibody in human rabies immunoglobulin preparations when these potencies were determined by the MNT and the RFFIT (77-93). Results obtained in these laboratories by the RFFIT were consistently lower than those obtained by the MNT. The potential implication in post-exposure treatment was an underestimation of the potency of immunoglobulin preparations with the RFFIT, which could lead to suppression of active rabies immunizatioii. In addition, in some of these laboratories the results of the MNT were influenced by species differences between reference and test preparations. Follow!ng these reports, a collaborative study (108) was initiated by the International Laboratory for Biological Standards at the State Serum Institute in Copenhagen, Denmark, and its results reviewed at the thirty-ninth meeting of the WHO Expert Committee on Biological Standardizat~on in October 1988 (109). The study did not confirm the difference between the two techniques and it was concluded that the International Standard for Rabies lmmunoglobulin of human origin should continue to be used to calibrate the other reference materials of human or equine origin required for estimating the potericy of preparations contain~ng rabies antibodies. Furthermore, the stiidy showed that the RFFIT was the best technique for evaluating the potency of rabies immunoglobulin and serum preparations. Laboratories using the MNT were consequently advised to verify that their results did noi differ from those obtained by the RFFIT I f differences were reported, these laboratories were advised to either switch to l i ie RFFIT or introduce a suitable correcting factor

Research techniques

Since the early 1980s, molecular biology techniques have been used extensivelv in sti ides on the rabies virus. Most of the ccirrerit research on the development of rabies vaccines including procedures for tt ier licensing and release, and on the pathogenesis o! the virus is based on these techniques


Cloning and sequencing techniques have led to the determination of the organization of the rabies genome (110-112) (see also Chapter 3) The nucleotde sequences of all five genes coding for the st~uctural proteins of the virus have now been determined Furthermore the complete amino acid sequences of these proteins have been deduced Antigenic sites on the virus proteins (espec~ally the G protein) have beer? mapped using specific panels of monoclonal antibodies (113-117) Antigenic determinants in the G protein and RNP recognized by virus- speclfic B and T cells have been identified ( 1 18 119) and svnthetic peptides incorporating these deterrnnants have been tested for antigenicity and mmunogenicity in mice (120)

In addition to diagnosis and epidenioloqy monoclonal antibodies have become essential tools in rabies research for mapping specific eptopes (Chapter 12) More recently monoclonal antibodies of murine origin have been tested for post-exposule treatment of animals (121) Studies are now under way to prepare chimeric (murne-h~iman) antibodies and also to humanize morioclonal anti bodes of r-iiurine origin (48 122)

Expression of the rabies proteins (particularly the G protein) is now w i d ~ l y used in the producliori of rabies vaccines for the oral immunization of foxes and racoons Both rabies G-protein and ni~cleoprotein baculovirus recombinants are now available and could be considered for the production of low-cost rabies vaccines for anirnals and of rabies reference preparations (123 124) (see also Chapters 34 and 35)

Many laboratories have applied techniques for the cloning and preparation of nucleic probes to the detection of viral nucleic acid target sequences in clinical specimens (125) Amplification of the viral RNA by the polymerase chain reaction (PCR) followed by analysis of the amplified sequences was used Studies of the variability of selected genomic areas between virus strains can be performed on PCR-amplified nucleic a c ~ d sequences by enzyme restrict~on analysis or direct sequencing The technique has potential for studying the molecular epidemiology of rabies and iabes-related viruses (50 126) (see also Chapter 13) It has recently been used in pathogenicity studies on rabies (127 128) but because of technical difficulties there are severe iinitations on its use as a routine procedure

Conclusion

Most ot the future developments in rabies vaccine production. diagnostic procedures and studies on the pathogencity of the virus will stem from research carried out using the above techniques. While it is not possible to describe all of these techniques in detail in this book, the following chapters describe those

considered essential for the implementation and evaluation of rab~es control act~vities i r i dogs and wildlife. Adoption of the techniques best suited to local

conditions should lead to a marked improvement in the diagnosis of rabies, the control of reference materials and virus strains, and the production of rabies vaccines.

References

1. WHO Expert Corninittee on Rabies. Eighth repori. Geneva World Health Organization. 1992 (WHO Technical Report Series, No 824). Annex 1 .

OVERVIEW

2. Goldwasser RA et al. Fluorescent antibody stain~ng of rabies virus antigens In the salivary glands of rabid animals. Bulletin of the World Health Organiza- tion, 1959, 20: 579-588.

3. Dean D J , Abelseth MK. The fluorescent antibody test, In: Kaplan MM, Koprowski H, eds. Laboratory techn~ques in rabies, 3rd ed. Geneva, World Health Organization, 1973 (WHO Monograph Series, No. 23): 73-84

4 Kissling RE. The FAT in rabies In: Baer GM, ed. The natural history of rabies,

Vol. 1. New York, Academic Press, 1975: 401-416.

5. Wachendorfer G , Frost JW, Frohlich T . Current diagnostic procedures of rabies and related viruses. In. Kuwert E et al.. eds. Rabies in the tropics. Berlin, Springer-Verlag, 1985. 40-46.

6. Rapid laboratory techniques for the diagnosis of viral infections. Geneva, World Health Organization, 1981 (WHO Technical Report Series, No. 661).

7. Schneider L. Alurninum phosphate method for rabies virus purification In: Kaplan MM. Koprowski H, eds. Laboratory techniques in rabies, 3rd ed. Geneva, World Health Organization, 1973 (WHO Monograph Series, No. 23): 179-1 81.

8. Trimarchi CV, Debbie JG. Production of rabies fluorescent conjugate by immunization of rabbits with purified rabies antigen. Bulletin of the World Health Organization, 1974, 51 : 447-449.

9. Schneider LG et al. Rabies group specific ribonucleoprotein antigen and a test system for grouping and typing of rhabdoviruses. Journal of virology, 1973, 1 1 , 748-755.

10. Wiktor TJ , Koprowski H Monoclonal antibodies aga~nst rabies virus prepared by somatic cell hybridization, detection of antigenic variants. Proceedings of the National Academy of Sciences of the United States of America, 1978, 75: 3938-3942.

11. Barnard BJH, Voges SF. A simple technique for the rapid diagnosis of rabies in formalin-preserved b ran . Onderstepoort journal of veterinary research, 1982. 49: 193-1 94.

12. Swoveland PT. Johnson KP. Enhancement of fluorescent antibody staining of viral antigens in formalin-fixed t~ssues by trypsin digestion. Journal of infectiocis diseases. 1979. 140. 758-764.

13. Umoh JU. Blendon DC. lmrnunofluorescent staining of rabies virus antigen in formalin-f~xed tissue after treatment with trypsin. Bulletin of the World Health Organization, 1981, 59: 737-744.

14, Wiktor TJ. Clark HF Growth of rabies virus in cell culture. In: Baer GM, ed The natural history of i.abies. Vol. 1. New York, Academic Press, 1975: 155-179.


15. Crick J, King A. Culture of rabies virus 111 vitro, In: Campbell JB, Charlton KM, eds. Rabies. Boston, Kluwer Academic Publishers, 1988: 47-66.

16. Largi 0, Nebel A€, Savy VL. Sensitivity of BHK cells supplemented with diethylaminoetliyl-dextran for the detection of street rabies virus in saliva samples. Journal of clii?ical microbiology, 1975, 1 : 243-245.

17. Rudd RJ et al. Tissue culture technique for routine isolation of street strari rabies viriis. Joiirnal of clinical microbiology, 1980, 1 2 : 590-593.

18 Smith AL et a1 Isolation and assay of rabies serogroup viruses in CFT cells lntervirology, 1977, 8 92-99

19. Smith AL et al. Isolation of f~eld rabies virus strains in CER and murine neuroblastoma cell cultures. lnterv~rology 1978, 9, 359-361

20. Rudd RJ, Trimarchi CV. Comparison of sensitivity of BHK21 and murne neuroblastoma cells in the isolation of a street strain rabies virus. Journal of clinical mi'crob~ology, 1987, 25: 1456-1 458,

21. Umoh JU. Blenden DC. Comparison of primary skunk brain and kidney and raccoon kidney cells with established cell lines for isolation and propagation of street rabies virus, /i?fection and immunity, 1983, 41 : 1370-1372.

22 Webster WA Casey GA Diagnosis of rabies infection In Campbell JB Charlton KM eds Rab~es Boston Kluwer Academic Publishers 1988 201 -222

23. Atanasu P. Savy V, Perrin P. ipreuve immunoenzymatique pour la detection rapide des anticorps antirabiques. [Enzyme immunoassay for the rapid detection of rabies antibodies.] Annales de l'lnstitut Pasteur: Microbiology. 1977, 128A: 489-498.

24 Atanasiu P P e i i n P Delagneau JF Use of an enzyme immunoassay with protein A for rabies antigen and antibody determination Developments in b~ological standardization 1979 46 207-21 5

25. Atanasiu P et al. lrnrnunofluorescence and immunoperoxidase in the d a g - nosis of rabies In: Kurstac E. Morriset R, eds V/ra/ immunod~agnosis. New

York. Academic Press, 1974: 141-155.

26. Atanasiu P et al. Les IgG anl~glycoproteines rabiques marques par la per- oxydase et Iisothiocyanate de fluoresceine-resultats. [Peroxidase-labelled and fluorescein isothiocyanatelabelled antirabies virus glycoproten IgG- results]. Annales de l'lnstitut Pasteur. M~crob~ology. 1975, 1266: 69-75.

27. Perrin P, Rollin PE, Sureau P. A rapid rabies enzyme imrnunodiagnosis (RREID). a ~issfiil and sin-lple technique for the routine diagnosis of rabies.

Journal of b~ological standard~zat~on, 1986, 14: 21 7-222.

OVERVIEW

28, Perrin P, Sureau P. A collaborative study of an experimental kit for rapid rabies enzyme immunodiagnosis (RREID). Bulletin of the Woi-Id Health Organization, 1987, 65: 489-493.

29. A comparative study of the rapid rabies enzyme immunodiaynosis (RREID) technique. Bulletin of the World Health Organization, 1989, 67: 93-94.

30 Jayakumar R Ramadass P, Raghavan N Compar~son of enzyme immuno- diagnosrs with imrnunofluorescence for rapid diagnosis of rabies in dogs Zentralblatf fur Bakterioiog~e 1989 271 501 -503

31. Bourhy H et al. Comparative field evaluation of the fluorescent antibody test, vlrus isolation from tissue culture and enzyme immunodiagnosis for rapid laboratory diagnosis of rabies. Journal of clinical microbiology, 1989, 27: 51 9-523.

32. Torres-Anlel MJ, Tshikuka JG. Preliminary evaluation of an experimental rapid rabies enzyme (ELISA) immi~nodiagnosis (RREID) kit in a rabies virus tissue culture/weanliny rat system. Rabies information exchange, 1988, 17: 38-39.

33. Flamand A. Wilttor T J Koprowsk H. Use of hybridoma monoclonal antibodies in the detection of antigenc differences between rabies and rabies- related virus proteins. I. The niicleocapsid protein. II. The glycoprotein. Journal of general virology, 1980, 48: 97-1 09.

34. Lafon M, Lafage M Antiviral activity of monoclonal antibodies specific for the internal proteins N and NS of rabies virus. Journal of general virology, 1987, 68: 31 13-31 23.

35 D~etzschold B et al Antigenic diversity of the glycoprotein and nucleocapsid proteins of rabies and rabies-related viruses implications for epdem~oloyy and control of rabies Review of infecbous diseases 1988, 10 785-798

36 Vincent J Bussereaii F, Sureau P Immunological relationsh~ps between rabies virus and rabies-related viruses studied with monoclonal antibodies to Mokola virus Annales de l'lnstitut Pasteu! Virology 1988. 139 157-1 73

37 Smith JS Monoclonal antibody studies of rabies in insectivorous bats of the United States Review ol infect~ous diseases 1988 10 637-643

38. Smith JS, Rabies virus epitopic variation: use in ecologic studies. Advances in virus research 1989, 36: 21 5-253.

39 Rtipprecht C et a1 Epidemiology of rabies virus variants Americari jourr~al oi epiden~~ology 1987 126 298-309

40. Smith JS et al. Epidemiological analysis of street rabies viruses from en7ootic areas of the United States, In: Kuwert E et al., eds. Rabies iri the tropics. Berlin, Springer-Verlag. 1985: 604-61 0.


41. Schneider LG et al. Application of rnonoclonal antibodies for epidemiological investigations and oral vaccination studies, l . African viruses. 1 1 . Arctic viruses. Ill. Oral rabies vaccine. In: Kuwert E et al.. eds. Rabies in the tropics. Berlin, Springer-Verlag. 1985 47-59.

42. Selimov M et al. Antigenic variation in rabies viruses in the USSR. Rabies information exchange, 1988, 17: 40-42.

43. Report of the First WHO Consultation on rnonoclonal antibodies for rabies diagnosis and research, Geneva, 16- l8 September 1982. Geneva, World Health Organization, 1982 (unpublished document WHO/Rab.Res./82.15; available on request from the Division of Communicable Diseases, World Health Organization, 121 1 Geneva 27, Switzerland).

44 Report of the Second WHO Consultation on monoclonal antibodies for rabies diagnosis and research, Tonbach 27-28 May 1984 Geneva World Health Organization 1984 (unpublished document WH0:Rab Res 184 20 and corri- gendum available on request from the Division of Communicable Diseases World Health Organization 121 1 Geneva 27 Sw~tzerland)

45. Report of the Third WHO Consultation on rnonoclonal antibodies for rabies diagnosis and research, Paris, 1-2 June 1985. Geneva, World Health Organl- zation, 1985 (unpublished document WHO/Rab.Res./85.22; available on request from the Division of Communicable Diseases, World Health Organi- zation, 121 1 Geneva 27, Switzerland).

46. Report of the Fourth WHO Consultabon on monoclonal antibodies for rabies diagnosis and research, St Sirnons Island, Georgia, 26-28 May 1986. Ge- neva, World Health Organization, 1986 (unpublished document; available on request from the Division of Communicable Diseases, World Health Organi- zation, 121 1 Geneva 27, Switzerland).

47. Report of the Fifth WHO Consultation on rnonoclonal antibodies for rabies diagnosis and research, Geneva, 3 March 1989. Geneva, World Health Organization, 1989 (unpublished document WHO/Rab.Res./89.33; available on request from the Division of Communicable Diseases, World Health Organization, 121 1 Geneva 27, Switzerland).

48. Report of the Sixth WHO Consultation on monoclonal antibodies for i-abies diagnosis and research, Philadelphia, PA, 2-3 April 1990. Geneva, World Health Organization, 1990 (unpublished document WHO,:Rab.Res,/90.34;

available on request from the Division of Communicable Diseases. World Health Organization, 121 1 Geneva 27, Switzerland).

49. Lumio J et al. Human rabies of bat origin in Europe. Lancet, 1986, i: 378

50. Smith JS et al. Unexplained rabies in three immigrants in the United States. A

virologic investigat~on. New England journal of medicine, 1991. 324: 205-21 l .

OVERVIEW

51. WHO Expert Committee on Rabies. Seventh report. Geneva, World Health Organization, 1984 (WHO Technical Report Series, No. 709). 58-60.

52. Report of a WHO Consultation on requirements and criteria for field trials on oral rabies vaconation of dogs and wild carnivores, Geneva, 1-2 March 1989. Geneva, World Health Organization, 1989 (unpublished document WHO/ Rab.Res./89.32; available on request from the Division of Communic- able Diseases, World Health Organization, 1211 Geneva 27, Switzerland),

53 Schneider LG. Spread of virus from the central nervous syste.m. In Baer GM, ed. The natural history of rabies; Vol. I . New York, Academic Press, 1975: 273-301

54. Dierks RE. Electron microscopy of extraneural rabies infection, In: Baer GM, ed. The natural histor-y of rabies, Vol. l . New York, Academic Press, 1975. 303-31 7.

55. Schneider LG. The cornea test; a new method for intra vitam diagnosis of rabies. Zentralblatt fur Veterinarmedizin, Reihe B, 1969. 16. 24-31

56. Cifuentes E, Calderon E, Biglenga G. Rabies in a child diagnosed by a new intra vitam method: the cornea test. Journal of tropicalmedicine and hygiene, 1971, 74: 23-25.

57. Mathuranayagam D Vishnupriya Rao P. Antemortem diagnosis of human rabies by cornea1 impression smears using irnmunofluorescent technique Indian journal of medical research, 1984. 79: 463-467.

58. Bryceson ADM et a1 Demonstration during life of rabies antigen in humans. Journal of infectious diseases, 1975, 131 : 71 -74.

59. Blenden DC et al. lmrnunofluorescent examination of the skin of rabies- infected animals as a means of early detection of rabies virus antigen, Journal of clinical microbiology, 1983, 18: 631 -636.

60. Blenden DC. Use of irnmunofluorescence examinat~on to detect rabies virus antigen in the skin of humans with clinical encephalitis. Journal of infectious diseases, 1986, 154: 698-701

61. Anderson LJ et al. Human rabies in the United States, 1960-1979. Epide- miology, diagnosis and prevention. Annals of internal medicine, 1984, 100: 728-735.

62. Sureau P. Diff~culties in intra vitam laboratory diagnosis of human rabies Rabies bulletin Europe, 1988. January-March: 8-9.

63. Atanasiu P. Quantitative assay and potency test of antirabies serum and immunoglobulin. In: Kaplan MM, Koprowski H, eds. Laboratory techniques in


rabies, 3rd ed. Geneva, World Health Organization, 1973 (WHO Monograph Series, No. 23): 314-318.

64. Wiktor TJ. Tissue culture methods. In: Kaplan MM, Koprowski H, eds. Laboratory techniques in rabies, 3rd ed. Geneva, World Health Organization, 1973 (WHO Monograph Series, No. 23): 101-123.

65. Smith JS, Yager PA, Baer GM. A rapid tissue culture test for determining rabies-neutralizing antibody. In: Kaplan MM, Koprowski H, eds. Laboratory techniques in rabies, 3rd ed. Geneva, World Health Organization, 1973 (WHO Monograph Series. No. 23): 354-357.

66. Fitzgerald EA et al. A collaborative study on the testing of rabies immune globulin (human) by the mouse neutralization test (MNT) and the rapid fluorescent focus inhbition test (RFFIT). Journal of biological standardization. 1979, 7: 67-72.

67. Guillemin F et al. Comparison of two methods of titration of rabies neutralizing antibodies. Journal of biological standardization, 1981, 9: 147-156.

68. Garnham JZ et al. Application of [tie soluble antigen fluorescent antibody (SAFA) test to the serodiagnosis of rabies Journal of immunological methods, 1977, 14: 147-162

69. Dierks R E . Gough PM Passive haemagglutinat~on test for rabies antibodies. In. Kaplan MM, Koprowski H, eds. Laboratory techniques in rabies, 3rd ed. Geneva, World Health Organization. 1973 (WHO Monograph Series, No. 23) 147-1 50.

70. Wiktor TJ. A radioimmune assay for rabies binding antibody. In: Kaplan MM, Koprowski H, eds. Laboratory techt~iques in rabies, 3rd ed. Geneva, World Health Organization, 1973 (WHO Monograph Series, No. 23). 182-185

71. Haase M , Seinsche D. Schneider W. The mouse neutralization test in comparison with the rapid fluorescent fociis inhibition test: differences in the results in rabies antibody deterrnnations. Journal of biological slandardiza- tioii. 1985. 13: 123-1 28.

72. Kurz J et a1 Comparative studies of the two potency tests for anti-rabies serum: neutralization test it? mice (MNT) and rapid fluorescent focus inhibi- tions (RFFIT) Developtnenrs h biolog~cal sfandardiraiion. 1986, 64 99-1 07.

73 Gluck R et al. Human rabies immunoglobulin assayed by the rapid fluorescent focus inhibirion test suppresses active rabies immuni7ation Joi~rna! of biological standardization, 1987, 15 177-1 83.

74 Zalan E. Wilson C, Pukits D. A microlest for the quanttation of rabies virus neutralizing ant~bodies. Journal of biological standardizaiion, 1979, 7 : 21 3-220.

OVERVIEW

75. Guillernin F et al. Resultats compares des titrages des antcorps antrabiques par deux methodes utlisant I'immunofluorescence [Comparison of results of rabies antibody titres by two immunofluorescence tests.] Journal of b ~ o - logical standardization, 1981 9: 157-1 61.

76 Perrin P et al Application d un methode mmunoenzymatique au titrage des anticorps rabiques neutrallsants en culture celluia~re [Application of an

enzyme immunoassay to the ttration of rables-neutrali7ing antibodies In cell culture ] Journal of biological standard~zation 1985 13 34-42

77. War-ideler Al. Crassi M. Peterhans E. Enzyme-linked immunoadsorbent assays (ELISAs) for rabies antibodies. In: Thraenhart 0 et al., eds. Progress ~n rabies control. (Proceedings of the Second International IMVI Essen/WHO Symposium on '.New Developments in Rabies Control': Essen, 5-7 July 1988 and Report of the WHO Consultation on Rabies, Essen, 8 July 1988.) Royal Tunbridge Wells, Wells Medical, 1989: 381-383.

78. Thraenhart C, Kuwert EK. Enzyme imm~inoassay for demonstrailon of rabies virus antbodies after immunization. Lancet, 1977. i : 399-400.

79. Thraenhart 0 et al. Antibody nduct ion determined by the mouse neutralization test (MNT), r a p d fluorescence focus-nhibition test (RFFIT) and the Essen-enzyme-linked immunoadsorbent assay (Essen-ELISA) is correlated. In Thraenhart 0 et al.. eds. Proyress in rabies control, (Pro- ceedings of the Second lnternatioi'ial IMVI Essen,'WHO Sy~nposium on "New Developments in Rabies Control". Essen. 5-7 July 1988 and Report of the WHO Consultation on Rabies, Essen, 8 July 1988.) Royal Tunbridge Wells, Wells Medical, 1989. 384 -394.

80. Smith JS, Suniner WJ, Roumillat LF. Enzyme immunoassay for rabies antibody in hybridoma culture fluids and its application to differentiation of street and laboratory strains of rabies vrus. Journal of clinical microbiology, 1984, 19: 267-272

81 Perrin P et al The influence of the type of imniunosorbent on rabies ant,body EIA advantages of purified glycoprotein over whole virus Journal of biological standardization, 1986 14 95-1 02

4 u ose 82. Heberling RL, Kalter SS. Rapid dot-immunobinding assay on nitroceI1 l for viral antibodies, Journal of clinical microbiology, 1986, 23: 109-1 13.

83. Heberling RL et al. Serodiagnoss of rabies by do t -mmunobndng assay. Journal of clinical microbiology, 1987, 25: 1262-1 264.

84. Thraenhart 0. Dot-ELISA for fast determination of the Immune status against rabies with one drop of blood only. In: Thraenhart C et al., eds. Progress in rabies control. (Proceedings of the Second Internat~onal IMVI Essen//WHO Symposium on "New Developments in Rabies Control': Essen, 5-7 July 1988 and Report of the WHO Consultation on Rabies, Essen, 8 July 1988,) Royal Tunbridge Wells, Wells Medical, 1989 395-402.


85. Sureau P, Perrin P Rollin P. The newly developed rabies agglutination test (RAT) for rapid detection of post-vaccination antibodies. Journal of biologicai standardization, 1989. 16: 281-286.

86 Thraenharl O Results of an inquiry about the use of the NiH test for rabies vaccine potency testing Suggestions for further improvements Geneva World Health Organization, 1986 (unpublished document available on request frorn the Dlvision of Communicable Diseases World Health Organi- zation 121 1 Geneva 27 Switzerland)

87 Barth R. Jaeger 0 NIH potency tests for inactivated rabies vaccines Zentralbiatt fur Bakterioiogie 1979, 169 488-494

88. Barth R. Diderricll G, Weinniann E. NIH test, a problematic method for testing potency of inactivated rabies vaccine. Vaccine. 1988, 6' 369-377.

89. Report of discussions on improvement of potency tests for all types of rabies vaccines and of potency requirements for animals rabies vaccine, Geneva, 23-24 September 1982. Geneva, World Health Organ~zation 1982 (unpublished document WHO'Rab.Res.:'82 17; available on request from the Division of Communicable Diseases, World Health Organization. 1211 Gen- eva 27. Sw~tzerland).

90 Report of d WHO Workshop on NIH potency test for rabies vaccine Geneva 4-6 November 1985 Geneva World Health Organization 1986 (unpublished document VPH 86 63 available on request from the Division of Communi- cable Diseases World Health Organization 121 1 Geneva 27 Switzerland)

91. Report of a WHO Consultation on potency tests for rabies vaccines. Geneva. 7-8 December 1987. Geneva, World Health Organization, 1988 (unpublished docurrlent VPHi88.73: available on request from the Division of Communi- cable D~seases, World Health Organization, 1211 Geneva 27, Switzerland).

92 Report of a WHO Consiiltation on rabies Essen B July 1988 Geneva, World Health Organization 1988 (iinpublished document WHO/Rab Res 8830 available on request from the Divlsion of Communicable Diseases World Health O r g a n ~ a l i o n 121 1 Geneva 27 Switzerland)

93. Aubert M F , Andral L, Blancou J. Activity control of inactivaled rabies vaccines. Critical study of the NIH test using the so-called "peripheric test" as an experimental model. Journalof bioiogicalstaiidardiraiion, 1981.9. 35-43.

94 Barth R et al The antibody-binding test a useful method for quantitative determination of inactivated rabies virus antigen Journal of biological stdr~dardization 1981 9 81 -89

95. Barth R et al Validation of an in v~t ro assay for the determination of rabies antigen. Developments in bi'oiog~cai standardization, 1986, 64 87-92.

OVERVIEW

96. Ferguson M, Schld GC. A single radial mmunodiffusion technique for the assay of rabies virus glycoprotein. Journal of general virology, 1982, 59. 197-201.

97 Ferguson M , Seagroatt V Sctild GC Single radial imrnunodiffusion assays for the standardization of the antigenic content of rabies vaccines Develop- ments in biologica! standardization, 1984, 64 81 -86

98. Ferguson M , Seagroatt V SchId GC. A collaborative study on the use of single radial immcinodiffusion for the assay of rabies virus glycoprotein Journal of biological standardization, 1986, 12: 283-294.

99. Ferguson M et al. The effect of strain dfierences on the assay of rabies virus glycoprotein by single radial immunodiifusion. Journa! of biological standardizat~on, 1987, 15: 73-77.

100 Atanasiu P et al Titrage mmunoenzymatlque de la glycoproteine Une technique in vitro pour Iappreciat~on de 'activite des vaccins antl- rabiques' [Enzyme imniunoassay of glycoprotein An in vitro technique for measuring the actlvlty of rabies vacclnes] Journal of biological standardization 1980 17 291-309

101, Lafon M et al. Use of a monoclonal antibody for quantitation of rabies vaccine glycoprotein by enzyme immunoassay. Journa! of biological standardization, 1985, 13: 295-301

102 Thraenhart 0 Ramakrishnan K Standardization of an enzyme irnmuno- assay for the in vitro potency assay of inactivated tissue culture rabies vaccines determination of the rabies virus glycoprotein with polyclonal antisera Journal of bioiogica! standardization 1989 17 291-309

103. Report of a WHO Consultation on rabies vaccine potency testing, Nancy, 2-3 May 1991, Geneva, World Health Organization, 1991 (unpublished document; available on request from the Division o i Communicable Diseases, World Health Organization, 121 1 Geneva 27, Switzerland).

104. Dietzschold B et al. Induction of protective immunity against rabies by immunization w ~ t h rabies virus ribonucleoprotein. Proceedings of the Na- tiona! Academy of Sciences of the Un~ted States of America, 1987, 84: 91 65-91 67

105 Toills M et a1 Immunization of monkeys with rabies rrbor~ucleoprotein (RNP) confers protective immun~ty agalnst rabies Vaccine 1991 9 134-136

106. WHO Expert Committee on Biological Standardization. Eighth report. Geneva, World Health Organization, 1955 (WHO Technical Report Series, No 96).


107 WHO Expert Coinmittee on Biolog~cal Standardization. Thirty-hfth report Geneva, World Health Organizatori, 1985 (WHO Technical Report Series, No. 725).

108 Unitage of the International Standard of Rabies Irnmunoglobulin. Weekiy epidemiological record, 1989, 3 15-1 6

109 WHO Expert Committee on Biolog~cai Standardi7ation Thirty-ninth report Gciieva, World Health Organi7ation 1989 (WHO Technical Report Series, No 786)

110. Tordo N et al. Primary structure of leader RNA and nucleoproiein genes iri the rabies genome, segmented homology with VSV. Nucleic aods research, 1986, 14. 2671 -2683

1 1 1. Tordo N et al. Walking along the rabies genome. is the large G-L intergenic region a remnant gene7 Proceedings of ti le NationalAcademy of Sciences of the United States of America, 1986. 83: 3914-3918.

112 Tordo N et al. Completion of the rabies virus genome sequence determination: highly conserved domains among the L (polymerase) proteins of unsegmented negative-strand RNA viruses. Kroiogy, 1988, 165: 565-576.

113. Lafon M, Wiktor TJ, MacFarlan RI. Antigenic sites on the CVS rabies virus glycoprotein, analysis with monoclonal antibodies. Journal of general viroi- O ~ Y , 1983, 64: 843-845.

114. Lafon M, ldeler J, Wunner WJ, Investigation of the antigenic structure of the fables virus glycoprotein by monoclonal antibodies. Developments in biological standardizahon, 1984, 57: 21 9-225

115. Lafon M , Wiktor TJ. Antigenic sites on the ERA rabies virus nucleoprote~n and on structural protein. Journal of general virology, 1985, 66: 2125-2133.

116 Lafon M Lafage M Antiviral activity of monoclonal antibodies specific for the internal proteins N & NS of rabies virus Journalofgenerai virology, 1987, 68 31 13-3123

117. Dietzschold B et al. Localization and imrnunolog~cal characterisation of

antgenic domains of the rabies virus internal N & NS proteln. Virus research, 1987, 8: 1-25.

118. Wunner WH et al. The molecular biology of rabies viruses. Review of infectious diseases, 1988, 10 771 -784.

119, Ert H et al. Iriduction of rabies virus specific T-helper cells by synthetic pept~des that carry dominant T-helper cell eptopes of the viral ribonucleo- proteln. Journal of virology, 1989, 63: 2885-2892.

OVERVIEW

120 Dietzschold B, Ertl HC New developments in the p r e and post-exposiire treatment of rabies. Cr!tical reviews in immunology, 1991, 10. 427-439.

121. Schumacher CL et al. Use of mouse antirabies monoclorial antibodies in postexposure treatment of rabies Joi~rnal of clinical investigation, 1989, 84' 971 -975

122 Lafon M et al. Hurnan monoclonal antibodies specific for the rabies virus

glycoprotein and N protein. Journalofgeneral v~rology, 1990, 71: 1689-1696.

123. Prehaud C et al. lmmunogenic and protective properties of rabies virus glycoprotein expressed by baculovirus vectors. Virology. 1989. 173 390-399.

124 Prehaud C et al. Expression, characterization and purification of a phosphorylated rabies n~icleoprotein synthesized in insect cells by baculovirus vectors. V~roiogy, 1990, 178. 486-497.

125. Richman DD et al. Rapid viral diagnosis. Jourr?alof~nfect~ous diseases, 1984, 149: 298-31 0.

126. Ermine A, Tordo N Tsang H. Rapid diagnosis of rabies infection by means of a dot hybrid~zation assay. Molecular and celiular probes, 1988, 2. 75-82.

127. Shankar L', Dietzschold B, Koprowsk~ H. D~rect entry of rabres virus rnto the central nervous system without prior local replication. Journal of virology, 7 991, 65' 2736-2738.

128. Jackson AC, Wunner WH. Detection of rabies virus genomic RNA and mRNA in mouse and human brains by using in situ hybridization. Journaiof virology, 1991. 65: 2839-2844.

CHAPTER 3

Characteristics and molecular biology of the rabies virus N Tordo '

Introduction

Since the early 1980s the availability of molecular biology techniques has made possible a fundamental examination of the causative agents of many d~seases, including rabies The purpose of this chapter IS to review current knowledge of the characteristics of the rabies vlrus to summarize the progress made during the past decade and to atternpl to delineate the main challenges for the future

The viruses that cause rabies encephalitis belong to the Lyssawrus genus of the Rhabdoviridae family They were originally classified in four serotypes on the basis of serological and antigenic relationships serotype 1 comprised the classical rabies vlruses notably the ' street (wild) and vaccinal strains sero-

types 2 3 and 4 were rabies-related viruses prototyped by Lagos bat Mokola and Duvenhage viruses, respectively The vaccinal strains of serotype l have l~ttle or no protective effect against rabies-related viruses Genetic studies have conf~rmed and extended thls classif~calion (I) Four genotypes corresponding to the four serotypes have been character i~ed In addition the recently identified European bat lyssaviruses (EBLI and EBL2) have been classified in genotypes 5 and 6

Morphology and structure

Morphology

The virions or virus particles have a bullet-shaped structure (Fig 3 l ) , with a diameter of 75 nm and a length of 100-300 nm (2-6) Variations in the length can be observed between rabies strains (e g CVS IS usually longer than PV) or can reflect the presence of defective interfering (D!) particles which occur when the multiplicity of infection is high The D1 particles possess a truncated genorne and are therefore defective in various viral functions and must depend upon infectious virions to comple~nent their deficiency Since their smaller genomes replicate rapidly these particles compete efficiently with normal genomes for encapsidatlon into the virlon

The virion can be dvided into two structural un ts ( F g 3 2) a central and dense cylinder formed by the helical ribonucleocapsid and a thin surrounding envelope (8 nm wide) covered with spike-like projections which are 10 nm in length and 5 nm apart The helical r~bonucleocapsid is extremely compact as indicated by the huge random ribbon escaping from the flat end of partially degraded virions

' Head, Lyssavlrus Laboratory Pasleur Institute P a r ~ s , France

28

CHARACTERISTICS AND MOLECULAR BIOLOGY OF THE VIRUS

A Fig. 3.1 Morphology d the rabies virus (negative-staining electron micaoscopy)

A Rabies vinon (genotype I PV stra 11)

B Intact Mokola vri~ons (genotype 3 Mok 5 s t ia~n) C Part~aIly dtsrupied Mokola v i~ ions jgenotvpe 3 Mok 5 stratn)


Structure

Biochemical studies have demonstrated that each vr ion is composed of a single molecule of genomc RNA and five proteins showing post-translational modifications:

1. The RNA-dependent RNA polymerase (L protein). which has a high relative molecular mass.

2. The glycoprotein (G), which exists in two forms with different extents of glycosylation, both of which are fatty acylated with palmitic acid.

3. The nucleoprotein (N). which is phosphorylated. 4. A phosphoprotein (M 1 ) . which has two forms w ~ t h different extents of ptiosptlo-

rylation. 5, The matrix or membrane protein (MZ), which has two conformations differing in

the number of internal disulfide bridges Both forms are palmitoylated.

The relative molecular masses and the number of copies of each protein per v~rion vary slightly between studies, owing to differences in both the estimation techniques used and in the viral strains studied. The values given in Fig. 3.1 are averaged from numerous reports (6). A more recent report proposes slightly different values (7).

Proteolytic treatment of the vr ion with trypsin removes the spike-ike projections and selectively affects the G protein (8. 9). Only a small G peptide remains linked with the spikeless particle, indicating that the G protein is membrane- anchored and constitutes the main component of the viral spikes. Each spike consists of three associated G proteins (10).

Following treatment with hypotonic buffer and edetic acid' or non-onic detergents such as octoxinol or octylglucoside, the viral envelope is solubilized and becomes permeable to trypsin. The associated G protein is released first, followed by fractions of the M2 protein. This finding initially suggested that the M2 protein was embedded in the inner layer of the membrane. However, recent studies on vesicular stomatitis virus (VSV) indicate that only part of the M2 protein is in contact with the membrane, the rest being embedded in the centre of the ribonucleocapsid coil ( 1 I). Subsequent treatment of the remaining ribonucleocapsid with an ionic detergent such as deoxycholate releases the M1 and L proteins, but does not affect the association between the RNA genome and the N protein. This latter is so intimate that the genome remains insensitive to digestion by ribonucleases.

Despite their parting in distinct iinits, the five viral proteins are close enough to be experimentally linked by chemical reagents. Whatever its exact position, the M2 protein plays a crucial intermediate and catalytic role withjn and between the viral envelope and the ribonucleocapsid. The M1 protein was originally considered as a component of the viral membrane before b e ~ n g reassessed as belong~ng to the ribonucleocapsid. However, the nomenclature "MI " (for membrane) has not been modified.

' Also known as ethylened~amne tetraacetate or EDTA.

31


Functional analysis of the infection

A productive ~nfection results from a chain of events, consisting first of encounter between the virus and the host, then (heir ~nteraction and finally multiplication of the virus.

Host species

The niaintenance of rabies is ensured by several wild animal species that serve as reservoirs and vectors of the disease in nature. A vector must be highly susceptible to the virus and able to develop interactions favouring the transmission of ihe disease before death. Several particularly efficient hosl-virus cornbinatioi~s have emerged in different geographic regions. Currently, the red fox is the main vector ot rabies in Europe while in North America, the racoon and the skunk are also important In developing countries, dogs remain the major vectors. although wildlife species are also involved Bat species serve as the main reservoirs of lyssavrus genotypes 2, 4. 5 and 6 ancl are also involved in the transmission of genotype 1

Routes of infection

The rabies viius is usually introdiiced by a bite wound although ~nfections have been reported froin aerosols and licks on broken skin or mucous membrane (12) (see also Chapter 1) The virus is neurotropic and rapidly enters the sensory and motor nerve endings ot the peripheral nervous system (13 14) Therefore the rabies virus IS only transently exposed to the inimune systeni although a recent paper suggests that ant~body mediated clearance of rabies from the central nervous system inay occur (15) Once in the neuron the virus is transported in the axons by retrograde axoplasinic flow to the perikaryon where i t undergoes replication The brain stem 1s infected first followed by the thalamus and then the cortex (13) During the later stages of infection however the entire central nervous system is infected as well as certain external tissues such as the salivary glands that ensure the transmission of infection It is not yet clear when the virus starts to replicate Experimental data support either a primary multiplication in the muscular cells (16 17) or on the contrary no multiplication outside the neuronal cells (14 18)

This plogressive infection explains why rabies has such a variable incubation period (generally 1-3 months) corltrasted by a short syrnptomatc period (less than 1 week) The latter ~nvariably leads to death because of the absence of effective therapy (19) The causes of death remain oDscure as only minor histopathological changes appear to occiir In mice the virus has been shown to ca i~se electroptiysiological dysturicton and alterations In sleep notably at the level of rapid eye movement sleep (20)

Penetration into the host cell

The binding of the rabies virus to the cell membrane is thought to be mediated by the G protein However a rabies-specific receptor remains to be characterized On the bass of sequence hornoogy between the external part of the G proteln and the receptor-bind,ng site of snake venom neurotoxns it was postulated that the raoies


virus could bind to the nicotinic acetylcholine receptor (21-23). However, this hypothesis may apply only to muscular cells. For neuronal and fibroblastic cells, it has been shown that oligosaccharides and Ipoprotein components such as the sialic acids of gangliosides may also be involved (13, 24). The rabies receptor appears to be complex and may vary from one cell type to another.

Once fixed on the cell, the rabies virion penetrates it by pinocytosis. The viral membrane then fuses with that of the lysosomal vacuole, releasing the ribonucleocapsid into the cytoplasm The riboni~cleocapsid is sufficient to assure the

transcription of the virus.

Transcription, replication and budding

The rabies virus possesses an unsegrnented negat~ve-stranded RNA genome All viruses presenting this genome structuie belong to three families the Rhabdoviri- dae the Paramyxovir~dae and the Filoviridae (25) Despite a ubiquitous host distribution these viruses have evolved an identical expression strategy The negative polarity means that the genome IS unable to be directly translated into viral proteins by the cell machinery Therefore a preliminary autonomous transcription step IS necessary to produce the complementary positive-stranded mes- seriger RNA (mRNA) as soon as the ribonucleocapsid is liberated in the cytoplasm This step IS assured by a genome-encoded enzyme the RNA-dependent RNA polymerase, which is included in the ribonucleocapsid Typically the viral polymerase does not recognize a naked RNA template but rather one encapsi dated with the N protein as the genome does not undergo uncoating into the cytoplasm

The mechanisms of transcription replication and expression of unsegmented negative-stranded RNA genomes (including that of the rabies virus) were originally established from studies on VSV (6 26-29) It is thought that a single promoter for polynierization is recognized by the transcriptase near the 3'-end of the genome (Fig 3 3) From that point the transcriptional complex progresses towards the 5 - end producing conseci~tive monoc~stronic transcripts first one small uncapped non-polyadenylated leader RNA then five mRNAs coding successively for the N M1 M2 G and L proteins To control this sequential progression the transcriptional complex recognizes start (S) and stop or polyadenylation (P) transcription signals flapking the cistrons These consensus sequences are approximately ten nucleotides in length The complex IS thought to dissociate from the template at each stop signal and to re-initlate poorly at the next start signal T h ~ s may be partly due to the size of the non-transcr~bed intergenic region which could impair the accessibility of the complex to the start signal This mechanism results in a progressive decrease in the rate of transcription from the 3'- to the 5'- end of the genome suggesting that the genornc location of a cistron directly influences its rate of transcription

A typical feature of transcription of the rabies genome is the phenomenon of alterriative termination due to the presence of two consecutive stop signals for the M2 arid G cistrons (30) Both signals can be alternatively used to produce either a large or a small messenger Because the transcription complex is thereby released nearby or far upstream from the next start signal alternative termination influences the efficiency of transcription of the distal gene by modifying the size of the non- transcribed intergenic region The ratio between the large and the small messen-


Fig. 3.3 Transcription and replication of the rabies genome

[~romoter for 1 S S S S S

, 8

P b a 3 # 8 , 8 , 8 D , P promoter for 8 8 M1& : : 4 8 0 8

8 ,

P small ~ 2 4 , , , , large M 2 4 , ,

8 ,

P: stop (polyadenylation) L W / 4 3 '

a: capping

,' y genome

5 ' : '"I"' "

I n I m

antigenome I 2 j 0

i 3 I P

genomes

TRANSCRIPTION

ger varies during the course of infection, and is different in fbroblast~c and neuronal cells. This suggests that alternative termination is a mechanism for regulating the expression of the rabies genome. which could be Influenced by factors of viral or cellular origin, such as proteins or peptides. The latter hypothesis suggests that the neurotropism of rabies should be investigated at the trans- criptonal level. lnterest~ngly, alternatve terminaton is only observed with certain fixed laboratory strains of rabies virus. such as PV, Pasteur and ERA, It is absent from the wild isolates sequenced up to now, and also from other fixed strains such as CVS, PM and HEP (31).

After transcription, the rabies messengers are processed through the cellular translation machinery. Most of them are expressed in the cytoso by free poly- ribosomes. However, the G protein is produced as a transmembrane molecule in the rough endoplasmic reticuum and is then transported via the Golgi apparatus to the cytoplasmic membrane of the cell, the glycosylated part facing the exterior.

It IS only after trailslation of the rnRNAs into rabies proteins that the replication step can begin. This leads to the synthesis of a positive-stranded antigenome that serves, in turn, to amplify negative-stranded genomes for the progeny virions. To


become functional templates for future transcription, the genome and antigenome have to be coated with N proteins. This requires an exact coordination between polymerization and encapsidation Therefore, a promoter for encapsidation must exist near their 5'-end to initiate the concomitant encapsidation of the growing RNA.

Before viral budding, transcription arid replication are inhibited. Simulta- neously, there is intense condensation and coiling of the ribonucleocapsids. These events take place in speclfrc areas of the cell membrane where transmembrane G

prote~ns have been concentrated. The vir~on leaves the cell by budding, the lipidic envelope being pulled out from the host-cell membrane.

The ribonucleocapsid complex, which consists of the RNA genome and the N,

M1 and L proteins, is the niininial virion structure exhibiting RNA synthesis. The respective role of each viral polypeptide in this process has been studied less deeply than the mechanisms themselves. On the bass of studies on VSV, it is known that the M1 and L proteins are the catalytic elements involved in the polyrnerase function. The L protein is the actual RNA-dependent RNA polymerase possessing most of the required enzymatic activities, including RNA synthesis, capping, polyadenylaton, and partial kination of the N and M1 proteins. The role of the M1 protein is probably more regulatory: it helps the RNA-dependent RNA polymerase so that it binds correctly to the promoter for polymerization; it possibly uncoats the RNA template upstream of the polymerization complex; and it checks the amount of N protein available for encapsidation.

The N protein plays a major structural role in encapsidating the RNA genome and antigenome. It is therefore involved in the switch between transcription and replication, as suggested by several studies on VSV. These studies indicate that replication cannot begin in the absence of sufficient quantities of N protein to encapsidate the growing template. As long as the amount of N protein remains below a certain level, the transcription complex recognizes a stop signal (T) at the. end of the leader gene and releases the leader RNA. It then recognizes the start and stop signals flanking the N, M l , M2. G and L genes and produces the naked mRNAs When there are sufficient amounts of N protein, the transcription complex participates in a concurrent assembly of the growing template and somehow is responsible for the polymerase ignoring the stop signal. This event corresponds to the switch between transcription and replication The viral polymerase then becomes unable to recognize the other start and stop transcription signals. Since the leader RNA is no longer produced, the promoter for encapsidation remains linked to the 5'-end of the growing genome, which is progressively encapsidated in a cooperative manner.

The M2 protein plays an important role during the latter stages of infection, notably dur~ng morphogenesis. Its position, between the ribonucleocapsid and the membrane, permits its interaction with both internal (N protein) and external (cytoplasmic tail of the G protein) proteins. It is able to inhibit RNA synthesis, mediate the coiling of the ribonucleocapsid, and concentrate the G protein, all of which are required before the virion buds out of the infected cell.

Viral proteins involved in the host immune response

Although all the viral proteins show antigenicty, they do not all play the same role in protection (32). The purified G protein has been shown to protect against an


intracerebral challenge with rabies virus, while the purified ribonucleocapsid only protects against a peripheral challenge (33). The G protein is the only rabies antigen that consistently induces virus-neutralizing antibodies (34). This property mainly depends on the preservation of its three-dimensional structure, although a linear neutralizing epitope has been identified (35). On the other hand, it shares the capacity to induce a cellular immune response involving both T-helper cells and cytotoxic T cells with the N and M 1 proteins. The T-cell response is thought to play an important role in the immune response to rabies (36).

These studies suggest that the G protein is the most important antigen for immunization. However, the N protein is also important for two principal reasons: (i) because of its capacity to substantially enhance the T-helper cell immune response to rabies vaccination; and (ii) because it is less variable than other antigens. This suggests that the N protein is the best candidate to Increase the protection spectrum of vaccines, notably to distant rabies-related viruses ( 1 , 33).

Molecular biology of the rabies virus

Since 1981, when the glycoprotein mRNA of the ERA strain became the first rabies gene to be cloned and sequenced (37), considerable advances have been made in determining the sequence of the rabies genorne (Fig. 3.4). Two basrcally distinct strategies were adopted: (i) cloning the viral mRNAs; (ii) cloning the RNA genome directly. The latter strategy also enabled the non-transcribed intergenic regions to be studied. As shown in Fig. 3.4, the only rabies (genotype I ) genomes that have been completely cloned and sequenced are those of the PV strain (1 1 932 bases) (38-40) and the SAD-B19 strain (11 928 bases) (41). To date the rabies-related strains have been only partially sequenced (Fig. 3.5).

With the development of the polyrnerase chain reaction (PCR), it has now become possible to amplify any rabies gene (42, 43). The sensitivity of this technique permits the isolation of the original rabies strains present in the infected tissue, without any cell culture or brain adaptation. A new technique combining PCR amplification and direct sequence analysis has been developed (42) (see also Chapter 13). This powerful tool of molecular epidemiology is currently leading to a dramatic increase in our fundamental knowledge of the lyssaviruses.

Considerable progress has also been made in expressing the rabies genes, since they are all now available. Fundamental aspects have been invest~gated through the expression of the L protein, in order to evaluate its role during transcription and replication (44). With this exception, most of the studies carried out have focused on the development of a recombinant vaccine expressing the genes coding for the G and N proteins, which are involved in the immune response to rabies infection (45) (see also Chapter 35).

The analysis of sequence data has led to substantial progress in understanding the structure-function relationships of the various elements of the rabies virus (6, 24). It has also led to an analysis of the evolution of the Lyssavirus genus (I), and more generally of the Mononegavirales order (46).

The leader RNA

This is a small RNA (57-58 ribonucleotides), which is very rich in A residues transcribed at the 3'-end of the genome. During transcription it carries the

fig. 3.5 Wild isolates of rabies (genotype 1) and rabies-related viruses (prototypes of genotypes 2-6) that have been studied at molecular levela, r B

Genotype 1 Genotype 2 Genotype 3 Genotype 4

World Nigeria Zimbabwe

1 ?&$ cultures

Genotype 5 Genotype 6 European bat European bat lyssavirus 1 lyssavirus 2

South South Poland France Finland Netherlands Africa Africa 1971 1981 1985 1989 1986 1989

Genes Rab~es Lagos Mokola 5 Duvenhage l Duvenhage 2 EEL 1 EEL l EEL 2 EBL 2 bat (Mok5) (Duvl) (Duv2)

Leader 1 N 1 , 4 1 1 1 1 1 1 1 1

M1 1 M2 1

G 1,5 1 1 1 1 1 1 1 1

W 1 1

L , l P 5' -end 1 X

1 lnteraene 8 1. WHO CollabMating Centre for Reference and Research on Rabies, Pasteur Institute, Pans. France. 7. Genentech, San Francisco, CA. USA. 2. WHO COllabMEiting Centre far Reference and Research on Rabss, The Wlstar lnstnute of Anatomy and Biology, Philadelphia. PA. USA. 8. Faculty of Pharmaceutical Sciences, Kyoto University. Kyoto. Japan. 3. WHO Cdlaborating Centre for Rabies Suwetllance and Research. Federal Research Institute for Animal V~rus Dtseases. Tijbingen, Germany. 9. Medical College, Oita. Japan. 4. WHO Collaborating Centre for Reference and Research on Rabies, Center for Infectious Diseases. Centers tor Disease Control. Atlanta. GA. USA a The strains listed are those that have been partially or totally sequenced 5. Centre National de la Recherche Scientaique, Gif-sur-Yvene. France. b Numbers refer to laboratories where the sequences were petformed. 6. Connaught Centre for Biitechnolcgy Research, Willowdale, Ontario. Canada.


promoter of encapsidation and cleaves it from the distal mRNA transcripts. In VSV, the leader RNA has also been implicated in the shut-off of the host-cell macro- molecular synthesis (47). However, this effect is not usually observed during infection with the rabies virus (48).

The N protein

This is a 450-amino acid long polypeptide which is phosphorylated on a serine

residue in position 389 (49). It shows a segmented homology with the N protein of VSV, ~nvoiving rna~nly the central part of the protein (38) The stretches of

conserved amino acids are probably those directly interacting with the RNA genome, since one function of the protein is to encapsdate and protect the genome.

Monoclonal antibodies directed against the N protein have been used to differentiate and classify rabies wild isolates into types of lyssaviruses (50-52) (see also Chapter 12). Three antigenic sites have been characterized along the protein, but only two of them are mapped. Sites l and Ill involve the stretches of amino acids in position 374-383 and 313-337, respectively Several immunodominant T-helper epitopes have also been characterized (53), one of which (in position 404-418) is protective for mice wheii coupled with the linear epitope of the G protein (54).

The M1 protein

This highly hydrophilic protein is 297 and 303 amino acids long in rabies and Mokola virus respectively It presents different phosphorylation states (55) and possesses numerous serine and threonine amino acids which anchor the phosphate residues Phosphorylation provides an overall negative charge which IS

increased by the very high content of acidic (aspartate and glutamate) amino acids The primary nucleotide sequence of the M 1 protein is very poorly conserved between rabies virus and Mokola virus particularly in the central part (position 55-200)

Two antigenic sites have been characterized along the M1 protein both located in position 75-90 Recently immunodominant cytotoxic T epitopes and T- helper epitopes were identified in position 191-206 (56) Interestingly, they are located partly in the poorly conserved central region (position 191-197) and partly in the conserved carboxy side (position 198-206)

The M2 protein

This is a 202-amino acid long polypeptide. It plays an intermediate role between the r~bonucleocapsid and the viral membrane, presupposing the presence of regions interacting with both elements. As with VSV, the 40 amino terminal residues, rich in charged and prolne residues (57), could mediate the inhibitory effect on transcription and replication before the ribonucleocapsid is coiled. This region seems to be involved in the host immune response to rabies, since a major antigenic determinant was recently located between residues I and 72 (58). A 19-residue central segment (in position 89-107) appears sufticiently hydrophobic to anchor


the protein into the virion membrane (39) The palmitoylation site of the protein which probably involves a cysteine residue remains to be characterized (59) However it does not appear to be In dlrect proxtmlty to the presumed mernbrane- binding site As discussea earlier recent studies on VSV suggest that the M2 prote~ri extends from ihe inner layer of the viral membrane to the internal core of the ribonucleocaps~d coil ( l l ) Regardless of its exact location the M2 protein is believed to play an important role in the morphogenesis of the virus

The G protein

The G proteiri (Fig. 3 6 ) is the best studied of the rabies proteins, at both the structural and irrirnunological levels (6. 24, 32. 50, 60). It is 524 and 522 amino acids long in rabies and Mokola virus, respectively, and contains two hydrophobic segments typical to its transmembrane nature. The amino terminal signal segment (first 19 residues) initiates the translocation of the nascent protein through the rough endoplasmic reticulum membrane, before being cleaved in the mature protein The translocation process continues up to the transmembrane segment (position 440-461), which remains anchored in the membrane as indicated by the palmitoylation of the cysteine residue in position 461 (59). This segment separates the cytoplasmic carboxy terniinal domain (position 462-505), which interacts with the internal proteins, from the external glycosylated amino terminal domain (position 1-439). The glycosylation and the fatty acylatlon of the protein take place during its transport frorn the rough endoplasmic reticulum to the Golgi apparatus and to the cytoplasmic membrane. One glycosylation site (in position 319) appears to be of major importance, both because it is present (and probably used) in all the lyssavirus strains sequenced up to now, and because it is the only region that shows homology with the G protein of VSV (61). The other sites vary between different strains and are not systematically used.

At least eight antigenic sites have been located on the external domain of the G protein of different rabies strains (I-VI, "a" and G I ) (35,60,6245) . With the exception of sites V1 and G1, which were also identified under denaturating conditions, these sites depend on conformation and the folding of the protein. Only five sites have been definitively mapped. Sites l. Ill, V1 and "a" involve the amino acids in position 231, 330-338, 264 and 342-343, respectively Site l1 is discontinuous and involves two separate stretches in position 34-42 and 198-200 linked by a disulfide bridge. It has been suggested that the current nomenclature should be modified according to the relative importance of the G protein regions in stimulating the B-cell response (60). The term "antigenic site" would be reserved for regions described by numerous different monoclonal antibodies (sites I!, Ill and

"a"), while the word "epitope" would be used for sites recognized by only one nionoclonal antibody (epitopes l. IV. V, V1 and G I ) . I t is noteworthy that the sequences corresponding to the antigenic sites of the rabies G protein appear to diverge in the G protein of Mokola virus.

T epitopes have also been located along the G protein using chemically cleaved or synthetic peptides (66, 67).

The G protein is also involved in the pathogenesis of rabies and is believed to assume at least part of the neurotropism of the virus. In genotype 1 viruses, neurovirulence seems to be directly related to the maintenance of an arginine (or lysine) residue in position 333 (site Ill) (68). Mutants with other amino acids in this


Fig. 3.6 Structure, antigenic sites and epitopes of rabies glycoproteina Antigenic sites and epitopes

, 0

(-1 9) -0 'v R W 440461

Signal g Transmembrane segment m 8 segment

NH2

Ectodomain 1-439

Cytoplasmic domain 462-505

37 158 247 319 333 (Arginine) PV n 504

37 247 319 333 (Arginine) ERA v 504

37 158 31 9 333 (Glutarnine) HEP v 504

37 204 319 333 (A~inine) cvs V 504

37 319 333 (Glularnic acid) Kelev V 504

202 31 9 333 (Aspartic acid) Mokola v 502

- 9 ; l I I l L

100 200 300 400 500 G glycoprotein amino acids

v Potential glycosylation sites.

V Residue involved in neurovirulence.

a The numbering is made on the assumption that the signal segment is cleaved in the mature protein.

position cannot infect certain types of neurons presumably because they are unable to recogrirze the receptor (14, 69) However several attenuated vaccinal strains such as HEP or Kelev are muted in position 333 (70, 71) Furthermore Mokoia virus (genotype 3), which is highly neuropalhogenic in mice and causes more severe encephalitis than rabies virus in dogs and monkeys possesses an aspartate residue in position 333 (72) This suggests that tissue specificity is probably very complex

The L protein

This giant protein of 2142 and 2127 amino acids in the PV and SAD-B19 strains of rabies virus respectively occupies more than half (54%) of the rabies genome Because only minute amounts are present in the virion and in the infected cell t is the least studied of the rabies proteii-is at the b~ochemical and immunological ievel but the best analysed at the theoretical level


One of the main characteristics of the L protein is the sequence homology that it shows with the L protein of other members of the Mononegavirales order which suggests that they have evolved from a common ancestor (40 73) The homology is not randomly distributed along the protein however and some domains are highly conserved with the amino acids in identical positions wh~le others are more variable Such a distiibution of independent conserved domains joined by more variable areas is consistent w ~ t h the multifunctional nature of the L protein (40, 73) It suggests that different activities are linked within the final polypeptide where they may retain a certain degree of independence, as indicated by the comple mentation observed between VSV L protein mutants Several activities were tenta- tively attributed to certain domains providing the first available guidelines for the future dissection of the functions of the L protein by site-directed mutagenesis (73)

It is noteworthy that the key domain catalysing the polymerization of the rabies genome (' polynierase module ) has been mapped The ' polymerase module appears ubiquitous in the animal kingdom since it is similar in all the RNA- dependent polymerases (RNA polymerases and reverse transcriptases) and partly conserved in the DNA-dependent polymerases (74 75) These results illustrate how rabies research can have influence beyond its own domain

Genomic signals

The start and stop transcription sigrials (internal signals) flanking the cistrons of the rabies and Mokola genome have been determined by S1 nuclease mapping experiments (6,39) They are composed of 9 nucleotide long consensus sequences closely related to those of VSV The stop signal is terminated by a sequence of 7 uridine residues which are reiteratively copied by the transcriptase in order to produce the polyadenylation tail of each mRNA before re-initiating at the next start signal

The promoters for polymerization and for encapsidation (external signals) are assumed to be present in the first 11 nucleotides of the 3'- and 5'-ends of the genome respectively (40 72 76) These sequences are strictly conserved and are inversely coniplementary However the genome is not likely to generate a pan- handle structure during transcription or replication since the genornes and antigenomes are encapsidated as soon as they are synthesized

Within a viral genus (Lyssawriis or Ves~culov~rus) all the signals are strongly conserved However when the whole Rhabdoviridae family is considered the internal signals appear more stable than the external ones (Fig 3 7)

The intergenic regions

The intergenic regions of the rabies and Mokola virus genomes (Fig. 3.7) vary both in length (2-24 nucleotides) and nucleotide composition ( l , 39, 72), while those of the VSV genome (Indiana strain) are conserved (mostly GA). This variation could modify the progressive loss of transcriptional efficiency occurring between cistrons, and partly explain the weaker performance of rabies virus compared with VSV during cell infection. slower cycle of transcription and replication, less production, and virtually no inhibition of cellular synthesis.


Fig. 3.7 Genomic signals of rhabdoviruses

Transcription stop signal lntergenic Transcription start signal reglon

A C U U U U U U U (U) Var~abie U U G U G Pu N G A 5' A U A C U U U U U U U G U U G U C N N U A G 5

Signals at the 3' -and 5' -ends of the genome and antigenome Encapsidafion

Lyssav~rus 1 S A C G C U U A A C A A . * . . . . . . . . . ' 3 . U G C G A A U U G U U

Polyrnerizat~on Ves~culovtrits 31 U G C Pv U Pv U N N U N S

Consensus 3' U G C - - - U - - U - 5' rhabdoviius

Fig. 3.8 Comparison of the genome of rabies virus (PV strain) with those of Mokola virus (Mok 5) and vesicular stomatitis virus (Indiana strain)

kb

j2 t

Prolein-coding region ---- Homology

1 2 3 4 5 6 b 1 2 3 4 5 6 7 8 9 1 0 1 1 1 2 k b Rabies genorne (PV straln) Rabies genome (PV strain)

Evolution of the rabies virus

A comparison of the genorne of the rabies virus (PV strain) with those of the Mokola virus (Mok 5) and VSV (Indiana strain) clearly indicates that unequal selective pressuies were imposed on the viral sequences during evolution (Fig 3 8) Between rabies and Mokola genornes representing the two more divergent genotypes of the Lyssavirusgenus the homology decreases in the following order N protein (80°/0 of conserved amino acids) M2 protein (76%) G protein (58%) and M 1 protein (45%) As the Mokoia poiymerase has not yet b ~ e n totaily sequenced ~t is not currently possible to evaluate its divergence rate However from a larger evolutionary distance such as that existing between rabies and VSV genomes representing prototypes of the two genera of the Rhabdoviridae family the

CHARACTERlSTlCS AND MOLECULAR BIOLOGY OF THE VIRUS

polyrnerase remains the only coriserved protein (33'76 of conserved amino acids), together with stretches of the N and G proteins. as discussed previously

It seems, indeed, that the need for conservation of a precise scenario for genome expression has led to a stringent maintenance of the major controlling elements, such as.

- the polymerase encoding the enzymatic activities required for transcription and

replication; - the genon?ic signals (external and internal): - the organization of the genorne, since the decreasing rate of transcrptioii

results in a direct correlation between gene location and expression rate.

Fig 3 9 shows that the organization of the unsegrnented negative-stranded RNA genomes is sirnlar witl? the major structural genes (N M1 and M2 proteins) encoded at the 3 ' sde while the 5 half codesfor the L protein (46) These locations are adapted to the stoichioinetric and catalytic requirements of the respect~ve gene products

Unsegmented negative-stranded RNA genomes generally use most of their length for coding purposes Iri this context the Lyssav~rusgenome is atypical since two substantial non-coding regions lollow the M2 arid G genes respectively The proximal region separates the 3' block of structural genes and the glycoprotein gene (G for Rhabdovridae F for Paramyxovridae) while the distal one separates the glycoprote~n and the polymerase genes A comparison of unsegmented negative stranded RNA gerlonles indicates that both these regions are located in very plastic areas during evolution (46) (Fig 3 9) This suggests that rabies virus is an evolutionary I n k and that the regions may represent remnant protein genes This assumption was mainly developed on the basis of the analysis of the rabies G-L intergene (6 39) which encodes

1. The HN (haemagglutinin-neuraminidase) protein in paramyxoviruses. 2. The NV protein (unknown function) in the infectious haematopoietic necrosis

virus (IHNV), a salmonod rhabdovrus. 3. The remnant $ pseudogene in lyssaviruses, which is similar in si ie to the NV

gene. 4, The dinucleotide GA in VSV, which appears at the time of size restriction.

From a mecl?anistic point of view it is noteworthy that the two M2-G and G-L intergenes of the rabies virus (PV strain) genome are precisely those where the phenomenon of alternative termination is observed during transcription

References

1 Bourhy H et al. Molecular diversity of the Lyssavirusgenus. Virology, 1993, 194. 70-8 l

2 Roots E Elektronenmikroskopische Untersuchungen an Gehrnen bei der expermentellen Tollwutinfekt~on [Electromicroscopic studies on brains in experimental rabies infection] Zeitschrift fur Naturforschung, Section C, Jourra l o f biosciences 1962 17 198-202

3. Matsumoto S. Electron microscopy of nerve cells infected with street rabies virus. Virology, 1962, 17: 156-158


4. Atanasiu P et al. Etude morphologique du virus rabique des rues en culture de tissu. [Morphological study of rabies street virus in tissue culture.] Comptes rendus cfe IXcadkmie des Sciences, Serie /l!, Sciences de !a Vie, 1963, 256: 321 9-3221.

5. Davies MC et a1 The electron lnicroscopy of rabies virus in cultures of chicken embryo t~ssues. Virology, 1963, 41 : 821-832.

6. Tordo N, Poch 0. Structure of rabies virus. In. Carnpbell JB. Charlton KM, eds. Rabies. Boston, Kluwer Academic Publishers, 1988: 2 5 4 5 .

7. Flamand A et al. Mechanisms of rabiesvirus neutraliraton. Virology, 1993, 194; 302-31 3.

8. Dietzschold B et al. Isolation and purification of a polymeric form of the glycoprotein of rabies virus. Journal ot general virology, 1978, 40: 131-139.

9. Delagneau JF, Perrin P. Atanasiu P. Structure of rabies virus: spacial relaton- ships of the proteins G, M1, M2 and N. Annales de l'lnstitul Pasfeuc Virologie, 1981, 132E: 473493.

10. Gaudin Y et al. Rabies virus glycoprotein is a trimer. Virology, 1992, 187: 627-632.

11. Barge A et al Vesicular stomatitis virus M protein may be inside the ribonucleocapsid coil. Journal of virology. 1993, 67: 7246-7253.

12. Constantine DG. Rabies transmission by air in bat caves Atlanta, Public Health Service Publication, 1967.

13. Tsang H. Interactions of rabies virus and host cells. In. Carnpbell JB, Charlton KM, eds. Rabies. Bostori, Kluwer Academic Publishers, 1988: 67-100.

14. Coulon P et al. Invasion of the peripheral nervous systems of adult mice by the CVS strain of rabies virus and its avirulent derivate AvOl. Journal of virology. 1989, 63: 3550-3554

15 Diet7schold B et al Dellneaton of putative mechanisms involved ~n antibody- mediated clearance of rabies virus from the cenlral nervous system Proceed-

ings of the National Academy of Sc~ences of the United States of America 1992 89 7252-7256

16 Fekadu M. Shaddock JH Peripheral d~stribution of virus in dogs inoculated with two strains of rabies virus Amencan journal of veterinaryresearch 1984, 45 724-729

17. Murphy FA et al. Comparative pathogeness of rabies and rabies-like viruses. Viral irlfection and transit from inoculation site to the central nervous system

Laboratory investigation. 1973, 28. 361-376.


18. Shankar V, Dietzschold B, Koprowski H. Direct entry of rabies virus into the central nervous system without prior local replication. Journal of virology, 1991, 65: 2736-2738.

19. Baer GM, Bellini WJ, Fishbein DB. Rhabdoviruses. In: Fields D, Knipe M, eds. Virology. New York, Raven Press, 1990: 883-930.

20. Gourmelon P et al. Electrophysiological and sleep alterations in experimental mouse rabies Brain research, 1986, 398: 128-140.

21. Lentz TL et a1 Amino acid sequence similarity between rabies virus glycoprotein and snake venom curaremimetic neurotoxins. Science, 1984, 226: 8477848.

22. Becker Y. Putative antigenic domains in glycoprotein G of rabies vlrus: is the RGK sequence involved in virus adsorption receptors? Virus genes, 1990, 3: 277-284.

23. Baer GM et al. Rabies susceptibility and acetylcholine receptor. Lancet, 1990, 335: 664465.

24. Wunner WH et al. The molecular biology of rabies virus. Reviews of infectious diseases, 1988, 10: 771-784.

25. Franck RIB et al. In: Classification and nomenclature of viruses. Fifth report of the International Committee on Taxonomy of Viruses. Vienna, Springer-Verlag, 1991 : 239-262.

26. Banerjee AK. Transcription and replication of rhabdoviruses. Microbiological reviews, 1987, 51: 68-87.

27. Emerson SU. Transcription of vesicular stomatitis virus. In: Wagner RR, ed. The rhabdoviruses. New York, Plenum Press, 1987: 245-269.

28. Wertz GW, Davis NL, Patton J. The role of proteins in vesicular stomatitis virus RNA replication. In: Wagner RR, ed. The rhabdoviruses. New York, Plenum Press, 1987: 271 -296.

29. Wagner R R . Rhabdoviridae and their replication, In: Fields BN, Knipe DM, eds. Virology New York, Raven Press, 1990: 867-881

30. Tordo N , Poch 0. Strong and weak transcription signals within the rabies genome. Virus research, 1988, Suppl. 2 : 30.

31. Sacrarnento D et al. Molecular epidemiology of rabies virus in France: comparison with vaccinal strains. Journal of general virology, 1992, 73: 1 149-1 158.

32. Celis E, Rupprecht CE, Plotkin SA. New and improved vaccines against rabies.


In. Woodrow GC, Levine MM, eds. Newgeneration vaccines. New York, Marcel Dekker. 1990: 419-439.

33. Dietzschold B et al Induction of protective immun~ty against rabies by immunization with rabies virus ribonucleoprotein. Proceedings of the Nat~onal Academy of Sciences of the United States of America, 1987, 84: 9165-9169.

34. Wiktor TJ et al. Antigenc properties of rabies virus components. Journal of immunology, 1973, 11 0: 269-276.

35 Bunschoten H et al. Characterization of a new virus-neutralization epitope that denotes a sequential determinant on rabies virus glycoprotein. Journal of general virology, 1989, 70: 291-298.

36. Kawano H et al. Protection against rabies in mice by a cytotoxc T cell clone recognizing the glycoprotein of rabies virus. Journalofgeneral virology, 1990, 71: 281-287.

37. Anilions A, Wunner WH, Curtis P. Structure of the glycoprotein gene in rabies virus. Nature, 1981, 294: 275-278.

38. Tordo N et al. Primary struciure of leader RNA and nucleoprotein genes of the rabies genome: segmented homology with VSV. Nucleic acids research, 1986, 14: 2671 -2683.

39. Tordo N et al. Walking along the rabies genome: is the large G-L intergenic region a remnant gene? Proceed~ngs of the National Academyof Scjences of the United States of Amer~ca, 1986, 83: 391 4-391 8.

40 Tordo N et al Cornpietion of the rabies virus genome sequence determination highly conserved doniains along the L (polymerase) proteins of unsegmented negative-strand RNA viruses V~iology, 1988 165 565-576

41. Conzelmann KK et al. Molecular cloning and complete nucleotide sequence of the attenuated rabies virus SAD-B19. Virology. 1990, 175: 485489.

42. Sacrarnento D. Bourhy H. Tordo N. PCR technique as an alternative method for diagnosis and molecular epidemiology of rabies virus. Molecular and cellular probes, 1991 , 6: 229-240.

43. Tordo N, Bourhy H, Sacramento D. PCR technology for Lyssavirus diagnos~s. In: Clewley J, ed. The polymerase chain react~on (PCR) for human v~ra l diagnosis. London, CRC Press, 1995: 125-145.

44. Conzelmann KK et al. An L (polyn1erase)-deficient rabies viros defective interfering particle RNA is replicated and transcribed by heterologous helper virus L proteins. Virologi.: 1991. 184: 655-663

45, Tordo N et al. Structure and expression in baculov~riis of the Mokola virus

glycoprotein: an efficient recombriant vaccine Virology, 1993, 194: 59-69.


46. Tordo N et al. Evolution of negatrve-stranded RNA genomes. Sern~nars ~n virology, 1992. 3 . 341-357.

47. Grinnell BW, Wagner RR. r ihb i t ion of DNA-dependent transcription by the leader RNA of vesicular stoniatitis virus: role of specific nucleotide sequences and cell protein b~nding. Moiecular and celliiiar biology. 1985, 5: 2502-2513.

48. Tuffereau C, Martinet-Edelist C. Effets de l'infection rabique sur le mbiabolisme de la cellule-h6te: y-a-t-il inhibition de la synthese des ARN ceilulaires? [Effects

of rabies infection on host-cell metabolism: is there inhibition of cellular RNA synthesis?] C'ompfes rendus de l'Acad6mie des Sc~ences, Series Ill, Sciences de ia Vie 1985. 300: 597400.

49. Dietzschold B et al. Localization and immunological characterization of antigenic dorna~ns of the rabies virus rnternal N and NS proteins. Virus research, 1987, 8: 103-1 25.

50. Dietzschold B et al. Antigenic diversity of the glycoprotein and nucleocapsid proteins of rabies and rabies-related viruses: implications for epidemiology and control of rabies. Reviews of infechous diseases. 1988, 10. 785-798.

51. Smith J. Rabies virus epitopic variation: use in ecologic studies. Advances in viriis research, 1989, 36: 21 5-253.

52. Montano Hrose JA, Bourhy H. Lafon M. A reduced panel of ant-nucleocapsid monoclonal antibodies for bat rabies virus identification in Europe. Research in virology. 1990, 141. 571-581

53. Ertl HC et al. Induction of rabies virus-specific T-helper cells by synthetic peptides that carry dominant T-helper cell eptopes of the viral rbonucleo- protein. Journai of viroiogy, 1989. 63: 2885-2892.

54. Dietzschold B et al. Structural and immunological characterization of the linear virus neutralizing epitope of the rabies virus glycoprotein and its possrble use in a synthetrc vacc,rne. Joiirnal of virology 1990, 64: 3804-3809.

55. Tuffereau C, Fischer S. Flamand A. Phosphorylation of the N and M1 proteins of rabies virus. Journai of yeneral v~roiogy, 1985 66: 2285-2289.

56 Larson JK et al. Identification of an immunodominarit epitope within the phosphoprotein of rabies virus that is recognized by both class l- and class 1 1 - restricted T cells. Jour i~a i of virology 1991, 65. 5673-5679

57 Poch 0, Tordo N. Keirli G. Sequence of the 3306 3'-nucler~lides of tllc genome of the AvOl strain rabies virus: structural siniilar!ies of the protein regions involveci in transcriplion. Bioch~n~ie. 1985, 70: 1019-1029.

58. Hiramatski K et al. blappir~g of the anrlgenic d?terrci:ian;s recognzcd by monocional antibodies against the M2 protein of I-abies virus Virology, 1992. 1 e 7: 472-479.


59. Gaudin Y et al. Fatty acyiation of rabies virus proteins. Virology, 1991, 184: 441-444.

60. Benmansour A et al. Antigencity of rabies virus glycoprotein. Journal of virology, 1991, 65: 41 98-4203.

61. Rose TK et al. Homology between the glycoproteins of vesicular stomatitis virus and rabies virus. Journal of virology, 1982, 43: 361-364.

62. Lafon M , Wiktor TJ, MacFarlan RI. Antigenic sites of the CVS rabies virus glycoprotein: aiialysis with monoclonal antibodies. Journal of general virology, 1983, 64. 843-851

63, Lafon M, ldeler J, Wunner W. Investigation of antigenic structure of rabies virus glycoprotein by monoclonal antibodies. Developments ~n biological standardization, 1984, 57: 21 9-225,

64. Prehaud C et al. Antigenic site II of the rabies virus glycoprotein: structure and role in viral virulence. Journal of virology. 1988, 62: 1-7.

65. Self l et al. Rabies virulence: effect on pathogenicity and sequence characterization of rabies virus mutations affecting antigenic site Ill of glycoprotein. Journal of virology, 1985, 53: 926-934.

66. MacFarlan R et al. T cell responses to cleaved rabies virus glycoprotein and to synthetic peptides. Journal of immunology, 1984, 133: 2748-2752.

67. MacFarlan RI, Dietzschold B, Koprowski H. Stimulation of cytotoxic T- lymphocyte responses by rabies virus glycoprotein and identification of an immunodomnant domain. Molecular immunology, 1986, 23. 733-741

68. Tuffereau C et al. Arginine or lysine in position 333 of ERA and CVS glycoprotein is necessary for rabies v~rulence in adult mice. Virology, 1989, 172: 206-21 2.

69. Lafay F et al. Spread of the CVS strain of rabies virus and of the avirulent mutant AvOl along the olfactory pathways of the mouse after intra-nasal inoculation. Virology, 1991, 183: 320-330.

70. Wunner WH, Dietzschold B. Rabies virus infection. genetic mutations and the impact on viral pathogenesis and immun~ty. In: Cruse JM, Lewis RE, eds. Antigenic variation: molecular and genetic mechanisms of relapsing disease. Basel, Karger. 1987. 103-1 24.

71. Morimoto K , Ohkubo A, Kawai A. Structure and transcription of the glycoprotein gene of attenuated HEP-Flury strain of rabies virus, Virology, 1989, 173: 465-477.

72. Bourhy H , Sureau P, Tordo N. From rabies to rabies-related viruses. Veterinary microbiblogy, 1990, 23: 1 15-1 28.


73 Poch 0 et al. Sequence comparison of five polymerases (L proteins] of unsegmenled riegative-strand RNA viruses: theoretical assignments of functional domains. Journal of general virology, 1990, 71: 1153-1 162.

74. Poch 0 et al. ldentificat~on of four conserved motifs among the RNA- dependent polymerase encoding elements. EMBOlournai, 1989,8: 3867-3874.

75. Delarue M et al. An attempt to un fy the structure of polynierases. Prote;n

engineering, 1990, 3: 461 -467.

76. Bourhy H et al. Complete c lonng and molecular organrat ion of a rabies-

related vrus: Mokola virus. Journal of general virology, 1989, 70: 2063-2074.

Routine laboratory procedures

CHAPTER 4

Rapid microscopic examination for Negri bodies and preparation of specimens for biological tests' E. S. Tierke12 & P. Atanasiu3

Rabies virus causes the appearance of specific inclusion bodies, known as Negri bodies, in the cytoplasm of infected nerve cells or cell cultures. These inclusion bodies can be detected histopathologically by microscopy or by the fluorescent antibody (FA) test (see Chapter 7). Thls chapter describes how to prepare samples for these tests.

The techniques employed in the laboratory diagnosis of rabies should embrace optimum conditons of accuracy, speed and economy. The method employing microscopic examination for Negri bodies, uslng the simple application of brain tissue to a slide and Sellers' technique for staining (see Annex), fulfils these requirements.

It has been found that Negri bodies, when present, are most readily demonstrated in Ammon's horn (hippocampus major) of the brain and also in the pyramidal cells of the cerebral cortex and Purkinje's cells of the cerebellum; they are found to a much more limited extent in the neurons of the thalamus, pons, medulla, spinal cord and sensory ganglia.

Dissection of the brain

Avery simple operation is required to expose Ammon's horn, which IS generally the best area for demonstration of Negri bodies in most rabid animals. With a pair of sterile scissors, a longitudinal incision is made into the dorsal surface of each cerebral hemisphere, about 2 cm lateral to the longitudinal fissure or midline of the brain (see Fig. 4.1). The incision is made from the region of the occiptal pole of the hemisphere and is extended forward for 3-5 cm and downward through the grey matter, and then completely through the white matter until a narrow space, the lateral ventricle, is reached. The opening is then widened by spreading the incised hemisphere, and Ammon's horn will be revealed as a semi-cylindrical, white, glistening body bulging laterally from the ventricle floor (see Figs. 4.2 and 4.3). It has a spiral contour and, on cross-secton, a characteristically rolled surface.

Preparation of slldes

Slides should be made first from Ammon's horn, then from the cerebral cortex, and f~nally from the cerebellum. Tissue samples (at least six) should be taken from

'Based on the chapter In the prevlous edition, whlch was prepared by the late E. S Tierkel and updated by the late P. Atanas~u. Deceased. Former Assistarlt Surgeon-General, US Public Health Service, Department of Health, Educat~on and Welfare, Washington, DC, USA. Deceased Former Honorary Professor, Pasteur lnstltute Paris, France.


Fig. 4.1 Site of incision Tor lowfjng Ammon 'S horn

By courtesy of d. Barrat Laborafo!re d'Etudes sur /a Rage et /a Pathologie des An~rnaux sauvages. Centre Naf!ufidl d Efudss vkler;na,re.~ st al~n~efltaires (CNEVA) Malzev~/le, France.

Fig. 4.2 incision to the lateral ventricle, showing Ammon's horn

By courtesy of J Barrat Laborato~re d'Etildes scir /a Rage el /a Pathologie des An~maux saiivages, CNEVA Malze~,ille, Fra~icc.

56

EXAMINATION FOR NEGRl BODIES

Fig. 4.3 Close-up of Ammon's horn bulging from the floor of the lateral ventricle

By coilrlesy of J Bariat. Laborafo~re d'Etudes sur /a Rage et /a Patholog~e des An~maux sauvages, CNEVA Malzev~lie. France.

these three areas on each side of the brain and examined microscopically before the brain is reported to be negative for rabies. It is always wise to select another area from each hippocampus for good measure.

The following three methods of applying fresh brain tissue to slides are recommended.

Impression method

With a pair of scissors small transverse sections (2-3 mm in thickness) of brain tissue (Ammon s horn, cerebrum or cerebellum) are cut and placed on clean blotting-paper or a wooden tongue-depressor cut surface facrng upward (Fig 4 4) A clean microscope slide is then touched against the cut surface of the section and pressed gently downwards with lust enough pressure exerted to create a slight spread of the exposed surface of the tissue against the slide According to the sire of the section 3-4 impressions can be made on one slide (Fig 4 5) While sti//rnoist the slide is flooded with Sellers stain for a few seconds rinsed under the tap and dried at room temperature without blotting The preparation is then ready for examination The impression may be examined directly under oil or covered with a coverslip mounted in balsam This method is preferred to the others described here because it maximizes the amount of nerve tissue that can be examined and minimizes the amount of cellular damage


Fig. 4.4 Transferring a specimen (from Ammon's horn and the rachi- dian bulb) to a wooden spatula (tongue-depressor) before making impressions

By co~~rtesy of J Barrat, Laboratoire d'Etodes sur /a Rage et /a Pathotogle des Anlinaux sauvages, CNEVA, Malzev~lle France

Fig. 4.5 #mpression method of slide preparation

By courtesy of J Bdrrat, Laboratoire d'Etudes sur /a Rage et /a Pathologie des An~maux sauvagrs CNEVA Malzev~ile france

58


Smear method

The smear method consists of placing a very small section of brain tissue on one end of the slide. Another slide is used to crush the section of tissue against the first slide and is then drawn across the length of the slide. The result is a fairly homogeneous thin film of tissue covering about three-quarters of the area of the slide. In this technique the tissue is spread over a rather extens~ve area. Care should be taken not to use too large a tissue-section, as this will result in an excessively thick film, making proper staining and microscopic examination

impossible. The iinpression method, however, gives superior results.

"Rolling" method

The last method, the "rolling" technique, consists of cutting a piece of brain tissue about 5 mm in diameter, and rolling or teasing it gently (cut surfaces downward) over the entire surface of the slide w ~ t h a toothpick or wooden applicator.

The staining procedure of Sellers is recommended here because of its accuracy and s~mplicity. In this technique, no preliminary fixation is required, since the tissue film is fixed and stained simultaneously, making it one of the most rapid and easily handled methods

The Negri body: differential diagnosis

Although generally rounded in form, the Negri body may be found to assume any shape. At various times in different laboratories it has been demonstrated to be round, oval, spheroid, amoeboid. elongate, triangular, etc. By the same token, there is great variation in size: generally it is found to be between 0.24 and 27.0 pm, It is characteristically acidophilic in staining reaction, and takes on the pink to purplish pink/magenta colour in differential stains that use basic fuchsin or eosin with methylene blue as their base (see Plate 5.1,C, page 70).

The Negri body is found within the cytoplasm of the neuron, typically between the nucleus and one corner of the neuron. or in the prolongation of the cell body. However, it should be stressed that the intracytoplasmic position of the Negri body can be expected with reasonable consistency only in histological sections of the brain. In the simple tissue-application techniques described above, the histological pattern is disturbed and one may very often see well-formed Negri bodies that appear to be entirely outside the neuron. Thus, in methods such as the impression, smear or rolling techniques, the intracellular position of the Negri body is not required as a diagnostic criterion, and Negri bodies that satisfy the requisites of morphological identification, whether inside or outside the neuron, are sufficient to establish a positive diagnosis.

The most characteristic feature of the Negri body is its internal structure. It is this feature that serves as the essential criterion for positive identification in the techniques described in this chapler. The matrix of the Negri body has an acidophilic staining reaction, and contained within this magenta-red structure are small inner bodies (lnnerkorperchen), basophilic granules that stain dark-blue to black. The size of these inner granules generally varies from 0.2 pm to 0.5 pm. Classically, the well-formed Negri body-the so-called textbook picture-will have its inner granules arranged in rosette fashion, with one large centrally placed body


and a series of smaller granules arranged neatly around the periphery of the Negri body. It should be pointed out, however, that this picture is the exception rather than the rule, and it is very rare that such an orderly arrangement oi the inner granules is seen. For diagnostic purposes. it is sufficient to establish the presence of these dark-blue/black staining granules, regardless of their numbers or pattern of distribution within the matrix of the Negri body (see Plate 5.1,C, page 70).

There is universal agreement that the Negri body is specific for rabies and its presence always indicates this infection, Furthermore, a fully formed Negri body cannot be confused with other inclusion-like bodies. However, other types of inclusion body are sometimes found in animal brains submitted for diagnosis and, because of certain siniilarities, may be mistaken for Negri bodies. For example, the acidophilic incliisiori bodies of canine distemper or Rubarth's disease (canine infectious hepatitis, fox encephalitis) are occasionally encountered in the brains of dogs and foxes. These seem to occur inore often In the thalamus and lentiform nuclei than in the hippocampiis. By the same token, the brains of non-rabid cats and laboratory white mice occasionally contain nonspecific acidophilic inclusion bodies when submitted for rabies diagnosis. All these non-rabies inclusion bodes have the same staining ctiaracteristics with Sellers' stain (see Annex). and they cannot be differentiated frorn each other with the techniques described above. However, these non-rabies inclusioli bodies, as a group, can be differentiated from Negri bodies with the use of Sellers' stain. The following characteristics may be used as a guide in this differentiation,

Negri bodies Non-rabies inclusion bodies

Presence of basopli~lic inner granules Heteroger~eous matrix Less refractive Magenta (heliotrope) tinge

Absence of internal structurea Homogeneous rnatr~x More refractive Colour more acidophiic (pinker)

"See Plate 5.1,D. page 70

Small atypical intracytoplasmic inclusion bodies are sometimes found in animals killed during the early stages of rabies. For that reason, it is imperative to hold dogs suspected as rabid in quarantine, rather than to kill them immediately and send the brain to a laboratory for diagnosis (see Appendix 1) . There are two reasons for this. First, it will permit observat~on for symptoms of rabies, which may make possible a clinical diagnosis of the disease. Secondly, the longer the animal is allowed to live, the better the chance of obtaining a positive microscopic diagnosis. The length of clinical illness in rabies is directly related to the size, abundance and development of Negri bodies. Thus, i f the disease runs its full course, Negri bodies are likely to be more abundant and fully formed with good internal structure.

The mouse inoculation test

Since Negri bodes cannot always be found in the brains of animals dying of rabies, it is important that negative specimens be tested by animal inoculation.

EXAMINATION FOR NEGRI BODIES

Extensive surveys of large numbers of rabies cases have shown that 10-15% of cases proved positive by n-iouse inoculation had been missed by direct-smear microscopic examination for Negri bodies It is therefore strongly recommended that laboratories that provide rabies diagnostic services be equipped to carry out animal inoculation tests on Negri-negative b r a ~ n t~ssues as well as tests such as the FA test

In the past the guinea-pig aiid rabbit were considered the most suitable test animals for this purpose Sirice the intracerebral injection of rabiesvrus into white mlce lias been shown to produce typical and constant ~nfection, the white mouse is now considered the test animal of preference The chief advantages of the mouse are the low cost making it possible to use several animals for one specimen the

relatively short inciibation period for street virus and the consistency of production of Negri bodies in the bra ns of mice ~noculated ntracerebraily with street virus Suckling inice (less than 3 days old) are more susceptible to rabies than adult or weanlng mice and should be used wherever possible (see Chapter 6)

A positive microscopic diagnosis is sufficient proof of the presence of rabies When the microscopic examination proves Negri negative or questionable however samples should be immediately taken froin the cerebral cortex cerebellum and Ammon s horn on each side of the brain and from the brain stem (medulla) These should be pooled in the emulsifier in preparation for the mouse inoculation test described in Chapter 6 It is important to achieve completely representative sampling of all those parts mentioned on each side for pooling since there is often great variability in the virus distribution through the brain

Antimicrobial agents for contaminated and decomposed specimens

It is often difficult to ob ta~n from the field animal brains that are bacteriologically sterile. The head may have been in transit for a long time or the animal may have been picked up long after death. shot. or clubbed on the head. Also, the cause of death may have been a bacterial encephalitis.

lntracerebral injection of bacteria may cause the death of rnoculated mice in 1-3 or more days. before any rabies virus that may be present in the inoculum has had its full incubation period. On the other hand, inoculated mice may live long enough for rabies incubation to be complete, may pass through the typical rabies symptoms of tremors, paralysis and prostration, followed by death, and may show typical Negri bodies and many bacteria in their brain smears.

When bacterial contamiriation is suspected-for example, if the animal brain is decomposed or when many bacteria are demonstrated on the original brain smears-it is best to treat the brain suspension with an antimicrobial before inoculating it into mice. Of the following agents, penicillin and streptomycin will give the best results, If these antibiotics are not available, any of the other agents may be used.

Pen~cillin and streptomycin. Add 500 IU of benzylpenicillin sodium and 2 mg (1560 IU) of streptomycin per ml of tissue suspension. Allow to stand for 30 minutes at room temperati~re before ~njection. This amount is usually enough, but for very heavily contaminated brains or salivary glands, as many as 1000 IU of penicillin aiid 3 mg (2340 IU) of streptomycin may be used.


Giycerol. Place the brain specimen In pure glycerol for 48 hours. Phenol 0.5%. Make up 0.5O/' phenol in physiolog~cal salt solution, This is used as the diluent for making up the 10% tissue inoculum. Hold for 6 hours or overnight. If held for 6 hours, keep the suspension at room temperature. If held for over 6 hours, keep the suspension overnight in the refrigerator. Thiornersal' 1:5000. Make up a 1 :5000 solution of thiomersal in physiological salt solution. This is used as the diluent for making up the 10% tissue inoculum. Hold for 6 hours or overnight. If held for 6 hours, keep the suspension at room temperature. If held for over 6 hours, keep the suspension overnight in the refrigerator.

In order to determine whether contaminating bacteria are present, a portion of all tissue emulsions should be cultured in dextrose iniusion broth or similar media, and streaked on a blood-agar plate. The recommended amount is about 0.1 m1 of emulsion in 3 ml of broth. The plate should be incubated overnight at 37.5"C.

Early deaths (1-3 days) among inoculated mice may be attributed to the presence of contaminating bacteria i f the cultures show moderate to heavy growth and if many bacteria are found in the brain smears of the dead mice.

Annex Preparation of Sellers' stain2

Examination of slide

In order to save time, the stained slide should rnitially be studied under low power and areas containing numerous large neurons selected for examination under oil immersion (Figs. 4.6 and 4.7).

Sellers' stain shows the Negri body well differentiated in magenta (heliotrope) to bright red, with well-defined dark-blue to black basophlic inner bodies. All parts of the nerve cell stain blue, and the interstitral tissue stains pink. Erythrocytes stain copper red and can be easily differentiated from the magenta-tinged red of the Negri bodies.

Stock solution 1. Methylene blue

Methanol (absolute acetone-free) 2 . Basic fuchsin

Methanol (absolute acetone-free)

10 g to make 1000 ml

5 9 to make 500 rnl

The stock s o l i ~ l ~ o n s are stored in bottles with screw tops or ground glass stoppers. Certified b~ological stains are preferable. The dry dyes should preferably contain no less than 85% methylene blue, and no less than 92% basic fuchsin. While absolute acetone-free methanol is recommended, chemically pure (C.P.) methanol meeting American Chemical Society specifications may be substituted i f

desired.

' Alsa known as th!merosai and rnercurothioiate

'This annex was klndly contrbuted by the late T. F. Sellers, Former Director Ernerrtus, Georgia Department of Public Health Atlanta, GA USA.


Fig. 4.6 Lo W-po wer vie W of impression, showing field (upper halo rich in neurons for examination under high power

X 200 By courtesy of United States Department of Health, Educat~on and Welfare, Public Health Service, Centers for Dtsease Control and Prevention (US DHEW-PUS-CDC), Atlanta. GA, USA

Staining solution Methylene blue (stock solution No. 1 ) Basic fuchs~n (stock solution NO. 2)

2 parts 1 part

M I X thoroughly but do not filter. Store in a container with a screw cap or ground glass stopper. The stainng soluton should be left to stand for 24 hours before use, and may be kept indefinitely if protected from evaporation.

Adjustment of stair)

When the stock solutions have been accurately prepared, the above proportions will usually produce a stain that will glve the desired colour differentiation. However, t is advisable to make a trial stain, and i f the results obtained do not equal those illustrated in Plate 5.1 ,C (page 7 0 ) , the stain may readily be adjusted. If the stroma IS a bright red rather than rose-pink in the thinner areas, and the overall staining effect is reddish, the fuchsin is too dominant. Add methylene blue stock


Fig. 4.7 High-power vie W of impression, showing Negri bodies

indicated by arrow X 900 By courtesy of US DHEW-PHS-COG Atlanta. GA. USA

solution in measured aniounts, checking with a trial stain after each addition until the desired colour balance is obtained. When the methylene blue is too dominant, the Negri bodies are a deep muddy maroon colour and the nerve cells stain too deeply. Adjustment may be made with the fuchsin stock solution in this case.

I f the stock solutions are protected from evaporation, more stain may be subsequently prepared using the stock solutions in the adjusted proportions.

Staining procedure 1. Prepare smears or irnpress~ons in the usual manner (see page57): no fixation is

required. 2, Immediately. while the preparation is still moist, immerse it in the staining

solution for 1 5 seconds, depending on the thckness of the smear. 3. Rinse quickly in running tap-water, and air-dry without blotting.

In some regions tap-water is not satisfactory for rinsing purposes The suitability of the water may be determined by comparing preparations rinsed with tap-water and oihers rinsed with distilled water containing 0.66 mol!l phosphate buffer, pH 7.0.

When not in use, the stain must be kept in a tightly closed container to prevent evaporation, which tends to make the fuchsin too dominant. The addition of


absolute methanol will restore the proper balance, It is convenient to keep the staining solution in a screw-capped Coplin jar for dally use. i f this is not available, the stain may be stored in a dropper bottle with a ground glass stopper and the smear flooded with the stain. Staining in thls manner will not be satisfactory unless the entire process can be completed within a few seconds.

The best results with the stain are obtained when the brain tissue is fresh. If the tissue is decomposed, the characteristic colour d~fferentiation is affected, and although the Negri bodies retain their staining quality, the smear as a whole becomes too red, or at times too blue, and identification of the bodies becomes more difficult.

CHAPTER 5

Histopathological diagnosis' P. Lepine2 & P. ~ t a n a s i u ~

In principle, the histopathological diagnosis of rabies consists in recognition of the presence in the animal of acute encephalomyelitis which can be ascribed to a specific agent, namely the rabies virus.

If the animal has died from rabies, it is normally easy to detect the specific lesions; if the animal has been killed, death may have occurred before the appearance of specific lesions (Negri bodies). Consequently, any animal in which the cerebrospinal axls (brain, medulla or ganglia) shows the slightest sign of lesions. particularly infiltrations, should be regarded as suspect, no matter how small the lesions may be.

After correct removal, the brain should be carefully dissected. Smears or impressions should first be examined for Negri bodies (see Chapter 4) or by the FA test or electron microscopy (see Chapter 18). However, while a positive result is indicative of infection, a negative result does not rule out the possibility of infection. A regular histopathological examination should be made of sections stained after embedding by a rapid method. At least six samples should be examined, corresponding respectively to Ammon's horn (both sides), the cerebral cortex (motor area), the cerebellum, the medulla, and a ganglion (gasserian or upper cervical).

The sections should be examined for:

Signs of meningoencephalomyelitis, i.e. meningitis, meningeal infiltration, peri- vascular cuffing, parenchymatous infiltration, formation of encephalitic nod- ules (Babes' tubercles), and gangl~on infiltration with satellitosis and neuronophagia (lesions of van Gehuchten and Nelis) (Figs. 5.1 and 5.2). These lesions may be detected by any staining method (e.g. haematoxylin-eosin, polychrome methylene blue). They siiow the existence of encephalomyelitis and enable a tentative diagnosis of rabies to be reached. Specific lesions. The different types of neurons shoula be examined for Negri bodies and lesions of fixed rabies virus.

The Negri bodies are f o u n d especially in t h e central pyramidal layer of Ammon's horn (Fig. 5.3) and the hippocampus, in the lower loop and the middle layer of the ganglioneurons of Ammon's horn and, less frequently, in the neurons of the cerebellum, the motor area of the cerebral cortex, and the medullary nuclei. They may be present in very large numbers in the ganglia but are generally small in size.

' Based on the chapter In the previous cd~t lon wh~ch was prepared by the iate P Lepine and updated by the iate P Atandslu

Deceased Former Director, V~rus Department Pasteur Instilute, Paris, France Deceased Former Honorary Professor, Pasteur lnst~fute Paris France

HISTOPATHOLOGICAL DIAGNOSIS

Fig. 5.1 Normal gasserian ganglion

Gangiioneurons witti a singie layer of sateilife cells

Fig, 5.2 infiltration of a gasserian ganglion with satellitosis and neuronophagia (lesions of van Gehuchten & Nelis)

The lesons of fixed rabies virus are found exclusively in the middle zone of the external layer of the cells of Ammon's horn. They always coexist in varying proportions with the lesions of street rabies virus.

These lesions can be detected only by special staining methods (Mann, Sellers, etc., see Plate 5 1) . Their presence enables a definite diagnosis of rabies to be made.


Fig. 5.3 Dog: street rabies virus

(Mann stain. ~20001 Negri bodies i r? neurvr?s of the central pyramidal layer o iAmn~or i s horn (The brain speclm was partially decomposed)

Removal of the brain and preparation of tissue samples for examination1

Large animals (dogs, foxes, cattle)

1 . Secure the animal firmly to the autopsy table or, better, decapitate the animal and hold the head tightly with forceps (Fig. 5.4). Thick rubber gloves and goggles should be worn to protect the hands and eyes.

2. Make an incision on the ni~dline of the skull through the skin, push aside the flaps of skin, and reflect the muscles and fascia as far as the base of the skull, proceeding from the ciown to a horizontal line passing through the eyes (Figs. 5.5 and 5.6).

3. Usirig a harnrner and bone chisel, open the skull at eye level and cut along the temporal bone (Figs. 5.7 and 5.8).

In the case of very large animals (such as large dogs, cows, etc.) a different method is preferable: make a longitud~nal saw-cut on each side of the midline at about 1.5 cm from i t and join up these cuts by one or two transverse saw-cuts above the orbits and at the occiput so that the calvaria can be removed in two symmetrical pieces.

4. Once the calvaria has been removed, use fresh instruments to open the meninges, including serrated dissection forceps and a pair of fine, sterile

' See also Appei3dx 1


Fig. 5.4 immobilization of a fox's head wifh forceps

By courtesy of J. Barrat, Laboratoire d'Etudes sur /a Rage et /a Pathologie des Animaw sauvages. CNEVA, tvlaizdv~/le, France.

scissors. Make an incision in the meninges, starting from the median region, along and on each side of the longitudnal snus.

5. Make a second incision perpendicular to the first and push the rneningeal flaps upward and backwards.

6. Using fresh instruments, cut through the medulla w ~ t h a scalpel as low as possible and lift up the brain, proceeding from back to front and successively severng the pairs of cranial nerves.

7. At the end of the operation, roll the brain gently forwards into a large sterile Petri dish so that it rests on its upper surface (Fig. 5.9). Note: In hot weather, or f the brain is soft (cadaver~c brain), i t should be placed in a refrigerator set at 5°C for 2 hours before dissection.

Examination of the brain Note whether or not there is congestion of the cerebral vessels or exudate in the meninges, etc. Dissect the brain as follows:

1 . Using a brain knife, separate the two hemispheres Iongtudinally, after having detached the cerebellum and the medulla.

2. Look for the hippocampus or Amrnon's horn. This may be done in two ways (see also Chapter 4):

(a) Cut across the brain transversely, starting from the base behind the optic chiasma and proceeding towards the lower third of the cerebral hemisphere. The third ventricle appears on the cut surface; Ammon's horn is seen as a whitish fold resembling a large bean cut transversely, and can easily be removed.


(b) Alternatively, a longitudinal incision may be made externally in the posterior third of each cerebral hemisphere about 1.5 cm from the midline. The incision is continued through the grey matter and the white matter until a narrow groove, the th~rd ventricle, is reached. Ammon's horn will be seen on the floor of the ventricle in the form of a glistening white, semi-cylindrical bulge, extending laterally on each side (see Fig. 4.5, page 58).

3. Cut transverse sections 1-2 mm in thickness from each Ammon's horn. Take similar samples from the cerebral cortex (motor area), the cerebellum, and the

medulla.

When using the impression method, at least six slides (two for each Amrnon's

horn, one for the cerebral cortex, and one for the cerebellum) should be carefully examined for Negri bodies before deciding that the results are negative. If the results are negative, however, histological examinations are carried out.

Preparation o f the tissue samples for histological examination. If the tissue is soft and difficult to section, prepare pieces of filter paper slightly larger than the tissue sample to be collected. Apply the piece of filter paper to the cut brain surface, grasp the edge of the filter paper with fine forceps held in the left hand, and with the right hand make a cut with a scalpel parallel to the filter paper and 2-3 mm from it, so as to remove the piece of brain. Immediately submerge the tissue sample and the fragment of filter paper to which it is adhering in the fixing agent.

Removal o f material for inoculation

During the operation, care should be taken to reserve material from the same areas (cortex, Ammon's horn, cerebellum and medulla) for use in animal inocula- tlon (see Chapter 6). If the brain is received in good condition and can be assumed to be sterile, the samples are removed before any examination is made. When the brain is infected, the samples may be removed at any time and anti- microbia l~ added, as described on page 61.

Plate 5.1

A Street rabies virus in dog brain. Many typical Negri bodies are situated in the cytoplasm and in the prolongations of the neurons, some of which contain haloes and "lnnerkorperchen " (Mann stain: X 1550).

B. Fixed rabies virus in tissue culture. lntracytoplasmic inclusions of different shapes and s~zes are observed (Mann stain: X 1550).

C. Street rabies virus in bovine brain. An inclusion in the cytoplasm of a degenerated neuron (to the right of the nucleus) is a typical Negri body with basophilic "lnnerkorperchen" (Sellers' stain: X 1550).

D. Cat brain section with nonspecific inclusions in the cytoplasm. No "lnnerkorperchen" (Mann stain: X 1550).

E. Street rabies virus in monkey ganglia. Infiltration and neuronophagia of a gasserian ganglion: chromatolysis and Negri bodtes (haematoxylin-eosin: X 660).

F Fixed rabies vtrus in tissue culture. Cytoplasmic inclusion (immunoperoxidase: X 1550).

Photographs kindly supplied by the late P. Atanasiu & J. Sisman, Pasteur Institute, Paris, France.


Fig. 5.5 Reflection of the skin, exposing the temporal muscks

By co~irtesy of J. Barrat, Laboratoire d'Etudes sur /a Rage et /a Pathologie des Animaux sauvages, CNEVA, Ma/zev~/le, France

Fig. 5.6 Dissection of the temporal muscles of the cranium

By courtesy of J, Barrat, Laboratoire d'Etudes sur /a Rage et /a Patliologie des Animaux sauvages. CNEVA, Ma/zBvi'/le, France.


Fig. 5.7 Opening the cranium

By courtesy of J. Barrat, Laboratoire d'Etudes sur /a Rage et /a Pathologie des Animaux sauvages, CNEVA, Ma/zevii/e. France

Fig. 5.8 Removal of the calvaria, exposing the brain

By courtesy of J Barrat, Laboratoire d'Etudes sur /a Rage et /a Pafholog~e des Animaux sauvages, CNEVA, Malzev~ile France

73


Fig. 5.9 Brain removed from the skull

After the b ran has been removed, it should he placed on a plate labelled with the sample no., species and sex of the animal. The scalpei and forceps should be chanyed after each operation.

By courtesy of J Barrat, Laborato~re d'Etudes sur /a Rage et /a Pathologie des Animaux sauvages, CNEVA, Malz~viile, France.

Small animals (rabbits, guinea-pigs, hamsters, mice)

Rabbit 1. Place the animal on its stomach and secure it to the autopsy tray, with the

head at the edge of the tray. 2. Using a serrated dissection forceps, scalpel and scissors, completely scalp the

head from the nape to the muzzle, removing the ears and the upper eyelids. 3. Moisten the exposed surface of the head with 70% ethanol and rapidly flame it

with a gas burner. 4. Holding the muzzle of the animal with a Farabeuf forceps in the left hand, open

the brain pan with three cuts of the bone forceps. Make the first two in the front part of the head, from each orbit to the rnidline, opening the brain pan (parietal

and temporal bones) in two flaps, to the right and to the left. The third cut is made at the occiput, lifting the bone flap backwards, which completes the clearance of the field of operation. The Farabeuf and bone forceps are then laid aside.

5. Using the fine forceps and scissors, free the meninges and section the anterior part of the brain at the olfactory lobe: after severing the medulla behind the cerebellum, raise the brain in order to cut the optic chiasma and then place it in a sterile Petri dish.


6. With the brain lying on its dorsal surface in the Petri dish, cut the cerebral trunk at the peduncle; then make a transverse cut through the brain substance, starting from the optic chiasma and going towards the convexity, parallel to the posterior surface of the hemispheres and the cut surface of the cerebral trunk. A second cut made in the same way, parallel to the first and 5-8 mm behind it, gives a transversesection of the brain that includes the hippocampal gyrus, Arnmon's horn and the basal optic ganglion-areas of choice for the detection of Negri bodies-in addition to the cortical motor area (Fig. 5.10).

7. Cut a transverse section of the cerebellum in order to examine Purkinje's cells and the peduncular region.

8. Finally, cut a s l c e from the end of the cerebral trunk to o b t a ~ n a section of the

medulla. 9. To reach the gasserian ganglion, cut through the petrosal bone with the bone

forceps at its insertion into the sella turcica: force apart the cut surfaces by everting the temporal bone downward and outwards. The gasserian ganglion is easily recognized from its whitish nodular appearance and its almost fibrous consistency.

10. Carefully remove the ganglion, place it on a piece of filter paper, and immerse it in the fixing agent, together with the other samples.

Guinea-pig, hamster 1. Secure the animal to the autopsy tray and scalp the head as described above. 2. Wipe the skull with 70% ethanol and flame it rapidly. 3. Using the second pair of large sterile scissors, open the skull by means of four

incisions encircling the skull cavity, the first one joining the two orbits and the other three made successively at the sides and the occipital bone.

4. Remove the brain with the fine forceps and scissors and dissect it as described above for the rabbit. The gasserian ganglion is more difficult to locate than in the rabbit; however, with care, it can be found.

Mouse 1 . Place the mouse, ventral surface downwards, on a sheet of cork. Spread the

limbs out and secure them with pins. Also secure the base of the tail and the anterior extremity of the muzzle, which should be pulled well forward.

2. Remove the skin from the head. 3. Wipe the skull with 70% ethanol and flame it very gently with the pilot flame of

the gas burner. 4. Open the skull with fine scissors. first by joining the orbits, then cutting laterally

through the skull fairly low, and finally pushing back the flap thus obtained. 5. Sever the medulla and the optic chiasma. 6. Remove the whole of the brain and place it in a sterile Petri dish. 7. Remove the cerebellum. 8. Cut through the brain transversely at the optic chiasma and along a plane

parallel to the posterior surface of the brain, so as to obtain a cross-section; this is placed in the fixing agent, the anterior portion being reserved for smears and inoculations. Also fix a section of the cerebellum. If examination of a ganglion is needed, the upper part of the spinal cord should be detached and removed together with the upper cervical ganglion.


Embedding, staining and examination for Negri bodies

Fixation

Bouin-Dubosq-Brazil mixture When rapid fixation is not necessary, the best fixing agent is Bouin- Dubosq-Brazil mixture. This can be used for all examinations not requiring a special fixing agent or immunohistochemical technques and is particularly suitable for the staining of nuclear and cytoplasmic inclusions. The formula is as follows:

Formaldehyde, 40% solulion Ethanol, 96% solution Dist~lled water Glacial acetic acid Picric acid

Prepare the tissue sections and fix for 24 hours in this mixture. When fixation is complete, dehydrate the tissue in 96% ethanol and then in absolute ethanol. Finally, embed the specimens in paraffin, using the following technique.

Embedding techniques

Rapid method for histologicai embedding For slices of brain tissue 1 mm in thickness, fixation is complete in 15-30 minutes if a rapid fixation method is employed. The tissue is transferred directly to absolute ethanol, with which it is treated for 20-30 minutes in two baths, followed by two changes of toluene and two of paraffin, each lasting 15 minutes, Including the time necessary for cutting and staining the sections, the preparation is ready for microscopic examination about 3.5-4 hours after autopsy of the animal.

Staining techniques

Sellers' method Sellers' method may be used for staining sections prepared using standard techniques. Thus, the simplicity of Sellers' method of staining is combined with the

Fig. 5.10 Sections of rabbit and mouse brain

A. Mouse or rabbit brain: ( l ) cerebrum: (2) cerebeilum; (3) meduiia oblongafa. B. Mouse brain: ( I ) cerebral cortex; (2) laterai ventricle (choroid plexus); (3) Ammon's horn;

(4) rnesencephaion; (5) inner nliclei of grey matter; (6) median venti~icle; (7) corpus caliosum.

C. Rabbit brain: ( l ) third ventricle; (2) Ainmon's horn (middie iayer); (3) rhinai fissure; (4) choro~d plexus; (5) ~nfundibuium; (6) hippocainpal gyrus; (7) laterai genicuiate ganglion; (8) ventral nucleus of thaiamus.

D Rabbit brain after removai from craniai cavity (inverted, lateral aspect): left-anterior portion; centre-temporai and parietai iobes; right-cerebellum. Section is performed anterior to a (through opbc chiasma) and posterior to b; the portion a-b is immersed in fixative.

E Rabbit brain, transverse section a-b: ( l ) corpus caliosum, (2) third ventricie; (3) lateral ventricle: (4) choroid piexus; (5) Amrnon's horn (middie layer); (6) Ammon's horn (inner iayer); (7) Ammon's horn (outer iayer); (8) iaterai gen!cu/ate ganglion; (9) ventral nucleus of thalamus; (10) iateral choroid piexus; (1 11 rhinai fissure; (12) infundibulum.

By courtesy of the late P. Atanas~u


greater accuracy afforded by histological examination, so that the Negri bodies can be observed in situ in the cells, and all possibility of error is excluded.

The preparation of Sellers' stain is described on page 62. After the paraffin is removed, the sections are stained by immersion in a mixture of 6 ml of basic fuchsin stock solution, 20 ml of methylene blue stock solution and 50 ml of absolute methanol.

The time required for staining depends on the thickness of the section; it is usually between 2 and 10 minutes. The stained sections are washed in tap-water, dried with filter paper without washing with alcohol, and mounted in balsam.

Results: Negri bodies, deep magenta red; neurons, blue-violet; inner bodies of the Negri bodies, dark blue to black; nucleoli of the neurons, dark blue; erythrocytes, copper red (see Plate 5.1,C and Chapter 4).

Mann's method This classical method gives sections that are permanently stained, with very fine differentiation of the Negri bodies. It is an excellent method for demonstrating the presence of Negri bodies, but it requires time and a certain dexterity for full success.

Prepare the following mixture immediately before use:

Methyl blue (not methylene blue), 1% aqueous solution 18 ml Eosin, 1 % aqueous solution 23 ml Distilled water 49 ml.

Stain the sections for 24 hours at room temperature, or for 6-14 hours at 38°C (in this case first treat the sections with a 1 : 1 mixture of absolute methanol and 10% formaldehyde in distilled water to render the gelatine insoluble and prevent the sections from becoming detached).

Wash with tap-water and then rapidly with absolute ethanol. For differentiation, use the following solution:

Sodium hydroxide in ethanol, 1.5% solution Absolute ethanol

1 rnl 30 ml.

Leave the section in this solution until it is stained pink (about l 0 minutes). As soon as this stage is reached, wash the preparation well with tap-water. The section should take on a sky-blue colour; if not, treat it with water conta~ning acetic acid (2 drops of acetic acid in 40 ml of distilled water) for 1 minute.

Dehydrate rapidly (absolute ethanol), treat with xylene, and mount in balsam.

Results: Negri bodies, vermilion red; nucleoli of the neurons, violet-red; chromatin, blue; cells, dark blue; stroma, pale blue; erythrocytes, p ~ n k (see Plate 5.1, A, B, page 70).

If phloxin B is substituted for eosin in the same proportion, the preparations obtained are less attractive (purplish-blue or mauve background instead of sky- blue), but the inclusions (Negri bodies) are more numerous and more striking.

Fuchsin-safranine-blue method After fixation, the tissue 1s embedded n paraff~n, cut into thin sections, and freed from the paraffin.


1. Stain for 10 minutes with the following mixture:

Solution 1: Basic fuchsin Ethanol, 50% solution

Solution 2: Safranine, 0.2% aqueous solution.

Mix equal parts of the two solutions and store in a dropping bottle; the mixture is fairly stable and keeps for some time.

2. Discard the stain, cover the section with a mixture of ethanol and acetone (equal parts) to remove excess stain, and wash rapidly: the section is coloured red.

3. Stain for 15-60 secondswith Unna's polychrome methylene blue (10% dilution) or with permanganate blue prepared by Stkvenel's method and used undiluted.

4. Discard the stain: the section is deep violet in colour. 5. Differentiate in ethanol-acetone for a few seconds: the section becomes blue. 6. Wash the preparation immediately in running tap-water to remove excess stain,

and again treat with ethanol-acetone. 7. Without washing the preparation, partially dehydrate it by shaking it in a Borrel

tube filled with absolute ethanol. The remaining stain is removed from the section which becomes differentiated, taking on a pink-lilac tint varying in paleness according to the thickness of the section.

8. Rapidly complete dehydration in absolute ethanol, carefully remove the alcohol in several changes of xylene or toluene, and mount in balsam.

Results: Stroma, very pale pink with nerve fibres a deeper pink; neuroglia and leukocytes, purplish-blue; neurons, light blue; chromatin, deep purple with the nucleolus a vivid red; nuclear inclusions and oxyphilic substances, bright pink; Negri bodies, poppy-red to mauve pink, with the internal structure lilac.

CHAPTER 6

The mouse inoculation test H. Koprowski '

The mouse inoculation test, in spite of its simplicity, depends greatly on the accuracy of its performance for dependable results.

Choice of mice

Strain

White mice of any breeding strain may be considered suitable, although preference should be given to the Swiss albino strain since it is very susceptible to rabies virus and it is easy to maintain the breeding stock in the laboratory. If the Swiss albino stock is not available, however, almost any breed of mice, except grey wild mice, can be used, because a genetically resistant strain has not yet been found. Grey wild mice should be excluded because of the difficulty in restraining these animals in cages during the observation period.

Suckling mice (preferably less than 3 days old) are more susceptible to intracerebrally inoculated rabies than weanling or adult mice and should be used wherever possible. The use of suckling mice has permitted indivdual virus iso- l a t i o n ~ that would have been missed if only young adults had been inoculated. If sufficient mice are available, the mouse inoculation test should be performed routinely, in addition to other diagnostic tests. Since confusion may arise from nonspecific deaths, the combined use of the mouse inoculation test and the FAtest is recommended.

Sex

Mice of both sexes are equally susceptible to rabies virus. It is inadvisable to keep older mice of the same sex in one cage since they are apt to kill each other in fights before the observation period is completed. This applies especially to males.

General health

It is imperative that the animals chosen for inoculation be in good health. It is important to know the history of the breeding colony, and it is advisable to inspect the animals closely before inoculation.

' D~rector, Center for Neurovirology Jefferson Cancer lns t~ t i~ te , Thornas Jefferson Un~vers~ty, PA. USA.

80

MOUSE INOCULATION TEST

Any animals found to have ectoparasites, ruffled fur or signs of diarrhoea

should be withdrawn from the test immediately. If the mice are sent to the laboratory from a considerable distance, it is advisable to postpone inoculat~on for at least 3 days in order to let them rest and become adjusted to changed conditions. In such cases, it may be equally important to leave a few animals from the shipment uninoculated in order to observe the death rate among "normal" mice as compared with inoculated animals.

Preparation of suspect material for inoculation

Choice of tissue

Either the brain or the salivary-gland tissue of a suspect animal may be used for virus isolation. Detection of the virus is more frequently possible in the brain than in the salivary glands. However, from the epidemiological and epizootiological points of view, it IS important to examine the salivary glands for the presence of virus.

Although it is relatively immaterial which part of the brain tissue is chosen for the preparation of the suspension, preference is usually given to Ammon's horn, the cerebellum and the cerebral cortex. When salivary-gland tissue is used, the submaxillary glands are those most likely to show the presence of rabies virus. In addition, it is always advisable to mince salivary-gland tissue before grinding.

Grinder

The choice of grinder depends to a certain degree on the amount of tissue available. If more than 3-4g of material are available, a small Waring-type blender should preferably be used. If the amount of tissue is less than 3 g (which is more usually the case), or if a Waring-type blender is not available, the following grinders or grinding devices may be substituted, in the order of preference listed.

(a) TenBroeck grinder: This grinder is easy to assemble, manipulate, clean and sterilize, but only brain tissue may be used in it since salivary-gland tissue is too tough for it to grind properly. A slight disadvantage is its fragility. If it is not used properly, it may break while being manipulated and it may also be damaged during cleaning, sterilizing, etc. This grinder and the Waring-type blender, if properly used, have advantages over the other grinding devices.

(b) Pestle andmortar: This is a time-honoured method of grinding and it has the advantage that, with the help of an abrasive such as sterile sand, even the toughest tissue can be properly ground up. The pestle and mortar cannot be operated under such sterile conditions as can the Waring-type blender or the TenBroeck grinder; however, it can be easily cleaned and sterilized, and stands wear and tear for a long period of time. It is also cheaper than the other grinding devices.

Diluent

The choice of diluent may be left to the user, but an isotonic salt solution is preferable, although mice have been known to withstand easily an intracerebral inoculation of sterile distilled water. The following diluents may be considered in order of their availability:


(a) A physiological salt solution containing varying amounts of animal serum (10-50% concentration): This is by far the commonest diluent used. It should be carefully ascertained, however, that the donor animal has never been vaccinated against rabies. it is therefore advisable to avoid the use of dog, cat or cattle serum. Normal sheep serum seems to possess some "antiviral" properties that are absent from rabbit serum. The rabbit should therefore be considered the preferred source of blood. The serum should be inactivated by heating it for 30 minutes at 56'C before using it as a component of the diluent. Diluent-containing serum should be sterilized by filtration through bacteria- retaining filters.

(b) Other diluents:

- skimmed milk; - 2% bovine serum albumin in buffered salt solution; - physiological salt soiution in distilled water.

These are not particularly recommended, although rabies virus seems to be reiatively non-susceptible to the inactivating action of salt solution, in contrast to such viruses as eastern or western equine encephalomyelitis.

Note: If the tissue suspension is to be frozen and stored, a 50% solution of serum in distilled water should be the preferred diluent.

Bacterial sterility

There is no need to add antimicrobials to suspensions of brain tissueif the material has been handled with reasonable precautions at the autopsy and dispatched in a sterile container. A 50% glycerol solution is highly recommended for preserving the material since, in addition to its preserving qualities, it exerts a strong bacteriostatic action (see Chapter 4. page 62). If bacterial contamination is suspected, however, streptomycin and penicillin should be added to the suspension to give a final concentration of 1560 IU of streptomycin and 500 IU of penicillin per ml. If antimicrobials are added, the suspension should be left to stand for at least 30 minutes before the animals are inoculated.

It is always advisabie to add antimicrobials to suspensions of salivary glands. The salivary-gland tissue suspension should also be cultured for possible bacterial contamination. Beef-infusion broth, thioglycolate medium, and blood-agar are considered to be good culture media for this purpose. If bacterial growth is observed, an attempt should be made to identify the bacieral agent. If the results of the mouse inoculation test are equivocal (see page 86), it may be advisable to

test the pathogenicity for mice of the bacterial contaminant by intracerebral inoculation.

Concentration of infected tissue in the suspension

This is optional. I f the suspension is for storage, prepare a 20% suspension by weight. The weight of the tissue in grams multiplied by 4 gives the required volume of diluent in millil~tres. However, if the suspension is to be used for inoculating mice intracerebrally, a 10% suspension by weight should be prepared. Either dilute the


20% suspension by adding an equal volume of diluent, or prepare the suspension by multiplying the weight of tissue in grams by 9 in order to obtain the required volume of diluent in millilitres.

In attempting to isolate certain strains of rabies street virus, interference phenomena may be suspected (see page 86). In such cases it is advisable to dilute the tissue suspension more than 109/0.

Centrifugation and filtration

I f the equipment is available, centrifuge the tissue suspension for 5 minutes at 150-200 g to remove the gross particles. However, if no centrifuge is available, mice may be inoculated ntracerebrally with a 10% uncentrifuged brain suspension. Salivary-gland suspensions, if uncentrifuged, must be filtered through one or two layers of sterile gauze in order to prevent the animals dying from trauma.

Inoculation of mice

Choice of syringe

Syringes that can measure accurately 0.03 ml (a single mouse dose) should be used. Thus 0.25-ml tuberculin syringes should be considered first, followed by 0.5-m1 or l -ml tuberculin syringes. For intracerebral inoculation, the needles should be 0.40-0.45 mm in diameter (27- or 26-gauge) and 1.0-1.5 cm long. Larger needles cause trauma to the brain substance.

Anaesthesia

The mice should be anaesthetized before they are inoculated, preferably using ether. A battery jar with a specially fitted wire bottom may be used for this purpose. I f no such device is available, pentobarbital sodium injection should be considered. Ideally, the table used for mouse inoculation should be placed well away from the wall, so that an assistant can etherize the mice from a position opposite the operator.

lnoculation technique

There is a wide choice of methods of inoculation. For a right-handed operator the following technique has been found satisfactory.

1. Place the anaesthetized mouse on its left side, with the legs pointing towards you.

2. Support the lower jaw of the animal with the left thumb, and place the left index finger behind the skull of the animal. Very little pressure should be exerted, otherwise the animal may be asphyxiated and die. Hold the syringe in your right hand in a horizontal position parallel to the table surface and perpendicular to the head of the mouse, with the needle pointing towards you. With a quick thrust, push the needle through the skull of the animal at the place that can best be described as the apex of an imaginary angle, the arms of which point to the animal's right eye and right ear. The needle should easily penetrate the bone


and should then be further inserted for about 0.1-0.2 cm into the brain tissue. If a 1.5-cm needle is used, care should be taken not to penetrate too far, otherwise the injection may be given into the base of the skull. Push the plunger to the next 0.03-ml mark and then gently withdraw the needle. Move the inoculated mice away from the syringe hand (i.e. right-handed persons pass the inoculated mice to the left). This is to prevent crossing hands, which may result in catching a finger on the needle held in the opposite hand.

3. Place the inoculated mice immediately in a can or box previously prepared and identified by a tag bearing the mouse-group number or any other particular identification mark. I f several groups of mice are inoculated, it is advisable to check the number of living mice in each group after the inoculations are finished. If any animals are found dead, an equal number of new animals should be inoculated and added.

Under no conditions should the same syringe be used for inoculation of two different suspensions. If an adequate number of sterile syringes is not available, each syringe should be boiled between inoculations, and care should be taken to let it cool before filling it with inoculum.

Strict safety precautions should always be followed whenever the rabies virus is handled (see Chapter 1). For example, the rapid emptying of the syringe used for inoculation into a pan of water will produce an aerosol that can cause infection in the operator, or in the animals with which he or she is working. Virus may be spread from the table used for animal inoculations to the hands, and if these are not washed properly it is possible to contaminate subsequent specimens during grinding of tissue in a mortar - for example, virus may be transferred from the hands or sleeves into the mortar. Mice laid on the table after etherization may awake and have to be put back in the same ether jar, thus contaminating it. Should other studies be carried out using the same ether jar, virus may be deposited on the head of normal mice and be carried into the brain by the inoculation procedure. The table used for the inoculations should therefore be considered contaminated until it has been washed with soap and water and treated with an appropriate disinfectant (see Chapter 1).

When inoculation of all the mice is finished, the syringe and needle may be rinsed in water, provided this is done gently with the point of the needle well below the surface of the water in the sterilization tray. The inoculation of mice and harvesting of mouse tissues is best done on trays that can be sterilized and covered with absorbent paper backed with polyethylene or aluminium foil. Instruments used for individual specimens should be sterilized by autoclaving (121 "C for at least 20 minutes), dry heat (170°C for 2 hours: or by continuous boiling (for 20-30 minutes); sterilization by flaming between taking specimens may not kill the virus.

Observation of inoculated mice

Although rabies virus will only rarely cause signs of illness in mice before the fifth day after intracerebral inoculation, it is advisable to check mice daily, beginning with the first post-inoculation day. The number of mice found normal, sick or dead is recorded on a mouse-history card which remains on file as a permanent record of the experiment. The observation period should extend for at least 21 days


after inoculation. Only rarely will rabies virus be detected in the inoculum later than 21 days after the inoculation.

The following signs should be noted on the mouse-history card, using the appropriate letter(s):

A. Ruffled fur. B. Tremors when held in the air by the tail with a pair of forceps. C. Lack of coordination of hind legs- note gait when placed on table and made

to move. D. Paralysis. E. Prostration (near death).

Deaths occurring 24-48 hours after intracerebral inoculation are attributable to causes other than rabies virus, such as trauma, bacterial contamination or other viruses. For diagnostic purposes, 1 or 2 mice may be killed each day, beginning on the fifth day, and their tissues examined for Negri bodies as well as for the presence of rabies antigen as demonstrated by immunofluorescent antibody staining. Frequently, an early diagnosis is thus obtained, particularly in instances where certain strains of street virus might take between 1 and 3 weeks to kill the mice

Note: Rabies should not be diagnosed on the basis of clinical signs of the disease alone. Although signs of paralysis 5 days or more after inoculation may give grounds for suspecting the presence of rabies virus, the same signs may be observed in numerous other viral, bacterial and protozoa1 infections that involve the central nervous system of the mouse. Definite evidence of the identity of the virus is obtained with the mouse neutralization test.

Further passages of infected material

If desired, brain tissue from mice that have succumbed to infection after noculation with the original virus may be made into a suspension as described above. It can then be stored, used in neutralization tests, or inoculated into another group of mice.

Removal of the brain

The brains of all mice that have died during the experiment, or that have been killed when prostrate, should be removed and examined for the presence of Negri bodies and rabies antigen by immunofluorescent antibody staining (see Chapters 4 and 7). The mice showing symptoms of prostration (terminal symptoms of rabies) should be killed with chloroform.

The mouse should be pinned, ventral surface downwards, to a dissecting board. Only two pins are necessary, one through the nose and one through the base of the tail. Three pins or spring clips may also be used: one through each of the forelegs, and one through the back of the tail.

After disinfection with 70% ethanol, the skin of the head and neck is cut away with forceps and scissors, exposing the skull. The skull is grasped in the orbits with mouse-tooth forceps and the calvaria of the skull is cut away with curved scissors, thereby exposing the brain. The brain is removed with curved scissors and is


transferred to a sterile Petr dish A thin section of the brain is cut out just anterior to the cerebellum, and is transferred to a wooden tongue-depressor or a paper towel A clean microscope sl~de is then pressed lightly against the cut surface of the section the pressure should be sufficient to create a slight spread of the exposed surface against the slide Negri-body stain and immunofluorescent antibody stain should then be applied to the slide (see plate 5 1, B and C, page 70)

Complications

Bacterial contamination of inoculum

If bacterial contaminants have caused the death of the mice in spite of the addition of antimicrobials, and if the original suspension has been preserved, the following methods may be tried in an effort to overcome the interfering action of the bacteria.

(a) Filtration through bacteria-retaining filters: The supernatant liquid of a suspension centrifuged at 400 g for 15 minutes should be used for this purpose. However, since rabies virus is a fairly large particle and the concentration of virus in specimens submitted from the field is not usually very great, the virus may be lost in the process of filtration.

(b) Dilution method: Sometimes the suspension can be diluted beyond the end- point of bacterial contamination with retention of viral activity, but this happens very rarely.

(c) Prolonged storage, In some instances, it is easier to overcome the effects of bacterial contaminants after the tissue suspension has been stored for a period of time either at freezing temperatures or in glycerol (see pages 62 and 82).

(d) Parenteralinociilation: The Syrian hamster, which is the animal most susceptible to parenteral infection, may be chosen for this purpose. Mice are relatively insusceptible to parenteral infection with rabies.

Presence of two viruses

This is particularly confusing if the second virus has pathogenic properties similar to those of rabies. Again, intracerebral or parenteral inoculation of animal species other than the mouse may be attempted, particularly in view of the extremely wide host-range of rabies virus.

Interference phenomena

In certain instances, it may be necessary to dilute the ir~oculum 10-100 times, or even more. This may be because of the properties of a particular strain of rabies virus or because large amounts of inactive virus particles, serum antibody, or other unidentified factors may interfere with the living virus. There is no rule for determining when this should be done. However, i f failure to isolate rabies virus is consistently encountered in the same species of animal in a particular geographical area, the possibil~ty of an interference phenomenon should be considered, and tissue suspensions should be diluted beyond 10% concentrations.

CHAPTER 7

The fluorescent antibody testi D. J. Dean, M. K Abelseth3 & P. Atanasiu4

The fluorescent antibody (FA) test is one of the most accurate microscopic tests available for the diagnosis of rabies and should be employed by all laboratories undertaking such work. Apart from a satisfactory microscope, the main requirements for success in using this technique are well trained staff and conjugated serum of good quality. When properly performed, the test is fast, comparatively inexpensive, and more accurate than either the examination of films or sections by recommended procedures or mouse inoculation tests. Fresh, frozen or glycerola- ted material may be examined. Diagnosis can be made accurately in most instances in minutes or hours, whereas the examination of sections or animal inoculation requires days or weeks.

Principle

The fluorescent antibody procedure was developed by Coons & Kaplan in 1950 and subsequently modified for the diagnosis of rabies by Goldwasser & Kissling in 1958. The procedure consists in labelling antibody with a fluorochrome, allowing the labelled antibody to react with specific antigen if present, and observing the product of the reaction under the fluorescence microscope. A substance is said to fluoresce if, upon absorbing light energy at a certain wavelength, it emits light of another wavelength.

Antigens reacting with antibodies tagged with fluorescein isothiocyanate, the dye most frequently used with rabies, appear under ultraviolet light as brightly coloured, apple-green or greenish-yellow objects against a dark background, which may or may not contain nonspecific fluorescing material. The character and intensity of the colour may be modified by the use of filters. Fluorescein isothiocyanate is available commercially in powdered form and may be stored for long periods without deterioration.

The fluorescence microscope consists of a standard microscope of satisfactory quality equipped with a darkfield condenser and an ultraviolet light source. Since only a small portion of the incident radiant energy is converted to fluorescent

light, examinations should be made in a darkened room employing the most intense source of light available; both excitation and barrier filters are used.

' Based on the chapter ~n the prevlous e d t ~ o n , which was prepared by D. J Dean & M. K. Abeiseth and updated by the late P. Atanas~u. Formerly Drector, Divislon of Laboralor~es and Research, New York State Department of Health, Albany, NY, USA. Formerly D~rector , Laborator~es for Veterlnary Sclence, Divlslon of Laboratories and Research, New York State Department of Health, Albany, NY, USA. Deceased. Former Honorary Professor, Pasteur lnst~tute, Par~s, France.

FLUORESCENT ANTIBODY TEST

Although a monocular microscope will yield more light, a good quality binocular microscope will provide sufficient light and is preferred in diagnostic laboratories where a large number of specimens are being processed. Light used for excitation is of a shorter wavelength than that emitted by the preparation. Blues and greens are excited only by ultraviolet light, while yellow and red fluorescence may also be excited by blue-violet light.

Several manufacturers provide complete kits for fluorescence microscopy and equipment is available for converting standard microscopes of satisfactory quality for use in fluorescence work. Guidance in the selection of a microscope should be obtained from those experienced in fluorescence microscopy and the manufac- lurer should be requ~red to demonstrate its capabilities prior to purchase. Apochromatic or fluorite objectives allow higher apertures and therefore present brighter images. Achromatic objectives can be used but provide less brilliance. The use of a dry sub-stage darkfield condenser should be considered since it eliminates the use of oil between the slide and front lens of the regular condenser. Large amounts of oil on the microscope may be a hazard since the methods described to process rabies specimens do not inactivate the virus.

An epifluorescence microscope with a x40 objective is recommended. The mercury or halogen lamp should be checked regularly and changed i f necessary.

Materials and methods

Conjugate

Once suitable microscopic equipment has been obtained, the quality of fluorescence is primarily dependent upon the quality of the conjugate used. Some laboratories use sera from hamsters immunized according to schedules similar to those developed by the California State Department of Health in the USA. Other laboratories use equine or goat antisera for conjugation. Since conjugates of acceptable quality are now available commercially, only laboratories familiar with such procedures should attempt to make their own. Such conjugates should be checked and calibrated against reference sera before they are used for diagnosis. Conjugates prepared in the laboratory should be lyophilized to prevent loss of potency. Good quality conjugates should have a minimum of nonspecific fluorescence at low dilutions and should be capable of being diluted as much as 1 :20 or more to prevent this problem. After a conjugate lot has been prepared, dilutions should be made and tested to determine which one is most satisfactory for routine diagnosis.

As needed, lyophilized conjugate is removed from the deep freeze and reconstituted to its original volume with sterile distilled water. It should then be filtered to remove extraneous materials, The desired amount of reconstituted conjugate is added to the appropriate volume of normal or infected mouse-brain suspension prepared with egg yolk as described on page 90. For example, if satisfactory staining is obtained with a final dilution of conjugate of 1 :40, 0.1 ml of conjugate is added to 3.9 ml of mouse-brain suspension. The remaining undiluted conjugate is stored at 4°C until use. After storage, the conjugate may need to be centrifuged at 200 g for 5 minutes to remove precipitate before it is added to the mouse-brain suspension. Diluted conjugate may be used for a period of a week or so, provided precautions are taken to safeguard against undetected decay of the


product or contamination. If the conjugate IS prepared as outlined in Appendix 2, the final dilution may be made with sterile distilled water rather than with the normal mouse-brain suspension as described. This is possible since very little nonspecific fluorescence is found in conjugates prepared in this manner. It should be noted that conjugate prepared in the laboratory may contain nonrabies fluorescein-labelled antibodies if multiple antigens are present in the antigen used in the preparation of antisera. Such conlugates may react when autogenous tissues (from different animals of the same species) are being examined. Care must be taken therefore to select pure rabies antigen.

An antinucleocapsid conjugate has recently been prepared in the rabbit. This conjugate has no nonspecific fluorescence.

Preparation of slides

Two methods of preparing slides are comnionly used, the smear method and the impression method (see Chapter 4). Uniform, thin films can be prepared by grinding hippocampal or other brain material without alundum l o a uniform paste with a sterile pestle and mortar or by cutting this material into small pieces with scissors. Films provide a greater number of cells for examination. This method is preferred for preparing brain material that has partially decomposed. Impression smears of such material would be too thick for successful examination.

Control slides

Control films or smears are usually prepared from the entire brains of mice or young hamsters inoculated intracerebrally with rabies street virus and killed when moribund. After fixation in acetone, slides can be stored under dry conditions at -20°C for long periods before use; however, slides no older than l 0 days are preferred.

Moose-brain suspension

When infected mouse brain is used as antigen for preparing antisera, it is usually necessary to dilute conjugates with a 20% normal mouse-brain suspension in PBS, pH 7.4 (see Chapter 8, Annex l ) , supplemented with 10% suspension of the yolk of chicken eggs embryonated for 6-7 days. After centrifugation for 10 minutes at 1000 g, the supernatant IS dispensed in glass containers in al~quois sufficent for 1-7 days' work and stored at - 20°C until used. Suspensions of infected mouse brain are similarly prepared using brains froni young adult mice inoculated intracerebrally with a 1 : l 0 0 to 1 : l000 suspension of the CVS st~.ain of fixed rabes virus and harvested when moribund.

lmmunofluorescent staining

Many variations have been developed in the procedure for immunofluorescent staining; however, the direct method is the most satisfactory and practical for rabies diagnosis and will be the only one described.


The direct test The direct method consists in applying a suspension of rabies antibody, prev~ously labelled with fluoresce~n isothiocyanate, directly to films, impression smears or sections of tissue under examination (Fig. 7.1).

No less than four films of salivary-gland tissue andlor brain tissue should be examined before a field specimen is reported as being fluorescent-negative. They should include two made from tissue from Ammon's horn and two from a paste prepared by grinding in a mortar, without diluent or alundum, equal portions of tissue from Arnmon's horn, the cerebellum and medulla oblongata (brain stem). Each of these parts may be examined separately i f desired. In examining bovine speclmens, the cerebellum a n d brain stem are tissues of choice. When t h e mouse

inoculation test is used, the remaining ground tissue IS diluted with distilled water contaning 5% normal horse serum to obtain a 10% tissue suspension and then injected intracerebrally. A direct comparison of the results of fluorescence and mouse inoculation is thus obtained.

When specimens are received in 50% glycerol-saline (preservative), it is imperative that the tissue be washed several times in saline. Glycerol may interfere with the FA test because acetone combines with glycerol to mask fluorescence. In such cases satisfactory results may be obtained by omitting the acetone fixation step.

Films are air-dried, placed in a Copln jar or other suitable container, covered with cold acetone, and held in a freezer at -15°C to - 20"C for 2-4 hours; excellent results are also obtained with overnight fixation. After fixation, slides are removed from the acetone and air-dried. If films are used, two suitable uniformly thin areas 2.5 cm long are demarcated on the slide with a wax marking pencil. Impression smears, two per slide, are similarly demarcated. Slides, including previously prepared controls, are then placed on glass rods strateg~cally arranged across the top of a small sterilizing tray or other suitable container.

One area of f~ lm or one impression smear is stained by placing two drops of the diluted conjugate within the area marked by the wax pencil. The other film or smear is stained in the same way with conjugate similarly diluted with infected mouse- brain suspension. Conjugate should be spread uniformly, without disturbing the

Fig. 7.1 Direct method of detecting rabies antigen

Labelled + unlabelled = labelled anti body antigen anti body-antigen

WHO 94920

Schematic representation o f reachon oi unlabelled rabies vitiis (antigen) with fl~of?Scein- tagged antibody to yield fluoiesce~n labelled antigen antibody product


Plate 7.1 Phofomicrographs of brain material treated with hamster conjugate diluted 1 : 30

A Cerebellum--cow. Note fiiie particles (dust) as weli as antigen in nerve cell. B, Neuron-dog. C Neuron--cow. D. Hippocampus-fox. Typical variation ~n size. Suitable as positive control slide. E Amrnon's horn-+at. F Cerebeiium-mouse (CVS strain)


film, either by rotating the slides or with an applicator stick or toothpick. A fresh stick should be used for each area stained. The slides are placed on a rack in a pan (a shallow boiling pan with suspended glass rods similar to that used for bacterial staining is suitable). Water is added to the pan, which is then covered and placed in an incubator at 37°C for 30 minutes. This procedure creates sufficient humidity to prevent drying of the test material.

After incubation the slides are washed by dipping in PBS, pH 7.4, then further rinsed by immersing for two successive 10-minute periods. Coplin jars are convenient for this purpose. The slides are then removed and air-dried in a vertical position. When dry, one drop of 50% buffered glycerol (pH 7.6) is added and coversl~ps are placed over the areas to be examined. When properly stained, the positive control film and test films containing rabies antigen will contain brilliantly fluorescing apple-green or greenish-yellow structures, varying in size from tiny bodies, which are commonly called sand or dust and are barely visible, to those comparable in size to Negri bodies (Plate 7.1). Fluorescing antigen of diagnostic significance may be seen in nerve fibres and cells from both the central nervous system and the salivary glands. Control slides should always be examined before and after scanning the test films to ensure that the equipment is operating satisfactorily and that films are properly stained. When large numbers of slides are examined consecutively, control slides should also be interspersed with the test slides, e.g. every tenth slide.

No fluorescence should be seen on films or smears prepared from conjugate diluted with uninfected mouse-brain suspension as described above; this control procedure is therefore important in determining the specificity of the fluorescence and minimizing the number of false positives.

A number of techniques are available for the intra vitam diagnosis of rabies in humans and animals. Viral antigen may be detected by the FA test in cornea1 impressions, skin biopsies or saliva samples from patients with rabies. In addition, the test may be used to detect rabies virus-neutralizing antibodies in serum samples from patients with rabies. However, the technique is of limited sensitivity for intra vitam diagnosis since a negative result does not rule out the possibility of infection. Accordingly, samples diagnosed as negative should be inoculated into suckling mice (see Chapter 6).

Discussion

The FA test for the detection of rabies antigen has been perfected to such an extent that it is one of the quickest and most reliable methods available, both for diagnostic and research purposes. Errors can usually be traced to faulty equipment, unsatisfactory conjugate, lack of control slides, and lack of experience in reading slides.

The FA test may occasionally yield positive results when the mouse inoculation test is negative, since the former detects inactivated as well as live antigen. In this respect, fluorescence tests are more sensitive than inoculation tests in mice. All test slides submitted to 50 state health department laboratories in the USA in 1968 were diagnosed correctly. Some laboratories have reported, on occasion, mouse- positive, fluorescent-negative specimens, but such reports have declined as staff have gained experience in performing the FA test.


From 1966 to 1970, nearly 15000 specimens were examined in the Laboratories for Veterinary Science of the New York State Department of Health, Albany, NY, USA. Of these, 802 of a totai of 804 were found positive by the FA test. In only 2 cases was the mouse test positive and the FA test negative. Insufficient sampling was responsible for these results. Since samples found positive in the FA test are not routinely injected into mice, data are not available on the reverse situation. It is perhaps significant that the New York and Californian state laboratories, which reported almost perfect correlation, employed a conjugate of hamster origin that could be diluted as much as 1 :50 prior to use, thus minimizing or eliminating the problem of nonspecific fluorescence. It is advisable to test in mice all fluorescent- negative specimens involving human exposure in order to monitor the accuracy of the FA test. It should be kept in mind that suckling mice (less than 3 days old) are more susceptible to rabies than weanling or adult mice and therefore should be used where rapid results are important. Since antigen can be detected by the FA test as early as 3 days after inoculation, one mouse per day can be killed beginning on day 3.

Errors may also result from the use of unsatisfactory or inadequately adjusted equipment, nonspecific fluorescence, gradual and unrecognized loss of light from a fluorescent lamp, failure to use the adequate controls, and colour blindness on the part of the operator. In addition, care should be taken to ensure that cross- contamination of slides does not occur.

Adequate sampling is an important factor. Since rabies virus is usually transmitted from the site of inoculation to the central nervous system via the peripheral nerves, we routinely examine brain stem material either separately or as part of the composite sample. On repeated occasions, the brain stem has been positive when films prepared from the hippocampus, cerebellum or cerebral cortex were negative. Brain-stem sampling is particularly advantageous when animals are killed during the early stages of the disease.

With satisfactory conjugate, counterstains such as rhodamine-conjugated albumin are rarely needed. When rhodamine is used in dilutions of less than 1 :80, loss of specific fluorescence occurs in direct proportion to the concentration of rhodamine used. Dilutions of 1 :80 or greater are sufficient to stain background material a light brick red, leaving the greenish-yellow rabies antigen in sharp contrast. There is no apparent correlation between the condition of the specimen examined and the degree of nonspecific fluorescence. Good to excellent fluorescence with little or no nonspecific fluorescence has been observed in badly decomposed specimens and, occasionally, is encountered in fresh specimens. With experience, however, it is possible to differentiate between specific and nonspecific fluorescence wlthout great difficulty. The importance of using inhibited conjugate (diluted as described earlier) cannot be overemphasized since this procedure will eliminate nonspecific fluorescence.

Examination of the salivary glands should be made routinely in the case of bats involved in human exposure and may yield valuable information with other species as well. Salivary glands are more difficult to sample adequately and to process than brain material. Since grinding with a pestle and mortar is not usually satisfactory owing to the character of the tissue involved, films are customariiy prepared from the salivary-gland tissue by the impression technique (see page 57) and stained in the same way as brain material used for the direct microscopic test.


The fluorescence seen in salivary-gland preparations is more diffuse than that seen in brain tissue. Nonspecific fluorescence, varying from blue to red, is often seen and appears to be especially characteristic of glandular materal. With adequate sampling and routine use of control slides, examination of the salivary glands by the FA test can provide accurate information for the diagnosis of rabies in bats.

CHAPTER 8

Virus isolation in neuroblastoma cell culture W. A. ~ebster ' & G. A casey2

Techniques for the isolation of field strains of rabies virus in cell culture are now well developed and are widely used for the diagnosis of rabies. In many laboratories, the rabies tissue-culture infection test (RTCIT) has replaced the mouse inoculation test (MIT). The RTClT is relatively easy to perform, is much less expensive than the MIT and, most important, can substantially reduce the time required for obtaining results.

The early use of cell cultures for the isolation and growth of various rabies virus strains was reviewed by Crick & King (I) in 1988. The first attempts at isolating field virus used baby hamster kidney (BHK-21) cells. However, the results were not encouraging, since the sensitivity of this cell type was often less than that of the MIT. Numerous other cell types, including chick embryo-related (CER), McKoy, bovine, bat, skunk, dog and racoon cells have been used for the isolation of rabies field virus (2-4). While BHK-21 cells are still used in certain serological tests and in studies on fixed rabies virus strains, murine neuroblastoma cells are now generally accepted as being superior to all others tested for the isolation of field virus strains

(4, 5). Before any new technique can be accepted as a recognized diagnostic test, its

sensitivity must be rigorously compared with that of established tests. Numerous comparisons have now been made between the RTCIT and both the fluorescent antibody (FA) test and the MIT (3, 5-9). Results of these studies indicate that the RTClT is at least as sensitive as the MIT in demonstrating the presence of rabies field virus in animal and human tissues, In studies with inocula containing very small amounts of field virus, the MIT in 21-day-old mice demonstrated virus in only 50%, whereas the RTClT detected virus in 99% of the same inocula (9).

The RTCIT procedure outlined below has been developed to enable a small number of staff to examine a relatively large number of specimens (9). Between June 1986 when the test was formally adopted by the rabies unit of the Animal Diseases Research Institute and December 1990, 52600 specimens suspected of being infected with rabies were examined by a staff of five. A total of 27500 specimens diagnosed as negative by the FA test were tested by the RTClT and 34 of these were found to be positive. Under these conditions, the test must be designed in such a way that it is easily performed, reliable and inexpensive. The RTClT is as quick to perform as the MIT and the costs associated with the latter are up to 5 times higher. Most importantly, results are obtained within 4 days with the RTCIT, as compared with 30 days in the MIT (see also Chapter 6).

' Formerly Biologist, Rab~es Unit, Agriculture Canada, Food Production and Inspection Branch. Animal Diseases Research Institute. Nepean, Ontar~o. Canada Blolog~st, Rab~es Unlt, Agr~culture Canada, Food Producl~on and lnspectlon Branch, Animal Diseases Research Institute, Nepean Ontario, Canada.

ISOLATION IN NEUROBLASTOMA CELL CULTURE

Rabies tissue-culture infection test (RTCIT)

Tissue preparation

Brain tissue is mashed on a paper towel using a wooden tongue-depressor and approximately 0.5 g is added to 5 m1 of phosphate-buffered saline (PBS) containing antibiotic (see Annex 1) to make a 10% (w/v) suspension. This suspension is shaken or vortex-mixed vigorously and allowed to settle for at least 1 hour at 4 "C. The upper clear layer is then withdrawn and diluted 10-fold with Eagle's minimum essential medium supplemented with 10% fetal calf serum (EMEM-10) (see Annex 1) to make a 1% (w/v) suspension.

Cell suspension

Murine neuroblastoma (NA-Cl300) cell cultures are grown at 35-36°C in 25-cm2 plastic culture flasks containing 5 ml of EMEM-10. A flask with a nearly confluent monolayer will provide sufficient cells to inoculate one 96-well microtitration plate. Monolayers are detached by trypsinization and resuspended in EMEM-10 to give a concentration of approximately 5 X 105 cells/ml. Immediately before use, diethylaminoethyl-dextran (DEAE-dextran) is added to give a final concentration of 25 pg/ml.

Test procedure

1. For each specimen, add 0.1 m1 of the cell suspension to each of 4 wells of a 96- well microtitration plate. A total of 24 specimens can be tested on one plate.

2. Add 0.2 m1 of the 1 % brain-tissue suspension (in EMEM-10) to each of the 4 welts containing cells

3. Mix the wells on a mixer. 4, Incubate at 35-36°C in a humid chamber with 5% carbon dioxide (CO,) for 4

days. The plates should be checked daily for cell growth. If contaminated wells are noted, that specimen should be re-mixed in saline containing twice the original amount of antibiotic and re-inoculated into new cell cultures.

5. Prepare a positive field virus (3 wells) and negative (2 wells) control plate. This should be done on a separate microtitration plate and after all the test plates have been prepared in order to eliminate any possibility of contaminating a negative specimen with rabies-positive material.

6. After incubation, remove the media from the wells and wash once with 0.01 mol/l PBS (Annex 1).

7. Fix the cells by adding cold 70-80% acetone to each well and incubate at room temperature for 30 minutes.

8. Shake out the acetone and air-dry the plates. 9. Add 1 drop of fluorescein-antibody conjugate to each well and incubate at

37°C for 30 minutes. 10. Shake out the conjugate and wash once with 0.01 mol/l PBS. 11. Counterstain with Evans Blue (1 :40000) for 15-30 seconds. 12. Rinse once with 0.01 mol/l PBS. 13. Exam~ne the plates under an ultraviolet microscope using a I O x

objective.


Factors affecting sensitivity

Various factors such as specimen contamination, preservation and storage may affect the sensitivity of the RTCIT. These are discussed briefly below.

Cell culture Various strains of field rabies virus (as identified with monoclonal antibodies) may differ in biological properties such as their isolation and growth characteristics in cell cultures. While murine neuroblastoma cells have been shown to be more sensitive than BHK-21 cells to infection by most field viruses, some virus strains grow equally well in both cell types (10). Moreover, not all neuroblastoma cell lines are equally sensitive ( I I ) . Differences have also been noted between cultures of the same cell line obtained from different sources (W.A. Webster, unpublished observation). In the latter study, two cultures of murine neuroblastoma cells (NA- C1300) were obtained, each from a different laboratory, and inoculated with 10% brain suspensions of the Canadian Arctic field strain (12). Results indicated average infection rates of 65% and 10% with the same inocula in the two cultures. Before setting up a diagnostic procedure, extensive testing must therefore include the com~ar ison of cell cultures from various sources.

Tissue preparation The 10% brain-tissue suspensions can be clarified either by allowing the larger debris to settle or by centrifugation. Depending upon various factors, centrifugation of the suspension may reduce the amount of virus demonstrable by cell culture (5). Allowing the tissue to settle in PBS containing antibiotic for at least 1 hour has several advantages. It reduces the costs of the test, since a centrifuge is not required. In addition, the lack of nutrients and the presence of antibiotics in the saline can eliminate a certain amount of bacterial contamination in the tissue, which is often received in various stages of decomposition. However, the use of saline-antibiotic diluent is not recommended for long-term storage of tissue suspensions. Cell-culture medium containing 10% fetal calf serum (EMEM-10) is fairly effective for preservation of rabies virus (including fixed strains), although some f~eld viruses lose infectivity even at - 70°C.

The use of a further 10-fold dilution in EMEM-10 of the 10% brain suspension as inoculum removes most toxic substances associated with brain tissue. Cell growth is improved and the resulting infection rates are at least as high as those of cultures inoculated directly with the 10% brain suspension (9).

While brain is the preferred tissue for testing, other tissues or fluids may be used when there is damage to the cranium or for intra vitam diagnosis. Salivary-gland tissue or saliva is often used and virus can be easily isolated in cell culture if serial dilutions in EMEM-10 are used as the inoculum. The use of a 1 % tissue suspension is recommended, since inocula containing large amounts of salivary-gland tissue often produce severe cytolysis in cell culture. Rabies virus may also be isolated from spinal-cord tissue, which may be positive when brain tissue is negative, as well as from fluids (saline) used to wash the cornea, although trials with the latter have been limited (4) .

Specimen contamination Bacterial and fungal contaminants can have a deleterious effect on cell cultures. The preparation of tissue suspensions in cold PBS containing antibiotic reduces

ISOLATlON IN NEUROBLASTOMA CELL CULTURE

the effects of bacterial contaminants (9). Of some 3700 specimens tested during a perod when ambient temperatures ranged between - 10°C and + 30°C only 4 of these were unfit for a diagnosis to be reached. In severely contaminated cases, extra antibiotic may be added. A fungicide may also be added to control fungi, although growth of these organisms does not always result in cell destruction.

Filtration of tissue suspensions to remove contaminants reduces the amount of demonstrable virus; in some cases, the number of positive cells may be reduced by 50-100% (W.A. Webster & G.A. Casey, unpublished observation).

Test plates The RTClT can be performed on 96-well microtitration plates (5, g ) , 60-well or 96- well liunian leukocyte antigen (HLA) plastic plates (3) , coverslips (8) , or on multi- chambered glass slides (6, 7). The use of 96-well microtitration plates is preferred to other types for a variety of reasons. Although the volume (13 ~ t l ) of suspension used in each well of an HLA plate I S sufficient for suspensons containing large amounts of virus, the results obtained with suspensions containing very small

Fig. 8.1 Rabies-infected neuroblastoma ceDs stained by immunof/uorescence


Fig. 8.2 Rabies-infected neuroblastorna cells stained by the avidin- biotin techniaue

amounts of virus (as would be expected in these test suspensions) are inconsistent. Coverslip cultures grown in flasks pose special problems with handling because of their fragility. The use of multi-chambered glass slides can increase the cost of the test by a factor of 5-10.

Fixation and staining Fixation of cell monolayers can be accomplished using either acetone, various alcohols or formalin. Both 70-80% acetone and an absolute methanol:forrnalin (1 . l ) mixture have been used on plastic microtitration plates and produce excellent results (5). Formaldehyde (10% solution) may also be used, but tends to diminish fluorescence when the virus is stained with immunofluorescent stains.

The usual method of examining cell cultures for the presence of rabies virus is by immunofluorescence staining. Counterstaining with Evans Blue increases the contrast by decreasing background and nonspecific fluorescence. Care must be taken. however, not to overstain with Evans Blue since it can also reduce specific fluorescence. The use of a 100 X magnification (10 X objective) gives excellent results (Fig. 8.1) and allows a much more rapid examination than at higher magnifications. Staining cells by immunoenzymatic (13, 14), peroxidase- antiperoxidase ( 1 5 ) and avidin-biotin (Fig. 8.2) methods also gives satisfactory results (A. Bourgon, unpublished results), but is more time-consuming and more technically complicated (see Annex 2). Infected monolayers on glass slides can be stained with histological stains (see Chapter 4), such as van Giesen's stain for Negri bodies; however, this method is not recommended, since only the larger inclusion bodies are seen and many moderate to light infections would be misdiagnosed.


incubation period It is now generally accepted that the test culture should be incubated for 4 days (5, 9). Shorter periods have been suggested (3, 6-8). However, when only small amounts of virus are present, a longer incubation period is required. Most specimens which are negative by the FA test and subsequently positive by the RTCIT infect only small numbers (1-5%) of cells in culture (9). Examination of these cultures on successive days post-infection indicated that a positive diagnosis would have been made in only half of the specimens examined at 2 days. whereas all cultures were rabies-positive at 4 days post-infection (9).

References

1. Crick J, King A. Culture of rabies virus in vitro. In: Campbell J B , Charlton KM, eds. Rabies. Boston, Kluwer Academic Publishers, 1988: 47-66.

2. Consales CA et al. Cytopathic effect induced by rabies virus in McCoy cells. Journal of virological methods, 1990, 27: 277-285.

3. Tollis M et al. Sensitivity of different cell lines for rabies virus isolation. Zentralblatt fur Veterinarmedizin, Reihe B, 1988, 35: 504-508.

4. Webster WA, Casey GA. Diagnosis of rabies infection. In: Campbell JB, Charlton KM, eds. Rabies. Boston, Kluwer Academic Publishers, 1988: 201 -222.

5. Rudd RJ, Trimarchi CV. Development of an in vitrovirus isolation procedure as a replacement for ihe mouse inoculation test in rabies diagnosis. Journal of clinical microbiology, 1989, 27: 2522-2528.

6. Barrat J et al. Diagnosis of rabies by cell culture. Comparative immunology, microbiology and infectious diseases, 1988, l l : 207-21 4.

7. Bourhy H et al. Comparative field evaluation of the fluorescent-antibody test, virus isolation from tissue-culture and enzyme immunodiagnosis for rapid laboratory diagnosis of rabies. Journal of clinical microbiology, 1989, 27: 519- 523.

8 . Chitra L, Pandit V, Kalyanararnan VR. Use of murine neuroblastoma culture in rapid diagnosis of rabies. Indian journal of medical research, 1988, 87: 113-116.

9. Webster WA. A tissue-culture infection test in routine rabies diagnosis. Canadian journal of veterinary research, 1987, 51: 367-369.

10. Webster WA, Charlton KM, Casey GA. Growth characteristics in cell culture and paihogenicity in mice of two terrestrial rabies strains indigenous to Canada. Canadian journal of microbiology, 1988, 34: 19-23.

11. Tsiang H et al. Neurotropism of rabies virus. An in vitro study. Journal of neuropathology and experimental neurology, 1983, 42: 439-452.


12. Webster WA, Casey GA, Charlton KM. Major antigenic groups of rabies virus in Canada determined by anti-nucleocapsid monoclonal antibodies. Compara- tive immunology, microbiology a n d infectious diseases, 1986, 9: 59-69.

13. Mannen K et al. Micro-neutralization test for rabies virus based on an enzyme imrnunoassay. Journal o f clinical microbiology, 1987, 25: 2440-2442.

14. Steece RS, Calisher CH. Evidence for prenatal transfer of rabies virus in the Mexcan free-tailed bat ( Tadarida brasiliensis mexicana), Journal o f wildlife diseases, 1989, 25: 329-334.

15. Bourgon AR, Charlton KM. The demonstration of rabies antigen in paraffn- embedded tissues using the peroxidase-antiperoxidase method: a comparative study. Canadian journal of veterinary research, 1987, 51: 117-120.

Annex I Media

Eagle's minimum essential medium (EMEM), as modified by MacPher- son & Stoker, 1962

Component Sodium chlorde (NaCI) Glucose Potassium chloride (KCI) Calcum chloride (CaCI,) Sodium phosphate, nionobasic.

monohydrate (Na,H2P0,. H20) Magnesium sulfate (MgSO,) Phenol red Sodium succinate, hexahydrate Succinic acid L-Arginne hydrochloride L-D~sodium cystine L-Histidine hydrochlorde, rnonohydrate L-lsoleuclne L-Leucine L-Lysine hydrochloride L-Methionine L-Phenylalan~ne L-Threonne L-Tryptophan L-Tyrosine L-Valine Calcium D-pantothenate Choline bitartrate Choline chloride Folic acid i-lnositol


Nicotinamide Pyridoxal hydrochloride Riboflavin Thiamin hydrochloride

EMEM- 10

EMEM-10 is prepared from EMEM, as above, supplemented with:

Fetal calf serum Tryptose phosphate broth

l -Glutarnine Neomycin suifate

Phosphate-buffered saline (PBS), 0.01 moll/, pH 7.4

Sodium chloride (NaCI) Sodium phosphate, dibasc (Na,HPO,) Potassium phosphate, monobasic (KH,PO,) Distilled water to make

150 g 20.5 g

4.9 g 18 litres

The pH should be adjusted to 7.4 by adding either Na,HPO, or KH,PO, as required.

PBS with antibiotic

PBS with antibiotic is prepared from PBS, as above, supplemented with:

Streptomycin sulfate Benzylpenicillin

Annex 2 Avidin-biotin staining method'

1 . Fix cell cultures in cold 70-80% acetone and incubate at room temperature for 30 mnutes. Shake out acetone and air-dry.

2. Rinse each well (3 times) with tromethamine2 buffer (0.01 rnol/l, pH 7.5). 3. Cover cells with normal goat serum diluted 1 :20 with tromethamine buffer and

incubate at 45 "C in a humid chamber for 20 minutes. 4. Repeat step 2. 5. Cover cells with primary antisera (rabbit anti-N protein serum diluted in

tromethamine buffer, preferably at a high dilution if titre permits) and incubate at 45 "C for 20 minutes. Rinse with tromethamine buffer and repeat incubation.

6. Repeat step 2. 7. Cover cells with bridge or link antisera (biotinylated goat antirabbit serum

diluted 1 :20 in tromethamine buffer) and incubate at 45°C for 20 minutes.

' W.A. Webster, unpublished. *A lso known as tr~s(hydroxymethyi)aminomethane


8. Repeat step 2. 9. Immerse cells in 30% hydrogen peroxide (H202) diluted 1 :25 with absolute

methanol and leave at room temperature for 10 minutes. Repeat. (No rinse between.)

10. Repeat step 2. 11. Cover cells with ABC complex (streptavidin-biotinylated horseradish peroxi-

dase) and incubate at 45°C for 20 minutes. This should be prepared 30 minutes in advance (mix 0.1 ml of streptavidin with 10 m1 of trornetham~ne buffer and then add 0.1 ml of biotinylated peroxidase).

12. Repeat step 2. 13. Cover cells with fresh diaminobenzadine tetrahydrochloride (DAB) solution

(0.05% DAB in trornethamine buffer (10 ml) with 3 pi of H202) and leave at room temperature for 2 minutes. Mix Immediately before use (at late stage of ABC complex incubation). Repeat.

14. Rnse with tromethamne buffer. 15. Rnse with distlled water. 16. Counterstain w ~ t h haernatoxylin for 1 minute an.d change to fresh haerna-

toxylin for 5 minutes. Wash twice in tap water. Blue in tromethamine buffer. 17. Dehydrate rapidly (to absolute alcohol), treat with xylene, and mount.

CHAPTER 9

Rapid rabies enzyme immunodiagnosis (RREID) for rabies antigen detection H. Bourhy ' & P. Perrin2

Introduction

Enzyme immunoassays have recently replaced many of the traditional techniques for diagnosing viral infections in animals and humans. These new procedures do not require expensive equipment and are quick and easy to perform. Moreover, they are considerably safer than many of the traditional techniques and, because of their high sensitivity and specificity, can also be applied to large numbers of specimens under field conditions. In 1986 a solid-phase enzyme-linked immunosorbent assay (ELISA) called rapid rabies enzyme immunodiagnosis (RREID) was developed for the diagnosis of rabies, based upon the detection of rabies virus nucleocapsid antigen In brain tissue (I). During extensive trials, it was found easy to perform and showed high precision and reproducibility (2-6). Subsequently, an RREID test using enzyme-labelled avidin-biotin (RREID-biot) was developed. It appeared to increase slightly the sensitivity of the RREID without affecting its specificity. However, the most recently developed enzyme immunoassay (RREID- lyssa), also based on an avidin-biotin amplification system, further increases the detectability of rabies-related strains (lyssavirus serotype 2, 3 and 4 and European bat lyssavirus) in infected specimens (7).

Method

Principle of the RREID

Rabies nucleocapsid antigen is extracted from infected cells and purified on a caesium chloride (CsCI) gradient (8, 9). The viral antigen is emulsified in Freund's adjuvant and administered intramuscularly to outbred rabbits. After several injections, blood samples are taken from the rabbits. Rabies antinucleocapsid IgG is then purified from the serum using chromatography (10) (see also Appendix 2).

The immunoassays described here employ rabies antinucleocapsid rabbit IgG as solid phase ( I ) . However, the RREID-lyssa uses a mixture of rabbit IgG directed against lyssavirus serotype 1 (PV strain), serotype 3 (Mokola virus) and European bat lyssavirus (EBLI), while the RREID and RREID-biot are performed with IgG directed against only the PV strain. The incubation of a positive specimen results in the binding of rabies nucleocapsid antigen to the IgG. In the RREID, the bound viral antigen is revealed with a peroxidase conjugate (rabies antinucleocapsid rabbit IgG conjugated with horseradish peroxidase), while the RREID-biot and RREID-

' Rabtes Un~t, Pasteur lnst~tute, Paris, France Lyssavirus Laboratory, Pasteur lnst~tute Par~s, France


lyssa employ an IgG blotin conjugate (botinylated rabies antinucleocapsid rabbit IgG conjugated with a streptavdin-peroxidase complex (7)) (Fig. 9.1). A yellow- orange colour appears after the addition of the substrate and o-phenylenediarnne (chrornogen); negative specimens are colourless. In both tests, the coloration can be evaluated qualitatively, with the naked eye (Fig. 9.2), or quantitatively, using a spectrophotometer (Fig. 9.3).

Safety precautions

The isolation of rabies nucleocapsld antigen should preferably be performed accordng to Biosafety Level 3, particularly when rabies-related strains are used ( 1 1 ) . All specimens and the positive antigen control should be regarded as potentially virulent, even after they have been inactivated by P-propiolactone or by heating. They should therefore be carefully handled with gloves and iluids should be collected in a recipient containing 5% sodium hypochlorite (bleach) solution. All chemicals should be used according to good laboratory practice and any contact w t h s k n or mucosae should be avoided.

Preparation of reagents

Rabies antinucleocapsid rabbit igG The techniques for producing and purifying rabies antinucleocapsid rabbit IgG directed against rabies serotype 1 (PV strain), serotype 3 (Mokola virus) and

Fig. 9.1 Principle of the RREID, RREID-biot and RREID-lyssa

RRElD RREID-lyssa

Y Y ? lmmunoglobulin G (IgG) directed against rabies virus (PV strain), Mokola virus and European bat lyssavirus 1 (EBL 1) nucleocapsid.

R A C Anti-nucleocapsid-biotin conjugate.

K Anti-PV nucleocapsid-peroxidase conjugate.

0 Rabies nucleocapsid (RNP).

-0 Streptavidin-peroxidase conjugate. WHO 94720

Fig. 9.2 Diagnosing rabies by the RRElD and RREID-lyssa (qualitative method)

Antigen distribution plan on the plates

Antinnn icnrntype)a

1 2 3 4 5 6 7 (2) (3) (4) (4) (1 (1)

l l \ l l -2

c n, .-

Ill

RREID RREID-lvssa

1 2 3 4 5 6 7 1 2 3 4 5 6 7

a Serotype 1 refers to rabies virus (PV strain); serotypes 2, 3 and 4 refer to Lagos bat, Mokola and Duvenhage viruses respectively.

b Refers to European bat lyssavirus 2.

107


Fig. 9.3 Diagnosing rabies by the RREID and RREID-lyssa (quantitative method)

IgG anti-PV -m- RNPPV --e RNPMok -A- RNP EBL

0.01 0.1 1 10 100 1000 RNP concentration (nglml)

PV = rabies virus (PV strain). Mok = Mokola virus. EBL = European bat lyssavirus.

European bat lyssavirus (EBLI) and conjugating the rabbit IgG to peroxidase or biotin are described in Appendix 2.

Sensitization of microtitration plates Microtitration plates with 96 wells divided into 6 strips of 16 wells each should be used. They should be sensitized as follows:

1. Add 1.0 pg (200 pi) of purified rabies antinucleocapsid IgG in carbonate buffer, pH 9.6, to each well. Incubate for 3 hours at 37°C.

2. Add 300 p1 of 0.3% bovine serum albumin and 5% sucrose in carbonate buffer, pH 9.6, to each well and incubate for a further 30 minutes.

3. Vacuum-dry the microtitration plates and store at - 20 "C. Alternatively, plates that are sealed under vacuum may be stored at 4°C.

Preparation of solutions The preparation of a stock solution of PBS-polysorbate, pH 7.4 for the washings is described in the Annex. The buffer for peroxidase substrate and the stopping solution (4 mol/l sulfuric acid) for the enzymatic reaction are made ready for use.

Preparation of control antigens 1. Remove the brains of uninfected (negative control) and rabies-infected (posi-

tive control) mice under aseptic conditions and homogenize in PBS, pH 7.0, 10- fold concentrate (PBS 10X).

2. Centrifuge the brain suspensions for 30 minutes at 2000 g and inactivate the supernatants by adding P-propiolactone to a final concentration of 1 : 4000 and incubate for 24 hours at 4°C. The control suspensions may be stored at - 20°C or (preferably) lyophilized in l -m l aliquots. The rabies strain used for

the positive control should belong to the same serotype as that against which the conjugate is directed.

Preparation of samples 1. Homogenize a sample of the brain stem, cortex and Ammon's horn from each

brain submitted for diagnosis at 30% (w/v) either in PBS 10X, pH 7.4, or in cell- culture medium at 4 "C (if virus isolation in cell culture is also performed on the same specimen; see Chapter 8). The test suspension may also be prepared from brain specimens collected by the retro-orbital or occipital foramen routes (12) (see also Chapter 4 and Appendix 2).

2. Centrifuge the suspension for 30 minutes at 2000 g. 3. Inactivate the supernatants by heating in a water bath for 2 hours at 56 "C.

Performing the test

Reconstitution of reagents 1. Open the vacuum pack containing the sensitized microtitration plates and take

out the required number of plates. Reseal the pack and keep it at 4°C. 2. Wash the strips5 times in PBS-polysorbate, pH 7.4 and dry on absorbent paper

at room temperature. 3. Reconstitute the negative and positive control antigens, if they are lyophilized,

using 1 ml of distilled water. 4. Dilute the required quantity of conjugate 1 : l 0 in PBS 10X, pH 7.4. 5. Prepare the substrate--0-phenylenediamine (chromogen) solution-just be-

fore use and keep t in the dark. 6. Place all the reagents at room temperature approximately 10 minutes before

starting the test.

lmmunocapture 1. Prepare a blank control for the photometric readngs by adding 200 p1 of

PBS-polysorbate-bovine serum albun~in (BSA) to the first well ( IA) of the microtitration plate (in some automatic plate readers, a blank control should be made in all the wells of the first line, 1A-IH).

2. Add 200 pi of the positive and negative control antigens to the second and third wells (1B and I C ) .

3. Add 200 pl of the supernatant of each sample to the remaining wells. 4. Seal the microtitration plate with adhesive plastic film and incubate for 1 hour at

37 "C. 5. Remove the film and aspirate the contents of each well into a recipient

containing 5% sodium hypochlorite (bleach) solution.


6. Fill the wells with washing solution (PBS-polysorbate). lnvert the microtitration plate and blot onto absorbent paper. Repeat the procedure 5 times.

RREID 1 . Add 200 p1 of peroxidase conjugate to each well 2. Seal the microtitration plate with adhesive plastic film and incubate for 1 hour at

37 "C. 3. Remove the film and aspirate the contents of each well. 4. Fill the wells with washing solution. lnvert the plate and blot onto absorbent

paper. Repeat the procedure 5 times. 5. Add 200 p1 of the substrate-chromogen solution to each well. Leave the

microtitration plate in the dark for about 20 minutes at room temperature. 6. Add 50 111 of stopping solution to each well. Do not expose the microtitration

plate to direct light before reading

RREID-biot and RREiD-Iyssa l. Add 200 pI of IgG-biotin conjugate to each well. 2. Seal the microtitraton plate with adhesive plastic film and incubate for 1 hour at

37 "C. 3. Remove the film and aspirate the contents of each well. 4. Fill the wells with washing solution. lnvert the plate and blot onto absorbent

paper. Repeat the procedure 5 times. 5. Add 200 p1 of IgG-biotin conjugate to each well and incubate for 30 minutes at

37 "C. 6 . Aspirate the contents and repeat step 4. 7. Add 200 p1 of substrate-chromogen solution to each well and leave the

niicrotitration plate in the dark for 10 minutes at room temperature. 8. Add 50 pl of stopping solution to each well. Do not expose the microtitration

plate to direct light before reading.

Interpretation of results

Reading with the naked eye The colour can be easily evaluated with the naked eye. The blank and the negative antigen control should appear colourless, while the positive antigen control should be yellow-orange. Specimens that show an orange coloration are considered positive, while those that are colourless are considered negative. This reading is often sufficient for diagnostic purposes.

Reading with a spectrophotometer Absorbency measurements should be taken within 30 minutes of stopping the reaction (see above, step 8). 1. Carefully wipe the bottom of the microtitration plate and place it in the

spectrophotorneter. 2. Determine the optical density (OD) at 492 nm of the blank(s), controls and

samples. 3. Calculate the absorbency by subtracting the OD of the blank from that of the

controls and samples. The absorbency of the negative and positive antigen controls should be below 0.1 and above 1.5 units respectively, otherw~se the test

is not valid. Samples that have an absorbency of more than 0.08 units (in the case of the RREID) or 0.1 units (RREID-blot and RREID-lyssa) above that of the negative control are considered positive.

Evaluation of the technique

The RREID has been evaluated in various epidemiological situations. When used for the routine diagnosis of rab~es on soine 10 000 specimens collected in France, the sensitivity and specificity of the technique were 98.5% and 99.996, respectively (3). The resiilts obtained under field corlditions by 17 laboratories in Africa, America, Asia and Europe also showed a good correalion, especially considering that the technique was being performed for the first time (2, 3) One major characteristic of the RREID is that its reliability is not affected by decomposition or heat inactivation of the specimen. Furthermore, testing by RREID of brain specimens collected through the occipital or retro-orbital forarnen routes provides a very useful tool for epidemiological surveys of rabies (12). Tlie detection limit of nucleocapsid antigen of lyssavirus serotype 1 is in the range of 0.8-1.0 ng;'ml for RREID Nevertheless, the sensitivity is lower when specimens infected with rabies- related strains are tested.

To further increase the sensitivity of the immunocapture of rabies antigen, an enzyme-labelled avidn-biotin method was developed. This amplified technique, called the RREID-blot, increases the detection limit of rabies nucleocapsid antigen (serotype 1) to 0.2 ng/ml without affecting the specificity Similarly, a mixture of rabbit IgG directed against serotype 1 (PV strain), serotype 3 (Mokola virus) and European bat lyssavirus (EBLI) can be used for the immunocapture and for the conjugate to increase the detectability of rabies-related strains. This technique, called the RRED-lyssa, increases the detection limit of nucleocapsid antigen to 0.2 ng,!ml, whatever the serotype of the strain considered (7).

In conclusion, the RREID and especially the RREID-blot and RREID-lyssa tests are excellent tools for the detection of all the currently known lyssaviruses relevant to public health. They are simple, rapid, sensitive, specific and economical Use of these tests can be recommended either for confirming the results of the fluorescent antibody test or for epidemiological surveys in association with simple methods of brain collection such as the retro-orbital or occipital foramen procedures (12). They can also be easily applied to the quantification of rabies antigen.

References

1. Perrin P. Rollin PE, Sureau P. A rapid rabies enzyme immunodiagnosis (RREID): a useful and simple technique for the routine diagnosis of rabies. Journal of biological standardization. 1986. 14: 21 7-222.

2. Perrin P, Sureau P. A collaborative study of an experimental kit for rapid rabies enzyme immunodiagnosis (RREID). Bulletin of the World Health Organization, 1987. 65: 489493.

3 Bourhy H et al. Comparative field evaluation of the fluorescent ant~body test, virus isolation from tissue culture, and enzyme immunodiagnosis for rapid


laboratory diagnosis of rabies. Journal of clinical microbiology. 1989, 27: 519-523.

4. Jayakumar R, Ramadass P. Raghavan N. Comparison of enzyme immunodiagnosis with immunofluorescence for rapid diagnosis of rabies in dogs Zentralblail fur Bakteriologie 1989, 271 : 501-503.

5. Saxena SN et al. Evaluation of the new rapid rabies immunodiagnosis tecli- nique, Indian journal of medical research, 1989, 89: 445-448.

6. Morvan J, Mouden JC, Coulanges P. Le diagnostic de la rage par methode ELISA. Son application a Madagascar: avantages et inconvenents. [Diag- nosis of rabies by the ELISA test. Application to Madagascar: advantages and disadvantages.] Archives de l ' l~~shtut Pasteur de Madagascar, 1990. 57: 193-204.

7. Perrin P et al. A modified rapid enzyme immunoassay for the detection of rabies and rabies-related viruses: RREID-lyssa, Biologicals, 1992, 20: 51-58.

8. Sokol F. Pur~fcat ion of rabies virus and isolation of its components. In: Kaplan MM, Koprowski H, eds. Laboratory techniquesin rabies, 3rd ed. Geneva, World Health Organization, 1973 (WHO Monograph Series, No. 23): 165-178.

9. Ataiiasiu P et al. lmmunofluorescence and irnmunoperoxidase in the diagnosis of rabies. In: Kurstak E, Morisset R, eds. V~ral immunodiagnosis. New York, Academic Press, 1974: 141-155.

10. Joustra M, Lundgren H. Preparation of freeze-dried, monomeric and mmuno- chemically pure IgG by a rapid and reproducible chromatographic techn~que. In: Peeters H ed. Prohdes of the biological fluids, (Proceedings of the 17th Colloquii~m.) Bruges, Arrtchap, 1969. 51 1-51 5.

11. United States Department of Health and Human Services Biosafety in inicrobiological and biomedical laboratories. Washington. DC, US Govern- ment Printing Office, 1984 (Department of Health and Human Services Publication No. (CDC) 84-8395)

12. Montano Hirose J, Bourhy H Sureau P. Retro-orbital route for brain specimen collection for rabies diagnosis. Vetennaiy record, 1991, 129. 291-292.

Annex Preparation of reagents

Carbonate buffer, pH 9.6 Sodium bicarbonate (NaHCO,) Sodlum carbonate (Na,CO,) Distilled water to make

Phosphate-buffered saline (PBS), pH 7.4, 10-fold concentrate (PBS IOX) Sodium chloride (NaC) 80.00 g Potassium chloride (KCI) 2.00 g Sodium phosphate, dibasic, dodecahydrate (Na2HP0,.12H20) 11.33 g Potassium phosphate, monobasic (KH,PO,) 2.00 g Distilled water to make 1000.00 ml

PBS-polysorbate, pH 7.4 PHS 10X

Polysorbate 20 Distilled water to make

PBS-polysorbate-BSA, pH 7.4 PBS 1OX Polysorbate 20 Bovine serum albumin (fraction V) Distilled water to make (pH adjusted to 7.4 using 4 mol,'l sod~um hydroxide)

Buffer for the peroxidase substrate, pH 5.6 (citrate buffer) Sodium citrate dihydrate 11.67g Citric acid, monohydrate 2.17 g Hydrogen peroxide, 309'0 solution (H,O,) 1 .OO m1 Distilled waier to make 1000.00 m1

Substrate-chromogen solution o-Phenylenediamne Citrate buffer (prepared as above)

CHAPTER 10

Cell culture of rabies virus A. A. King '

The rabies virus is highly neurotropic, but in the infected animal after replication in the central nervous system (CNS) it may spread centrifugally to most organs of the body, in which it is often able to replicate efficiently. It is not surprising, therefore, that the rabies virus can be cultivated in a wide variety of host cells. Such cultivation is extremely important, not only for obtaining knowledge about the virus itself. but also for producing large quantities of virus for the preparation of vaccine.

Susceptible cells, cell lines and strains

Tissue-culture techniques were first applied to the study of rabies virus in 1913 by Nogiichi ( 1 ) and Levaditi (2). Levaditi (2) also reported the first successful propagation of the virus in spinal ganglia maintained in a medium containing coagu- lated inonkey plasma. Ir 1930, Stoe (3) carried a strain of the virus through five passages in embryonic chicken brains and hearts explanted in rabbit plasma clots without loss of infectivity. The studies with primary cell expants did not, however provide much information about the kinetics of virus replication.

Atanasu et al. cultivated street and fixed strains of virus in non-neural cells of a mouse ependymoma tumour cell line and their work led to the first report of intracytoplasmic inclusions resembling Neyri bodies in a cell-culture system (4-6).

The susceptibility of a variety of primary kidney-cell cultures (of mouse, hamster, pig, dog and monkey origin) and avian embryo fibroblasts to infection with rabies virus demonstrated that these cells could potentially be used for rabies vaccine production. For example, a modification of the first rabies vaccine prepared in primary hamster kidney-cell culture (7). incorporatlng the Belling virus s l ran IS used for the preparation of human vaccine in China (8) (see also Chapter 30) The Vnukovo-32 strain is used for the production of a similar vaccine in the former USSR (9) (see also Chapter 31). In addition, vaccines for human use have been produced from fetal bovine kidney cells (10) dog k~dney cells ( 1 7 ) (see also Chapter 29) and chick embryo fibroblasts infected with Flury LEP C25 rabies virus (12) (see also Chapter 27)

Not un t~ l the development of several susceptible cell lines and cell strains, however was a systematic approach to the study of rabesvrus and its interaction with host cells made possible. There are now a number of continuous cell lhnes which are also used for the production of rabies vaccines for animals, such as baby hamster Itdney cells, line 21 (BHK-21) (13) hamster kidney fibroblasts (Nil-2) (14) and cliiclt embryo-related cells (CER) (15). Because of the heteroploid

' Raa~cs Hcsearcr~ Leader Ce~tral Vcternary Laboratory Ministry of Agriculture, Fisheries arid Food, New Haw W s y h r r ! g ~ S ~ ~ r r e y . Ergland

114

CELL CULTURE

characteristics and carcinogenic potential of these cell lines, however, only Vero cells have been considered suitable (or the production of human vaccines (76) (see also Chapter 26).

The production of vaccine in a human diploid cell line (WI-38) was first described in 1964 (17). Since then, other human diploid cell strains have been developed, such as human embryonic lung (HEL) (18) and human lung (MRC-5) (19) (see also Chapter 25). and a rhesus monkey diploid cell line vaccine has also been produced (20, 21) (see also Chapter 28).

Rab~es virus has been successfully propagated in several cell lines of poiklo-

thermic origin (22). It has also been propagated in embryonic chick myotubes, mouse macrophages, rat sensory neurons, rat pituitary turnout- cells, black goat

and guinea-pig kidney cells, fetal bat cells, human synovial fluid (McCoy) cells, Japanese quail embryo cells and Chinese hamster ovary (CHO) cells (see Table 10.1 for references), In addition, when the relationship between plasma membrane organization and cell susceptibility in vitro was investigated in mammalian, avian, fish and arthropod cell lines using CVS virus, specific viral antigen was detected in insect (Aedes albopictus) cells, although at a lower level than in the other cells tested (23).

Ne~iroblastoma cells of hurnari or murine origin are now widely used in rabies virus investigations. Mouse neuroblastoma (NA-C1300) cells (hypoxanthine- guanine phosphoribosyl transferase-deficient) are especially useful, since they share a number of characteristics with human neurons, including gross microscopic and fine-structural neuron-like morphology, and the presence of micro- tubular protein, neurotransmitter synthetic enzymes and electrically excitable cell membranes with acetylcholine receptors (24, 25).

Methods of virus propagation

Most conventional methods of virus propagation can be employed for rabies virus propagation in tissue culture Monolayers in stat~onary or revolving vessels and in suspension cultures have been used s~~ccessfully Some cell types require a specialized medium but for most purposes the propagation medium is based on Earle s salt solution supplemented with Eagles amino acids and vitaniins or with lactalbumin hydrolysate Many commercial companies will supply such media either ready to use or in concentrat~d or powder form The addition of serum appears to protect against thermal inactivation of the virus rather than provide necessary ingredients for virus replication (26)

Optimum growth is obtained at a low incubation temperature (32°C) (27) and during virus production the medium should be maintained at a basic pH (7 4-7 6) by the addition of extra sodium bicarbonate (26) Open culture systems also require the addition of extra carbon dioxide

In many cell systems considerable periods of adaptation and prolonged passaging rnay be required before substantial yields of virus can be obtained Diploid cells such as the normal human diploid fibroblast line WI-38 (28) are particularly difficult in this respect as exemplified by the original adaptation of the Pitman-Moore strain lo Wl-38 cells (17 22 29 30) Prolonged and regular serial passage of infected cells initially with the addition of fresh ~~n in fec ted cells was required before sufficient infectiousvirus was released into the culture supernatant fluid This virus was then used to infect naive cells Even so the yield of infectioiis


Table 10.1 Tissue-culture systems for rabies virus

Tissue-culture system Author(s) and year

Primary cultures Mouse kidney cells Hamster kidney cells Pig kidney cells C t i ck errit~ryo fibroblasts Duck embryo fibroblasts Dog kidney cells Monkey kidney cells Fetal calf kidney cells Cells of poikilothermic origin Ch!ck erribryo rnyotubules Mouse dorsal root ganglia Racoon kidney cells Skilnk brain cells, skunk kidney cells Japariese quail embryo cells Rat sensory neurons Rat dorsal root ganglia rat rnyotubules Rat tr igeninal ganglion neurinoma Rat primary cortical neurons Human dorsal root ganglia

Diploid cell lines Human ~ i n y W 3 8 Hdman embryondc lung (HEL) Rhesus monkey lung KLL-2 Humari u r l g MRC-5

Heteroploid cell lines Mouse ependymoma cells Baby hamster kidney cells line 21 (BHK-21 Chick embryo fibroblasts (CEF) Harnster kidney fibroblasts (NI-2) Chick embryo related (CER) cells Green monkey hdney cells line 4647 Vero monkey kidney cells Rabb t non- td rno~r cells Mouse neuroblastoma (NA-1300) cells Mouse neuroblastoma (NA-2) cells Black goat kidney cells Guinea-pig kidney cells

Fetal hat cells Rat tr igemina neurinoma (NGUK-I) cells Human synovial fliiid (McCoy) Humari rieuroblastoma (MR-32) cells Chinese hamster ovary (CHO) cells

Vieuchange et a l . 1956 (81) Kissling, 1958 (27) Abelseth. 1964 (88) Kondo. 1965 (89) Kondo, 1965 (89) Hronovsky et al., 1966 (90) Larig et a l . 1969 (91) Atanasiu et al.. 1974 (10) Wiktor & Clark, 1975 (22) Lentz et a 1 1982 (75) Tsiang et a l . 1983 (92) Urnoh & Blenden. 1983 (93) Umoh & Blenden, 1983 (93) Seroka et al.. l986 (94) Lycke & Tsiang, 1987 (78) Tsiang et al , 1989 (95) Yusupov et a1 , 1989 (96) Tsiang et a l . 199la (97) Tsiang et a l . 1991b (98)

Wiktor et al. 1964 (17) Hronovsky et al., 1973 (18) Walace et al., 1973 (20) Johnson et a l , 1985 (19)

Atanasio & Laurent, 1957 (4) lS13) MacPherson & Stoker, 1962 (13)

Yoshino et al., 1966 (32, 33) Diamond, 1967 (14) Smith et a l , 1977 (15) Aksenova et al., 1985 (99) Montagnon et al., 1985 (16) Sureau et al.. 1985 (100) Webster 1987 (43)

Barrat et a l , 1988 (101) Kim et al.. 1988 (102) Kim et al., 1988 (102)

Steece & Cal~sher. 1989 (103) Yusupov et al., 1989 (98) Consales et a l . 1990 (40) Conti et al., 1990 (104) Burger et al.. 1991 (105)

Lyrnphocytes

Mouse macrophages (p388D1) King et al.. 1984 (86)

CELL CULTURE

virus from Wl-38 cells was only about one-tenth of that prodiiced when BHK-21 and Vero cell I i~ies were used.

The quality of the cells at the time of initial infection is critical and in order to obtain the maximum yield of virus from monolayer cultures, it is essential to use those that are just confluent, having been passaged not more than 2-3 days previoilsly (31). Alternatively, high yields of virus can be obtained when cells in sirspension are infected and then allowed to form monolayers. High input muliiplicities of infection (MOI) rnay occasionally lead to autointerference (32-35), although this is not normally observed. Nevertheless, efforts to produce single growth curves in rabies-infected cells have not been particularly successful, possibly because virus replication proceeds relatively slowly and because ot the difficulty of producing synchronous cell populations (29, 30).

Cytopathology

In general no clearly definable cytopathological effect accompanies the production of rabies virus in cell cultures In infected BHK-21 cell monolayers for example

the cells merely begin to age and detach more quickly from the supporting surface than uninfected control cells (31) In monolayers of chick embryo fibroblasts however the viability of infected cells is sufficiently affected to permit plaque formation (32 33 35)

Similarly plaques were formed in a sublne (S13) of BHK-21 cells adapted to growth in agarose suspension (34) and plaque-forming viruses were recovered from persistently infected BHK-21 cells (36) This led to the development of the first reliable plaque assay system for rabesvirus Plaque formation was also induced in Vero cells infected with rabies virus and rabies-related (Lagos bat and Mokola) viruses (37) and a reproducible plaque assay system for rabiesviruses in CER cells was described (38)

In murine neuroblastoma cells infected with fixed rabies virus and rabies- related (Lagos bat Mokola and Duvenhage) viruses the cytopathological effects are frequently more severe than in BHK-21 cells Duvenhage virus induced the formation of syncytia (39) and kotonkan and Obodhiang viruses after adaptation could be titrated by plaque assay

In McCoy cells both fixed and street rabies viruses caused cytopathological changes from 24 to 72 hours after infection dependent upoii MO1 (40) The cytopathological effect was easily recognized and resembled that induced by vesicular stomatitis virus (VSV) infection Higher titres of the PV virus strain were found in supernatants of McCoy cells than in those of Vero cells

Persistent infection

Although rabies virus-infected cells can be maintained in cultore for extensive periods of time without noticeable cytopathological effect, persistent infect~on may give rise to different effects, depending upon the cell system used. For example, endosymbiotic infection, as found in rabbit endothelial cells infected with the CVS strain, was characterized by the accumulation of only small intracytoplasmic inclusions in the cells, with no apparent interference with the mechanism of cellular replication throughout a period of more than 2 years (41). On the other hand. in


persistent infections of Nil-2 cells, the production of new virus was observed to reach its peak in 3-4 days and then to fall off sharply during the iiext 6-7 cell transfers (42, 22) Fluorescent intracytoplasmic inclusion bodies remained present in all the cultured cells after the decline in virus production. The cultures became resistant to superinfectioii with VSV and an inhibitor with interferon-like properties was recovered from the culture mediiim

After this phase of infection, there was a marked decrease in the number of antigen-containing cells and in the production of infectious virus, followed again by a yradual increase In the proportion of infected cells and in the production of infectious virus. These cycles of high and low levels of rabies infection and corresponding resistance to challenge with VSV recurred periodically (26)

A somewhat different outcome of persistent infection has been described in NA-C1300 cells infected with a rabies virus of skunk origin ( 4 3 ) . Infected cells were subcultured every 3-4 days and the supernatants were monitored for the presence of iiifectious virus Eventually no infectious virus could be detected in the supernatant by cell culture or animal inoculatior? tests, even though 95-100% of the infected cells remained positive for viral nuceocapsd antigen A similar study using persistently infected BHK-21 C13 cells, which were passaged l 0 0 times without re-nfection, found that infectious virus continued to be released into the supernatant (44).

Following the discovery by von Magnus that serial undiluted passage of influenza virus prodiiced incomplete virus particles that interfered with the growth of infectious virus (45) a similar phenonienon was identified in cell cultures infected with rabies LEP Flury virus (46) These "defective interfering ' (DI) particles have since been described in many rabies cell systems and with many virus strains, suggesting that their replication is influenced by host-cell factors and the virus strain (47-49).

The particles are between one-third and two-thirds as long as infectious particles and require the presence of homologous infectious particles to undergo replication (49). They are readily produced: for example, detectable levels of DI particles could be generated in the first undiluted passage of a cloned pool of HEP Flury virus (50). They are probably irivolved in the establishment of persistent in

vitro infections (49. 51). Although long-term persistent infections with rabies virus in cell culture have

been studled In cons~derable detail so tar there 1s not a clear understanding of the effects of the evolutionary changes found in the viral RNA genome and proteins the role of DI particles the effect of temperature-sensitive mutants or the effect of changes in the synttiesis of vriis-specific RNA and proteins on the course of rabies irnfections in nature (52).

Virus in infected cells

Although inclusion bodies associated w ~ t h rabies infection of the brain cells were first described in 1903, it was not untrl much later that tliese "Negr~ bodies" were shown to contain a viral antigen (53) and viral particles were demonstrated within the ir1cli~sion bodies (54, 55). In a study using ferrilin-labelled antiserum, it was shown that slmilar inclusion bodes in BHK-21 cells, although containing virus particles, were largely composed of ribonucleoprotein (56).

CELL CULTURE

Conventional histolog~cal stalning techniques for rabies (see Chapters 4 and 5) have now beer1 largely replaced by the fl~iorescent antibody (FA) techniyiie which stemmed from the work of Coons & Kaplar? (57) and was later adapted for tlie diagnosis of rabies (53) (see also Chapter 7) The FA technlqiie has also become the method of choice for following the progress of infection in cell culture With fixed cells the antigen predominantly stained is the nucleoprotein (N protein) of the nucleocapsid whereas staining of unfixed cells reveals mainly viral glycoprotein (G protein) located on the cell plasma membrane (30) Although the FA test is a reliable and sensitive technique for demonstrating the presence of rabies virus in cells i t does not necessarily provide a measure of the amount of infectious virus within the ells or ot ther potential to release virus i r ~ i o the culture nledium

The introduction of monoclonal antibodies (MAbs) to the rabies field (58) (see also Chapters 11 and 12) l ias provided specific probes for the determination of antigenic variation among rabies and rabies-related viruses These antibodies are directed mainly against either the N protein (MAb-Ns) and are used on fixed infected cell cultures or the G proiein (MAb-Gs) for use in mouse neutralization tests When employed for the laboratory analysis of isolates in an integrated systeni of rabies case surveillance such probes can provide a great deal of information about the d~stribution prevalence and transmission of antigenic var- ants both within and between annial species (59)

Application of cell-culture methods

Diagnosis

A diagnostic capability that includes virus isolation and identification in cell cultures is highly des~rable, even in countries where rabies is not endeniic. Following numerous reports of the isolation of fixed and street viruses in BHK-21 cells, the finding in 1975 that these cells could be successfully i~Sed for both post- mortem and ante-mortem diagnoses and that results could be obtained far more quickly than with the mouse inoculation test was a fundamental step forward (60); in 1980 the use of these cells for routine diagnosis was proposed (61).

BHK-21 cells are now used in many diagnostic laboratories. They are hardy, require a relatively simple medium supplemented with inexpensive bovine serum. and since they do not require additlonal carbon dioxide to support growth, they are safer than open culture systems By regular (3-4 day) transfer of stock bottles at a split ratio of 1 :4, cells remain healthy and an 80-cmZ bottle at 3 days usually contains sufficient cells to give 100 ml of cell suspension. Culture bottles surplus to immediate requirements can be stored at room temperature or at 4°C and passage levels can be kept low by returning to these cells up to 3 weeks later, when after one or two passages they regain their rapid growtt? characteristics.

The method used at Weybridge over many years for isolating and identifying the rabies virus was established by an experiment in which a pool of BHK-21 cells treated with diethylaminoethyl-dextran (DEAE-dextran) was infected with the supernatant fluid from a 1056 suspensiori of infected braiti which had been vigorously shaken (not centrifuged) in a leakproof bottle and ef t for a few minutes to allow coarse particles to settle.


The mixture was divided between six 25-cm2 bottles and 12 Leighton tubes containing coverslips (used to monitor the adaptation process), which were then incubated at 35'C. Each bottle was used as a series of passages at 1 , 2 , 3 , 4 , 5 or 6 days, the cells in the bottles were trypsinzed and a portion (25%) was used to infect a further bottle and two Leighton tubes. The most rapid adaptation took place when infected cells were passaged either daily or every 2 days.

Following the adaptation of several hundred virus strains from many different species and geographical locations, some general observations can be made:

- at the first passage, many "Negri-body"-sized and smaller inclusions are observed in the cells;

- at the second passage, some cells contain large amorphous masses of iiucleocapsid antigen (presumed to be within infected cells transferred from the first culture), while others appear to be "newly infected";

- at the th~rd passage, a sim~lar picture is observed, except that at least 50% of the cells are infected. The addition of an excess of uninfected cells at this stage allows the preparation of microtitration plates for analysis using MAbs and of virus stocks. These stocks are frequeritly obtained in less time than is taken by the sanie inoculum to kill mice following intracerebral inoculation of street virus. Not all viruses adapt at the same rate, however, therefore passaging is continued cintil at least 50% of the cells are infected.

The potential of CER and mouse neuroblastoma cells for use in the diagnosis of rabies infection has also been evaliiated (15 62) Murne neuroblastoma cells have been shown to be more susceptible to rabies virus infection than any other cell lines tested and in some laboratories virus isolation in cell culture (with neuroblastoma cells) has replaced intracerebral inoculat~on of mice for diagnosis (63, 64) (see also Chapter 8)

Rabies virus assay

As already mentioned a number of plaque assays have been developed for the quantification and cloning of rabies virus The most widely used method is based on agarose-suspended BHK-21 S13 cells in which most culture-adapted fixed strains produce plaques after 5-7 days of incubation (26 34) The plaques obtained in chick embryo fibroblast (CEF) cell cultures were regarded more as proliferative foci than the type of dead cells of which plaques are usually composed (32 33) Other methods of plaque assay include BHK-21 cell monolayers with a cross Inked dextrari overlay (65) and pig kidney cell monolayers with a carboxymethylcellulose overlay (66)

Plaque assays however do not always give consistent results even when well- adapted fixed rabies virus stra~ns are used For virus titration they have been largely superseded by fluorescent focus assays in which cells are incubated with ser~al dilutions of virus then fixed and treated with fluorescent serum or MAb-Ns 1-4 days later In these tests titration end-points compare well with those obtained by intracerebral inoculation of mice (62) Similarly nhibltion of fluorescent focus forniation can be used to measure virus-neutralizing antibody levels in serum (see Chapter 15) This method has been extended to antibody-binding tests

for determining the potency of rabies vaccnes (67) (see also Chapter 43)

CELL CULTURE

Infected cell-culture fluids also include non-infectious virus particles and other virlon components. The complement fixation test can be used to measure total virus antigen and the haernagglutinaton test using goose erythrocytes can be used to measure intact viral particles. However, the latter test requires a concentra- tiot? of at least 106 plaque-forming units (PFU)/ml and can be carried out only in serum-free medium (68).

Pathogenesis

Cell culture of fables viruses can be used to study the processes of infection, transcription, translation, replication, assembly and release. In addition, many of the effects of the virus on cellular metabolism can now be exaniiiied at the molecular level (see Chapter 3).

Most of the earlier studies of rabies infection in cell culture employed thin- section electron microscopy to visualize how infection occurs, the fate of the virus once it has entered the cells, and the processes of morphogenesis and release of the virus. However, the majority of these studies were made in BHK-21 or CER cells, despite the fact that in the infected animal virus replication occurs predornnantly in nervous tissue. Stud~es of persistently infected NA-C1300 cells examined by scanning and freeze-fracture electron microscopy may provide a more accurate picture of what happens in natural infection (69).

The development of plaque techniques has allowed the selection of virus "clones" and mutants. The availability of suitable virus-neutralizing MAb-Gs has made it possible to select virus variants (70, 71) with altered pathogenic (70) or protective potential (72) whose antgenic changes can be subsequently mapped. A comparison of the cell-to-cell spread of pathogenic parenteral virus and non- pathogenic variants in vivo (in the brains of infected adult mice) and in vitro (in BHK-21 and neuroblastoma cells) demonstrated that the differences in the pathogencity of the two types of virus were correlated in both in vivo and in vitro systems (73). The differences between pathogenic and non-pathogenic virus infections were observed only in the neuroblastoma cells which, as mentioned above, share a number of characteristics with neurons and which may therefore serve as a more appropriate model for in vitro pathogenicity studies.

Despite intensive research, a specific cell receptor site for rabies virus has not yet been identified, although muscle spindles and motor end-plates in striated muscle have been shown to be involved ( 1 4 ) . The role of the acetylcholine receptor (AChR) has also been the subject of much debate, following the observation that antigen could be detected by the FA test at neuromuscuar junctions - an area rich in AChRs- shortly after rabies virus was added to mouse diaphragm muscle (75). However, tests In a number of cell-culture systems, some of which lack h g h - density AChRs, have indicated that this receptor is not necessary for rabies virus infection and that the suscept~b~lity of different cell types does not depend on a single specific type of receptor (76, 77).

A highly sophisticated compartnientalized technique has been developed for the culture of dorsal root ganglion cells This method allows the infection arid manipulation of neuronal extensions without exposing the neuronal sorna to the rabies virus (78). The h g h binding affinity of the v~rc~s to unmyelnated neurtes and its transfer by the neiirites to the neural sorna supports the view that, in vivo,


sensory nerves may be involved in the centripetal transfer of virus to the central nervous system.

Other determinants, however. may be involved in the attachment of rabies virus to cell surfaces. Studies using CER cells suggest that phospholipids (79) and carbohydrate moieties (80, 81) may play a role in the early interaction between the cell membrane and virus. It has also been postulated that sialylated gangliosides (82) and sialic acid (83) may be involved.

The possibility of an immunological involvement in at least some cases of rabies virus infection, such as the "early death" phenonienori has been explored (84). In a test using the P388DI mouse rnacropiiage cell line (85), it was shown that rabies virus antiserum diluted beyond its neutralization end-point enhanced the ability of the virus to infect these cells (86).

References

1. Noguchi H. Contributions to the cultivation of the parasite of rabies. Journal of experirnenta! medicine, 191 3. 18: 31 4-31 6.

2. Levaditi C. Virus rabique et culture de cellules in vitro. [Rabies virus and in vitro cell culture.] Cornptes rendus des seances de la Sociktk de Biologie et de ses f//iales. 191 3 , 75 : 505.

3. Stoel G. Syrnbiose du virus de la rage avec les cultures cellulaires. [Symbiosis of rabies virus with cell cultures.] Comptes rendus des seances de /a Societe de Biolog~e et de ses f~iiales, 1930, 104: 851-852.

4. Atanasiu P, Laurent C. Multiplication du virus de la rage des rues sur une cellule gliale (ependyrnorne de a souris) en culture de tissus; aspect histo- logique. [Multiplication of rabies street virus on a g la l cell (mouse ependy- rnoma) in tissue culture; histological appearance.] Comptes rendus de !'Academic des Sciences, Serie l/!, Sciences de /a vie, 1957, 245: 2562-2564.

5. Atanasiu P, Lepine P. Multiplication du virus rabique des rues sur la turneur ependyma~re de la souris en culture de t~ssus. Eftet cytolytique. [Multiplication of rabies street virus on mouse ependymorna in tissue culture Cytolytic effect.] Annales de I'lnslitul Pasteur, 1959, 96: 72 78

6 Atanasiu P Favre S, Collombier M Multiplication du virus de la rage fxe sur cellules gliales en culture et apparition d ~riclusons specif~ques ntracytoplas-

miques Application au diagnostic de la rage [Multiplication of rabies fixed virus on glial cells in culture and appearance of specific cytoplasmic inciusions Applicat~on to rabies d~agnosis ] Coinptes rendus de I'Academie des Sciences, Sene l//, Sciences de /a vie 1961 252 2029-2031

7. Fenje P, A rab~es vaccine from harnster kidney tlssue cultures, preparation and evaluation in animals. Canadian journal of n?icrobio!ogy, 1960, 6: 605-609.

8. Liri FT et al. The primary hamster kidney cell rabies vaccine: adaptation of

CELL CULTURE

viral strain, production of vaccine, and pre- and postexposure treatment. Journal of infectious diseases, 1983, 147: 467-473.

9. Selimov M et a1 Evaluation of the ~nactivated tissue culture rabies vaccine froni the Vnukovo-32 strain, results of its industrial production and field use for postexposure immunization of man. Developlnents in biological stand- ardizabon, 1978, 40: 57-62.

10. Atanasiu P. Tsiang H. Garnet A. A new rab~es vaccine for human use prepared in primary tissue culture. Annales de l'lnstitut Pasteur: Microbiology. 1974, 1258. 41 9-432.

11 Van Steeiiis G et al, lmmunogenicity of dog kidney cell rabies vaccine (DKCV). In: Kuwert E et al., eds. Rabies in the tropics. Berlin, Springer-Verlag, 1985: 172- 180.

12. Barth R et a1 A new inactivated tissue culture rabies vaccine for use in man. Evaluation of PCEC vaccine by laboratory tests. Journal of biological standardization. 1984, 12: 29-46.

13. MacPherson I, Stoker M. Polyoma transformation of hamster cell clones-an investigator- of genetic factors affecting cell competence. Virology. 1962, 16. 147-151.

14. Diamond L. Two spontaneously transformed cell lines derived from the same hamster embryo culture International journal of cancer, 1967, 2 : 143-152.

15. Smith AL et al, Isolation and assay of rabies serogroup viruses in CER cells. Intervirology, 1977, 8: 9 2 99.

16 Montagnon BJ. Fournier P. Vincent-Falquet JC. Un nouveau vaccin antirabique a usage human: rapport preliminaire. [A new rabies vaccine for human use: preliminary report.] In: Kuwert E et al., eds. Rabies in the tropics. Berlin, Springer-Verlag, 1985: 138-~143.

17. Wiktor TJ. Fernandes MV. Koprowski H. Cultivation of rabies virus in human diploid cell strain WI-38. Jourrial of immunology, 1964, 93: 353-366.

18. Hronovsky V. Cinatl J, Benda R, Cultivation of the fixed rabies virus strain HEP FIury 1. Adaptation to the human diploid cell line HEL. Journal of hygiene. epidemiology, microbiology and immunology, 1973, 17: 285-293,

19. Johnson SE, Pearson EW, Ing WK. Clinical responses in humans to rabies vaccine prepared in MRC-5 diploid cells from Canadian seed virus. In: Kuwert E et al., eds. Rabies in the tropics. Berlin. Springer-Verlag, 1985: 99-105.

20. Walace RE et al. Development of a diploid cell line from fetal rhesus monkey lung for virus vaccine production. 111 vitro, 1973, 8: 323-332.


21. Buryoyne GH et al. Rhesus diploid rabies vaccine (adsorbed). a new rabies vaccine using FRhL-2 cells. Journai of infectious diseases, 1985, 152, 204-21 0.

22. Wiktor TJ, Clark HF Growth of rabies virus in cell culture. In: Baer GM, ed. The natuial history of rabies, Vol. 7 . New York, Academic Press, 1975: 155-179.

23. Seganti L et al. Susceptibility of mammalhan, avian, fish and mosqu~to cell lines to rabies virus infection. Acta virologia, 1990, 34: 155--163

24. McMorris FA, Ruddle FH. Express~on of neuronal phenotypes in neuroblastoma cell hybrids. Developments in biology, 1974, 39: 226-246.

25. Cark HF. Rabies viruses increase in virulence when propagated in neuroblastoma cell culture. Science, 1978, 199: 1072-1075.

26. Wiktor TJ. Tissue culture methods. In: Kaplan MM, Koprowski H, eds. Laboiatory techniques 117 rabies, 3rd ed. Geneva, World Health Organization, 1973 (WHO Monograph Series. No. 23): 101-123.

27. Kisslng RE. Growth of rabies virus in non-nervous tissue culture. Proceedings of the Society for Experimental Biology and Medicine, 1958, 98: 223-225.

28. Hayfick L, Moorhead PS, The serial cultivation of human diploid cell strains. Expeiirnental cell research, 1961 25. 585-621.

29. Wiktor TJ, Clark HF. Chronic rabies virus infection of cell cultures, infection and immunity, 1972. 6: 988 995.

30. Clark HF. Systems for assay and growth of rhabdovirus. In: Bishop DHL, ed. Rhabdoviruses, Vol. 1. Boca Raton, CRC Press, 1979: 23-41,

31. Crick J, King A. Culture of rabies virus in vitro. In: Campbell J B Charlton KM, eds. Rabies. Boston, Kuwer Academic Publishers, 1988: 47-66.

32. Yoshino K, Taniguchi S, Arai K. Autointerference of rabies virus in chick embryo fibroblasts. Proceedings of the Society for Experimental Biology and Medicine, 1966, 123: 387-392

33. Yoshino K Taniguchi S, Arai K. Plaque assay of rabies virus in chick embryo cells. Archiv f i r gesamfe Virusforschung, 1966, 18: 370-373.

34. Sedwick WD, Wiktor TJ. Reproducible plaquing system for rabies, lympho- cytic choriomeningitis and other ribonucleic a c ~ d vlruses in BHK-21I13S agarose suspensions Journal of viioiogy, 1967, 1: 1224-1226.

35. Yoshino K, Morishima T. An improvement in the plaque assay of rabies virus in chick embryo cells. Aichiv fur gesarnle Virusforschung, 1971, 34: 40-50.

CELL CULTURE

36 Kawai A, Matsurnoto S lnterfer~ng and non-interfering defective particles generated by a rabies small plaque varlant vrus Virology 1977 76 60 71

37. Buckley SM, Tignor GH. Plaque assay for rabies serogroup viruses in Vero cells. Journai of clinical microbiology, 1975, l : 241-242.

38. Bussereau F, Flamand A. Pese-Part D. Reproducible plaqung syslem for rabies vrus in CER cells Journal of viroiogical methods, 1982, 4: 277-282.

39. Clark H F . Rabies serogroup viruses in neuroblastoma cells: propagation. "aulointerferer~ce" arid appareintly rando~n back-tnutat~on of attenuated viruses to the virulent state. lnfectiori arid immunity, 1980, 27: 1012-1022.

40. Consales CA et al. Cytopathic effect induced by rabies virus in McCoy cells. Journai of virolog~cal methods, 1990. 27: 277-285.

41. Fernandes M, Wiktor TJ. Koprowski H. Endosymbiotic relationship between animal viruses and host cells. A study of rabies virus in tissue culture. Jouniai of experimentai medicine, 1964, 120. 1099 1 1 16.

42. Wiktor TJ, Koprowski H. Cyclic appearance of viral inhibitor in tissue cultures chronically infected with rabies virus. [Abstract.] Bacter~ologicaiproceedings. 1967. p. 166.

43. Webster WA. A tissue culture infection test in routine rabies diagnosis. Canadian )ournai of veterinary science. 1987. 51. 367-369.

44. CokiC-Danjanovik J. Hronicno inficirane kulture celija virusom besnila. [Chronic infection of cirltured cells by rabies viruses.] Medicinski pregled, 1989, 42: 4 5 4 7 .

45. Vori Magnus P. Propagation ot the PR8 strain of influeriza A virus in chick embryos. Ill. Properties of the incomplete virus produced in serial passages of undiluted virus. Acta pathologica et microbioiogica Scandinavica, 1951, 29: 156 181.

46. Crick J , Browri F. An interfering component of rabies virus which contains RNA. Jourrial of general virology, 1974, 22: 147 151

47. Huarig AS. Defective interfering viruses. Arinual review of microbioiogy, 1973. 27: 101-1 17.

48. Huarig AS, Baltimore D. Defective ~nterfering animal viruses In: Fraenkel- Conrat H. Wagner RR, eds. Con7prehensive viroiogy, Voi 10. New York, Plenum Press, 1977: 73-1 17.

49. Holland JJ . Defective interfering rhabdoviruses. In. Wagner RR, ed. The rhabdoviruses. New York, Plenum Press, 1987: 297-360.


50. Kawai A, Matsurnoto S, Tanabe K. Characterization of rabies viruses recovered from persistently infected BHK cells. Virology, 1975, 67: 520-533.

51. Holland JJ, Villareal LP. Breindl M. Factors involved in the generation and replication of rhabdovirus defective T particles. Journal of virology. 1976, 17: 805-81 5.

52. Wunner WH. Rabies viruses-pathogenesis and inimunity, In. Wagner RR, ed. The rhabdoviriises. New York, Plenum Press. 1987: 361-426.

53. Goldwasser RA, Kisslirig RE. Fluorescent antibody staining of street and fixed rabies vaccine antigens. Proceedings of the Society for Experimental Biology and Medicine, 1958, 98, 21 9-223.

54. Matsurnoto S. Electron microscopy of nerve cells infected with street rabies virus. Virology 1962, 17: 198-202.

55. Matsurnoto S. Electron niicroscope studies of rabies virus in niouse brain. Journal of cellular biology, 1963, 19: 565-591

56. Humrneler K, Koprowski H. Wiktor TJ. Structure and development of rabies virus in tissue culture. Joornal of virology, 1967, 1 152-170.

57 Coons AH, Kaplan MM Localization of antigen in tissue cells. II. mprove- ments in a method for the detection of ar-it~gen by means of a fluorescent antibody. Journal of experimental medicine, 1950, 91 : 1 1 3.

58. Wiktor TJ, Koprowski H. Monoclonal antibod~es against rabies virus produced by somatic cell hybridization: detection of antigenic variants. Proceed- ings of the Nattonal Academy of Sciences of (he Urlited States of America, 1978, 75: 3938-3942.

59. Smith JS. Sanderlin DW. Yager PA. The application of monoclonal antibodies to epidemiologic studies of lyssaviruses. In: Thraenhart 0 et al.. eds. Progress in rabies coritroi. (Proceedings of the Second International IMVl Essen/WHO Symposium on "New Developments in Rabies Control': Essen, 5-7 July 1988 and Report of the WHO Consultat~on on Rab~es. Essen, 8 July 1988.) Royal Tunbridge Wells, Wells Medical 1988: 115-125.

60 Larghi OP et al. Sensitivity of BHK-21 cells supplemented with diethylaminoethyl-dextran for detection of street rabies virus in saliva samples. Journal of clfnical microbiology, 1975, 1 243-245.

61. Rudd RJ, Trimarchi CV, Abelseth MK. Tissue culture technique for rout~ne isolation of street strain rabies virus. Journal of clinical microbiology, 1980, 12 590-593.

62. Smith AL et al. Isolation of field rabies virus strains in CER and rnurine neuroblastorna cell cultures. lntervirology. 1978, 9: 359-361

CELL CULTURE

63 Porlno D, Favre S. Sureau P Use of neuroblastoma cells (MNB) for the isolation of street rab~es virus from fixed speclrnens Rabies information exchange 1982, 6 35-36

64. Webster WA, Charlton KM, Casey GA. Persistent infections of a field strain of rabies virus in murine ~ieuroblastoma (NA-C1300) cell cultures. Canadlan journal of veterinary research, 1989, 53. 445-448.

65 Schneider L. Cell monolayer plaque test. In, Kaplan MM, Koprowski H, eds. Laboratory techniques in rabies, 3rd ed. Geneva, World Health Organization, 1973 (WHO Monograph Series, No. 23 ) : 339 342

66 Strating A, Mitzel JR, Slater EA. Plaque assay of rabies virus on porclne kidney cell monolayers. Journal of cli~lical microbiology, 1975, 2. 126-128.

67. Barth R et al. The antibody-binding test, a useful method for quantitative deterrnlnation of inactivated rabies virus antigen. Journal of b~ologica/ standardization, 1981. 9. 8 1 89.

68. Kuwert E. Haemagglutination and t-iaernagglutination-inhibition tests. In. Kaplali MM, Koprowski H, eds, Laboratory techniques in rabies, 3rd ed. Geneva, World Health Organization, 1973 (WHO Monograph Series, No, 23): 135-1 46.

69. lwasaki Y, Mnamoto N. Scanning and freeze-fracture electron mcroscopy of rabies virus infection in murne neuroblastorna cells. Comparative immu- r?o/ogy, /lncrobio/ogy and infectious diseases, 1982, 5: 1-8.

70. Coulon P et al. Molecular bass of rabies virus virulence. 1. Selection of avirulent mutants of the CVS strain with anti-G monoclonal antibodies. dour-nal of general virology, 1982, 61. 97-100.

71. Lafon M Wiktor TJ, MacFarlan RI. Antigenic sites on the CVS rabies virus glycoprotein: analysis with rnonoclonal antibodies. Journal of general virology, 1983, 64: 843-851.

72. Wunner WH, Dietzschold B. Rabies virus infection: genetic mutations and the impact on viral pathogenic~ty and immunity. In: Cruse J M Lewis RE, eds. Antigenlc variation: molecular and genetic mechanisms of relapsing disease. Basel, S. Karger, 1987: 103 124.

73. Dietzschold B et al. Differences in cell-to-cell spread of pathogenic and apathogenic rabies virus in vivo and in vitro. Journal of virology, 1985, 56: 12-1 8.

74. Harrison AK, Murphy FA. Lyssavrrus infection of muscle spindles and motor end-plates in striated muscle of hamsters. Archives of virology, 1978, 57: 167-1 75.


75 Lentz TL et al Is the acetylcholine receptor a rabies virus receptor? Science, 1982 215 182-184

76. Reagan KJ. Wunner WH. Rabies virus interaction with various cell lines is independent of the acetylcholine receptor: brief report. Archives of v~rology, 1985. 84: 277-282.

77, Tsiang H. An in vitro study of rabies pathogenesis. Rullehn de l'lnstitut Pasteur, 1985. 83: 41-56.

78. Lycke E, Tsiang H. Rabies virus infection of cirltured rat sensory neurons. Journal of v~roiogy, 1987, 61. 2733-2741.

79. Superti F et al. Role of phosphoiipids in rhabdovirus attachment to CER cells. Archives of virology. 1984, 81 321 3 2 8 .

80. Conti C, Tsiang H. Effect of concanavalin A on the early events of rabies virus infection of CER (chick embryo-related) cells. lnterviroiogy 1985,24: 166-169.

81. Conti C, Superti F, Tsiang H. Membrane carbohydrate requirement for rabies virus binding to chicken embryo-related cells. Intervirology, 1986, 26: 164-168.

82. Superti F et al. Invovernent of gangios~des in rabies virus infection. Journaiof general virology, 1986, 67. 47-56

83 Tsiang H. Rabies virus itifection of myotubes and neurons as elements of the neurornuscular junction. Review of infectious diseases, 1988, 10. 733-738.

84. Blancou J, Andral B, Andral L. A model in mice for the study of the early death phenomenon after vaccination and challenge with rabies virus. Journal of general virology, 1980, 50: 433-435

85. Koren HS, Handwerger BS, Wunderlich JR. Identification of macrophage-like characteristics in a cultured rnurine tumor line. Journalof immunology, 1975. 1 14: 894-897.

86. King AA. Sands JJ, Porterfield JS. Antibody-mediated enhancement of rabies virus infection in a mouse macrophage cell line (P388D1). Journai ofgeiierai virology, 1984, 65: 1091 -1 093.

87 Vieuchange J et al. Affinite du virus rabique fixe poui les cellules d'or~gine renale ciiltivees in vitro. [Affinity of rabies fixed virus for cells of renal origin cultivated in vitro.] Bulletin de /P?cademie nationale de Medenne, 1956, 5 : 77 7 9 .

88. Abelseth MK. Propagatiori of rabies virus in pig kidney cell culture. Canadian vetertnary jourral 1964, 5: 84-87.

89. Kondo A. Growth ctiaracteristrcs of rab~es virus in primary chick embryo cells.

Vjrologji 1965, 27: 199-204.

CELL CULTURE

90. Hronovsky V, Benda R, Cinatl J. Propagaton of streei rabies virus in primary cultures of dog kidney cells, Acta virolog~ca, 1966, 10: 181

91. Lang R et al Adaptation d'crn virus rabique fixe aux cellules renales de singes. Utilisation de ce virus pour la preparation d'un vaccin antirabique inactrve. [Adaptation of a fxed rabies virus on monkey kidney cells. Use of this virus for the preparation of an inactivated rabies vaccine.] Comptes rendus de

IXcad6mie des Sciences, Skrie Ill, Sciences de la v i , 1969, 269. 2287-2290.

92 Tsiang H et al. Neurotropism of rabies virus, an in vitro study. Journal of rleuropatl~ology and experimental neurology, 1983, 42: 439-452.

93 Umoh JU, Blenden DC. Cornparrson of primary skunk brain and kidney and raccoon kidney cells with established cell lines for isolation and propagation of street rabies virus. Infection and immunity, 1983, 41: 1370-1372.

94 Seroka D, Karpnski S, Labunska E Adaptation of strains of the rabies virus Flury LEP and HEP to primary culture of fibroblasts of the embryo of Japanese quail (Coturnix coturnix japonica) Medycyna doswiadczalna I

mikrobioiogica, 1986, 38 228-233

95. Tsiang H et al. The anterograde transport of rabies virus in rat sensory dorsal root ganglia neurons. Journal of general virology, 1989, 70: 2075 2085.

96. Yusupov R et al. Relative efficacy of methods of isolating rabies virus. Veterinariya Moskva, 1989, No. 4: 27-29.

97. Tsiang H et al. Inhibition of rabies virus infection in cultured ,-at cortical neurons by an N-methyl-D-aspartate non-competitive antagonist MK801 Antimicrobial agents and chen~otherapy, 1991, 35: 572-574

98. Tsiang H. Ceccaldi PE, Lycke E. Rab~es virus irifection and transport in human sensory dorsal root ganglion. Journal ofgenerai vit-oiogy, 1991, 72 1191-1194.

99. Aksenova TA et al. Cultivation of the rabies virus in the continuous kidney cell line 4647 from the green marmoset. Voprosy vircisologii. 1985, 30 180-182.

100. Sureau P. Perrin P. Horaud F. A rabies vaccine produced in a non- tumorigenic rabbit cell line Developinents in bioiogicalstandardization, 1985, 60 133-139.

101. Barrat J et al. Diagnostrc de la rage sur culture cellulaire. Comparaison des resultats de I'inoculation au neuroblastome rnurin et de I'inoculation a la souris [Diagnosis of rabies in cell culture. Comparison of results of the inoculation of murlne neuroblastoma cells and rnoiise inoculation.] Com- parative immunology, microbiology and infectious diseases. 1988. 11 207-2 14.


102. Kim D H et al. Development of a new assay method for the ERA strain of fables virus. Research Reports of the Rural Development Administration - Veterin- ary - Korea Republic, 1988, 30. 60-64.

103. Steece US, Calisher CH. Evidence for prenatal transfer of rabies virus in the Mexican free-tailed bat (Tadarida brasillensis mexlcana). Journal of id id life diseases, 1989, 25. 329-334.

104. Conti C et al. Effect of inhibitors of cytoplasmic structures and funct~ons on rabies virus infection in vitro. Comparative immunology, microb~ology and infectious diseases, 1990, 13: 137-1 46.

105. Burger SR et al. Stable expression of rabies virus glycoprote~n n Chinese hamster ovary cells. Journal of general v~rology, 1991, 72: 359-367.

Special diagnostic and research techniques

CHAPTER 11

Techniques f ~ r the production, screening and characterization of monoclonal antibodies M ia fon '

In 1975. Kohler & Mlstein described the technique for the production of monoclonal antibody-secreting cell lines (hybridomas) formed by fusion of myeloma cells with B lymphocytes ( I ) . Since then, monoclonal antibodies (MAbs) have been widely used in Studying the immunology and epidemiology of viruses, particularly the rabies virus, Immortalization of B cells is achieved by fusion with myeloma cells and by using a selective medium (see below) in which liybridomas distinct from rnyeloma cells can survive.

This chapter describes a technique for producing mouse rnonoclonal hybri- dornas. The technique Inay also be used to prepare human or rat hybridonias by using rat myeloma, human myeloma or heteromyeloma cells (resulting from the fusion between human B cells arid myeloma cells of rodent origin) cultivated in appropriate media, as described elsewhere (2, 3). (For details of the safety precautions required for prepariiig MAbs directed against rabies virus, see Chapter 1 . )

In a well-equipped cell-culture laboratory, MAbs can easily be produced once'

animals are properly immunized: -- mycoplasma-free myeloma cells having good viability are available; - sufficient antigen is available to screen supernatants of hybridomas and

ascites.

Immunization of animals

The animal (mouse) is primed with at least three injections of rabies antigen given intramuscularly subcutaneously or intraperitoneally every 7-10 days A booster injection is then given intravenously 3-4 weeks later

Rabies vaccine (1 dose per mouse) or sucrose-purified rabies virus (100 pg per mouse) can be injected without adjuvant However purified rabies proteins peptides or infected cells should usually be administered with adjuvant e g Freund s adjc~vant For the first injection the antigeii (5-10 pg of rabies protein or 20 X 10" infected cells) should be mixed with complete r ieund s adjuvant and injected subcutaneously Incomplete Freund S adjuvant must be used for intramuscular inoculations The intravenous booster is always administered without adjuvant

Fusion is carried out 3-4 days afterwards Antibody titres specific for the antigen can be monitored by enzyme imrnunoassay or the ne~itralization test on serum samples collected from the retro-orbital plexus of animals

' Laboratory Head. Rabies U q i t Deoarlrnent of Virology, Pasteiir nstilulc. Paris France

133


Myelomas

Myeloma cells used for the production of MAbs are deficient in the enzyme hypo xanthne guanlne phosphoribosyltransferase (HGPRT) Unlike lymphocytes and hyblidomas these cells are unable to grow in selective hypoxanthine- aminopterin-thymidine (HAT) medium To avoid myeloma revertants no longer susceptible to aminopterin rnyelomas should be treated once per month with 8- azaguanine a poison which is incorporated by HGPRT in DNA and kills cells that do not show HGPRT deficiency

Sublines of myelomas that do inot produce heavy or light chains (such as (SP,-0) are preferred to producer myelomas (such as X63) in order to facilitate screening and purification procedures

Myelomas are maintained in Dulbecco s modified Eagle s medium (DMEM) (see Annex) supplemented with 10 ml of 100 mmol'l sodium pyruvate 10 m1 of 200 nimol l L-glutamine 4 5 g of glucose and 1 X 105 IU of gentamicin per litre and 10% fetal calf serum (DMEM-1OFCS) in a humid incubator with 5% carbon dioxide (CO,) One week before fusion the cells are transferred to DMEM contain~ng 10% horse serum to improve the eff~ciency of fusion The cells are maintained in stationary suspension culture at concentrattons of 1 X 106 cells/ml and 5 X 106 cells/ml for fusion Only cells in a logarithmic phase are suitable for fusion

Fetal calf sera

Fetal calf sera (FCS) are screened for their ability to support growth of hybridomas and myelomas and to form contiguoiis growth sheets in plates, while horse sera (HS) are screened for their ability to support fusion.

All sera are heated at 56'C for 30 minutes in order to inactivate coniplement.

Fusion

The technique is described using splenocytes as the source of B lymphocytes. However, fusion can also be performed with lymph node or peripheral blood lymphocytes.

The PEG 1500 (polyethylene glycol 1500 or macrogo1)-mediated fusion technique is the most commonly used method. Alternatively, electrofusion can be applied ( 4 ) .

Materials

Phosphate-buffered saline (PBS).

FCS and HS. 0 DMEM without serum. 0 DMEM with 20% FCS: 9 ml of DMEM-l0 FCS, 1 m! of FCS. r, Hypoxanthine-aminoptern-thymidine (HAT) medium, 50-fold concentrate

(ready to use). a DMEM-10 FCS-HAT: 98 ml of DMEM 10 FCS, 2 ml of HAT, 50-fold concentrate. 0 DMEM-l0 FCS-HAT, 2-fold concentrate: 96 m1 of DMEM-I0 FCS, 4 ml of HAT,

50-fold concentrate.

PRODUCTION OF MONOCLONAL ANTIBODIES

a PEG 1500 in 75 mmol/l hepes. a 96-well sterile flat-bottom microtitraton plates. a Conical-bottom centrifuge tubes, 15 nil and 45 ml. a Stock solutions: - 0.16 molll ammonium chloride (NH,CI): 8.3 gllitre. - 0.17 molll trometamol, pH 7.65: dissolve 20.6 g of trometamol base in 900 ml

of distilled water, adjust pH to 7.65, and make to 1000 ml. e Working solution: mix 90 ml of 0.16 mol/l NH,CI and 10 ml of 0.17 mol/l tro-

metamol, pH 7.65. Adjust pH to 7.2 with dilute hydrochloric acid (HCI). Filter through a 22-nm membrane.

All media should be incubated at 37°C for 1 hour

Method

1. Kill the animal by dislocating the neck and immerse the body in 70% ethanol. 2. Open the skin of the abdomen. Wash the exposed tissue with 70% ethanol.

Open the peritonea1 cavity. 3. Remove the spleen and transfer it into a Petri dish containing 10 ml of PBS and

a sterile metal grid. 4, Disrupt the spleen by pressing it through the grid. 5. Transfer the cell suspension into a 15-m centrifuge tube. Centrifuge the spleen

cells at 200 g for 10 minutes and wash them twice in PBS. 6. Remove the red blood cells by resuspending the pellet in 5 m1 of sterile NH,CI

(haemolysing solution). Allow to stand at room temperature for 5 minutes. 7. Add 7 ml of FCS, centrifuge at 200 g for 10 minutes rid wash twice with DMEM

without serum. 8. f?esuspend in 10 m1 of DMEM without serum. Count the number of spleen cells

(approximately 80-100 X 106). 9. In the meantime, centrifuge the rnyeloma cells grown in DMEM-l0 HS at 200 g

for 10 minutes and wash them twice with DMEM without serum. Resuspend in 10 m1 of DMEM withoc~t serum and count.

10. Mix the myeloma cells and spleen cells in different ratios: 1 : 1, 1 : 4 and 1 : 10. 11. Centrifuge the myeloma cells and spleen cells together at 200 g for 5 minutes

in a 45-rnl conical centrifuge tube. 12. Discard the supernatant and gently tap the tube to spread the pellet. Add 1 m1

of PEG 1500 over a per~od of 1 minute and mix by gently tapping the tube. DO not resuspend the cells with a pipette.

13, Incubate the tube at 37°C for 2 minutes. 14. Add 1 ml of DMEM without serum. Do not mix. 15. Add 2 ml of DMEM without serum arid mix by gently shaking the tube. 16. Add 4 ml of DMEM-20 FCS and mix with a pipette. 17. Add 20 m1 of DMEM-10 FCS and centrifuge at 200 g for 5 minutes. 18. Discard the supernatant. Resuspend the pellet ir 100 rnl of DMEM-I0 FCS and

distribute 100pl of the cell suspension into each well of the m~crotitration plates.

19. Incubate the plates for 24 hours at 37 "C in a humid incubator with 5% CO,. 20. Add 100 p1 of DMEM-10 FCS-HAT, 2-fold concentrate, to each well.


21 Replace 100 111 of supernatant with l 0 0 JLI of fresh DMEM-l0 FCS-HAT twice a week

22 After 2 weeks screen the culture supernatants for the presence of MAbs with the desired specificity Positive cell clones should be expanded in DMEM-10 FCS-HAT deep-frozen, cloned and eventually inoculated into mice for the production of ascites (see page 140)

23 Well established cultures should be maintained in DMEM-10 FCS

Screening of hybridorna supernatants

Enzyme immunoassay

Enzyme minunoassay (EIA) is the most convenient test for screening hybridoma supernatants because Iianogram amounts of antibody can be detected and up to 100 samples can be tested at the same time

Microtitraton plates coated with the immunizing antigen are incubated with supernatants of hybr~doma colonies Mouse MAbs are detected with specific sera conjugated to peroxidase

Materials e 96-well microtitraton plates.

Carbonate-bicarbonate buffer, 0.05 rnol,'l, pH 9.5. Antigen (rabies vaccine, purified rabies virus or purified rabies proteins). PBS-polysorbate 20, 0.05 rno/I.

e PBS-polysorbate 20, 0.05 molil, supplemented with 10% normal serum of the species in which anti-mouse IgG was prepared (blocking medium). I g G biotin conjugate (biotinylated rabies antinucleocapsd mouse IgG conjugated wrth a streptovidin-peroxidase complex). 2,2-Azno-di-3-ethyl-benzthiazolin-sulfonate (ABTS) and substrate buffer for ABTS.

a Positive and negative controls. (In the absence of rabies-specific MAbs, rabies- specific mouse serum can be used as positive control.)

Method 1 . Sensitize the microtitration plates by adding 200 ng (100~11) of antigen in

carbonate-bicarbonate buffer, 0.05 mill, pH 9.5. to each well. Seal with adhesive film. Incubate overnight at 4°C.

2. Empty the plates and wash once with PBSpoIysorbate 20. 0.05 mol/l 3. Add 300 i l l of blocking medium to each well and incubate the plates for 30

niinutes at 37'C in a humid incubator with 506 CO,. 4. Empty the plates and add 100pl of an appropriate dilution of the IgG-biotin

conjugate to each well. lncubate for 30 minutes at 37" in a humid incubator with 5% CO,.

5. Enipty the plates and wash 5 times with PBS-polysorbate 20, 0.05 mol/l. 6. Add 100 p1 of an appropriate dilution of peroxidase conjugate to each well and

incubate for 30 minutes at 37°C in a humid incubator with 556 CO,. 7. Repeat step 5. 8. Add 100 i l l of ABTS to each well and incubate the plates for 15 minutes at room

temperature.


9. Place the plates in a spectrophotometer. Determine the optical density at 405 nm of the blank(s), controls and samples.

Indirect fluorescent antibody staining

The specificity of antibodies secreted by hybridomas can be determined by indirect

fluorescent antibody staining of infected cells. MAbs directed against the nucleocapsid (MAbs-Ns) can be distinguished from those directed against the glycoproten (MAb-Gs) according to the localization of rabies antigens in infected cells. G proteir-i is the major surface antigen, whereas N antigens accumulate only in the cytoplasm. Non-fixed cells with intact plasma mernbraries are only stained by MAb-Gs (so-called membrane fluorescence). On the contrary. acetone-fixed cells with solubilized membranes are stained by MAb-Gs and by MAb-Ns (so-called nucleocapsid fluoresceiice).

Therefore, supernatants that stain only acetone-fixed cells correspond to hybridomas secreting MAb-Ns, and those that stain fixed and non-fixed infected cells correspond to MAb-Gs. MAbs directed against the matrix protein (MAb-M2s) also stain the surface and the cytoplasm of infected cells. However, they can be distinguished from MAb-Gs as the fluorescence is dff i ise (so-called cytoplasmic fluorescence).

Infected cells can be prepared in humari leukocyte antigen (HLA) plates, tissue culture-chamber slides or 96-well microtitration plates. The technique is described with HLA ~ l a t e s .

Materials HLA plates. Rabies virus Fibroblast cells (BHK-21 BSR or CER), trypsinized and adjusted to a concentration of 1 X 106 cells, ml Eagle's minimum essential medium (see Chapter 8, Annex 1) supplemented with gentamicin glutamine and 8% heat-nactivated FCS (EMEM-8 FCS) (culture medium for fibroblast cells and rabies virus). Anti-IgG mouse immunoglobulins conjugated to fluorescein. Acetone (add 20 ml of distilled water to 80 m of acetone, store at 4'C)

Method 1. Seed the HLA plates with a suspension of fibroblast cells and virus (10 ~ L I per

well) prepared with 100 pI of diluted virus and 400 ~ t l of trypsinized cells (cell concentration 1 X 106 cells/ml), The virus dilution should be appropriate to obtain, under the conditions described above, 8056 of infected cells in 24 or 48 hours.

2. Close the lid of the plate tightly and incubate at 37°C in a humid incubator with 5% CO,.

Preparation of acetone-fixed plates:

1. Fill the plate with 10 ml of PBS and empty. 2. Wash once with 10 ml of acetone


3. lncubate the plates with 10 ml of acetone at room temperature without the I d for 30 minutes

4. Empty the plates and allow the acetone to evaporate.

Preparation of non-fixed plates:

1. Fill the plates with 10 ml of PBS and empty gentiy. 2. Remove any excess PBS with a tissue. Do not allow the non-fixed cells to dry out.

Staining of fixed and non-fixed cells:

1. Put 10 p1 of each superriatant of hybridoma in two wells of each fixed and non- fixed plate. Include positive and negative controls. Close the plate lid tightly.

2. Incubate the plate for 30-60 minutes at 37°C in a humid incubator with 5% CO,.

3. Wash the plates by immersing them carefully in PBS. Handle non-fixed plates very gently. Do not allow the non-fixed cells to dry out.

4. Add 10 /11 of fluorescein-conjugated ant-mouse IgG serum to each well. 5. Incubate for 30 minutes in a humid incubatoi at 37°C with 5% CO,. 6. Repeat step 3. Examine the plate under ultraviolet light. 7. Score for positive wells. Record the intensity and nature of fl~rorescence

(membrane, cytoplasmic or nucleocapsid).

The same procedure can be used or? acetone-fixed uninfected cells in order to detect MAbs directed against specific cellular components.

Neutralization assay

An in vitro neutralization test can discriminate between neutralizing and non- neutralizing MAbs by those which stain the surface of ~nfected cells, usually G protein-specific. This test requires at least 0.5 ml of hybridoma supernatant and can only be used after the cultures are expanded Virus titres in the presence of hybridoma supernatant and In the presence of culture medium are compared. A 10-fold or greater reduction in virus titre is considered as evidence of neutralization. Results are expressed in terms of the neutralization index, which is the difference between the logarithmic titres of virus in the presence of hybridoma supernatant and In the presence of medium. Hybrdoma supernatant contains neutralizing antibodies when the index is 1.0 or more.

Materials Microtitration plates.

0 Rabies virus. 0 Fibroblast cells (BHK-21, BSR or CER), trypsirized and adjusted to a concentra-

tion of 1 X 106 cells/ml. 0 EMEM-8 FCS (culture medium for fibroblast cells and virus) e Positive control: neutralizing antibody or serum. e Negative control: culture medium or supernatant of non-producer myeloma

cells. e Acetone (80 rnl of acetone, 20 m1 of dstilled water). e Rabies-specific fluoresceiri-conjugated IgG


Method

1. lncubate 50 /LI of hybridoma supernatant with 50 pi of serial 10-fold dilutions of virus (e.g. 1 :10, 1 : 100, 1 : l000 and 1 : 10000) in duplicate, for 60 minutes at 37°C in a humid incubator with 5% CO,, Include positive and negative controls.

2. Add 50 pi of the fibroblast cell suspension (1 X 106 cells/ml) to each well. 3. Incubate for 3 days at 37°C in a humid incubator with 5% CO,.

4, Fix the plates with acetone (see page 137). 5. Add 10 p1 of fluorescein-conjugated anti-mouse IgG to each well. 6. lncubate for 30 minutes in a humid incubator at 37'C with 5% CO2. 7. Wash the plates by immersing them carefully in PBS 8. Examine the plates under ultraviolet light. 9. Score the percentages of infected cells and calculate the neutralizing index.

Cloning of hybridomas by limiting dilution

Materials

DMEM-20 FCS. Supernatants of myeloma culture (conditioned medium), previously centrifuged and filtered to remove myeloma cells. Conditioned medium can be stored frozen.

0 96-well microtitration plates.

Method

1. Prepare the cloning medium (10 m1 per plate) by mixing 1 part of conditioned medium and 2 parts of DMEM-20 FCS.

2. Prepare suspensions of hybridomas containing 300, 30 and 3 cells per ml in cloning medium

3. Distribute 100 p1 of hybridoma suspension to each well of the microtitration plate. Set up at least one plate for each hybridoma concentration.

4. Incubate for 10-15 days at 37°C in an incubator with 5% CO,. 5. Score for positive hybridomas.

Hybrdomas secreting relevant antibodies are expanded in a second 96-well plate and then transferred to a 24-well plate. To ensure that they are monoclonal. hybridoma cells can be cloned again following the same protocol.

Production of large amounts of monoclonal antibodies

After positive hybridomas have been cloned and frozen in liquid nitrogen. they may be propagated in bulk in order to produce large amounts of MAbs. Two procedures are possible: production of ascitic fluid by injection of hybridomas into the peritonea1 cavity of syngeneic mice, or production of large volumes of hybridoma supernatants of cell cultures. The protein content of ascitic fluid precipitated by ammoniuni sulfate is 10-30 mg per ml, whereas cell-culture supernatants are usually 100-1000-fold less concentrated. Protein impurities in cell-culture supernatants can be largely reduced by the use of serum-free medium, whereas ascites (even those precipitated by ammonium sulfate) often contain other immunoglobulins of the same subclass.


Cell-culture supernatants

Hybridoma cells are inoculated into cell cultures (stationary, revolving or suspension) at a concentration of 2-5 X 106 cells per ml. After 6 days, the cell-culture supernatants are centrifuged at 200 g for 10 minutes, tested for desired antibody specificity and frozen at - 20°C. FCS (2-596) is added in order to facilitate purification procedures.

Ascitic fluid

1 . Inoculate syngeneic 12-20-week-old mice intraperitoneally with 0.5 ml of 2,6,10,14-tetramethylpentadecane at least 1 week before the hybridoma cells are injected. It is preferable to use females that have had several litters.

2. Inoculate the mice intrapertoneally with 2 X 106 hybridoma cells in 0.5 ml of saline buffer.

3. Tap mice with a distended abdomen with a 1.2 X 40-mm hollow needle (usually 10-20 days after the hybridoma injection). A local anaesthetic should be administered before this procedure is carried out. Repeated tapping may be carried out before animals are killed.

4. Remove the cells by low-speed centrifugation at 200 g for 10 minutes and remove the fibrin with a pipette.

5. Check individual samples for relevant antibody activity. Mix samples of similar titre. Discard any samples of low activity.

6. Store the samples at - 20cC or - 8OCC. 7. Precipitate the immunoglobulins using ammonium sulfate (see Appendix 2,

page 436) and dialyse. Determine the protein content by colorimetry (5).

Freezing and thawing of hybridomas

Materials

Sterile dimethyl sulfoxide (DMSO) 20% FCS. 2-ml cryotubes. Isotherm box. Freezer, set at - 80°C. Liquid nitrogen container. 24-well tissue-culture plates.

Method

Hybridoma clones or bulk colonies can be frozen and thawed successfully by the following procedure.

Freezing method Note: Only healthy cells in mid-log phase culture should be frozen

1 . Centrifuge the hybridoma cells at 200 g for 10 m~nutes. 2. Resuspend the cells with a pipette in 90% FCS and 10% DMSO to give a final


concentration of 2-5 X I Q 6 cells per ml it is preferable to use a glass pipette for this step

3 Distribute the cell suspenslon in l -ml amounts into labelled cryotubes Place the tubes In an isotherm box and store for 1-3 days at - 80°C

4 Transfer the cryotubes in the liquid nitrogen container

Thawiiig melhod 1 Place the cryotubes in a warm (37 C) water bath until the contents have

thawed 2 Rinse the tubes with 70% alcohol before opening, and dilute the contents to

10 m with warm culture rnedum 3 Centrifuge the cell suspenslon at 200 g fo r 10 minutes and wash twice with PBS

suppleniented with calcium and magnesium ions to remove DMSO 4 Resuspend in 6 m of medlum supplemented with 20% FCS Distribute l ml of

the resulting suspension into a 24 well tissue-culture plate and incubate at 37 "C for 3 days

Note: Any hybridomas that did not reconstitute efficiently can be used, immediately after thawing, in the production of ascitc fluid (see above) After collection, hybridoma cells are plated in Petri dishes or in 6-well tissue-culture plates for 24 hours. Non-adherent cells are transferred to a new plate and su~ernatants checked for relevant antibody activity.

Characterization of monoclonal antibodies

lsotyping

Identification of antibody isotype is important for the interpretation of immunological assays such as immune lysis, binding to Staphylococcus aureus protein A or the coinpetitive binding assay. Accurate isotyping for heavy and light chains can be performed by the enzyme immunoassay (EIA) with anti-isotype mouse antibody preparations. A number of EIA kits are available commercially, most of which use typlng sticks spotted with antibodies specific for the different immunoglobulin chains (e.g IgG,, IgG,,. IgGzb, IgM, K and 2 ) . The sticks are incubated with the hybridoma supernatants or diluted asctes fluid, and then treated with peroxidase-labelled goat anti-mouse antibody.

lmmunoprecipitation

lrnmunoprecipitation is the most widely used method to determine the specifcity of antibodies secreted by hybridoma cells.

Materials c Rabies antigen (sucrose-purified rabies virus or lysale of infected cells), labelled

by supplementing the vircs culture medium with 14C-labelled amino acids or 35S-labelled methionine (3-5 pCi per ml). Lysis buffer: 0.5 mol,'l t rometamol HCI, pH 7.8: 1 mol: NaCl; 1 C% octoxinol; 0.5% sodium deoxycholate; 100 units per ml of aprotinin (protease inhibitor). Staphylococcus aureus protein A


Gel electrophoresis buffer containing sodium dodecyl sulfate (SDS) and niercaptoethanol (3).

Method 1 , Incubate rabies virus-infected lysate containing 6 x 106 counts per minute

(cpm) in 100 pI of lysis buffer and 2 LLI of ascitic fluid at 4°C for 18 hours. 2. Add 50 { L of 10% (v/v) protein A and incubate for a further 2 hours. 3. Rinse 3 times in lysis buffer, and then add 100 p l of gel electrophoresis buffer

(see above). Incubate at 95 "C for 3 minutes. 4. Cerltrifuge the suspensiori at 2000 g for 15 minutes arid collect the super-

natants. 5. Separate the immunoprecipitated antigens by SDS-polyacrylamide gel electro-

phoresis (SDS-PAGE).

For further details of this technique, see reference 3

lmmunoblotting

While immunoblotting does not require labelled rabies antigens, it is restricted to MAbs that recognize epitopes resistant to denaturation by SDS. Rabies virus proteins or the Iysate of infected cells are separated on SDS-PAGE and transferred onto a nitrocellulose membrane. Nitrocellulose strips are incubated with hybridoma supernatants. Bound antibodies are detected with an appropriate secondary antibody and chromogen. For further details of this technique, see reference 6.

Use of monoclonal antibodies

MAbs can detect even minor antigenic differences among closely related antigens. Therefore, they have been widely used in identifying rabies virus strains and in diagnosing rabies in humans and animals (see Chapter 12). The detection of virus strains resistant to neutralization by murine neutralizing MAbs ( 7 , 8) has enabled functional antigenic maps of rabies G protein to be constructed (9-11). MAbs have also been used to locate antigenic sites on the rabies virus N protein ( 1 2 ) . Candidate vaccines have been prepared using anti-idiotype MAbs directed against antigenic determinants in the rabies protein (13) Possible replacement of human rabies immunoglobulin (HRIG) or equine rabies immunoglobulin (ERIG) by MAbs has been investigated in hamsters and rnice treated after infection with field strains of rabies and lyssaviruses; preliminary results appear to be promising ( 14, 15).

References

1 Kohler G, Milstein C Continuous cultures of fused cells secreting antibody of predefined specificity Nature, 1975, 265 495.

2 Ueki Y et al. Clonal analysis of a human antibody response Quantitation of precursors of antibody-producing cells and generation and characterization of morioclonal g M , IgG and IgA to rabies virus. Journal of experimental medicine, 1990, 171 : 19-34.


3 Lafon M et a1 Human monoclonal antibodies spec~frc for the rabies virus glycoprotein and N protein Journal of general virology 1990 71 1689- 1696

4. Zimmermann U Scheurich P. High frequency fusion of plant protoplasts by electric fields. Planta. 1981, 151. 26 32.

5. Lowry OH et al. Protein measurement with the Folin phenol reagent. Journalof biological chemistry, 1951. 193: 265-275.

6. Dietzschoid B et al. Local~zatiori and ~rnrnunological cliaracterizatiori of

antigenic domains of the rabies virus internal N and NS proteins, Virus researcii, 1987, 8: 103-1 25.

7. Wiktor TJ, Koprowski H. Monoclonal antibodies against rabies virus produced by somatic cell hybr~dization, detection of antigenic variants. Proceedings of the National Academy of Sciences of the United States of America, 1978, 75: 3938-3942.

8. Wiktor TJ, Koprowski H. Antigenic variants of rabies viruses. Jo~irnal of experimental medicine, 1980, 152: 99 1 12.

9. Lafon M, ldeler J, Wunner WH, Investigation of the antigenic structure of rabies virus glycoprotein by monoclonal antibodies. Developments in biological standardization, 1983, 57: 21 9-225.

10. Lafon M , Wiktor TJ, MacFarlan RI. Antigenic sites on the CVS rabies virus glycoprotein: analysis with monoclonal antibodies. Journal ofgeneral virology, 1983, 64: 843-851.

11. Bunschoten H et al. Characterization of a new virus-neutralising epitope that denotes a sequential determinant on the rabies virus glycoprotein. Journal of general virology, 1989, 70: 291 2 9 8 .

12. Lafon M , Wiklor TJ. Antigenic sites on the ERA rab~es virus nucleoprotein and non-structural protein. Journal of general virology, 1985, 66: 2125-2133.

13. Reagan KJ et al. Anti-idiotypic antibodies induce neutralizing antibodies to rabies virus glycoprotein. Journal of virology, 1983, 48: 660-666.

14. Schumacher C et al. Use of mouse anti-rabies monoclonal antibodies in post- exposure treatment of rabies. Journal of clinical investigations, 1989. 84: 971 -975.

15. Montano-Hrose JA et al. Protective activity of murine monoclonal antibodies against European bat lyssavirus infection in mice. Vaccine, 1993, 11: 1259-1 266.


Annex Dulbecco's modified Eagle's medium (DMEM)

Component

inorgan~c salis Calcium chloride, dihydrate (CaC1,.2H20) Ferric nitrate, nonahydrate (Fe(NO3);9H,O) Magnesium sulfate (MgSO,) Potassium chloride (KCI) Sodium chloride (NaCI) Sodium phosphate, monobasc (NaH,PO,)

Amino acids L-Argnine hydrochloride L-Cystine dihydrochloride L-Glutamine Gycine L-Hstidine monohydrochloride, monohydrate L-Isoleucine L-Leucine L-Lysne rnonohydrochloride L-Methionne L-Phenylaanine L-Serine L-Threonine L-Tryptophan L-Tyrosine L-Valine

Vitamins Calcium pantothenate Choline chloride Folic acid i-lnositol Nacinamide Pyrdoxal hydrochloride Riboflavin Thiamine hvdrochloride

Other D-Glucose

Phenol red

After combiriing the above components, the pH should be adjusted to 6.3 Sodiuni bicarbonate (Na,CO,) should then be added to a f~nal concentration of 3700 n-igll and the pH adjusted to 7.8.

CHAPTER 12

Monoclonal antibodies for the identification of rabies and non-rabies lyssaviruses J. S. Srn~th ' 6: A. A. K/ng2

Since I978 when rnonocloridl antibodies (MAbs) directed ayciirist the rabies virus were first described ( I ) the ability to identify and characterize an tgenc variants of rabies has become an essential part of almost all rabies investigations This chapter describes some of the methocs used to perform these analyses together with examples of research f ~ i i d n g s arising from the~r use

Materials and methods

MAbs directed against one or more Lyssavirus isolates are prepared using standard procedures, as described in Chapter 11. The immunoglob~rlrn subclass and the rabies protein specificity are determined.

Preparation of virus isolates for analysis with MAbs directed against the rabies ribonucleoprotein (MAb-RNPs)

Preparation of virus isolates from brain t~ssue 1. Inoculate two 3-week-old mice w ~ t h 0 03 m of supernatant from rabies-infected

cell cultures or 0.03 m1 of a IO?" suspension of rabes-infected b r a n tissue. 2. After ~nocii lation, the mice should be observed daily. When they become

completely paralysed. kill both mice and remove their brains. 3. Roll one brain along the surface of a wooden tongue-depressor so that excess

moisture such as blood is removed and the lower surface of the brain adheres to the wood. The second brain should be frozen as virus stock.

4. Using scissors, remove the upper surface of the cerebral hemispheres in order to expose the antigen-laden white matter.

5. Using the impression method, prepare serial sections of brain tissue in wells on labelled polytetrafluoroethylene3 (PTFE)-coated glass slides (see Chapter 4). Sufficient slides should be prepared to allow repeat tests to be made. Care should be taken to avoid contact between the rim of the well and the infected brain tissue, which may lead to the spread of MAb-RNPs between wells.

6. Leave the slides to dry at room temperature for at least l hour, and then fix n acetone at - 20 "C for a minimum of l hour. Leave to dry at room temperature.

' Research Microbiologist. Rabies Labolatory. Division of Viral anii Rickettsia Diseases, Centers for Disease Control and Preventton. Atlai!a G A USA Rabies Research Lcaoer, Ct?ri:ral Veterinary Laboratory Ministry of Agriculture, Fisheries and Food, Weybrdge, Surrey. Er~glarid. Also knowri as politef


7 Stair1 one slide with a fluorescein sothlocyanate (FITC)-conjugated antiserum which has a broad cross-reactivity with Lyssav~rus ~solates and record the degree of positivity At least 50% of the microscope f~elds should contain rabies- specific inclusion bodes, otherwise the sample should be rejected

When correctly prepared, slides can be stored at - 20°C for at least 1 year.

Preparation of viriis isolates in cell culture The reaction with MAb-RNPs may be easier to interpret with rabies-infected cell cultures than with rabies-infected brain material. If the distribution of the antigen in the b r a n impression slides is uneven, or if the reaction with the MAbs cannot be determined conclusively, a 10% suspension of the brain ~naterial in Eagle's minimum essential medium (EMEM) is prepared and used to inoculate cell cultures (see Chapters 8, 10 and 15).

Virus for MAb analysis and neutralization tests may be prepared as follows.

1. Suspend 6 X lo6 BHK or MNA cells in 1 rnl of EMEM. 2. Add 0 5 nil of b r a n suspension and incubate for at least l hour at 37°C. 3. Add 12.5 m1 of EMEM, and transfer 7.0 m1 of the d~luted suspension (containing

3 X 106 cells) into a 25-m1 flask. 4. Distribute the remaining cells into the wells of tissue-culture chamber slides,

PTFE-coated microscope slides or microtitration plates (50000-100000 cells per well).

5. Incubate the flask and slides or plates in a humidified incubator with CO, at 37'C for 2 days.

6. Decant the fluid, and then fix the slides or plates in acetone at - 20°C for 5-10 minutes, Leave to dry at room temperature.

7. Stain one well with the FlTC-corijugated antiseruni and observe under an immunofluorescence microscope. At least five of the observed microscopic fields should contain infected cells (i.e. there should be at least five foci,/well), otherwise the sample should be rejected.

Isolates with insufficient infectivity should be passaged at 2-day (BHK) or weekly (MNA) intervals by dispersing the cell monolayer in the culture flask and seeding new cultures as above. This process is repeated until cultures contain at least five foci/well. Repeated passage does not usually affect the MAb-RNP reactivity pattern.

Preparation of virus isolates for analysis with MAbs directed against the rabies glycoprotein (MAb-Gs)

Virus for neutralization tests with MAb-Gs is prepared by continuing passage in cell culture until the culture supernatant fluid contains more than 10000 infectious units of virus per ml. This usually occurs at the second or ttiird passage level. Alternatively, if cell-culture facilities are unavailable, virus stocks may be prepared by intracerebral inoculation of mice and preparation of a 10% mouse brain-tissue suspension containing more than 10000 MICLD,,,/ml (the median lethal dose for mice inoculated by the intracerebral route).

IDENTIFICATION OF RABIES AND NON-RABIES LYSSAVIRUSES

lrnmunofluorescence tests using MAb-RNPs

A panel of MAb-RNPs is prepared. Each MAb is diluted in EMEM supplemented with HEPES (4-(2-hydroxyethy1)-l-p~perazineethane-sulfonic acid) buffer and l mmolj l sodium azide to produce a working dilution which gives intense (4 + ) fluorescence with the hon1ologous virus immunogen. Using the correct dilution of MAb is essential: if the dilution is too low, it may not be possible to distinguish between closely related isolates; if it is too high, inconsistent results may be obtained. Working dilutions will remain active for several weeks if kept at 4'C.

The working dilution of the MAb-RNP panel and a 10-fold dilution of this preparation are used for IFA staining of virus isolates as follows.

1 . After air-drying the brain-impression or cell-culture slides at room temperature for 10 minutes, add each MAb in the panel in turn to one or more impressions or culture wells. Care should be taken to avoid cross-contamination.

2. Incubate the slides at 37'C in a humidified incubator for 30 minutes. 3. Remove any unbound antibody by washing the slides in PBS. Care shoirld be

taken to avoid transferring MAbs from one impression or well to another. In some instances, this may require removing the MAbs using a pipette before the washing step. Rinse the slides briefly in distilled water and leave to dry at room temperature.

4. Add FITC-conjugated anti-mouse ~mmunoglobulir~ G (pre-t~trated to the optimum dilution) to each impression or well. Return the slides to the incubator for a further 30 minutes.

5. Rinse the slides briefly in distilled water and observe under a fluorescence microscope Magnification should be 200-400 times. A slide should be recorded as positive i f fluorescent intracytoplasmc inclusions are observed.

Samples are tested first with the working dllutron of the MAb-RNP panel. Negative or weak positive reactions are confirmed by repeat tests witli the 10-fold dilution of the working dli it ion. Patterns of reactivity for each isolate with each MAb-RNP of the panel are established according to the following criteria:

A positive reaction signifies that the intensily of fluorescence and amount of antigen stairled with the working dilution of MAb is identical to that of the homologous virus control. A weak pos~tive reaction signifies that the intensity of fluorescence with the working dilution of MAb is much less than that of the homologous virus control. B ran impressions containing very little antigen and cultures from early cell passages often may not react with the working dilution of MAb, and react only weakly with the 10-fold dilution of the working dilution. A negative reaction signifies that there is no specific fluorescence with the working dilution or the 10-fold dilution of the working dilution

Antigenic analysis using MAb-Gs

The reaction of MAb-Gs with a vlrus isolate may be deterrnned by a varying virus and constant antibody method or a constant virus and varying antibody method In the first niethod serial 5-fold or 10-fold dilutions of the virus are mixed with a constant amount of MAb usually diluted to contain sufficient antibody to


neutralize 1000 or 10000 infectious units of homologous virus. In the second method, dilutions of each MAb-G are made in 8-well slides and mixed with 100 infectious units of virus.

1, lncubate the slides in a humidified incubator witli 5% (BHKj or 0.5~9'~ (MNA) CO, at 37 "C for 60 minutes.

2. Add 0.1 m1 of indicator cells (BHK or MNA cells) to a concentration of 6 x 106 cells; m1 to each well of the slides.

3. Incubate the slides in a humidified incubator with 5% (BHK) or 0.5% (MNA) CO, at 37°C for 2 days.

4. Decant the fluid and then fix the slides in acetone at - 20'C for 5-10 minutes. Leave to dry at room temperature.

5. Add FlTC-conjugated anti-mouse immunoglobulin G (pre-titrated to the optimum dilution) to each well. Reiurn ihe slides to the incubator for a further30 minutes

6. Rinse the slides briefly in PBS and distilled water and observe under a fluorescence microscope

A 100-fold (10') difference in neutralization titre signifies an antigenic difference in the viral glycoproten

If cell-culture facilities are unavailable, a neutralization test in niice may be performed instead.

Applications

Identification and analysis of laboratory and vaccine rabies strains

Antigenic differences in nucleocapsid proteins divide the laboratory vaccine strains into five groups. The rabies vaccine strains ERA,'SAD ( 2 ) and LEPIHEP (3, 4j, which originated from North American field strains of rabies. can be

Table 12.1 lmmunofluorescence reaction of 7 MAb-RNPs with laboratory strains of rabies virus

Virus strain M A ~ - R N P ~

TU187 W377-7 C3 C8 C10 C17 C18

ERA + + + + + + SAD + + + + + + LEP + + + + + + HEP t t PV + + + + + + + PM + + + + + + + cvs + 0 + + + + + + = strong posltive reaction 0 = weak positive reaction No characler = r ~ o reac!lon. aThesotirces of the MAb-RNPs were as follows. MAb TU187 Schileiaer e! al ( E ) . MAb W377-7. Wiktor et a1 (7). MAbs C3. C8 C10 C17 and C18 Smlth (8)


dst lngu~shed from each other and from the strains CVS PV and PM (5) w h c h originated from the Pasteur laboratory strain of rabies virus however within the Pasteur group only the mouse-adapted CVS strain can be distinguished (Table 12 1)

Neutralization tests with MAb-Gs confirm the differences noted for the ribo- niicleoproteins and ale helpful i i i distinguishing the Pasteur vaccine strains (Table 12 2) The main advantage of the MAb-G panels however IS in their ability to ill~istrate the differences between the G proteins of the vaccine strains of rabies and the G proteins of the non-rabies lyssaviruses (Table 123) Although the

vaccine strains of rabies used for prophvlaxs share sufficient cross reactivity with all fieid strains of rabies virus to give protection to exposed individuals the identification of non-rabies lyssaviruses such as Mokola Lagos bat and Duven- hage and the recognition of Eulopean bat lyssaviruses with antlgen~c characteristics quite different from the vaccine strains illustrates the antigenic diversity wlthin the Lyssav~rusgenus Recognition of this diversity is essential to the interpretation of comparative tests of vaccine efficacy

Identification of non-rabies lyssavirus isolates

The first isolates of non-rabies Iyssaviruses were recognized because of their poor reaction with standard rabies diagnostic reagents Hyperimrnune serurn reagents specific for these viruses have been prepared in some laboratories but they have not been widely used and it has not yet been possible to firmly establish their epidemiology The orgrn of some of these viruses is shown in Table 12 4

Recently a WHO Consultation on monoclonal antibodies for rabies diagnosis and research prepared a panel of three MAb-RNPs which should greatly facilitate the identificaton of these viruses(l0 1 1 ) The first W502-2 reacts with all known Lyssav~rus isolates the second C15-2 reacts with rabies virus stralns (laboratory and field isolates) but does not react with the non-rabies lyssaviiuses (Lagos bat Mokola Duvenhage and European bat lyssaviruses) and the third W422-5 reacts weakly with the non-rabies lyssaviruses Lagos bat Mokola and Duvenhage but does not react with rabies virus or the European bat lyssaviruses Ampoules of each MAb will be supplied to national laboratories upon approval of a written protocol of intention submitted to the Veterinary Public Health unit WHO

Once the preliminary identification of tlie isolate has been made a more precise identification can be made using other MAbs For example solates from each of the vlrus groups in Table 124 have been tested with a panel of MAbs induced by using a representative isolate from each of these groups as the irnmuriogeii

The original distinctions between tl ie non rabies lyssavirus groups recognized by hypermmune serum reagents were confirmed in the MAb-RNP analysis In addition the MAb panel recognized multiple variants within some of these groups (12) Although isolates of Moltola virus were more closely related to each other than they were to Isolates of Lagos bat virus and vice versa three different variants of Mokola virus and five different varar ts of Lagos bat virus were recogn~zed by the MAbs Unfortunately there are too few isolates from these two virus groups to permit the association of a parlicular variant with a particular species or geographical area It is possible however to use the distinctive MAb patterns observed ~n tests of the European bat isolates as a means of predicting an


Table 12.2 lmmunofluorescence reaction of 32 MAb-Gs by antigenic site with laboratory strains of rabies virus

MAb-G Virus straina

Number Antigenic ERA LEP HEP CVS-11 CVS-24 PV PM site

+ = =. 1!12 reduc:~on n vlrus !itre No cha:ac.tpr = no reduciiorl iii wrus !itre a ERA LEP and HE? vaccines orignated :rom fleld isolates in !he USA CVS-11, CVS-24, PV and PM

vaccines or~glriated from the Challenge Virus Standard Pasteur lnst~tute Pars France. Soiirce. Adapted from Rupprecht et a (9) . Used by peririlssiorl


assoclatlon between a particular var~ani and a part~cular bat species Two variants were f o u ~ d in tests on 44 isolates from bats in Europe (see Table 12 4) The fiist and most common variant was found in 41 of 44 isolates from at least 4 bat species from seven European countries including all of 36 isolates from Eptesicus serotir~us bats and from a humali rabies case of bat origin in the former USSR The second variant was found in ali 3 isolates from Myotis dasycrlenie bats and a human rabies case in Finland suspected to be of bat origin, but not in any of 41 isolates from other bats in Europe

Differentiation of field isolates of rabies

Because of the ease with which immunofluorescence tests can be performed most epidemiological surveys of antigenrc variants of rabies virus are conducted with panels of MAb-RNPs Studies have shown that isolates taken from epidemiolog- cally distinct outbreaks of rabies often possess antigenically distinct rbonucleo- proteins making it possible to identify a particular variant and to study its distribution and transniission in nature (8)

In the United States data obtarned from antigenic analyses are combined with a well-established system of case surveillance which identifies the animal species affected by rabies in a given area Rabies in terrestrial animals in the United States is primarily a disease of two species the striped skunk (Mephitis meph~tis) and the racoon (Procyon lotoi) For example of over 8000 cases of rabies in terrestrial an~mals in 1992 77% were in these two species (13) Rabies occurs in geographically distinct outbreaks (enzootics) where either skunks or racoons are responsible for the majority of cases Within these species defined enzootics cases in other specles occur infrequently and when widely distributed geographically are thought to result from contact with the major host specles and to be unirnportant for the marntenance of the enzootic In some outbreaks however a few recurring cases involving a second species in an area are noted These cases may represent a traiismission cycle independent of the dominant hosts and a role for other species in the maintenance of the enzootic

The ability to identify antgenic variants of rabies and to map their distribution has led to a better iinderstanding of the above frndings For exaniple if a single antgenic variant is found in all isolates from an area from both the predominant host species and other species a slngle dominant host species may be responsible for the maintenance of the enzootic However if isolates from species other than the main hosts contain a different variant multiple independent cycles of rabies transmission in different species may contribute to the enzootic

Exaniples of both of these possibilities were found in studies conducted in Florida and Texas (14 15) In Florida 84% of the rabies cases in terrestrial aninials were in racoons and the aniigenic variant commonly found in racoons was aiso detected in all other cases in terrestrial animals Since cases in species other than racoons occurred rnfrequently and were widely distributed it was concliided that the racoon was responsible for the maintenance of the enzootic in terrestrial animals in this area

In Texas 76% of the rabies cases in terrestriai animals occur in skunks and the 10-20 cases of fox rabies which occur each year are clustered in a few countles near the centre of the state Although these findings might also seem to descrlbe a single host species enzootic data from MAb-RNP analysis suggest that the

Table 12.3 lmmunofluorescence reaction of 33 MAb-Gs by antigenic site with CVS and non-rabies lyssaviruses

Virus strain -- ---

MAb-G CVS-11 a African origin European bat origin - -

Number Antigenic site Lagos bat Mokola Duvenhage Germany Poland Denmark


Table 12.4 Origin of 64 non-rabies lyssavirus isolates

Virus group Reference Country Year Species no. of virus isolated (no. tested)

1.agos bat 1 Nigeria 1956

43 Ceritrdl African 1974 Republic

190 South Africa 1980

134 South Africa 1982 3 Soiith Africa 1983

4 1 Senegal 1985

113 Zimbabwe 1986

Eptesic us heivum (1 ) Myoli,; pus1lIus (1 ) Eplesicus wahIOergi (1 ) Cat ( l ) Bat (1) Epfesicus helvum (1 ) Cat ( 1 )

Mokola 4 Nigeria 1968 Crocldura sp (1) 5 South Africa 1970 Cat ( 1 )

39 Carneroon 1971 Crocldura sp (1) 40 Central Africarl 1981 Lemmus

Hepublic s ikapus~ (1 ) 174 177 Zinibabwe 1981 Cat (4)

Duvenhage 6 South Africa 1970 Hurrian (1) 139 South Africa 1981 Bat ( I ) 131 Z~nibabwe 1986 Bat ( l )

European bat V a r i o ~ s Europe 1968-89 Bat"41) 8 Finland 1985 H ~ r n a n (1)

154 Forrner USSR 1985 Hunian (1) 29 Netheriar-ids 1987 Myotis

dasycne~ne (l ) 30 Netherlands 1987 Myot~s

dasycneme ( 1 ) 228 Netherlands 1989 My0 trs

dasyrneme ( l )

a Mos:Iy Eptesicus serotinus, in :he Ukraire, isolates were also obtained from l%sper!ibo munnos and Nyc!alus noctuia ( 2 ) and 3 ~nideqtifed species

occurrence of rabies in these foxes may be the result of a transmission cycle independent of that maintaining the disease in skunks. Two antigenc variants were found in rabies isolates from terrestrial animals in Texas. The first variant, found in all of 33 rabies isolates from skunks, was widely distributed across the state (66 of 88 isolates in 51 counties). The second variant, while a minor variant in the state at large (found in 22 of 88 isolates), was the most common variant in the five counties that reported primarily fox rabies (22 of 27 isolates) and was found in 13 of 14 isolates from foxes in this area.

lDENTIFlCATlON OF RABIES AND NON-RABIES LYSSAVJRUSES

Discussion

The adaptability of rabies to a variety of warm-blooded hosts has resulted in a complex interactive zoonoss involving multiple hosts By simply controlling canine rabies or by imniunizng a single wildlife species in areas where there IS more than one host the risk of rabies transmission to humans is greatly reduced but as long as an underlyiiiy enzootic exists in wildlife the threat of reintroduction of the

disease to domestic animals and humans will remain Since the ntroduclion of monoclonal antibody techniques to the rabies field

considerable progress has been made in understanding the movement of rabies in and arnoriy its various reservoir hos t s Tile availab~lity of panels of MAbs now allows dentif icatio~i of the various Lyssaviriis types and subtypes and the differentiation of the strains used for vaccine production from lield virus isolates

Much more remains to be done however part~culaily in those areas where relatively few isolates have been examined Worldwide better surveillance is needed to estimate the potential for both intraspeces and interspecies transms- sion wrthin fables enzootrcs Bat rab~es, for example though clearly of relatrvely insignificant public health importance at present may pose some problems for the future as rabies in terrestrial animals cornes under control The greater role for rabres MAbs may prove to be in the f~eld of vaccrnation and/or posi-exposure treatment for humans using chimeric (n~urine-human) or genetically engrneered MAbs For the foreseeable future however there can be no doubt that MAbs will be an essential tool in identifying those factors that affect the maintenance and soread of rabies in nature

References

1. Wktor TJ, Koprowsk H Monoclonal antibodies against rabies virus produced by somatic cell hybr~dizaiion: detection of antigenrc variants. Proceedings of the National Academy ol Sciences of the United States of America, 1978, 75: 3938-3942.

2 Abelseth MK An attenuated rabies vaccine for domestic animals produced in tissue culture. Canadian veterinary journal, 1964, 5: 279-286.

3. Koprowski H, Cox HR. Studies on chick embryo-adapted rabies virus. l. Culture characteristics and pathogencity Joiirnal of immunology, 1948, 60: 533-554.

4. Koprowsk H, Black J, Nelson DJ, Studies on chick embryo-adapted rabies virus. VI. Further changes in pathogenic properties following prolonged cultivation in developing chick embryo. Journal of irnniiinology, 1954, 72: 95-~ 106.

5. Wrght JT, Habel KA. A comparison of antgenicity and certain biological characterisiics of 6 sub-strains of Pasteur fixed rab~es vrrus. Journai of immunology, 1948, 60. 503-51 5.

6. Schneider LG. Barnard BJH, Schneider HP. Applrcatron of monoclonal antibodies for epidemiological investigations and oral vaccination studies. I .


African viruses. In: Kciwert E et a1 , eds. Rabies in the tropics. Berlin, Springer- Verlag, 1985. 47-53.

7. Wiktor TJ. Flamand A, Koprowsk H. Use of monoclonal antibodies in diagnosis of rabies virus infection and differentiation of rabies and rabies-related viruses. Journal of virological methods, 1980, l: 33-46.

8. Smith JS. Rabies virus epitopic variation: use in ecological studies. Advances~n virus research, 1989. 36.215-253.

9. Rupprecht CE, Dietzschold B, Wunner WH. Antigenic relationships of lyssaviruses. In: Baer GM, ed. The natural history of rabies, 2nd ed. Boca Raton, FL, CRC Press, 1991: 69-1 00.

10. Report of the Fifth WHO Consultation on monoclonal antibodies for rabies d ~ a g n o s ~ s and research, Geneva, 3 March 1989. Geneva, World Health Organization. 1989 (unpublished document WH0,'Rab. Res./89.33; available on request from the Division of Communicable Diseases, World Health Organization, 121 1 Geneva 27, Switzerland).

11 Report of the S~xtli WHO Consultation on rnonoclonal antibodies for iabies diagnosis and research Philadelphia PA 2 3 April 1990 Geneva World Health Organization 1990 (unpublished document WHO /Rab Res 90 34 available on request from the Division of Communicable Diseases World Health Organization 1211 Geneva 27 Switzerland)

12 King AA. Studies of the antigenic relatioilships of rabies and rabies-related viruses using antiniicleocapsid inonoclonal antibodies [Dissertation]. Guildford, Universily of Surrey, 1991

13. Krebs JW, Strine TW. Childs JE Rabies surveillance in the United States during 1992. Joiirnai of the American Veterinary Medical Association, 1993, 203. 1718-1731.

14. Smith JS et al. Demonstration of antigenic variation among rabies virus isolates by using monoclonal antibodies to nucleocapsid proteins. Journal of clii-iical microb~ology, 1986, 24, 573-580.

15. Smith JS e l al. Surveillance and epidemiolog~c mapping of monoclonal antibody-defined rabies variants in Florida. Journal of wildlife diseases, 1990, 26: 473-485

CHAPTER 13

The polymerase chain reaction (PCR) technique for diagnosis, typing and epidemiological studies of rabies

N. Tordo,' D. Sacran7ento2 & H. 5ourhy3

Introduction

Until recently routine diagnosis of rabies even after isolation of the viliis from animal or cell culture was based exclusively on the detection of viral antigens in the suspected specimen using specific polyclonal or rnonoclor~al antibodies (PAbs or MAbs) or w t l i special stains (see Chapters 4-10) However the recent clor~ing of the entire rabies genolne Ihas provided diagnostic laboratories with nucleic acid probes for all the rabies virus genes ( 1 3 see also Chapter 3) Besides the diagnosis of infected samples, the other main problem for diagnostic laboratories is to characterize the infecting rabies virus strain Currently the distinction between the fixed rabies virus strains and the wild rabies or rabies-related isolates is essentially based on antigenic differences using MAbs (see Chapters 11 and 12) Recent progress in the understanding of the molecular structure of rabies and rabies related viruses (1-5) has also led researchers to use molecular biology techniques for typing purposes and molecular epideniiological studles (see 6 for a recent review)

The molecular biology techniques are described here as piecsely as possible except for the very basic procedu~es that are eas~ly found In standard laboratory manuals (7) Particular attention is given to describing simplified methods using commercially available enzymes kits or apparatus A list of the various buffers and solutions required is given in the Annex

Amplification of the rabies transcripts

Principle

Since rabies tianscrlption and replication produce RNA exclusively the amplification procedure consists of the reverse transcr~ption of the target vlral transcript Into complementaly DNA (cDNA) followed by the amplif~cation of the cDNA by the polymerase chain reaction (PCR)

Techniques

Collection of samples Samples must be collected under aseptic conditions in order to avoid contamna- tion that could then be amplified by the high sensitivity of the PCR technique. Brain

' Head. Lyssavirus Laboraloiy Pasteur Institute, P a r ~ s , France. L.yssav~rus Laboratory, Pasle~ir lnst~tute Palls France ' Rab~es Unit Pasteur nstltute. Paris France


samples should be collected via the occipital foramen (8) or the retro-orbital route (9). using a sterile disposable plastic pipette. These methods are rapid and do not require opening of the skull. After collection, the samples should be stored at - 80°C in 1.5-ml microtubes.

RNA extracbon This technique can be adapted to infected cell cultures or samples of infected brain tissue In tile former case, a 75-ml culture flask of confluent inonolayer cells (e.g. MNA or BHK-21 cells) is washed 3 times with 25 ml of PBS and covered with 5 m1 of extractiorl butfer (see Annex). In the latter, samples of brain tissue weighing approxilnately 2 g each are homogenized in 0.5 ml of extraction buffer using a plastic pestle adapted to the microtube.

1 Extract the protein by adding an equal volume of solvent to the cell suspension, mixing vigorously and centrifuging at 13000 g fo r 30 minutes at room temperature, to separate the aqueous and organic phases. Repeat the extraction until mater~al 1s no ionger visible at the aqueous-organic interface. Generally, extraction with I-phenol, 2-phenol/chloroform (1 : 1) and 1-chloroform should be sufficient.

2. Dilute the aqueous phase with sod~um acetate, pH 5.2, to give a final concentration of 0.3 mol,!l.

3. Precipitate the RNA by adding 2 volumes of 70% ethanol and storing at - 20 'C for 1 hour.

4. Centrifuge the precipitate at 13000 g and room temperature for 10-30 minutes. 5. Resuspend the pellet in 70% ethanol and centrifuge again Repeat. 6. Dry the pellet and resuspend in distilled water to a final concentration of 1 mg

per ml, as estimated by measuring the absorbence of the suspension at 260 nm.

cDNA synthesis Total brain RNA (1 pg) is annealed with the primer (100 ng) in 3 p l of distilled water at 65 C for 3 rn~nutes After ch~l l ing on ice, the reactlon m~xture is diluted with 7 pI of reverse transcriptase (RT) buffer (see Annex) contain~ng 200 units of Moloney murine leukaemia virus reverse transcriptase and incubated for 90 minutes at 42 C it IS then d~luted 10-fold in trometamol edetic acid' (TE) buffer (see Annex)

cDNA amplif~cation by PCR The reaction mixture (10pl) is diluted with 9 0 ~ ~ 1 of PCR buffer (see Annex) containing 100 ng ot each prlmer and 2 units of Taq polymerase. For amplification, the mixture is covered with 100 p1 of paraffin oil and subjected to consecutive cycles of denaturation (D), annealing (A) and elongation (E) (see pages 160 and 165) in an automatic thermal reactor.

Precautions Different primer sets can be selected. Depending on the length and cornple- mentarity of the primers to the target region, it may be necessary to adjust the setting of the amplification programme.

'Also know11 a.; ethylenedamnetetraacettc acid or EDTA

158

POLYMERASE CHAIN REACTION

To avoid the problem of contamnaton, gloves should be worn whenever samples or reagents are handled. High-speed mixing and centrifugation procedures should be carried out in tightly closed containers and in a biological safety cabinet (type l1 or Ill). Other operations that might cause aerosols (e.g. pipetting) should also be carried out in biological safety cabinets. Pipetting by mouth should be avoided, and the reaction mixtures should be stored in a freezer reserved for that purpose (see also Chapter 1) . Ideally, work areas should be separated for setting up reactions. Each test should include both a positive (infected brain) and a negative (uninfected brain) control. A control where the nucleic acid to be amplified is replaced by water should also be included to detect contaminants in [he PCH buffer

Diagnosis

Principle

Rabies virus and Mokola virus are the prototypes of the two most divergent serotypes or genotypes (1 and 3) of the Lyssavirus genus, and their comparative analysis is illustrative of the maximum variability within this genus (5). Fig. 13.1 compares the genome sequence of both viruses over the first 6000 nucleotides froni the 3-end, covering the N. M 7 , M2 G and Ygenes and part of the L gene (see Chapter 3, pages 39-42 and 43-45). These different genes are obviously subjected to unequal selective pressures, the protein-coding regions being more conserved than the non-coding ones, with the exception of the atypically variable M1 protein gene. The most highly conserved region is the N protein gene. Since this gene is transcribed in highest aniounts during transcriptioi? of the rabies genome (see Chapter 3), the resulting rnRNA (NmRNA) could be a suitable nucleic target for rabies diagnosis, fulfiiling the two main prerequisites: sequence stability between isolates and presence in large amounts.

Technique

The investigator will be confronted with three independent bimodal choices. First, the presence of viral nucleic acids can be demonstrated either directly, by simple hybridization with viral probes complementary to the targeted region, or indirectly, after a preliniinary PCR amplification step. Secondly, the presence of viral products can be determined either on filtered dot blots, or on Northern or Southern blots after separation by agarose gel electroplioresis. Thirdly, the hybridized probe can be visualized either directly, by autoradiography (using a radiolabelled probe), or indirectly, by a coloured enzyme-linked ~mrnunosorbent assay (ELISA) (using a digoxgenin-labelled probe) or by enhanced chem~luminescence (using a horseradish peroxdase-labelled probe).

These three choices are unrelated and can be combined independently.

O/igodeoxynuc/eotides and PCR conditions As the efficiency of PCR is inversely related to the length of the sequence, the selected set of primers should be separated by a short distance (here 443 base pairs or bp). Different primer sets have been described in the literature (6).


Fig. 13.1 Comparison of the genome sequence of rabies virus (PV strain) and Mokola virus, based on the first 6000 nucleotides from the 3'-end

- l Y I EPIDEMIOLOGY [

A

Rabies genome (PV strain)

~ K D .

P r o t e ~ n - c o d ~ n g region

Diagonal lines outline the homologous regions The pr~tners selected either for diagnosis or for typing and molecular epidern~ologiral piirposes are ind~cated

N1 ( + ) sense. (587) 5'-TTT GAG ACT GCT CCT TTT G-3' (605) N2 ( - ) sense. (1029) 5'-CC CAT ATA GCA TCC TAC-3' (1013).

#

L L primer IZJ /

Both primers were chosen in particularly stable regions of the N gene to allow suitable amplification of most rabies and rabies-related isolates They were synthes~ied and numbered according to the sequence of the Mokolavirus The N I primer shows 2 3 msmatches with the sequence of the known fixed rabies strains, while the N2 primer shows 0-2 mismatches This heterogeneity ensures that amplification is efficient even in extreme conditions that could naturally occur with the wild rabies isolates

The N I primer is used to prime cDNA which is thereafter arnplif~ed by the NI-N2 set

The PCR conditions are five initial cycles of denaturation (D 60 seconds at 94°C) annealing (A 90 seconds at 45°C then 20 seconds at 50cC) and

I TYPING I


elongation (E: 90 seconds at 72'C) and 30 additional cycles where D and E are reduced to 30 seconds arnd 60 seconds, respectively. The final elongation is always carried out at 72 "C for 10 minutes.

Dot-blot analysis of extracted RNA or PCR prod~icts The extracted RNA (5-10 ;L = 5-10 pg), the PCR products (5 i l l ) or serial 10-fold dilutions of the control DNA are diluted to 1 0 0 ~ ~ 1 with TE buffer, denatured by heating at 95'C for 10 minutes and chilled on ice. Samples are then filtered onto a dry nylon membrane using a mi~lti-well vacuum filtration unit. After air-drying at

room temperature, the nucleic acids are covalently bound to the membrane by exposure to ultraviolet light (312 rirn) for 3 minutes.

Northern blot analysis of extracted RNA The extracted RNA ( 5 - 1 0 ~ ~ 1 = 5-10pg) is denatured in 10 mmol I sodium phosphate dibasic-sodium phosphate monobasic (Na,HPO,-NaH,PO,) buffer pH 7 5 supplemented with 0 5 mmol/l edetic acid and 50% formamde for 5 minutes at 65-C separated by electrophoresis on a denaturing 0 goh agarose gel buffered with 10 mmol l Na,HPO,-NaH,PO, buffer pH 7 5 and 1 1 mol ' l formal dehyde and then transferred onto nylon membranes using sodium-sodium citrate (SSC) buffer 20-fold concentrate (see Annex)

Southern blot analysis of PCR products The PCR products (10 PI) are separated on a 1.246 agarose gel containing 0.5 pg/ml of ethidum bromide buffered with trometaniol-acetateedetc acid (TAE) (see Annex), photographed and transferred onto nylon membranes i i sng 0.4 mol/l sodium hydroxide (NaOH).

Radiolabelled or non-rad~olabelled probes A purified RsaI,,,-Pvull,,, DNA fragment (100 rng) internal to the amplified segment is labelled by random priming using either a-32P-labelled deoxynucleo- tide triphosphate (dNTP) or digoxigenn-labelled deoxyuridne triphosphate (dUTP) The respective procedures are carried out according to the manufacturers nstructlons Briefly the DNA is heat-denatured and chilled on ice The synthesis is random primed with hexanucleotdes and takes place for 5 hours at 37 "C in the presence of a suitable buffer the dNTP labelling mixture (see above) and Kleenow DNA polymerase ( a fragment that has no exonuclease act~vity) The labelled DNA is separated from unbound deoxynucleotides by cross-linked dextran chromatography or precipitation with 70°4 ethanol

Hybridizat/on with rad~oiabelled probe 1 lncubate the membranes (100cm2) in 10 m1 of hybridization butfer 1 (see

Annex) in a sealed plastic bag for 1 hour at 65°C 2 Without allowing the membranes to dry discard the buffer and add 5 ml of fresh

buffer containing 10, courits per minute (cpm) of denatured 32P-labelled probe Incubate at 65'C overnight

3 Discard the buffer and wash the membranes 3 times for 15 minutes at 65°C in 40 mmol/l Na,HPO,-NaH,PO, buffer pH 7 5 supplemented with 1 mmol/l edetic acid and l o/b sodium dodecyl sulfate (SDS)

4 Leave the membranes to dry and expose them to scientific imaging film with intensifier screens at - 70 "C for periods ranging from 1 hour to overnight


Hybridization with non-rad~olabelled probe 1. incubate the membranes (100cm2) with 10 ml of hybridization buffer 2 (see

Annex) in a sealed plastic bag for 1 hour at 65'C. 2. Without allowing the membranes to dry, discard the buffer and add 5 ml of fresh

buffer containing 5 p l of denatured digoxigenin-dUTP-labelled probe. Incubate at 65 "C overnight.

3. Discard the buffer. and wash the membranes twice for 5 minutes at room temperature with SSC buffer, 2-fold concentrate (see Annex) and 0.1% SDS, and twice for 15 minutes at 65 "C with SSC buffer, diluted I@-fold and 0.1 % SDS. From that point, the membranes can be stored air-dried until immunological detection, which is carried out at room temperature.

4. Wash the hybridized membranes for 1 minute in detection buffer 1 , immerse for 30 minutes in l00 ml of detection buffer 2 and wash again in detection buffer l .

5. Soak the membranes for 30 minutes in 20 ml of solution containing 3.0 units of polyclonal sheep anti-digoxigenin Fab fragments conjugated to alkaline phosphatase.

6. Discard the solution and wash the membranes twice for 15 minutes with 100 ml of detection buffer 1.

7. Equilibrate the reaction mixture above pH 7 for 2 minutes with 20 ml of detection buffer 3 (see Annex).

8. Immerse the membranes in 10 nil of freshly prepared colour solution (see Annex). Do not shake the mixture. This step should be performed in the dark. A brown precipitate will usually start to form within a few minutes and continue to be produced for up to 24 hours.

9. After 24 hours, stop the reaction by washing the membranes for 5 minutes in TE buffer (see Annex). Evaluate the colour of the precipitate qualitatively with the naked eye.

Discussion

A prel~m~nary ~nvestigation of the above techniques has been described in detaii (6, 10). Studies of the different combinations of nucleic acid samples, diagnostic support and probe labelling indicate the following:

1 Diagnosis by direct detection of viral RNA either on dot or Northern blots is not sufficiently sensitive in its present form. Although this method had been cons~dered to be sensitive enough when a 32P-labelled probe was used ( 1 l ) , a more detailed evaluation showed that it detects only the most infected samples. The results were similar wheri a non-radiolabelled probe was used.

2. The direct observation of the amplfied segment on agarose gel (Fig. 13.2 (a)) is also not sensitive enough for diagnostic purposes, since nonspecific bands may CO-migrate in highly degraded samples (not shown). However, probing of the corresponding Southern blot with a 32P-labelled probe (Fig. 13.2 (b)) clearly discriminates between specific and nonspecific amplifications and is an efficient method of diagnosis. The characterization is more uncerta~n when a non-radiolabelled probe is used (not shown).

3. Dot-blot analysis of the amplified products is appropriate for diagnosis, regardless of the method used for labelling the probe (Flg. 13.2. (c) and (c l ) ) . It is highly sensitive, quick and easy to perform, and could be simplified fur-


Fig. 13.2 Comparison of different diagnostic techniques for amplifying nucleic acid samples by PCR using the Nl-NZ primer set

Sample

(a) Technique Probe

Eiectrophoresis through an No ethidium bromide agarose gel

443 bp t

Southern blotting 32P-labelled

104 103 102 10' 100 10-1 Dilution of N gene cDNA (pg)

1 0 4 1 0 3 102 $01 100 10-1 Dilution of N gene cDNA (pg)

WHO 94919

N~ne suspect samples ( T I T 9 6posit1ve and 3 negahvel and one negative control (C, water) were tested The ainpl~f~ed products (443 bp) were analysed. by direct iiluminatlon ~11th ultraviolei light after eiectrophoresis through an ethidiom b~ornide agarose gel (a); by Southern biotiiilg using a 32P-labelled probe (b); by dot-blotting using either a "P-labelled (c) or a digoxiyeniii-labelledprobe id). In the iower line of each dot-blot. serial dilutions of ihe N gene cDNA indicate the sensitiviiy of the technique

ther by direct dotting without filtration. Since the 32P-labelled probe and the dgoxigenn-labelled probe are stable for less than 2 weeks and several months respectively, the latter method should be favoured. It allows detection of as little as 1 pg ( lo5 molecules) of amplified products (Fig. 13.2 (d)).

Comparison with classical diagnostic techniques

One hundred samples received in the National Reference Centre for Rabies, Pars, were checked in systematic blind trials both by PCR (using dot-blots and a dgoxigenn-labelled probe or Southern blots and a 32P-labelled probe) and by routine diagnostic techniques (the fluorescent antibody (FA) test, the rapid lissue- culture infection test (RTCIT) and rapid rabies enzyme immunodiagnosis (RREID)) (10). The rescrlts obtained by PCR were identical to those obtained by routine dlagnostlc techniques, proving that the PCR technique could be used for diagnostic purposes. In Thailand, similar results were obtained by PCR i~s ing nested primers (12). Studies are under way to improve the statistical significance of the results when a larger number of samples are examined by PCR.


As with RTCIT, there was a delay of about 18 hours before results were obtained by the PCR (using dot-blots and a digoxigenin-labelled probe), compared with 2~-3 hours In the FA test and 3-4 hours in RRElD (13). At present, PCR should be considered as an alternative method for confirming results obtained by other diagnostic techniques. However, studies are being carried out to reduce the time required for the PCR technique. In order to be used for iroutine diagnostic purposes, the PCR technique would have to be shown to be more sensitive than the existlng routine techniques or suitable for the mtra \/itan? diagnosis of rabies using saliva cerebrospinal fluid or blood samples Since the N1-N2 primer set IS

efficenl in ampl fy~ng the N gene of rabies and rabies-related viruses as distant as Mokola, the PCR technique could also be used to diagnose most lyssavirus infections.

The amplification level o i cDNA to the N gene was found better than that of cDNA to NmRNA, although the latter has been shown to be produced in large quantities during infection (see Chapter 3). This unexpected difference is probably due to the protective effect of nucleocapsid structures for the viral genome, while the messengers are more accessible to nucleases. Accordingly, the use of the positve-sense N I primer for the synthesis of cDNA, resulting in the amplification of the negative-sense N gene. is preferred. In the future, the conibination of the cDNA synthesis and PCR amplification in a single step would both reduce the time required for the PCR technique and increase its sensitivity by simultaneously amplifying the N gene and the NrnRNA. In this case, the positive-sense N I primer and the negative-sense NZ primer are used together as the cDNA synthesis step.

If the N1-N2 primer set is shown to be efficient for diagnosis. other primer sets could also be selected in different conserved regions of the genonie (6)

Typing and molecular epidemiological studies

Principle

For typing studies in contrast to diagnosis it is important to find sensitive criteria to differentiate virus isolates The most suitable genoniic region for this purpose depends on both the method of different~at~on (e g hybridlratlon restriction fragment length polymorphisin seyuencing) and the evolutionary distance between the viruses to be identified The highly variable areas of the genome (see Fig 13 1) such as the remnant rabies pseudogene (Q) (a non-protein coding region highly susceptible to rnutalons) are more likely to represent the natural evoiutioi'7 of the virus outside any external seleclive pressure and are most suitable for differentiating closely related isolates ( 1 6 14) In contrast the conserved protein regions are suitable for typing distantly related lyssaviruses (5 6)

Technique

In contrast to diagnosis typing and molecular epidemiological studies are performed only on the PCR amplified nucleic acid sainples The samples are dffereiitiated either by their susceptibility to typical DNA probes or restriction enzymes (typing) or by direct sequencing (molecular epidemiology)

Oligodeoxynucleot~des and PCR conditions To amplify the Q region in the majority of rabies ana rabies-related viruses two conserved primers are selected in the flanking genes Owing to the variability


of the proximal G gene, notably in the transmembrane and cytoplasmic peptides. a suitable 23-mer G primer could only be located upstream from the transmembrane peptide. In the distal L gene, the closest conserved 24-mer sequence is selected as second primer (L primer):

G ( + ) sense, (4665) 5'-GAC TTG GGT CTC CCG AAC TGG GG-3' (4687) L ( - ) sense. (5543) 5'-CAA AGG AGA GTT GAG ATT GTA GTC-3' (5520)

Bolti primers are synthesized and numbered according to the sequence of the PV rabies virus strair The G and L primers show 0-5 and 0-2 mismatches respectively with the sequence of the other known rabies and rabies-related viruses

The G primer IS used to prime cDNA which is thereafter amplified by the G-L set produciiig a DNA fragment of approximately 880 bp The PCR conditions are 30 cycles of dena tu~a ton (D 50 seconds at 94-C) annealing (A 90 seconds at 45 C) arid elongation (E 120 seconds at 72 'C) the last elongation being always com~le ted at 72 C for 10 minutes

Typing by restriction en7yn7e analysis of PCR-amplified products Alquots (2-10 j i ) of PCR products are digested by various restriction enzymes (e g BamHl Hindll H~ndlll Pstl Rasl or Taql), following the manufacturers conditions for incubation The restriction fragments are separated by electrophoresis on a 1 2% agarose gel containing 0 5 jcglml of ethidium bromide buffered with TAE The pattern is analysed on a print of the gel under ultraviolet light (312 nm) (Fig 13 3)

N~~c leo t ide sequencing of PCR-amplified segments for molecular epidemiology The PCR products (30~11) are separated by electrophoresis on a 0 7 % agarose low-melting gel buffered with TAE During electrophoresis the gel should be kept at 4 C to preveni it from melting When sufficiently separated from the excess primer and nonspecific products, the ainplified segments are cut out from the gel melted at 95°C and distributed in 10- j~ l aliquots into l 5-m1 microtubes and stored at - 20 C

The amplified segments are sequenced using the following technique known as the dideoxy-chain termination technique

1. Denature one tube by heating at 95 -C for 3 minutes, then keep it at 37cC throughout the sequencing reaction to prevent polymerization of the agarose gel.

2. Add 2 p l (50 ng) of either G or L primer and 2 jtl of annealing buffer (see Annex) and incubate at 37 'C for 20 minutes

3 Add 6 ~ t l of iabelirig solution (see Annex) containing the four dNTPs (one of which should be labelled with x - ~ ~ S ) and T7 DNA poymerase Mix gently and incubate for 2 minutes at 37 C

4 D~stribute 4 5 j i I of the reaction mixture into each of four pre-warmed 1 5-m1 microtubes each containing 2 5 jil of sequencing solution (see Annex) consisting of a dNTP1ddNTP mixture Incubate the tubesfor5 minutes at 37'C

5 Stop the reaction by adding 5 jil of stopping solution (see Annex) to each tube 6 Freeze-dry the samples to a final volume of 1 3 pi 7. Denature the samoles at 90 'C for 3 minutes and then chill on ice


8. Prepare a sequencing gradient gel (60 cm X 0.1 mm) consisting of: on the top, TBE buffer 2-fold dilution. and 5% acrylam~de-urea (see Annex); on the bottom, TBE buffer 5-told concentrate, and 7% acrylamide-urea (7, 10). The 2-fold dilution of TBE buffer should also be used as buffer for the top gel; however, the buffer used for the bottom gel is not diluted.

9. Load the samples onto the gel, and perform electrophoresis at 50 watts (approximately 2000 volts) until the bromophenol blue reaches the bottom of the gel (after about 5 hours).

10. Fix the gel in 109,n ethanol and 10% acetic acid. Dry on filter paper and expose to film overnight at roorn temperature.

Discussion

The G-L primer set has been shown to be efficient In ampl~fying the JI gene of all

rabies fixed virus strains or rabies virus isolates of genotype 1 (10) Its true specificity is evidenced by the amplification of Mokola W gene (genotype 3) highly divergent frotn the rabies J, gene (genotype 1) such that internal probes are unable to show reciprocal hybridization on Southern blots (6, 10) Such a differential hybridization could be regarded as a possible approach to typing unknown strains of rabies and rabies-related viruses

Typing with restr~ction enzymes By visual observation of the cleavage pattern of the G-L amplified fragment uslng a panel of only four major restriction enzymes (BamHl, Hindll. Hindill and Pstl) (Fig. 13.31, it is poss~ble to distinguish between:

1 The main groups of the fixed rabies virus strains used as vaccine seeds, the lineage of which has been described (70, 15) (see also Chapter 3) These include the follow~ng

-Louis Pasteur virus (PAS), Pasteur virus (PV- l?) , Evelyn Rokitniki Abelseth (ERA) and a therrnosensitive mutant of the Street-Alabama-Dufferin isolate (SAD-B1 9):

-Fury High Egg Passage (HEP); -Challenge Virus Standard (CVS) and Pitman-Moore (PM) strains.

2. The French wild rabies isolates from foxes (FWR) 3. Mokola virus (MOK).

Additional restriction enzyrnes can then be used to differentiate the different strairns within each of these groups. For example, Rsal differentiates the ERA and

SAD strains from PV-11 and PAS, and Taql further distnguishes ERA from SAD- B1 9

Since typing with restrict~on enzymes enables the principal fxed rabies virus stra~ns to be distingu~shed from field strains circulating In western Europe, it is extrenlely usefill for evaluating the effectiveness of the current oral rabies vaccination programmes.

In addition, different panels of restriction enzymes have been used on different reglons of the rabies genome, which have made i t possible to differentiate rabies variants on the basis of their taxonomy, geographical distrbution and host species (6) .


Fig. 13.3 Typing by restriction enzyme analysis of PCR-amplified products (a) Restriction patterns of the G-L amplified fragment of the

PV and ERA fixed rabies strains using a limited panel of restriction enzymes

(b) Restriction map of the G-L fragment of the main fixed rabies virus strains used for vaccine production, fhe French wild rabies isolates (FWR) and Mokola virus

WOK) P V ERA

ERA - T a a 1580)

SAD I I

Taql (290) + Taql (290) BamHl . . FWR

I l

Barn H l

HEP I

l ' ~ ~ ~ ~ 3 k i l

base pairs

In the sL1!nmary table sensittvity to an enzyme !s indicated by an astensk

Molecular epidem~ologicai studies The protocol described is highly effcient for direct sequencing of any double- stranded PCR product separated on low-rneltng agarose gel without any additional purification. It has unlmited appl icab~l~ty, ever1 when an ~nternal sequencing primer different from tliose employed for ampifcat ion IS used. Its fdelity has been repeatedly verified by comparng the sequences obtained with cloned cDNA templates.


Fig. 13.4 Pair-wise comparison of the G-L fragments (position 4722-5396 in the PV strain) of the main rabies virus strains used for vaccine production and 12 French wild rabies isolates

FWRz

FWRa

ERA

PV

HEP

PM

cvs ERA PV HEP PM

FWRa and FWRz represent the two most u'ivergent French wild rabies isolates The hgures indicate the percentage of nucleotide divergence The shaded bloch iiighlights the contrast between the !ntrins,c conservation of the ~ / i l d isolates and their divergence from the vaccine strains

This technique has been employed to sequence the W gene (position 4722-5396 in the PV strain) in most fixed rabies strains as well as in 12 French wild isolates (15). The results (Fig. 13.4) show that French wild isolates form a relatively homogeneous group, exhibiting less than 3.5% of divergence in the considered genomic region In contrast, they substantially diverge (about 15%) from the fixed rabies strains used in vaccnes, even though the latter still provide full protection in France Such divergence is not surprising since most ot the classical vaccine seeds were isolated 40-100 years ago

Within the strains used for vaccine production, the PAS/PV strains are strongly related to the ERAiSAD strains. The CVS;'PM strains show a high degree of similarity. The HEP strain is distinct, although much closer to the CVSjPM strains than to the other fixed strains. These molecular relationships are somewhat contradictory with the lineage of the strains. since the PAS, PV, CVS and PM strains

are thought to be derivatives of the original Louis Pasteur strain (isolated in 1882). while ERAISAD and HEP are derivatives of strains isolated in the USA in 1935 and 1939, respectively (15) (see also Chapter 3). This strongly suggests that the history of these strains was in part wrongly recorded.

Molecular epidemiological studies have been carried out in which partial or total sequences of the N, G and UJ genes have been conipared. Such studies have led to

considerable progress being made in understanding the genetic variability of and phylogenic relationships within the Lyssavirus genus (5, 6, 15-19)


Compar~son with other typing techr~iques The typing of rabies stra~ns by visual Inspection of the restriction pattern is simple and rapid, and can usually be completed within 24 hours of receiving the sample. Furthermore, it does not require previous cell adaptation, a possible source or selector of mutations, but instead preserves the natural situation. All these qualities provide a substantial advantage over the classical analysis of antigenic differences using panels of MAbs directed against either the rabies nucleocapsid or the glycoprotein (see Chapter 121.

Direct sequer-icrig rernans the rnost complete technique for characterizing rabies isolates It is usually completed within 2-3 days, the results being available on the third or fourth day for immediate analyss by suitable computer programs.

Beside its suitability for routine diagnostic purposes, the PCR technique appears as the most conveneni and powerful tool for typing and molecular epidemiological studies of lyssaviruses (6, 10) even on highly degraded samples. An analysis of the diversity of the lyssaviruses is currently in progress; particular attention is being glven to members of the divergent genotypes 2-6.

References

1. Tordo N, Poch 0. Structure of rabies virus. In: Campbell JB. Charlton KM, eds. Rabies. Boston, Kuwer Academic Publishers, 1988: 25-45.

2. Wunner WH et a . The molecular biology of rabies viruses. Review of infectious diseases, 1988, 10 771-784.

3. Bourhy H, Sureau P, Tordo N. From rabies to rabies-related viruses. Veterinary microbiology, 1990, 23: 1 15-1 28.

4. Bourhy H et al. Complete cloning and molecular organization of a rabies- related virus: Mokola virus. Journal of general virology, 1989, 70. 2063-2074.

5. Bourhy H Kissi B, Tordo N. Molecular diversity of the Lyssavirus genus. Virology, 1993, 194: 70-81

6. Tordo N. Bourhy H. Sacrament0 D. PCR technology for Lyssavirusdiagnosis. In. Clewley J, ed. Tlie polynlerase chain reaction (PCR) for human viral diagnosis. London, CRC Press. 1995: 125-145

7. Sambrook J, Fritsch EF Maniatis T. Molecular cloning; a laboratory manual. Vol. 2. New York, Cold Spring Harbor, 1989.

8 Barrat J , Halek H Sirnpltfied and adequate sampling and preservation techniques for rabies diagnosis in Mediterranean countries Comparative ~mmunology microbiology and infectious diseases 1986, 9 10

9. Montaiio Hrose JA, Bourhy H, Sureau P. Retro-orbital route for brain specimen collection for rabies diagnosis. Veterinary record, 1991, 129: 291 - 292.


10. Sacramento D, Bourhy H, Tordo N. PCR technique as an alternative method for diagnosis and molecular epidemiology of rabies virus. Molecular and ceilular probes, 1991, 5: 229-240.

I l . Ermne A, Tordo N, Tsiang H. Rapid diagnos~s of rabies ~nfection by means of a dot hybridization assay. Molecular and cellular probes, 1988, 2: 75-82.

12. Kamolvariri N et al. Diagnosis of rabies by polymerase chain reacton with nested primers. Journal of ~nfectious diseases, 1993, 167: 207-210.

13. Bourhy H et al. Comparative field evaiuation of the fluorescent-antibody test, virus isolaton from tissue culture, and enzyme inimunod~agnosis for rapid laboratory diagnosis of rabies. Journal of cliriical microbiology, 1989, 27: 51 9-523.

14, Tordo N et al. Walking along the rabies genome: IS the large G-L intergenc reglon a remnant gene? Proceedings of the National Academy of Sciences of the United States of America. 1986, 83: 391 4-391 8.

15. Sacramento D et al. Molecular epidem~ology of rabies virus in France: comparison w ~ t h vaccne strains. Journal of general VII-ology, 1992, 73: 1149-1158.

16. Smth JS et al. Epidem~ologic and historical relationship among 97 rabies virus isolates as determined by iim~ted sequence analysis. Journal of infect~ous diseases. 1992, 166 296-307.

17. Nadir-Davis SA, Casey GA, Wandeler A. Identification of regional varlants of the rabies virus withln the Canadian province of Ontario. Journal of general virology, 1993, 74: 829-837.

18. Tordo Net al. Molecular epidemiology o i Lyssavirus. focus on the glycoprotein and pseudogenes. Onderstepoort journal of veterinary researcii, 1993, 60: 315 323.

19. Bourhy H , Kissi B, Tordo N. Taxonomy and evolutonary studies on Lyssavirus w t h special reference to Africa. Onderstepoortjournal of veterina1:y research, 1993, 60: 277-282.

Annex Preparation of buffers and reagents

Acrylamide, 5% solution/urea Acrylarnide 61s-acrylarnide Urea TBE buffer, prepared as below, 2-fold dilution, to make


Acrylamide, 7% solution/urea Acrylamde 66 5 g 61s-acrylamide 3 5 g Broniophenol blue 0 3 9 Saccharose l00 g Urea 480 g TBE buffer, prepared as below, 5-fold concentrate, to make 1000 m1

Annealing buffer 1,4-Dithiothreitol Magnesium chloride (MgCI,)

Trometamol hydrochloric acid (HCI). pH 7.6

160 mm01 100 mm01

10 mmol

Colour solution Detection buffer 3 , prepared as below, supplemented with 0.33 mg/ml of ntroblue tetrazolium salt and 0.175 mg/ml of 5-bromo-4-chloro-3-indoiyi phosphate, toliu- dinium salt.

Detection buffer 1, pH 7.5 Sodium chloride (NaCI) Trometamol-HCI Distilled water to make

150 rnmol l00 mm01

1000 m1

Detection buffer 2 Detection buffer 1, prepared as above, supplemented with 0.5% blocking reagent

Detection buffer 3, pH 9.5 WC12 NaCl Trometamol-HCI Distilled water to make

Extraction buffer, pH 8.0 Dextran sulfate Edetc acid Sodium dodecyl sulfate (SDS) Tyloxapol Distilled water to make

Hybridization buffer 1, pH 7.5 Bovine seruin albumin (BSA) Edetic acid SDS Sodium phosphate, monobasic-disodum

phosphate, dibasic (NaH,PO,-Na,HPO,) Distilled water to make

50 mm01 iOO minol 100 mmol

l000 ml

50 g 1 mmol

1% 1%

1000 m1


Hybridization buffer 2, pH 7.0 Block~ng reagent SDS Sod~urn N-Iauroylsarcosinate Sod~um-sodium c~trate (SSC) buffer prepared as below,

20-fold concentrate D~stilled watet to niake

Labelling solution, pH 7.5 d-Adenine [ ~ - ~ ~ ~ ] t r i p h o s p h a l e ( [ c L - ~ ~ S I ~ A T P ) (1000 Ci,:rnmol) BS A d-Cytosine triphosphate (dCTP) 1,4-Dithiothreito (DTT) T7 DNA polymerase Glycerol d-Guanine triphosphate (dGTP) Na Cl d-Thyrnid~ne triphosphate (dTTP) Tiometamol-HCI Distlied water to niake

PCR buffer, pH 8.3 d-Adenine triphosphate (dATP) dCTP Dimethylsulfox~de (DMSO) Gelat~n dGTP

MgCI2 Potasslum chloride (KCI) Trornetarnol-HCI dTTP Distilled water to tnake

Reverse transcriptase (RT) buffer, pH 8.3 dATP dCTP DTT dGTP KC1

MgCl l RNAsln Trornetarnol-HCI dTTP Distilled water to make

Sequencing solution A (adenosine), pH 7.6 d ATP

l pmol 30 mg

0.4 prnol 1.5 mrnol

500 000 1U 1.5%

0.4 prnol 85 mm01 0.4 pmol

7 mm01 l000 ml

0.2 rnrnol 0.2 rnmol

10% 0.01 ?/o

0.2 mm01 1 5 rnmol 50 rnmol 10 mrnol

0.2 rnrnol 1000 m1

1 mmol 1 rnrnol

10 rnmol 1 rnrnol

75 rnrnol 3 mrnol

4 X I Q 5 units 50 rnrnol

1 rnmol 1000 nil


ddATP dCTP dGTP NaCl Trornetamo-HCI dTTP

Sequencing solution C (cytosine), pH 7.6 dATP dCTP ddCTP dGTP NaCl Trometamol-HCl dTTP

Sequencing solution G (guanine), pH 7.6 d ATP dCTP dGTP ddGTP NaCl Trometamo-HCI dTTP

Sequencing solution T (thymidine), pH 7.6 dATP dCTP dGTP NaCl Tronietamol-HCI d TTP ddTTP

14 pmol 840 p n o l 840 pm01 50 mm01 40 mm01

840 pn?ol

840 pm01 93.5 pmol

14 pm01 840 pmol 50 mmol 40 mmol

840 pmol

840 pm01 840 pm01 93.5 jlrnol

14 pm01 50 mm01 40 mm01 840 pmol

840 pm01 840 pmol 840 prnol 50 mm01 40 mm01

93.5 pmol 14 pmol

Sodium-sodium citrate (SSC) buffer, 20-fold concentrate, pH 7.0 NaCl 3 mol Sodurn cttate 0.3 mol Dist~lled water to make 1000 ml

SSC buffer, 2-fold concentrate, pH 7.0 SSC buffer prepared as above d~luted 10-fold w ~ t h d~st l led water

SSC buffer, 10-fold dilution, pH 7.0 SSC buffer, prepared as above d~luted 200-fold w ~ t h dist~lled water


Stopping solution Brornophenol blue Deionized formamide Edetic acid, pH 7.5 Xylene cyanol

Trometamol-acetate-edetic acid (TAE) buffer, pH 7.7 Edet~c a c d Trornetarnol acetate Distilled water to make

Trometamol-borate-edetic acid (TBE) buffer, pH 8.3 Edet~c acid Trometarnol-borate D~stilled water to make

Trometamol-edetic acid (TE) buffer, pH 8.3 Edetc acid Trometamol-HCI Distilled water to make

1 mrnol 40 rnrnol l000 m1

2 mrnol 89 mmol 1000 ml

1 mrnol 10 rnrnol l000 m1

CHAPTER 14

Techniques for the purification of rabies virus, its subunits and recombinant products B. Dietzschold '

Introduction

In recent years considerable progress has been made in understanding the molecular structure and function of rabies virus proteins An essential part in deterniining structure-function relat~onships IS the purification of rabies particles and separation of their structural components This has been ach~eved primar~ly by the application of new technqiles in protein purification This chapter describes various basic techniques su~tabie for the purification of rabies virus particles, subunits, structural proteins and genetically engineered proteins

Purification of rabies virus particles

The following technique is suitable for purifying small amounts of the virus for laboratory use,

Concentration of the virus from tissue-culture supernatant

1 Remove the cells and cell debris from the tissue-culture supernatant by low- speed centrifugation at 3000 g for 20 minutes.

2. Separate the virus particles from the tissue-culture supernatant by ultracentri- fugatiol? for 120 ~ninutes at 50000 g and 4'C.

3. Discard the supernatant and resuspend the pellet in a small volume (1-2% of the original volume) of NTE buffer, pH 7.5 (see Annex).

Purification of the virus

The conceritrated rabies virus suspension can be purified by isopycnic centrifugation 111 a sucrose density gradient as follows:

1 Prepare a 15 50% sucrose dens~ty gradient (in NTE buffer) using a gradient maker Alternatively a step gradient can be prepared Incubation of the step gradient for 2 hours at 37 'C or for 78 hours at 4 C will result in the forniat~on of a continuous gradienl

2 Layer the virus suspensloll on top of the gradient (the volume of the virus suspension should not exceed loo4, of the gradient volume) and centrifuge for 1 hour at 100000 g and 4°C After centrifugation the virus particles will form a

' ProIeSSor Centei for Neu~oiogy. Department of Immunology and M~crob~o iogy , Jefferson Med~cal College, Thornas Juffcrson Un~verslty. Ph~ladelpli~a, PA, USA


single band at a buoyant density of 1.17 g/cm3. The virus band can be visualized by light scattering. using a focused light source.

3. Collect the virus band using a hypodermic needle and dialyse the virus against an appropriate buffer (phosphate-buffered saline or NTE).

Further purification car? be achieved by rate velocity centrifugation in a 5-3O0Io sucrose density gradient, which is described ~n detail elsewhere ( I )

Purification of rabies virus subunits and structural proteins under non-denaturing conditions

Purification of rabies virus glycoprotein

The rab~es virus glycoprotein (G protein) is a typical membrane protein containing a hydrophobic domain which helps it to anchor in the l i pd biayer of the viral membrane In order to preserve its antigenic structure the G protein must be soli~bilized with mild ionic deteigents such as octoxinol oi octyl-[I-( + )- gucopyranoside (OGP) Furthermore detergents must be present during all the stages of purification in order to prevent aggregation and precipitation of the G proteiri

Isopycnic sucrose density centnfugatlon The rabies G protein can be extracted and purified by sopycnic centrifugation as follows (see reference 2 for further details)

1. Solub~l~zat~on of the G prote~n Dissolve 2 mg of purified rabies virus in 5 ml of 0.3 m o / I sodium chloride (NaCI). 50 mniol/l trometamol-HCI, pH 7.6 and 2% OGP, Incubate the mixture for 20 miriutes at room temperature and then centrfirge for 70 minutes at 120000 g. The supernatant will contairi the G protein, while the pellet will consist mostly of ribonucieoprotein (RNP) and matrix protein.

2, Isopycnic centrifugat~on on a sucrose gradient. Prepare a sucrose gradient consisting of 3 zones: on the bottom, 1.0 ml of 3 % sucrose: then 8 m1 of 5-2546 sucrose (linear gradient): and on the top, 1 m1 of 3% sucrose. The sucrose solutions should be prepared with 0.5 mol/l NaCI, 50 mmol/l trometamol-HCI, pH 7.6 and 2% OGP. Layer 1 m1 of solubilized G protein on top of the gradient and centriflige for 36 hours at 150 000 g and 4'C. After centrifugation. collect 0.5-m fractions from the bottom of the tubes using a hypodermic needle. Fractions containing G protein can be identified by the enzyme-linked imrnunosorbent assay (ELISA), as described in Chapter 9.

/soe/echc focusing ~n a sucrose gradient 1. Add 0.5 m1 of a 20% aqueous solution of octoxinol to a suspension containing

2 mg of purified rabies virus In 5 rnl of NTE buffer. 2, Incubate the mixture for 20 minutes at room lernperat~ire and then centrif i~ge for

60 minutes at 120000 g and 4°C. 3. Dialyse the supernatant against distilled water containing 1 % glycerol (gly-

cerin). 4. Add carrier ampholytes (pH 3.5-10) to a final concentration of 1 YO and layer the

sample on a 5 40% sucrose gradient contain~ng 1% carrier arnpholytes (pH 3.5-10) in a I 10-m1 electrofocusing column.

PURlFlCATlON TECHNIQUES

5. Perform electrofocusing at 800 V (constant voltage) for 48 hoirrs. 6. After isoelectric focusing. collect l -ml fractions from the bottorn of the coliimn

Measure the pH of each fraction. Fractions containing G protein can be identified by the ELISA, as described in Chapter- 9.

Purification of rabies virus ribonucleoprotein

Large quantities of rabies virus ribonucleoprotein (RNP) can be purified from

~nfected cells as follows

1. Infected cells should be scraped (never trypsin~zed) After washing the cells once in PBS, suspend approximately 2 ml of the cell pellet in 5 ml of ice-cold distilled water containing aprotinin (1000 kallikrein inhibitor units). Homogenize the cells in a glass homogenizer and centrifuge for 20 minutes at l000 g. Collect the supernatant.

2 Suspend the pellet containing the cell debris as described above and centrifuge for 20 minutes at 1000 g. Collect the supernatant and repeat this step.

3. Combine the siipernatariis from steps 1 and 2. Centrifuge for 10 minutes at 12000 g and collect the supernatant.

4. Add 6 g of caesium chloride (CsCI) to 15 m1 of the resuliing supernatant. After the CsCl is dissolved, n ix well and centrifuge for 18 hours at 150000 g a n d 4 'C. After centrifugation, the RNP will form a sharp band at a buoyant density of 1.31 g/cm3, which may be visualized by light scatterilig. Collect the viriis band using a 23-gauge needle (approximately 0 7 ml per band).

5. Dissolve 4 g of CsCI iri 9 m1 of NTE buffer containing 2% OGP. Mix 4 volurnes of this solution with 1 voltiine of RNP collected in step 4 and centrifuge for 18 hours at 150000g and 4°C

6. Collect the band as described in step 4 and dialyse agalnst an appropriate buffer (e.g. PBS).

Purification of recombinant proteins

Rabies G protein nucleoprotein (N protein) and phosphoprotein (M1 protein) have been expressed in insect cells using baculoviruses as vectors These expressed proteins have been shown to possess the same antgenicity and mmunogenicity as the corresponding virus-associated proteins However at present only N protein can be effectively purified from insect cells infected with baculovirus recombinants using imrnunoadsorpt~on chrornatoglaphy

Preparat~on of iminunosorbenls 1 . Mix5 m of rabbit antsera directed against the RNP protein with 5 ml of 0.4 mol/l

trornetamol-HCI, pH 80 and load onto a protein A-Sepharose (cross-linked agarose) coliimn calibrated with 0.2 niol,;l trometamol-HCI, pH 8.0.

2. Circulate the antibody through the column for 4 hours at room temperature using a peristaltic pump.

3. Wash the column with 50 ml of 02 mol/l trometamol-HCI, pH 8.0. 4. Elute the adsorbed immunoglobulin with 0.1 m o / I citric acid, pH 3.0 and dialyse

against 0.2 mol, l sodium bicarbonate (NaHCO,). pH 8.5.


Preparation of affinity gel 1 Transfer 8 ml of affinity gel (N-hydroxyscrccinamide ester) to a small glass fluted

funnel and wash the slurry with 5 m1 of ice-cold distilled water 2 Immediately after washing, combine the gel with 5 m1 of piirf ied antibody The

antibody concentration should be 10-20 mg'ml 3 Vortex-mix the gel for 18 hours at 4'C 4 Add 0 5 ml of 1 mol j l ethanolamine hydrochloride, pH 8 0 and vortex-mix for a

further hour at 4 - C The gel IS now ready for immurioadsorption chromatography

lrnrnunoadsorpt~on chromatography 1 . Prepare the cell lysate from the baculovirus recombinant-nfected cells as

described on page 175. 2. Adjust the pH and NaCl concentration of the lysate to 7.5 and 0.1 mol/l,

respectively. 3. Transfer the affiniy gel (see above) inlo a 10-ml column and calibrate with NT

buffer (see Annex). 4. Load the lysate onto the column and circulate it through the column for 4 hours

at room temperature using a peristaltic pump. 5. After adsorption, wash the column with l 0 0 bed volumes (approximately l 0 0 ml)

of NT buffer and elute the N protein with 0.15 mol j l ethanolamine, pH 11.0 and 10% glycerol.

6. Collect l - m l fractions from the bottom of the column, as described above. Fractions containing N protein can be identified by measuring the adsorption under ullraviolet light at 280 nm.

7. Combine the fractions containing N protein and dialyse against 1 mol/l trometamol-HCI, pH 7.8 and l mol/l NaCl at 4'C.

Purification of rabies virus proteins under denaturing conditions

To prepare proteins for chem~cai analysis or for T-helper ceil assays, it is not essential to preserve their entire antlgenic structure Rabies vlrus proteins can be effectively separated and purrfed by sodium dodecyl sulfate-polyacrylamide gel electroplioress (SDS-PAGE) as described below

Solubilization of rabies virus

1 Suspend 1 mg of rabies vlrus in 1 ml of NTE buffer and add 100 jil of 10% SDS to the suspension

2 Heat the sample n a boiling water-bath (100 C) for 1 minute and then c h ~ l l in an Ice bath for 15-30 m~nutes

3 Precip~tate the virus protelns by adding 5volumes of ce-cold absolute ethanol 4 Store the sample overnight at -20 C and centrifuge for 10 minutes at

12000g 5 Suspend the precipitate in 500 pi of 6 5 mmol/ l trometamol-borate buffer pH

84 , containing 1 % SDS 5% 2-mercaptoethanol and 10% glycerol 6 Heat the sample in a boiling water bath until the precprtate is dissolved 7 To v~sualize the protein In the gel dissolve 10-20 pg of the virus n 100 pi of

0 05 rnol l tron?etamol-borate buffer pH 7 8 containing 20io SDS

PURIFICATION TECHNIQUES

8. Mix with 2 0 ~ 1 of 0.2 rnol/l borate buffer, pH 9.2 and lop1 of a solution contalning 5 rng of fluorescarnlne in 1 m1 of dimethylsulfoxide.

9. Incubate the mixture for 20 minutes at room temperature, then stop the reaction by adding 5 p1 of I mol/l methanolamne.

10. Combine the resultng protein sample w t h the untreated protein sample.

Eiectrophoresis

Electrophores~s is performed on a 12.5Y0 polyacryamide separating gel with a 5% stacking gel. The gels are prepared as follows.

m Separating gel solution. M I X 10 m1 of 1.5 mol/l trometamol-sulfate, pH 8.4, with 17.5 m1 of distilled water, 12.5 m1 of 4056 acrylamde blsacrylamide (50: l ) , 60 pi of 10% ammonium persulfate and 24 pI of N, N, N, NI-tetramethylethylene dia- mine (TEMED).

m Stacking gel solution. Mix 3 ml of 0.75 rnolil trometarnol-sulfate, pH 8.4, with 7 m1 of distilled water. 2 m1 of 30% acrylamide-bisacrylamde (50: l), 60 ~ t l of 1046 ammonium persulfate and 5 111 of TEMED.

1 . After the gels have been stacked, leave them to stand for 24 hours 2. Perform electrophores~s on the polymerized gel (before loading w ~ t h a sample)

at l00 V (constant voltage) for 2 hours. 3, Load the protein sample on top of the gel and continue electrophoresis for 16

hours at 5-C and 100 V (constant voltage). The buffer for electrophoress I S

6.5 m m ~ l , ~ l trometamol-borate, pH 8.6, contaning 0.1 SDS, 1 mmol,/l dithiothreito and 1 mmol/'l sodium thioglycolate.

4. To v~sualze the p!-oten bands. observe the gel under ultraviolet light. 5 Reinove the proten bands using a razor blade and immerse for 5 minutes in

distilled water. 6. Purify the protens by electroelution overnght at 80 V (constant voltage) in 50 mmol;l ammonlum bicarbonate (NH,HCO,) containing 0 1% SDS and 0.001 moll l dithothretol.

7. Precipitate the eluted protein with 5 volumes of ice-cold absolute ethanol.

References

1. Wiktor TJ et al. lnducton and biological properties of defective interfering par tces of rabes virus. Journal of virology, 1977, 21: 626-635.

2. Thibodeau L, Naud P. Boudreault A. An nfueriza imrnur~osome t s structure and antigenc properties. A model for a new type of vaccne, In: Nayak DP Fox CF, eds. Geiietic variatior? airlong iniIuen7a viruses. New York. Academlc Press, 1981 587-600.


NTE buffer, pH 7.5 Soduni chlorde (NaCI)


Trometamol-hydrochloric a c ~ d (HCI) Edetic acid'

NT buffer, pH 2.5 NaCl Trometamol HCI

' Also Kriowr as e:~ivlened~am~ne !c!raacctlc a c d (EDTA;

180

CHAPTER 15

A rapid fluorescent focus inhibition test (RFFIT) for determining rabies virus-neutralizing antibody J. S. Smith.' P. A. Yager2 & G. M. Baer3

Virus neutralization tests are the most widely used assays of rabies antibody There IS great variation in the way the tests are conducted but in all assays dilutions of heat-inactivated serum are incubated with a fixed amount of rabies v~rus for 60-90 minutes at 37°C Residual virus infectivity is then determined by inoculating laboratory animals or cell cultures with the virus The sensitivity of the system for growth of residual non-neutralized virus and the time available for virus replication def~ne the amount of challenge virus that is added to each serum dilution The precision with which small amounts of residual infectious virus can be measured also varies with the test procedure

A virus neutralization test in mice was developed in 1935 ( I) It has been widely used and has become the standard by which other tests are now evaluated Cheaper and less time-consuming alternatives to mouse inoculation tests evolved with the adaptation of the Challenge Vlrus Standard (CVS) strain of rabies virus to growth In cell culture in 1958 (2) Several serolog~cal tests were developed for measuring rabies virus-neutralizing ant~bodies in cell cultures including plaque assays (3 4) fluorescent antibody tests (5-10) and enzyme immunoassays (11)

The most widely used cell-culture techniques are those in which foci of virus- infected cells are observed by fluorescent antibody staining (5-10) This chapter describes a rapid fluorescent focus inhibition test (RFFIT) that is used at the Centers for Disease Control and Prevention ( 12)

Standard procedure

Preparation of seed virus suspension

1 . Trypsinze one 3-day-old 150-m1 flask culture of mouse neuroblastorna (MNA) cells. At the Centers for Disease Control and Prevention. an MNA cell line (13), originally obtained from the Wistar Institute, Philadelphia, is used.4 These cells prefer an acidic medium, supplemented with vitamins A similar cell line

' Research Microbiologist Rables Laboratory, Division of Viral and I?ickettsai Diseases, Centers for Disease Control and Prevention. Atianta. G A USA Rabies Laboratory. Div~sion of Viral and Rickettslal Diseases Centers for Disease Cortroi and Preven!ion Atianta. G A USA Direclor Baer Laboratories. Mexico C ~ t y , Mexlco

4Available on request from the Rabies Laboratory Divlson of Viral and Ricket ts~al Diseases. Centers for D~sease Conrrol and Prevention, Atianta G A USA


(CCL131) may be obtained on request from the American Type Culture Collection (ATCC).'

2. Resuspend 30 X 10" cells in a 50-m1 conical centrifuge tube in 2.7 m1 of Eagle's mnlmum essential medium supplemented with 10% fetal bovine serum (EMEM-10) (see Chapter 8, Annex 1 )

3 Using standard rabies safety procedures add 10 X 106 infectious units of CVS- 11 rabies virus and vortex-mix once The virus strain used at the Centers for Disease Control and Prevention (VR959) is obtained from the ATCC ' Incubate the cells and virus for 15 minutes at 37°C: vortex-mix the cells once during this time.

4. Add 10 ml of EMEM-10, vortex-mix, and centrifuge the cells at 500 g for 10 minutes.

5. Discard the sopernatant. Resuspend the cells in 30 m\ of growth mediuin (see Annex 1) and transfer to a 150-nil flask.

6 Gently rock the flask to mix the cell suspension and then prepare three 8-well tissue-cultiire chamber slides (Fig 15 1) by pipetting 0 2 m1 of the cell suspension into one well of each slide

7 Incubate the flask and slides at 37 C in a humidfled incubator with 0 5 % carbon dioxide (CO,) The flask should be incubated as a closed culture (tighten the cap)

8 At 20 40 and 64 hoiirs after infection acetone-fix and stain one slide using an imm~rnofluorescence technique (14) to determine the virus infectivity (see page 137; see also Chapter 7). The supernatant should be harvested 24 hours after the cells reach 100% infectivity (typically 40 hours after infection)

9. Transfer the supernatant to a 50-nil centrifuge tube and centrifuge at 4000 g for 10 niinutes.

10. Distribute the supernatant into 0.5-ml alquots and store at --70-C

Titration of seed virus suspension

1. Thaw one aliquot of the seed vlrus and prepare serial 10-fold d~lutions (10- '-10-~) in EMEM-10.

2. Distribute 0 1 m1 of each virus d l u t o n into one well of an 8-well tissue-culture chamber slide. Add 0.2 m1 of MNAcelIs suspended in EMEM-10 (concentration 5 X 10" cells per 0.2 ml) to each well.

3. Mix the cells and virus by gently rocking the slide, then incubate at 37'C in a humidified incubator with 0.5Oio CO, for 40 hours

4. Acetone-fix and stain the slide using an ~mrniinofiuorescence technique (see page 137). Evidence of virus infection should be observed at the 10 fi dilut~on of virus, indicating a virus stock suspension containing at least 1 X 106 infectious ur i ts per 0.1 ml. Prepare sufficient seed virus so that frequent serial passage of

the virus is unnecessary.

Preparation of stock virus suspension

1, Infect 30 X 1o6 MNA cells with 1 0 x 106 infectious units of the seed virus preparation (see above).

' Amer~carl Type C~l\!ure C o l l e c t o ~ (ATCC1 17301 par i t lawr Drve . Roc~v i l l e , MD 30852 USA

182

Fig. 15.1 An &well tissue-culture chamber slide and a 4-well polytetrafluoroethylene (PTFE)-coated microscope slide

Phoiographed b y D. W Sanderl~n; used by per~ri~ssion.

2. Harvest the supernatant 24 hours after the cells reach 10094 infectivity (typically 40 hours after infection).

3 Distribute the supernatant into 0.5-ml aliquots and store at -70°C

Titration of stock virus suspension

1 . Thaw one aliquot of the seed virus and prepare serial 10-fold dilutions (10-'-10-~) in EMEM-10.

2. Distribute 0.1 m1 of each virus dilution into one well of an 8-well tissue-culture chamber slide. Add 0.2 m1 of MNA cells suspended 11 EMEM-10 (concentration 1 X l o 5 cells per 0.2 ml) to each well.

3. Mix the cells and virus suspension by gently rocking the slide, then incubate at 37'C in a tiumidified incubator with 0.5% CO, for 20 hours

4. Acetone-fix and stain the slide using an immunofluorescerice techn~que (see page 137).

Each we!! of an 8-well tissue-culture chamber slide contains 25-50 distinct inicroscopc f~elds when observed at 160-200 times magnification (Fig 15 2) One unit of virus for the RFFIT 1s determined as the dilution at which 5OD/0 oot the observed microscop~c f~elds conta~n one or more foci of infected cells (the focus- forming dose, FFD,,). The stock virus suspension should contain at least 1 X 104 FFD,, per 0.1 ml (i.e., the well with cells infected with the 10-4 dilution of the vlrus should contain at least one focus of Infected cells in 50% of the


Fig. 15.2 One microscopic field from a well infected with approximately 50 FFDS0 (200 X)

Photographed by D W S a ~ d e r i ~ n used by permiss~on

observed microscopic fields). A stock virus suspension of this titre can then be diluted to 10-2.3 to obtain a challenge virus containing 50 FFD,, (see Annex 2).

Reference sera

A national or international reference serum standard diluted to a potency of 2.0 IU per ml should be included in each test. The reference serum used at the Centers for Disease Control and Prevention is the first International Standard for Rabies lmrnunoglobulin (151, which may be obtained from the State Serum Inst~tute. Copenhagen, Denmark.' The reference serum should be rnainta~ned as frozen aliquots in amounts sufficient for one week of tests. A positive seruni control standard diluted to a potency of 0.5 IU per rnl and a negative seruni control standard with a potency of less than 0.1 IU per rnl should also be prepared by the laboratory and included in each test.

Test sera

Serum samples should be heated at 56'C for 30 minutes before testing in order to inactivate coniplement, If sera are frozen. they should be reheated after thawing. Serial d~lutions o i lest sera may be prepared i n an 8-well tissue-culture chamber slide Screening dilutions of 1 :5 and 1 .50 are sufficient for routine evaluation of vaccination eff~cacy and may be made as follows.

' Intsrrat iona Laboratory of Biolog~cal Standards, Slate Serum lnst~t~lte, 80 Amager Boulevard. OK-2300 Cnperltiagen S Denmark

1. Prepare a 1 :2.5 dilution by adding 0.1 ml of inactivated serum and 0.15 m of EMEM-10 to one well of the side. Mix by gently rocking the slide

2. Transfer 0.05 ml of the 1 :2.5 dilution to a second well containing 0.45 nil of EMEM-10. Discard all but 0.1 m1 from the well containing the 1 :2.5 dilution.

3. Mix the second well and discard all but 0.1 ml. 4. Add 0.1 ml of the challenge virus preparation (containing 32-100 FFD,,) to all

serum dilutions. 5. Mix and incubate at 35'C in a hi~midified incubator with 0.5?'0 CO, for 90

minutes.

Addition of cells

1. During the incubation period, trypsnize a stock culture of 3-5-day-old MNA cells.

2. Resuspend the cells in EMEM-10 to give a final concentration of 1 X 105 cells per 0.2 rnl

3. Distribute 0.2 ml of the cell suspension into each well of the slide and incubate at 35% in a humidified incubator with 0.5% CO, for a further 20 hours.

Acetone fixation and staining by immunofluorescence

1 After 20 hours remove the slides from the incubator and pour off the medium into a virucidal solution (see Chapter 1 page 5)

2 Rinse the slides once in PBS and then fix for 10 minutes at room temperature in cold acetone ( - 20 C)

3 Leave the slides to dry for 10 minutes before adding fl~iorescein isoth~ocyanate- conjugated anti-rabies serum The coi-ilugate may be prepared in EMEM-I0 or PBS, there is no need to adsorb the conjugate with tissue or cells The working dilution of the conjugate should be determined by titration The slides should be stained for 20-30 minutes at 37°C and then rinsed in PBS and d~stilled water respectively

6 Observe the slides under a fluorescence microscope

Calculation of virus-neutralizing antibody titres

Residual virus is detected using a standard fluorescence microscope. The serun? neutralization end-point titre is defined as the dilution factor of the highest serum dilution at which 50% of the observed microscopic fields contaiii one or more infected cells ( I e a 97% reduction in the virus inoculum). This value rnay be obtained by mathematical interpolation (see Annex 2) Alternatively, a 100% neutralization titre may he determined by recording the highest serum dilution at which 100?/0 of the challenge inoculum is neutralized and there are no infected cells in any of the observed fields. For both titraton methods, the titre of antibody in the test serum (in IU per ml) can be obtained by comparison wlth the titre of the national reference standard included in each test (see Annex 2).


Alternative test procedures

Use of BHK-21 S13 cells

If preferred, the baby hamster kidney cell line, BHK-21 S13 ( 16), may be used.' The S13 cell line is der~ved from a clone of BHK-21 cells, C13 which grows well in suspension culture. The S13 cells do not form a smooth, contiguous monolayer and are more resistant to detachment during acetone fixation than the C13 cells. The following changes to the procedures should be made.

Preparation of seed virus suspension

1 Remove the growth med~um (see Annex 1 ) from one 3-day-old flask culture of BHK cells

2 Rinse the monolayer with 5 ml of virus medium (see Annex 1) and add 5 rnl of virus medium containing 25 X 106 rnfectious unlts of CVS-11 rabies virus to give a multiplicity of infection of l 3

3 Incubate the cells and virus for 1 hour at 37'C 4 Discard the supernatant and rinse the monolayer with 5 m1 of virus med~um 5 D~scard the medium and add 30 m1 of flesh virus med~um 6 Seal the flask and incubate at 32 C in a humidified incubator with 3% CO, The

flask should be incubated as a closed culture The progress of the infection is riot monitored the optimum day of harvest is determined empirically (typically 72 hours after infection)

7 Transfer the supernatant to a 50-ml centrifuge tube and centrifuge at 4000 g fo r 10 m~nutes

8 Distribute the supernatant into 05-nil aliquots and store at -70JC

Preparation of stock virus suspension

The stock virus suspension is prepared in the same manner as w ~ t h MNA cells except that the noculum coritains 25 X 106 infectious units of the seed virus preparation

Titration of virus suspensions and addition of cells

The steps are performed as described for MNA cells, except that a different growth medium is used (see Annex l ) , the cell concentratiori is 5 X l o 4 cells per 0.2 ml, and the plates are incubated in 3% CO,.

Use of polytetrafluoroethylene-coated slides

If tissue-culture chamber slides are not available polytetrafluoroethylene2 (PTFE)- coated niicroscope slides niay be used I f so, serial dilutions of test sera and virus should be carried out in test tubes The final mixture of serum, virus and cells is

' Avavable on request from the A!nericar, Tyoe CuitL!,e Coliect~o!i (ATCC) 12301 Parklawn Dr~ve . Rocicville, U0 20852 USA Also rnown as polltet

Table 15.1 Concordance between the mouse neutralization test (MNT) and the rapid focus fluorescent inhibition test (RFFIT)

RFFlT MNT

Positive reaction Negative reaction

Positive reaction 366 25 Negat~ve reaction 0 121

-

Adapred from retererce 8 Used by permsston

added to the slide as well-rounded droplets, al~quots of one serum dilution are added to each of four wells The slides should be placed on moistened paper towels and covered with the i d of a microtitration plate or some similar device to prevent the cells from drying out


The RFFIT has been shown to be slightly more sensitive than the mouse neutralization test (MNT) in detecting virus-neutralizing antibodies in post- vaccinal sera (see Table 15 1)

Several studies have shown that strict attention to two items will result in an RFFIT that is more reliable and tepioducible than a neutralization test in mice (17-20) The first of these is the titre of the challenge virus with the number of FFD,, varying rio more than twofold from the selected challenge dose For example if 60 FFD,, I S selected as the challenge dose the virus for a given RFFIT must contain between 30 and 90 FFD,, for the test to be within the acceptable range The second item is the determination of a range of acceptable titres for the reference standard serum or immunoglobulin This range may be calculated from titrations performed over several months

The RFFIT may also be used for measuririg (he potency in terms of IU per m1 of human rabies immunoglobulin (HRIG) preparations to be used for post-exposure treatment (9 19-21) However these tests which frequently involve collaboration between several laboratories require attention to two addit~onal details First it IS

important that all the laboratories involved in an evaluation are using the same straln of CVS-l1 challenge virus Monoclonal antibody studies of antigenic determinants in the rabies glycoproten have revealed several strain differences even within the CVS rabies strains (22) (see also Chapters 11 and 12) which may give rise to differences 11) the virus-neutralizing antibody titre

Secondly collaborating laboratories must use the same reference sercrrn standards In one study HRIG preparations had similar values whether assayed by the RFFIT or the MNT Values for equine rabies ~mmunoglobulin (ERIG) preparations, however were much higher in the MNT than in the RFFIT (23) Although not exarnined in this study antigenic differences in the challenge virus in the different laboratories r i g h t also have contributed to the discrepancy between the two tests


References

1 . Webster LT Dawson JR. Early diagnosis of rabies by mouse inoculation. Measurement of humoral immunity to rabies by moilse protection test. Proceedings of the Society for Experimental Biology and Medicine. 1935, 32: 570-573.

2. Kissling RE. Growth of rabies virus in non-nervous tissue cultiire. Proceedings of the Society for E~perimental Biology and Medicine, 1958, 98. 223-225.

3, Wktor TJ, Cark HF. Application of the plaque assay technique to the study of rabies v~rus-neutraliz~ng antibody ~nteractions. Annales de I'lnstituf Pasteur: Microbiology, 1973, 124: 271 -282.

4. Bussereau F. Flamand A, Pese-Part D. Reproducible plaquing system for rabies vlrus in CER cells. Journal of viroiogical methods, 1982. 4: 277-282.

5 K ~ i g DA Croghan DL Shaw EL A rapid quantitative in vitroserum neutralza- t o n test for rab~es antibody Canadian veterinaryjournal 1965 6 187 193

6 Lennette EH, Einmons RW The laboratory diagnosis of rabies review and perspective In Nagano Y Davenport F eds Rabies Proceedings of the working conferer~ce on rabies Tokyo University of Tokyo Press, 1971 77 90

7. Debbie JG. Andrulonis JA. Abelseth MK Rabies antibody determination by irnmunofluorescence 111 tissue cultirre. Infection and immunity. 1972, 5' 902-904.

8. Smith JS, Yager PA. Baer GM, A rapid reproducible test for determining rabies neutralizing antibody. Bulletin of the World Health Organization. 1973. 48: 535-541

9. Cho H C Fenje P. Rabies neutralizing antibody determination in tissue cult i~re by direct fluorescent antibody technique. Journal of biological standardization, 1975, 3 101 -105.

10. Zalan E, Wlson C, Pukitis D. A microtest for the quant~talion of rabies virus- neutralizing antibodies. Journal of biological standardization, 1979, 7: 213 220.

1 1 . Mannen K et al. M~croneutralization test for rabies viriis based on an enzyme immunoassay. J o u r ~ a l of clinical microbiology, 1987, 25: 2440-2442

12. Velleca WM, Forrester FT. In, Laboratory metiiods for detecting rabies. Atlanta. GA United States Department of Health and Human Services, Centers for Disease Coiitrol 1981. 1-153.

13. Wiktor TJ, Doherly PC, Koprowski H. In vitro ev~dence of cell-mediated

immunity after exposure of mice to both live and inactivated rabes virus

RFFIT

Proceedings of the Natiorial Academy of Sciences of the United Stales of America, 1977, 74. 334 -338.

14. Goldwasser RA. Kssling RE Fuorescerit antibody staining of street and fixed rabies virus antigens. Proceedings of the Society for Experimeiital Biology and Adedicine, 1958. 98: 21 9-223.

15 WHO Expeit Corninittee on Biological Standardization Thirty-fifth report Geneva World Hedlth Oryanrzarron 1985 (WHO Technical Report Series No 725)

16. MacPherson l. Stoker M. Polyoma transformation of hamster cell clones-- an investigation of genetic factors affecting cell conipetence. V!rology, 1962, 16, 147-151.

37 Frtzgerald EA et a1 A collaborative study on the potencv testing of antrabies globulin JoiirnaI of biological standardiiation 1975 3 273-278

18 Ftzgerad EA et al. Laboratory evaluation of the irnmune response to rabies vaccine. Journal of biological standardization, 1978, 6: 101-1 09.

19 Fitzgerald EA et a1 A collaborative study on tile testing of rabies immune globulin (human) by the mouse neutralization test (MNT) and the rapid fluorescent focus inhibition test (RFFIT) Journal otbiologicalstandardiration, 1979 7 67-72

20 Louie RE el a1 Measurement of rabies antibody comparison o l tile mouse neutralization test (MNT] with the rapid fluorescent focus inhib~tron test (RFFIT) Journal of biological standardiration. 1975 3 365-373

21 Guilleinin F et a1 Comparaison de deux methodes de titrages des antcorps anlirabiques neutralisanls [Comparison of two methods for the determination of rabies neutralizing antibodies] Journalof biologicalstandardirat~on 1981 9 147 156

22 Wiktor TJ, Flamand A Koproivski H Use of rnonoclonal antibodies In diagnosis of rabres vrrus rnfect~on and differeritiaton of rabies and rabies-related viruses Journalof virologicalmethods 1980 1 33-46

23 Haase M, Seinsche D, Schneider W. The mouse neutralization test in comparison witli the rapid fluorescent focus inhibition test: differences in the results in rabies antibody determinations. Journal of biological standardiza- tiori, 1985. 13: 123-128.


Annex 1 Growth media for MNA and BHK-21 S13 cells

Component Concentration (in mgll unless stated otherwise)

MNA BHK-21 S13

Caciuni chlori~Je anhydrous (CaCl,) Ferr'c nitrate nonahydrate (re(N03),.9H,0j Potassium chloride (KCI) Magnesium sufate, heptahydrate (MgSO4,7H,O) Sodium chloride (NaCl) Sodium phosphate, rnonobasic, monohydrate

(NaH,PO,.H,O] D-Glucose Phenol red L-Arginine hydrochloride I--Cystine L-Glutamne L-Histdine hydrochloride. monohydrate ~ - I s o l e ~ ~ c i n e

L-Lysine hydrochloride L-Methionne L Phenylaan~ne L-Threonne L-Tryptophan L-Tyros~ne L-Valine D-Calcium pantothenate Choine chloride Fo lc acid lnositol N iacnamde Pyr:doxal hydrochloride R:boflavn Thiamine hydrochloride Tryptose phosphate b io lh Fetal bovine sera (inactivated) Sodi~irn bicarbonate (NaHCO,) Pencilliii (base)

Streptomycin (base) Arnphoter~cin B

Virus medium for BHK-21 S13 cells

G r o w t h r n e d u m f o r BHK-21 S13 cells, a s above , wi th n o feta l b o v ~ n e serum o r t ryptose p h o s p h a t e b ro th , a n d supp lemented w ~ t h 0.1% bov ine serum a l b u m i n a n d

an add i t i ona l 2.8 g o f sod ium b ica rbona te

Annex 2 Calculation of titres

Determination of FFD,, of challenge virus ' 1 . Calculate the percentage of fields containng infected cells.

Cumulative totals Percentage NO, of fields of fields

Virus containing Fields containing Fields containing containing dilution infected Cells infected cells no infected cells infected cells

2. Uslng the method of Reed & Muench (see Chapter 38, Annex), calculate the difference between the logarithm of the startng point dlut ion and the logarithm of the 50% end-pont dilutlon (difference of logarthms) from the formula:

50% - (infectlv~ty next below 50%) logarithm of - X (~nfect~vty next above 50%) - (nfectlvity next below 50%) dilution factOl

In this example the starting point di luton (the dllution showing an Infectivity next below 50%) IS and the dilutlon factor is 10.

Hence, the "difference of logarithms" is:

3. Since the infectivity decreases with increasing dilution, the 50% end-point di luton is lower than the starting point dilution and IS calculated by subtracting the difference of logarithms as follows:

log (reciprocal of 50% end-pont dilution)

= log (reciprocal of starting point dilution) - difference of logarithms

= log 104 - 0.091 = 3.91 (approx.)

Hence log (50% end-point dilution) = -3.91 and the 50% end-point dllution (FFD,,) = 1 0 - ~ . ~ '

' S e e also Appendx 3


Determination of 50% end-point titres of serum

1. Calculate the percentage of fields containing infected cells

Cumulative totals Percentage No. of fields of fields

Serum containing Fields containing Fields containing containing dilution infected cells infected cells no infected cells infected cells

2. In this example the dilution factor is 10 and the starting point dilution (showing an infectivity next below 50%) is 1 :5.

Using the formula above, the difference of logarithms is:

Since the infectivity is increasing as the dilution increases, the 50% end- point dilution s Iiigher than the starting point dilution and is calculated by adding the difierence of logarithms as iollows:

log (reciprocal of 50% end-point dilution)

= log (reciprocal of starting point dilution) + difference of logarithms

= 0.699 + 0.667 = 1.366 = 1.37 (approx )

Hence log (50% end-point dilution) = -1.37 and the 50% end-point dilution = 10-l.~' ( = 1 :23)

The 50% end-point dilution of the reference serum controls is determined similarly, except that serial 5-fold dilutions are used, hence the logarithm of the dilution factor is 0.699 = 0.70 (approx.)

Determination of the relative potency of a test serum in international units (/U) per m1

No of end-point titre of test serum X 2 1U

the test serum end-point titre of reference serum diluted to contain 2 0 lU,ml

For example, the number oi IU/ml of a tesl serum of end-point titre 1 :23 is 0.23 when compared with a reference serum of end-point titre 1 :200.

CHAPTER 16

An in vitro virus neutralization test for rabies antibody C. V Trimarchi, ' R D. Rudd2 & M. Safford. Jr3

Diagnostic laboratories perform rabies virus neutralization tests lor numerous purposes

- to confirm the adequacy of the antibody response following post-exposure vaccination, part~cularly in cases where there has been a departure frorn recommended schedules or routes of administration, or when the patent is suspected of being immunocomprom~sed;

- to determine the need for vaccinations to boost pre-exposure immunization; - to detect rabies antibody in serum and cerebrospinal fluid to aid in the intra

vitam diagnosis of the disease; - in conjunction with vaccine trials, to determine the efficacy of a particular

vaccine or vaccine regimen in stimulating a humoral antibody response.

In the past, the standard method for assaying rabies antibody titres consisted in neutralizing a constant dose of the previously titrated challenge virus with a series of different dilutions of antirabies serum, using ince as the indicator system ( I ) . Following the successfiil propagation of rabies virus in cell culture, these methods have been replaced by in vitro methods.

Resdiral virus has been demonstrated in inoculated cell rnoriolayers by plaque reduction, the enzyme-linked immunosorbent assay (ELISA; see Chapters 41 and 42) and immunofluorescence (2). Currently, the most widely used method is the rabies fluorescent focus inhibition test (RFFIT), as described in Chapter 15.

Method Precautions

The procedures described below should be carried out in accordance with recommended laboratory practices for handling infective materials (3, 4, see also Chapter 1 ) . All procedures involving rabesviri is should be conducted in b~ological safety cabinets (Class Ill). and protective clothing should be worn. During all dilutions the pipette tip should be replaced between serial transfers.

' D~rector Rables Laboratory Wadsworth Center for Laborator~es and Research New York State Departmel: ot Heal:h A loa ly New York NY USA

'Assoc~are Bac'er~oiog~st Rabies Laboratory Wadsworth Cen!er for Laboratories and Research New Yo lk State Department of Health Aibany New York NY USA ~ a ~ ' e r l ~ l 0 ~ 1 S t Raoies Laboratory Wadsuorth Center 'or iaoora to r~es a r d Researc i hew York state Department of Health Albany New York h Y USA


Production of challenge virus

Produce a seed virus pool by propagating the Challenge Virus Strain (CVS-11)' in mouse brain or cell culture to contain approximately 105 TCID,, (median tissue- culture infective dose) per ml. Dispense in l-m1 aliquots and store at -70°C.

It has been shown that the comparability of neutralization test results can be affected by the selection of the challenge virus strain and the source of the standard reference serum. When these factors are controlled, the RFFIT has been shown to be reproducible and sensitive when performed properly (2). It is particularly sensitive for detecting low levels of rabies virus-neutralizing antibody, However, some aspects of the procedure may affect the comparability of results between laboratories. In the RFFIT, the antibody titre is determined by statistical analysis of the reduction In the number of fluorescent foci found at two dilutions of serum. This requires mathematical analysis of the proportion of microscope fields containing infected cells. The skill and experience of the mcroscopist and the tedium of counting the number of infected fields by fluorescence microscopy can affect the precision of this method Reproducibility of results is also very dependent on maintaining the challenge dose within narrow limits. The challenge dose is determined by a quantitative method that can be affected by subtle changes in cell growth characteristics. Moreover, this challenge dose is determined without the benefit of titration against a reference serum.

This chapter describes an alternative in vitro neutralization test that defines the antibody titre as the highest dilution of serum that produces complete neutralza- tion of the challenge virus This permits less quantitative and tedious microscopic examination and eliminates the need for complex statistical estimation of titre The titre is expressed in International Units (IU) and is determined by comparing the titre of the sera under test with that of a standard rabies serum reference preparation The challenge virus dose is determined not only by virus titration, but also by neutralization with the reference serum. This standardization is validated on each test by titration against the reference serum

1. Seed a 75-cm2 plastic cell-culture flask with 3 X 103 BHK-21 C13 cellsZ in Eagle's minimum essential medium (EMEM) supplemented with 10% fetal bovine serum and 10% tryptose phosphate broth, together with 2 mmol of glutamine, 2.2 mg of sodiurn bicarbonate (NaHCO,), 200 IU of penicillin and 0.4 mg of streptomycin per ml (Eagles growth medium or EGM)

2 After 24 hours, wash the cells once with EMEM 3 Thaw one alquot of the seed virus pool and dilute in EGM to contain 10-15

TCID,, per m1 Add 30 ml to the cell-culture flask and incubate at 34°C in a hirmidified chamber with 5% CO, until a complete monolayer is formed (approximately 4 days).

4. Aspirate the EGM from the flask and add 10 ml of EMEM at 37°C. Allow to stand for 5 minutes, and then aspirate the EMEM. Add 10 ml of trypsin diluted to 3 nig/ml in Hank's balanced salt solution without calcium, magnesium or sodiurn bicarbonate. Aspirate all but approximately 0.5 ml of the trypsin from the flask and incubate for 4 minutes

' Stock number VH959, available on request from the Arner~can Type Culture Coiectio!i (ATCC) 12301 Parklawn Drive, Rockville. MD 20852, USA.

'Stock number CCL-10: ava~iable on reques! from ATCC

VIRUS NEUTRALIZATION TEST FOR RABIES

5. Resuspend the trypsinized cells in 30 m1 of EGM and cool in an ice-water bath. Remove a 0.5-ml aliqiiol of cells and seed 5 wells of a 96-well microtitraton plate with 0.1 ml each. lncubate the plate at 34°C in a humidified chamber with 5% CO, for 2 hours.

6. Remove the plate from the incubator and fix the cells with a 1 : 1 mixture of 4% formaldehyde in distilled water and absolute methanol (5). Examine the cells by the fluorescent antibody (FA) test (6). If the proportion of infected cells is significantly less than 1009~b, seed another 75-cm2 flask with 3 X 106 of the cells in 30 ml of EGM, and repeat the above steps.

7 When all the cells are infected, seed five flasks with 1 X 1G6 infected cells per flask and incubate at 34°C in a humidified chamber with 5% CO, until a complete monolayer is formed (about 7 days).

8. Remove the flasks from the incubator and freeze at -70'C. Place the flasks at room temperature until the contents have completely thawed, then agitate and re-freeze. Allow the flasks to thaw at room temperature, and then centrifuge at l000 g for 30 minutes.

9 Pool the resuit~ng supernatant fluid, distribute in 0.5-m! aliquots, and store at - 70 "C

Determination of virus titre

1 inoculate 90 pi of EGM into each of 9 wells in 5 replicate columns of a 96-well microtitration plate. Perform a titration of the virus by distributing 10 p of the challenge virus pool into the first well and transferring 10 p1 serially into each subsequent well, lncubate at 34°C in a humidified incubator with 5% CO, for 1 hour.

2, Add 50 p1 of a suspension of mouse neuroblastoma (MNA) cells, ~ 1 3 0 0 , ' to each well (concentration 6 X l o 5 cells/ml). lncubate the plate for 48 hours at 34OC in a humidified incubator with 5% CO,.

3, Fix the wells as described above, and then perform the FA test. Calculate the TCID,, of the seed virus pool using the method of Reed & Muench (7; see also Chapter 38, Annex). The challenge virus pool should contain between 105.5 and 106 TCID,, per ml.

Determining the working dilution of challenge virus

1 . Dilute the standard reference serum preparation to contain 1 IU per ml. At the New York Department of Health, the United States Standard Immune Globulin is used as reference serum.2 Alternatively, the International Standard for Rabies lmrnunoglobulin or a national reference preparation which has been calibrated against the International Standard, may be used. Prepare 3 columns of serial twofold dilutions of the standard reference serum to 1.128. This is

'Stock number CCL-147, available on reqbesl from ATCC '(Lot R-3, ava~iable on request from the Federal Drug Ad~n~: is t ra t~on, Center for B ~ o i o g c s Evaluat~on and

Research. Bethesda MD 20892, USA 3Ava~iable on request frorn the Deparlrnenl of B~oioglcal Standardlzaton State Serum l l is t tu te, 80

Arnager Boulevard, DK-2300 Copenhagen S. Denmark


achieved by dispensing 50 p1 of EGM into each well, adding 50 p1 of the standard reference serum to the first well, and serially transferring 50 p1 to each subsequent well. Discard 50 pi from the last well in each column.

2. Inoculate each well of the 3 columns with 50 111 of the challenge virus, diluted to contain 102, 103 or 104 TCID,, per ml respectively.

3. Cover the plate and incubate at 34°C in a humidified incubator with 5% CO, for 1 hour.

4. Add 50 y l of the MNA cell suspension (see above) to each well. 5. Cover the plate, incubate as above for 48 hours, and perform the FA test. The

dilution of the challenge virus pool that is completely neutralized by the 1 :32 dilution of the reference serum, and for which virus is present at the 1 :64 dilution, is selected as the working challenge virus dilution. It may be necessary to adjust the dilution of the challenge virus pool slightly to achieve the desired results.

Test samples

The test may be used to detect antibodies in serum, plasma or cerebrospinal fluid samples, which should be stored at - 20°C. The samples should be thawed immediately before the test is carried out and inactivated by heating in a water bath at 56°C for 30 minutes. Light to moderate haemolyss has not been observed to affect cell growth or reproducibility of results.

Serum- virus titration

1. Distribute 50 p1 of €GM into each well of a 96-well microtitration plate. 2. Add 50 pi of the inactivated serum to the first well in each column and mix by

stirring with the pipette tip or by trituration. 3. Transfer 50 pi of the resulting mixture to the next well. Repeat the process for a

total of six wells for the reference serum (five replicates), two wells for the negative control serum, and six wells for each serum under test (Fig. 16.1). Discard 50 pi from the last well of each colunin. When more than one plate is required to accommodate the number of samples under test, it is not necessary to duplicate controls on each plate.

4. Prepare a virus titration column by distributing 50 pi of EGM into the first well and 90 p1 into each of four subsequent wells (see Fig. 16.1).

5. Thaw a vial of the pooled challenge virus and dilute in EGM to the working dilution (see above). D~str~bute In 50-111 volumes into each well containing s e r u m and into the first well of the virus titration column (see Fig. 16.1). Incubate the plates at 34°C in a hiimidifed incubator with 5% CO, for 1 hour.

Preparation of cells

While the serum-virus mixture is incubating, prepare the MNAcelIs for use. Using a 25-cm2 flask containing MNA cells that have grown to 75-950h confluence, dissociate the cells with trypsin (see page 194. step 4). Resuspend the cells in EGM at room temperature and cool rapidly in an ice-water bath. Repeated trituration during cooling will prevent the formation of cell aggregates. Count the cells in a


Fig. 16.1 Plate set-up for the in vitro virus neutralization test 7 -

--p

l 2 3 4 5 6 7 8 9 1 0 1 1 1 2

f i s ) (a l @ @ @ @ @ @ @ ~ @ w ~ y @ 006666(30@@(;:3@ 1 128 I 128 1128 1128 1128

0000003000'30 f7 '

H O O c > O O O : ~ o O c C ; O ~ , ~ f

CVT = challenge virus tltiatlon RS - reference srrurn NS - negatlve serum PS - patient serum

slandard haemocytometer and adjust the concentratio11 to 6 X 10" cells per m by adding cold EGM. Store at 4'C until use

Virus titration and addition of cells

Immediately after incubation transfer 10 i l l of the challenge virus from the first well of the virus titration column to the second well Mix by stirring with the pipette tip or by trituration Repeat the process for the remaining wells Add approximately 50 p1 of the cell preparation to each well (one drop from a 1 m1 pipette) Cover the plate and incubate at 34 C in a humidified incubator wlth 5% CO, for 48 hours

Fluorescent antibody test

1. Using a pipette, aspirate the media from all wells and l111 each with phosphate- buffered saline (PBS), pH 7.6 (see Chapter 20, Annex 2).'

2. After2 minutes, aspirate the PBS and f i l l each well with a 1 . 1 mixture containing 4% formaldehyde in distilled water and absolute methanol. Leave to stand for 1 hour at room temperature.

3. Aspirate the fixative, and wash the plate twice (for 5 minutes) with PBS. Do not reduce the time or number of washes because residual flxative will affect the fluorescein sothiocyanate (FTC)-labelled antibody.

4. Empty the plate, but do not allow the wells to dry. Add 50 y l of FITC-labelled rables anti-nucleocapsid conjugate to each well and incubate the plate for 45 minutes at 34°C in a humidified incubator.

5. Empty the plate and wash twice for 3 minutes with PBS. Empty the plate again,

'The pH should be ad lus t~d to 7 6 u s n g d lute sodlum hydroxide (NaOH)

197


and then distribute 0.05 rnl of PBS into each well and examirie w ~ t h an inverted fluorescence microscope.


Virus should be present in the challenge virus column to 10 ', representing a challenge virus dose of 100 to 1000 TCID,,. If significant departure from this persists, the working dilution of the challenge virus must be redetermined. Virus should also be present in the wells containing the negative control serum at the 1 : 4 and 1 :8 dilutions.

The titration of the reference serum determines the dilution that is the equivalent of 1 IU per m S~nce the challenge virus dose was selected by titration against the reference serum to achieve complete neutralization at 1 :32 but not at 1 : 6 4 this is the usual outcome. However, a variation of one dilution in the neutralization end-point of the reference serum is acceptable. Therefore the highest dilution of the reference serum that completely inactivates the challenge virus in at least three of the five replicate titrations determines the dilution that is equal to 1 IU for each test. If there is greater variability than this, the challenge virus titrat~on must be repeated and the working dilution of the challenge virus adjusted.

The highest dilution of the seruin under test that completely neutralizes the challenge virus, as evidenced by the absence of infected cells in the FA test, s the antibody titre. This is expressed in IU and is determined by comparing the neutralization end-point of the serum under test with that of the standard reference serum. For example, in a test in which 1 IU of reference serum neutral~zes virus to 1 : 3 2 a patient whose seruni also neutralizes virus to 1 :32 has an antibody titre of l IU, while a patient whose serum neutralizes virus to 1 : 16 has an antibody titre of 0.5 IU.

References

1. Koprowski H. The mouse inoculation test. In: Kaplan MM, Koprowski H, eds. Laboratory techniques in rabies, 3rd ed. Geneva, World Health Organization, 1973 (WHO Monograph Series, No. 23): 85-93.

2. Smith JS. Rabies serology. In: Baer GM, ed. The naturalhistory of rabies, 2nd ed. Boca Raton, CRC Press, 1991: 245-247.

3. Biosafety in microbiological and biomedical laboratones. 3rd ed. Washington, DC, United States Public Health Service, 1993.

4. Laboratory biosafety manual, 2nd ed, Geneva, World Health Organization. 1993

5. Rudd RJ, Trimarchi CV. Development and evaluation of an in vitrovirus isolation procedure as a replacement for the mouse inoculation test in rabies diagnosis. Journal of clinical microbiology, 1989, 27: 2522-2528.

6, Trimarchi CV, Debbie JG. The fluorescent antibody test in rabies. In. Baer GM,


ed. The natural hislory oi rabies. 2nd ed. Boca Raton, CRC Press, 1991: 21 9-233.

7. Reed LJ, Muench H. A smple method of estimat~ng ffty percent endponts. American journal of hyg~ene, 1938, 3. 493-497

8. WHO Expert Committee on Bioiogical Sfandard~zation. Thirty-fifth report. Geneva, World Health Organization, 1985 (WHO Technical Report Seres, No. 725).

CHAPTER 17

Competitive ELISA for the detection of rabies virus-neutralizing antibodies L. D. Elmgren' & A. l. wande1er2

One of the mair-i am5 of rabies seiology is to deterrnine the level of virus- neutralizing antibodies in an individual Current WHO guidelines ( I ) indicate that neutralizing antibodies may be measured in miLe or in tissue culture Rabies serology is usually carried out using in vitro ne~~trdl izat ion tests specifically Ihe rapid fluorescent focus inhibition test (RFFIT) (2 see also Chapter 15) The RFFIT being a biological test system has certain disadvantages Specific antibody in a sample is measured by its ability to inhibit viral replication in cell culture therefore any nonspecific factors in the sample that inhibit viral growth will also be measured as neutralizing antibody In addition biological assays require specialized equipment and laboratory faci l~t~es Neutralizing antibody levels may also be determined using mouse neutralization tests however these require the use of an~mals and specialized housing faci l~t~es and are beyond the capabilltles of many laboratories

In recent years many enzyme-linked ~nmunosorbent assays (ELISAs) have been developed for measuring antbod~es directed against both v i ~ a l and bacterial pathogens in serum A number of ELSAs have been described for the detection of rabies antibodies ($5) These tests are both simple and quick to perform with results being available within a few hours Nevertheless, the specificity of these tests is limited to achieve a high degree of spec~fic~ty a highly purified form of antlgen must be employed (6, 7) This often requires elaborate purification procedures which many laboratories are not equipped to per fo~m The cost of such procedures also makes these tests impractical for many laboratories The consequence of not using a highly purified antigen is that antibodies other than those of interest may be measured giving inaccurate results An enzyme-conjugated species- specific antibody is used to detect antibody in these tests thereby limiting their usetulness to sera of a few related species In a laboratory processing sera from many different animal species such a test may not be suitable

Many of the disadvantages of an indirect ELISA can be overcome with a competitive ELISA (C ELISA) C-ELISAs are based on competition between two reactants for a llmited amount of antigen or ant~hody-binding s~tes In a d~rect C-ELISA for measuring antibody levels a known amount of specific indicator antibody (monoclonal antibody conjugated wrth an en7ymej is mixed with the serum under test and allowed to compete for a limited amount of antigen bound to a so ld matrix When the enzyme substrate is added to the system (in the presence of a chromogen) a decrease in colour intensity indicates competition and therefore the presence of antibody in the test serum The decrease In colour

' B~ologist, Rab~es Unit. Agriculture Canada , Animal Diseases iiesearci- institute Nepean, Ontario Canada. Head. Rabies Unit A g i ~ c u l t ~ i r e Canada. Animal Diseases Research Institute, Nepesn Onlario, Canada.

COMPETITIVE ELlSA

lntenslty is proportional to the levei of ant~body unless there 1s complete ~ n h b ~ t o n of the labelled indicator antibody

A C-ELISA using a rabies virus neutralizing monoclonal antibody (MAb) conjugated to an enzyme is ideally suited to rabies serology If a neutralizing MAb is used then only neutralizing antibody in the test serum will be measured The niain advantage of using a MAb in a C-ELISA is that the antigen need not be highly purified because of the highly specific nature of MAbs However a disadvantage of using only a part~ally purified antigen I S that it niay still he infectious

Thts chapter describes a C-ELISA for the detection of rabies virus-neutralizing antibodies using a horse~adish peroxidase-labelled neiitralizing MAb and a partially purlfled rabies virus (ERA strain) antigen The test is both spec~fic and

sensitive and takes about 4 5 hours to complete Using software and techniques developed at the Animal Diseases Research Institute (ADRI) Nepean Canada (L?), the test is semi-quantitative and fully computer automated Moreover the test car? be used to determine the virus-neutralizing antibody levels in the serum of any animal species including humans

Although the test as described here is fully automated (using an automatic plate washer plate reader and computer) it can easily be adapted to the field All that is required are antgen-coated plates conjugated MAb and substrate Simple graduated pipettes may be used to dispense reagents into wells however it may be advantageous to use a single-channel micropipette and a repeating niut-channel micropipette Samples should be handled and appl~ed in the same way as for the automated procedure (see below) Plates may be washed manually by immersing the plates in a tray containing wash buffer and then inverting them over a waste container to remove the buffer Following the addition of substrate qualitative results can be determined by comparing the colour intensity of the test sera with that of the control sera after a predetermined incubation time (approximately 10 minutes) Only sera that have h ~ g h titres will show a clearly positive reaction, however perhaps more importantly negative sera will be readily identified as having no visual colour difference from the negative control

Method

Preparation of antigen

The ERA strain' of rabies virus is grown in BHK-21 C13 cells as described in Chapter 10. Cell-culture supernatants containing high titres (at least 10' TC/DSO) of rabies virus are clarified by low-speed centrifugation at 500 g for 5-10 minutes to remove cellular debris. After clarification, the virus is partially purified by ultracentrifugation at 70 000 g and 4 % for 2 hours. The pelleted virus is resuspended in phosphate-buffered saline (PBS) (see Chapter 20, Annex 2) to a final volume of 1/100 of the original and stored at -70;'C until required.

Preparation of microtitration plates

The rabies antigen is diluted in coating buffer (0.5 moll1 carbonate, pH 9.6; see Annex 1) to a final protein concentration of approximately 50 ~cg/nil. The virus

' Other virus straiqs may be ureferabe, depending on the purpose of :hc !es!


sirspension is then dispensed in 100-~11 aliquots into each well of a 96-well microtitration plate and incubated overnight at 4 "C. Care should be taken to avoid contact with the virus (see Chapter 1 ) . In order to achieve even coating of the wells with the antigen, the microtitration plates should not be stacked during incubation. After 18 hours' incubation. sealed plates may be stored at -70°C with the coating fluid in the wells until required.

Selection and labelling of monoclonal antibodies

The monoclonal antibodies (MAbs) selected for this test should be neutralizing antibodies (i.e. directed against an eptope on the rabies glycoprotein recognized by neutralizing antibodies) of the IgG class.' Purification and labelling (of highly concentrated MAbs) iscarried out by standard methods (9, 10). Purified MAbs may be labelled with an appropriate enzyme (peroxidase or phosphatase). In the C- ELISA described here, a neutralizing MAb, 10EC9 (7), of the IgG, subisotype IS

labelled with horseradish peroxidase (see Annex 2).

Assay parameters

The assay described here is modelled after similar assays developed at the ADRI (8). Controls are measured in quadruplicate and test sera in duplicate (Fig. 17.1). The arrangement of the samples minimizes the effects of within-plate variations due to photometer function and other physical variations associated with microtitration plates ( I 1 ). I f the ADRl ELISA software is used, initial optical density (OD) readings are taken at precisely 4 minutes after the addition of the substrate. Using a standard curve that is part of the ELISA software. the computer calculates (using the 4-minute readings ot the buffer control) the "target" time for reading the plate. The "target" time is the calculated time for the buffer conl ro to achieve an optical dens~ty of 1.000.

Assay procedure

1. Remove the plate from the freezer and thaw the contents of the wells. Discard the well contents and wash the plate 5 times with wash buffer (see Annex 1) using an automatic plate washer

2. Block all nonspecific binding sites in the wells by adding 200 p1 of wash buffer supplemented with 5% horse serum (free of antirabies imm~~noglobul ins) to each well. Incubate for 2 hours at 28°C.

3. During the incubation period for blocking, prepare dilutions of the sera under test and the control sera in wash buffer. Each test serum should bediluted 1:10 in a final volume of 250 pi Control sera should also be diluted 1 10 in a final volume sufticient for the day's testing. Controls consist of 4 types: (I) diluent buffer (TG); (ii) low-titre positive sera (01); (111) negative sera (02); and (iv) high-titre positive sera ( 0 3 ) . All dil~rtions should be preparea and niaintaned

P-

' lnfor-nation or? sources of su~tabie MAbs Tay be o b t a i ~ e d lroirl Veternary Pubiic Health, D v l s o n of Cornrnunlcable Dlseases World Healt'l Oryar l~ ra t~on l 21 l Geneva 27, Sw!tzc:and.

COMPETITIVE ELISA

Fig. 17.1

/ Quad 4 -~

TG = dlluent buffer Q1 = low-tltre pasltlve sera a2 = negatlve sera Q3 = hlgh.ttre positive sera.

Plate set-up for the C-ELISA

at room temperature. If the ADRl ELISA software is used, input the sample identification numbers (see Annex 3).

4. Prepare sufficient quantities of the appropriate dilution (previously determined by checkerboard titration) of the horseradish peroxidase-conjugated MAb in wash buffer.

5. Before applying the samples, wash the blocked, antigen-coated plate 5 times with wash buffer. Remove any residual wash buffer by vigorously tapping the inverted plate onto a I~nt-free absorbent cloth. Wash one plate at a time and do not proceed to the next plate until all the samples have been applied.

6. Dispense 50.~1 volumes of the test and control sera into the wells of the microtitration plate (see Fig, 17.1). Apply the controls first.

7 Immediately following the application of the controls and test sera, add 50 pI of diluted, horseradish peroxidase-conjugated MAb to each well. Seal the plate and place on a rotary shaker for 1 minute to mix the samples.

8. Remove the plate and place in an incubator at 28°C for 2 hours. Proceed to the next plate if more samples are to be tested.

9. I f the test is automated, ensure that the plate reader is switched on 20 minutes before reading. The plates should be read at 414 nm. Turn on the computer and load the C-ELISA program. The plate readers should be compatible with the ELISA software.

10. Between 5 and 10 minutes before the incubation time for the first plate has expired, prepare enough substrate solution (see Annex 1) for all plates (10 ml per plate).

P n f 7

000C>0@0 Test sera 1 - 20

Test sera 21 - 40

2 3 4 5 6 7 8 9 1 0 1 1 1 2

0 0 0 ~ ~ . ~ 0 0 3 ~ v 00000@00C03

a @ @ @ @ @ @ @ @ @ l m Quad 3

00000@00000 00000000000 O C C O O O ~ O C C 3 @ @ @ @ @ @ O O @ @ @

Quad 2

nn

,? r U LJ \d

Test sera 21 - 40

1 Test sera 1 - 20


11. Wash an uncoated microtitration plate 5 times in the plate washer and distribute 100 irl of substrate solution into each weil. Shake this plate on the rotary shaker so that a uniform meniscus is formed in the wells. Use this plate as a blank control for the plate reader (may be reused iridefinitely).

12. Wash the first test plate 5 times in the plate washer Immediately distribute 100 p i of substrate solution into each well and simultaneously begin timing the reaction for precisely 4 minutes. Place the plate on the rotary shaker: after 4 minutes. transfer the plate to the plate reader for the 4-minute reading. Replace the plate on the shaker for the duration of the target time as indicated by the computer. When prompted by the computer, return the plate to the plate reader for the targeted reading.

13. Repeat step 12 for all subsequent plates to be read.


For the test to be valid the control sera for the assay must fall within specified limits Limits may vary between laboratories using d~fferent control sera therefore specific contiol limits should be established in each laboratory At the ADRl the C- ELlSA software uses two standard deviations as the control limits (8) if two or more controls fall outside the specified control limits then the test must be repeated If only one of the controls falls outside the llmits then the plate may be acceptable provided that the other controls are within the specified limits and do not show excessive variability (denoted by an asterisk in the status column of the computer print-out in Annex 3) The optimum development time is approximately 10 minutes but may vary as long as the controls fall w~thin the specified limits

Diagnostic validation

For a test sample to be regarded as positive (P see Annex 3) the intensity of substrate coloration should be inhibited by at least 20% The percentage inhibition is calculated as follows

Percentage inh~bition=lOO-{(OD of test sample1OD of buffer control) X 100)

Samples showing excessive variability should be re-tested. The cut-off threshold for determniiig the status of a sample was identified by

comparing the C-ELISA percentage inhibition values with serum neutralization titres for over 800 sera. At a 20% inhibition cut-off pon t , the C-ELISA showed a

sensitivity and specificity (12) of 99% relative to the virus neutralization test. These findings were subsequently validated further by ROC analysis (13, 14). Such statistical analyses should be performed whenever the C-ELISA is introduced In a particular laboratory or the test conditions are changed

References

1. WHO Expert Committee on Rabies. Eighth report. Geneva, World Health

Organization, 1992 (WHO Technical Report Series No. 824).

COMPETITIVE ELlSA

2. Smith JS, Yager PA, Baer GM. A rapid reproducible test for deterrnning rabies neutralizing antibody. B~illetin of the World Health Organization, 1973, 48.

535-541.

3. Atanasiu P, Savy V, Perrin P. Epreuve immunoenzymatique pour la detection rapide des anticorps anti-rabiqiies. [lrnmunoenzymatic test for the rapid detection of rabies antibodies.] Annals of microbiology, 1977, 128A: 489-498.

4. Nicholson KG, Prestaye H. Enzyme-linked immunosorbent assay, a rapid reproducible test for the measurement of rabies antibody. Journal of medical virulc~yy, 1982. 8. 43-45

5. Barton LD, Calnpbell JB. Measurement of rabies-specific antibodies in carnivores by an enzyme-linked imniunosorbent assay, Journal of wildlife diseases, 1988, 24: 246-258.

6. Perrin P et al The influence of the type of immunosorbant on rabies antibody EIA: advantages of glycoprotein over whole virus. Journal of biological standardization, 1986, 14. 95-1 02

7. Grass M, Wandeler Al. Peterhans E. Enzyme-linked immunosorbent assay for determination of antibodies to the envelope glycoprotein of rabies virus. Journal of clinical microbiology, 1989, 27: 899-902.

8 Nielsen K et al Erlzynle immunoassay application to diagnosis of bovine brucellosis Ottawa Supply and Services Canada 1992

9. Harlow E Lane D. Storing a ~ i d purifying antibodies. In Antibodies:a 1aboratc1r.v i ~ x ? n i ~ a / . Cold Spring Harbor, New York, Cold Spring Harbor Laboratories. 1988: 309-31 2.

10. Harlow E, Lane D, Labelling antibodies, lr!: Ant~bodies: a laboratory manual. Cold Spr~ng Harbor, New York. Cold Spring Harbor Laboratories, 1988: 31 9-358.

11. Stemshorn BW et al. A computer-interfaced photometer and systematic spacing of duplicates to control within-plate enzyme imrnunoassay varatioii. Journal of immunological methods, 1983, 61 : 367-375.

12 Martin SW Meek AH, Willeberg P Measurement of disease frequency and production In Veterinary epidemiology principles and methods Ames, A Iowa State University Press, 1987: 48-76

13 Kraerner HC Assessment of 2 x 2 associations generalization of signal- detection methodology The Amencan statistician, 1988 42 37 49

14. Metz C€. Basic principles of ROC analysis. Seminarsin nuclear medicine, 1978, 8: 283~-298.


Annex 1 Preparation of buffers and reagents

Coating buffer: 0.5 moI,'/ carbonate buffer, pH 9.6 Sodium bicarbonate (NaHCO,) Sodium carbonate (Na,CO,) Distilled water to make

Dissolve the sodium bicarbonate in distilled water and add sodium carbonate until the desired pH is reached. Prepare fresh as required.

Phosphate buffer, pH 7.2 Sodium chloride (NaCI) Sodium phosphate, monobasic (NaH2P0,.H20) Sodium phosphate, dibasic (Na2HP0,.7H,O) Distilled water to make

Wash/dlluent buffer Phosphate buffer, pH 7.2, prepared as above, supplemented with 0.15 rnol/l sodium chloride and 0.05% polysorbate 20.

2,2'-Azino-bis (3-ethylbenzthiazoline-6-sulfonic acid) (ABTS), 40 mmol/l ABTS 233 mg Distilled water to make 10 ml

Substrate buffer: 0.05 movl citrate buffer, pH 4.0 Citric acid 9.6 g Distilled water to make 1000 ml

Dissolve the citric acid in 900 ml of distilled water and adlust the pH to 4 0 by adding sodium hydroxide (NaOH) pellets Make up to l litre and store at 4°C Warm to room temperature before use

Substrate solution ' Substrate buffer, prepared as above ABTS. prepared as above Hydrogen peroxide (H,O,) 3% stock solution

100 m1 0 25 ml 0 05 rnl

This gives a final concentration of l mmoI/'l s~ibstrate (H,O,) and 4 rnrnol;l chrornogen (ABTS). The solution should be prepared nimediately before use and kept in a dark bottle until dispensed

' Oual;tlt~es refer lo !he amount requ~red o e ~ p'ate

206

COMPETITIVE ELlSA

Annex 2 Conjugation of monoclonal antibodies

Monoclonal antibodies (MAbs) are conjugated to horseradish peroxidase using the periodate-oxidation method as modified by Henning & Nielseri.' Briefly, the procedul-e is as follows:

1. Dssolve 10 mg of horseradsh peroxidase type VI. in 2.5 m1 of distilled water. Add 0.5 ml of 0.1 mol:' sodium nietaper~odate (NalO,) and stir for 20 minutes at room temperature.

2. Dialyse the resulting horseradish peroxidase aldehyde (green in colour) against 1 nirnol:'l acetate butfer, pH 4.4 at 4°C for 2 days.

3. After dialysis, add 50 j ~ l of 0.5 mo1I carbonate buffer, pH 9.5, to the horseradish peroxidase followed by 2 mg of MAb in carbonate buffer (coi?centration 1 mg/mI) to give a final volume of 5 ml.

4. Gently stir the mixture for 2 hours at room temperature before stopping the reaction with the dropwse addition of 250 111 of ascorbic acid (4 mgiml).

5. Allow the mixture to stand for 2 hours at 4 % and then dialyse against PBS for 2 days at 4°C.

Annex 3 ELlSA software

The ELISA software used is a BASIC program that automates the reading of plates (Titretek plate readers only) and the management of test results for many different ELlSAs developed at Agrcuture Caiiada's Animal and Plant Health Laborator~es.~

The table below contains the C-ELISA test results for 40 patient sera examined at the ADRI.

Sample results obtained by the C-ELISAa

File name: 9401 1401 Plate no.: 1 Development time: 9 84 rn nutes

Test date: 01-14-1994 Test time: 10 53 56 Technician: LE

CONTROLS

ID OD of sample Mean SD Status % Inhibition

Quad 1 Quad 2 Quad 3 Quad 4

' He.it.o(i DM N e s e n K t i Peioxiiiasc-iaaelicd r r~orocor ia i a r~ t lbod~es for ~ s e 11 enzyrre imrndroassay Journal of fi~lr~iuilodss,iy 1987 8 297 307

2 F u ~ t l i e v infotmatton aboul ! t ~ e sof:ware rna) ae obtalried t r o l l the Methods Deveoamer t and Transfer i lnlt l . l- lr~~noiog); S e c t ~ o r ~ . Ar~itr lai Osease R e s ~ : ~ r c i i ~,is!~tu'c 3851 Fa ldi?.ileio Road PO Bok 11300 Sta!or. H Nepear~ O! i ta io K2 t i 8P9 Caridcia

TEST SAMPLES

I D OD of sample Mean SD Status % Inhibition

Quad 1 Quad 2

ID OD of sample Mean SD Status % Inhibition

Quad 3 Quad 4

7 1 0 755 0 700 0 727 0 039 P 27 22 0 833 0 862 0 847 0 021 15 23 1 050 0780 0915 0 191 8 24 0 839 0670 0755 0 120 P 24 25 1012 0 875 0 944 0 097 5 26 0 440 0 488 0 464 0 034 P 53 27 0 154 0 037 0 096 0 083 P* 90 28 0 209 0181 0195 0 020 P 80 29 0 379 0241 0310 0 098 P 69 30 0 238 0737 0235 0 004 P 76 3 1 1161 0795 0978 0 759 2 32 0813 0739 0776 0 052 P 22 33 0916 0832 0874 0 059 17 34 0 953 0 952 0 953 0 001 4 35 1 073 0723 0898 0 247 10 36 0 887 0864 0876 0016 17 37 0 803 0 793 0 798 0 007 P 20 38 0 646 0722 0684 0 054 P 3 1 39 0 632 0384 0508 0 175 P 49 40 0 735 0854 0795 0 084 P 20

I'=p~,certage of ~nhib i t io i n~ less b a r Z o o 0 01 = I O A 'itrp ~;.os!ive sera 0 2 = negatlvi. se-d Q3= tiigh ti'rp positive sera l C=dilueni bu fer (buf f? ' ~ o r i t r o l ) aResbl+s are cornDilger generated and have been rounded to 3 decimal places *Indicates sample showin6 exressive va rab i ! y

CHAPTER 18

Electron microscopyl K ~uni rne ler* & P. ~ t a n a s i u ~

The electron niicroscope is essentially a research tool and only rarely will it be necessary to employ it for diagnostic purposes in preference to the light microscope. It is of practical value in the examination of vaccine strains for the detection of contaminants not readily recognized by other means, In rabies research, the electron microscope has played a key role in elucidating the structure and the morphogenesis of the virus in experimental animals and in tissue cultures.

Structural studies In longitudinal section and in negative contrast, the intact virion resembles a bullet (Fig. 18.1). It measures 180 X 75 nm. A fringe of surface spikes can be seen and the surface itself appears as a dense mass with occasional particles exhibiting a distinct honeycomb-like arrangement. The spikes are about 6.0-7.0 nm long.

The internal nucleoprotein exists in helical form (Fig. 18.2). When isolated from disrupted vr ions it appears as a single-stranded right-handed helix 1 pm in length and 16 nm in diameter, with a periodicity of 7.5 nni. The helical structure of the nucleocapsid is extremely labile. The uncoiled strand of the helix is about 4.2 pm long with a varying width of 2-6 nm, depending on the angle of viewing.

Studies of morphogenesis

Experimental animals

Two virus strains were iised: the Pasteur strain passaged in rabbit brain and a wild strain isolated from the brain of a fox. Suckling mice 4-6 days old were inoculated intracerebrally. The incubation period for fixed rabies virus was 4-6 days and for street rabies virus 10-20 days. The animals were killed when they showed signs of imminent death.

The Aininons horn from each side of the brain was removed after cooling to 4°C. One part was kept for direct examination, after emulsification in phospho- tungstic acid, and the other part was used for light microscopy of inclusion bodies and thin sectioning.

Fragments of the Ammon's horn were fixed for 2 hours at 4OC in 4% glutaral (glutaraldehyde), pH 7.2. The fragments were cut into small cubes and then fixed

' Based on the chapter n the previous edition, w i i c h was prepared by K Hum~neier and updated by the late P. A t a n a s ~ ~ i Director D~v is~on of Expermental Pathology. The Children's Hosp~tal of Pli~iadeiphia School of Med- cine. University of Perir~sylvari ia, Philadelphia PA IJSA Deceased Former Honorary Professor Pasteur I ns t~ tu ie , P a r s France


Fig. 18. I Rabies vifion in negative contrast

for 18 hours in isotonic osniium tetroxide (osmic acid). 1 % solution, pH 7.2. After dehydration, the cubes were embedded in a mixture of an epoxy resin and an epoxy resin adhesive. Thin sections were stained for 60 minutes in a 596 solution of urariyl acetate and for 20 seconds in absolute lead citrate and then examined under the electron microscope.

Virions were rarely seen in negatively stained en7usions of Animon's horn cells. These virions otten possessed a double membrane, projections and transverse striations. Occasionally the nucleocapsds extri~ded from the caudal part of the virion (Fig. 183) The size and shape of these virions were identical with those derived from tissue cultures (see Fig. 18.1).

The evolution of lhe infection appeared to differ considerably, depending on the virus strain used. Fixed virus cai~sed tlie appearance of specific inclusions, or matrices, in the pyramidal cells of the Animoii's horn. The fine structure of these inclusions was iiomogeneous, consstitig of thin filan~ents (3.0-5.0 nm). Virions were found only at the perphery of the ncusions, where they formed on

membranes of the endoplasmic retculum and ergastoplasmic vesicles. These virions can be seen in transverse and longitudinal sections in Fig. 18.4.

This liind of morphogenesis was rarely observed with street virus. The niorphology of the niatrix or cytoplasm of the pyran~idal cells tended to be heterogeneous. Besides the characteristic filaments--presumably the nucleoproten-coiisderable condensailon of the material in the matrix was observed. The density increased with time and membranous structures appeared. These struc-

ELECTRON MICROSCOPY

Fig. 18.2 Nucleoprotein isolated from rabies virions

tures were 25-30 nm in darneter and frequently exhibited a transverse striation Distinct double membranes without dny relation to the pre-existing cell membrane structures also emerged These double membranes often aligned in pairs and the gap between then? (50 60 nm) corresponded to the internal diameter of a complete virlon (Fig 18 5)

These findings suggest that during the morphogeness of street virus in the Ammon s horn of the rnouse not only repl~cation of the nucleocaps~d strands but also the formation of the envelope and the assembly of the virus particle take place within the matrix

Tissue-culture studies

The fixed rabies stra~ns CVS arid Flury HEP were used in these studies Monolayers of baby hanister kidney line 21 (BHK-21) cells were ma~ntained in

Eagle S basal medium (see Annex) in Hanks solution supplemented with 10% fetal calf serum (EBM IOFCS)

LABORATORY TECHNIOUES IN RABIES

Fig. 18.3 Rabies virion after emulsification of Ammon's horn pyramidal cells (negative staining)

After removal of the niedum, the cell rnonolayers were washed and exposed tor l hour at 37°C to the virus at an input multiplicity of 5-10 plaque-forming units (PFU) per cell. After virus adsorption the cells were washed, EBM-IOFCS was added, and the cultures were incubated at 34 C io be harvested at predetermined ~ntervals.

Tlie harvested cells were washed and fixed in 3% glutaral for 1 hour. Subsequently they were fixed in 1% osmium tetroxide pH 7.2, dehydrated and embedded n epoxy resin T h n sectlons were s taned wlth u rany acetate and lead

cltrate (see page 210) and then examined under the electron microscope. In [issue cultures, the first structilral changes became obvious 6-7 hours after

infection. These changes consrsted in the appearance of granular material, whlch later became i~larnentous and replaced the cell organelles. Virus particles later appeared in or around these matrices (Fig. 18 6) This morphological evidence of virus replication in the cytoplasm is identical with the inclusions seen in neurons in tthe brain of infected animals The filanientous strands are present in the cytoplasm in helical form, similar to those in the virus particles, arld thus constitute the nucleoprotein of the virus. Dur~i iy virus inaturation on pre-existing cell membranes. these strands are incorporated into the virions. With some attenuated vlrus strains,

ELECTRON MICROSCOPY

Fig. 78.4 Fixed virus: thin section of a pyramidal cell of the Ammon's horn (6 days after infection)

X 50 000 The stiuctuie of the rnatnx is homogeneoiis The v~ r~ons form on pre-ex~shng cell niembranes

however a different morphogenesis can be observed Virus particles emerge on the per~phery of the matrices and inside them without obvious involvement of pre- existing cell membranes At present it is riot clear whether these differences in morphogenesis which ale also evident in brain cells of experimental animals reflect differences in biological characteristics or whether they represent different degrees of attenuation to a given cell Nevertheless the experimental results described in this chapter do siiow that all the distinct features of the rnor phogenesis of rabies viruses seen in neurons of the animal brain can also be observed in cells maintained in culture


Fig. 18.5 Street virus: thin section of pyramidal cells of the Ammon's horn (6 days after infection)

X 80 000 Double membranes arid complete wnons are observed In the matrix.

ELECTRON MICROSCOPY

Fig. 18.6 Replication of rabies virus in pyramidal cells in in vitro culture

X 32 000 M = matnx; VP = virus particles; N = nucleus.


Annex Medium for agarose: Eagle's basal medium' (EBM)

Final composition of medium Stock solutions *

Coniponeni Sodium chlorde (NaCI) Potassiurn chloride (KCI) Sodrum phosphate, nionobasic.

rnonohydrate (NaH,PO,.H,O) Magnesum sulfate heptahydrate

(MSSO,.'IH,OI Calcium chloride, anhydrous

(CaCl2) Glucose

L-Argnine hydrochloride L-Cystine L-Tyrosine L-Histidi~ie L-Isoleucine L-Lysne L-Methonine L-Phenylalanine L-Threonine L-Tryptop han L-Valne

Biotin Folrc acid Choline chloride Ncotnamide Calc~um D-pantothenate Pyridoxal hydrochlorde Thrarnine hydrochloride Rrboflav~n I-lnositol

rng/litre 6800.0 Soiution A. Balanced salt 400.0 solution (Earle's), 10-fold

concentrate (withoiit 140 0 sodium brcarbonate and

phenol red) 200.0

21 0 Solufion B. Amno acids, 12.0 100-fold concentrate 18.0 8.0

26.0 26.0

7.50 16.50 24.0

4 0 23 50

1 .O Solution C. Vitamns. 1 .O 100-fold concentrate 1 .o 1 .o 1 .o 1 .o 1 .o 0.10 1 80

292.0 Soiution D: Glutamne 100-fold concentrate

Sodurn bicarbonate (NaHCO,) 2200 0 Soluhon E Sodium brcarbonate 5 6 g]l00 m1

' E a y c H Nulritton needs of ~ ~ ~ a ~ ~ r n a l i a r i cells n cu!ure SClRfici?. 1055 122. 501 504 The ~ t o c ~ c sou!ions s i e d here arc available in the i o r r of concen!rates. :he strenyths r~d ica teo are rel;itive to t i le f n a co!mpost~on of tPe medium shown on t i le left

ELECTRON MICROSCOPY

Quantities required for preparation of a 2-fold concentrate ' of EBM

S t e r ~ l e demlneralized water S t o c k s o l u t i o n A S t o c k s o l u t i o n B S t o c k s o l u t ~ o n C S t o c k s o l u t i o n D S t o c k s o l u t i o n E P e n i c i l l i n and s t r e p t o m y c i n . l00000 lU/ml

M y c o s t a t i n 50000 IUiml Fetal calf s e r u r n

616 ml 200 m1 20 m1 20 rnl 20 m1 80 m1 2 m1 2 ml 40 m1

' When diluted with an eauaivolu~ne of I O . 6 agarose in sterile demineralized water, ihisconcentra!e gtves a solut~on having Ihe final compos~l ion shown opposlte

m Methods of vaccine

production

Brain-tissue vaccines

CHAPTER 19

General considerations in the production and use of brain-tissue and purified chicken-embryo rabies vaccines for human use F -X Mestin ' & M. M. Kaplan2

Introduction

In 1988, of the 112 countries that responded to the WHO World Rabies Survey ( l ) , 32 (28%) reported produc~ng vaccine for human use. Among those countries. all those in Africa and Asia and 90% of those in Central and South America reported producing rabies vaccines from b r a n tissues, either from adult or suckling mammals. Human vaccines prepared from b r a n tissues represented 85% of the doses produced or imported by the countries that responded to the survey.

It should be noted that neither China nor tile former USSR replied to Ihe above survey. In the early 1980s, these two countries abandoiied brain-tissue vaccine production in favour of cell-culture vaccine production, It is estimated that between 4 and 5 million doses (2, 3) of vaccines prepared in primary hamster kidney cells were used by these two countries in 1988. This suggests that cell- culture vaccines may have represented about 35% of the vaccine doses used worldwide in 1988. In the countries that responded to the above survey, more than 60% of the doses of veterinary vaccines produced or imported in 1988 were prepared in cell colture.

Thus, rnost of the rabies vaccines used for post-exposure treatmenl of humans are still produced from brain tissue. The Semple-type vaccine remains the most widely available rabesvaccine in the world. In 1973. the WHO Expert Committee on Rabies recommended that no vaccine containing living virus should be employed in hc~rnans and that the production of Fermi-type vaccines. which contain residual living v~rus, should be discont~nued (4, 5) The use of Hempt vaccine was discont~nued at the end of the 1970s following a report of post-exposure treatment failures (6).

Post-exposure treatment with brain-tissue vaccines may induce severe neurological complicat~ons, and vaccine failures are more frequent than with the new generation cell-culture and highly purified duck-embryo vaccines. However, highly potent and entirely safe rabies vaccines prepared on cell culture are more difficult to produce than bra~n-tissue vaccines, and are more expensive. even when they are produced locally (7).

' Chief Veler~riary P ~ ~ b l r Health, D~vls ion of Comniunicabie Diseases, Worid Heaitti O r g a r ~ z o t ~ o n , Geneva. Sw1!7erland Fornicr Drector. Research Promotion and Development World Heaitti Organiia!~on. Geneva Swit7er- larlci


Adverse effects of brain-tissue vaccines

Vaccines prepared from adult brain tissue

Neurological reactions such as enceplialomyelilis and poyneuritis have been reported in humans following administration of multiple doses of rabies vaccine prepared from the biains of adult animals Rates of neurological complications associaled with the use of adult brain-tissue vaccines mainly Semple-type reported in the literature vary greatly from 1 case per 142 treatments to 1 case per 7000 treatments (8-16) In the best surveys conducted among large numbers of patents (17000-24000) in one institution in charge of post-exposure treatment complcatior? rates varied from 1 per 1200 to 1 per 3000 ( 9 15) The case-fatality rate among these pat~ents was reported to vary from 0 02% to 10% (9 11)

In 1938 it was siiggested that such neurolog~cal reactions were associated with an inimune response to neural antigens contained in the vaccine To study this phenomenon further numerous trials were carried out in which experimental allergic encephalomyelitis (EAE) was induced in several laboratory animal species by injecting brain material (17-20) These animals developed clinical signs and histopathological lesions similar to those observed in humans following vaccna tion The encephalitogenic antigen II? EAE was subsequently identified as myeln basic protein (21-23) Recent studies of allergic encephalomyelitis nd i lced by vaccination have suggested that other brain substances may also play a role as enceplialitogens (24-26)

Vaccines prepared from suckling-animal brain tissue

In order io prevent such neurological complications a number of scientists attempted to develop nerve-tissue vaccines using newborn suckling mammals as their brains were considered to be free of myeliri In 1955 in Chile Fuenzaida & Palacios developed a technique for the production of nerve tissue vaccine using suckling n ice aged 3 and 5 days at the time of inoctjlation (17) Other groups worked on sucklirg rabbits and rats Suckling rat b r a n vaccines were developed In the 1960s in thp USSR and were widely used u n t ~ l the late 1970s (28) In 1980 these vaccines were replaced by a coricentrated piirified cell culture vaccine produced on primary Syrian hamster kidney cells (5) A suckl~ng-rabbit b r a n vacune was developed by the National lnst~tute of Public Healtki r i the Netherlands in 1964 The immi~nogencrty of this vaccine was found to be appropriate for the post exposure treatment of humans when administered in 14 15 daily injections (29 30) However the vacc1ne was never used on a large scale On the other hand s~ckl iny-mouse b r a n (SMB) vacLries continue to be widely iised for the post- exposure treatment of humans in Latin America About 2 5 million doses of such vaccines were produced i r i eight Latiri Arnerican countries in 1988 (2)

The use of SMB vaccine for post-exposure treatment was introduced in Chile in 1963 By 19G5 this vaccine was also being used in other Latin American countries In 1967 11 cases of neurological complications associated with SMB vaccine were reported in Venezuela and led to a thorough study of the possible causes

These studies ruled out a viral etology and demonstrated that myeln was present in preparations of brairi from mice aged 9 days The preparation was found to be encephalitogenic in guinea-pigs Histopathological lesions of EAE were dlso found in guinea-pigs inoculated with this material These results strongly

BRAIN-TISSUE VACCINES

suggested that SMB vaccines were potentially encephaitogenc for humans (31, 32).

In addition, a survey was carried out of the incidence of neurological complications associated with the use of SMB vaccine in Latin America. Between 1964 and 1969. a total of 40 cases were reported in eight countries. On average, the rate was one case per 8000 patients in the countries covered by the survey (33).

Most (69%) of the neurological complications reported in the above survey involved the peripheral nervous system. Among 32 patients with such coni- plicatons, the case-fatality rate was 22'10 In contrast, ihe case-fatality rate among 90 patients with ne~irological complications associated with rabies vaccines prepared from adult brains (Fermi, Hempt and Sempe types) was only 4.8% (34).

Under such conditions, although the incidence of neurological complications was lower with SMB vaccir~es than adult-bran rabies vaccines, the former could be considered only marginally safer

Vaccines prepared from embryonating eggs

The followir~y sectior? refers to vaccines produced on embryonatlng eggs. Vac- cines prepared on primary cells from avian embryos (e.g. chick erribryo fibroblasts) are discussed in Chapter 27.

Duck-embryo vacone In 1955 Peck et a1 attempted to develop a rabies vaccine from duck embryos (34) A duck embryo vaccirie first becanie commercially available in 1957 and represented most of the rabies vaccine doses administered in the USA in 1973 Principles for the production of duck embryo vaccine were described in the previous edition (35) However this vaccine still contained neural and other duck- embryo antigens a r ~ d reports of reactions among vaccinees (36-38) drew the attention of the United States health authorities Local reactions were reported in all of those receiving a full course of 14 daily doses However a study carried out by the Centers for Disease Control (CDC) showed that the incidence of neilrological complications among vaccinees in the iJSA was very low (3 1 per 100000) (39) A slightly higher rate ( 4 4 per 100000) was reported among vaccinees who had received the complete course of mmunrat ion (40) Between 1958 and 1975 a total of 21 cases of major neurological complicat~ons with two deaths were reported in a series of 595000 recipients of duck-embryo vaccirle (41) Because of its relative safety this vaccir~e was used for pre-exposure immunization of professional groups at risk (42) However, a review of the use of duck-embryo vaccine for pre- exposure immunization and posl-exposure treatment revealed a lack of consensus in the scientific community regarding its efficacy (43 44) A number of reports cast doubts on the efficacy of this vaccine and the vaccine was eventually withdrawn from the United States market The tirsi hiiman trials using h i ~ m a n diploid cell (HDC) raD8es vacLine in the USA were initiated by the Centers for Disease Control in 1978 The safety and efficacy of HDC vaccine contributed to the disappearance o f non-purified non-concentrated duck-embryo vaccine from the United States market (45) However a group still confident in the mmunogenicity of this product (46) continued working on its improvement by increasing its potency and conconiitantly reducing the number of doses requlred from 14 to 5 or 6 and increasing its safety by extracting most of the non-essential duck-embryo antigens


(47, 48). The production process for concentrated, purified duck-embryo vaccine is described in Chapter 22

Chick-embryo vdccir~e Studies on the low egg passage (LEP) Flury strain of rabies virus adapted to grow in chick embryo (49) showed that most of the virulence was lost between the 176th and 182nd passage The higher passage did not induce rabies wheri given intracerebrally to adult mice rabbits and dogs This high egg passage (HEP) strain was also shown to be effective and safe in dogs and cattle Accordingly a rabies vaccine for human use was prepared using this HEP Flury strain

However studies showed that the efficacy of this vaccine was strictly dose related and that it contained large amounts of foreign proteins constituting an inoculation hazard (50) At its s~xth meeting the WHO Expert Committee on Rabies recommended that live attenuated rabies vaccines be restricted to animals (4)

Efficacy of brain-tissue vaccines

Evaluation of the efficacy of adult brain-tissue rabies vaccine under field conditions led to equivocal results (51). There have been only two large-scale studies carried out in persons exposed to rabid animals, which included untreated "control groups" (16, 52). The tirst study was carried out by the Pasteur Institute of Southern India, Coonor in the 1960s and summarized results acquired on about 900 persons treated (conipletely or incompletely) after exposure to rabies (52). The second was conducted in Thailand in 1984 and included data on 661 persons treated after exposure to animals with proven rabies (16). The studies showed that a death rate of 0-13.4% could be expected with adult brain-tissue vaccines, depending on the severity of exposure and whether or not the full course of treatment was completed in people exposed to rabies. In contrast, Ihe death rate among untreated "control groups" was 45-57%.

In addtiorl to severe neurocomplications, vaccine failures related to the low immunogenicity ot some nerve-tissue vaccines (Semple and Hempt types) have been reported. In Germany, at least 18 out of the 49 cases of human rabies reported dur~ng the period 1951-1976 occurred in spite of timely and complete post-exposure treatment with Hempt and Semple vaccines (6). In Thailand, a total of 169 cases of treatment failure were reported with Pasteur and Semple-type vaccines between 1913 and 1970, representing slightly more than 1 failure per 1000 treatments (53).

On the bass of the inducl~on of v~rus-rieutraliz~rig antibodies in pre- and post-

exposure treatments and the absence of focal or general reactions in vaccinees, the superiority of cell-culture vaccines and purified avian embryo vaccines over

brain-tissue vaccines and non-purlfled avian embryo vaccines has been clearly demonstrated (51. 54-59).

Recent developments in brain-tissue vaccine production

The risk of neurological complications following administration of brain-tissue

vaccines IS clearly related to the number of doses njected (17, 21, 33, 60), In addition, local reactions to the vaccine rnay prevent as many as 60% of patents


from receiving the full course of treatment (61). Efforts to improve these vaccines therefore aimed at reducing the quantity of encephalilogenic substances and non- essential proteiiis in the vaccines without affecting their potency, and at using fewer doses of vaccine of an increased potency if possible.

During the past four decades, many efforts have been made to improve SMB vaccine through the development of purification techniques and the definition of the most iminunogenic routes and schedules of vaccine admiriistration. Relatively little has been acliieved, however. to improve adult brain-tissue vaccines.

In India, the qiiantity of Semple-type vaccine administered to patients has recently been reduced (62) (see also Chapter 20)

The Pan American Zoonoses Center (CEPANZO) demonstrated thdi reduced schedules of SMB vaccine of high potency comprising seven injections and two or three boosters induced virus-neutralizing antibody titres equivalent to those associated with a conventional schedule of 14 doses and two boosters (63, 64) In addition CEPANZO developed a technique for the reduction of the myelin content by centrifugation which did not affect vaccine potency (65)

The Fuenzalida-Palacios technique has been modified in an attempt to reduce further the level of encephaitogenc substances in the final product. Vaccines for human use are now prepared using mice no older than one day at the time of inoculation (see Chapter 21). Centrifugation of the brain suspension at 17000 g is still recommended. Other techniques for the preparation of purified concentrated SMB vaccine combining centrifugation and chromatographic procedures were evaluated earlier (66). Although proinisng, production of this purified vaccine did not go beyond the experimental stage (67).

Attempts were also made to develop techniques for removing encephalitogenic substances frorn rabies-infected adult brains. Although encouraging results were obtained in the laboratory using chlorofluorocarbons no industrial applications were ever made (68 69) as the first reports on the safety and efficacy of the HDC vaccine were being published (70 71) which discouraged further research

AI its eighth meet~ng the WHO Expert Committee on Rabies reconimended Lhat the use of adult brain-tissue vaccine be discontinued wherever possible (72). With regard to countries where Lhe demand for human rabies vaccine is likely to decrease in the near future, such as those in Latin America, the use of SMBvaccine may still represent the most cost-effective choice for the prevention of human rabies (73). The World Health Organization, considering economical and techno- logical constraints assoc~ated with inajor changes in production techniques, cooperated with a number of cocintries in Africa and Asia to develop "intermediate technologies" aimed at increasing the saiety and potency of Semple-type vaccines. When compared with the original process for the preparation of Semple- type vaccine, the major suggested changes were:

1 Use of p-propiolactone (BPL) as an inactivating agent (see Chapter 20) instead of phenol.

2. Use of young adult animals, which increases virus yield and subseqi~ently gives vaccines of higher potency.

3. Freeze-drying of the final product, and bottling in single-dose vials.

An increased potency should perniit reduction of the volume per dose (e g. from 5 m to 2 ml per dose of 5 % sheep brain vaccine) and also of the number of


injections required for a full post-exposure treatment (from 14 to 10 daily ooses) However no modifications should be rnade unless the new dosage andtor reduced vaccination protocol have been shown to induce in ethically acceptable trials appropriate levels of virus neutralizing antibody in volunteers and after- waids to confer piotection under f~eld conditions Whenever possible vaccnes prepared from adult animal brains should be replaced by vaccines prepared In cell cilltiire

References

1. Worid rabies sui-vey. Geneva, World Health Organization, 1991 (unpublished document WHO/Rabes/91.202; available on request from the Division of Communicable Diseases, World Health Organization, 1211 Geneva 27, Switzerland).

2 i n FT. The protective effect of the large-scale use of PHKC rabies vaccine in humans in China. Bulletin of the Wofld Health Organizatioil, 1990.68 449-454.

3 Selmov M et al. Specific activity of concentrated and purlfled cell culture rabies vaccine (CPCRV) from the strain Vnukovo-32-107 in an experiment with lherapeutical immunization of humans. Journal of hygiene, epidemiology, microbiology and immunology. 1982. 26: 83-94.

4. WHO Expert Committee on Rabies Sixth report. Geneva World Health Organi- zation, 1973 (WHO Technical Report Series. No. 523).

5. Lepine P. Fermi-type vaccine. In: Kaplan MM, Koprowski H, eds, Laboratory techrliques in rabies. 3rd ed. Geneva, World Health Organization, 1973 (WHO Monograph Series. No. 23): 199-200.

6. Kuwert EK et al. Introductory remarks: stumbles and leaps in rabies vaccination of man with special regard to the FR Germany. In: Kuwert EK, Wiktor TJ. Koprowski H, eds. Cell-cuitui-e rabies vaccines and their protective effect in man. (Proceedings of WHO Coi~su/tat io~?s, Essen, 5 7 March 1980.) Geneva, International Green Cross. 1981 31-51

7, Horack HM. Allergy as a factor in the development of reaction to antrahies treatment American journal of the medical sc;iences. 1939, 197 672.

8. Pait CF, Pearson HE. Rab~es vaccine encephalomyelitis in relation to the incidence of rabies in Los Angeles. Americanjournal of public health, 1949, 39: 875-988.

9. Greeriwood M. Tenth report on data of anti-rabies treatments supplied by Pasteur Institutes. Bulletin of the Health Organisation, 1945146. 12: 301

10. Sellers TF. Complicatiori of rabies treatment. Journal of the Medical Associa- tion of Georgia, 1947: January.


11. Sellers TF Rabies, the physician S dilemma. American journal of tropical inedicine, 1948, 28: 453-456.

12. McFadzean AJS. Choa GH. The neuroparalytic accidents of antirabies vaccination. Transactions of the Ro)/al Society of Tropical Medicine and Hygiene, 1953, 47 372--386

13. Blatt NH et al. Reactions following anti-rabies prophylaxis. Americanjournaloi diseases of children, 1953, 86. 395-402.

14. Appelbaum E, Greenberg M Nelson J. Neurological complicatiorls following antirabies vaccinat~on Journal of the American Medical Association. 1953, 151: 188-191.

15 Hemachudha T et a1 N e u r o l o g ~ a cornpiications of a Semple type rabies vaccine c l ! n c a and immunological studies Neurology 1987 37 550-556

16. Sitthi-Amorn C et al. The diagnostic properties of laboratory tests for rabies. International journal of epidemiology, 1987. 16: 602-605.

17 Hurst EM The effect of injection of normal brain emulso~? into rabbits with special reference to the aet~ology of the paralytic accidents of anti-rabic treatment Journal of hygiene 1932 32 3 44

18. Kabat EA, Wolf A, Bezer AE. Rapid production of acute disseminated encephalomyelitis in Rhesus monkeys by injection of brain tissue with adjuvants. Science, 1946. 104: 362 363.

19. Kabat EA. Wolf A. Bezer AE. Tile rapid production of acute disseminated encephalomyelitis in Rhesus monkeys by injection of heterologous and homologous brain tissue with adjuvants Journal of experimental medicine, 1947, 85: 117-1 28.

20. Jervis GA. Experimental allergic encephalitis in animals. and its bearing upon the etiology of rneur-oparalytic accidents following antirabies treatment in man. Bulletin of the World Health Organization, 1954, 10: 837-844.

21 Eylar EH, Hashim GG. Ailergic encephalomyelitis: the structure of the encephalitogenic deterrnnant. Pr-oceediiigs of the National Academy of Sci- ences of the United States of America, 1968, 61: 644-650.

22. Eylar EH et al. Experimental allergic ericephalomyelitis (EAE)-an encephalitogenic basic protein from bovine myelin. Archives of biochemistry and biophysics. 1969. 132: 34-48.

23. Eylar EH, Caccam J, Jackson JJ. Experimental allergic encephalomyelitis: synthesis of disease-indijcing site of the basic protein. American Association for the Advancement of Science, 1970, 168: 1220-1 223.


24. Hemactiudha T et al. Myelin basic protein as an encephaltogen in encephalo- inyelitis and polyneuritis following rabies vaccination. New Englandjournal of medicine, 1987. 31 6: 369-374.

25. Heniachudha T et al. Neurologic complications of Semple-type rabies vaccine: clinical and immi,nologic sti~dies. Neiirology 1987, 37: 550-556.

26. Heniachudha T et al. Inimunologic studies of paiients with chronic encephalitis induced by post-exposure Sernple rabies vaccine. Neurology, 1988, 38: 42-44.

27. Fuenzalida E, Palacios R. Un rnetodo mejorado en la preparacibn de la vacuna antirabica. [An improved method for preparation of rabies vaccine.] Eoletin del lnshtuto de Eacteriologia de Chile, 1955. 8: 3-10

28. Svet-Moldavaskil GJ et al. An allergen-free antirabies vaccine. Bulletin of the World Health Organization, 1965, 32. 47-58.

29. Gispen R, Schmittmann GJP Saathof B. Rabies vaccine derived from suckling rabbit brain. Archiv fur gesamte Virusforschung. 1965, 15: 366-376.

30. Gispen R. Saathof B. Neutralizing and fluorescerlt antibody response in man after rabies treatment with suckling-rabbit bran. Archiv fur gesamte Virus- fo i~chung. 1965, 15: 377-385.

31. Trejos A et al. Laboratory investigations of neuroparalytic accidents associated with suckling-mouse brain rabies vaccine. Part l. Encephalitogenicity and virological studies. Annales de I 'lnstitiit Pasteur: Immunologic, 1971, 125: 9 17-924.

32. Varela Diaz VM et al. Laboratory investigations of neuroparalyt~c accidents associated with suckling-mouse brain rabies vaccine. Part 1 1 . Encephalito- genicity of rnurine brain rnyelin preparations. Annales de l'lristilut Pasteur. Irr~rr~unologie, 1971 125: 925-938.

33. Held RJ. Adaros LH. Neurological disease in man follow~ng administration of suckling-mouse brain antirabies vaccine. Bulietin of the World Heaith Organi- zation, 1972, 46. 321-327.

34. Peck FB. Jr, Poweli H M Culberston CG. A new antrabies vacclne for human use. Clinical and laboratory resillts using rabies vaccine made from embryonated duck eggs. Jouriial of laboratory and clinical methods, 1955, 45. 679-683.

35. Hoskns JM. Duck-embryo vaccine 111: Kaplan MM, Koprowski H, eds. Labora-

tory tec/?niques in rabies, 3rd ed. Geneva, World Health Organization, 1973 (WHO Monograph Series, No. 23): 243-255.


36 Prussrn G, Katabr G Dorsolumbar rnyelrt~s followrng antrabres vaccinatron w ~ t h duck-embryo vacclne Annals of iiiteii7al inedicine, 1964 60 114-1 16

37. Perine PL, Harrs D, Kirkpatrick CH. lm~nunologic reaction to duck-embryo rabies vaccine. Journal of the American Medical Association. 1968.205: 87-90.

38. Harrngton RG, OIin R, Incomplete transverse myelitis following rabies duck-

embryo vaccination. Journal of the American Medical Associaljon, 1971 216: 2137-2138

39. Rubn RH et al. Adverse reactions to duck-enibryo rabies vaccine. Arlrlals of

internai medicine, 1973, 78. 643-649.

40. Kaiser HB et al Unusual reaction to rabies vaccine. Journal of /he American Medical Association, 1965, 193: 11 9-1 20.

41. Recommendations of tile Public Health Service. Advisory Committee on Immunization Practices Morbidity and mortality weekly repori. 1977, 25 (No. 51 1.

42 Farrar WE Jr Warner AR Jr Vivnna S Pre-exposure immunization against rabies using duck-enibryo vacclne Military rnedicine 1964 129 960 965

43 Turner GS. Rabies vaccines, British medical hulletin. 1969, 25: 136-142

44 Turner GS. An assessment of the current position of rabiesvaccination in man. In: Waterson AP, ed. Recent advancesin c/~nicalvirology Edinburgh, Churchill Livingstone, 1977: 79-92.

45. Haenzel I, Schlnler M, nderbtzin TM. Effectiveness and acceptability of the rabies vaccine Lyssavac. Schweizerische mediz~nische Wochenschrift 1976, 106: 1637-1641

46. Scliell KR et al. A highly purified and concentrated duck-embryo vaccine, a preliminary report. Journal of biological staridardization, 1980, 8. 97-106.

47. Gluck R et a1 New aspects concerning the inini~~nogenicity of rabies vaccines produced in ariirnal brains (duck embryo). In. Kuwert E et al.. eds. Rabies in the tropics. Berlin. Spririger-Verlag, 1985' 181 -1 88.

48. Wiktor TJ, Plotkin SA. Human cell-culture rabies vaccine. Antibody response in man Journal of the American Medical Assooalion, 1981, 224. 1170-1 171.

49. Koprowski H, Cox HR. Studies on chick embryo-adapted rabies virus. Part 1. Culture and cl?a~.acteristics and pathogenicity. Journai of ~mrnonology, 1948. 60. 533-554

50 Schwab MP et al. Avianized rabies virus vaccination in man. Bulletin of the World Health Organization. 1954, 10: 823-835


51. Clark HF, Wiktor TJ, Koprowski H. Human vaccination against rabies. In: Baer G, ed. The natural history of rabies, Vol. 2. New York, Academic Press, 1975: 343-365.

52. Veerarag havan N. Annual report, Director, Pasteur institute oiSouthern India, Coonor. Madras, Diocesan Press. 1970 35-42.

53. Tliraenliart 0, Marcus I. Introductory remarks: new developrrients of rabies vaccines and potency assay. In: Thraenhart 0 et al., eds. Progress in rabies controi. (Proceedings of the Second international IMVi Essen"WH0 Sym- posiuin on "New Deveiopments in Rabies Control", Essen, 5-7 July 1988 and Report of the WHO Consultation on Rabies, Essen. 8 July 1988.) Royal Tunbridge Wells. Wells Medical, 1989 27-36.

54 Largh O P Diaz AM. Arrossi JC. Comparison of the immunological response of humans to suckling mouse brain and human diploid cell vaccnes. In: Kuwert E et a l . eds. Rabies in the tropics. Berlin, Springer-Verlag. 1985: 189-195.

55. Sikes RK et al. Effective protection of monkeys against death from street virus by post-exposure administration of tissue-culture rabies vaccine. Bulletin of the Worid Heaith Organization, 1971. 45: 1-1 1

56. Bahmanyar M et a1 Successful protection of humans exposed to rabies infection. Postexposure treatmeiit with the new human diploid cell rabies vaccine and anlirabies serum. Jounial of the American Medical Association, 1976, 236: 2751 -2754.

57. CDC Veterinary Public Heaith Notes, Virai Diseases. August 1978. Atlanta, GA, United States, Department of Health, Education and Welfare. Public Health Service, Centers for Disease Control, 1978.

58. Warrell MJ et al. Economical multiple-site intradermal immunisation with human diploid-cell-strain vaccine is effective for post-exposure rabies prophylaxis. Lancet, 1985. i . 1059-1062.

59. Report of a WHO Consultation on rabies vaccine prepared in human dipioid cell ciiitures. Teheran, 5-8 May 1975. Geneva, World Health Organization. 1975 (unpublished docunient VPH/75.3: available or1 request from the Division of Communicable Diseases, World Health Organization, 1211 Geneva 27,

Swtierland).

60. Kopro~vski H, LeBel I. The presence of complement-fixing antibodies aganst brain tissue in sera of persons who had received antirabies vaccine treatment. Arnericar~ lournal of hygiene, 1950, 51 : 292-299.

61 Nchoison KG Improving rabies post-exposure treatment in developing countries-some suggestions In Kuwert EK, Wiklor TJ Koprowski H eds Cell-culture rabies vaccines and the~r protective effect ~n man


( Proceed~i~gs of WHO Consiilta tions, Essen, 5-7 March 1980.) Geneva, International Green Cross. 1981: 288-293.

62. Tripalhi KK, Madhusudana S N Sahu A. Reduction in the dosage and schedule of BPL-nactivated neural tissue vaccine for rabies prophylaxis in man, Indian journal of medical research, 1990, 91 : 334-339.

63. Held ?R et al. imuniracibn humana con vacuna antirabica de cerebro de rat6n lactante. [lmrnur?ization of humans w t l i suckling-mouse brain rabiesvacc~ne.] Boletin de /a Of~ciria Sanitaria Panamericana, 1972. 72 : 565-575

64. Diaz AM et al. Vacuna antirabica de cerebro de raton lactante. Esquemas reducidos de imunizac16n post-exposicion. [Rabies vaccine from suckling- mouse brain Abbreviated schedules for post-exposure immunization.] Revista Argentina de rnicrobiologia. 1979, 1 1 . 42-44

65. Larghi OP et al. Laboratory suckling-mouse brain rabies vaccine. Part Ill. Preservation of vaccine potency after elimination of murne brain myelin by centrifugation. Annales de l 'lnst~tut Pasteur: Microbiologie, 1976, 1278: 567-572

66 Thomas JB et a1 Purification of fixed rabies virus Virology 1965, 25 271-275

G7 Skes KR Larghi OP Purified rabies vaccrie developnlent and comparison of potency and safety with two human rabies vaccines Journal of immunology, 1967 99 545-553

68 Turner GS, Kaplan C Some properties of fixed rabies virus. Journalofgeneral virology, 1967. 1. 537-551.

69. Kaplan C, Turner GS. Rernoval of encephalogenicty from extracts of normal rabblt cectral nervous system by treatment with fuorocarbons. Nature, 1968, 219. 445 446

70 Wiktor TJ Koprowsk H Sc~ccessf~il irnrnuniiatlon of primates with rabies vaccirie prepared in human diploid cell strain W1 38 Proceedings of the Society for Expet~mental Biology and Medicine 1965, l 1 8 1069-1 073

71. Wiktor TJ et al Immu~iogenicity of concentrated and purified fables vaccine of tissue culture origin Proceedings of the Society for Experimental Biology and Medicine, 1969, 13 1 : 799-805

72 WHO Expert Committee on Rabies Eighth report Geneva World Health O r g a n r a t o r 1991 (WHO Techr- cal Report Series No 824)

73 Tra~s fe r of technology for production ot rabies vaccine memorandum from a WHO meeing Bullelin of the World Health Organizat~on 1985 63 661-666

CHAPTER 20

P-Propiolactone-inactivated sheep brain vaccine H. Singh '

A pher~olized nerve-tissue vaccine was originally developed in India in 191 1 by Sir David Semple at the Central Research Institute. Kasauli. Procedures for the preparation of Semple-type vaccines were described by various authors in 1973 (1-3). These procedures were later modified and p-propioactone (BPL) was used as an inactivating agent instead of phenol. Since 1972, this vaccine has been manufactured at the Central Research Institute. Several field trials have confirmed the efficacy of the vaccine. On the bass of studies in humans and animals, the dosage schedule of this vaccine has recently been modified (see below).

Composition

The vaccine is a 5% suspension of sheep brain-tissue infected with fixed rabies virus inactivated by BPL and contains 0 01 % thiomersal' and 0 25% phenol as preservatives

Preparation of the seed virus

Virus strain

The vaccine is prepared with the PV-11 strain of Pasteur fixed rabies virus. The strain has been maintained by passaging exclusively through rabbits. It is obligatory to limit the passage level to 10 passages from the original strain. This is achieved by using a seed lot system. i.e. primary and secondary seed lots.

Preparation of seed lots

Healthy rabbits weighing approximately 2 kg are lightly anaesthetized with ether and are inoculated ntracerebrally with 0 2 ml of a 10-2 dilution of the seed virus When they are completely paralysed (after about 8 days) the rabbits are killed by intravenous inject~on of air and their bralns are harvested All these procedures are carried out under strict aseptic conditions as described in Chapter 1

Sterility tests are carried out on each brain for aerobic and anaerobic organisms The harvested brains are washed with sterile distilled water and stored in separate containers at - 20 C or lower for up to 1 year After the brains have thawed they are homogenized in 0 5 inoli l sodium-potassium phosphate buffer

' Head. Department of Med~cal M~crobioiogy Postgraduate inslltute of hledica Education and Research Chandiga~h, lncl~a Also known as th~omersaia!e and rnercuioth~olate

INACTIVATED SHEEP BRAIN VACCINE

pH 7.6 (see Annex l ) , to give a 20% suspension (wlv). The suspension is then distributed in aliquots of 1 or 2 ml and fieeze-dr-ied

An appropriate number of vials of the freeze-dried suspension are reconstituted and subjected to quality control tests s i ~ c h as sterility. virus titre. identity and moisture content (see page 237); wlien the tests are satisfactory, these aliquots constitute primary seed lots. Secondary seed lots are prepared in a similar way froni the prlinary seed lots.

Preparation of the vaccine

Inoculation of sheep with the seed virus

Healthy sheep weighing between 7 and 12 kg are kept in quarantine under veterinary supervision for 7 days, during which time they are prepared for inocua- tion. The head and neck of the sheep are shorn and the operation site is thoroughly disinfected with povdone-iodine or 7006 ethanol. The inoculum is prepared just before use by reconstituting a vial of the secondary seed lot with sterile distilled water to give a final concentration of 1 :600. The sheep are inoculated intracisternally with 0.8 m1 of the seed virus, under strict aseptic conditions.

Harvesting and storage of brains

After inoculation (he sheep are placed in isolation pens (not the same pens as those used for cluarantrne) After about 5-7 days when they are completely paralysed the sheep are killed by intravenous injection of air and decapitated The heads are transferred to the dissection laboratory The brains are removed under aseptic conditions washed with 0 05 mol/l phosphate buffered saline (PBS) pH 7 0 (see Annex 2) and placed in a pre-weighed sterile aluminium or stainless steel container Representative samples of the b r a n tissue are taken and tested for bacterial sterility in duplicate nutrient agar slants The brains are weighed suitably distributed in various batches and stored at - 20 C or lower for up to 6 months until needed for vaccine production Any b r a n showing appreciable bacterial contamination is discarded

Preparation of l0 % brain-tissue suspension and inactivation

Frozen brains of a particular batch are pared to give a combined weight of approximately 160 g. After the brains have thawed, they are placed in a sterile Waring-type blender containing a sufficient volunie of 0.5mol1I sodium- potassium phosphate buffer, pH 7.6, to give a 10% suspension (wlv). Each paired brain-tissue suspension is collected in a separate flask. An aliquot is removed from each flask for determ~nat~on of the virus titre. V~rus t i l rat~on is carried out by intracerebral inoculation of mice; the titre should be at least 106 LDS,, per 0.03 m1 of inoculum Thiomersal is added to a concentratiori o f 0.02% and BPL is added to a final concentration of 1 :do00 and the flasks are transferred to a 37 'C incubator for 2 hours, durirlg which time they are shaken every 10-15 minutes. Phenol to a concentratiori of 0.25% is added afterthis period, because it will react with BPL. thus reducing its activity. The flasks are subsequently transferred to a cold room


for a period not exceeding one month. Samples are taken from each flask for sterility testing. Suspensions that show an adequate virus titre and pass identity and sterility tests satisfactorily are taken for filtration, pooling and dilution.

Pooling and dilution

Suspensions from each flask are passed through a ster~le nylon cloth filter with a fine mesh and collected in stainless steel containers of suitable capacity. An equal volume of 0.05 niol/l PBS, pH 7.0 (see Annex 2), containing 0.125% phenol is theri added to make it a 5% suspension containing 0.25% phenol and 0.01% thiomersal.

Tests on the bulk vaccine

The bulk vaccine is distributed in suitable quantities in sterile 2-litre glass bottles. Representative samples are taken for the following tests.

1 . Virus inactivation test. 2. Innocuity test. 3. Potency test. 4. Sterility test. 5. Determination of the pH. 6. Test for the presence of thiomersal and phenol

A sample of bulk vaccine is retained for a period of 1 year After all the tests performed on the bulk vaccine have been completed satisfactorily the vaccine is distributed into n~ulti-dose bottles each containing 30 m1 of vaccine Representa- tive samples of the finished product alorig with a summary protocol are then sent to the quality control division for certification and release

Tests on the finished vaccine

Representative samples are taken for the iol low~ng tests

1. Potency test ( i f not done on the bulk vaccine) 2. Innocuity test 3. Virus inactivation test. 4, Identity test.

5. Sterility test. 6. Determination of the thiomersal content. 7. Determination of the phenol content. 8. Determination of the pH.

Expiry date

The date of mariufactiire IS that of the last satisfactory potency assay. The expiry

date is usually not more than 6 months after this date


Labelling

The vaccine is labelled according to the requirements of the national control authority. The label should show,

- the name of the vaccine:

- the name and address of the manufacturer; - the batch number assigned by the manufacturer: - a list of the active ingredients and the amount of each present, with a

statement of the net contents. e.g. number of dosage units;

- the manufacturing licence no.: - the date of rriariufacture a n d the date of expiry;

- recommended storage conditions or handling precautions that may be necessary;

- instructions for use, and warnings or precautions that may be necessary; - the nature and amount of ureservatives.

Quality control tests

Identity tests

The identity of the seed virus is established by performing a virus neutralization test The constant serLim and varying virus dilution method is employed For establishing the identity of virus in sheep brain-iiss~ie suspensions a mixture of equal quantities of a I O - ~ dilution of bran-tissue suspension and a l F 3 dilution of hypermmune antirabies serum rased against a different rabies virus strain is incubated at 37 C for 1 hour Eight niice are inoculated intracerebrally with the mixture and observed for 14 days Another group of n i ce is treated similarly with a mixture prepared with normal horse serum All mice in the latter group should die whereas all those inoculated with the antirabies serum should survive

Virus titration

Three serial 10-fold dilutions ( 1 0 5 1 0 6 and i O - 7 ) of the material to be tested are prepared with 2% chilled inactivated normal horse serum in distilled water Each dilution is administered intracerebrally to a group of eight mice each mouse receiving 0 03 ml The ihree groups of mice are kept under observation for 14 days and a record is made of the number of paralysed animals that die from the 5th day onwards The virus titre is calculated by the Spearman-Karber method (see Appendix 3) or the method of Reed & Muench (see Chapter 38, Annex)

The virus titre should be no less than 105 LD50 per 0 03 ml for the seed virus and no less than 106 LD,, per 0 03 m1 for the sheep brain-tissue suspension

Virus inactivation test

The vaccine is inoculated intracerebrally into a group of 20 mice weighing 11-14 g, each mouse receiving 0.03 ml. The mice are observed for 14 days: any deaths occurring during the first 4 days are disregarded, but there must be no deaths after this date, otherwise the batch of vaccine must be relected.


Potency test

The potency of each batch of vaccrie is determined by the NIH test (see Chapter 37). According to the seventh report of the WHO Expert Committee on Rabies (4), vaccines that are administered in at least 14 daily injections and booster inocillations require a minimum potency of 0.3 IU per rnl, irrespective of the volume given in each injection. Generally the vaccine described above will have a potency of at least 1.0 IU.

innocuity test

Eight mice weighing 17--20 g are irioculated intrapertoneally with 0.5 ml of the vaccirle and observed for 7 days. All the mice should survive, otherwise the batch of vaccine must be rejected.

Sterility test

Sterility tests are performed at each stage of manufacture as well as on the bulk vaccine and the finished vaccine using standard procedures All samples shoiild be free from bacteria (both aerobic and anaerobicj and fungi.

Determination of the pH

This is done using a pH meter with glass and calomel reference electrodes, which are standard~zed with standard buffer solution at pH 5.0 and 9.0. The electrodes are rinsed with distilled water dried and the pH of the solution recorded.

The pH of the vaccine should be between 7.0 and 7.2.

Identity test

An identity test is performed on at least one labelled container from each tilling lot. At the Central Research Institute. the potency test described above is also used as an identity test.

Biochemical tests

Test for the presence of thiomersal

A sarnple (2 5 rril) of the bran-tissile suspension under test is mxed w ~ t h an

equal volume of distilled water In a test tube and 0 1 m1 of concentrated (3aoh) hydrochloric acid (HCI) is added A control is prepared with 10°/o normal sheep brain suspension without thiomersal The tubes are shaken and heated to boiling point The test 1s positive ! f a white precipitate appears wittiir? 30 seconds No such precipitate is seen ~n the controls

Test for the presence of phenol

A sample (0.05 ml) of the 109h bran-tissue suspension is mixed thoroughly with 7.5 rnl of distilled water In a test tube. Then 0.5 rnl of a 1 : 8 dilution of Folin-


Ciocalteau reagent is added and mixed well, followed by 2 rnl of 15% sodium carbonate (Na,CO,) solution. A control is prepared with 10% normal sheep brain suspension without phenol. The tubes are placed in a boiling water bath for 60 seconds, The test is positive if a blue colour of much higher intensity develops in the test than in the control.

Titration of ,&propiolactone

BPL is a condensation prodiict of ketone and formaldehyde It is an unstable llquid tha t readly polymerizes uporl exposure to elevated temperatures and hydrolyses in the presence of moistule A preparation that contains a significant amount of polymerized BPL will precipitate during use and may give rise to irregularities during virus inactivation The best results are obtained with the freshly distilled compound and it should be stored in tightly closed containers (preferably in sealed ampoules) at approximately - 10 "C to - 20 C and kept away from light It should never be stored at temperatures above 0 "C or below its freezing point It also read~ly reacts with hydroxyl phenolic amino sulfadryl and carboxyl groups and loses its virucidal activity Accordingly phenol should not be added to the 10% brain-tissue suspension until the inactivation with BPL is completed

The BPL used must be ttrated on receipt and at yearly intervals thereafter to ensure that it contains at least 80% of active BPL As already mentioned BPL is very rapidly hydrolysed in a saline medium and is entirely absent from the finished vaccine The titration is therefore carried out on the container of BPL itself using the technique of Tyler & Beesing (4) as described below

Reagents Sodium thiosulfate Potassium phosphate, d b a s c (K,HPO,) solution Sodium acetate solution Starch solution Iodine solution

Techn~que With a pipette measure out 25 ml of 0.4 molll sodium thiosulfate and add 2 m1 of the potassium phosphate, dibasic solution. Add not more than 0.5 g of the sample to be tested, containing about 5 mmol (0.3-0.4 g) of BPL. Shake gently to dissolve and wail 10-12 minutes. When the reaction is complete, wash the walls of the vessel with about 25 m1 of water and titrate by means of the 0.1 mol l l iodine solution, using the starch solution as indicator.

Control titrations Blank titration must be carried out at the same time and urder the same conditions to correct the small errors that may occur in the t t ra ton of thiosulfate in a slightly alkaline solution The BPL content (Oh) is given by the formula (3)


where.

A = volume (in m!) of iodine solution required for the sample B = volume (in ml) of iodine solution required for the blank N = concentration of the iodine solution (molll) W = weight (in g) of the sample 0.995 = correction factor.

Preparation of standard vaccine

There are various methods for testing the potency of rabies vaccines. The NIH test is the method of choice (see Chapter 37). Since this involves incorporation of a standard rabies vaccine, a detailed method of standard preparation is given below

Healthy sheep weighing 7-12 kg are inoculated ntracisternally with 0 8 ml of the PV strain of rabies fixed virus ( 1 600 dilution) and kept under observation When they are completely paralysed (after about 8 days), the sheep are killed and their brains are removed under aseptic conditions. The brains are homogenized in a suitable stabilizer (see Annex 3) to give a 20% suspension (wlv). An aliquot is removed for determination of the virus titre and identity tests Sufficient thiomersal and BPL are then added to give a final concentration of 001% and 1 2000 respectively and the suspension placed in an incubator at 37 C for 2 hours Sterility and virus inactivation tests are performed on the emulsion

When all the tests (virus titre, identity, sterility and inactivation) are satisfactory. the vaccine is distributed in suitable quantities into glass vials or ampoules and ireeze-dr~ed.

The freeze-dried product is tested for sterility and moisture content Its potency is then determined by the NIH test using the national reference preparation of iabies vaccine The potency of the final product should be no less than 0 8 IU per ml

The Iyophilized vaccine should be stored at 4'C or lower It maintains its potency for about 2 years when stored at 4 C and for about 5 years i f stored at - 25 C or lower

Dosage schedule

The vaccination schedule recommended by tile WHO Expert Committee on Rabes

n 1984 (5) n cases of Class ll and Class Ill exposure consisted of 14 dally doses, plus booster doses 15 and 65 days after the end of the primary series The schedule was modified in April 1988 following several animal experiments and field studies

(6) which showed that it was possible to reduce the number of vaccine doses given to patients The schedule currently used at the Central Research Institute consists of 10 daily niections of 5 m1 of vaccine given subcutaneously over the anterlor abdomen followed by one oooster dose of the same quantity after 3 weeks for Class l1 exposures and two booster doses 1 and 3 weeks after the last primary dose in cases of Class Ill exposure The vaccine dose is reduced to 2 ml for children weighing less than 30 kg


References

1 Seligrnar~n B Jr Semple-type vaccine In Kaplan MM Koprowski H eds Laboratory iechriiqiies 1r1 rabies 3rd ed Geneva World Health Organization 1973 (WHO Monograph Series No 23) 192-198

2 Selmov MA Morogova VM Phenolzed freeze-dried sheep brain vaccine A Method used in the USSR In Kaplan MM Koprowski H eds Laboratory techniques irl rabies 3rd ed Geneva World Health Oiganizaton 1973 (WHO Monograph Series No 23) 201-204

3 Lepine P et al. Phenolized, freeze-dried sheep brain vaccine. B. Method used at the Pasteur Institirte, Paris In: Kaplan MM, Koprowski H, eds. Laboratory techniques in rabies. 3rd ed. Geneva, World Health Organization, 1973 (WHO Monograph Series, No. 23). 204-212.

4 Tyler WP, Beesing DW. Chemical and cryoscopic analysis of /l-propiolactone Ar~alytical che/nistr.y, 1952, 24: 151 1-1 51 3.

5 WHO Expert Committee on Rabies Seventh report Geneva, World Health Organization 1984 (WHO Technical Report Series, No 709) Annex 1

6 Trpathi KK Madhusudana SN Sahu A Reduction in the dosage and schedule of BPL-inactivated neural tissue vaccine for rabies prophylaxis in man Indian journal of medical research 1990 91 334-339

Annex 1 Preparation of 0.5 molll sodium-potassium phosphate buffer, pH 7.6

Reagents Sodium phosphate dibasc (Na,HPO,) Potassium phosphate, monobasic (KH,PO,) Distilled water to make

617.7 g 88.4 g

l 0 0 litres

Technique

1. Dissolve both the salts separately in flasks using about 2.0 litres of distilled water for each: warm the contents of the flasks over a i a m e until the salts go into solution

2 Pour the solutions into a 100-l~tre stainless steel container and add the remainder o i the d~stilled water Add 7 0 litres of distilled water more to compensate for loss due to evaporation during autoclaving

3 Check the pH 4. Pass the solution through G-2 porosity filters (pore size 40.9 pm) to remove any

particulate matter and distribute in suitable quantities into glass containers. 5 Sterilize the solution by aiitoclaving at 121 'C for 1 hour.


Annex 2 Preparation of 0.05 mslll phosphate-buffered saline (PBS), pH 7.0

Reagents Na,HPO, KH,PO, Sodium chloride (NaCI) Distilled water to make

61 7.7 g 88.4 g

1773 0 g 100 litres

Technique

1 Dissolve all the above salts separately in flasks using a sufficient quantity of distilled water for each warm the contents of the flasks over a llame until the salts go into solution

2 Pour tlie so~ i t i ons into a 100-litre stainless steel container and add the remainder of the distilled water Add 7 0 litres of dist~lled water more to compensate for loss due to evaporation during autoclaving

3 Check the pH 4 Pass the solution through G-2 porosity filters (pore size 40 9 pm) to remove any

particulate matter and distribute in suitable quantities into glass containers 5 Sterilize the solution by autoclaving at 121 C for 1 hour

Annex 3 Preparation of stabilizer for rabies vaccine, pH 7.2

Reagents Casein hydrolysate Sucrose

Technique

1. Prepare a 20% solution of casein hydrolysate (acid digest) in distilled water and adjust the pH to 7.2 using 10 m o / I sodium hydroxide (NaOH) solution. Sterilize by autoclaving at 103 kPa for 20 minutes.

2 Prepare a 10% solution of sucrose in distilled water and adjust the pH to 7.2 using 10 moll l sodium hydroxide solution. Sterilize by filtering through a sintered glass G-5 porosity filter (pore size 1.2 pm).

3 Mix the two solutions in equal proportion under aseptic conditions.

CHAPTER 21

Suckling-mouse brain vaccine1 A. M Diazr

The suckling-mouse b r a n (SMB) vaccine was developed a l the Bacteriological Institute o i Ch~le in 1954 ( 1 ) T h ~ s vaccine was originally used in dogs: use In humans was begun experimentally In 1960 (2). Since that time, SMB vaccine has been used In most Car~bbean and Latin American countries for rmmunization of humans. Currently, about 5 i~ii l l ion doses are prepared for human use annually as well as more than 20 million doses for veterinary use.

By 1972, most of these countries had adopted a pre-exposure vacclnation schedule consisting of 3 doses of SMB vaccine and a reduced post-exposure treatment sctiedule of 7-10 doses.

The Fuenzalda-Palacios technique for the producton of SMB vaccne has been modified in an attempt to reduce further the level of potential encephalto- genic substances In the f~na l product. Vaccines are now prepared using mice no older than 1 day at the time of ~noculation and the brain suspension is centrfuged at 17000 g for l0 minutes. The safety of post-exposure treatment using SMB vaccine has been improved further by a reduction in the number of doses required.

Since 1980, the marked increase in the production of SMB vaccine for use in dogs has permitted the implementation of control programmes within the frame- work of the PAHOIWHO programme for the elimination of urban rabies in Latin America by the end of this decade.

Formula

Each 2-m1 dose for lhuman use contains inactvated virus in 20 mg of sucklng- mouse bran, phenol in a concentralion of l 1000, and thomersa13 in a concentration o l 1 10000

Each 2-ml dose for use in dogs contalns inactivated virus in 50 mg of suckling- mouse brain wlth the same preservatves

Preparation of the inoculum

Virus strains

Three strains of fixed rabies virus are used for the preparation of this vaccine. 2 strains isolated in Chile (strains 51 and 91, of dog and human origin respectively)

' Eased on the chapter by E Fbenzal~da r - the previous edi ton ' ~ t l ~ e f , ouaity Con i rc Unit. PAHO WHO Pan Aner ican Institute for Food Protecton and Zoonoses

(INPPAZ) Martinez Argent~na 'Also known as thmerosal and rnercuroth~oldlr


and the CVS strain.' Strains 51 and 91 have been maintained since their isolation by intracerebra passages in adult mice (1).

Seed preparation

"Seed stocks'' are 20% suspensions of bra~ns obtained from 3-4-week-old mice, previously inoculated with the respective strains. These suspensions are prepared in sterile distilled water, pH 7.2 7.4, supplernerited with sucrose and gelatin to final conceritrations of 7.5% and 0.5*/6 respectively. The suspensions are Iyophilzed in 0.5-rnl volumes and kept at - 20°C for up to 5 years.

Preparation of the working stock suspension

Adult mouse brains infected with ihe seed of each strain are used to prepare a 20% suspension (wlv) in 2% horse serum or with 7.5% sucrose in double-distilled water, both coniaining 200 IU of penicillin and 200 p g of streptomycin per ml. The suspension is stored in vials at - 30°C to - 70'C and represents the "working stock suspension ' , from which dilutions are prepared for mouse inoculation, as described below.

Test for viral contaminants

To determine whether the work~ng stock suspension is coniamnated with a neurotropic virus other than rabies virus 0 5 m1 of the supernatant from a 1 10 dilution of this suspension is incubated for 90 minutes at 37 C with 0 5 ml of undiluted antirabies tiypermmune seruni The serum is prepared in horses or rabbits m m u n i e d with virus grown in animals other than mice The mixture is inoculated intracerebrally in doses of 0 01 ml into 10 suckling mice (aged about 1 week) and in doses of 003 m into ien 3-4-week-old mice All the inoculated animals must remain healthy for 30 days

lnoculum

To prepare the inoculum the supernatant from the working stock suspension which must be sterile is diluted until it conta~ns about 100 1000 LD,,

Inoculation and harvest

If the vaccine is for human use suckling mice no older than I day are inoculated intracerebrally with 0 01 m1 of the inoculum If the vaccine is for veterinary use mice aged 2-4 days are used The ~noculum is also injected intracerebrally In doses of 0 03 m1 into 10 mice weighins 11-14 g each to determine the viab~lity o i the virus Within 14 days after inoculatioii all mice must develop signs of rabies If some mice

' At i ts ~lgh:ti rnce!?q in 1991 t ~ e WHCl Expert Co,ri~rit!ee on Rabies stressed that rabies vir~;s slrairis iised for the prodict ion of vacciries mds: be c a r e f u i : ~ selected arid t!iat periodic checks must oe carried odt on their an!igenic idei i?ty (:o e m u r e :ha! straips other !bar 'hose used for vaccirie productloq are not n!roaucedj

SUCKLING-MOUSE BRAIN VACCINE

remain healthy during this period. the working stock suspension must be renewed, but the harvest can be ~ ~ s e d ; if none becoines ill, the mice inoculated for harvest should not be used for virus product~on.

Approximately 96 hours after inoculation, i.e., a day before the usual end of the incubation period, all mice are killed using ether or chloroform, placed in a gauze package and submerged in a 70% ethanol solution. Harvest at this time avoids loss from early deaths and ensures that the virus titre is at maximum.

The dead mice are secured to a dissection board by means of a rubber band. The skirl of the head and neck of each animal is disinfected with povidone iod ine (100i0) or any other appropriate disinfectant. The brains are removed under a negative draught hood by suction. using a 2-nil syringe with a 25-40-mm X 1.8-mm (15-gauge) needle The needle is inserted into the cranial cavity in the frontal area The brain material is aspirated gently, without lateral ~novements and dispensed into a flask of mor-e than 100-ml capacity The flask should have a screw cap. During harvest the bottom of the flask is submerged in crushed ice. If the collected material is not to be used immediately it is stored at 7 0 ' C . The average weight of a single brain obtained in this way is 0.20-0.25 g.

For large-scale production it is convenient to use a vacuum pump connected to a thick-walled flask by means of a rubber hose (internal diameter 5 mm) with a cotton filter. The needle (see above) is connected to the flask by another tube that extends about 3 cm into the flask. All mice are placed in a gauze package which is submerged for 10 minutes rn 70% ethanol. These mice are put on a tray under a negative draught hood and are held with the forefinger and thumb while the needle is introduced The vacuum is regulated by means of a Mohr clamp.


Immediately before prepaiation of the vaccine the brains are partially thawed at room temperature If the vaccine is to be inactivated by exposure to ultraviolet light a 10% suspension (wlv) of tissue in cold double-distilled water is prepared To obtain a homogeneous suspension a blendei or some similar apparatus 1s used and 3 runs of 1 minute each are made allowing intervals of 10-20 seconds between the runs If the vaccine is to be inactivated using p-propolactone the 10% suspension (w v) of tissue 1s prepared in stabilizer solution A (see Annex 2) When the vaccine is prepared for use in humans the 10% suspension is centrifuged for 10 niinutes at 17000 g and the supernatant liquid is used

Control tests

Virus titration

The virus titre of the supernatant liquid is determined by ntracerebral inoculation of serial tenfold dilutions into niice (4-5 weeks old, 11-14 g). It is advisable to perform this titraton as soon as possible, Titres ranging from lo7.0 to 1 0 ' . ~ are considered acceptable.

Test for viral contaminants

This test IS done in the same way as for the working stock suspension (see above)

245


Test for bacterial contarninanis

Before inactivation a sterility test IS performed to detect bacterial contamination Each of 5 tubes containing 9 m of thioglycolate broth is inoculated with 1 ml of the whole suspension One tube is agitated and 1 ml of the suspension is transferred to each of a second series of 5 tubes T h ~ s procediire is continued to the 10-b dilution All the tubes are incubated at 37'C and observed daily for 14 days Bacterial growth must not occur Although the sterility test on the bulk vacclne should detect any bacterial contaninants that would make the vaccine unaccept- able it IS advisable to perforni the bacterial count test as an in-process control because the detection of a t i gh level of ~ontaminat ion is indicative of deficiencies in good mani~facturing practices during the iri ital stages of production that should be corrected at that point

By ultraviolet light A twofold dilution of the 10% suspension is prepared by adding cold, sterile, double-distilled water to obtain a 5% concentration. The virus is inactivated by ultraviolet light using a con t in~ io~ is flow plasma sterilizer as described in Annex 1 The material is exposed to a set of 4 virucidal-type 30-W ultraviolet lamps. 91.5 cm in length, and is processed at a rate of about 250 ml per minute.

By /i-propiolactone A fivefold dilution of the 10% suspension is prepared by adding stabilizer solution A (see Annex 2) to obtain a 2% concentration. The 2% suspension is diluted with an equal volume of stabil~zer solution B (see Annex 2) to glve a final concentration of 1 %. The virus is inactivated by the addition of a small volume of cold, sterile, double-distilled water containing sufficient p-propiolactone to give a final concentration of 1 :4000 lnaciivation is allowed to proceed for 3 hours to ensure total hydrolysis of P-propioactone.

Safety test

Ten n ice, each weighing 18-20 g, are inoculated intracerebrally with 0.03 ml of the inactivated bulk vaccine, Ten suckling mice are similarly inoculated with 0.01 m1 of the inactivated bulk vaccine. All test animals must remain free from rabies symptoms and frorn other diseases of the central nervous system for at least 2 weeks.

Final dilufion of the vaccine

Immediately after irradiation, the 5% suspension is diluted 5-fold by adding 4 volumes of double-distilled water coritaining sufficient glucose, phenol and thiomersal to achieve final concentrations of 5[% 1.1000 and 1 : 10000 respectively. I f the vaccine is to be used in dogs, the 5% susperision must be diluted in 1 volume of the same diluent. If the vaccine for canine use is to be lyophilized, the glucose diluent is replaced by a diluent containing 7.5% sucrose and 0.5% gelatin.


Phenol and thiomersal are used to protect the antigen aga~nst the action of enLymes ~n the nervous tissue as well as to preserve sterility.

Sterility test on bulk vaccine

All flasks containing vaccine must be tested for sterility A l-m1 sample from each flask is inoculated into ai least 5 tubes of thioglycolate broth trypticase soy

medium arid Sabouraud s agar The tubes are incubated for 14 days, the first group at 30-36 C and the remaining two at 22-25 'C

The product should not be dispensed into vials until all the above tests have been satisfactorily completed

Potency test

lmmunogenrc potency IS determined by the NIH test (see Chapter 37) A rabies reference vaccine of SMB origin' prepared by the PAHO WHO Pan

American Institute for Food Protection and Zoonoses (INPPAZ) was recently evaluated in a collaborative study r r i the Caribbean and Latin Arner~ca The study demonstrated that the vaccine could be used as a regional standard to determine the potency of SMB vaccines (3)

Sterility tests on final container

Once the product has been dispensed, sterility tests are conducted on a representative number of flasks or vials of each batch of vaccine, as recommended by the WHO Expert Committee on Biological Standardization (4).

As a safety measure it is essential that the producing laboratory should preserve the following samples of each batch of vaccine until 6 months after the expiry date for possible future reference:

30 ml of 10% virulent suspension frozen at - 7OCC: 30 m1 of 546 inactivated suspension frozen at - 70°C; 30 m1 of vaccine refrigerated at 4°C or 7°C.

Expiry date

The exprry date of the i iq~ i id vaccine IS 1 year from the date of issue provided that it is stored at 4 C as rccommerlded In the case of lyoptiilized vaccine for canine use with residual moisture levels of 2-3% the expiry date is 18-24 months after the time of manufacture

References

1. Fuenzalida E, Palacios R. Un metodo mejorado en la preparacion de la vacuna antirabica. [An improved method for the preparation of rabies vaccine] Boletin del lnstituto de Bacteriologia de Chile. 1955, 8: 3-10.

' Avalable on rcques! i r o n the PAHWWHO pan Amerlean inst!t;te ior Fooo Pro!ec!on a n d Zoonoses (INPPAZ), Calle Ta lca l i~ iano 1660 1640 Martiner Argent~na


2. Fuenzalida E, Palacios R , Borgofio JM. Antirabies antibody response in man to vaccine made froni infected suckling-mouse brains. Buileiin of the World Health Organ~zation, 1964, 30. 431 -436.

3. Diaz AM0 et al. Rabies reference vaccine for use as a regional standard for Latin America and the Caribbean countries. Bioiogicais. 1990. 18. 281-287.

4. WHO Expert Comm~ttee on B~ological Standardization. Twenty-fifth report. Geneva. World Health Organ~zation. 1973 (WHO Technical Report Series, No. 530). Annex 4.

Annex 1 Ultraviolet-light irradiation for inactivation of vaccinesl

Principles and methods

The use of ultraviolet (UV) light to inactivate microorganisms without impairing their antgenicity is a well-established procedure

UV energy has little ability to penetrate biological substances since it is absorbed rapidly Therefore for effective exposure the material being irradiated must be presented to the incident UV rays in a very thin film In addition as the amount of UV energy absorbed by a biological substance increases the degree of chemical change in that substance increases the length of exposure must therefoie be as short as possible while accomplishing the purpose of nact ivat~on otherwise bredkdown of the aiitigen will occur The biological effect of UV irradiation on viruses is for practical purposes instantaneous and once this imnieddte etfect has taken place no further action ensues There IS no ev~dence that any secondary by-products that would subsequently be deleterious to the antigen are formed in ttie irraddtion process

Besides a controlled length of exposure and irradiation of the material in a thin film the apparatus used should allow for a continuous flow of the antigen suspension through the equipment Although quartz is a good conductor of UV energy and has been used extensively for making various types of container in which an antigen could be exposed in general an apparatus designed to eliminate the passage of ttie UV rays through quartz gives more efficient radiation The apparatus should also be easy to operate and use a readily available inexpensive source of UV light

Apparatus

Several types of apparatus have been designed that in some way incorporate the principles listed above.

Standardization of apparatus

Froni the practical standpoint the only standardization necessary is the determna- tion of the rnlnlmum exposure (or maximum (ate of flow) necessary to inactivate the

' Based 0,- !tie cqapter ny 'ho late K. Haoe In !he p;evous e r j ~ l o n

248


working stock virus suspension. In general. a single run w ~ t h the following rates of flow should give a range of exposure from one coinpletey inactivating the virus to one in which virus is still viable

50 rnl per minute 100 ml per minute 200 ml per minute 400 m1 per minute

Once the minimum exposure consistently causing virus inactivation has been

established, the rate of flow should be set so that, for routine production. twice this amount of exposure w ~ l l take place, T h ~ s gives a sat~sfactory safety factor beyond the point of inactivation of the virus, and yet well within the limits of exposure possible before destruction of the antigen with loss of rnrnunogenc potency. It has been found that the minimum inactivating exposure must be increased fivefold before a marked drop in antigenic potency occurs.

Under ideal operating conditions there should be continuous monitoring of the amount of UV energy to which tlie virus suspension is being exposed.

Virus suspension

As with other methods of inactivation the virus titre of the original virus suspension should be as high as possible but variation of as much as orie in the log trtre w~ l l not noticeably change the necessary exposure time Since UV light will not penetrate biological substances to any appreciable depth, it is important that the emulsion be a un~formly fine one To ensure this the virus suspension should be f~ltered through several layers of sterile surgical gauze or a wire screen before being irradiated

In general a 5% suspension of infected tissue is used with the irradiation equipment although suspensiorls up to 10°/~ in concentration may also be inactivated

Preservation and storage of irradiated vaccine I f the irradiated vaccine 1s to be kept in the liquid state before use a chemical preservative should be added Thometsal at a concentration of 1 8000 or 0 25% phenol appears to be superior to formaldehyde Irradiated vaccine because it contains no substances deleterious to viral antigen may be kept frozen or may be dried from the frozen state If the vaccine is to be freeze-dried substances such as sucrose or glycine should be added to prevent a loss of potency After freeze- drying the vaccine will maintain its potency indefinitely provided it is kept suff~cently dry

Testing of ~rradia ted vaccine Sterility safety and potency tests should be carried out or1 the irradiated vaccine in the same manner as with other types of inactivated rabies vacclne Since it is difficult to monitor the effectiveness of the radiation to which the vaccine is being exposed it is especially important that both potency and safety tests be carried out routinely on every batch of the final product This is the only way to check whether irradiation has proceeded accordirg to plan


In general, it has been found that highly potent vaccines can be made consistently by the lrradlation method There is no evidence to date that the ncidence of post-vaccinal coniplicat~ons is any greater after the use of irradiated vaccines than with other types.

Annex 2 Preparation of stabilizer solutions

Stabijizer solution A, pH 7.4 Sodium chloride (NaCI) Sodum phosphate, monobasic, dhydrate (NaH2P0,.2H,0) Sodi im hydroxde (NaOH) Distilled water to make

Stabilizer solution B, pH 7.4 Glucose Stablllzer solution A, prepared as above. to make

Embryonating egg vaccines

CHAPTER 22

Purified duck-embryo vaccine for humans R Glijck'

G~ven the excellent yields of rabies virus obtained from embryonated duck eggs and the continuing need for low-cost rabies vaccines, techniques have been developed for purifying rabies virus from avian proteins on an industrial scale.

Purified duck-embryo (PDE) rabies vaccine offers the same immunogenlclty and safety as other rabies vaccines of high potency, including human diploid cell (HDC) rabies vaccine, and has the advantage of low production costs ( l ) . Rabies virus is extracted from infected e~nbryonated eggs without using mechanical shear forces in order to reduce the release of soluble viral and avian antigens. Purfica- tion is accomplished by density gradient centrifugation and removal of non-viral p i d s by extraction in a water-miscible organic solvent. After chemical precipitation, the virus is ~nactivated w ~ t h 8-prop~olactone The final product contains no detectable rnyelin basic protein and only trace arnounts of avian antigens (2). The average potency of a vaccine dose is 5 IU perml. Of equal irnportance is the fact that the vast niajorty of rabies antigens incorporated in the vaccine consist of intact virus (3). The relatively high content of rabies nucleoproten (N protein) further enhances the protection offered by PDE vaccirle compared w ~ t h other tissue-culture vaccines (4) .


Preparation of the seed virus suspension

The Pitman-Moore (PM) strain of f~xed rabies virus adapted to human diploid cells (8HDCS) from the Wstar Institute Philadelphia PA, USA is used It was adapted to the embryo cells by intracerebra passage in mice and repeated passage by inoculation in duck eggs which have undergone initial ~ncubation The blrus strain used for the preparation of the vaccine is from a passage with a particularly high trtre whrch was used for the preparation of rabies vaccine in accordance with the method descr~bed in the previous edition (5)

Fertilized duck eggs from healthy stocks incubated for 5 days at 35 37'C and 65-70% relative humidity are used

1. On the 6th day of incubation. candle the eggs Discard any eggs that are cracked, infertile, contain displaced air sacs. or contain dead or weak embryos.

2. On the 7th day, inoculate 0.25 m1 of the rabies virus directly into the yolk sac (Fig. 22.1; see Chapter 23). Continue incubatiorl for 10-14 days.

3. At the end of the incubation period, caiidle the eggs again. Discard any eggs containing dead or weak embryos.

' Head Departrnei.: of Virology Swiss S e ~ u n and Vaccine 1rsti:u:e Berne, Sw~tzeriand

LABORATORY TECHNlQUES IN RABIES

Fig. 22.1 Vaccination of duck eggs

By courtesy of R Giuck

4. Open the eggs under aseptic conditions, and remove the live embryos (see Chapter 23).

5. Decapitate the embryos, and store the heads separately under sterile coriditions in the vapour phase over liquid nitrogen until the sterility tests are complete.

6. Place groups of 40-60 of the sterile heads in a sterile meat mincer containing sufficient stabilizer to glve a 33% suspension. Phosphate-buffered saline (PBS) (see Chapter 20, Annex 2 ) other appropriate saline solutions, or even desalnated water may also be added, provided the pH remains between 7.0 and 8.0. Remove samples for sterility testing.

7 Wash the suspension twice ~ i h a phosphate-containing buffer. Remove the tissue fragments by centr fuging the suspension twice at 10000-15000 g and

2-8'C for 17 minutes. and then f~ltering through several layers of gauze. 8. Resuspend the pellet in a phosphate-coniaining buffer and stir for at least

1 hour at 1-4'C. Repeat step 7. The resulting suspension should contain approximately 30% of infectious virus. The ~nfectious titre should be within the range 10'-10' LD50 per ml.

Nimination of non-viral iipids

The non-viral l ipds are removed by mixing the resulting virus suspension with an inert liquid hydrocarbon solvent of relatively low density, such as n-heptane.


Fig. 22.2 Large-scale ultracentrifugation of rabies virus in a sucrose gradient

By courtesy of R Gliick

1 Add an appropriate amount of the BPL solution to the bulk vaccine preparation to g ve a final concentration of 1 4000

2 Stir the mixture at 4 C for 5 minutes and transfer ~t to a fresh vessel Allow the mixture to react at 4 C for a further 40 hours The mixture should be stirred throughout this period and the pH and temperature should be measured cont~nuously

3 Measure the pH of the mixlute at the end of the inactivation period It should drop trom 8 0 to approxinately 7 4

4 Finally add thromersal' to the inactivated vaccine to give a final concentralion of 1 10000 Rernove sdrnpes for innocuity testing (see below)

' Also known as nieicurottiolate and th~ornersaiate

256

PURIFIED DUCK-EMBRYO VACCINE

Lyophilization

1. Dispense the inactivated virus into 3-1-17 vials in single doses of 1 ml. 2. Place rubber stoppers loosely on top, arid freeze-dry the vial contents under

vacuum. 3. When the lyophilization process is completed, push the stoppers in tight and

close the vials with flip-off caps to ensure that tthe vials are airtight.

4. Store the lyophilized vaccine at - 20°C.

Reconstitution of lyophilized vaccine

The lyoph~lized vaccine is reconstituted for use by injecting 1 ml of sterile distilled water through the rubber stopper into each vial The vial is then shaken gently without formirig a foam until the vaccine is completely dissolved The entire contents of the vial are then injected subcutaneously into the upper arm of the patient

Control tests

Tests on the bulk vaccine

Innocuity test Three rabbits (20-2.5 kg) and 30 mice (14-16 g) are inocolated intracerebrally with 0.2 m of the final bulk vaccine and observed for 14 days. All animals must remain free from signs of disease.

Tests on the final vaccine

The tests listed below are performed on the final lyophilized product reconstituted with sterile distilled water

Potency test The vaccine should have a minimum potency, as measured by the NIH test (see Chapter 37), of 2 5 IU per dose.

Sterility test The vaccine should be shown to be sterile and free from any adventitious agents (especially avian leukosis viruses and adenoviruses)

Giycoprotein content The vaccine should have a minimum glycoprotein content. as measured by the single radial immunodiffusion test ( 6 ) . of 5.0 IU per dose.

Analytical tests Appropriate tests should be performed to determine the nitrogen, cholesterol. sodium chloride. BPL and thomersal content ot the vaccine.

Safety test Three guinea-pigs (250-350 g) are inoculated intraperitoneally with 5 0 m1 of the reconstituted vaccine In addition, three inice (17-22 g) are inoculated


intrave~iously with 0.5 m of the reconstituted vaccine. All test animals must remain free from signs of rabies and other diseases.

Stability test The potency of the vaccine should be routinely tested every 6 months.

Expiry date

The lyophilized vaccine may be used up to 4 years from the date of release from storage. In spite of very high stability (the vaccine is stable for up to 12 months at 37 "C), it should be kept between 2 and 8 "C after release.

References

1. Glijck R et al. A riew highly mmunogenic duck-embryo rab~es vaccine. Lancet, 1984, i. 844-845.

2. Glijck R et al. Absence of myelrn basic protein in an improved purified duck- embryo rabies vaccine. Neurochernical pathology, 1986, 4: 69-75.

3. D~etzschold B et al. Chemical and immunological analysis of the rabies soluble glycoprotein Virology, 1983, 124: 330-337.

4. Ertl HCJ et al. lndcicton of rabies virus-specific T-helper cells by synthetic peptdes that carry dominant T-helper cell epitopes of the viral ribonucleoprotein. Joiirnal of virology, 1989, 63: 288552892,

5. Hoskins JM. Duck-embryo vaccine. In: Kaplan MM, Koprowski H, eds. Labora- tory techniques in rabies, 3rd ed. Geneva, World Health Organization, 1973 (WHO Monograph Series, No. 23): 243-255.

6. Ferguson M, Schild GC. A single radial immunodiffusion technique for the assay of rabies glycoprotein antigen. Journal of general virology, 1982, 59: 197- 201.

Annex Preparation of stabilizing medium

Solution 1 Potassium phosphate, dbasic (K,HPO,) Cysteine hydrochloride Lactose Distilled water to make

1. Dissolve the potassiuni phosphate, cysteine hydrochloride and lactose in the distilled water. Add each constituent separately in the order listed and dissolve completely before adding the next.

2. Sterilize the solution by positive pressure filtration, 3. Store in completely filled, tightly stoppered bottles at approximately 4°C.

Solution 2 Gelatin Distilled water to make

PURIFIED DUCK-EMBRYO VACCINE

1 Dlssolve the gelatin in the d is t~ led water which should be warmed to 43-45°C 2 Sterilize the warm solut~on by positive pressure filtration, using a filter and

pressure vessei pre-warmed to 36 37 "C

3 Store at approximately 4 C

Preparation of working stabilizing medium

l . Melt the gelat~n (solution 2) in a water bath maintained at 36-37 "C. Cool to approximately room temperature.

2 MIX equal volumes of solutions 1 and 2. 3 Measure the pH of the mixture and adjust it to 7.6 by adding approximately 30 mi

of 1.0 molli sodium hydroxide (NaOH) solution. 4. Store in completely filled, tightly stoppered bottles at approximately 4 "C.

CHAPTER 23

Chicken-embryo vaccine for dogs H. Kopro wski '

At present the Flury and the Kelev strains of rabies virus adapted to and modified in the developing chicken embryo are used for the production of chicken-embryo rabies vaccines for the immunization of dogs Vaccine made with the Fury strain ( l ) has been used on a worldwide basis and therefore the description of the production procedute will be mainly confined to the Flury-strain vaccine The same principles however are followed in the preparation of chicken-embryo rabies vaccine with the Kelev strain (2)

The Flury-strain vaccine for dogs represents the 40th-50th egg-passage level of the virus At this egg-passage level the virus is pathogenic for mice rats and hamsters when given by the intracerebral and intramuscular routes Guinea-pigs appear to be susceptible when inoculated inlracerebrally but not ~ntramuscularly On the other hand the virus does not produce any signs of infection in rabbits when given by the intracerebral or intramuscular routes or in dogs when given by the intramuscular route

Further modification of the Flury strain has been achieved by continued egg passage (180th passage) and this has reduced its pathogenicity for experimental animals while its antigenicity has apparently been retained At present, the high egg-passage (HEP) level vaccine is recommended for use in cattle and cats and the low egg-passage (LEP) level vaccine described here for use in dogs only

The Kelev chicken-embryo vaccine represents the 60th-70th egg-passage level of the virus At this egg-passage level the virus does not produce any signs of infection in hamsters guinea-pigs or rabbits inoculated either intramiiscularly or intracerebrally Dogs inoculated intramuscularly with concentrated suspensions of the virus fail to show signs of infection


Preparation of seed virus

Seed stocks are prepared from 6094 suspensions of chicken embryos previously ~noculated with the respective stra~ns. The suspensions are tested for potency by the guinea-pig potency test (see Chapter 39) and then either frozen or lyophilized

and kept a1 - 60°C.

' Director, Ce~!e r for Neurovrology Jeffersni. Cancer i:istiiuIe Thornas Jeflerson Unveislty. Phiadelph~a, P A USA

'These s l rans are ava~labie to natonal iaboralorieson request from the W o r d Health Organlrat~or~. 121 1 Geneva 27 Swtzeriancl

CHICKEN-EMBRYO VACCINE FOR DOGS

Preparation of the working stock suspension

Chicken embryos infected with the seed of each stra~n are used to prepare a 20% suspension (w.'v) in distilled water. The suspension is prepared immed~ately before use and forms the working stock suspension which is used to infect the embryos for vaccine production, It shoiild be tested for viability as follows.

Inoculate a group of six mice ntracerebrally with 0.03 rr~l of a 10-I dilution of the working stock suspension, using a 0.25-ml tuberculin syringe with a 0.40-mm X 6-mm (27-gauge) needle. All mice should develop signs of rabies within 6-8 days after inociilation, I f some mice remain unaffected by the 9th day, drscard t h e inoculated eggs .

In the case of the Kelev strain, tile viability test must be carried out in suckling mice.

inoculation of eggs

The number of eggs to be inoculated will depend on the amount of vaccine required and on the storage facilities. On a large production scale, the average yield is 4-6 vaccine doses per Iharvested embryo Embryo mortality due to nonspecific causes does not usually exceed 15%

Fertile hens' eggs incubated for 7 days at 36 5 ' C are used, It is best to obtain fertile incubated eggs for vaccine production from a commercial hatchery and transfer the eggs to the laboratory 1 day before inoculation.

1 On the 7th day of incubation candle the eggs using a concentrated light source. Discard any eggs that are cracked infertile, contain displaced air sacs,

Fig. 23.1 Piercing the air sac

By courtesy of R Gldck Head Department of Virology SWISS Ser~im and Vaccine Iiistrtute, Berne Switzerland

261


or contain dead or weak embryos (i.e. embryos that are poorly vascularized or slow-moving). Mark the boundar~es of the air sac on the shell with a pencil and place the eggs with the air sac upperrriost in a tray (e.g. a commercial cardboard egg-tray).

2. Disinfect an area around the top of the air sac with 70% ethanol and flame it briefly.

3. Using a carborundum disc attached to a rotary motor tool or a dental drill, make two crosswise cuts, or penetrate the shell directly at the top and centre of the air sac, I f this type of cutting machine is not available. pierce the shell at the point indicated with any sharp instrument (see Fig. 23.1).

4. Using a syringe with a 090-1.00-mm X 20-25-mm (19-20-gauge) needle, inoculate 0.25 m1 of the working stock suspension directly into the yolk sac. Fill the syringe with the inoculum and hold it in the right hand; insert the needle vertically into the egg for its entire length (see Fig 23 2) Withdraw the needle carefully after inoculation

5. Seal the ~noculation hole with a few drops of a melted mixture of paraffin and petroleum jelly (two parts paraffin, one part petroleum jelly), or 4% collodion. Test the bacterial sterility of the work~ng stock suspension.

After the eggs have been inoculated they should be incubated at 3 6 5 - C for 9-10 days

Harvest of infected embryos

1 . At the end of the incubation period, candle the eggs and mark the edge of the air sac Discard any eggs contain~ng dead embryos.

Fig. 23.2 Yolk-sac inoculation of egg

By courtesy of R Gluck. Head, Department of V~rology, Swiss Seru~n and Vaccine Inslitufe, Berne, Swt/tzet land.


2. Pace the eggs in a tray, and disinfect the shell with 70% ethanol and briefly flame it.

3. Remove the live embryos by one of the following methods:

(a) Using either a sterile scalpel or sharp-pointed scissors, remove the shell over the air sac and discard (see Fig. 233) Tear the chorioallantoic membranes, so tliat the embryo is exposed. Pass a wire hook attached to a handle under the neck of the embryo, and pull the embryo slowly iipward. Tile embryo will be pulled out of the egg, free from the yolk and extra- embryonic material. Alternatively, i t l t the embryo out with orle p a r of forceps whle separatng the membranes with another (see Fig. 23.4).

(b) Put on sterile rubber gloves. U s i ~ g a sharp sterile instrument. crack the sheii of the egg at the point indicated. Open the shell with gloved hands and pour the contents over a sterile wire gauze placed or] top of a sterile container The fluid and yolk should drop into the sterile container, while the embryos can be collected from the layer of gauze.

4. Place the embryos in a chilled weighed container. Discard all other materials, such as egg fluids. yolk, yolk-sac and extra-embryonic tissue

5. Rewegh the container with the collected embryos and note the weight.

Preparation of embryonic sospension

The infected embryos should be homogenized immediately after harvesting This is best done using a Ihoniogenzer of the Warng blender type a 1-litre stainless steel blend~ng jar mounted on the standard motor is suiable for this purpose

Fig. 23.3 Cutting off shell cowering air sac

By courtesy of R. Giiick, Head, Department of Virology, SWISS Seriini and Vaccine insi~tute, Berne, Sw~tzeriar~d.

263


Fig. 23.4 Separating embryo from membranes

Bv courtesy ot R. Gluck, Head, Department o f Virology Swiss Serun-i and Vacc~ne Institute. Berne, Switzerland

1. Place one group of weighed embryos in a chilled blending jar. 2. Add sufficient chilled sterile distilled water to make a 66% w,'v suspension

(approximately 5 ml of distilled water per 10 g of embryos). I f bacterial contamination is suspected, penicillin (1000 IUtmI) and streptomycin (1560 lU/ml) should be added to the water.

3. Close the blending jar and transfer it to the power unit. 4. Switch on the motor at rriaximum speed (approximately 19000 revlmin) and run

for 60 seconds only. Wait 1 minute and then run at maxnium speed for a further 60 seconds Wait again for 1 minute and run for a final 60 seconds, also at maximum speed.

4. Filter the resulting susperislorl itlrough two or three layers of sterile wire gauze

into a chilled sterle container. Measure the volume, and d~lute w t h an equal volume of stabilizing solution (see Annex) to produce a 33% wlv tissue suspension.

5. Distribute the suspension in 3-m1 amounts into single-dose vials. 6. ILyophIize the vial contents using the procedure recommended by the manufac-

turer of the Iyophilizer. The additon of stabilizing solution to the vaccine before freeze-drying greatly increases the resistance of the virus to thermal nactiva- lion.

7. Seal the vials under vacuum or under an atmosphere of nitrogen and store at 4 'C.


Control tests

Control tests on the raw virus suspension

Control tests on the raw virus suspension are carried out according to standard procedures Consequently details are not given here

Tests for sterjljty

Tests for the presence of pathogenic organisms are made in mice. Eight young adult rr~ice are each inoculated ~ntraperitoneally or subcutaneously with 0.5 m of the reconstituted vaccine. At least seven of these nrce must rernain healthy during a 7-day observation period.

A sample of bulk material collected from each batch before the addition of antibodies should be tested by standard procedures for the presence of Salmo- nella.

Specific safety test

A specific safety test on the lyophilized product is made in at least two I-3-month- old dogs. Each dog is inoculated ntramuscularly with the recommended dose and observed for 21 days. For the test to be satisfactory, all the dogs must remain free of any symptoms that might be attributed to the vaccine. The dogs used in this test should not be used in any subsequent safety test.

Assay of virus infectivity

1 Each of two single-dose vials of final vaccine from each lot is reconstituted by adding 3 0 m of sterile distilled water containing 2% of normal horse serum Withdraw half a dose (1 5 ml) from each of the two vials of reconstituted vaccine or a full dose (3.0 ml) i f the two vials are combined, and add this to 7 m1 of sterile distilled water containing 2 % of normal horse serum in a test tube This gives 10 ml of diluted vaccine representing a 10- I dilution

Penicillin (500 iU/ml) and streptomycin (780 lU/ml or 100 icg~ml) may be added to the diluent

2 Using recommended laboratory procedures (see Chapter l ) , make further tenfold drlut~ons from the 10- ' dilution prepared above A suggested method is as follows

(a) Prepare an appropriate number of test tubes with 4.5 m1 ofthe sterile diluent referred to above.

(b) Using a l - m l pipette transfer exactly 0.5 m1 of the reconstituted, well mixed sample (10-' dilution) to the first tube without touching the diluent with the pipette. Discard the pipette in an appropriate receptacle.

(c) Take another I-m1 pipette and mix the suspension in the f~rst tube by drawing and expelling the suspension in and out of the pipette at least 5 or, preferably, 10 times. The mixing should be done as vigorously as possible withoiit causing bubbles to form extensively. Using tile same pipette, transfer exactly 0.5 ml of the sample ( 1 0 - ~ dilution) to the second tube


Volume of diluent Volume (m1 per dog dose transferred

Tube no. of dried vaccine) (ml) Resultant dilution

without touching the diluent with the pipette. Discard the pipette. Continue this procedure until all the tubes have been used.

(d) The reconstituted vaccine (10-I dilution) is mixed with the sterile diluent as described above to give the dilutions shown in the table.

Tubes containing the desired dilutions are placed in an ice-bath. The highest dilutiori to be used is optional; however, the lowest dilution used should kill at least 80% of the mice. Each dilution is administered intracerebrally to a group of at least 10 young (4-6-week-old) adult mice weighing 15-19 g each mouse receiving 0.03 ml. At least 80% of the mice inoculated with each dilution must survive longer than 4 days or the test is considered invalid and must be repeated. Any mice that die within 1 hour of inoculation may be replaced.

The 6 groups of mice are kept under observation for 14 days and a record is kept of the number of paralysed animals that die from the 5th day onwards. The LD,, titre is calculated by the method of Spearman-Karber (see Appendix 3) or Reed & Muench (see Chapter 38, Annex). The virus titre should be no less than 103.3 LD50 per 0.03 rnl This level should be maintained throughout the expiration period. However, it is suggested that the minimum titre for release should be 0.5 log higher, i.e. IO3.' LDS0 per 0.03 ml.

References

1. Kornarov A, Horenstein K. Studies on the pathogencity of an avianized street rabies virus. Gornell veterinarian. 1953, 43: 344-361.

2. Koprowski H, Cox HR. Studies on chick-embryo-adapted rabies virus; culture ctiaracteristics and pathogenicity. Journal of immunology, 1948, 60: 533-544.

Annex Preparation of stabilizing solution, pH 7.6

Potassiurn phosphate, monobasic (KH,PO,) Peptone Lactose Distilled water to make


The mixture must be heated slightly In order to dissolve the lactose. Cool the mixture to room temperature and adjust the pH to 7.6 with sodium hydroxide (NaOH), 1 mol/l. The soluton should be slerilized by filtration through Seitz- or Millipore-type filters under positive pressure; it should not be autoclaved.

Cell-culture vaccines

CHAPTER 24

Cell-culture vaccines for human use: general considerations K G, Nicholson'

The production of rabies vaccines in cell cultures has resiilted in higt~ly purified vaccnes with greatly improved potency and safety. The potency of cell-culture vaccnes is determined by the N H test and is expressed in International Units (IU) per ml; to satisfy the requirements for human rabies vaccine published by WHO (7 -3) cell-culture vaccines should have a minimum potency o i 2.5 IU per dose.

Human diploid cell vaccine

Human diploid cell (HDC) vaccine, the first commercially prepared cell-culture vaccine to gain widespread acceptance has become the standard against which all other rabies vaccines prepared in cell culture have been compared The HDC vaccines produced by the Merieux Institute in Marcy l Etoile, France and the Behr~ng Institute in Marburg Germany are both whole virion preparations grown in MRC-5 human diploid cells and inactivated with P-propiolactone (BPL) The vaccine produced by the Behrng Institute is concentrated and purified by rate zonal ultracentrifugation whereas that produced by the Merieux Institute is concentrated by ultrafiltration

Early trials in volunteers rapidly established that HDC vaccnes were superior to nerve-tissue and avlan preparations Not only were virus-neutralizing antibody levels far higher than with other vaccines but they appeared earl~er and in practically 100% of vaccinees with fewer doses and only m i d local reactions In 1976 trials In the Federal Republic of Germany and Iran showed that HDC together with purriied equine or human rabies immunoglobulins provided complete protection to persons bitten by animals with proven rabies The vaccination schedule consisted of six doses of 1 ml given on days 0 3 7 14 30 and 90 This regimen is now ofiicialy recommended by WHO although the last booster dose (on day 90) has been dropped (4)

The major disadvantage of HDC vaccine is its relatively high cost compared with other cell-culture vaccnes which is parily due to the difficulty of handling diploid cells In addition the virus harvests must be concentrated 10-20-fold to achieve vaccnes o i adequate potency None the less more than 10 nii l ion doses had been used to treat several million patents worldwide by December 1994

Other cell-culture vaccines

In view of the expense of HDC vaccine, manufacturers have explored new cell substrates culture systems virus strains and techniques In order to reduce costs

' Ser ior Lcc!u'er in infec!ious Diseases, Departrricr?; of Mirlnblology b lecca l Sciences Bti idir lg Leicester Un~versity Ur>iveis~ty I ioad Leicester Englario


and yet retain the high standards of potency and efficacy of HDC vaccine. In the former USSR an ultraviolet light-nactivated vaccine has been prepared from primary hamster kidney cells (PHKC) infected with the Vnukovo-32 strain of rabies virus (see Chapter 31). A PHKC vaccrie has also been prepared using the Beijing strain of fixed rabies virus inactivated with forinaldehyde (see Chapter 30). Approximately 50 million doses have been produced primarily for post-exposure treatment

Vaccines have been prepared in prirnary chick embryo cells using the HEP and LEP Flilry strains of fixed rabiesvirus inactivated with BPL A purified chick-embryo cell (PCEC) rabies vaccine was developed by Kondo et al (5-7) it? the 1960s and l~censed for Lise in 1980 Since then, this vaccine has been introduced inlo many countries, especially in south-east Asia, manly for post-exposure treatment A similar vaccine was developed in Germany by the Behrng Institute (see Chapter 27) When applied in trials involving 4247 doses in 935 vaccinees, the vaccine was shown to induce high at i tbody titres comparable to those produced by HDC vaccine. Over 2 million doses of this vaccine have been used for post-exposure treatment in Africa, Europe, the Indian subcontinent, and south-east Asia (Thailand)

Vaccines have similarly been prepared on primary cultures of fetal bovine kidney cells and dog kidney cells. Primary fetal bovine kidney cell (FBKC) vaccine, which is prod~iced using the PV-31 strain of Pasteur fixed rabiesvirus, was licensed for post-exposure treatment in France in 1984, but discontinued after 1987. Several reports indicate that FBKC vaccine is well tolerated and highly immunogenic with antibody responses similar to those elicited by HDC vaccine Details of the dog kidney cell vacclne w h ~ c h is p~epared using the PM strain oi fixed rabies vlrus adapted to human diploid cells, are given in Chapter 29

Recently, a vaccine has been prepared using Vero cells (a continuous cell line derived from vervet monkeys) grown on a microcarrier system and infected with the PM strain of rabies virus (see Cliapter 26) Tile use of conlinuous cell lines for the production of human vaccines was not permitted until recently because i t was thought that potentially carcinogenic DNA might be released from the cells during viral replication. However, provided that appropriate carcinogenicity tests are carried out and the product is free from cellular DNA, continuous cell lines are now widely accepted for human vaccine production. The purified Vero cell rabies (PVR) vaccine is well tolerated and has been shown to be highly imrnunogenic in humans, again with antibody responses comparable to those induced by HDC vaccine. Approximately 11 million doses of PVR vaccine had been used worldwide by December 1994.

Safety

Minor adverse reactions

HDC and the more recently available cell-culture vaccines are well tolerated The freqcrency with which minor adverse reactions have been reported varies widely from one study to another primarily because of differences in the methods used to collect information However comparative studies have shown that there are no significant differences between HDC PCEC and PVR vaccines In terms of m n o r local and systeniic side-effects In one such study local pain was noted after

CELL-CULTURE VACCINES FOR HUMAN USE

22.5% and 21% of intramuscular vacciiiations with PCEC and HDC vaccines respectively. In most cases, the pain was mild and usually subsided within 2-3 days (8 j . Local erythema, swelirig and pruritus were noted after 2%. 5% and 2% respectively of intramuscular vaccinations with PCEC vaccine and 0%, 3% and 1 % of those with HDC vaccine. Intradermal adniii?istration of PCEC vaccine was less painful than intramuscular administration (incidence 11 % vs. 22.5%), but was followed by a higher iricidence of erythema (71°h vs. 2%), swelling (35% vs. 5%) and itching (12 5% vs. 2%) (8). Occasional sysleniic reactions were noted after both inlrarnuscular and intradermal injections. the most common were malaise. generalized aches and headaches, which were recorded among 3-5% of vaccnees. In a further comparative study local pain occurred after 2 2% and 7.8%

of vaccinations with PVR and HDC vaccines respectively; syslemic reactions were noted in less than 1 % of vaccinees (L. Teulieres, personal communicatiori)

Hypersensitivity reactions

Between May 1980 and April 1984, 108 allergic reactions to HDC vaccine were reported in the USA (overall incidence 11 per 10000 vaccnees). The reactions ranged from urticara to anaphylaxis and usually occurred after booster injections (9). No significant associations were demonstrated between persons who reported presumed Type Ill hypersensitivity reactions and age, route of admins- tration of the vaccine (intramuscular or intradermal), timing of booster after primary immunization, history of other allergies, or history of previous immunization with rabies vaccines other than HDC vaccine. These reports of systemic allergic reactions included nine cases of presumed Type l hypersensitivity (1 :10000), 87 cases of presumed Type Ill delayed hypersensitivity (9:10000), and 12 cases of indeterminate types of allergic reactions (1 .10000) (10) Most patients did not require admission to hospital and there were no fatalities. Skin testing of five patients who had reported reactions showed that all reacted strongly to the HDC vaccine from the Merieux Institute, but only one showed any reaction to the purified vaccine from the Behring Institute and that was weak (71). Radioallergo- sorbent tests on sera from four patients who had experienced allergic reactions showed that they all reacted positively to tlie Merieux HDC vaccine and to BPL- treated human serum albumin, but not to HDC vaccine not inactivated by BPL, to duck-embryo vaccine inactivated by BPL or to human serum albumin (12). It was subsequently found that adrnnistraion of homologous serum albumin treated with BPL induced anaphylaxis in guinea-pigs (73), and this led to the conclusion that the allergic reactions were caused by the BPL-treated human serum albumin componerit of the HDC vaccine.

Three allergic reactions were noted following immunization with an early batch of PCEC vaccine (8): however, post-marketing surveillance has not revealed any further hypersensitivity reactions to the vaccine There have been no reports of hypersensitivity reactions to PVR vaccine.

Neurological reactions

In contrast to the high incidence of neuroparalytic reactions to vaccines prepared from brain tissues (see Chapter 19) the incidence of neurological reactions to


HDC vaccine is extremely low. By June 1990, six cases of neurological reactions associated with the use of HDC vaccine had been reported (14-18) In five cases weakness or paraesthesia was transient although one patlent suffered residual wasting of the deltoid muscle The sixth patient developed a relapsing neurological illness which resembled multiple sclerosis Even if all six cases were caused by vaccination with HDC vaccine, the rate of neurological complications would be about 1 per 500000 patients, similar to that for vaccines such as oral poliomyelitis, yellow fever and tetanus toxod.

One case of peripheral neuropathy has been reported in association with FBKC vaccine ( 19).

Efficacy

A study conducted during 1980-82 in the United States showed that of 51 1 patients who had been bitten by animals with proven rabies and treated with five doses of HDC vaccine plus rabies immunoglobulin (RIG) none died from rabies (10). The vaccine has also been reported to be effective in at least 400 other cases (see page 271 ).

Post-exposirre treatment trials of other tissue-culture vaccines that satisfy the minmurn potency requirements (see page 272) have yielded similar results. The efficacy of PCEC vaccine was tested in 69 patients who had been bitten by animals with proven rabies in Thailand (20). The vaccine was given intramuscularly according to the schedule recommended by WHO (21); 32 patents were also given human rabies immunoglobulin (HRIG). None of the patients developed rabies. Similarly, no cases of rabies were found among 47 patients who were treated with PCEC vaccine and HRIG in Yugoslavia, most of whom had only mild exposure (category 11) to rabies (22). FBKC vaccine was also used successfully for the post- exposure treatment of 153 patients who had been exposed to animals with rabies, nine of whom had major bites The vaccine was given intramuscularly according to the recommended schedule; two patients were also given HRIG. All patients scrrvived the 6-month period of follow-up.

The efficacy of PVR vaccine was tested in 106 patients who had been bitten by animals with proven rabies (23). The vaccine was given intramuscularly according to the scliedule reconimended by WHO; 47 patients with severe exposure (category Ill) were also given HRIG. A I patients were alive and well after 1 year. The vaccine has also been used successfully in at least 500 other cases.

Post-exposure "treatment failures"

By June 1988 a total of 17 people had developed rabies in spite of post exposure treatment with HDC PCEC or PVR vaccine 11 had been treated with HDC vacclne (24) A further 12 deaths were reported between 1988 and 1991 (25-27) A review of the first l 7 deaths at an informal WHO meeting revealed that 13 had severe (category Ill) exposure I e licks on miicosal surfaces or single or multiple trarisdermal bites or scratches At least 11 of these 13 patients had not received passive immunization with RIG as recommended by WHO and only one had received treatment with anti-rabies imrnunoglobulin around the wound Treatment had also been delayed In most cases After careful anays~s by the expert group it


was agreed that only one could defintely be considered a treatment failure. Similarly, treatment had been delayed by 6 days in one of the subsequent cases (26), while in another (involving a category l l exposure), the patient had not rece~ved RIG or local treatment of the wound (25).

By early 1988, 17 cases of human rabies had been reported ii i spite of post- exposure treatment with the Viiukovo-32 primary hamster kidney cell (PHKC) vaccine. However, most, if not all, of these cases were associated with delayed or inadequate therapy. and should not be considered as treatment falures. In China, four people died after being bitten by a rabid woif on the head (three cases) or the forearm (one case), none received antirabies serum and treatment with PHKC vaccrle was either delayed 01- incomplete (28). In a further report, 10 deaths

occurred among 110 patients who had been bitten by rabid dogs and were treated with PHKC vaccine (29). Nine of the patients who died had severe (category Ill) exposure: however, none of these patients were considered treatment failures, since treatment was inadeauate

Economical post-exposure treatment regimens

Unfortunately the countries most in need of safe and effective rabies vaccines are the least able to afford them For example in Thailand in 1987 a post-exposure treatment course with HDC vaccine cost US$ 1 5 6 a b o u t 17 times the cost of sucking-mouse brain (SMB) vaccine and 48 times the minimum daily wage Although the cost of HDC vaccine has severely limited its use in developing countries there is corivincng evidence that as little as 1 4 m1 of the vaccine given intradermally in multiple sites induces adequate antibody titres A study was carried out among 78 patents who had been bitten by a rabid dog or cat (30) 36 of whom had severe exposure All were given a regimen of 0 1 ml intradermally at each of eight sites (on both sides of the deltoid lateral thigh supracapsular and lower abdominal req~ons) on day 0 at four sites on day 7 (the deltoids and lateral thighs) and at one site (the deitoid) on days 28 and 91 patients with severe exposure also received RIG Although one patlent was lost to follow-up the remaining patents were reported to be alive and well 1 year after exposure

Multisite ntradermal vaccination has been used in India Kenya and Thailand (26) By July 1988 PVR vaccine had been given to more than 10000 patients in Bangkok using a variation of the above regimen known as the 2-2-2-1-1 reglmen This regimen has been endorsed by the WHO Expert Committee on Rabies (4) l i consists of one dose (0 1 ml) of vaccine given at each of two sites on days 0 3 and 7 and at one site on days 30 and 90 To date only one death has been reported with this regimen which was n a patient who was treated 6 days after exposure There are reports of a furiher 157 patients who were treated with multisite ntradermal regimens using HDC PVR and PCEC vaccines following severe exposure to rabies and all survived (31 32) These regimens therefore appear to be at least as effect~ve as the intramuscular regimens

The multisite intraderrnal regimens lower the cost of vaccination with cell- culture vaccines by around 70%) compared with the conventional intramuscular schedules Multis~te intradermal regimens using PVR or PCEC vaccines cost US$ 16-20 2-3 times as much as intramuscular schedules using SMB vaccine However since fewer c l in~c visits are required than with tradit~onal treatments, there are direct savings on needles and syringes as well as indirect savings on lost


earrings and travel expenses These savings more than compensate for the higher costs of cell-culture vacclnes The incidence of vacclne related compiications and treatment failures is also rnuch lower than with nerve-tissue vaccnes In addLion the rnultsite inlradermal regimens are less panf i l l than the tradltional intramuscular schedules and tend to be preferred by patients


The availability of HDC and other cell-culture vacciries of high immunogenicity and safety has enableu persons who are at high risk of exposure to be protected by pre-exposure ~mrnun~zation. Pre-exposure immunization is also used to protect individuals who are living in or travelling to areas where rabies is endemic.

Recorninendations regarding pre-exposure immunization have been published by WHO and by some national authorities, At its eighth meeting, the WHO Expert Committee on Rabies (4) recommended that such immunization should preferably consist of three full intramuscular doses of cell-culture vaccine of potency at least 2.5 IU per dose given on days 0, 7 and 28. For adults, the vaccine should always be adninistered in the deltoid area of the arm. For children. the anterolateral area of the thigh is also acceptable. Alterriatively, the vaccine may be administered intradermally in 0.1-ml volumes on days 0, 7 and 28. (A few days' va r~a t~on 1s not important.)

Intraderma application is of particular interest in areas where economic coristraints limit vaccine availability. In 1982 the public health authorities in the United States endorsed the use of cell-culture vaccnes given by the intradermal route after reviewing data from 11 studies in Europe and the USA; satisfactory antibody responses were seen in all vaccinees given three full intradermal doses. Several studies showed that vaccinees should be protected against challenge even if all three doses are given percutaneously However, pre-exposure immunization with HDC vaccine administered intradermally should, whenever possible, be performed before starting antimalarial prophylaxis, since virus-neutraliz~ng antibody levels have been shown to be lower in patients rece~ving chloroquine phosphate. When this is not feasible. HDC vaccine should be administered intramuscularly.

All persons who work with live rabies virus in a diagnostic, research or vaccine production laboratory should have a serum sample tested for rabies virus- neutralizing antibodies every 6 nionths and a booster administered when the titre falls below 0.5 lU/ml. All other persons at continuing risk of exposure to rabies should have a serum sample tested for rabies virus-neutralizing antibodies every year; a booster should be administered when the titre falls below 0.5 1U;ml. It has been shown that two doses of cell-culture vaccine, given by the intramuscular or intradermal route, induce adequate antibody titres in virtiially all vaccinees. This suggests that it may not be necessary to check the antibody response of persons at low risk of exposure to rabies. However, many authorities, including WHO, recoinmend that a serologcal test sliould be performed 1 3 weeks after the last dose, to ensure that virus-neutralizing antibody titres are adequate (> 0.5 lU,!mi).

References

1 . WHO Expert Committee on B~o lo~ i ca i Standardization. Thirty-seventh report.

276


Geneva, World Health Organization, 1987 (WHO Technical Report Series. No 760), Annex 9.

2. WHO Expert Committee on Biological Standardization. Thirty-first report Geneva, World Health Organization. 1981 (WHO Technical Report Series, No 658). Annex 2.

3 WHO Expert Committee on Biological Standardization, Forty-third report. Geneva, World Health Organ~zatiori, 1994 (WHO Technical Report Series. No. 840), Annex 4; Annex 5.

4. WHO Expert Co~nmittee on Rabies. Eighth report. Geneva, World Health Organization, 1992 (WHO Technical Report Series. No. 823).

5. Kondo A, Takashin-~a Y, Suzuki M, Inactivated rabies vaccine of chick embryo cell culture orgin. Syinposia series in immunobiologicalstandardization 1974. 21: 182-189.

6. Kondo A. Pre-immunization and post-exposure treatment with inactivated rabies vaccine of chick embryo origin (CEC). Developments in biological standard~zation, 1978, 40: 147-1 53.

7 Kondo A The mlnmum requ~rement of dried inactivated t~ssue culture rabies vaccine (CEF rabies vaccine) In Kuwert EK Wiktor TJ Koprowski H eds Cell- culture rabies vaccines and their protective effect in man (Proceedings of WHO Consultations Essen 5 7 March 1980) Geneva International Green Cross 1981 245-248

8. Nicholson KG et al. Pre-exposure studies with purified chick embryo cell- culture rabies vaccine and human diploid cell vaccine: serological and clinical responses in man. Vaccine, 1987, 5: 208-210.

9. Systemic allergic reactions following immunization with human diploid cell rabies vaccine. Morbidity and mortality weekly report, 1984, 33 185-1 87.

10. Winkler G. Current status of use of human diploid cell strain rabies vaccine in the U.S.- May 1984. In. Vodopija l et al., eds, Improvements in rabies post- exposure treatment. Zagreb, Zagreb Institute of Public Health, 1985. 3-8.

11. Changes recommended in use of human diploid cell rabies vaccine. Journal of the American Medical Association. 1985, 245: 1 4 15.

12. Baer H , Anderson MC, Ouinnan G. Beta-propiolactone treated human serum albumin (BPL-HSA): an allergen for humans receiving rabies vaccine. Journal of allergy and clinical immunology, 1985, 75: 137.

13. Levenbrook IS et al. Sensitization induced in guinea pigs with beta- propiolactone treated serum albumin: experimental evidence for the cause of


allergic reactions in humans receiving human diploid cell rabies \/accines. lnteriiational archives of allergy and applied immunology, 1986, 80: 110-1 11

14. Gardner SD. Prevention of rabies in man in England and Wales In: Pattison JR, ed. Rabies: a growing threat. Wokingham, van Nostrand Reinhold, 1985: 39-49.

15. Boe E, Nyland H. Guillain-Barre syndrome after vaccinahon with human diploid cell rabies vaccine. Scandinavian journal of infectious diseases, 1980, 12: 231 -232.

16. Bernard KW et al. Neuroparalytic illness and human diploid cell rabiesvaccine. Journal of the Amei-ican Medical Association, 1982, 248. 31 36-31 38.

17. Knttei T et al. Guiilain-Barre syndrome and human diploid cell rabies vaccine. Lancet, 1989, i: 1334 1335.

18. Tornatore CS, Richert JR. CNS demyelination associated with diploid cell rabies vaccine. Lancet, 1990, 335. 1346-1 347.

19. Courrier A et al. Peripheral neuropathy following fetal bovine cell rabies vaccine. Lancet, 1986, i: 1273.

20. Wasi C et a!, lmmunogenicity and reactogenicty of the new tissue culture rabies vaccine for human use (purified chick embryo cell culture). In: Vodopija l et al., eds. lmprovements in rabies post-exposure treatment. Zagreb Zagreb Institute of Public Health. 1985: 85-94.

21. WHO Expert Committee on Rabies. Seventh report. Geneva, World Health Organization, 1984 (WHO Technical Report Series, No. 709).

22. Ljubicic M et al. Efficacy of PCEC vaccine in post-exposure rabies prophylax~s. In: Vodopija I et al., eds. Improvements in rabies post-exposure treatment. Zagreb, Zagreb Institute of Public Health, 1985: 95-101.

23. Suntharasamai P et al. New purified Vero cell vaccine prevents rabies in patients bitten by rabid animals. Lancet, 1986. ii: 129-131.

24. Sureau P. Analysis of human rabies cases despite vaccination, In: Thraenhart 0 et al., eds. Progress in rabies control. (Proceedings of the Second international IMVI Essen,/WHO Symposium on "New Developments in Rabies Control': Essen, 5-7 July 1988 and Report of the WHO Consultation on Rabies, Essen, 8 July 1988.) Royal Tunbridge Wells, Wells Medical, 1988: 421-424.

25. Discussion. In: Thraenhart 0 et al., eds. Progress in rabies control. (Proceed- ings of the Second International IMVl Essen/WHO Symposium on "New Deveiopments in Rabies Control", Essen, 5-7JuIy 1988 and Report of the WHO Consultation on Rabies. Essen, 8 Juiy 7988.) Royal Tunbridge Wells, Wells Medical. 1988: 448-450


26 Wilde H. Chutivongse S. Rabies in Thailand. economic perspectives and the intraderrnal vaccine regimen. In: Thraenhart 0 et al.. eds. Progress in rab~es controi. (Proceedings of the Secorid International IMVi Essen/'WHO Sym- posium on "New Developments in Rabies Control': Essen, 5-7 July 1988 and Report of the WHO Consultation on Rabies, Essen. 8 Juiy 1988.) Royal Tunbr~dge Wells, Wells Medical. 1988. 529-535.

27. Fescharek R et al. Post-exposure rabies prophylaxis, when the guidelines are not respected. Vaccine, 1991, 9: 868-892.

28. Lln FT et al. Further s tudy on the stability arid etf~cacy of t h e pririiary hamstet-

kidney cell rabies vaccine. In: Vodopija I et a l . eds, Improvements in rabies post-exposure trealment, Zagreb. Zagreb Institute of Public Health, 1985. 37-45.

29, Lin FT et al. The protective effect of PHKC rabies vaccine used in man on a large scale. In. Thraenhart 0 et a1 , eds. Progress 117 r.abies control. (Proceed- ings of the Second Inten'lationat /MV/ Essen/'WHO Syinposium on "New Developments in Rabies Control". Essen. 5-7 Jiily 1988 and Report of the WHO Consultation 017 Rabies, Essen, 8 Jiily 1988.) Royal Tunbridge Wells, Wells Medical. 1988. 5055515.

30. Warrell MJ et al. Economical rnuitiple-srte intradermal immunization with human diploid cell strain vaccitie is effective for post-exposure rabies prophylaxis Lancet, 1985. I: 1059-1062.

31. Madhusadana SN et al. Multisite intraderrnal vacciriation using tissue-culture vaccine as ari economical prophylactic regimen against rabies. Indianjournal of medical research, 1988, 87: 1-4

32. Chutivongse S et al. Postexposure prophylaxis for rabies with antiserum and intradermal vaccination. Lancet, 1990, 335: 896-898.

CHAPTER 25

Vaccine for humans prepared in human diploid cells R, Branche '

An inactivated rabies vaccine for human use was first prepared in cell culture in 1964 ( l ) In 1966 it was shown that the human diploid cell (HDC) strain WI-38 was a suiiabe substrate for the propagation of the Pitman-Moore (PM) strari of fixed rabies virus (2) The original procedure for the production of t h ~ s vaccine was described in the previous edition (3) Since 1967 research and development have been carried out on this vaccine at the Merieux Institute The vaccine was first licensed for use in France in 1974 (4) and commercial production started in 1978


Cell cultures

Thevirus is now cultivated in MRC-5 human diploid cells ( 5 ) which are propagated and controlled according to the recommendations published by WHO (6 8 ) and the regulations of national authorities.

Seed lot of virus

The working seed lot is prepared from the PM strain and frozen in aliquots at - 70 C Each aliquot is used to prodi~ce a single batch of vaccine Infectious

titrations of the virus are performed by intracerebral inoculation of adult mice When kept at - 70 "C the virus suspension is perfecily stable for at least 11 years The ~r l fect~v~ty titre of the virus In the seed lot should not be lower than 106' LD,, per nil

Infection of cultures and propagation of virus

Every batch of vaccine is produced starting from a working cell bank consisting of a s ngle arnpoule of the cell seed at the sixteenth population doiibllng level (PDL) The cells are expanded by serial subculture up to the 29 5th PDL (see Annex)

infection of ceils it? susperision 1 Inoculate freshly trypsnized cells at the twenty-seventh PDL in suspension with

the seed virus at an input multiplicity of 1 LD,, per 25 50 cells and stir with a magnetic stirrer for 15 minutes at 37 C

' Fornwr Heaa, Rabies Vacc~ r i e Produc:ion Oepartnieni, Sera and Vacc~ i l es . Mer~eux I?sti!ute Marcy ,'Etoi.e Fpanre

HDC VACCINE FOR HUMANS

2 Seed the cells in culture flasks and add culture rned~um The culture rned~um consists of the standard Eagle s basal medium (EBM) (see Chapter 18 Annex) containing 2 5 g of sodiuni bicarbonate (NaHCO,) and 5 mg of neomycln sulfate per litre and supplemented with 10% fetal calf serum It should be gassed w,th carbon dioxide (CO,) before use

3 lncubate at 37 C for 3-4 days until a complete rnonolayer is formed The cells should be at the 285th PDL Remove samples for testing for bacterial contaminants and adventitious agents

4 Wash the cells with phosphate-buffered saline (PBS)

5 Trypsinize the cells and replant in equal port io~is in two new culture flasks 6 Incubate at 37 C for 3-4 days in culture rned~um (see step 2)

7 Discard the supernatant containing fetal calf serum and wash the cells three times with a proteln-free medium such as EBM to remove any remaining bovine proteins

8 Replace the culriire medium by virus propagation niedium consisting of EBM containing 4 g of sodium bicarbonate and 5 mg of neomycin sulfate per litre and supplemented with 0 3% human albumin It should be gassed with CO, before use

9 Incubate the culture flasks for 3-4 days at 35 C Remove samples for testing for bacterial contaminants and adventitious agents

10 Harvest the culture supernatants three times at Intervals of 3 4 days

Clarification of infectious medium

The harvested virus is clarified by filtration through a 08-pm membrane Virus samples are pooled to constitute the b i ~ l k virus suspension

Concentration and purification of the virus

The bulk virus suspension is concentrated and purified by ultrafiltration through a membrane with a relative molecular mass cut-off of 10000. followed by flitration through a 0.45-pm membrane

Control tests

Testing of the cells

The cell seed should be approved by and registered with the national control authority and should comply with the requirements for rabies vaccine for human use published by WHO (6, 8) for freedom from extraneous agents, lack of tumorigencty, normal karyology ihrougliout approximately the first two-thirds o f its normal life-span, and identity

Testing of the virus strain

The virus strain to be propagated in diploid cells should be a fixed strain and should be ident f~ed by historical records The strain should have a short stable and reproducible incubation time when administered intracerebrally to suitable


animals and shocrld not form Negri bodies. In addition, it should be shown, by tests in ani~nals and cell cultures, to be free from extraneous agents.

Tests on the cell culture

Absence of extraneous agents At least 10% of the cell suspension at the twenty-sixth PDL should be used to prepare control cultiires. These cultures should be processed in the same way as the production cell cultures, but not inoculated with Lhe virus. They should be observed for at least 2 weeks for evidence of any cytopathological changes.

At the end of the observation period, the control cell cultures should be examined. I f [his examination shows evidence of the presence in a control culture of any extraneous agent, the virus grown in the corresponding inoculated cultures should not be used for vaccine production.

Other tests At the time of the harvest of the production cultures, the supernatants of the control c~~ l tu res should be collected, pooled and tested for mycoplasmas and for sterility A small quantity of cells should be tested for normal karyology

Tests on inactivated virus

Test for absence of live virLls This test is carried out on the inactivated virus 10, 12, 14 and 26.5 hours after the addition of /l-propiolactone. A sample (0.03 ml) of the virus suspension is inoculated intracerebrally into a group of 20 adult mlce which are observed for 21 days. The test is satisfactory if none of the mice show any signs of rabies.

Tests on bulk vaccine

Test for absence of live virus At least 25 m of the bulk vaccine is ~noculated into human diploid cell cultures, which are observed for 21 days. At day 14 and 21, supernatant samples are removed and inoculated intracerebrally into a group of 20 adult mice, which are observed for 21 days. The test is satisfactory if none of the mice show any signs of rabies.

Test for glycoprotein content The glycoprotein content of the bulk vaccine IS determiried by if? viiro tests such as single radial immunodiffusion, radioimmunoassay (9) and the antibody-binding test (10) (see Chapters 40 and 43).

Other tests Tests for sterility and for freedom froni mycoplasmas are performed

Tests on final vaccine

Potency test The potericy of each final lot of vaccine is determined by the NIH test (9) (see also

HDC VACCINE FOR HUMANS

Chapter 37) The potency should be no less than 2 5 1U per human dose and should be measured after- 1 month a l 4 "C and 37'C

Test for glycoprofei~i content The glycoprotein content of each final lot of vaccine is determ~ned by ~n vitro tests (see above).

Other tests Each final lot of vaccine is iested for appearar~ce, residual nioisture content. total protein content, sterility and safety.

Conclusion

Since 1974, more than 10 million doses of this vaccitie have been used in humans, the majority for post-exposure treatment. It has been shown to be safe and effective when used properly.

References

1. Wiktor TJ, Fernandes MV, Koprowsk H Cultivation of rabies virus in human diploid cell strain W-38. Journal of immunology, 1964, 93. 353-360.

2. Koprowski H. In vitro production of antrabies virus vaccine. Symposia series in imrnunobi'ological sfandardization, 1966. 1 : 357-366.

3. Koprowski H. Vaccine for man prepared in human diploid cells. In: Kaplan MM. Koprowski H, eds, Laboratory techniques in rabies, 3rd ed. Geneva, World Health Organization, 1973 (WHO Monograph Series, No. 23): 256-260

4. Petermann HG, Lang R, Branche R. Real~zation of a new rabies vaccine from cellular culture. Comptes rendus des seances de !a Socikte de Biologie et de ses filiales. 1967, 265: 2143-2144.

5. Jacobs JP, Jones CH, Bailie JP. Character~st~cs of a human dlploid cell designated MRCS. Nature, 1970. 227: 168-170.

6. Requirements for rabies vaccine for human use. WHO Expert Committee on Bio/ogica/Star7dar~diration. Thirty-firstreport. Geneva, World Health Organiza- tion, 1981 (WHO Technical Report Series. No. 658), Annex 2.

7. Acceptability of cell substrates for production of biologicals. Report of a WHO Study Group. Geneva, Worid Health Organ~zation, 1987 (WHO Technical Report Series. No 747).

8. Requirements for rabies vaccine for human use (amendment 1992). WHO Expert Comm~tfee on Biological Standardizat~on. Forty-third report. Geneva, World Health Organization, 1994 (WHO Technical Report Series, No. 840), Annex 4.


9 W k t o r TJ. A radioimmune assay for rabies binding antibody. In- Kaplan MM, Koprowski H, eds. Laboratory techniques ~n rabies, 3rd ed. Geneva, World Health Orgar?~zat~on. 1973 (WHO Monograph Series, No. 23): 182-185.

10 Arko RJ Wiktor TJ Sikes R l i The antibody binclng test for vaccine potency In Kaplan M M Koprowsk H eus Laboratory techn~ques in rabies 3rd ed Geneva World Health O r g a n i ~ a t ~ o n 1973 (WHO Monograph Series N o 23) 292-294

Annex Flow chart for the production of HDC vaccine using the MRC5 cell strain

Passage or population doubling

Day level (PDL) Step Container

One thawed ampoule of seed vlrus

One 75-m1 culture flask Two 75 mi c,!lture flasks Culture flasks tip to 1 2 l~tres C ~ l t ~ r e flasks d p to 4 8 l'tres Culture flasks up to 1 9 2 Itres

lnocuiatlon of the Cultore ilasks up to 38 4 Itres seed virus

Culture flasks up to 115 2 litres

30 or 31 29 5 Frst 2 rnsngs Culture flasks up to 230 4 litres

31 or 32 29 5 3rd rns ?g 34 or 35 29 5 I st harvest 38 or 39 29 5 2nd harvest 42 or 43 29 5 3rd harvest 43 29 5 Concentration of

the virus 44 19 5 Inactvaton with

BPL

45 29 5 Inactvation wth BPL

Control tests Distr~but'on of the

ina l vaccne Into ampodles

Lyophrlizaton

CHAPTER 26

Purified Vero cell vaccine for humans B. Montagnon ' & B. Fanget2

The previous chapter described the rabies vaccine for human use prepared in human d i p o d cells In spite of its safety and high immunogenicity the relatively low titre o l virus productiori by these cells constituted a limitation to the large scale production of a coniparatively cheap rabies vaccne of equal quality The inactivated poliomyelitis vaccine was the first vaccine made using this cell substrate and led to the revision of the requirements for the vaccine by the WHO Expert Committee on Biological Standardization ( I )

The technique developed by van Wezel (2) consisting of the culture of cells on microcarriers stimulated large-scale cultures of cells for human vaccine preparations Following the production of the inactivated poliomyelitis vaccine in Vero cells (3) studies were carried out to develop a human rabies vaccine The resulting vaccine which required a purification step in order to remove the residual cellular DNA is known as the purified Vero cell rabies vacci?e (PVRV) (4-6)

Cell cultures

The Vero cell line was established in 1962 starting froni a primary culture of vervet monkey (Cercop/t/?ecus aethiops) kidney cells ( 7 ) After further passages the cell line was transferred to the American Type Culture Colleclion (ATCC) and in 1979 a prlmary cell bank (PCB) was establrshed starting from one ampoule of the cells at the 124th passage level (8 9) The PC6 was used to prepare several manufacturers working cell banks (MWCBs) The former was tested according to the existing requirements for continuous cell lines used for inactivated virus vaccine production ( I ) for the latter the tests were conipleted taking into account the latest requirements available (10)

As potential tumorigencity was the main concern associated with the use of these cell lines studies were r i t ia ted using athymic nude mice and Syrian hamsters The tests were negative in the hamsters while the mice showed evidence of nodule formation at the inoculation site However even with well-known tumorigenic cells such as HeLa or Hep 2 cells no metastasis was observed in inoculated animals When immunosuppressed newborn rats were inoculated with the tumorigenic cell lines over 90% of the animals were positive for progressively growing tumours and over 4090 showed evidence of rnetastases In the lungs 3 weeks later (11) Inoculation with Vero cells had no effect until the 169th passage level Accordingly the 137th passage level was i ~ s e d for the MWCB and the 142nd passage level was used for the production cell culture

' Head, Reguia1o:y Atiairs, Mer~eilx lnsitute. Marcy l E:oi!e, France Head, Pharrnaceu!cal Depar!meri! Merleuv ivstltl;te Marcy ; 'Eto~ie. F r a m e

285


The Vero cell line has been characterized according to the requirements for continuous cell lines used for inactivated virus prociuction ( 7 ) The identity of the cell line was tested using isoenzyrne analysis and DNA fingerprinting after restriction enzyme digestion Tests for the presence of bacteria, fungi mycoplasmas and viral contaminants (including retroviruses) in the MWCB and the production cell cultures were constantly negative


Seed lot of virus

For the production of the seed lot, the Pitman-Moore (PM) strain of fixed rabies virus is used. after adaptation to growth in WI-38 cells. The infectivity titre of the virus in the seed lot should be tested by intracerebra inoculation of mice and should be no less than 10"' LD,, (median lethal dose) per ml.

Preparation of cells

Each batch of vaccine is produced starting from a working cell bank, consisting of a single ampoule of the MWCB at the 137th passage level. The cells are expanded by serial subculture up to the 142nd passage level. The area of microcarrier available to the cells increases from about 0.6 m2 at the 137th passage level to 900 m* (3 g/litre) at the 142nd passage level. The culture mediurn consists of Eagle's minimum essential medium (EMEM, see Chapter 8, Annex 1 j, supplemented with 4410% fetal calf serum (FCS) and neomycin, polymyxin B and dihydrostreptomycin sulfate.

After each passage (equivalent to a population doubling level of 2-41, the cells are removed from the microcarrier by trypsin~zat~on and hoinogen~zed in fresh EMEM supplemented with 4-10% FCS The resulting cell suspension is then used as the inoculum for the next passage. At the 141st passage level, samples should be removed for testing for adventitious agents (see page 287).

Infection of cells

1 Inoculate freshly trypsinized cells at the 142nd passage level with the seed virus at an inpiit multiplicity of infection of about 1 LD,, per 1000 cells

2 Seed the cells in culture flasks containing microcarriers and add culture niedum (see above) lncubate at 37'C for about 3 days until a complete monolayer is formed Remove samples for testing for bacterial contaminants and adventitious agents

3 D~scard the ciilture medium and rinse the microcarrlers several t~mes with

EMEM without serum 4 Replace the culture med~um by EMEM supplemented with 0 3 % human

albumin 5 lncubate the flasks for a further 6 days at 37°C Remove samples for testing for

bacterial contaminants and adventitious agent5 6 Harvest the cult~ire supernataiits

Generally, 5-6 harvests can be collected over a 3-week period

PURIFIED VERO CELL VACCINE FOR HUMANS

Clarification of the virus

The harvested virus suspension is clarified by filtration through a 0.45-pm membrane filter.

Concentration of the virus

The virus suspenson is concentrated 10-25-fold by ullrafltration through membranes with a relative molecular inass cut-off of 10000.

Inactivation of the virus

The virus suspension is inactivated using P-propiolactone (see Chapter 20). Samples are removed for testing for the absence of ~ v e virus (see page 288). After inactivation, the suspension is again concentrated by ultr-afiitration and stored at - 40 "C.

Purification of the virus

When the above tests are completed satisfactorily the alquots of every individual harvest are thawed pooled and purified by sucrose-density centrifugation at 90000 g (B15 rotors) for 3 hours The purified fractions are diluted with phosphate- buffered saline (PBS) and clarified by filtration through a 0 45-pm membrane The samples are slored at - 40 C

Production of the bulk vaccine

An al~quot of the bulk virus suspension is thawed and diluted with EMEM supplemented with 5% human serum albumin to a potency of 2 5 IU per ml, and then filtered through a 045-pm membrane

Production of the final vaccine

The bulk vaccine is distributed in I-m1 amounts ~ n t o vials, freeze-dried and sealed.

Control tests


Absence of extraneous agents At leasi 1096 of the cell suspension at the 141st passage level should be used to prepare control cultures. These cultures should be processed in the same way as the production cell cultures, but not inoculated with the virus. They should be observed for at least 2 weeks (until the firial harvest of virus is completed) for evidence of ariy cytopathological changes.

At the end of the observation per~od, the control cell cultures should be examined I f thls examination shows evidence of the presence of any extraneous agent, the virus grown in the rioculated cultures should not be used for vaccine production. The cell-culture supernatants should be pooled and tested for


infectivity in fresh Vero cells. A sample should be removed for haemadsorption tests using guinea-pig erythrocytes.


The final vaccine should meet the requirements for human vaccines prepared in continuous cell lines published by WHO (10 12 13) Each final lot of vaccine shouid be tested for purity safety residual cellular DNA and potency

Tests for residual DNA Acceptable litnits of cellular DNA per dose of the final vaccine should be determined by the nalional control authority, taking into account the effect of the inactivation procedure on the biological activity of DNA as well as previous experience with DNA levels in rabies and other vaccnes produced in various systems (12, 13). The quantity of residual cellular DNA is generally expressed In picograins (pg) per dose.

A recent collaborative study to examine the reliability of assays for the measurement of residual DNA in biolog~cal products derived from continuous cell lines revealed both a nrarked degree of inaccuracy and large differences in the estimates made by different laboratories (14) . Accordingly care is required in interpreting data obtained using hybridization techniques (15).

Potency test The potency of each final lot of PVRV is determined by the NIH test (see Chapter 37) I t should be no less than 2 5 IU per dose The potency of the vaccine has been shown to be rnaii-itaned tor at least 36 months at 4-C 30 months at 37 "C, and at least 12 months at 45 'C

Expiry date

The vaccine may be used up to 36 months from the date of release from storage. It should be kept betweeri 2 ° C and 8°C after release.

References

1. Requirements for poliomyelitis vaccine (inactivated). WHO Expert Committee on Biological Standardization. Thirty-second report. Geneva, World Health Orgarization. 1982 (WHO Technical Report Series. No 673). Annex 2

2. Van Wezel AL. Growth of cell strains and primary cells on microcarriers in homogeneous culture. Nature, 1967, 216: 64-65.

3 Montagnon BJ Fanget B, Vincent-Falquet JC Industrial scale production of inactivated polio virus vaccine prepared by culture of Vero cells on microcarrier Reviews of i n fec t~ow diseases 1984 6 (Suppl 2) 341-344

4 Fournier P et al A new vaccine produced from rabies virus cultivated on Vero cells In Vodop~ja l et a1 eds Improvements in rabiespost-exposure treatment Zagreb Zagreb Institute of Public Health 1985 115-121

PURIFIED VERO CELL VACCINE FOR HUMANS

5. Montagnon BJ. Fournier P Vincent-Falquet JC. Un nouveau vaccin antirabique a usage humain, rapport prelminare. [A new rabies vaccine for human use preliminary report.] In Kuwert E et a l eds Rabies in the tropics. Berlin, Springer-Verlag, 1985. 138-143

6. Roumiantzeff M et al. Rabies vaccine produced in cell culture production, control and cliriical results, In: Kurstak E et al., eds. Applied virology. New York, Academic Press. 1984. 241 -496.

7 Yasumura Y, Kawakita Y. [Studies on SV-40 in tissue culture-preliminary step for caricer research irr viiro.] Nitrun rinsho, 1963, 21. 1201 1215.

8 Montagnon BJ. Fanget B N c o a s AJ The large-scale cultivation of Vero cells in microcarrier culture for virus vaccine productiori Preliminary results for killed poliovirus vaccine Developments in biologicalstandardization, 1981, 47 55 64

9 Montagnon BJ Polio and rabies vaccines produced in continuous cell lines a reality for Vero cell line, Developments in biological standardization, 1989, 70: 27-47.

10 Requirements for continuous cell lines used for biologicals production WHO Expert Comm~ttee on Bio/og!ca/ Standardization Th~rty-sixth report Geneva World Health Organization 1987 (WHO Technical Report Series, No 745) Annex 3

11. Van Steens G Van Wezel AL. Use of ATG-treated newborn rat for in vivo tumorigenicity testing of cell substrates. Developnlents in biological standardization, 1982, 50: 37-46.

12. Req~iirements for rabies vaccine (inactivated) for human use produced in continuous cell lines. WHO Expert Committee on Biological Standardization. Thirty-seventh iaport. Geneva, World Health Organization. 1987 (WHO Tech- nical Report Series, No. 760) Annex 9.

13 Requirements for rabies vaccine (inactivaied) for human use produced in continuous cell lines (amendment 1992) WHO Expert Committee on Bio/ogica/ Standardization Forty-third report Geneva World Health Organization 1994 (WHO Technical Report Series No 840), Annex 5

14. WHO Expert Committee on Biological Standardization. Forty-first report. Gerieva, World Health Organizat~on, 1991 (WHO Technical Report Series, No. 814).

15. Acceptabil!ty of cell substrales for production of biologicals. Report of a WHO Stlidy Group. Geneva, World Health Organization, 1987 (WHO Technical Report Series, No. 747).

CHAPTER 27

Purified chick-embryo cell vaccine for humans R. Barth' & V. Franke2

This vaccine is prepared in primary chick embryo cells derived from specific pathogen-free (SPF) eggs, It is a free7e-dried preparation consisting of purified and concentrated rabies virus antigen iiiactivated with 8-propiolactone.

History

Early tissue-culture studies revealed that the low egg passage (LEP) Flury strain of rabies virus showed favourable characteristics as a vaccine strain (1) On the bass of these results an inactivated vacclne for veterinary use was developed in 1973 with the Flury LEP strain propagated in a primary chick embryo cell system The first steps towards the production of a tissue-culture vaccine for human use were begun with the concentration and purification of rab~es vlrus by ultracentrifugation in a sucrose-density gradient (2) The antibody-binding test was modified to permit quantitative dete~mination of Inactivated rabies virus antigen (3) During the 1980s avariety of other laboratory tests were carried out on p~ lo t batches of the purified chick-embryo cell (PCEC) vaccine (4 5) and clinical trials were ~nltiated in humans (6 7) The efficacy of the vaccine was also extensively tested under field conditions (8-13)


Seed lot of virus

For the production of the master seed lot, the Flury-LEP strain of fixed rabies virus is used, after adaptation to growth in primary SPF chick embryo cells.

The origin of this virus is shown below.

59th passage level in primary chick embryo cells (obtained from ATCC)3

l t

primary hamster kidney cells (90 passages)

I t

human diploid ceii strain Wi-38 (12 passages) l

+ primary chick embryo cells (25 passages) (LEP-C25)

'Former Head, Rabies Vaccinc Development and Product~on, B c l ~ r ~ n g lnsttute, Marburg, Germany. 'Head, Veterinary Vaccne Product~on. Behring lnsttute, Marburg. Germany. 3Arner~cari Type Culture Coilectlon

PURIFIED CHICK-EMBRYO CELL VACCINE

The master seed lot (LEP-C25) represents the 25th passage level of the virus in primary SPF chick embryo cells. The working seed lot (LEP-C26) represents the 26th passage level of the virus.

Both the master seed lot and the working seed lot have been shown to be free from foreign viruses, mycoplasms, bacteria and fungi by procedures recommended for rabies vaccines for human use (14, 15). The master seed lot and the working seed lot are stored at - 80 "C and - 190°C respectively.

Cell cultures

The virus is cultivated in primary SPF chick embryo cells, which are propagated and controlled according to the recommendations published by WHO (14-16) and the regulations of national authorities.

Fertile hens' eggs incubated for 7-9 days at 36.5"C are used. The eggs are obtained from an SPF flock and transferred to the laboratory 1 day before inoculation.

1. On the 7th day of incubation, candle the eggs and mark the edge of the air sac. Discard any eggs that contain dead or underdeveloped embryos.

2. Disinfect the shell with 70?& ethanol and briefly flame it. 3. Place the eggs in a tray in a biological safety cabinet (class I ) , and remove the

live embryos by one of the methods described in Chapter 23 (page 263). 4. Decapitate the embryos (discarding the heads) and place them in a chilled

container, 5. Weigh the container with the collected embryos. Sufficient chick embryos

should be collected to provide cell culture for one batch of vaccine. 6. Add 0.25% trypsn solution to the embryos. Wash the resulting cell suspension

with PBS and resuspend in Eagle's minimum essential medium (see Chapter 8, Annex 1) supplemented with 0 3% human albumin to a final concentration of 0.8-1.2 X 10" cells per ml. Reserve at least 5% of the cell suspension to prepare control cultiires (see page 293).

Infection of cells and harvest of the virus

With the exception of step 3 the following steps should be carried out in a biological safety cabinet (class 11).

1 Inoculate freshly trypsinized cells in suspension with the working stock virus at an input multiplicity of ~rlfection predetermined to infect most cells after 3-4 days

2 Seed the cells into flat-bottomed culture flasks and Leghton tubes and add culture medium

3 Incubate for 3-4 days at 34-36 "C until a complete rabies infected monolayer can be observed by fluorescent microscopy on the Leghton tubes

4 Harvest the culture supernatant containing the virus Remove samples for testing for bacterial contaminants and adventitious agents

5 Add fresh culture medium and incubate the flasks for a further 3-4 days at 34-36'C Harvest the ci~l ture supernatant Remove samples for testing for bacterial contaminants and adventitious agents



The harvested virus should be filtered to remove cell debris


The virus is inactivated with P-propiolactone (see Chapter 20, page 235). After inactivation, the virus suspension is kept between 2 and 6 "C.


The virus suspension is concentrated and purified by centrifugation at 90000 g in a sucrose-dens~ty gradient (0-60%). A band is formed at about 36% sucrose. This band, which contains the virus, is collected in fractions at the end of the centrifuge run. The virus is stored at - 70°C until all the control tests are completed (see below). The potency of the concentrated and purified virus suspension should be between 150 and 500 IU/ml.

Preparation of the final vaccine

The purified and concentrated virus suspension is mixed with a stabilizer solution containing degraded gelatin in TEN (trometamol-edetic acid1-sodium chloride) buffer solution and distributed in 1 .O-ml amounts into vials, which are freeze-dried under vacuum and sealed. The final vaccine should have a minimum potency, as determined by the NIH test (see Chapter37) and the modified antibody-binding test (see Chapter 43), of 2.5 IU per dose.

Control tests

In-process controls

Representative samples are removed at each stage of the production process ("in- process controls") for testing according to The 1nternat;onal Pharmacopoeia (17) and the requirements for rabies vaccines for human use published by WHO (14, 15).

Potency tests

Studies of tests for evaluating the potency of fables vaccines have shown that the NIH test measures only a fraction of the potency of PCEC vaccine compared with other potency tests. Since Challenge Virus Standard (CVS), a derivative of the Pasteur strain, is recommended as the challenge strain for the NIH test, vaccines prepared with viruses other than the Pasteur strain (such as the PCEC vaccine) may appear to be of lower potency than expected when tested in mice challenged intracerebrally with CVS. However, this finding does not appear to influence the

'Also known as elhylened~amine tetraacetate ol EDTA

292


efficacy of PCEC vaccine in the field Various comparative studies on PCEC vaccine have revealed that the ratio of potency values as determined by the NIH test and other tests is 1 2 2 (18-20)

Stability test

The stability of the PCEC vaccine was evaluated by the N H test using different lots

of vaccine which were stored at dfferent ternperalures The results are shown in Table 27 1

The shelf-life of the vaccine is controlled by storing samples of the final vaccine a t d~fferent lernperdlures arid determining t h e potency by the NIH test and t h e modified antibody-binding test conducted in parallel


At least 5% of the working cell suspension is used to prepare control cultures. These cultures are processed in the same way as the production cell cultures, but not inoculated with the virus. They are tested as described in Chapter 25.

Administration of the vaccine

The vaccine dose is 1 0 m1 administered intran?usculatly into the deltoid region of the arm or in the case of small children into the anterolateral aspect of the thigh


One dose of vaccine is administered intramuscularly on days 0, 7 and 28

Post-exposure treatment

One dose of vaccine is given intramuscularly on days 0 , 3 7, 14 and 30.

Expiry date

The vaccine may be used up to 3 years from the date of release. It should be kept between 2 and 8'C.

Table 27.1 Potency of the PCEC vaccine after storage at different ternperalures

Period of Number of storage vaccine Change in

Temperature ( "C) (in months) lots tested potency (X)

2-8 57 20 12 31 (with 75% relative

humidity; tropcal conditions) 24

37 6 55 24


Laboratory tests

Numerous laboratory tests were carried out on the vaccine before clinical trials in humans were begun They included

1 Tests for freedom from residual chick-embryo cell protein -agar gel precipitation -~nirnunoprecipitation and SDS-PAGE, -studies in guinea-pigs (immunizat!on w ~ t h the vaccine, followed by intra-

verious administration of chick-embryo cell protein). 2. lnriocuty tests in monkeys arid dogs. 3. Tests for the absence of pyrogens in rabbits. 4. Stiidies on the iriduction of neutraliring antibodies in sera from mice and

monkeys immunized with the vaccine. 5 Post-exposure efficacy sti id~es wilt1 the vaccine in mice and guinea-pigs

compared with HDC vaccine 6 Potency tests ~ncluding the NIH test and in vitro tests such as the antibody-

biiiding test 7. Stability tests following storage at high temperatures

References

1. Barth R, Jaeger 0. Untersuchiingen mit eingen Tollwutvirusstammen in verschiederien Gewebekultursystemen. [Studies with several rabies virus strains in various tissue-culture systems.] Zentralblatl fur Velerinarmedizin, Reihe B, 1970, 17: 363-380.

2. Hilfei?haus J et al. Large-scale purification of animal vlruses in the RK-model zonal ultracenlrfuge. Journai of bioiogical standardization, 1976. 4: 263-271

3. Barth R et al. The antibody-binding test: a useful method for quantitative determination of inactivated rabies antigen Journal of bioiogica! standardi/ati'on, 1981 . 9: 81 -89.

4 Barth R et al A new inactivated tissue-culture rabies vaccine for use in man Evaluation of PCEC vaccine by laboratory test Journal of biological standardization 1984 12 29-46

5. Barth R et al. Purified chick-embryo cell (PCEC) rabies vaccine for human use-laboratory data. In: Kuwert E et al.. eds. Rabies ~n the tropics. Berlin, Springer-Verlag. 1985: 11 7-124.

6. Bijok U et al. Clinical trials in healthy volunteers with the new purified chick- embryo cell rabies vaccine for man. In: Kuwert E et al.. eds. Rabies in the tropics. Berlin. Springer-Verlag. 1985. 125-132.

7. Bijok U. Purified chick-embryo cell (PCEC) rabies vaccine: a review of the clinical development 1982-1984 In: Vodopija I et a l , eds. Improvements in

rab~es post-exposure lreatnlent Zagreb, Zagreb lnsttute of Public Health, 1985' 103-~111.


8. Vodopija l, Smerdel S, Bijok U. PCEC rabies vaccine and HRlG in post- exposure protection against rabies. In. Kuwert E et al.. e d ~ . Rabies iii the tropics. Berlin, Sprnger-Verlag, 1985: 133-137.

9. Petrovik M, Petrovik M, DamjanovC J. Laboratory examination of PCEC rabies vaccine produced by the Behring Institute. Marburg, FRG, In: Vodopila 1 et al., eds, /mprovemenis in rabies posl-exposure treatment. Zagreb. Zagreb Institute of Public Health. 1985: 63-69,

10. Sehgal S Report of the trials of PCEC (purified chick-embryo cell) rabies vaccine in India. In: Vodopila l et al., eds. Improvements in rabies aosl-

exposure treatme~lt. Zagreb. Zagreb lnstitute of Public Health, 1985: 71-75.

11, LjubiEiC M et al. Experience w ~ t h PCEC rabies vaccine in healthy adults. In: Vodopila I et al., eds, Iinprovements in rabies post-exposure treatmeni Zag reb. Zagreb lnstitute of Public Health, 1985: 77 8 4 .

12. Wasi C et al. Immiinogenicity and reactogenicity of the new tissue culture rabies vacclne for human use (purified chick-embryo cell culture). In: Vodopija I et al.. eds improven~ents in rabies post-exposiire treatment. Zagreb Zagreb Institute of Public Health, 1985: 85-94.

13 Ljubieik M et al. Efficacy of PCEC vaccine in post-exposure rabies prophylaxis. In: Vodopija l et al.. eds. Improvements in rabies post-exposure treatment. Zagreb Zagreb Institute of Public Health. 1985: 95-101.

14. Requirements for rabies vaccine for human use. WHO Expert Committee on BiologicalStar?dardization. Thirty-firstreport. Geneva. World Health Organiza- tion, 1981 (WHO Technical Report Series, No 658), Annex 2.

15. Requternents for rabies vaccine for human use (amendment 1992) WHO Expert Committee on Biological Standardization. Foriy-third report. Geneva, World Health Organizatiori 1994 (WHO Technical Report Series. No. 840). Annex 4.

16. Acceptability of cell substrates for productiori of biologicals. Report of a WHO Study Group. Geneva, World Health Organization. 1987 (WHO Technical Report Series, No. 747).

17. The lnternat~or~ai P/~arn7azopoeic?. 3rd ed. Geneva, World Health Organi7a- tion, 1979 (Volume 1: general inethods of analysis); 1981 (Volume 2. quality specifications), 1988 (Voiurne 3: quality specifications); 1994 (Volume 4: tests, methods and general requirenients: quality specifications for pharmaceutical substances. exc,pents, and dosage forms).

18 Barth R Diderrich G Weinmanrl E NIH test a piobematic method for testing potency of inactivated rabies vaccine Vaccine 1988 6 369 377


19. Baith R et al. Purlfled chick-embryo cell (PCEC) rabies vaccine. its potency performance in different test systems and in humans. Vaccine, 1990. 8: 4148.

20. Barth R. The NIH test and its problems. In: Thraenhart 0 et al.. eds. Progress in rabies control. (Proceedings of the Second Internationa! IMV! Essen/WHO Symposium on "New Deveiopments in Rabies Control", Essen, 5 7 July 1988 and Report of the WHO Consultation on Rabies, Esserl, 8 July 1988.) Royal Tunbridge Wells, Wells Medical, 1988 281-287.

CHAPTER 28

Fetal rhesus monkey lung diploid cell vaccine for humans R Barlh' & V Frar?ke2

This vaccine is prepared in a diploid cell line (fetal rhesus monkey lung cells, line 2 (FRhL-2)). It is a liquid preparation corislsting of rabies virus inact~vated wlih ,L- propiolactone and adsorbed to alurnniurn phosphate.

History

The fetal rhesus monkey diploid cell (FRhMDC) vaccine was developed in the 1970s by the Biologic Products Division of the Michigan Department of Public Health in the USA. Trials in human volunteers were begun in 1979 ( 1 ) and the vaccine was licensed for use in the USA by the Food and Drug Administration (FDA) in March 1988 for both pre- and post-exposure treatment.


Cell cultures

The FRhL 2 cell line is used, which is obtained from the American Type Culture Collection (ATCC) The cells are used for the production of the vaccine at the thirty-first passage level ( I )

Seed lot of virus

For the production of the seed lot the Kissling strain of fixed rabies virus 1s used after adaptation to primary hamster kidney cells The Michigan Department of Public Health obtained the Kissing strain at the 122nd passage level After adaptallon to primary hamster kidney cells (4 passages) the strain was transferred to FRhL-2 cells (13 passages) (2)


1 . Prepare a cell suspension from rnonolayer cultures at the thirty-first passage level using 0.025% trypsin and 0.01 94 edetc acid3 to detach the rnonolayers.

2. Inoculate the cell suspension with the seed virus at an input multplic~ty of infection of 1 LD,, per 10 20 cells.

3. Seed the cells in 150-m1 culture flasks at a concentration of 8 10 X 106 cells per flask and add 50 ml of culture medium. The culture medium consists of Eagle's

' Former Heas Rabies Vaccine CIeveio~n:el: and Pfodbctior. Behriqg Inst i tu te Marburg Gerrnany Head, Veterinary Vaccine Produc?ion Eehrng ins!itu!e Marburg. Gemany . Also known as ethvienediamine tetraacetate or EDTA.


minimum essential medium (EMEM, see Chapter 8 Annex 1) supplemented with 10% fetal bovine serum and 0 1% sodium bicarbonate (NaHCO,)

4 Incubate at 37'C for 24 hours until a complete monolayer is formed Remove samples for testing for bacterial contaminants and foreign viruses

5 Discard the culture niedium and wash the cells three times with EMEM supplemented with 0 1 % NaHCO,

6 Add 50 mi of maintenance medium consisting of EMEM supplemented with 0 22% NaHCO,

7 Incubate the culture flasks for 12 13 days at 33-35'C Remove samples for testing for bacterial contaminants and foreign viruses

8 Harvest the culture suDernatants


The harvested virus is clarified by filtration through a 0.45-pm membrane. Virus samples are pooled to constitute the bulk virus suspension.


The bulk virus suspension is inactivated using P-propiolactone (see Chapter 20).


1. Add aluminium phosphate to the bulk virus suspension to give a final concentration of 0.1 mg per litre.

2. Stir the mixture for 4 hours at 4'C and leave it to stand overnight. 3. Wash the precipitate three times with EMEM supplemented with 0.1 O h NaHCO,

(2).

Preparation of the bulk vaccine

The bulk vaccine 1s prepared by resuspending the washed aluminium precipitate in about one-seventeenth of the original volume of the inactivated bulk virus suspension. The resuspension medium contains 1 % human serum albumin, 1 % lactose and 1 : l 0000 thiornersal.'

Control tests

In-process and final controls

Tests for in-process control and tests on the final vaccine are carried out according to the requrrements for human rabies vaccines published by WHO (3,4).

Potency test

The potency of each final lot of vaccine should be determined by the NIH test (see Chapter37) and should be no less than 2.5 1U per dose In 1985 nine lots of the final vaccine were tested by the NIH test; the mean potency was 3.6 IU per dose ( l , 2).

' Also known as thiorr~ersalate and rnercurotli~olate

298

FETAL RHESUS MONKEY LUNG DIPLOID CELL VACCINE

Expiry date

The vaccine may be used up to 1 year f r o ~ n the date of release, It should be kept

between 2 and 8°C after release. However. the potency of the vaccine appears to be maintained for up to 6 months at room temperature and up to 2 months at 37 "C.


Pre-exposure vaccination

One dose (1.0 ml) of the vaccine is administered on days 0, 7 and 28. The vaccine should be given into the deltoid area of the arm or, in the case of small children, into the anterolateral aspect of the thigh muscle.

Post-exposure vaccination

One dose of the vaccine is given intramuscularly on days0,3,7,14 and 30. In cases of severe (category Ill) exposure, human rabies imrnunoglobulin (HRIG) (20 lU/kg of body weight) should also be given (5). lntradermal application is not recommended.

Persons who have previously received full pre- or post-exposure treatment with vaccine and who have an acceptable rabies virus-neutralizing antibody titre should be given only two booster doses intramuscularly, on days 0 and 3, but no rabies imrnunoglobulin.

Laboratory tests

Antibody response in guinea-pigs

The antibody response in guinea-pigs immunized with FRhMDC vaccine was found to be similar to that induced by HDC vaccrne and PHKC vaccine. The antibodies were of the IgG class (2).

Post-exposure treatment trials in guinea-pigs

Three groups of guinea-pigs were inoculated with 200 LD,, of rabies street virus on day 0. On day 1, the three groups were given antiserum (20 or 40 IU), vaccine (0.025 or 0.05 ml), or a combination of the two, followed by one dose of vaccine on days 3 , 7 . 14 and 28. The experiment showed that post-exposure immunization with FRhMDC vaccine provides protection against rabies street virus and that protection is primarily provided by the vaccine (2).

References

1. Berlin BS et al. Rhesus diploid rabies vaccine (adsorbed). a new rabies vaccine. Results of initial clinical studies of pre-exposure vaccination. Journal of the American Medical Association, 1982, 247: 1726-1 728.


2. Burgoyne GH et al. Rhesus drploid rabres vaccine (adsorbed): a new rabies vaccine uslng FRhL-2 cells. Jourr~ai of infectious diseases, 1985, 152: 204-210.

3 WHO Expert Committee on Bioloyicai Standardization Thirty-first report Gen- eva, World Health Organzat~on, 1984 (WHO Technical Report Serres, No 658) Annex 2

4. WHO Expert Committee on Biological Standardization. Forty-third report. Geneva, World Health Organizat~on, 1994 (WHO Technical Report Series, No. 840), Annex 4.

5 . WHO Expert Comm~ttee on Rabies. E~ghth repart Geneva, World Health Organization. 1992 (WHO Technical Report Seres. No. 824).

CHAPTER 29

Dog kidney cell vaccine for humans R. Barth. ' V Franke2 & G. van Steenis3

This vaccine is prepared in dog kidney cell cultures on microcarriers (1-3) It is a freeze-dr~ed preparaton consst~ng of flxed rabies virus inactivated with p- prop~olactone and adsorbed to a lu~nin~uin phosphate atter reconstitution


Cell cultures

The kidneys are obtained from 6 9-week-old beagles which are bred in a closed colony The cells are trypsinized according to the perfuslon technique of Kammer (2) wlth slight modifications At the National Institute of Public Health and Environmental Protection in Blttioven Netherlands vaccine production is based on a cell bank of primary dog kdney cells which are trypsinized and preserved n liquid nitrogen at the end of the first passage The cells are used for vaccine product~on at the second passage level

Seed lot of virus

A working seed lot is prepared from the Pitman-Moore (PM) stra~n of fxed rabes virus adapted to hunian diploid cells


1. Seed the cells in a 40-litre culture vessel containing microcarriers at a concentraton of 150 x 103 cells per m1 and add culture medium. The culture medium conssts of Eagle's minimum essential medium (EMEM: see Chapter 8, Annex 1) supplemented with 8 % bovine seruni and 0.25'/0 lactabumin hydrolysate.

2, Incubate the vessel at 37°C for 6-8 days until the concentration reaches approximately 106 cells per nil.

3. Discard the medium and inoculate the cells with the working seed lot, Incubate at 35°C for 6 days in medium 199 (see Annex) supplemented with 19'0 calf serum.

4. Discard the medium and wash the cells with rnedium 199 without serum.

' Former Head, Rab~es Vaccine Development and Productiori Betiring institute, Marbi i rg Germany Head, Veternary Vaccine Produc!ion Behring institute Marburg, Germany Head, Quality and Regu!atory Affairs National institute o! Public Heath and Enviroi~mental Protection BiltOoven, Netberiands.


5. Replace the culture medium by virus propagation medium (medium 199 without serum) and incubate at 35°C.

6. Harvest the culture supernatants at intervals of 4-5 days. Remove samples for testing for potency (by intracerebral inoculation of adult mice) and for bacterial contaminants and adventitious agents. The infectious titre of the harvested supernatant should be between 105.' and IO'.~ LD50 per ml.


The virus harvests are pooled (about 250 litres) and clarif~ed by filtration through 1.2-pm membranes. The pooled harvests constitute the bulk virus suspension.


The bulk virus suspension is concentrated by ultrafiltration through membranes with a relative molecular mass cut-off of l00000 to give a final volume of 1 litre. The concentrated virus suspension is purified by gel filtration on a 20-litre column of Sepharose 6B (beaded agarose) in phosphate-buffered saline (PBS) w ~ t h 55 mg of polysorbate 80 per litre. The first fraction of 2.5 litres is collected.


The purified and concentrated virus is inactivated with p-propiolactone, using the procedure described in Chapter 20 (page 235).

Preparation of the final bulk vaccine

After inactivation with p-propiolactone, the virus suspension is diluted with PBS to obtain the final bulk vaccine. Lactose (5%) is added as stabilizer.

Preparation of the final vaccine

The final bulk vaccine is distributed into vials in 1.0-ml amounts and freeze-dried.

Control tests

Tests for in-process and final controls are carried out according to the requirements for rabies vaccines for human use published by WHO (4. 5).


Potency test The potency of each final lot of vaccine should be no less than 2.5 IU per dose, as determined by the NIH test (see Chapter 37).

Test for protein content

The protein content of the f~na l vaccine should be between 9 and 27 pg per dose

DOG KIDNEY CELL VACCINE FOR HUMANS


The freeze-dried vaccine is reconstituted before use by adding 1.0 ml of distilled water containing 2 mg per m1 of al i rrnini~~m phosphate (AIPO,) as adjuvant.


One dose (1.0 ml) of the vaccine is administered intraniuscularly on days 0. 7 and 21. Alternatively. the initial dose of vaccine may be given on day 0, followed by a booster injection after 1 and 7 months. The injection should be given into the deltoid region or, in the case of small children n t o the anterolateral aspect of the

thigh muscle.


One dose of the vaccine is given intramuscularly on days 0, 3, 7, 14 30 and 90

References

1. Van Wezel A L van Steenis G. Produc to~ i of an inactivated rabies vaccine in primary dog kidney cells. Developments in biologicalstandardization, 1978, 40: 69-75.

2 Kammer H Cell d~spersal methods for increasing yield from animal tissue Applied microbiology 1969 17 524

3. Van Steenis G et al. Dog kidney cell rabies vaccine: some aspects of its control and efficacy in man. In: Kuwert EK. Wiktor TJ, Koprowsk H, eds. Cell-culture rabies vaccines and liieir protect;ve effect in man (Proceedings of WHO Consu/tations. Essen, 5-7 March 1980.) Geneva, International Green Cross, 1981 : 78-86

4. WHO Expert Committee on Biological Standardization Thirty-first reporl. Geneva, World Health Organization, 1981 (WHO Technical Report Series. No. 658), Annex 2.

5 . WHO Expert Committee on Biological Standardization. Forty-third report. Geneva, World Health Orga~iization, 1994 (WHO Technical Report Series. No. 840), Annex 4.

Annex Preparation of medium 199

Component

Organic sails Calcium chloride, dhydrate (CaCl,, 2H,O) Ferric nitrate, nonahydrate (Fe(NO3),.9H,O) Magnesium sulfate, heptahydrate (MgSO;7H20)


Potassium chloride (KCI) Sodium chloride (NaCI) Sodium bicarbonate (NatiCO,) Sodium phosphate. rnonobasic, dihydrate (NaH2PO,.2H,O)

Other components Adenine sulfate Adenosine triphosphate (ATP) 5'-Adenylic acid Cholesterol D-2-Deoxyri bose Glucose Glutathione (reduced) Guanne Hypoxanthne Phenolsulfonphthalein (phenol red) Polysorbate 80 D-Ri bose Sodium acetate, trihydrate Thyrnine Uracil Xanthine

Amino acids DL-Alanine L-Arginne hydrochloride DL-Aspartic acid L-Cysteine hydrochloride L-Cystine DL-Glutamic acid L-Glutamine Glycine L-tiistidine monohydrochloride, rnonohydrate 4-Hydroxy-L-proline DL-isoleucine DL-Leucine L-Lysine rnonohydrochloride DL-Methionne DL-Phenylaanine L-Proline OL-Serine DL-Threonine, hernihydrate DL-Tryptophan L-Tyrosine DL-Valine

Vitamins p-Aminobenzoic acid Ascorbic acid (+)-Biotin

Calciferol Calc~um D-pantothenate Choine chloride D~sodium tocopherol phosphate Folic acid i-lnos~tol Menadione Niacinamide Nicotinic acid Pyridoxal hydrochlorde Pyridoxine hydrochloride

Riboflavin Thiamine hydrochloride V ~ t a m ~ n A (acetate)

DOG KIDNEY CELL VACCINE FOR HUMANS

CHAPTER 30

Primary hamster kidney cell vaccine for humans R, Barth,' V. ~ranke ' & F. T. Li'n3

This vaccine is a freeze-dried or liquid preparation consisting of fixed rabies virus inactivated with formaldehyde (formalin). The virus is propagated in primary kidney cells of Syrian hamsters.


Seed lot of virus

The Belling strain of fixed rabies virus is used after adaptation to growth in primary Syrian hamster kidney cells (PHKC) ( I ) The strain was originally isolated in 1931 from the b r a n of a dog that died flom rabies in Belling China It was fixed by successive passages in rabbit brain and was used for the production of a Semple- type nerve-tissue vaccine until 1980 when the PHKC vaccine was iritroduced

The adaptation of the Belling strain to propagate in PHK cells comprised two stages The first stage involved five passages in PHK cells with 15-30 days of n c u b a t ~ o n at 37°C The infected nionolayer was trypsinized mixed with fresh PHK cells and seeded into new containers After the fifth passage the virus was propagated in successive passages by transfer of the virus-containing supernatant During passages 1 1 5 5 the virus titre gradually increased from 1 0 to 4 5 log LD,, per 0 03 ml in mice inoculated intracerebrally and the period to reach the peak titre was reduced from 15-30 days to 5-7 days The 55th passage level was inociilated into giiinea-pig brain and back into PHKC monolayers After three alternative passage cycles the virus t~tre in the tissue-culture supernatant had increased tenfold and was used for the production of the seed material ( I )


1 , Infect freshly trypsnzed PHK cells in suspension at an input mult!piicity of infection predeterniined to infect inost cells after 2-3 days.

2 Seed the cells n culture flasks and add culture medium (Eagles minimum essential medium (EMEM see Chapter 8 Annex l ) supplemented with 8-10%

bovine serum). 3 Inciibate the flasks at 37'C for 2-3 days until a coniplete monolayer is formed. 4. Remove the culture medium. 5. Wash the cells three times with Hanks' balanced salt solution (see Annex).

' Former Heaci Haoes Vaccine Deveiaprricr~; ririd Proc~~c ! ion Behrng Insttute, M a r b ~ ~ r g Gemany Head, Vefernary Vacclne P ~ o d u c t i o r Beiiri!ig lns;i:l~!e, IVldrburg, Gerl l i i i r ly Former Head Departr:ent of Viral Vacr~nes Wuhan ins?~'.~+e for Bo iog ica p -odbc ts Mnis!ry of Pubic Hcalih Wuct~ang, C h n a

PHKC VACCINE

6. Add virus propagation medium (medium 199; see Chapter 29. Annex). 7. Incubate at 37 'C for 6-7 days. 8. Harvest the culture supernatant.

The infectivity titre of the virus should be 106.0-107.0 LDS0 per m1


See Chapter 25 .


For the production of the lyophilized vaccine the bulk virus suspension is concentrated tenfold by ultrafiltration The virus suspension is not concentrated for the liquid vaccine


Formaldehyde (formalin) is added to the bulk virus suspension to a final concentration of 1 :5000 vollvol. The suspension is heated at 37°C for 48-72 hours.

Preparation of bulk vaccine

In the case of the liquid vaccine, the inactivated virus is adsorbed to aluminium hydroxide (final concentration 0.5 nigiml). Thiomersal' (0 01 %) is added as a preservative, plus human serum albumin (3 mg/ml). In the case of the freeze-dried vaccine, human serum albuniin (3 mglml), sucrose (50 mglml) and gelatin (10 mglml) are added to the inactivated virus before freeze-drying.

Preparation of final vaccine

The bulk vaccine is distributed in 1.0-m1 amounts into vials and either sealed or freeze-dried and sealed.

Control tests

In-process controls

In-process controls are carried out according to the requirements for biological products of the Ministry of Health. China. Before inactivation, the virus bulk suspension is tested for sterility and for freedom from foreign viruses (1 ) .


Potency test The potency of each final lot of vaccine is determined by the NIH test (see Chapter

' Also known as tt l~merosal and ?lerc~ro'hiolate

307


37). It should be no less than 2.5 \U per m\. A study demonstrated that the potency of five lots of liquid PHKC vaccine was between 105.0 and 105,5 LD50 per m1 by the Habel test. In another study, the potency of three lots of the ireeze-dried vaccine was shown to be within the range 105.7-105.9 per ml by the Habel test and 2.7-40 IU per ml by the NIH test. Until 1986, the Habel test was used for determining the potency of the final vaccine.

Test for stab~iity Six lots of freeze-dried PHKC vacciiie were incubated at 37<C for 2 and 4 months After storage or-ily one lot of vaccine showed a significant change in potency (2)

Test for protein content The protein content of the final vaccine should be no less than 3 rng per rnl.

Tests for freedom from anaphylactic and encephalomyelitic substances Several final lots of vaccine are tested for freedom from anaphylactic and encephalomyelitic substances in guinea-pigs ( l ) .

Expiry date

The liquid vaccine may be used up to 12 months from the date of release, while the freeze-dried vaccine may be used up to 2 years after release The vaccines should be kept at 2-6°C after release


The vaccine dose is 1 0 ml admin~stered intramuscularly.


One dose is given on days 0, 7 and 14


One dose is given on days 0. 3. 7, 14, 30 and 90

References

1 , Lin FT et al. The primary hamster kidney cell rabies vaccine adaptation of viral strain, production of vaccine and pre- and postexposure treatment. Journal of

infectious d~seases, 1983, 147: 467-473.

2. Lin FT et al. Further study on the stability and efficacy of the primary hamster kidney cell rabies vaccine In Vodopija I et al., eds. Improvements in rab~es post- exposure treatment. Zagreb, Zagreb Institute of Public Health, 1985: 37-45.

Annex Hanks' balanced salt solution

Calcium chloride (CaCI,) Glucose Magnes~um sulfate, heptahydrate (MgS04.7H,0) Phenol red Potassium chloride (KCI) Potassium phosphate, moriobas~c (KH,PO,) Sodium chloride (NaCI) Sod~um phosphate, dibasic (Na,HPO,) Streptomycin sulfate Dstilled water to make

After filtration, the solution is ready for use

PHKC VACCINE

0 14 g 1.0 g 0.2 g 002 g 0.4 g 0.06 g 8.0 g

0.0475 g 0.2 g (156000 IU)

1000 rnl

CHAPTER 31

Vnukovo-32 primary hamster kidney cell vaccines for humans R. Barth,' ii Frankez & M. A. Se/imov3

These vaccines are a non-concentrated and a concentrated and purified cell- culture vaccine, prepared in primary kidney cells from Syrian hamsters and inactivated by irradiation with ultravioiet light ( 1-3).

Preparation of the vaccines

Seed lot of virus

The vaccines are prepared with the Vnukovo-32 strain of SAD virus, adapted to primary hamster kidney cells (PHKC). The origin of this virus is shown below (2).

street virus from a dog that died in Alabama in 1935 (SAD)

+ PHKC (35 alternating passages between mouse brain and PHKC,

followed by 10 serial passages in PHKC at 37'C)

1 PHKC (33-178 passages at 32°C)

The master seed lot and the working seed lot are prepared from virus between the 33rd and the 178th passage level in PHKC. The seed lots are stored at - 60 "C or freeze-dried.

Preparation of cell culture

Kidneys from young Syrian hamsters (weight 40-50 g) are removed and digested using a 0.25% trypsin solution. The cells are suspended in Hanks' balanced salt solution (pH 7.0-7.1) (see Chapter 30, Annex) containing 0.3% lactalbumin hvdrolysate ( IAH) and 10?/0 normai bovine serum (NBS) to a final concentraiion of

' Former Head Rabies Vacc~ne Deveiopme~~t and Production Behring lnstltute Marblirg Germany Head Veterinary Vaccine Production Behiing lristltute Marburg Germany

Former Head Rab~es Prophylax~s Laboratory, lnstttute of Pol~ornyelilis and Viral Encephal~tic Diseases Academy of Medical Sciences of the Russian Federation Moscow Russian Federation

VNUKOVO-32 PHKC VACCINE§

4 X 105 cells per m1 and seeded in 300-m1 volumes into 1.5-litre flat-bottomed bottles.

Infection of cells and propagation of virus

inoculation of cells with the working seed lot is done either on rnonolayers or in

suspension. The size of the inoculum should be sufficent to infect the majority of cells by the fourth day.

Virus harvest

The virus is harvested on the fourth day after inoculation. With an interval of another 2 days, 3-4 harvests are possible. The cell supernatant is collected into 5-litre bottles and frozen at - 20°C. Samples are removed from each bottle for testing for sterility arid for determination of the virus titre. The virus litre should be no less than 1 0 ~ . ~ LDS0 per ml, otherwise the sample must be rejected.

Preparation of the bulk virus suspension

When the above tests are completed satisfactorily, the virus samples are pooled to form the bulk virus suspension. About 15 litres of bulk virus suspension are required for the production of one lot of vaccine The bulk virus suspension is clarified by fiitration through 1.8-2.0-pm membranes under positive pressure.


The bulk virus suspension is concentrated by ultrafiltration through a 50-nm membrane. The resulting filtrate is then purified by gel chromatography (4)

inactivation of virus

The virus suspension is inactivated by irradiation with iiltraviolet light (see Chapter 21, Annex 1). The unit consists of a closed chamber with four ultraviolet lamps suspended in its upper part. A sta~nless steel disc rotates under these lamps. Pressure in a closed system drives the virus fluid onto the rotating disc, where it forms a thin layer of about 1 mm.

Preparation of final vaccines

After addition of 1 % gelatose and 7.5% sucrose solution the bulk vaccines are distributed into vials (in 3-ml volumes for the non-concentrated vaccine and in l-m1 volumes for the concentrated and purified vaccine), deep-frozen at between - 60°C and - &'C, and lyophilzed. The vials are filled with argon containing

not more than 0.008% nitrogen, 0.001 % oxygen and 0.001 % moisture. The vaccines are distributed in 5-dose packs with solvent ( 1 ) .


Control tests

Potency tests

The potency of the vaccines IS determined by the NIH test (see Chapter 37). It should be no iess than 0.5 IU per rnl' for the non-concentrated vaccine and no less than 2.5 U per ml for the concentrated and purifred vaccine.

Administration of the vaccines

The vaccines should be reconsttuted immedately before use, by injectlng dstilled water into the vial (3 ml for the non-concentrated vaccine and 1 m1 for the concentrated and purified vaccine). The former is administered by subcutaneous injecton into the abdomirial fat, according to the rnstructons of the manufacturer. The treatment schedule usng t l i s vaccne consists of many more Injections than are recommended by the WHO Expert Committee on Rabies (7 ) . Furthermore, the dosage varies according to the level of exposure, contrary to the recornmend- ations of the above Expert Comm~ttee. For these reasons, the schedule will not be described here The concentrated and purfied vaccine is given intramuscularly ~ n t o t h e deltoid region o i the arm. For cases of mild exposure (not involving the head or linibs), one dose (1.0 ml) should be glven on days 0,3. 7, 14, 30 and 90 ( l ) .

References

1. Selimov M et al. Nonconcentrated and concentrated tissue-culture rabies vacclnes from the Vnukovo-32 and Vnukovo-32-107 strains. [Abstract.] In: Kuwert EK, Wiktor TJ, Koprowsk~ H, eds. Ceii-culture rabies vaccines and their protective effect in man. (Proceedings o f WHO Consuitations. Essen, 5-7 March 1980) Geneva, International Green Cross, 1981: 64

2. Selhmov MA, Aksenova TA. Tissue-culture antirab~c vacclne tor human use. in: Regamey RH et al., eds. Proceedings of the Twelfth International Symposium organized by the permanent section of Immunological Standardization, Tailoire, 27-30 May 1965. Easel. Karger, 1966: 377-380

3. Tissue-culture rabies vacclne for h i ~ m a n use (Rabivak-Vnukovo-32). Moscow, Foreign Trade State Co., 1985: 1-61. (Unpublished document; available on request from Foreign Trade State Co., 21 New Arbat Street, Moscow 121 906, R u s s i a n Federation.)

4. Bresler SE et al. Adsorption chromatography of vlruses. Journal of chromatography, 1977, 130: 275-280.

5. WHO Expert Cornntittee on Bioiogicai Standardization. Thirty-seventh report. Geneva, World Health Organization, 1987 (WHO Technical Report Seres. No. 760).

' Editor's no:e t'le WHO Expert Co.nmit!ee on Bo iog ica S+andardiza?ion recommends that cell-culture vaccines for human use s iou ld have a ;rnlrnurn co!ency of 2 5 IU per oose (5 C)

VNUKOVO-32 PHKC VACCINES

6. WHO Expert Comm~ttee o n Bio/ogicaI Standardization Thirty-first report. Geneva, World Health Organrat ion, 1981 (WHO Technical Report Seres, No. 658), Annex 2.

7 . WHO Expert Corn~riittee on Rabies. Eighth report. Geneva, World Health Organizaton, 1992 (WHO Technical Report Seres, No. 824). Annex 2.

CHAPTER 32

Cell-culture vaccines for veterinary use P. Reciilard '

Since 1958 when rabies virus was first grown on hamster kidney cells consder- able progress has been rnade in the field of in v~ l ro cell cultures and of techniques for adapting and replicating rabies virus in these cells which has led to the development of a new second generation of rabies vaccines (1)

Several types can be distinguished among the second-generation vaccines for veterinary use depending on whether they are live or inactivated, the strain of rabies virus used and the characteristics of the cell substrate chosen for viral replication

Substrates for the production of seed virus and vaccine

Strain of rabies virus

The main virus strains used for the production of rabies vaccines for veterinary use are listed in Table 32.1.

Whatever the type o i vaccine. before prociuction I S launched, it is important to determine the specific characteristics of the strain that has been selected:

e by studying the full history of the strain from its origin up to the latest passage constituting the ~nit ial inoculum by checking on its purity, in~munogenicity and identity, The best method of identification is to obtain an antigenic map by analysis using a panel of monoclonal antibodies.

e by measuring its residual pathogenicity, i f the strain is to be used in the production of a live vaccine.

Cell cultures

Two types of cell substrate are generally used in the production of rabies vaccines for veterinary use:

e Primary cells from the organs of adult or newborn animals or embryos. The main sources of such cells are hamster, fetal bovine dog or piglet kidney tissue or chick embryo tissue.

e Diploid cells or continuous cell lines. These cell lines are mainly derived from hamster kidney cells, such as the BHK and Nil-2 lines.

' Former Head. Departmen! 01 Research a rd Development, Pasteu: Ins!i!u:e-Productior: Marries-ia- Coque:te. France

CELL-CULTURE VACCINES FOR VETERINARY USE

Table 32.1 Fjxed rabies strains used for the production of inactivated rabies vaccines

Strain Origin Passage history

Paris Pasteur

PM (Pitman-Moore)

CVS (Challenge Virus Standard)

Fuenzalida S-51

Fuenzallda S-91

Kelev

SAD (Street- Adbama D u ' f e r ~ n )

ERA (Evelyn Rokitniki Abelseth)

Vn U kovo-32

Flury

LEP

lsolated from a rabid COW in France in 1882

Pastcur rabbit fixed

rabes vrus

PV fixed rabies vlriis

PV fixed rabies virus

PV fixed rabies virus (Japan, 1915)

lsolated froni a rabid dog in Chile in 1943

lsolated from a human in Chile in 1943

lsolated from a rabid dog in Israel in 1950

lsolated from a dog that died in Alabama, USA in 1935

SAD v.rus

SAD vlrus

lsolated from a human in Georg~a, USA in 1939

Flury strain

Flurv strain

Over 300 passages in rabblt b ran , also adapted

to Vero cells Adapted to rabbit brain

fetal bovine kidney baby hamster kidney line 21 (BHK 21) and Vero ells

Adapted to rabbit brain, human diplo'd primary dog kidney, Vero and Nil-2 cells

Adapted to mouse brain BHK 21 and chick embryo- related cells

Adapted to rabbit b r a n and suckling guinea-pig brain cells

Adapted to rnoirse b r a ~ n cells

Adapted to suckling mouse brain cells

100 passages in chick embryo cells; also adapted to mouse b r a n cells

Adapted to rnoirse b r a ~ n and BHK-21 cells

Adapted to porctne kidney and BHK-21 cells

Adapted to primary tiarnster kidney cells

Adapted to chick embryo cells

40 50 Dassages in chick embryo cells, also adapted to primary chick embryo cells and to BHK-21 cells

Over 180 passages in chick embryo cells, also adapted to primary chick embryo cells

Pnmary cells T h e p r imary cel ls a re ob ta ined b y incuba t ing the o r g a n s at 37°C n a water b a t h

c o n t a i n i n g a 0.25% trypsin solut ion wi th or w ~ t h o u t the add i t i on of 0.2% edetic


acid' in phosphate-buffered saline (P6S) free of calcium or rnagnesiutn ions The bulkof the trypsin is then removed by centrifuging the cell suspension at 300 600 g for 20 minutes The cell suspension is clarified by filtration through a fine mesh nylon net or a stainless steel screen and then dispersed in growth medium (e g Eagle S minimum essential medillm or EMEM) containing 2-5% calf serum which inactivates any remaining trypsin This suspension represents the master seed lot A cell count is carried out and an appropriate volume of growth medilim containing 5-10% calf serum is then added so that when the suspension is f~nally dispensed into culture containers the concentration is 0 5 2 0 X 105 cells per cmZ of the surface available for monolayer format~on

The cell cultures are incubated at 37°C for 4-7 days or until a monolayer is formed

The cells are multiplied by successive passage to obtain a pnmary seed and then a secondary seed and finally a production seed or working seed lot The master and working seed lots are frozen in l i qu~d nitrogen and used for the preparation of further seed lots

Continuous cell 11ne.s Although some BHK cell lines can grow in suspension cultures most of the rabies vaccines produced from continuous cell lines are now prepared uslng BHK-21 and Nil-2 cells which need to be attached to a support Starting from an ampoule of the seed virus the cells are multiplied by successive passage initially in small static bottles and later in roller bottles until the quantity of cells needed for producing a batch of vaccine is obtained I f high yields are required the cell concentration of the working seed lot can be increased to 4 5 X 105 cells per cm2 in roller bottles

The cells can also be grown on microcarriers in bioreactors The microcarriers are maintained In suspension in the culture medium The cells become attached to the microcarriers and divide gradually covering them This technique was introduced by van Wezel in 1967 (2) It IS suitable for producing large quantities of virus and it also has the advantage of al low~ng several parameters such as pH oxygen and carbon dioxide levels to be monitored and controlled The cells are separated from the microcarriers using trypsin as described above Studies are under way to develop a high density cell-culture system using bioreactors of reduced volume This technique involves continuous perfusion of the culture by fresh nledium and removal of spent biolin It is simpler and cheaper to perform than classical cell-culture techniques

These cell lines are incubated for 24-96 hours depending on the initial cell concentration and the temperature of incubation or until a monolayer is formed

Preparation of the vaccines

Production of the seed virus

Table 32.2 shows the main v ~ r i ~ s and cell systems that are used for the production of rabies vaccines for veterinary use. Each type of cell substrate has a corresponding strain of rabies virus adapted to it. This makes it easier to infect the cells, since the seed vlrus which serves as the i n o c u l ~ ~ m comes from a virus strain that has been previously adapted to those cells

' Also known as ethylenedtarni~e tetraace!ate or EDTA

31 6


Table 32.2 Cell-culture rabies vaccines for veterinary use

Type of vaccine Strain of virus Cell substrate

Live ERA P g kldney cells Hamster kidney cells BHK-27 cells

Inactivated

Fury

SAD

cvs-11

Chick ernbryo fibroblasts Dog kidney cells BHK-21 cells

Dog kiciney cells Hair~ster kidney cells BHK-21 cells

Hamster kidney cells BHK-21 cells

Nil-2 cells

BHK-21 cells

The cell cultures are infected by adding the seed virus to the cell suspension before the latter is seeded into culture bottles or by adding it to the culture containers after the cell monolayer has formed The viral inoculum is added in a quantity calculated to obtain the optimum multiplicity of infection which varies according to the cell substrate culture system and growth cycle used (e g 1 plaque forming unit (PFU) per 10 cells) Puring viral replication the growth medlum is replaced by a maintenance rnedum (e g EMEM) without serum and the cultures are incubated at 33 37°C depending on the specific characteristics of the viral strain used

The v~ra l harvests are made up of the maintenance medium into which the vrrus particles are released This production of virus is accompanied by a more or less marked cytopathic effect and even by cell lyss which may occur early when a high multiplicity of infection is obtained The viral harvests collected when the quantity of virus in the medium is at 11s peak should be immediately clarified to eliminate cell debris ether by centrifugation or by successive filtration through 8-pm 3-pm and 1 2-pm membranes

Production of the final vaccines

Live attenuated vaccines if the harvested virus is to be used for preparing a live vaccine the clarified viral suspension is diluted such that one dose of the final vaccine contains the amount of ~nfective virus necessary to immunize the host animals The potency of live vaccines is expressed in units of infection e g PFU or the median lethal dose for mice (LD,,) After calibration the vaccine is immediately freeze-dried in an appropriate protective excipient (e g albumin or serum)

inactivated vaccines Various methods can be used for inactivating the rabies virus, such as treatment with P-propiolactone IBPL), acetylethyenemine or other imines. or ultraviolet


irradiation. Phenol and formaldehyde are not recommended, since they may reduce the immunogenicity of the virus and release toxic or irritant residues. The most widely used inactivating agent is BPL, generally at a concentration of 1 :4000 and in a medium with the pH ma~ntained above 7 . BPL kills both bacteria and viruses without affecting the antigenicily of the rabies virus (see Chapter 20, page 235).

Several methods that combine the action of heat and BPL are often applied so as to ensure complete inactivation of the virus and total hydrolysis of the inactivating agent. Following inactivation, the viral suspension may be concentrated by ultrafiltration on semi-permeable membranes (with a relative molecular mass cut-off of 100000), depending on the antigen content. The latter can be determined by the enzyme-l~nked immunosorbent assay (ELISA) ( 3 , 4 ) or the single radial immunodiffusion test (SRD) (5 ) , using rnonoclonal or polyclonal antibodies directed against the rabies glycoprotein (see Chapters 40 and 41).

After tests for inactivation and potency have been carrled out, the viral suspension is either freeze-dried or stored in liquid form, In the latter case, it is recommended that stabilizers and adjuvant are added before distributing the liquid inactivated vaccine into ampoules. Table 32.3 gives the composition of a stabilizer for liquid inactivated vaccines.

Manufacturing requirements for cell-culture rabies vaccines and recommendations for their use

Live vaccines for veterinary use

The manufacturing requirements for live vaccines for veterinary use are generally considerably less than those for inactivated vaccines. The production techniques are simpler and the volumes of cell and viral cultures much lower. In regard to

Table 32.3 Formula of an additive for stabilizing liquid inactivated rabies vaccines

Substance

Glycine Hstdine Argin~ne Alanine Dextran 20000 (medical purity) Lactose Sorbitol Bovine album111 Saccharose Aiurnint~m hydroxde (adjuvant)

Concentration in final vaccine (gllitre)a

1 3 1.5 4.0

0.7 2.2

10

5.5 20 50

0.25b.c

aWhen all the components (except alurnlnluni tiydrox~de) have been cornblned the pH should be adjusted to 7 2-7 4

dry we~ght of aluminturn. CAfter adding the aurnin~um hydroxide. the pH should be adjusted to 80.


production costs, live vaccines may be considered as the best suited for the transfer of technology to developing countries. However, they have several disadvantages that should also be considered:

Attenuated viral strains possess a residue of pathogencity that varies according to the species of animal and the route and dose of inoculation. This residual pathogenicrty has been demoilstrated to be associated with vaccines prepared using the LEP FIury (6), ERA and SAD strains of rabies virus (7-10). These vaccines are strictly contraindcated for puppies and cats, and for any animal in poor health.

m Live vaccines are stable only under strict conditions of storage and use, even when freeze-dried.

Inactivated vaccines for veterinary use

The inactivated cell-culture vaccines for veteriilary use have been shown to be safe and highly effective Such vaccines have proved suitable for the mass Immuniza- tion of domestic carnivores and cattle at its eighth meeting in 1991 the WHO Expert Committee on Rabies recommended that inactivated rabies vacclnes should be used for mass canine vaccination campalgns (11)

In dogs inactivated vaccines containing the adjuvant aluminium hydroxide have been shown to confer immunity for up to 3 years after a single inoculation (12) To obtain this high degree of potency however it is essential to select a very productive cell-culture system and to apply a production protocol that not only ensures high yields but also conserves the antigenicity of the virus during the various stages of productron

Control tests

International requirernerlts for testing rabies vaccines for human and veterinary use are periodically reviewed by the WHO Expert Committee on Biological Standardization (13) and have recently been revised to include vaccines prepared in cell cultures (14, 15).

These tests, which are specified for each stage of production, relate mainly to seed viruses, the cell substrate, virus harvests, complete inactivation of the virus and the safety and potency of the finished vaccine. The methods used in these tests and the minimum requirements are also reviewed by the WHO Expert Committee on Rabies ( 1 l, 16).

Special attention must be paid to testing the potency of inactivated vaccines. While in vitro test methods for the determination of antigen content are useful for making potency estimates during production, they cannot substitute for in vivo test methods for the finished vaccine (15, 17). Each lot of finished vaccine for veterinary use must be tested by the NIH test (see Chapter 37) and should have a potency of at least 1.0 International Unit (IU) per dose, In addition, the final vaccine should not be licensed or released unless an adequately designed experiment has demonstrated a duration of immunity of at least 1 year in the species for which the vaccine is to be used (16, 17).


Fig. 32.1 Plan for a facility for the production of rabies vaccine for veterinary use *

l -- pp

7 Area l

Entrance hall Dressing rooms Room for storage in l~quid nitrogen Room for storage in freezers (- 80'C) Office Toilets Store Dry ice cabinets Crushed ice cabinets Assembiy room and sterilizers Sterilization autoclave Sterilization oven

Area 2

S!, S2 = Staff entrance airiock SM1 = Airiock for media containers 8 = Room for storage of sterile equpmenl 9 = Media filtration room CE1 = Media incubation room CFl = Media cold room 10 = Cell culture room CE2 = Celi culture incubator r = Sliding doors 11 = Room for transfer of inactivated

virus vaccine SM3 = Transfer airlock 12 = Room for addition of adjuvant CF2 = Cold room for bulk vaccine H I , H2, H3, H4 = Laminar airilow cabinets EP = Pure water production

Entrance

Area 3

53, S4 = Staff entry airlock SM2 = Equipment entry airlock A2 = Autoclave for decontamination of

outgoing equipment 13 = Viral culture room CE3 = Incubator for viral culture CF3 = Cold room for storage of

virulent harvests r = Roller doors H5 = Larninar airflow cabinet

Outside the building:

Production procedure: cultures in roller tubes LT1, LT2 = Maintenance centre (system: BHK cells infected with Pasteur PVIBHK and controls rabies virus)

Annual production capacity: 400 000 to 800 000 doses.

Dimensions are given in metres. WHO 97247

CELL-CULTURE VACClNES FOR VETERINARY USE

Planning a facility for the production of rabies vaccine for veterinary use'

In planning a fac i l i t y f o r t h e p r o d u c t ~ o n of rabies v a c c i n e , f a c t o r s such as t h e size o f t h e p r e m i s e s , the q u a n t i t y of e q u i p m e n t and t h e p r i n c i p l e s o f operation a r e

d e t e r m i n e d by the manufacturing c a p a c i t y r e q u i r e d . The description h e r e re fe r s to a fac i l i t y producrng between 400000 and 800000 doses of vaccine p e r y e a r

Fig. 32.2 Plan for a biological productions facility* (housed in a hangar)

I

l = Control laboratory 2 and 3 = Bacterial vaffiines '

4 = Central washing plant 5 = Media preparation 6 = Viral vaccines (except

rabies) 7 = Rabies vacclne for

veterinary use 8 = Bonling and packaging 9 = Storage of finished

products and dispatch 10 = Wo&shops I l = Essential common

services 12 = General store 13 = Incinerator 14 = Rables animal compounds 1

(a) mice (b) kenneis !

15 = Washing plant and supplies for animal house i 16 = Animal house (mice, , rabbits, guinea pigs) 1

17 = Administrative building I 18 and 19 =Watchmen's lodgings ' 20 = High-tension line and

transformer

Exit Entrance

Shaded areas =buildings under hangar Dimensions are given in metres W O er246

'See reference 18


(Fig. 32.1) at the semi-industrial level, using cell cultures in roller bottles, within a compact space. It is not necessary to have a mechanized production line for such a facility. The production of rabies vaccines from cell cultures is based on the use of a v~rus seed lot system and on the cell seed lot system.

The quality and quantity of rabies virus produced in culture can vary greatly, depending on the type of cell and strain of virus used. It is therefore important to note that for the above output the calculations are based on average yields obtained using one of the best known systemsfor the production of rabiesvaccine for veterinary use: BHK-21 cells infected with the Pasteur PV strain of rabies virus adapted to those cells. in that system, 3-5 X 106 infected cells provide suif~cient viral suspension in 1 ml of harvest to make a dose of finished vaccine that conforms to the requirements for rabies vaccine for veterinary use published by WHO (potency at least 1.0 IU) (73). This means that the facility should be designed and equipped for the annual production of 500 litres of cell and virus culture in roller bottles, divided into 20 batches of 25 litres, to be produced over 10 months of the year. In many cases, the facility for the production of rabies vaccine will be part of a larger structure in which other viral and bacterial vaccines are produced. This is because the infrastructure and general facilities required are such that their cost could not reasonably be covered by the production of rabies vaccine alone. Fig. 32.2 shows the location of a rabies vaccine production unit within a facility for the production of biologicals.

Similar cell-culture systems can be used for the production of rabies vaccine containing attenuated rabies virus (SAD, SAG) or poxvirus and vaccinia virus recombinants (VRG), which are being tested or used for the oral immunization of carnivores in bait-delivery systems.

References

1. Sureau P. Rabies vaccine production in animal cell cultures. Advances in biochemical engineering and biotechnology, 1987, 34: 11 -1 28.

2. Van Wezel AL. Growth of cell strains and primary cells on microcarriers in homogeneous culture. Nature, 1967, 216: 64-65.

3. Van der Mare1 P, van Wezel AL. Quantitative determination of rabies antigen by ELISA. Journal of biologica! standardization, 1981, 50: 267-275.

4. Perrn P, Morgeaux S, Sureau P. In vitro rabies vaccine potency appraisal by ELISA: advantages of the mrnunocapture method with a neutralizing anti- glycoprotein monoclonal antibody. Biologicals, 1990, 18: 321-330.

5. Ferguson M, SchId GC. A single radial immunodiffusion technique for the assay of rabies glycoprotein antigen. Journal of general virology, 1982, 59: 197-201,

6. Dean DJ, Guevin VH. Vacclnation of cats against rabies. Journal of the American Veterinary Medical Associalion, 1963, 142: 367-370.

CHAPTER 33

Modified live-virus rabies vaccines for oral immunization of carnivores J Blancoo ' & F -X Mes/in2

During the 1960s research workers in Canada and the USA showed for the first time that it was possible to vaccinate red foxes (Vuipes vulpes) by feeding them baits containing a modified live-virus (MLV) rabies vaccine (1-4). While this method was only studied experimentally in the USA, it was rapidly developed and used for the oral immunization of foxes in the field in Europe initially in Switzerland (5), and later In the Federal Republic of Germany ( 6 ) and in several other European countries (7)

It was subsequently shown that it was also possible to use these vaccines for the oral immunization of other species, including the wolf (Lupus lupus) (8, g), the corsac or Afghan fox (Vulpes corsac) (10 ) , the arctic fox (Alopexlagopus) (1 1, 12) , the racoon dog (Nyctereutes procyonoides) ( g ) , the domestic dog (10, 13), the racoon (Procyon lotor), and the cat (14, 15) However, species other than Vulpes sp, required higher dosesof MLVvaccine to be immunized orally. In addition, it was demonstrated that MLV vaccines were not able to protect all the major host species of rabies for example, oral immunization with ERA (Evelyn Rokitnik Abelseth) vaccine did not protect the striped skunk (Mephitis mephitis) against challenge

(76) This chapter examines the technical aspects of MLV vaccines that are of

~nterest to laboratory workers and rabies control authorities including their origin, identification production. safety and efficacy.

For further information on the use of these vaccines, their incorporation in baits and their distribution during vaccination campaigns in the field, the reader is referred to more comprehensive reviews. including the special issue of Revue soentifique et technique de /'Office International des Epizooties (17) the eighth report of the WHO Expert Committee on Rabies (18) the proceedings of the International WHO Symposium on Wildlife Rabies Control (19) and the reports of a number of WHO meetings (20 25).

Modified live-virus vaccines

All the trials published so far have shown the difficulty and very often the practical impossibility of vaccinating against rabies by the oral administration of an inactivated rabies virus (7, 26-34)

In this chapter only the MLV vaccines will be considered Recornbinant Iive- vlrus vaccines are discussed in detail in Chapters 34 and 35

' Director General Office Interrational des Epizootes (DIE) Paris France Chief Veterinary Pub ic Heal'h Division of Co?imdncabIe Diseases World Healtq Orga?zatior Geneva Switzerland

324

MLV VACCINES FOR ORAL IMMUNIZATION OF CARNIVORES

Several types of MLV vacclnes have been proposed for the oral ~mmun~zation of animals in the past 20 years including CVS (Challenge Virus Standard) ( 1 35), HEP (Flury High Egg Passage) and LEP (Flury Low Egg Passage) (36) a variant of HEP HEP 675 (37 38) AvOl (avirulent Orsay 1) (39) TSG (thermosensitive Gif 1) (40), GSC (salivary gland cell) (41) TAG (thermosensitive avirulent Gif) (42) and a variant of SAD (Street-Aabama-Dufferin) SADVa (43) However only six MLV vaccines have proved suitable for use in the field ERA SAD Bern SAD-B19 SAD-P5 88 SAG, arid SAG, A I these viruses are derivatives of the origir~al SAD virus a street rabies virus whch was isolated from a dog that died in Alabama in 1935 The virus was adapted to baby hamster kidney (BHK) cells and Dufferin was added to the name when the virus was obtaned by Connaught laboratores (44)

The sections that follow provide technical details on these vaccines The virus titres are expressed either in plaque-forming units (PFU) Iml MICLD,, m1 (the median lethal dose for mice inoculated by the ntracerebral route) or TCID,, m1 (the median tssue-culture infective dose)

ERA vaccine

Ong1n ERA virus is a dervatlve of the SAD virus adapted to BHK cells (45)

Identifica tion The virus can be distinguished from wild virus strains using monoclonal antibodies (MAbs) It has been extensively characterized using genetic mapping and used for the production of the recomb~nant v a c c n a rabies vaccine (46).

Production The virus is produced by Connaught laboratories in Ontario, Canada. It can be cciltvated on various cell lines, partculary BHK (and its clones BHK-21 and BHK- 21 C13) cells. Vero cells, arid primary porcine and canine kidney cells The mean titres obtained in the culture supernatant vary according to the cell line (106.' TCID5,/ml on primary porcine kidney cells; 106.7 MCLD,,/mI on primary canine kidney cells; 107.' TCID5,/m on BHK-21 C13 cells). The stability of the vaccine is increased by adding chemical substances containing proteins, such as actal- bumin, egg yolk or human serum albumin.

Safety The vaccine is pathogenic for adult mice (by the ntracerebral, intramusci~lar and oral routes) and for many other rodent species, including ferrets, hedgehogs, rats and muskrats (by the oral route) (47-50). It is also pathogenic for skunks when given by the nasal or enteric routes (15) and for cats when given by the intramuscular route (50) .

Efflcacy Efficacy has been demonstrated by many authors using either d~rect instillation in the mouth or oral administration of the vaccine in a b a t . The effective dose has been established at 106 MICLD,, in the fox (51). When given orally at higher doses (107.5-108 MICLD,,), the virus induces virus-neutralizing antibodies in the dog (13, 51) and the racoon, but not in the skunk or the cat (16).


Field trials Since 1990, about 2 million baits containing ERA vaccine have been distributed in south-eastern Ontario, Canada. The baiting campaigns have led to a dramatic reduction in the incidence of rabies in this area.

SAD-Bern (SAD-BHK) vaccine

Origin SAD-Bern (SAD-BHK) virus is a derivative of the SAD virus, which was originally produced on BHK-21 cells at the Institute of Veterinary Microbiology of the University of Bern, Switzerland (5) .

Identification The virus may be distinguished irom field fox strains using MAbs (43)

Production The virus is cultivated on BHK cells. The mean titres obtained in the culture supernatant are 107 TCID,,/ml. The stability of the vaccine is increased by adding 5-10% egg yolk. The infectivity titre then decreases by only half after 72 hours at 37 "C (52).

Safety Ingestion of the vaccine can be fatal in 1-5% of some rodent species (Apodemus sylvaticus, Microtus arvalis and mice) (23, 53), and also in baboons (Papio cynocephalus) (54). However, its residual pathogencity is similar to that of the ERA strain (see above). In foxes that have been art~fic~ally irnmunosuppressed with corticosteroids, the virus can be isolated from the salivary glands 8-14 days after oral administration (55).

Efficacy Oral administration of 106 TCID,, has been shown to protect red foxes against challenge (53). Ingestion of 107 TCID,, produced seroconversion in 27 of 36 foxes, 5 of 7 dogs and 4 of 14 cats (56).

Field trials The SAD-Bern vaccine has been used in plastic capsules stapled to chicken head baits (5) Between October 1978 and October 1990, 1.3 rnill~on such baits were distributed In Switzerland with great success (18 ) . Since then, however, the SAG, vaccine has been used instead.

SAD-B19 (BHK- Tii) vaccine

Origin BHK-B19 (BHK-Tu) virus 1s a derivative of the SAD virus. It was cultivated at the Federal Research Institute for Animal Virus Diseases in Tubingen, Germany The virus was selected for vaccine production on the basis of ~ t s thermostability and its low residual pathogenicity (57).


Identification The virus may be distinguished from field rabies virus strains and the ERAISAD derivatives using MAbs.

Production The virus is produced on a selected clone of BHK cells. The titre obtained in the culture supernatant is generally above 10' TCID,,/ml (57). The vaccine remains stable for up to 14 days in the field, even at high temperatures (40°C) (58). However, its stability can be increased further by adding 2% calf serum.

Safety

SAD-B19 vaccine is pathogenic for mice and rats when given by the intracerebral intramuscular and oral routes However it appears to be less pathogenic than the ERA or SAD-Bern vaccines for non-target species, particularly the muskrat (59) It is not pathogenic for cats, even when given after treatment w ~ t h immuno- suppressant drugs (55) or for racoons opossiims or dogs (15) Similarly no cl~nical signs of rabies were observed in chimpanzees following oral administration of 1 5 X 1 Oa focus-forming units (FFU) of the vacclne (60)

Efficacy The effective dose in the red fox has been estimated at I o ' . ~ TCID,, (58). When administered orally (in a bait) to racoons, a dose of TCIDBO was shown to confer protection (in 6 out of 8 animals) against challenge (16) . The virus also protected racoon dogs against challenge (57).

Field trials SAD-B19 vaccine has been widely used in the field. Between 1983 and 1993, over 20 million baits containing this virus were distributed in Europe (including Austria, Belgium, the former Czechoslovakia, France, Germany, Italy and Luxembourg) with no reported deaths among non-target species and with seroconversion rates among foxes of 50-75%, depending on the area (58, 61, 62).

SAD-P5/88 vaccine

Origin SAD-P5188 virus is a derivative of the SAD-Bern virus, which was originally cloned and adapted on baby hamster kidney cells (line BSR-P5/88) (63) .

Identihcation SAD-P5/88 virus may be distinguished from other rabies viruses using the Mabs NC 239.17' and NC 187.5.10.' It can also be identified by its defined residual pathogenicity for certain species of laboratory rodents when administered by the oral route.

Production The virus is cultivated on BSR-P5188 cell monolayers in revolving vessels or in suspension.

' Avalable on request from the WHO Colaborat~ng Centre for Rabies Surveiiiance and Research. Federal Research institute for Animal Virus Diseases, Postfach 1149, D-W 7400 Tub~ngen. Germany.


Safety Ingestion of the vaccine can be fatal in 5-10% of some rodent species (mice and muskrats). However, ~t is not pathogenic for foxes, cats, dogs, ferrets, mink, racoons, rats, hens, pigeons or wild boars. No cases of transmission of the virus have been noted. The virulence of the virus did not increase following 10-12 successive passages in mice.

Efficacy When administered orally to six foxes, a dose of 1.6 X 106 FFU was shown to confer protection against a challenge that killed all controls. A large-scale field trial carried out over an area of 29000 km2 and involving four immunization campaigns conducted at 6-month intervals led to the elimination of the disease in the target area (64).

Field tr~als SAD-P5188 vaccine has been widely used in the field Over 12 million doses of this vaccine have been distributed in Germany with no reported cases of vaccne- induced rabies or persistence of the virus In wildlife populations (65)

The stability of the vaccine was tested under field conditions at temperatures ranging from 5 5'C to 20'C (mean 16 7 C) The virus titre decreased from 106 FFU per dose to 106 FFU after 1 week and to 10" FFU after 2 weeks Storage for 2 weeks at 4 "C or 10°C led to only a slight reduction in virus titre (0 2 log FFU) however the vrus titre was reduced by 1 2 log FFU following storage at 20'C Repeated freezing and thawing (10 cycles) had no effect on the virus titre

Vnukovo-32 vaccine

Origin Vnijkovo-32 virus is a derivative of the SAD vlrus which was cultivated at 32 C on hanister cells in a laboratory located close to the town of Vnukovo Russian Federation

Idenb f~cat~on The virus may be d~stinguished from field rabies virus strains and other SAD- related strains using MAbs

Production The virus is cultivated on primary hamster kidney cells (34-40 passages) ( l l ) at the Institute of Poliomyelitis and Viral Encephalitis of the Academy of Medical Sciences. Moscow, Russan Federaton. The titres obtained in the culture supernatant vary from 106 to 10' TCID,,/ml.

Safety No data are as yet available about the safety of this vaccrne in non-target species.

Efficacy When given orally in chicken head ba ts to 86 arctic foxes, the vaccine elicited high titres (0.2-0.0016) of virus-neutralizing antibod~es (11). The immunizing dose was not specified,


Field trials Field trials are being conducted in the Russian Federation, in which tens of thousands of baits containing the Vnukovo-32 strain have been distributed.

SAG, vaccine

Origin

SAG, virus is a deletion mutant of the SAD-Bern virus, which was originally cloned on CER cells using the antiglycoprotein MAb 50AD1 (66) and cultivated on BHK cells in the presence of the same MAb.

ldentifica tion SAG, virus differs from SAD-Bern virus in possessing a serine residue in position 333 of the amino acid sequence of the glycoprotein and one different nucleotide in codon 333, It is resistant to neutralization by MAb 50AD1, It may also be identified using MAb 40DC2.

Production The virus is cultivated on BHK-21 cells, The mean titres obtained in the culture supernatant are usually 10' TCID,,/ml.

Safety SAG, vaccine is pathogenic neither for adult laboratory mice nor for several species of wild rodents tested by the oral, intramuscular or intracerebral routes; however, it is pathogenic for suckling mice when given by the ntracerebral route. Several studies have compared the residual pathogenicity of the SAG, vaccine with that of other vaccines intended for oral vaccination of animals (67, 68).

Efficacy The virus conferred protection against challenge to 9 foxes that had ingested a dose of 106 PFU and to 4 out of 5 dogs that had ingested baits containing IO'.' TCID,, (17) . In a study car!-ied out at the Centers for Disease Control and Prevention, Atlanta, G A USA. 3 groups of 6 dogs were each given a dose of log , 10' or 106 TCID,, orally and challenged with a virulent strain (Mexican dog strain) after 364 days. The immunized dogs were protected against challenge, whereas the unimmunized controls died from rabies.

Field trials Between 1989 and 1993, over 3.5 million baits containing the SAG, vaccine were distributed in France and Switzerland.

SAG2 vaccine

Or~gin SAG, virus is a deletion mutant of SAD-Bern virus, which was selected in two steps in the presence of two different antiglycoprotein MAbs.

ldentif~calion SAG, virus differs from SAG, and SAD-Bern viruses in possessing a glutamate residue in position 333 of the glycoprotein and two different nucleotides in the


codon for position 333. The replacement o i two nucleotides reduces the chances of reversion to the more virulent SAD-Bern virus.

Production The virus is cultivated on BHK-21 cell monolayers in revolving vessels in the Dresence of MAbs.

Safety SAG, vaccine is apathogenc for adult mice when given by the intramuscular or intracerehral route Its pathogenicity was determined in severe combined immunodeficient (SCID) mice All SCID mice inoculated orally and all adult outbred and immunocompetent ICR (white Swiss) control mice given the vaccine orally or parenterally survived the observation period without showing clinical signs

Efficacy Efficacy trials have been conducted in foxes and dogs When administered orally to 16 foxes a dose of 10' PFU of SAG2 elicited detectable titres of virus-neutralizing antibodies and conferred protection against a challenge that killed all controls Similarly 2 groups of 5 dogs given a dose of 10' or 10' PFU orally were protected against a challenge that killed 4 out of 5 controls (60)

Guidelines for assessing the safety and efficacy of MLV vaccines

Safety assessment for target and non-target species

Dogs are very closely associated with humans especially children in most cultures The likelihood of direct exposure and of passive transfer of rabies vaccine virus is considerably higher in programmes for the oral immunization of dogs than in those for the immunization of wildlife In view of the high probability of contact between young children and puppies the vaccine chosen for oral vaccination should not produce disease in such young dogs when administered orally at 10 times the dose for field use

The possibility of excretion of vaccine virus in the saliva of the animals described above shoiild also be examined Following immunization swabs should be taken daily for 10 days No virus should be present after 3-4 days Any virus recovered should be characterized using MAbs or other appropriate procedures

In addition where feasible at least 10 and i f possible 50 of each of the most common local rodent species should be given the field dose of vaccine (I e the dose which is contained in a bait) orally and lntramuscularly This may require the use of difierent vlrus concentrations and volumes for different speces depending on their we~ght and size If the animals so vaccinated develop sickness or die from rabies the use of the vaccine should be recons~dered

Relevant local wilc or domestic animal species that may take baits should also be given a dose of vaccine orally equivalent to 10 times the field dose in a volume adapted to body weight In Europe common species are wild boars, stone martens badgers and cats

Any rabies virus isolated from animals in vaccination areas should be examined using MAbs or other appropriate tests to ensure that no vaccne- induced rabies has occurred


An intense surve~llance system should be established to detect any possible

human exposure to vaccine. Persons who accidentally come into contact with the vaccine (via the mouth. nose, eye or a wound) should receive rabies post-exposure treatment. Similarly, persons worlting with the vaccine and at risk of exposure to it should receive pre-exposure inimunization.

Additional precautions

The use of live vaccines, particularly for ttie oral vaccination of dogs, should be discouraged when the risk of unintentional exposure of severely immunocompromised persons (such as those with acquired imniunodeficiency syndrome) is considered to be high, because of the chance of enhanced viral replication, altered tropism or untoward adverse events. Conversely, inactivated vaccines do not appear to represent a danger to immunocompromised individuals Research on inactivated rabies virus vaccines for the oral immunization of dogs is in progress, and some encouraging results have been obtained in racoons. Until an effective inactivated vaccine becomes available, however, several candidate MLV and recombinant vaccines appear lo offer safe, efficacious and economical solutions for the oral immunization of target species.

The candidate live vaccine should also be tested in primates, such as chimpanzees, baboons or rhesus monkeys. At least 10 animals of one species should be given 1 m or more of 10 times the intended field dose of vaccine by direct instillation into the oral cavity. If sufficient numbers of animals are available, the vaccine should also be tested in a similar number of immunocompromised primates. No vaccine-related deaths should occur during an observation period of at least 90 days. Tests for rabies and vector virus antibodies should be made before inoculation of the vaccine and at the end of the experiment.

Considering that incubation periods of rabies following MLV inoculation in primates may be in excess of 90 days, i t is suggested that primates be administered a dose of a modern, potent, inactivated rabies vaccine and serum samples be examined for signs of an anamnestic response (e.g. a sudden rise in rabies virus- neutralizing antibodies) or the "early death" phenomenon (69).

The candidate vaccine should also be given by the oral, intiacerebral, intramuscular and other relevant routes to nude and SClD mice or other immunodeficient laboratory animal models. Results should be compared with those observed in immunocornpetent laboratory animals.

Following oral immunization, saliva samples should be taken from the animals described above and examined for vaccine virus using an appropriate test. Studies should also be carried out on appropriate immunodeficient laboratory animal models to determine the effects of different VII-us titres.

1 Baer GM, Linhart SB, Dean DJ. Rabies vaccination of foxes. Annual report of the Divis~on of Laboratory Research. Albany, New York State Department of Health, 1963: 145.

' Uriiess ot"lrwisc stated, unpublished 0ocuinel:s have been prepared by the World Health Organizattori, Geneva, and are available or1 request from the Otvtsiori o f Comrnuntcablr~ Dtscascs, Word Health Organtzation 121 1 Geneva 27 Switzerland


2. Black J G Lawson KF. Sylvatic rabies studies in the silver fox (Vulpes vulpes). Canadian journal of conlparative medicine, 1970. 34: 309-31 1.

3. Black JG, Lawson KF. Further studies of sylvatic rabies in the fox (Vulpes vulpes). Canadian veterinary journal, 1973. 14: 206 21 1.

4. Debbie J G , Abelseth MK, Baer GM. The use of commercially available vaccines for the oral vaccination of foxes against rabies, American journal of epidemiology, 1972, 96: 231 -235.

5. Steck F et al Oral immunization of foxes against rabies. A field study Zentraiblatt f i r Veterinarmedizin, Reihe B, 1982, 29: 372-396,

6 Schneider LG. Oral iminunizaton of wildlife against rabies. Annalesde l'lnstitut Pasteur. Virology, 1985 136E: 161 -1 65.

7. Blancou J et al. La vaccination antrabique des animaux sauvages. [Immu- n~zal ion of wildl~ie against rabies.] Revue scientifique et technique de /'Office International des Epizooties, 1988, 7 : 989-1 003.

8, lvanovsltii EV. Specific prophylaxis of rabies in animals (GNKI ethanol vaccine for wild animals). In. Urban VP, ed. Problemy veterinarnoi imniunologia. [Problemsof veterinaryimmunology] Moscow, Agropromizdat, 1985: 141-143.

9. Kovalev NA, Davydenko VP, Shashenko AS. Effectiveness of oral immunization of carnivores against rabies. Veterinarnaya nauka-proizvodstvo, 1987, 25: 3-6.

10. Kerimbekov KK et al. Immunity following oral vaccination of carnivores against rabies. In: Urban VP, ed. Problen~y veterinarnoi immunoiogia. [Problems of veterinary in7munology.l Moscow, Agropromizdat, 1985. 143-145.

11. Selimov MA et al. Oral immunization of arctic foxes with live rabies vaccine prepared from the Vnukovo-32 strain propagated in cell culture. Akademia med~cina nauk, 1987, 32: 622-623.

12. Follmann E H , Ritter DG. Baer GM, lrnmunization of arctic foxes (Alopex lagopus) with oral rabies vaccine. Journal of wildlife diseases, 1988, 24: 477-483.

13 Blancou J et at Vaccination par vole orale du chien contre la rage et epreuve par un virus d origine canine [Immunization of dogs against rabies by the oral route and challenge with a virus of canine origin] Annaies de mkdecine veterinaire 1990 134 563-566

14. Rupprecht CE, Kieny MP. Development of vaccinia-rabies glycoprotein recombinant virus vaccine. In: Campbell JB, Charlton KM, eds. Rabies. Boston, Kluwer Acadeniic Publishers, 1988 335 -379.

15 Blancou J et a1 lnnocute et efficacite d un vaccn antirabraue recombinant


des virus de la vaccine et de la rage adminstre par voie orale au renard, au chien et au chat. [Safety and eificacy of a recombinant vaccinia-rabies vaccine administered by the oral route to foxes, dogs and cats.] Annales de recherche vetennaire, 1989, 20. 1955204,

16. Tolson ND et al. Studies of ERAiBHK-21 rabies vaccine in skunks and mice. Canadian journal of veterinary research, 1988, 52: 58-62.

17. Rage en Europe. [Rabies in Europe.] Revue scientifique et technique del'Office International des Ep~rooties. 1989, 8 (No. 4).

18. WHO Expert Committee on Rabies. Eighth report. Geneva, World Health Organization, 1992 (WHO Technical Report Series, No. 824).

19 Bogel K Mesl~n F-X Kaplan M eds Wildlife rabies control (Proceedings of the lnternational WHO Symposium on wildlife rabies control, Geneva 2-5 July 1990) Royal Tunbridge Wells Wells Medical 1992

20. Report of a WHO Consultation on oral immunization of dogs against rabies, Geneva, 26-27 February 1988 (unpublished document WHOjRab. Res./ 88.26).

21 Report of a WHO Consultation on fequirements and criteria for field trials on oral rabies vaccination of dogs and wild carnivores Geneva 1-2 March 1989 (unpublished document WHO, Rab Res ,89 32)

22 Report of a WHO NVI Worhst7op on arctic rabies Uppsala 24-27 April 1990 (unpublished document WHO Rab Res 9035)

23. Report of a WHO/APHIS Consultation on baits and baiting delivery systems for oral immunization of wildlife against rabies, Fort Collins, 10-12 July 1990 (unpublished document WHO/Rab.Res./90.36).

24. Report of a WHO Seminar on wildlife rabies control, Geneva, 2-5 July 1990 (unpublished document WHO/VPH;90.93)

25. Report of the Second WHO Consultation on oral immunization of dogs against rabies, Geneva, 6 July 1990 (unpublished document WHO/Rab. Res.jg1.37).

26. Nicholson KG, Bauer SP. Enteric inoculation with ERA rabies virus. Evaluation of a candidate wildlife vaccine in laboratory rodents Archives of virology, 1981, 67.51-56.

27. Metianu T. Vaccinat~on antirabique par voie orale avec des vaccins tues. Premiers resultats. [Vaccination against rabies by the oral route using inactivated vaccines. Initial results.] Bulletin de I'Academie veterinaire, 1981, 54.4815490.

28. Lawson KF et al, Immunization of foxes by the intestinal route using an inactivated rabies vaccine. Canadian journal of veterinary research, 1989, 53: 56-61.


29. Blancou J et al. Vaccination du renard roux (Vulpes vulpes L.). contre la rage par voie orale. Bilan d'essais en station experimentale. [Vaccination of red foxes (Vulpes vulpes L.) against rabies by the oral route. Assessment of trials at an experimental station.] Revue d'ecologie (terre et vie), 1985, 40: 249-253.

30. Brochier B et al. Vacciriat~on of young foxes (Vulpes vulpes L.) against rabies: trials with inactivated vaccine administered by oral and parenteral routes. Annales de recherche veterinaire, 1985, 16: 327-333.

31. Report of the Third WHO Consultation on oral immunization of dogs against rabies, Geneva, 21-22 July 1992 (unpublished document WHO/Rab.Res.; 91.38).

32. Aubert MFA, Blancou J, Hoffman M. Oral vaccination of foxes against rabies: trials with inactivated virus in enteric coated tablets. Rabies information exchange, 1982, 5: 37-38.

33. Maharaj l , Froh K J Carnpbell JB, Immune responses of mice to inactivated rabies vaccine administered orally, potentiation by Quillaja saponin, Cana- dian journal of m~crob~ology, 1986, 32: 41 4-420.

34. Chaval SR, Carnpbell JB. lmmunomodulatory effects of orally administered saponins and nonspecific resistance against rabies infection, International archives of allergy and applied inimunology, 1987, 84: 129 134.

35. Bussereau F, Aubert M, Blancou J. Temperature-sensitive mutants of rabies virus: behaviour following inoculation into mouse and fox. Annales de I'lnstitut Pasteur. V;rology, 1983. 134E: 31 5-325.

36. Dubreuil M et al. The oral vaccination of foxes against rabies. An experimental study. Annales de recherche vetkrinaire, 1979, 10: 9-21.

37. Kiefert C, Wachendorfer G, Frost JW. Unschadlichkeitspriifungen mit der geklonten Variante des Flury HEP-Virus (Stamm 675) bei wildlebenden Spe- zies. Ein Beitrag zur oralen lmmunisierung von Fijchsen gegen Tollwut. [Safety tests with the cloned variant of the Flury HEP virus (strain 675) in wild animals. Contribution to the oral immunization of foxes against rabies.] Tierarztliche Umschau, 1982, 37: 165-1 76.

38. Wachendorfer G, Friedrich H, Frost JW. Wirksamkeitspriifungen rnit der geklonten Variante des Flury HEP-Virus (Starnm 675) beim Fuchs (Vulpes vulpes L.). Ein Beitrag zur oralen Irnrnunsierung von Fiichsen gegen Tollwut. [Efficacy tests with the cloned variant of the Flury HEP virus (strain 675) in the fox (Vulpes vulpes L.). Contribution to the oral immunization of foxes against rabies.] Tjerarztliche Umschau, 1984, 39: 93-103.

39. Pepin M et al. Oral immunization against rabies with an avirulent mutant of the CVS strain: evaluation of its efficacy in fox (Vulpes vulpes) and its infect~vity in seven other species. Annales de I'lnstitut Pasteur; Vjrology, 1985, 136E: 65-73.

40. Prehaud C et al. Characterization of a new temperature-sensitive and avirulent mutant of the rabies virus. Journal of general virology, 1989, 70: 133-143.


41. Blancou J, Aubert MFA, Andral L. Studies on pathogenic, immunogenic and protective efficiency of fox rabies virus before and after adaptation to cell culture: application to vaccinat~on against rabies. Canadian journal of microbiology, 1983, 29: 77-83.

42. Tdke R et al. Characterization of a double avirulent mutant of rabies virus and its potency as a vaccine, live or inactivated. Vacone, 1987, 5: 229-233.

43. Schneider LG et al. Application of inonoclonal antibodies for epidemiological investigations and oral vacclnation studies. Ill. Oral rabies vaccine. In: Kuwert E et al., eds. Rabies in the tropics Berl~n, Springer-Verlag. 1985: 55-59.

44. Fenje P. Propagation of rabies virus in cultures of hamster kidney cells. Canadian journal of microbiology, 1960, 6: 479-484.

45. Abelseth MK. Propagation of rabies virus in pig kidney cell culture. Canadian veterinary journal, 1964, 5: 84-87.

46. Kieny MP et al. Expression of rabies virus glycoproten from a recombinant vaccinia virus. Nature, 1984, 312: 163166.

47. Bijlenga G, Joubert L. Haute pathogenicite pour la sours par vole digestive d'un composant viral du vaccin antirabique a virus vivant modifie ERAIBHK. Sur la prophylaxie medicale de la rage en France. [ H ~ g h pathogencity for mice of a viral component of the modified-live ERA/"BHK rabies vaccine administered by the oral route. Rabies prophylaxis in France.] Bulletin de I'Academie veterinaire, 1974, 47: 423-435

48. Forster U et al. Unschadlichkeitspriifungen von Tollwut-Lebendvakzinen an wildlebenden Saugern. [Safety tests of live rabies vaccines on wild rodents.] Fortschritte der Veterinarnledizin, 1976, 2 5 : 257-262.

49. Winkler WG, Shaddock JH, Wlliams LW. Oral rabies vaccine: evaluation of its infectivity in three species of rodents. American journal of epidemiology, 1976, 104. 294-298.

50 Lawson KF et a Safety and immunogen~city of ERA strain of rabies virus propagated In a BHK-21 cell line Canadlan journal of veterinary research 1989, 53 438-444

51. Lawson KF et al. Safety and immunogenicity of a vaccine b a t containing ERA strain of attenuated rabies virus. Canad~an journal of veterinary research. 1987, 51 : 460-464.

52, Hafliger U et al. Oral immunization of foxes against rabies: stabilization and use of bait for virus application. Zentralblatt fijr Veterjnarn7edizin, Reihe B, 1982, 29. 604-61 8.

53. Steck F et al. Oral immunization of foxes against rabies. Experentia. 1978, 34: 1662.


54 Bingham J et al Pathogenicity of SAD rabies vaccine given orally in Chacma baboons (Papio ursinus). Veterii'lary record, 1992 131: 55-56.

55 Ciuchini F et al Effects of corticosterods mediated mmunosuppression on the distribution of rabies vaccine virus in red foxes orally lmmunlzed against ra bies Journal of veterinary medicine Series B 1986, 33 628-631

56. Ciuchini F et a!. Richerche su di un vaccino aliestito con lo stipite SADjBHK-21 (Beriia) del virus de la rabbia da impiegare per la vaccinazione dela volpe. [Research on a vaccine prepared from the SAD, BHK-21 (Bern) rabies virus strain for the immunization of foxes.] Clinica veterinaria. 1985, 108: 219-230.

57. Wandeler Al. Oral immunization of wildlife. In: Baer GM. ed. The natural history of rabies, 2nd ed. Boca Raton CRC Press. 1991 485-503.

58 Schneider LG et a Current oral rabies vaccination in Europe an interim balance Rev~ews of infectious diseases 1988 10 654-659

59. Schneider LG, Cox JH. Ein Feldversuch zur oralen Immuriisierung von Fiichsen gegen Tollwut in der Bundesrepublik Deutschland. [A field trial of the oral immunization of foxes against rabies in the Federal Republic of Germany.] Tierarztliche U/nscl-iau, 1983. 38 31 5-324.

60 Report o l the Fourth WHO Consi~ltat~ori on oral immunization of dogs against rdbies Geneva 14 15 June 1993 (unpublished document WHO Rab Res /' 93 42)

61 Wachendorie~ G et al Erfahrungen mlt der oralen lmmun!sierung von Fbchsen gegeli Tollwui in Hessen [Experience with the oral immunization of foxes against rabies in Hessen ] Tierarztliche Praxis 1986 14 185-196

62. Pastoret PP et al. Campagne nteri-iationale de vaccination antirabique du renard par vole orale rnenee au grand-duche de Luxembourg, en Belgique et en France. [International campaign of oral rabies vaccination of foxes in Luxembourg, Belgium and France.] Annales de medecine veterinaire, 1987, 131 : 441 4 4 7 .

63 Sinnecker H et al Die Entwicklung des Tollwutlebendimpfvirus SAD Potsdam 5 88 zur oralen Fuchsimpfung sowie seine Charakterisierung am Mausmodell [Development of the SAD, Potsdam 5 88 live rabies virus vaccine for tile oral immunizaton of foxes as well as its characterzation in the mouse model] Monalshefte fur Veterinarmedizin 1990 45 77-80

64 K~ritscher M Bernhard J Lemke I Untersuchungen zilr Unschaalichkeit epizootiolog~schen Unbedenklichkert und Wirksamkeit des Tollwutirnpfstam- mes SAD Potsdam 5 88 [Studies on the safety ep~zootiological safety and effectiveness of SAD Potsdam 5 88 strain for rabies vaccination ] Monats- hefte fur Veterinarmedizin 1990 45 81 -84

65. Stohr K et al. Orale Immuriisierung freilebender Fijchse gegen Tollwut -

MLV VACClNES FOR ORAL IMMUNIZATION OF CARNIVORES

Vorberetung und Durchfuhrung der ersten Feldversuche in den ostdeutschen Landern. [Oral imniunization of wild foxes against rabies-preparation and implementation of initial field trials in the Germati Democratic Republic.] Monatshefte fGr Veterinarmedizin, 1990. 45 782-786.

66. Leblois H et al. Oral immunization of foxes with avirulent rabies virus mutants. Veterinary microbiology, 1990. 23: 259-266.

67. Flamand A et al. The antigenc structure of the rabies glycoprotein, application of basic research to oral vaccination of foxes. In: Thraenhart 0 et al., eds. Progress in rabies contmi. (Proceedings of the Second International lMVl

Essen,/WHO Syrnpos~iim on "New Developments in Rabies Control", Essen. 5-7 Juiy 1988 and Report of the WHO Coi~sul fa t~on on Rabies, Essen. 8 July 1988.) Royal Tunbridge Wells Wells Medical, 1989: 72 77.

68. Artois M et al. Potential pathogenicty against rabies: a comparison. Vaccine, 1992. 10: 524-528.

69. Blancou J. Andral B, Andral A. A model in mice for the study of the early death phenomenon after vaccination and challenge with rabies virus. Journal of general virology, 1980, 50: 433-435.

Genetically engineered vaccines

CHAPTER 34

General considerations in the use of recombinant rabies vaccines for oral immunization of wildlife C. E. Rupprechl, ' C. A ffan/on2 & H. Koprowski3

Advances in cloning and gene expression have led to the evolution of a new gelleration of rabies vaccines Strains of orthopoxvirus such as vaccnia virus have been extensively characterized at the moleci~lar level (1-3) Their large genome ease of production therrnostability broad host range and safety record make these viruses highly suitable for the expressori of foreign genes including hepatit~s herpes ~nfluenza and rab~es vttuses Expression of foreign genes by a self-replicating entity (I e vector virus) offers the potential for long-term immunity without the risk of disease from the foreign agent when used in its entirety as with certain attenuated viruses For example the recombiriant vaccnia virus that expresses the glycoprotein (G protein) gene of rabies virus (VRG) which was developed by inserting a plasmid-bearing cDNA copy of the G protein gene of the ERA strain irito [he thymidine kinase gene of the vaccnia virils (Copenhagen strain) (4 5) cannot cause rabies When administered orally to laboratory animals the VRG vaccine rapidly induces high levels of rabies virus-neutrali71ng antibodies and confers protection against challenge with several ar~tigencally distinct street rabies viruses

The VRG vdccine was first shown to induce rabies vrus-neutralizing antibodies when given orally to laboratory rodents (6) It was subsequently demonstrated that racoons (7) and other carnivores (8-10) could be vaccinated orally through the consumption of baits containing the VRG vaccine After 10 years of experiments under laboratory conditions and field trials in Europe (71) and North America (12) the VRG recombinant virus vaccine is the only recombnant vaccine that has been extensively used in the field for the control of rabies in wildlife Two additional recombnant vaccines have been prepared (I) using a racoon poxvirus vector expressing the G protein of the CVS strain (13) (11) using a human adenovirus type 5vector expressing the G protein of the ERA strain in its deleted E3 region (14) It is anticipated that Lhese and other recently developed recornbinant vaccines such as those prepared using herpes virus canine adenovirus avipoxvirus and baculovirus may undergo limited safety and efficacy trials in the field in the near fu t~ i re

' C h i e l Rabies Section. Division of Viral and Hicket tsa Diseases National Center for lnfectious Diseases Centers for Disease Control and Prevent~on. Atianta. C A USA iiabes Section. Uivis~on o? V~rai and Rickettsia Diseases. Na!ionai Center for Infectious Diseases. C e ~ t e r s for Disease Control aiid Prevention, Atlanta G A USA Director Center lot Neuroviroiogy, Jefterson Cancer instl tute Thomas Jefierson Un lve rs t y Philadelphia, PA USA


The majority of the safety requirements for modified live-virus vaccine$ are also applicable to recomonant vaccines (15-17 see also Chapter 33) The VRG vaccine has been extensively tested in over 10 avian and 35 mammalian species ncIudii?g the majority of rabies reservoir hosts

In addition studies have been carried out in primates in which the vaccine was admii?istered by the parenteral and oral routes (17) Groups of 7-8 squirrel monkeys were given VRG vaccine or one of two other vaccinia virus vaccines lntradermally by scarification No significant pathological differences were detected between the three groups Of 11 chmpanrees given VRG orally ( log PFU ml) 8 developed rabies virus-neutralizing antibodies No adverse effects were observed in the immunized animals

Regardless of the vaccine dose or roule of administration the vaccinated animals remained clinically norrnal with no overt gross or histopathological les~oris Following oral administration tonsils and local lymph nodes are key sites ior early v~ra l repl~cation However the VRG vaccine is cleared relatively quickly from the saliva (e g within 48 hours) and any residual virus is readily inactivated in the gastrointestinal tract True I~orizontal transmission has not been observed although mechanical transmission I S possible (e g a recently vaccinated lactating female racoon grooming her suckling kids) Passive immunity may be transferred to the progeny No abortifacient teratogenic or oricogenic side-effects of VRG virus hdve been found Large-scale field trials in foxes have been conducted with the VRG vaccine in Belgium and France and limited trials have been carried out in racoons in the United States no adverse effects have been reported to date

Studies in severe combined immunodeficient (SCID) mice have shown that the VRG vaccine is much less pathogenic than the parental vaccinia virus strain when given by the intramuscular intraperitoneai intracranal or intradermal route thereby confirming the results found in immunocompetent animals Adverse effects were minor in mice given the VRG vaccine and were dependent on the dose and the route of administration (16)

During a field trial of VRG vaccine in southern New Jersey in the United States up to 20% of stray cats in the vaccination area showed evidence of contact with baits containinq the vaccine Given their high risk for infection with feline leukaemia birus (FeLV) feline infectious perilonitis (FIP) or feline mmunodefi- ciency virus (FIV) and the close relat~onship between domestic animals and humans the patiiogenicity of the VRG vaccine for these potentially immunosuppressed animals was evaluated In two related studies domestic cats with and wthoi i t feline viral infections were given the VRG vaccine orally (1 m containing 10' PFU) or intraderinally (100 jil containing 10" PFU per site) There were no differences In the development and regression of mild intradermal lesons at the

inoculation sites regardless of whether the cats were infected or not No leslons were observed in animals given the vaccine by the oral route (16)

The VRG vaccine shares rnariy basic propert~es with parental vacclnia vlrus but differs in other ways which make the vector virus safer The deletion of the thymidne kinase gene dramatically reduces the pathogenicity of the vaccine for rodents when it I S administered by the intracerebral and ntraperitonea routes (18) In additoi-i no viral spread from currently known sites of viral replicalion has been observed Other methods of attenuation such as the deletion of genes accounting for the virulence of the virus or Ihe insertion of immunomodulator genes may also

prove ~iseful bearing in mind that additional atteriuation may alter the efficacy of

RECOMBINANT VACCINES FOR ORAL IMMUNIZATION OF WILDLIFE

the vaccine. Other safety concerns relating to the potential for the establishment of wildlife reservoirs of the disease, the possible implications for the public health sector and the probability of untoward environmental events resulting from recombination of the recombinant vaccine vector in the field with naturally occurring viruses should also be addressed. regardless of the viral vaccine species used as vector.

Although no longer in use, vaccinia was routinely used as a human vaccine for almost two centuries. Dur~ng the worldw~de campaigl? to eradicate smallpox, rnillions of doses of certain strains of vaccinia virus containing high concentrations of dermatotropic-adapted virus were administered by scar~fication, ofien under less than ideal conditions (19). Uriquestionahly, the ecolog~cal opporturlity for transfer of virus from humans to domestic animals frequently existed and transmission by contact was demonstrated (e.g. from cattle to humans). Consider- ing the intensity of the bond between humans and animals in certain cultures, cases of transmission were much less frequent than predicted and were self- limiting In spite of extensive studies, the evolutionary origin of vaccinia virus was never identified and no original natural hosts of vaccinia virus have ever been found (20, 21). Thus, vaccinia virus appears best regarded as a laboratory virus (non-existent in nature) for which humans were the intentional prrncipal host. Although vaccinia virus has a broad host range in experimental infection. there is no substantial evidence to suggest that i t can become widely established in animal populations by natural infection. Other candidate recombinant vaccines have not yet been evaluated in the field

No self-replicating system ether modified-live or recombinant-based, is totally risk-free. However. it is clear that vaccinla virus is a minor pathogen when given parenterally. When inoculated intradermally, it produces a lesion characterized by hyperplasia, proliferation and inflammatory infiltration that progresses from a papule, vesicle and pilstule to a crust, followed by healing. A transient viraemia is possible, but generalized lesions are extremely rare in irnrnunocompetent hosts. Occasional abnormai reactions range from mild, local ulceration to vaccine- induced deaths, broadly classified as dermal in origin or involving the central nervous system (e.g. post-infectious viral encephalitis). Considering the attenuated nature of the VRG recombnant virus, its demonstrated innocuity in animals. the proposed restricted nature of its use and the frequency with which humans have been exposed to baits containing the vaccine (less than one case of exposure per 100000 baits deployed), it is extremely unlikely that such vaccine-induced complications would occur in immunocompetent humans from field application. However, only widespread field tests will be able to document adequately these otherwise biologically rare events. both for vaccinia and other virus vector recombnant vaccines.

Compared with the high frequency of intramoiecular viral recombination that occurs in vitro under non-selective conditions, the likelihood of natural recombination is remote. For recolnbnation to occur, simultaneous infection with two viruses capable of genome interaction is necessary. The probability of recombination of vaccine virus with naturally occurring related viruses and consequent regenera- tion of a wild-type virus depends on the taxonomic relationship of the viruses and also on the prevalence of the recornbinant virus and that of the naturally occurring virus, The latter can be determined by examining sera from wild animals for various agents related to the virus vector of the vaccine prior to field tests. For orthopox-


viruses in general. there is no strong evidence to suggest that recombination has occurred under natural conditions (22)).

Safety requirements for recornbinant live rables vacclnes have been elaborated by various WHO consultations on oral immunization of carnivores against rabies (16 17).

It is likely that recornbrnantvaccines will eventually be used in the global control of many infectious diseases. Such vaccines have been shown to be safe, potent. versatile and reliable, and should gradually replace less effective first- and second- generation viral vaccines. Attention to characteristics of the vector, such as its origin, passage history. pathogenesis, epizootiology and any unique qualities of the resirlting recornbinant species, should be considered during vaccine development ( l 6 77). While vaccin~a virus vaccines were among the first to be extensively used in the laboratory and the field (23. 24), other viral vectors may also ultimately contr~bute to the control and elimination of rab~es in the future.

References

1 Mackett M Smith GL Moss B Recornbinant vaccinia viruses as new live vaccines (review) Biotechnology and genetic engirleering review 1984 2 383-407

2. Mackett J et al. Vaccinia virus recombinants: expression of VSV genes and protective inimunization of inice and cattle. Science. 1985, 227: 433-435.

3 Markett M Smith GL Vaccin~a virus expression vectors Journal of general virology 1986 67 2067 2082

4 Keny M P et a1 Expression of rabies virus glycoproten from a recomblnant vacciria virus Nature 1984 312 163-166

5 Wiktor TJ et a1 Protection from rabies by a vaccinia virus recombnant vaccine containing the rabies glycoprote~n gene Proceedings of the Nat~onal Acad- emy of Sciences of the Un~ted Slates of America 1984 81 71 94-71 98

6. W~ktor TJ et al. lrnmunogenic properties of vaccinia recombinant virus expressirlg the rabies glycoproten. Annales de l'lnstitur Pasieur. Virology. 1985 136E: 405--411.

7 Rupprecht CE et al. Oral lmrnunization and protection o l raccoons (Procyon lotor) with a vaccinia rabies glycoproten recombinant virus vaccne. Proceed- ings of the Naiional Academy of Sciences of the United States of America,

1986, 83: 7947-7950.

8. Blancou J et al. Oral vaccnation of the fox against rabies using a live recombnant vaccinia virus. Nature, 1986, 332: 373-375.

9. Tolson ND et al. Immune response I[? skunks to a vaccinia virus recombinant expressing the rabies virus glycoprotein. Canadian journal of veterynary research, 1987, 51 : 363-366.

RECOMBINANT VACCINES FOR ORAL IMMUNIZATION OF WILDLIFE

10. Rupprecht CE, Kieny MP. Development of a vaccina-rabies glycoprotein recornbinant vlrus vaccine In Carnpbell JB, Charlton KM, eds. Rabies Boston. Kluwer Acadeniic P:iblishers, 1988 335-364.

11. Pastoret PP et al. First field trial of fox vaccination against rabies using a vaccina-rabies recombinani virus. Veterinary record, 1988. 123. 481-483.

12 Hanon CA et a1 Proposed field evaluation of a rabies recornbnant vacclne for raccoons (Procyori lotor) site selection target species characterst~cs and placebo baiting trials Jouinal of wildlife d~seases 1989 25 555-567

13 Esposito JJ et ai Successful oral rabies vaccination of raccoons with raccoon poxvrus recombinants expressing rabies virus glycoprotein (technical note) Virology 1988, 165 313-316

14 Prevec L et a1 A recombinant human adenovrus vaccrie against rabies Journal of infectious diseases, 1990 161 27-30

15 Wandeler AI et al Oral immunization of widli le against rabies concept and first f led experime~its Revieu of ~nfectious diseases 1988. 10 649-653

16 Report of the Foiirth WHO Consultation on oral irnrnunoation of dogs against rabies Geneva 14-15 June 1993 Geneva World Health Organization 1993 (unpublished document WHOfRab Res 93 42 available on request from the Division of Cornm~~nicable Diseases World Health Organization 121 1 Geneva 27 Switzerland)

17 Repori of the Third WHO Consiiltation on oral immi~nization of dogs against rabies Geiieva, 21-22 July 1992 Geneva World Health Organization 1992 (unpublished docunient WH0,Rdb Res ,92 38 available on request from the

;?eneva Division of Communicable Diseases World Health Organ~zalion 121 1 C' 77, Switzerland)

18 Buller RMI- et a1 Decreased viriilence of recoriibinant vaccnia virus expression vectors is associated with a thymidine kinase-negative phenotype Nature, 1985 317 813-815

19. Fenner F et al. S n ~ a l l p i ? ~ and its eradication Geneva, World Health Organza tion, 1988.

20 Baxby D Vaccinia virus In Qunnan GV Jr ed Vaccinia viruses as vectors tor vaccine antigens (Proceed~ngs of the Workshop on Vaccin~a Virrises as Vectors for Vaccine Antigens Chevy Chase MD 13- 14 November 1984 ) New York Elsevier 1985 3-8

21 Dumbell KR Aspects of the biology o l orttiopoxviruses relevant to ihe use of recornbinant vaccines as field vaccines In Qunnan GV Jr ed Vaccinia viruses as vectors for vacone antigens (Proieediiigs of the Workshop on Vaccinia Viruses as Vectors for Vaccine Arlligens Chevy Chase MD 13- 14 November 1984) New York Esever 1985 9-13


22. Durnbell KR. Diverse family of orthopoxviruses: biological safety issues. In: Fox J, ed. Rabies: large-scale use of vaccines. Washington, D C National Audubon Society. 1990: 8.

23. Rupprecht CE et a1 Oral wildlife rabies vaccination, development of a recoinbinant vrrus vaccine. In: McCabe RE, ed. Transactions of the fifty- seventh Meeting of the North American Wildlife and Natural Resource Conference. Washington, DC, Wildlife Management Institute, 1992: 439-452.

24. Hanlor? CA et al. A vaccina-vectored rabies vaccine field trial: ante- and post- mortern bomarkers. Revue scient~fique et technique de /'Office Irilerriational des Epizooties, 1993, 12: 99-1 07.

CHAPTER 35

Expression of rabies proteins using prokaryotic and eukaryotic expression systems B Dietzschoid

To date, both the glycoproiein (G protein) and nucleoproten (N protein) genes of the rabies virus have been successfully expressed in a variety of prokaryotic and eukdryotic expression systems

Complementary DNA (cDNA) coples of the rabies G protein messenger RNAs (rnRNAs) o l the ERA and CVS-11 strains were inserted into the bacterial cloning vehicle pBR322 at the Pstl site (I) The hybrid plasm~ds were then transferred into Eschench~a coli and a clone containing the entire nucleotide sequence of the G protein gene was selected The cDNA contarnrng the complete sequence of the N protein gene was inserted into plasm~d pUCI9 (W H Wunner personal communi- catron)

Prokaryotic expression systems

Escherichia coli

Yelverton et a1 (2) used a vector containing the G protein cDNA for the direct expression of the complete sequence of the G prote~n of CVS-11 rabies strain in E c011 Lathe et al (3) used derivatives of the M13 bacterial plasmds to direct the expressron of the rabies G protein cDNA in E c011 In both constructs cells containing the rabies G protein cDNA synthesized a new protein that was recognrzed by antrsera raised agarnst rabres virus G protein However the G protein expressed in E coii was extremely insoluble, and did not react w t h ant -G protern monoclonal antibodies (MAb-Gs) directed against conformational determinants This G protein also fa~led to confer protection aga~nst rabies These findings suggest that the G protein was probably not folded or processed correctly Thereiore efforts have turned to eukaryotic systems that allow the correct processing of the G plotein molecule

Eukaryotic expression systems

Poxviruses

Poxviruses are excellent candidates for the expression o i foreign genes because of the large size of their genomes and their mode of replication. Recently, vaccinia virus has been developed as a cloning and expression vector, and immunization

' Professor, Center ior Neurology, Department of Immunology and M~crob~ology Jcfierson Medcal College, Thomas Jeffersari IJn~vers~ty . P+ladelpI i~a PA, USA

347


experiments havp shown that genetically engineered recon~binant vaccinia viruses would be suitabie candidate vaccines ( 4 5) Accordingly the vaccinia virus was used to express ttie rabies virus G protein The construction of the re- combinaiit vaccinia virus expressing the G protein gene of rabies virus (VRG) is desciibed in detail elsewhere (6) Briefly the cDNA sequence of the G protein was placed under the control of the early class vaccirlia virus promoter and inserted at the BarnHl site ot the thymdt?e Ihnase gene cloned in a suitable bacterial plasmid (see Chapter 34) The chimeric gene formed in this manner contained the vaccinia RNA stdrt site juxtaposed with the rabies translational codon to avoid the p r o d u ~ t u n of a fusion protein This plasmid construct was used to transfect vaccinia v i rus-nf~cted cells to prepare a recombinant virus contanirlg the cDNA ot the rabies G proteii? at the thynldine kinase locus Infection of BHK-21 cells with the VRG virus resiilted i r i ttie expression of rabies G protein on the cell surface as detected by immur~ofluorescence assay The protein expressed by the VRG virus reacted w ~ t h MAb Gs In a pattern identical io that of native rabies virus G protein suggesting that the newly expressed G protein had the rorrect conformation When administered intiadernially or ir~tramuscularly to mice and rabbits the VRG virus rapidly induced tiigh titres of rabies virus-neutralizing antibodies and conferred protection against challenge with street rabies virus strains and the rabies-related Duvenhage virus strat i (7) Moreover racoons immunized orally with VRG virus developed virus-neutrairr?g antibodies and long-term protection aga~rist rabies denlonstrating that this virus may be a suitable candidate for oral immunization of w~ldlife (8) Previous vaccination with VRG did not interfere with the immune response to a second inoculation with this recombinant For example there flas a draniatic iricrease in virus-neutralizing antibody titres when the vaccinated animals were given a booster injection 6 months after the primary imm~rnization (8)

The VRG vitus is also capable of inducing cell-mediated immunity A J mice immunized with VRG demonstrated a substantial secondary cytotoxic T- ymphocyte resporise in wtro after re-exposure of their lyrnphocytes to rabies virus or after inoculatior~ with a pathogenic strain of rabies virus (9)

Esposito et a1 (10) used racoon poxvrus to produce two infectious recombinant virus vaccines that express the rabies G protein Expressiori of the G protein was controlled by Ihe early late class vaccinia promoter P75 or the late class promoter P11 Bolh recombinant viruses were showri to express the G protein accurately and mice or racoons inimiinired with either recombinant developed high levels of virus-neutralizing antibodies and were protected against a lethal challenge tnfection with rabies street vlrus

Taylor et at ( 1 1 ) used d fowl poxviius (TPV) vector to immunize non avian species against raoies The cDNA of ttie rabies viriis G protein gerie was expressed urder the contiol of the early late class vaccinia promoter H6 The recombinant

virus was shown to express the rabies G protein accurately and although there was no evidence of a productive viral replication in a variety of mammals such as mire cats dogs rabbits and cattle ttie level ot G protein expression was sufficient to induce a protective immune response in these species The absence of production of ~nfect~ous progeny virus may represent an important safety feature for vaccination s r c e the potential transrnisson to other animal species and also the replication in irnmunoconipromised anlmals 1s greatly reduced

EXPRESSION OF PROTEINS

Adeno viruses

Prevec et a1 (12) inserted the cDNA of the rabies virus G protein o l the ERA strain between the promoter and poly-A addition site of the early region of simian vacuolating virus 40 (SV40) and placed this construct into the deleted E3 region of human adenovirus type 5 This recombinant virus contains the rabies G protein gene in the same orientation as the F3 gene in the parerit virus The recombinant vrus was found to be highly mrnunogenlc eliciting rabies v~rus-neutralizing antibodies in dogs and mice and protecting mice against a lethal intracerebra1 challergc infectiori W th rabies virus

Baculo viruses

The insect baculoviriis Autoyrapha californca (AcNPV) is a very potent expression system for foieign genes because the yields of the expressed proteins are very high dnd the biological properties of the expressed mater~al are conserved ( 13)

Prehaud et a / (14) have used the AcNPV to express the cDNA of rabies virus G prote~n cloned iron? the Gif-siir-Yvette CVS strain The G protein cDNA was inserted into the transfer vector pAcYM1 such that the cDNA was under the control of the AcNPV polyhedron proinoter Cotransfecton of Spodoptera frugiperda cells with this recombinant transfer vector and AcNPV DNA resulted in a recombinant virus that exhb~ted a polyhedron-negative phenotype and expressed the rabies G protein in those cells

The redctiv~ty pattern of the expressed material with several MAb-Gs was identical to that of the rlative rabies virus G protein and immunization of mice with the expressed G protein resulted in high levels of virus-neutralizing antibodies and also conferred protection against a lethal challerige infection with rabies virus The synthesis of large quantities of G protein in insect cells infected with the AcNPV recombinant makes this expression system a potential economical source of a low- cost rabies vaccine provided that effective methods of purification can be developed

In a later study Fu et a1 (15) cloned a cDNAcopy of the RNA gene that encodes the N protein of the ERA strain of rabies virus into baculovrus The recombinant baculoviius expressed the N protein abundantly in Spodoptera frugiperda cells The N protein was subsequently extracted and purified to near homogeneity using affinity chromatography The piirified N protein reacted with 31 of 32 MAbs directed against the native rabies virus ribonucteoprote~n (RNP) Like the RN? the purified N protein was a major aniigen capable of inducing virus-specific T-helper cells Priming of mice with the purified N protein prior to a booster inoculation with inact~vated ERA virus vaccine resulted in a 20-fold increase in the production of virus-ne~itralzing antibodies After immunization with the purified N protein mice developed a strong anti RNP antibody response and were protected against a letha! challenge with iabies vrrus These data ~ndicate that the N protein expressed in insect cells IS ani~gencal ly and immunogenically similar to native rabies virus RNP and could therefore be used for the production of a low-cost rabies vaccine for the large-scale immunization of humans and animals


References

1. Anitionis A. Wcrnner WH. Curtis PJ. Structure of the glycoprotein gene in rabies virus Nat~~ re . 1981. 294: 275-278.

2. Yelverton E et al. Rabies virus glycoprotein analogs: biosynthesis in Esche- richia coli. Science, 1983, 21 9: 61 4-620.

3. Lathe RF et al. M13 bacteriophage vectors for the expression of foreign proteins in Escherlchia coli: the rabies glycoprotein. Journal of molecular and appiied genetics, 1984, 2: 331 3 4 2 .

4. Panicali D et al. Construction of live vaccines by using genetically engineered poxviruses biological activity of recombinant vaccinia virus expressing influenza virus hemagglutinin. Proceedings of the National Academy of Sciences of the United States of America, 1983, 80: 5364-5368.

5. Smith GL, Mackett M Moss B. Infectious vaccinia virus recornbinants that express hepatitis B virus surface antigen. Nature, 1983, 302. 490-495.

6. Kieny M-P et al. Expression of rabies virus glycoprotein from a recombinant vaccinia virus. Nature, 1984, 321: 163-166.

7. W~ktor TJ et al. Protection from rabies by a vaccinia virus recombinant containing the rabies virus glycoprotein gene Proceedings of the National Academy of the United States of Amer~ca, 1984. 81: 71 94-71 98.

8. Rupprecht CE et al. Oral immunization and protection of raccoons (Procyon lotor) with a vaccinia-rabies glycoprotein recombinant virus vaccine. Proceed- ings of the National Academy of Sciences of the United Slates of America, 1986, 83: 7947-7950.

9. Wktor TJ, Kieny M-P Lathe R. New generation of rabies vaccine. In: Kurstak E et al., eds. Applied virology research, Vol. I . New York, Plenum Press. 1988. 69-90

10. Esposito JJ et al. Successful oral rabies vaccination of raccoons with raccoon poxvirus recombinants expressing rab~es virus glycoprotein. Virology. 1988, 165: 31 3-31 6.

1 1 . Taylor J et al. Recombinant fowlpox virus inducing protective immunity in non- avian species. Vacc~ne, 1988, 6: 497-503

12. Prevec L et al. A recombinant human adenovirus vaccine against rabies. Journal of infectious diseases. 1990, 161 : 27-30.

13. Luckow VA, Summers MD. Trends in the development of baculovirus expression vectors. Biotechnology, 1988, 6: 47-55.

Vaccine safety and tests for potency and antigen

quantification

CHAPTER 36

General considerations in testing the safety and potency of rabies vaccines' P. Sizaret2

Safety tests on rabies vaccines are designed to detect any material or property that may be harmful to the recipient, such as contamination by extraneous agents, incomplete inactivation of the virus, or toxicity. Details of these tests are provided in the requirements for rabies vaccines published by WHO (1-6). Specific minimum requirements and regulations should be established by national health authorities, It is desirable that such regulations be similar to those published by WHO.

As with other inactivated viral vaccines, avoiding the presence of residual virulent virus is of the utmost importance. For rabies vaccines the test for residual virulent virus is performed by intracerebra inoculation of mice, usually with a sample from the inactivated bulk, i.e. the material prior to the addition of preservatives and other substances. Furthermore, i f the vaccine is produced in cell culture, the rabies virus amplification test should be carried out in a cell substrate similar to that used for production. In the case of modified-live (attenuated) rabies vaccines for veterinary use, the problem is rather more complex because a minimum amount of infectious virus is required; it must be demonstrated, however, that each final lot is safe for the intended recipient.

Besides rabies virus, which is intentionally present in the vaccine in some form, inactivated and modified-live rabies vaccines may contain other infectious agents that were present in the original animal or cell-culture source, Inactivated rabies vaccines produced from a cell-culture source must meet the safety requirements for extraneous agents that are demanded for inactivated cell-culture vaccines in general. Procedures giving absolute assurance that a rabies vaccine is free of all possible extraneous infectious viruses can be very complex and costly. Such procedures can, however, be limited to what is reasonable and practical by taking advantage of the knowledge accumulated with a particular cell-culture or animal source.

Where the rabies vaccine is derived from a continuous cell line, the tests for the detection of extraneous agents in the substrate should be performed on each virus seed lot, a process which minimizes the overall costs of production. In the case of vaccine for human use, the purified bulk material should be shown to contain a quantity of residual cellular DNA not greater than that set by the national control authority (3).

For licensing purposes, it has always been difficult to evaluate the efficacy of candidate rabies vaccines for human use on the basis of results obtained in

I Based on the chapter by !he late K. Habei In ?he prevlous ed~tioii Former Scientific Officer, Biologicals, Division of Drug Management and Policies, World Health Organization. Geneva, Switzerland.


humans under field conditions because of (a) the lack of true controls (b) the relatively small number of hiiman cases of rabies in a given area and (c) the impossibility of accounting for all the factors that impair the comparability of human exposures In practice individuals are immunized according to the recommended schedule and serum samples are then examined at intervals to determine the rapidity of appearance and the amount of virus-neutralizing antibodies

Althougti the need for evaliiating the immunizing potencies of each final lol of rabies vaccine before release has been recognized since the early Pasteur days and practical star~dardized tests have been available and in use for over 45 years many laboratories producing rabies varcines still do not rnake a routine practice of periodically testing the potency of their products However merely follow~ng a standatd vaccine production procedctre does not necessarily ensure that the u r o d u ~ t will always have saiisfactory potency levels

There appear to be three Important considerations in assessing a potency test for the release of rabies vaccines First the test procedure should evaluate in a quantitative way the property of the vaccine that determines its effectiveness in the prophylaxis of rabies 111 humans or ariimals Using a naturally susceptible host the ideal test would sitnulate conditions of natural exposure and usual prophylactic treatment In the case of vaccines for human use this would mean the use of street virus introduced by the peripheral route in the most appropriate animal model followed by daily injections of the vaccine lot to be released This has been found to be impractical as have most types of tests where vaccine I S administered alone after experimental infection of the test animal The NIH potency test therefore involves two irijectons of vaccine followed by a subsequent challenge with fixed virus given intracerebrally-a more easily standardized type of challenge

While far from reproducing the conditions of natural exposure and the standard schedule of vaccine admin~stration, the NIH test does reflect In most cases the ability of a vaccine to protect the intended target species under natural conditions However for licensing purposes vaccines for veterinary use should have been demonstrated to protect vaccinated animals of the intended target species (e g dog or cat) against a field rabies virus challenge

On the other hand potency tests for modified-l~ve vaccines for veterinary use consist of measuring the titre of infectious virus in a sample from each fllling lot The vaccine lot is released if the t~tre is not less than that proved as efficacious in all species of animals foi which the vaccine is recommended (2)

The second important factor in a potency lest is its practicability-the ease with which it IS carried out the availab~lity of the materials needed and the cost and time involved The time factor is important since vacciries should not be released until potency tests are completed and the time required for t h ~ s test has to be deducted from their period of effectiveness

The third requirement is for standaid~zation of the test procedure so that the results obtained with different vaccines will be comparable, regardless of the laboratory where they have been tested

It has been found through experience that the two most important variables that can markedly influence the results of any type of potency test are the animals and the challenge virus employed For optimum results, the animals should be in good health and of the correct age Another factor that has proved troublesome in tests in mice is the variation in the ability of mice of different strains and from

TESTING THE SAFETY AND POTENCY OF VACCINES

different colonies to respond to a given vacciiie Obviously for test purposes, a strain of rnice should be used that has been shown to respond efficiently to a knowri potent vaccine Uniformity of response among individual animals making up the test group does not necessarily require highly inbred strains but all animals should be from the same closed colony stock

Variation in the degree of demonstrable immunity has also been shown to depend upori the strain of virus used to challenge the immunized animals This variation has been minimized bv the developnient of a standard challenge virus- the CVS fixed virus derived from the original Pasteiir stra,n ' The proper technique

for preparing and handling the slanddrd challenge virus is given in Chapter 37 In the following chapters of Part V several dilferenl p o t e n ~ y test5 are described

The first is the NIH test a potency test in mice which shoi~ ld be used for the purpose of releasing inactivated virus vaccines intended for human or veterinary use The inclusion of a standard vaccine which has been calibrated against the Interna- tional Standard for Rabies Vaccine and which is tested in parallel with the unknown vaccine is essential for the result to be expressed in International Units (IU) in the case of veterinary vaccines the inclusion of a reference vaccine is also necessary for demonstraling that the test product is not inferior to that of a vaccine which has beer1 shown to be efficacious in the species of animals for which the vaccine is intended The NIH test is the most standardized of the existing in vivo potency tests and tinder optimal conditions i t gives iesults of high comparability However lt has certain disadvantages when compared with potency tests involving peripheral challenge In v a r o ~ i s experiments in i i i ce and other species, modified live recombitiarit vaccines expressing rabies virus nucleoprotein did not confer protection against iritracerebral challenge but did elicit protection agaipst peripheral challenge (7-9) The NIH test which involves ntrdcerebrai challenge does nol Drov de a measure of the protector conferred by the rabies (ribo)nucleoproteii- conta ried in the complete vlron vaccine

The Habe test and the guinea-pig potency test are also included in Part V since these tests are still used by some laboratories

Studies have been initiated to develop potency tests in mice involving vaccination followed by peripheral challenge with rabies virus (10) In one such test mice are givei- an intramuscular injection conta n ng 0 1 m1 of vaccine followed 4 weeks later by a peripheral challenge Such challenge procedures more closely mimic natural exposure than does ntracerebral challenge and may in the future replace the NIH test

The other tesis described in Part V are in vilro tests for determining [he antigen contenl of rabies vaccines including the single radial ~rnmunodifiusion (SRD) test the enzyme-linked imrnurlosorberit assay (ELISA) and the modified aritibody- binding test (ABT) These tests are quicker and more economical to perform than the NIH test and have been found to be suitable for in-process controls It is hoped that some of these tests will eventually replace challenge tests currently used for [he release of inactivated rabies vaccines provided that the glycoprotein and nucleoproiein ont tent of the lot to be released is shown to be no less than that of

' CVS-77 IS a v a ~ d b l c !O ria?~oria' Iaboratoric~s or1 reql lest l r o r i :he Arnerca i~ Type Culture Coiiect~on (ATCC) 17301 Parklaw17 D f ~ v e Rockvi le MD 20852, USA and from t h e W o r d H e a t h Organrator: 171 1 Geneva 77, Switzerland


the manufacturers reference vaccine However consensus has not yet been reached on suitable tests based on antigenic content (4 6)

Modified live-virus recombinant vaccines expressing only the rabies virus nucleocapsid protein can be tested by measuring the level and persistence of antinucleocapsid protein ant~bodies However it may be important to measure the induction of cellular immunity against rabies viriis Although tests are available that can measure the in vitro proliferation of lymphocytes or specific cytokines such tests are difficult to perform and to interpret they are therefore not currently recommended for routine use (10)

References

1. Requirements for rabies vaccine for human use. WHO Expert Committee on Biological Standardization. Thirty-first report. Geneva, World Health Organ i -

zaton, 1981 (WHO Technical Report Series, No. 658), Annex 2.

2. Requirements for rabies vaccine for veterinary use. WHO Expert Committee on Biological Standardizaiion. Thirty-first report. Geneva. World Health Organi- zation, 1981 (WHO Technical Report Series, No. 658) Annex 3.

3. Requirements for rabies vaccine (inactivated) for human use produced in continuous cell lines. WHO Expert Committee ori Biological Standardaation. Thirty-seventh report. Geneva. World Health Organization. 1987 (WHO Tech- nical Report Series, No. 760), Annex 9.

4. Requirements for rabies vaccine for human use (amendment 1992). WHO Expert Committee on Biological Standardization. Foriy-third report Geneva, World Health Organization, l994 (WHO Technical Report Series, No. 840). Annex 4.

5. Requirements for rabies vaccine (inactivated) for human use produced in continuous cell lines (amendment 1992). WHO Expert Committee on Bio- logical Standardization. Forty-third report. Geneva, World Health Organira- tion. 1994 (WHO Technical Report Series, No. 840), Annex 5.

6. Requremenrs for rabies vaccine for veterinary use (amendment 1992) WHO Expert Cornn-iittee on Biological Standardization. Forty-third report. Geneva, World Health Organization, 1994 (WHO Technical Report Series, No. 840), Annex 6.

7. Tollis M et al. Immunization of monkeys with rabies ribonucleoprotein (RNP) confers protective immunity against rabies. Vaccine. 1991, 9: 134-136.

8. Lodmell DL et al. Raccoon poxvirus recombinants expressing the rabies virus nucleoprote~n protect mice against lethal rabies virus infection. Journal of virology. 1991, 65: 3400-3405.

9. Fu ZF et al. Rabies virus nucleoprote~n expressed in and purified from insect

TESTING THE SAFETY AND POTENCY OF VACCINES

cells is efficacious as a vacclne. Proceedings of the National Academy of Sciences o f the United States of America, 1991, 88. 2001 -2005

10. WHO Expert Cotnrnijtee o n Rabtes. Eigtilh reporl. Geneva, World Health Organization, 1992 (WHO Technical Report Seres, No. 824).

CHAPTER 37

The NIH test for potency' L. A. W//bur2 & M. F. A. ~uber t "

Standard test

The N H test for potency was originally developed at the National Institutes of Healtli Bethesda MD USA The test measures the degree of protection conferred by iiaclivated rabies vaccnes in i;lmui-iiied mice challer~ged with rabiesvrus The test is condiicted by vaccinating two groups ot mice twice 7 days apart with dilutions of a referen~e vaccine and the vaccine under test Seven days after the last vaccination the immunized animals and a control group of mice are challenged with the Challenge Virus Standard (CVS) mouse-brain strain of fixed rabies virus The mice are observed for 14 days and the median effective dose (ED,,) of the refereiice and test vacciries is determined based on the number of survivors The relative potency of the test vaccine is then calculated by comparing the ED,, of the test vaccine with that of the reference vaccine

Challenge Virus Standard (CVS)

CVS is usually supplied4 frozen as a 20% mouse-brain suspension in a diluent containing 2% of horse serum in distilled water. This should be shipped in solid carbon dioxide (dry ice) and stored at - 70-C; it should be used only if received in the frozen state, It is also distributed as a freeze-dried preparation when there is a possibility that the shipment will thaw before receipt. The ampoules of the freeze- dried preparation contain 0.5 m1 of a 20% mouse-brain suspension otherwise identical with the frozen preparation; they should be shipped with frozen ice packs and stored at 4'C.

Preparation of the working CVS

1 . Thaw the contents ot an ampoule of frozen virus rapidly under cold running water and dilute with 2% horse-serum diluent (CVS diluent see Annex) to give a looh mouse-brain suspenslori. Alternatively, reconstitute the freeze-dried preparation w ~ t h the recommended diluent or sterile water and then dilute with CVS diluent to obtain a 10% mouse-brain suspension. Higher titres will be obtained with freeze-dried preparations by passaging the seed virus once in mice before final expansion,

' Based on rhe chapter by t B Sei~gmai,n Jr 1.1 thc! previous edition. Biologies Virology Laboratory. National Vclcririary Services Laboratories, Animal and P a r i l Health inspection Service, United States Deparliiicrit of Agriculture (USDA) Ames. IA LJSA Labora!oire d'E!udes sur ia Rage et la Pathoiog~e des Animaux sauvages Centrtl National d'Etudes veternares et ailmentairss [CNEVA) Malzeviiie. France

4 A v ~ i l a b i e on reuues! from the Worid Heaitb Oraarization, 121 1 Geneva 27 S w i t i e r i a ~ d . Wl-l0 C ~ l l a b o - rating Cen!res (see Appendix 4 ) t i le Amei~can Type Culture Collection (ATCC). 12301 Parklawn Drive. Rockviiie MD 20852, USA, or national control a ~ t l i o r i t i c s

NIH TEST FOR POTENCY

2 Centrifuge the mouse-brain suspensiori at 200 g for 10 minutes. Dilute the resulting supernatant with CVS diluent to a final concentration of approx- mately l o 3 LDS, per ml. The dilution should be calculated from the reference insert received with the CVS.

3. Inoculate at least l 0 young mice, weighing 13-16 g . intracerebraly with 0.03 m1 of the diluted supernatant.

4. When they are completely paralysed (usually after 6-10 days), kill the mice hiimaneiy and remove their brains. Imrnediately freeze the brains with dry ice and store at -- 70 'C.

5. When the collection is complete, thaw. weigh and reduce the harvested brains to pulp using a sterile rnorlar and pestle, a tissue grinder, a mixer or another appropriate device Tt is procedure shoi~ ld be carried out under a negative draught hood to prevent the release of the virus in an aerosol (see Chapter 1) . Add sufficient CVS diluent during grinding to yield a 20% suspension by weight. Each mouse brain will yield about 1.5 ml of mouse-brain suspension.

6. Assign a lot number to the suspension, centrifuge at 200 g for 10 minutes. and immediately distribute into sterile ampoules. Seal the ampoules, freeze rapidly and store at - 70 'C.

Each step in preparing tlie working CVS rnust be carried out promptly and in an ice water bath or equvalerir so as to ensure the survival of the maximum nossiole amount of virus

Determination of the LD,, of the working CVS

Before use as a challenge virus the median lethal dose (LD,,) of each lot of the working CVS should be determined in 6-week-old mice as follows:

1 Remove oine arnpoirle of tlie pooled working CVS from storage at - 70 C and thaw rapidly under cold run i ing water

2 Prepare serial tenfold dilutions of the supernatant in CVS diluent 3 Inoculate groups of 10 mice intracerebraly with each diliition of the working

CVS each rnoirse receiving 003 m1 Obseive the mice for 14 days and record the number that die from rabies afterthe first 5 days Include any mice showing signs of fixed-virus rabies (paralysis convulsions) on the 14th day

4 Calculate the LD,, of the working CVS using the rnethod of Spearman & Karher (see Appendix 3) A lot is considered satisfactory i f the LD,, value 15 between the 10-" O and 1 0 - 8 0 dilutions, inclusive Tlie maximum variation from test to test in the titre obtained should not exceed one tenfold dilution when the same lot of challengevirus is used The lot of working CVS may be used for as long as full potency is maintained as shown by rnouse titration The period of storage should not exceed 2 years however it is recommended that the working CVS is used within 1 year

Reference vaccine

The fifth International Standard for Rabies Vaccine (B-propioactone-inactivated and freeze-dried) is available to national laboratories on request.' It was prepared

' iqrernational Laboratoiy for B~ological Standards, State Serum insrlrute 50Amager Boulevard DY-2300 Ccpenhage i Denmaik

361


in Vero cell culture and was establ~shed in 1991 with a deflried potency of 16 International Units ( U ) per ampoule ( I ) . When diluted in accordance with recommendations, the reference is eqcrivalent to 1 IU per ml.

Each ~ iat ional laboratory should prepare a national reference vaccine that is calibrated against the International Standard. The national reference preparation should be supplied to routine prodilction laboratories within the country. The national reference preparation should be an inactivated, treeze-dried vaccine that is routinely nont to red and cornpared with the International Standard. National reference preparations should be replaced when any departure from the lnlerna- tonal Standard occurs (usually after 4-5 years).

The reference vaccine is reconstituted (usually with distilled water) to a final potency of 1 IUjmI and divided into two portions. The first portiori is held in an c e - water bath or equ~valent and IS used, after dilution, for the ~ n i t ~ a l vacc~nation. The second portion is immediately froien and held at - 70'C for use in the preparation of dilutions at the time of the second immunization.

Immunization of mice

For inoculation o i mlce both the test vaccine ar,d the adjusted reference vaccine are diluted as follows.

1. Prepare three or more (usually five) fivefold dilutions (e.g. 1 5, 1 .25, 1 :125, 1 :625 and 1 :3125) of the vaccinejs) under test in phosphate-buffered saline (PBS) pH 7.6 (see Chapter 20, Annex 2).'

2. Prepare four fivefold dilutions of the reference vaccine in PBS, pH 7.6 (e.g. 1 .10, l :50, 1 :250 and l : 1250). The starting dilution of the reference vaccine will depend on the strength of the challenge dose and the potency of the vaccine under test.

3. Inject groups of at least 16 mice, weighing 13-16 g, intrapertoneally with 0.5 m1 of each dilution of the test vaccine or reference vaccine. The vaccine suspension should be injected rapidly into the posterior portion of the peritonea1 cavity (Fig. 3 1 . 1 ) away from the midline. using a 0.50-mm X 16-mm (25-26-gauge) needle Each mouse should be given two doses of vaccine 1 week apart. Set aside enough control mice for an adequate titration of the challenge virus to be made w ~ t h at least 10 mice for each dilution of virus (a total of 30-40 control mice)

A different needle and syringe should be used for lnoculat~ng each group of mice However if supplies dre I m ~ t e d a single needle and syringe may be used for inoculating the test vaccine or reference vaccine In that case mice receiving the most dllute vaccine stiould be inoculated first followed by those receiving successively mote concentrated vaccines Mice receiving different vaccine concentrations should be housed separately

Challenge of control and test mice

All mice are challenged intracerebrally 14 days after the first dose of vaccine as follows.

' Tilt pH stlould be adjusted to i 6 Irslng oilb!c sod~urn rvdroxlde (NaOHI

NlH TEST FOR POTENCY

Fig. 37. I lntraperitoneal inoculation of a mouse

1 Thaw an adequate number of ampoules of the pooled working CVS (usually one) rapidly under cold running water.

2. Dilute the supernatant in CVS diluent to contain between 12 and 50 LD,, of virus per 0.03 m1 (recommended titre 25 LD,,). based on previous titrations Prepare three terifold dilutions (10 l , 10-' and I O - ~ ) of the diluted supernatant in CVS diluent to condiict a titration of the working CVS. It is recommended that the dilutions of the challenge virus be held in an ice-water bath, or its equivalent. both before and during the test in orderto prevent loss of potency

3. Challenge the immunized mice intracerebrally with 0.03 ml of the dilution containing between 12 and 50 LD,,. Inoculate the control mice intracerebrally with 003 m of each dilution of the challenge virus. It is preferable to use a different syringe for each dilution of the challenge virus; however, if only one syringe is used, the 10-3 dilution must be injected first. followed by the 10-2 dilution and then the 10-' dilution.

4. Observe the mice for 14 days and record the number that die from rabies after the f~rst 5 days. Include any mice showing signs of fixed-virus rabies (paralysis, convulsions) on the 14th day.

The definitions of "paralysis" and "convulsions", as applied to mice following injection of the challenge virus, are as follows:


Paralys~s is the partial or complete loss of motor power of oine or more legs Convulsions are indicated by violent and abnormal muscular contractions of the body-often termed spasms These are brought about by external stimulation. such as a disturbing sound or handling

Calculation of potency

For the test to be considered valid, the results obtained after challenge of the immunized mice must show that the dilutions of reference vaccine encompass the 50% end-point - that is, at least 70°/6 of the mice receiving the lowest dilution (highest dose) of vaccine survive and at least 70% of those receiving the highest dilution (lowest dose) die. The virus titre of the challenge virus must fall between 12 and 50 LD,,, In practice, tile LD,, of the challenge virus should be determined by calculating the mean of several values obtained in successive tests. It is not worthwhile attempting to readjust it from test to test.

A volumetric method of calculation of potency should be used, This compares the 50°% end-point dilution (vaccine dilution protecting 50% of mice) of the vaccine under test w t h that of the International Standard (or equivalent national reference vaccine). The relative potency (RP) of the vaccine under test is determined by the formula:

reciprocal of ED,, of TV dose of TV RP = X

reciprocal of ED,, of RV dose of R V

where:

TV = test vaccine. RV = reference vaccine. dose = volume of a single vaccinal dose, as stated by the producer

For example i f the ED,, of the test vaccine is 1 90 and that of the reference vaccine is 1 70, the reciprocal values will be 90 and 70 respectively If it is assumed that a single human dose of the vaccine under test is 2 rnl and that 1 m of the reference vaccine represents a single dose for humans then

Minimum potency requirements

The relative potency of rabies vaccines for animals should be determined using a recognized rabies reference vaccine and the batch of rabies vaccine used in a valid vaccination challenge test In the target species. The test should be carried out at the end of the period of immunity claimed by the vaccine producer. The relative potency value obtained in the NIH test should become the minimum value for all subsequent batches of the vaccine.

At its eighth meeting, the WHO Expert Committee on Rabies (2) suggested that inactivated veterinary vaccines with a potency of less than 1.0 IU per dose, as measured by the NIH test, should nol be licensed or released unless an adequately designed experiment has demonstrated a duration of immunity of at least 1 year in


the species for which the vaccine is to be used The Committee recommended that h~ghly purified rnodern rabies vaccines for human use should have a minimum potency of 2 5 IU per dose (3, 4) and that suckling-mouse brain vaccines for hurnari use should have a rnir?imurr? potency of 1 3 IU per dose (5) , regardless of the number of doses required for full post exposure treatment

Modified NIH test'

This test is a simplification of the N H test The airn of t t is test is to delernline

whether a rabies vaccine satisties the minimum potency requirement without assigning a precise value to it The test prov~des qualitative results It is ~ a r r i e d out

in a laboratory and requires a homogenous stock of challenge virus well standardized methods of titration and laboratory animals of constant quality (consistent response to the vaccine and the challenge virus)

This test is particularly useful for testing multiple batches of a vaccine within a short time and for reducing the number of mice used in the NIH test However owing to its lack of precision the niinmurn potency requirement for vaccines tested by the modified test is higher than that forvaccines tested by the standard NIH test

Method

Before performing the modified test a laboratory rnust have titrated the reference vaccine (fifth International Standard for Rabies Vaccine or a national reference vaccine calibrated against the International Standard) several times in order to determine its ED,, When the weighted mean of the ED,, of the reference vaccine has been calculated the theoretical ED,, for a vaccine of the required potency can be determined using the forniula

where:

D = the minimum ED,, required for the vaccine uiidei test (decimal logarithm of the Inverse of ttie arithmetical dilt~tion)

D,,, = the weighted mean of the ED,,s obtained with ttie reference vaccine n = the required potency of the vaccine under test (IU ml) N = the potency of the reference vaccine (IU ml) This information is provided in

the insert supplied with the vaccine v = the volume (ml) of a s~ngle human dose of the vaccine under test as stated by

the manufacturer

This diluiron is then used for the vaccines under test. The standard NIH test protocol is followed. uslng 10 rnce vacciriated with the theoretical ED,, of the vaccine under test. To satisly the minlmum potency requirement. at least 8 of the 10 vaccinated mice should survive after challenge.'

'Th is sectlon has been prepared by M t A Aubert. ' I f one modse d e s before chaiie?ge the vaccive map ae cons~dered to sausfy the mlnmum potency

requlre-le?r provide0 that H ot the 9 re ra in ing m c e su,vive after challenge


Example

Suppose that after three independent titrations of the current International Standard fo i Rabies Vaccine, the mean ED,, is found to be 1.05. Since the potency of this standard is defined as 16 IU per ampoule and it is reconstituted with 8 ml of distilled water per ampoule, 1 m1 of the reconstituted vaccine will contain 2 IU. Thus. D, = 1.05 and N = 2.0. Assuming that the minimum requirement for the vaccine under test is 1 IU per m1 and the volume of the prescribed dose is 2 ml, the minimum ED,, required for the vaccine under test is calculated using the above formula as follows:

Therefore the dilution to use for the vaccine under test is 5 m1 of vaccine and 9 ml of diluent Assuming that the t t ra ton of challenge virus and the titration of the reference vaccine give results within the usual range the vaccine under test satisfies the minimilm potency requirement if at least 8 out of 10 of thevaccinated mice survive challenge

I f a more precise evaluation of the potency of the vaccine is required a diagram (Figs 37 2 and 37 3) may be used to determine the 95% contidence interval For this evaluation the mean slope of the vaccine t t ra ton curve obtained from probit analysis must be known This mean should be determined in each laboratory for each given type of vaccine to be tested However the range of slopes given in F ~ g s 37 2 and 37 3 covers all the values usually observed in laboratories

In the above example suppose that 9 of the 10 vaccinated mice survive challenge and that the mean slope of the vaccine titraton curve is 2 8 Then according to Figs 37 2 and 37 3 the potency of this vaccine will be between 1 3 IU and 8 3 IU (95% confidence interval)


Fig. 37.3 Determining the lower 95% confidence limit of the potency a vaccine by the modified NIH test

5.0

4.5

4.0

2 3.5 C .-

g 3.0 .- 8 3 2.5 - 0 X g 2.0 Q) +

2 1.5

1 .o D

0.5 2 0 0.7 1.0 1.3 1.6 1.9 2.2 2.5 2.8 3.1 3.4 3.7 4.0 4.3

Mean slope of vaccine titration curve

Statistical background The modified NIH test IS based upon binoniial distribution In this test the probability (P) of a given number ( s ) out of 10 mice surviving challenge IS

calculated using the following formula

where.

p = thp theoret~cal proportion of mice protected by the vaccine (varies according to its potency)

At dilut~on D as d e f ~ n ~ d above the probability that a vaccine with a potency below the minimum reqiiirement would protect more than 7 out of 10 vaccinated mice against challenge is less than 005 hence the limit for this test The above figures are established using the binom~al distribut~on and the probit analysis equation

y = Slog,, ( { l ) c 5

where:

y = probit S = rnean slope of the vaccine titration curve deterinined froni probit analysss n = the requreo potency of the vaccine under tesi ( n IU ml)


References

1 WHO Expert Committee on Bio!ogical Standardization Forty-second report Geneva World Health Organizatoi? 1992 (WHO Technical Report Series No 8221

2 . WHO Expert Comrnittee on Rabies. Eighth report. Geneva, World Health Organization, 1992 (WHO Technical Report Series, No. 824)

3. WHO Expert Committee on Biological Standardization. Thirty-seventh report. Geneva. World Health Organization. 1987 (WHO Technical Report Series, No. 760).

4 WHO Expert Con7mittee on Biologica! Standardization Thirty first report Gen- eva, World Health Organ i i a ton 1981 (WHO Technical Report Ser~es No 658)

5 . WHO Expert Committee ot? Rabies. Seventh reporl Geneva, World Health Organizat ion 1984 (WHO Techrical Report Series. No. 709).

Annex Preparation of Challenge Virus Standard (CVS) diluent, pH 7.6

Horse serum (inactivated at 5 6 - C for 30 minutes) Benzylpericillin Streptomycin Sterile de-onized water to make

20 m1 500000 units

l g 1000 m1

Adjust the pH to 7.6 with 7.5% sodium bicarbonate (NaHCO,) and store at 4'C Check the pH nirnedately before use and adjust i f necessary.

CHAPTER 38

Habel test for potency K. Habel'

Standard test

The test consists of the intrdperitoneal inociilation of whale Swiss mice aged 4-6 weeks and we ghiing approximately 12 g with the vaccine followed by challenge with ihe Challenge Viriis Standard (CVS) strain of fixed rabies virus (see Chap- ter 37)

Immunization procedure

Sixty mice are ~noculated with 025 ml of vaccine diluted to give a 0 5 % wiv suspension of b r a n tissue When vaccines prepared from material other than b r a n tissue are tested the vaccine should be diluted 1 10 before inoculation The inoculations are given intraperitoneally on Monday Wednesday and Friday during 2 successive weeks (a total of six doses) Thirty mice should be kept apart from the rest at the beginning of the test for use as controls at the time of challenge

Challenge

A challenge test is performed 14 days after the first dose of vaccine Two ampoules of the CVS rabies strain should be thawed out and diluted to a 10- ' suspension Serial tenfold dilutions ( l ~ ~ ' - l ~ ~ ' ) are then made using 2% heat-inactivated horse or rabbit serum in distilled watei as diluent To prevent loss of virus titre the dilutions of challenge virus should be kept in all ice water bath during the test Care should also be taken to use a separate pipette for transferring and mixing the material at each step in the serial dilution series

Groups of 10vaccinated mice are chaleriged ir~tracerebrally with 0 03 m1 of the l G - 5 l G - 4 10 -? 10-' and 10- ' dilutions of virus in that order The same syringe and needle may be used for all inoculations provided the empty syringe is rinsed several times in the next dilution before being refilled

With a new syringe the control mice are then inoculated wilh ihe 10-' 10-6 and 10 ' dilutions of the virus in order to determine what dilution represents the median lethal dose or LD,, I f the challenge virus is fully active these three dilutions will usually give survival rates ranging from 100% to less than 50% Scattered deaths at all three dilutions should be regarded with suspicion as ~ndlcating errors in the dilution of virus For the potency test to be valid the 50% end-point dilution (corresponding to a mortality rate of 50%) should be beyond the lK5 dilution in the control mice

' Dccoased Former Research t e l o w Department o f Exper~mentai Pattiaiogy Scripps C i~n lc and Research Foundat~on La J o l a CA, USA


Following challenge the mice are observed daily for 14 days for signs of rabies Any deaths occurring after the first 5 days should be recorded When the potency is calculated any mice surviving at the end of 14 days but showlng definite sigris of rabies should be considered as having died from rabies

Determination of degree of protection

The 501'0 end-porit dilutions in vaccinated and control mice are determined by the methods described in the Annex anri in Appendix 3, page 445. These inethods give the d l ~ ~ l i o r ~ s at which, theoretically. 50% of the mice would die from rabies, the calculaiions beirig made on the basis of the actual results in the test with the dilutions employed. By subtracting the log of the 50% end-point dilution in the vaccinated animals from that in the controls, the log of the "LD,, of protection'' of the vaccine is obtained. To satisfy the mlnimum potency requirements, the difference stiould be 3 or log 1000, indicating that the vaccine confers protection against l000 LD,,

Modified Habel test

For many laboratories producing rabies vaccine, the need for relatively large numbers of experimental animals makes the rouiirie use of the standard Habel test for potency impractical, This is especially true in those laboratories where multiple small volumes of vaccine are prepared at frequent intervals. In these situations it would be of value to have available a simplified potency test that would screen out vaccines of below-standard antigenic potencies and would at the same time be comparable with the more complete standard test in respect of minimum requirements.

In ctiecking the results of many potency tests with vaccines of high, low and intermediate levels of antigenicity, it has been apparent that approximately 500 LD,, as an ntracerebral challenge dose kills less than half the immunized mice i f the potency of the vaccine determined by the complete test is over 1000 LD,, of protection, which is the minimum req~iirernenl

The modified test procedure is the same as the standard procedure (see above), using six intraperitonea doses of a 0.5% nerve-tissue suspension (0 75 m ) , and ntracerebral challenge on the 14th day followed by 14 days of observation. Only 20 mice are r i~munired. arid 15 additional rrice are held as controls. On the 14113 day the CVS strain (see Chapter 37), diluted to contain 500

LD,, per 0.03 m/, basea on past experience with a frozen virus pool, IS given iniracerebrally to the vaccinated mice, and at the same time groups of five control

mice are given the chalerige virus diluted to 10 5, 1 0 and 1 0 ' (see above). For the test to be valid. the control litre should indicate that the challenge dose

given to the vaccinated mice was between 100 and 1000 LD,,: at least 50% of the vaccinated n ice should survive in order that the vaccine pass the screening poteiicy rest (tor calc~ilation of rhe challenge dose, see Chapier 47, page 418)

It is recorrirriended that at periodic intervals a complete standard potency test (see above; see also Chapter 37) be performed to obtain a more quantitative

evaluation of the vaccine over long periods of production in each laboratory.

HABEL TEST FOR POTENCY

Annex Calculating 50% end-point dilutions by the method of Reed & ~ u e n c h '

Although there are reasons for preferring the Spearman-Karber method for calculating 50% end-point dilutions, the method of Reed & Muench is still in use in some laboratories. The prerequisites are the same as for the Spearman-Karber method: a constant number of animals per dilution. a constant dilution factor, a range of dilutions covering both 100% and 0% of positive reactors, and no accidental deaths. If accidental deaths do occur, preference should be given to the Spearman-Karber formula (see Appendix 3, Examples 1 and 2, page 446).

The starling point for the calculation of 50% end-point dilut~ons (LD,, titres) by the Reed & Muench method is the dilution showing a mortality next below 50% ("starting point dilution"), The formula given below is used to determine the difference between the logarithm of the starting point dilution and the logarithm of the 50% end-point dilution ("difference of logarithms"). If the mortality decreases with increasing dilution (as in titrations of virus suspensions), the 50% end-point dilution will be lower than the starting p o ~ n t dilution. The "difference of logarithms" has therefore to be subtracted from the logarithm of the reciprocal of the starting point dilution. On the other hand, the "difference of logarithms" has to be added if the mortatity increases with rncreasing d~lution. This distinction, which is illustrated in the following examples, must always be borne in mind when calculations are made.

Example I. Titration of virus suspension

Suppose that in a Habel test for potency the titration of the virus suspension in control mice gives the following results:

Serum No. of mice Cumulative totalsa Mortality dilution

Survived Died Survived Died

10-5 0 10 1 1: 17117= 100% 10 4 6 711 1 = 64% 10-5 9 l 1/14 = 7%

a Totals are accumulated frorn 10-"o lo- ' for sdrvvors and froni 10-' to lV5 for mice considered as hav~nq died of rabies

In thls example the dilution factor is 10 and the starting point dilution (the dilution showing a mortaliiy next below 50%) is I O - ~ .

Calculate the "difference of logarithms" from the formula:

50% - (mortality next below 50%) ... X

logar~thm of

(mortality next above 50%) - (mortality rnext below 50%) dilution factor

' Based on the chapter by R, J Lorerir and K Bogel in the previous edit~ori

371


Hence, the "difference of logarithms"

S~nce, in this example, the mortality decreases with increasing dilution, the 50% end-point dilution is lower than the starting point dilution and is calculated by subtracting the "difference of logarithnls" as follows:

log (reciprocal of 50% end-point dilution) = log (reciprocal of starting point dilution) - "difference of logarithms" = log 107 - 0.75 = 6.25

Hence, log (5094 end-point dilution) = - 6.25 and 50% end-point dilution (LDS0 titre) = 10-6.25.

Example 2. Titration of rabies vaccine

Suppose that in the N H potency test the t t ra ton of a rabies vaccine gives the following results:

Dilution of 5% Brain No. of mice Cumulative Mortality brain-tissue tissue totals vaccine (mg)


In this example the dilution facior is 5 and the starting point dilution' (showing a niorlality next below 50%) is 5- l .

Using the same formula as in the previous example, the "difference of logarithms" is:

' In some lexfboohs the starting point dlution for titrations of varcines and sera (increasing rnortali!y with increasirig dilutions) is !aken as the dilution showing a tnortaly nex! above 50''. The formula for the "diffeieiice of ogari!hrns' then becomes.

(~ iu r ta i i t y at dilution next above 50%) - 50°/~ -- - X logarithm of diiu!ion factor

(mottaity next above 50°4) - (mortality next below 50?/0]

If this forniula is used, the "difference of logarithms" has to be subtracted from the logarithm of the reciprocal of the starting point dilutiori.

HABEL TEST FOR POTENCY

Since in this case, the mortality is increasing as the d~lution increases, the 50% end-point dilution will be higher t h a n the start~ng point dilution and will be calculated by adding the "difference of logarithms" as follows:

log (reciprocal of 50°6 end-point dilution) = log (reciprocal of starting point dilutior~) + 'dfference of logarithms = 0699 + 0491--1 2

Hence, log (50% end-point dil i i ton) = - 1.2 and 50% end-point diluton =

10 ( = 1 :16).

Example 3. Titration of antirabies serum

Suppose that a typical protocol of the ttration of a therapeutic serum gives the following results:

Serum No. of mice Cumulative totals Mortality dilution


In this example the dilution factor is 2 and the starting point diluton (showing a mortality next below 50%) is I o - ~ .

Using the same formula as in the previous example, the "difference of logarithms'' is.

Since in this case the mortality is increasing as the dilution increases. the 50% end-point dilution will be hgher than the starting p o n t dilution and wiil be calculated by addng the "difference of logarithms" as follows.

log (reciprocal of 50% end-point dilution) = log (recprocal of starting point dilution) + "difference of logarithms" = 3 + 0.19--3.2

Hence, log 50% end-point di lut ion=-3.2 and 50% end-point dilution = 10-3.2.

CHAPTER 39

Guinea-pig potency test for chicken-embryo vaccine H. K o ~ r o w s k i '

The yuiriea-pig potency test consists of the intramuscular inoculation of guinea- pigs with chicken-embryo vacciiie, followed 3 weeks later by challenge of the animals with street or fixed rabies virus. Animals weighing not less than 350 g at the beginning of the test should be used.

Immunization procedure

1 . Reconstitute each of two vials of vaccine representing each series or subseries in the f~na l containers with sterile distilled water c o n t a ~ n n g 2% normal horse seruni to give a l ~ n a l volume of 3 ml (irrespective o i the manuiacturer's recommended dosage)

2. Combine the two doses of the reconstituted vaccine, withdraw a full dog dose (3 ml) and add this to 17 ml of d~st~ l led water containing 2% normal horse serum, This will yield a 5% tissue suspension.

3 Inoculate at least 10 healthy guinea-pigs with 0.25 m1 of the diluted vaccine The vacclne should be injected into the yastrocnemius muscle on the inside of the leg as near to the nerve as possible At the same tirne, five or more healthy gilnea-pigs are set aside as vaccine controls.

Preparation of challenge material

Street rabies virus

1 Inociilate at least thiee adult dogs (over 6 months of age) with 0 1 m of a 20% susoensiorl of infected canine submaxillary t~ssup prev~ousiy stored at - 70 C (see step 6) The suspensiori should be injected bilaterally into the masseter muscle using a 1 nil syringe and a 090 mm X 25-mm (20-gauge) needle (Fig 39 1) The an~rnals should be observed ior signs of rabies and Killed when they are completely paralysed

2 After remoi~ing the submaxillary ylana separate a small plece by cutting with sc ssors a r d s t o r ~ the remainder b e r w e ~ n - 50 C avd - 70 C Using a

mortar grind the selected portion with enough 10% normal rabbit serum n phys ologcal saline solution to make a 10n/~ susperison by weight

3 Centrifuge the suspension in an angle-head centrifuge for 1 minute at 150-200 g and separate the sopernatant

' Cice~!or C e m r for Neurovirologv Jefterso~? Cdrlcer n s t i t ~ t e . Thornas Je+fe:son Ui ivers i ty Philadelphia PA USA

374


Fixed rabies virus

The Challenge Virus Standard (CVS) strain of fixed rabies virus can also be used to challenge the giiinea-pigs The preparation of the pools of virus in mice is desciibed in Chapter 37 page 360

To increase the pathogenicity of this strain for gtiinea-pigs by the intramuscular route, a few ir-itracerebral passages in young gi lnea-pigs (150-200 g) can be performed and a pool of guinea pig-brain material used for challenge in place o i mouse brahn material

Challenge of the guinea-pigs

Three weeks after vaccination, the vaccinated guinea-pigs and controls are c h a - lenged by an intramirscular injection of canine street or fixed rabies virus.

The dosage used is 0.5 m1 of a suitable dilution of the virus suspension injected intramuscularly into the leg opposite that used for vaccination. The dilution is one twofold d l i i t ion lower than that which kills 100% of the guinea-pigs in a preliminary t i t raton: i.e.. if 100% of the guinea-pigs are killed with the 1 : 160 dilution. a 1 :80 dilution should be used as the challenge virus for the potency test.


A sample of results obtained with chicken-embryo vaccines that have successfully passed the guinea-pig potency test is shown in Table 39.1. It will be seen that the vaccine has proved of definite ant igenc value for dogs challenged with street virus. In most instances, dogs were protected against street-virus challenge with a vaccine preparation that immunized guinea-pigs in 5% chicken-embryo tissue suspensiori As mentioned previously, the challenge virus should be of sufficient potency io cause the death o f most, i f not all, control animals. It has been found that the titre in mice must be at least l ~ - ~ . ~ ~ LD50 for the challenge virus to meet this requirement

In the actual test. 80% of the control animals should die of rabies within 21 days after challenge with street virus and within 14 days after challenge with fixed virus; 70% of the vaccinated girnea-pigs, and in any case no fewer than seven animals. should survive challenge inoculation without showing any signs of rabies. This test very closely parallels events occurring in nature and can be performed with reat~ve ly little labour. Only one inoculation with vaccine and one challenge inocii lation are recluired. By employing locally isolated street-virus strains for challenge purposes, Ihe test can be used to evaluate rhe potency of chicken- embryo vaccines 111 different geographical areas of the world against different strains of rabies A disadvantage is that dogs are required for the preparation of challenge virus. However, if a potent preparation of salivary gland tissue is

employed, a relatrvely small number of dogs w11I suffrce to yield a prepara- tor1 that can be used for a long period for challenge purposes

Tests w t l i large batches of Flury LEP raoies vaccine in guinea-pigs, using a CVS virus for challenge of vaccinated animals, have demonstrated that results of potency tests r i guinea-pigs can be correlated with the intracerebral t i t raton of the virus in mice (see Chapter 6). The paralysis pattern that is observed 4-10 days f o l l o w ~ ~ ? g the ctiallenge correlates well with the virus titre of the vaccine and can therefore serve to facilitate the selection of potent vaccines

Table 39.1 Results of the guinea-pig potency test for chicken-embryo vaccine

Mortality ratios after intramuscular challenge with rabies street virus

Guinea-pigs Dogs --

immunized with the following dilutions of Batch LD50 ~ a c c i n e : ~ non- non- of titre in vaccinated vaccinated vaccine mice 1:s 1:20 1:80 1 :320 1 :l280 1 :5120 controls vaccinatedb controls

lOil0 two tests performed

315 013 / 10110

l0110 'l9 ' two tests performed 819 016 6/10

" Wlttl Oalctlcs A and D, 1 rnl of 5% :Issue susoension bvas used: wlth all other batches, 0 5 n was used (Note. 025 ml of 5% t lss~ic suspenslori 15; now used for r o ~ j l n ? ti!SllrlQ J

5 rr' of 20% :ISSUE suspenson were dsed for vacclnat~orl

CHAPTER 40

Single radial immunodiffusion test for the determination of the glycoprotein content of inactivated rabies vaccines M. Ferguson '

Principle

The glycoprotein content of inactivated cell-culture or purified avian-embryo rabies vaccines can be determined by comparing their reaction with specific anti- glycoprotein serum In the single radial immunodiffusion (SRD) test with that of a reference preparation.

In this test, the vaccines are initially treated with detergent to release the glycoprotein antigen from the rabies virus particles. Serial dilutions of the reference and test vaccines are then prepared and distributed into wells in agarose gels. The solubilized glycoprotein diffuses radially and reacts with an anti-glycoprotein antibody incorporated in the gel to produce a zone of precipitation. At equilibrium, the area of the diffusion zone has been found to be proportional to the amount of glycoprotein added ( 1 , 2). Preparations of low potency, such as vaccines derived from brain tissue, and preparations in which the rabies virus IS adsorbed to aluminium hydroxide, are unsuitable for assay by this method. The SRD technique is rapid and enables highly reproducible estimates of glycoprotein content to be made without the use of infectious virus.

Materials

Standard preparation: The fifth International Standard for Rabies Vaccine, which was established in 1991 ( 3 ) , or a national reference vaccine, which has been calibrated aga~nst the International Standard, is used.

Anti-giycoprotein serum: Antisera prepared by immunizing an~mals with purified rabies glycoprotein should be used in this test. Antigenic differences between rabies virus strains can affect potency values obtarned by SRD assays if the reference and test vaccines are heterologous (4) . Antiserurn homologous to the test vaccine should be used. Suitable antisera are available from the National Institute for Biological Standards and ControlZ (ERA or LEP strains) or the Center for Biologics Evaluation and Research3 (PM strain).

'Senior Scientts!. National institute for 81oIogicai Standards and Control South Mimms Potters Bar. Herts, England

'Nat~onal l r is i i !~ te for Biological Standards and Gon!roi. Sou!ti Mirnrns. Potters B a r Herts EN63QG, England

3 D ~ v ~ s ~ o ~ i of Product Quality Control, Center for Biologics Evaualion and Research, Food and Drug Adrnlnstratlon, Bethesda. MD 20892, USA

SRD TEST FOR DETERMINATION OF GLYCOPROTEIN CONTENT

Detergent Mulgofen BC-720 or Zwlttergent 3 14, diluted to 20% or 10% respectively is used as detergent

Diluent: Dulbecco's phosphate-buffered saline soliition A (PBSA; see Annex), pH 7.2, is used as diluerlt

Agarose l % agarose in PBSA containirig 0 0596 sodium azde should be used for preparing the gel

Method

1 Prepare agarose gels containing a suitable amount of ant-rabies glycoprotein serum on a level surface in Petri dishes or in wells on glass plates so that the depth of agarose is 1 5-2 Omm lnimune sheep serum has been found to be satisfactory when incorporated at a concentration of 3-5 ILI per rnl of gel (2) When the gels have set, cut out wells 4 mm in diameter far enough apart so that zones in neighbouring wells do not overlap For example if the zone of precipitation is 8-9 mm in diametei the wells should be 1 1 cm apart

2 Reconstitute the standard and test vaccines from the freeze-dried preparations following the manufacturers instructions

3 Add detergent to duplicate al~qi iots of the standard and test vaccines to solubilize the glycoprotein antigen and prepare two independent dilution series of at least four dilutions as indicated in Fig 40 1 For most vaccines suitable

Fig. 40.1 Flow diagram of SRD operations

Step 1.

Step 2.

Step 3.

Treatment with detergent

Preparation of dilution series

Preparation of SRD plates

Reference vaccine

A duplication

Test vaccine

SRD plate A

SRD plate B

WHO 95038


dilutions are 1 1 3 4 1 2 and 1 4 which can be prepared as shown in Table 40 1

4 Distribute 20-111 volurnes o l the standard and test vaccines from one dilution series into the wells of one gel and those fiom the secoiid series of dilutions into the wells of a second gel (see Fig 40 1) The positions of all vaccine dilut~ons should be randornized Cover the gels with lids and place on wet absorbent material in a plastic box to prevent the gels from drying out Leave at room temperature for at least 18 hours

5 Place the agarose gels in a container of PBSA for 24-48 hours to remove extraneous proteins

6 Cover the gels with filter paper and absorbent material press flat and leave to dry at 37'C

7 When the gels are dry stain using Coomassie blue stain (0 3% wlv In a mixture of 12Oh acetic acid and 29% niethano) and de-sta~n in the methanol-acetic acid mixture until the zones of precipitation are clearly d~stinguished from background staining (Fig 40 2)

8 Measure the diameter of the diffusion zones at the widest part in two d~rections at right angles to each other and calculate the square of the mean diameter

Table 40.1 Preparing dilutions of reference and fest vaccines for the SRD test

Antigen Dilution Volume ( p l )

Antigen Diluent

Reference or test vaccine 1 . l 100 0 3:4 150 50 1.2 100 100 1 ' 4 50 150

Fig, 40.2 Reaction zones from SRD assay for rabies glycoprotein

Dilution 1:l 3:4 1 :2 1 :4

Reference vaccine

, <

Test f?*

vacclne ,%L d ,; l ! I .- -

a ,,a 3 ! !

WHO 95039

380

SRD TEST FOR DETERMINATION OF GLYCOPROTEIN CONTENT

Fig. 40.3 Calculation of dose-response slopes

Reference vaccine

20 1 l l I I

1 :4 1 :2 3:4 1:l Antigen dilution

Potency of vaccine X = X X potency of reference vaccine r


The rnelhod used to calculate (he results will vary according to the degree of accuracy required and the statstical resources available.

The most simple method involves plott!ng the valiies of the square of the mean diameter ( d 2 ) agairist the dllulion of antigen for the refere~ce and test vaccines from each set of duplicate plates. The graphs are constructed either by eye or preferably by least squares linear regression analysis. Each d o s e response curve should be Inear and the lines should meet at a common tntercept on the d2-axs (F1g.40 3). The slope of the dose-response curves may be calculated by regression a ~ a l y s ~ s or by measuring the height of each curve above the mean Intercept. The slope of each curve 1s proportlonal to t s helgtit above the intercept, The glycoprotein content of each vaccine (in IUIml) is calcuaied as follows.

slope of oose response curve for test vaccirle p-- - X glycoproteir coi7tent slope of dose-response curve for reference vacclne

of reference vaccine (111 I U l m )

For a more accut.ate analyss, the results may be calculated usng standard slope-rato assay methods (4, 5). A parallel line model niay also be used provided


that the dose-resporise curves for the reference and tesi vaccines are linear and parallel over the range of doses used in the calculatior~ (5).

References

1 . Fergusor~ M, Sch~ld GC. A single radial irrirnunodiffusion technique for the assay of rabies glycoprotein antigen: application to potency tests of vaccines against rabies. Journal of general virology, 1982, 59. 197-201

2. Ferguson M. Seagroatt V, Schild GC. A collaborative study on the use of single radial immiinodiffusion for the assay of rabies virus glycoprotein. Journal of biological standardization, 1984, 12, 283-294.

3. WHO Expert Coinm~ttee on Bio/ogica/ Standardization. Forty-second report. Geneva. World Health Organization. 1992 (WHO Technical Report Series, No 822).

4 Ferguson M et a1 The effect of strain differences on the assay of rabies virus giycoprotein by single radial imniunodiffusion Journal of biological standardization, 1987 15 73-77

5. Finney DJ. Statistical methods in biological assay. 3rd ed. London, Griffin & Co. Ltd. 1978.

Armex Preparation of Dulbecco's phosphate-buffered saline solution A (PBSA), pH 7.2

Sodium chloride (NaCI) Sodium phosphate, dibasic (Na,HP04) Potassium phosphate, monobasic (KH,P04) Potassium chior~de (KCI) Distilled water to make

Sodium azide. 0.5?/0 soliition, may be added as preservative

l 0 0 g 14.5 g 2.5 g 2.5 g

10 litres

CHAPTER 41

Enzyme-linked immunosorbent assay (ELISA) for the determination of the glycoprotein content of rabies vaccines P. Perrin, ' M. Lafonz & P. Sureau3

Principle

The enzyme-linked immunosorbent assay (ELISA) is one of several ~n vitro tests that can be used for testing the potency of rabies vaccines ( 1 , 2). The principle of this test is based on the nieasurement of the virus glycoprotein content in rabies vaccine by ELISA using an indirect method (3) or an irnmunocapture method (4, 5). This chapter describes the technique using the imrnunocapture method.

The vaccine to be tested is incubated in microtitration plates previously sensitized with anti-glycoprotein polyclonal or virus-neutralizing monoclonal antibodies (6) . Bound antigens are subsequently identified by adding the same antibody labelled with peroxidase, which is revealed in the presence of substrate and chromogen. The glycoprotein content of the test vaccine isthen determined by comparing its absorbency with that of a reference vaccine.

Preparation of antibodies

Anti-glycoprotein polyclonal rabbit immunoglobuiins (PAb-Gs) are collected, purified and conjugated with peroxidase as described in Appendix 2. Anti- glycoprotein rnonoclonal mouse imniunoglobulins (MAb-Gs) are collected, semi- purified and conjugated with peroxidase as described in Chapter 11 and Appen- dix 2.

Precautions

The quality of the results depends on compliance with certain laboratory practices. If an automatic plate-washer is not available, the microtitration plates should be washed carefully, by distributing washing solution into each well, and then inverted on absorbent paper and left to dry after the last washing step. All reagents should be held at room temperature for 10 minutes before use. Vaccines or infected cell supernatants to be tested are often inactivated; nevertheless, all samples should be considered infectious and safety precautions observed (see Chapter 1). Before starting the test, a plan for distribution and identification of samples must be established. Both negative and positive controls should be

' Lyssavir lrs Labora!ory I1asteur Instltute. Paris France Laboratory ticad Rab~es IUnit IDepartrnent of Virniogy Pasteur iqstltute Paris France Deceased Fo r rne r .~ Raoics Unt Depaltinept of V~rologv, paste^., Instltute. Paris. France


included in each test to check for possible nonspecific reactions Dilutions of the reference antigen or vaccine and test samples should be prepared in tubes and not in the sensitized plate

When PAb Gs are used centrifugalion at 80000 g for 90 minutes and resuspension of the pellet in phosphate-buffered saline (PBS) (see Chapter 20 Annex 2) are required to eliminate soluble glycoprotein When MAb-Gs are used it must have been shown that ( I) the MAbs neutralize the virus used for production of the test vaccine (11) the glycoprotein of the product-speciiic reference vaccine (PSRV) binds to the MAbs It is recommended that the PSRV used in the test and the vaccines under test are prepared with the same rabies strain

Sensitization of microtitration plates

1. Sensitize the microtitration plates by adding 200 111 of an appropriate dilution of antibodies (PAbs or MAbs) in carbonate buffer, 0.05 mol/l, pH 9.6 (see Chapter 9, Annex). to each well. There should be approximately 1 p g of antibody per well.

2. lncubate the plates for 3 hours at 37:C in a humidified incubator (or seal with adhesive film).

3. Aspirate the contents of each well into a recipient containing 5% sodium hypochlorite (bleach) solution. lnvert the plates and leave to dry on absorbent paper for 5 minutes at room temperature.

4. Fill each well with 300 pl of 0.3% bovine serum albumin (BSA) and 5% sucrose dissolved in carbonate buffer, Incubate the plates for 30 minutes at 37°C.

5. Aspirate the contents of each well as above, Invert the plate and leave to dry on absorbent paper for 1 minute at room temperature. The plates should be used immediately or sealed and stored at - 20°C until use. If they are stored for over 3 months, however, they should be tested before use.

Assay procedure

1 . Fill the wells of the sensitized plate with PBS-polysorbate 20 buffer, pH 7 (washing solution: see Chapter 9, Annex).' Invert the plate and leave to dry on absorbent paper for 1 minute at room temperature. Repeat the procedure 4 times.

2. Prepare a blank control for the photometric readings by adding 200pl of PBS-polysorbate 20-BSA buffer, pH 7 (see Chapter 9, Annex)' to the first well ( IA ) of the plate (in some autoniatic plate readers, a blank control should be made in all the wells of the first line).

3. Add 200 pl of serial twofold dilutions of the reference vaccine (PSRV) in PBS-polysorbate 20-BSA, pH 7, in duplicate to the wells of the second and third lines. The lowest dilution should contain about l pg/ml of rabies glycoprotein.

4. Distribute 200111 of serial twofold dilutions of each vaccine sample in PBS-polysorbate 20-BSA, pH 7, in duplicate, into the remaining wells. Seal the plate with adhesive film and incubate for 1 hour at 37 "C.

'The pH should be adjusted to 7 0 Llsing d~lute hydrochior~c acid ( H C )

384

ELlSA FOR DETERMINATION OF GLYCOPROTEIN CONTENT

5. Remove the film and aspirate the contents of each well into a recipient containing 5% sodium hypochlorite (bleach) solution.

6. Repeat step 1 . Distribute 200111 of an appropriate di iut~on of peroxidase- labelled antibodies in PBS-polysorbate 20BSA, pH 7, into each well. Seal the plate and incubate for 1 hour at 37°C.

7. Repeat step 5. Wash the plate 6 times as above. 8 Add 200 p1 of substrate-chromogen solution (see Chapter 9. Annex) to each

well. Seai the plate with adhesive film and incubate in the dark for30 minutes at room temperature.

9. Add 50 ~ L I of stopping solution (4 mol,'l sulfuric acid) to each well. Do not expose the plate to direct light before r e a d ~ n g

10. Carefully wipe the bottom of the plate and place it in the spectrophotometer. Determine the optical density (OD) at 492 nm of the blank(s), reference vaccine and samples.

Fig. 41.1 Determining the glycoprotein content of a rabies vaccine by ELBA using PAb-Gs

t - OD PAb

a .- (I) S

8 0.8

l 10 100 1000 Glycoprotein content of PSRV (nglml)



The blank and the negative control should appear colourless, while the reference vaccine should be yellow-orange. The OD of the blank should be subtracted from that of the contr-ols and samples. The glycoprotein content is then determined by comparing tlie OD values of the sample with those of the PSRV. In the examples shown in Figs 41.1 and 41.2, the PSRV consists of purified viral particles (PV strain) with a glycoprotein content of 10 ng/ml (determined by a calorimetric method (7) for total protein titration arid by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) for the determination of glycoprotein percentage (8)) The OD (mean of each duplicate) of the PSRV is recorded according to the different concentrations (or dilutions) (see Figs. 41.1 and 41.2). The glycoprotein content of the saniples can then be extrapolated directly from the graphs obtained for the PSRV.

Fig. 41.2 Determining the glycoprotein content of a rabies vaccine by ELlSA using MAb-Gs

- OD MAb

1.8

Glycoprotein content of PSRV (nglml)

386

ELISA FOR DETERMINATION OF GLYCOPROTEIN CONTENT

For example, in the assay using PAbs (Fig. 41.1), a PV vaccine sample diluted 8-fold w h ~ c h exhibits a mean OD of 1 .I20 contains 280 X 8 = 2240 ngjnil of glycoprotein.

Similarly, in the assay using MAbs (Fig. 41.2). a PV vaccine sample diluted 16-fold which exhibits a mearl OD of 0.950 contains 9 8 x 16 = 1568 ngiml of glycoprotein.

Evaluation of in vitro potency

For vaccines prepared with the PV strain of rabies virus, a correlation has been demonstrated between the glycoproteiri content measured with the ELISA using PAb-Gs (8, 9) or neutralizing MAb-Gs (4, 5) and the protective activity measured with the NIH test (10). Under such conditions, the in vitro potency can be determined by the ELISA and the results expressed as equivalent International Units per m1 (EIU/ml).

However, in vitro potency evaluation is only valid i f three conditions are fulfilled:

- the PSRV has been previously tested by the NIH test for its in vivo potency; - a satisfactory correlation has been previously deinonstrated between the NIH

test and the ELISA test results: - the non-immunogenic soluble glycoprotein has been removed by ultracentrifu-

gation (see Chapter 14) or by using a neutralizing MAb in the ELISA.

References

1. WHO Expert Committee on Rabies. Seventh report. Geneva, World Health Organization, 1984 (WHO Technical Report Seres, No. 709).

2. Report of the Joint WH0,'German Green Cross informai Discussion on Bat Rabies in Europe, Marburg, 5 4 May 1986. Geneva, World Health Organiza- tion, 1986 (unpublished document WHOjRab. Res.,/86.24; available on request from the Division of Communicable Diseases, World Health Organization, 121 1 Geneva 27, Switzerland).

3. Atanasu P, Perrin P, Delagrieau J Use of an enzyme immunoassay with a protein A for rabies antigen and antibody determination. Deveiopments in b~oiogicai standard~zation, 1980. 46: 207-21 5.

4. Lafon M et al. Use of monoclonal antibody for quantitation of rabies vaccine glycoprotein by enzyme immunoassay. Journai of bioiogicai standardization. 1985. 13: 295-301.

5. Perrin P, Morgeaux S, Sureau P. in v~t ro rabies vaccine potency appraisal by ELISA: advantages of the imrnunocapture method with a neutralizing anti- glycoprotein nionoclonal antibody Bioiogicais, 1990, 18: 321-330.

6. Libeau G, Lafon M, Rollin P. Etude de la specficite d'anticorps rnonoclonaux obtenus avec la so~iche de virus de rage Pasteur PV. [Specitcity of monoclonal antibodies obtained with the Pasteur PV rabies virus strain.] Revue d'eievage et de m6dicine veterinaire des pays tl-opicaux, 1984, 37: 383-394.


7. Lowry D et al. Protein nieasurement with Foliri phenol reagent. Journai of biological chemistry, 1951. 195 265-275.

8. Adamowicz P et al. The use of varlous irnrnunochernica!, b~ochemica! and biological methods for the analysis of rabies virus production in tissue cultures. Developments in biological standardizat~on, 1984, 55: 191 -1 97.

9 Atanasiu P et al. Titrage immunoenzymatique de la giycoproteine. une technique ~n v~iro pour l'appreciation de I'actvite des vaccins antirabiqiies [lm- munoenrymatic titration of gycoprotein, ari i r i vitro technique for determining the aciivity of rabies vaccines.] Journal of biologicai standardization, 1982, 10: 289-296.

10. Selgi-iiann EB, Jr. The NIH test for potency In. Kaplan MM, Koprowski H, eds. Laboratory techniques in rabies. 3rd ed. Geneva. World Health Organization, 1973 (WHO Monograpti Series. No. 23) 279-286

CHAPTER 42

The Essen-ELISA for the determination of the glycoprotein content of inactivated cell-culture rabies vaccines 0 Thraenharf '

Principle

A non-competitive enzyme-linked imrnunosorbent assay (the Essen-ELISA) has been used to assess the glycoprotein antigen content of inactivated cell-culture rabies vaccines ( 1). The essential steps of the Essen-ELISA are'

- - fixation of the rabies virus in different dilutions of vaccine on microtitration plates;

-incubation with polyclonal antisera to rab~es vlrus glycoprotein; -further incubation with a species-specific irnrnunoglobulin G (IgG)-peroxidase

conjugate; i n c u b a t i o n with either 22'-azino-di-[3-ethylbenzthiazolinsulfonate] (ABTS) or

phenylenediamine.

The test is used for estin~ating the potency of inactivated cell-culture rabies vaccines of different types, vaccines with adjuvant and in-process samples such as cell-culture supernatants containing live or inactvated rabies virus, and for testing the stability of rabies vaccines at different temperatures. It is highly specific and provides a quantitative estimate of the glycoprotein antigen content of inactivated rabies vaccines. Potency estimates obtained by the Essen-ELSA have been shown to be correlated with the NIH test results. The lower Iirn~t of detection of the Essen-ELISA is 0.015 lU/ml. The test can also be used to estimate the nucleocapsd antigen content of rabies vaccines. Studies are under way to modify the test so that rnonoclonal antibodies can be used instead of polyclonal antisera.

Method

Preparation of antisera

Polyclonal antisera directed against the rabies virus glycoprotein antigen of the ERA or LEP strains are produced in sheep, while antisera directed against the rabies virus glycoprotein and nucleocapsid antigens of Pasteur vlrus are produced in rabbits. The antisera are diluted in phosphate-buffered saline (PBS), pH 7.2 (see Chapter 20, Annex 2),' supplemented with 8% fetal calf serum and 0.2% tyloxapol (serum dilution buffer).

' Academ~c Director WHO Colaborat~ng Centre for Reference and Research on Neurolog~cal Zoonoses lnstltute for Medcal Vlrology Ur l~vc rs l y of Essen Essen Gerniany

'The pH should be adjusted to 7 2 usng diiule sodicim hydroxide (NaOH)


Preparation of conjugate

A species-specific anti-lgG-peroxidase conjugate is diluted with PBS to an appropriate dilution (see below). Alternatively, the conjugate IS prepared as described in Append~x 2. page 438.

Standard for rabies vaccine

The fifth International Standard for Rabies Vaccine, which was established in 1991 with a defined potency of 16 IU per ampoule (2) . or a national reference vaccine which has been calibrated against the International Standard, is used.

Determination of the optimal dilutions of the anti-IgG-peroxidase conjugate

1. Dilute the reference vaccine in bicarbonate buffer, pH 9.6 (see Chapter 9, Annex), to a potency of 1.0 U per m1

2. Distribute the vaccine in 50-ALI aliquots into the wells of a 96-well microtitration plate. Place the plate in an incubator at 37'C for 18 hours to dry

3. F ~ l i the wells of the plate with wash buffer. Invert the plate and leave to dry on absorbent paper for 1 minute at room temperature. Repeat the procedure 4 times.

4. Add 50jt l of serial twofold dilutions (1 :loo-1 :51 200) of the antiserum to the wells of lines 2-1 1. lncubate the plate at 37 "C for 1 hour.

5. Aspirate the contents of each well into a recipient containing 5% sodium hypochlorte (bleach) solutlon (see Chapter 1) Repeat step 3

6 Add 50 ~ t l of serial twofold dilutions (1 100-1 3200) of the conjugate to the wells of rows B-G lncubate the plate at 37°C for 1 hour

7 Repeat step 5 Add 150 j ~ l of phenylenediam~ne or ABTS solution (see Annex) to each well lncubate the plate in the dark for 30 minutes at 37'C Alternatively add loop1 of ABTS to each well and incubate the plate in the dark for 10 minutes at 37 "C

8. Carefully wipe the bottom of the plate and place it in a spectrophotometer. Determine the optical density (OD) at 405 nm of the blank and the different dilutions of the antiserum-conjugate solution.

To determine the optimal dilution of the conjugate or antirabies serum, dose-response curves are drawn for each d~lut ion of the conjugate or serum in a coordinate system where X = the logarithm of each dilution and y = the logarithm of the corresponding OD value multiplied by 1000. The optimal dilution is the dilution for which the resulting curves show linearity and a slope of approximately 50". The initial OD value should not be less than 0.7.

Assay procedure

1. Prepare a blank control for the photometric readings by adding 50 lt l of PBS to the first well (row A) of the plate (in some automatic plate readers, a blank control should be made in all the wells of the first line).

ESSEN-ELISA FOR DETERMINATION OF GLYCOPROTEIN CONTENT

2 Add 50 p1 of serial twofold dilutions (l 16-1 256) of the reference and test vaccines in bicarbonate buffer pH 9 6 in duplicate to the remaining wells of the plate Up to six different vaccines may be tested on one plate Place the plate in an incubator for 18 hours at 37 C to dry

3 F111 the wells with PBS Invert the plate and leave to dry for 1 minute at room temperature Repeat the procedure 4 times

4 Add 50 /cl of the optimal dilutiori of the rabies antiserum to each well lncubate the plate for 1 hour at 37°C

5 Aspirate the contents of each well into a rec~pient containing 5% sodiuni hypochorite (bleach) solution Repeat step 3

6 Add 50 / L of the optimal diluliori of the species specific anti IgG conjugate to

each well lncubate the plate for 1 hour at 37 C Repeat step 5 7 Add 150 pi of phenylenediamine or ABTS solution (see Annex) to each well

Incubate the plate in the dark for 30 niinutes at 37 C Alternatively add l 0 0 j~1 of ABTS to each well and incubate in the dark for 10 minutes at 37 'C

8 Place the plate in a spectrophotometer Determine the OD at 405 nm of the blank(s) and the different dil~ltions of the reference and test vaccines

Evaluation of results

The relative potency of each test varcine is calculated by the parallel lhrie b~oassay (PLBA) technique using a computer program kriown as Parin 8 after transformation of the recaprocai values of the dilutions and the optical density into logarithms This transformation IS carried out to meet the criteria of homoscedas- tcity (defined as constancy of the variance of a measure over the levels of the factors under study) (3) The computer program also includes determination of the fiducial limits of the relative potencies and a complete variance analysis of the regression However i f a computer with this program and a Fortran interpreter are not available, a detailed analysis can be easily carried out using a pocket calculator (3)

Alternatively the relative potency of a vacclne may be calculated using a standard curve (see Fig 42 1) The graph is based on the results obtained by the Essen-ELISA with a reference vaccine (R) and a test vaccine (A) using glycoprotein antiserum (see Table 42 l ) The X- and y-coordinates of the curve are calculated using the following formula

X = log,,(lUx1000) (where IU = potency in International Units of the reference vaccine at a given dilution)

and y = log,,(OD X 1000) (where OD = optical density)

The potency of vaccine A is determined by drawing a horizontal line from the y-axis at a point equal to the OD value obtained with a dilution of the vaccine For example the mean OD of the 1 16 dilution was 0581 giving a y-axis value of log,, (0 581 X 1000) = 2 76 (a, ) From the point where this line crosses the dose response curve of the reference preparation a vertical line is drawn and the log,, (IU X 1000) value lead from the x-axis (a, = 2 4 ) The potency of the


Fig. 42.1 Determination of the relative potency of a test vaccine (A) using a standard curve

3.2 r

l l l

l l A

l

a2 b2 1.8 - I I i z I ii -

2.8 2.5 2.2 1.9 1.6 log,,, (potency (in IU) of reference vaccine at a given dilution X 1000)

Table 42.1 Potency estimates of a test vaccine (Ay and a reference vaccineb determined by the Essen-ELISA using antiserum directed against the gYycoprotein antigen of the ERA strain of fixed rabies virus

Vaccine Dilution

Reference 1.217 0.71 7 0 439 0.181 0.093 1211 0.713 0.432 0.187 0.082

A 0.550 0.305 0.154 0.079 0 062 0612 0.280 0.159 0.1 14 0.079

'Dog vaccine prepared w l t i SAD virus. ' F IU :~ LEP vacctne ca!brated a g a ~ s t t i e ln:ec?at~oPal Staldard

ESSEN-ELISA FOR DETERMINATION OF GLYCOPROTEIN CONTENT

1 : 16 dilution is therefore 0.25 IU and that of the undiluted vacclne is 4.0 IU (16 X 0.25 IU). This value is also obtained with the OD values of the 1 :64 dilution of vaccine A: b, = log,, (0.157 X 1000) = 2.19 and b, = log,, (IU X 1000) = 1.8. The potency of the 1 : 64 dilution is therefore 0.063 IU and that of the undiluted vaccine is 4.0 IU (64 X 0.063 IU)

References

1 . Thraenhart 0, Ramakrishriari K Star?dardization o l an enzyme imrnunoassay for the in vi lrn potency assay of inactivated tissue-culture rabes vacclnes. determination 01 the rabies virus glycoprotein with polyclonai antisera. Journal of bioiogi'cal standard!zation. 1989, 17. 291 -309

2 . WHO Expert Committee on B~ological Standard!zation. Forty-second report. Geneva, World Health Organization. 1992 (WHO Technical Report Series, No. 822).

3. Fnrley DJ. Statisticalmeti~ods !n biolog~calassay. 3rd ed London. Griffin & Co. 1978.

Annex Preparation of buffers and reagents

ABTS solution, pH 4.2 Acetic acid Sodunl acetaie Distilled watet to make plus Sodum phosphate rnorooasic monoi.iydrate(NaH,P04~H,0) 6 o Y plus 2,2'-A71no-di[3-ethylhen~thiarolinsulfonate] (ABTS) 1 097 g

Mix together the sodiuin acetate in 240 m1 of distilled water and acetic acid in 760 rnl of distilled water and then add sodium phosphate. Stir until dissolved and then add ABTS.

Tyloxapol Hydrogen peroxide Tyloxapol PBS to make

CHAPTER 43

The modified antibody-binding test for in viiro quantification of rabies virus antigen in inactivated rabies vaccines R Barth'

Principle

The quantification of rabies virus antigen has been of great importance to those who are concerned with the manufacture or control of inactivated rabies vaccines. Numerous contributions have been made by many workers and many different methods have been described. The antibody-binding tesi (ABT) was developed by Arko et al. (1) and subsequently tnodfied by Barth et al. (2, 3). The test involves three steps:

1. Serial dilutions of the inactivated rabies vaccine are incubated with a defined concentration of rabies virus-neutralizing antibodies.

2. The dilutions of vaccine in which the inactivated antigen is completely or partially ineutralized by the concentration of virus-neutralizing antibodies are indicated by adding a defined concentration of live indicator virus. After a time interval required for neutralization of the indicator virus, the resulting mixture is inoculated into a suitable cell-culture suspension grown in microt~tration plates.

3. After incubation. the plates are stained using the fluorescent antibody staining method (see Chapter 7). The dilutions in which the inactivated antigen IS

completely neutralized by the rabies virus-neutralizing antibodies (step 1) are ident~fied by the growth of indicator virus, whereas those in which the antigen is only partially neutralized or is not neutralized are identified by the limited or total inhibition of growth of indicator virus. The results are then used to calculate the 50% end-point dilution of the test vaccine.

The procedure described below refers to the ABT that is used at the Behring Institute

Method

1. Prepare a ctiick-ernbryo cell suspension containing 1 . 0 ~ 106 cells/ml and distribute in 200-p1 amounts into each well of a sterile disposable 96-well microtitration plate. lnciibate the plate for 24 hours at 37-C in a humidified incubator with 5% CO,.

2 PI-epare ser~al twofold d l u i o n s of ihe vaccines or artigeris under test and a relerence vacclrie uslng Eagle's basal medium (EBM, see Chapter 18, Annex),

' Former t icad. Rables Vacci re Developmerit 3rld Produc!ior Behr1.9 1ns:tu;e M a r b ~ ~ r g . Gervsny

394

MODIFIED ANTIBODY-BINDING TEST

pH 7.2, supplemented with 0.3% human albumin. The potency of the reference Vaccine should have been previously calibrated against that of the fifth lnterna- tiorial Standard for Rabies Vaccine (4).

3. Inoculate 2 ml of each dilution of the test and reference vaccines into test tubes containing 2 m1 of rabies virus-neutralizing antibodies (adjusted to a concentration of 0.2 IUlml). Include a negative control (2 n?l of the lowest dilution of the test and reference vaccines mixed with 2 ml of normal rabbit serum). Mix and place the test tubes in a water bath at 37°C for 60 minutes.

4. Inoculate 2 ml of indicator virus (Flury HEP K34, diluted to a concentration of 5000 TCID,,/mI) into each test tube. Include both negative and positive controls. The negative control consists of 2 m1 of indicator virus (5000

TCID,,/mI) mixed with 1 ml of rabies virus-neutralizing antibodies (0.2 IUIml) and 1 ml of EBM, while the positive control consists of 2 m1 of indicator virus and 2 rnl of EBM. Mix and return the test tubes to the water bath for a further 60 minutes

5. Inoculate 200 y l of the resulting mixture (vaccitie + virus-neutralizing antibod~es + indicator virus) into each of 8 wells of the lnicrotitration plate. Repeat the process for each dilution of the test and reference vaccines and the controls. Incubate the plate at 37-C for 72 hours in an incubator with 5% CO,.

6. Aspirate the supernatant from each well. Rinse the well with phosphate-buffered saline, pH 7.2 (see Chapter 20, Annex 2),' and leave to dry at room temperature.

7. Fix the cells with acetone diluted in ethanol ( 1 : 1). and stain with fluorescein isothiocyanate (FITC)-labelled antirabies immunoglobulin (see Chapter 9).

8. Observe the plates under a low-power fluorescence microscope. Score the number of wells containing one or Inore fluorescent foci.


The 50% end-point dilution of the test and reference vaccines is determined by recording the highest antigen dilution at which 50% of the observed wells contain one or more fluorescent f o c ~ The glycoproleln content of the test vaccine (in IUIml) can then be calculated by comparing the values for the test and reference vaccines using the method of Reed & Muench (5) (see Chapter 38 Annex)

Evaluation of the test

In-process control

The modified ABT has been used for over 10 years for in-process control of inactivated rabies vacclnes Its usefulness has been demonstrated in a variety of studies (2 3, 6, 7) Results are usually available in 3 days The test is routinely employed during the manufacturing process to determine the antigen content of the ,nactivated virus harvests the purified concentrated rabies virus antigen and the final vaccine lots in parallel with lhe NIH test It is also used to determine the stability of inactivated rabies vaccines at different temperatures

' The pH should be ad!usted to 12 using dilute sodium hydsaxide (haOH) solution


Estimation of potency

The modified ABT has been correlated to the NIH potency test and problems related to strain-specific properties have not been reported ( 7 8) The test is suitable fot assaying adsorbed vaccines provided that the concentration of the adsorbent (e g aluminium hydroxide) is within certain liniits Vaccines containing preservatives which may alfect tissue culture must iinderyo treatment to prevent destruction of the tissue culture

Results with the inodifed ABT are inore reproducible and show less variability than those obta~ned by the NIH test In add~tion the use of 1n v/trotechniques such as the ABT for in-process control and estimation of potency will reduce the need for animal testing, provided that such tests have been calibrated against other established n-tethods of potency evaluation as outlined In the requ~rements for

human rabies vaccines published by WHO (9 10)

References

1. Arko RJ, Wiktor TJ, Sikes RK The antibody-binding test ior vaccine potency. In. Kaplan MM, Koprowski H eds. Laboratory techriiques in rabies, 3rd ed. Geneva. World Health Organizatior-I. 1973 (WHO Monograph Series. No. 23). 292-294.

2 Barth R Gross-Albenhausen E Jaegei 0 The arrt~body-binding test a useful method for quantitative determination of inactivated rabies antigen Jouriialof biological standardizatiorl 1981 9 81 -89

3 Barth R. Jaeger 0. Antibody-binding test 111 microtechnique system. Rabies information exchange, 1985. 13. 29-32.

4. WHO Expert Committee on B~ological Standardization. Forty-second report Geneva. World Health Organization, 1992 (WHO Technical Report Series. No. 822).

5. Reed LJ. Muench H. A sitnple rnethod of estirriating 50% endpoints. American jouri-ial o l hygiene. 1938, 2 7 : 493-497.

6. Barth R et al. Validation of an in v~t ro assay for the deterinnation of rabies antlyen. Developrnenls in btoiog~cal standardizaiion. 1986, 64: 87-92.

7. Barth R . Diderrich G. Wenman E. NIH test, a problematic method for testing potency of inactivated rabies vaccine. Vaccine. 1988, 6: 369-377.

8. Barth R et al. Purified chick-embryo cell (PCEC) rabies vaccirie: its potericy perforn~ance in different tesr systems and in humans, Vaccine 1990, 8: 41-48.

9. Requirements for rabies vaccine for human use. WHO Expert Committee 01:

Biological Standardizatiot-i. Thirty-first report. Geneva, World Health Organ~zation, 1981 (WHO Technical Report Ser~es, No. 658) Arinex 2.

MODIFIED ANTIBODY-BINDING TEST

0 . Requirements for rabies vaccine for human use (amendment 1992). WHO Expert Committee on Bi'o/ogica/ Standardization, Forty-third report. Geneva, World Health Organ~zation, 1994 (WHO Techn~cal Report Ser~es. No. 840). Annex 4.

m Antirabies serum

and immunoglobulin

CHAPTER 44

Production of antirabies serum of equine origin T Liiekralang, ' J Warlgsa~' & P. Ptiarluphak3

Introduction

Different types of equine antirabies imm~~noglobul in (ERIG) have been produced using various immunogenic preparations consisting usually of a combination of inactivated and fixed strains of rabies virus ( I ) A therapeutic antirabies immunoglobulin for human use is produced at the Queen Saovabha Memorial Institute (QMSI) Bangkok, Thailand by immunizing horses with a purified Vero cell rab~es (PVR) vaccine The animals are given a series of injections of the vaccine in increasing volumes All the in~ections are given subcutaneously into the lateral aspect of the neck The inimunization period lasts 105 days and the first bleeding is made 14 days later

Method

Immunization schedule

Healthy male or female horses aged 4-12 years and weighing 350-450 kg are immunized as follows.

Day 0 : 0.5 m (one human dose) of PVR vaccine in 0.5 ml of complete Freunds adjuvant subcutaneously at orie site. DayZl , 28,35and 42: 1 m (two human doses) of PVR vaccine adsorbed with 1 m1 of 2% bentonite subcutaneously at one site (4 injections). Day63. 70, 77,84,91,98and 105: 2 ml (four human doses) of PVR vaccine withoui adjuvant subcutaneously on both sides of the neck (14 injections). Day 119. first bleeding

From day 21 to day 56, serum samples are taken at weekly intervals and the antibody titre I S determined by the rapid fluorescent focus inhibition test (RFFIT) or the mouse neutralization test (MNT) (see Chapters 15 arid 47) Anilnals showing a titre of less than 70 lU/ml by day 56 are withdrawn from the donor population With the above immun~zation schedule i t is usually possible to obtain antibody titres of 150 lU/ml or higher

Bleeding of horses

The animals are bled from a jugular vein by plasmapheresis (3000 m1 of whole blood is collected at each session) on days 119, 121 and 123 After 5 weeks, they

' Ass~stant Director, Queen Saovabha Memorial Institute, Banglrok Thailand 2Sc ien t~s t , Serum Sect10.1 Queen Saovabha Memorial Iris!it,~!e Rarykok, T i i a l a w i

Deputy Director. Queen SaovabQa Memoriaa Insritute Bangkok i i-ailand


are given four booster injections of 2 rnl (four human doses) of PVR vaccine at weekly intervals. The bleedings are repeated 14, 16 and 18 days after the last booster injection.

Concentration and purification of antirabies serum

The relatively large amounts of antirabies serum necessary for the protection of persons exposed to rabies, as well as the risk of anaphylactic accidents and other reactions, have led to the development of a nurnber of methods for preparing a pitrifled. concentrated serum.

Protein fractionation was first attempted by Habe in 1945 (2) using ammoniuin sulfate, and various other methods have since been described (3, 4). The method of fractionaton and purification adopted at the Pasteur Institute, Paris, consists of two stages.

(I) enzymatic digestion of the proteins followed by precipitation with ammonium sulfate;

(ii) removal of the excess proteins by thermocoagulation (see also Chapters 45-47),

Whichever method is adopted it is advisable to determine the final protein content of the purified serum and relate this to its protective titre (see Chapter 47) Paper electrophoresis should also be performed to check the fractionation of the proteins In general a concentrated purified serum with a titre of 120 IU/ml should not contain more than 5 % of total serum proteins

Equine hyperimmune plasnla or serum with an antibody titre of 150 lUlml or higher can be easily concentrated and purified to give an ERIG with a titre of 200 IUIml (see Chapter 45)

Factors affecting the production of ERlG

Donor animals

The animals used must be carefully selected as even for the same breed of horse lhe suitability of any particular animal for serum productiori varies according to its age nutrt ior ia status general health and immunization history Antibody titres should be checked in donor aniinals at appropriate intervals (see above) and any animals that do not produce consistently high titres should be withdrawn from the donor populatiori or? day 56

Type of rabies vaccine

Different techniques for the producton of ERIG have used a vdriety of rabies vaccnes to induce high antibody levels including vaccnes prepared froni nerve tissue as well as those prepared from t,ssue cultdre The former have been assocated with the occurrence of anaphylactic and rieuroparalytc reactions in vaccinated horses (T Luekrajang et al personal commbnication) No such reactions have been reported following immunization with PVR vaccine prepared according to the requirements for human rabies vaccines pubitshed by WHO

(5 6) Indeed recent data n d c a t e that PVR vaccine is probably more mmuno-

PRODUCTION OF ERIG

genic in humans than other cell-culture vaccines of equal potency (P. Khawplod et al.. personal communication).

Use of vaccines containing adjuvants

Vaccine adjuvants of various types have been used in different institutions. It has not yet been established whether adjuvants are necessary. Studies are also

needed to determine whether non-ulcerogenic adjuvants such as alum~nium hydroxide or bentonite are as effective In inducing high antibody titres as complete Freund's adjuvant, which often causes severe local reactions, Studies have shown that aluminium hydroxide mixed with PVR vaccine and used for the intramuscular vaccinaiion of humans will enhance the antibody response significantly (7)

Vaccination schedules

Sim~larly controlled studies are also required to determine the optimal schedule for vaccinating horses in order to produce high titres of ERIG

Potency

The RFFlT and the MNT are the most widely used potency tests for antirabies serum and immunoglobui~n They are described in Chapters 15 and 47

When the potency of several lots of €RIG from different manufacturers was assayed at the Queen Saovabha Memorial Institute wide variations in potency were found ranging from 122% to 31200 of the stated amount [H Wilde et a1 personal communication) Since high doses of ERIG may suppress the rabies virus-neutralizing ant~body response in vaccinees manufacturers must ensure that ERIG preparations are calibrated against the International Standard for Rabies lmmunoglobulin (8) or a national reference preparation and that the potency (in IU per ml) IS clearly stated on each ampoule or vial The leaflet accompanying the package should include information about the storage cond- lions and specify the expiry date of tlhe product

Safety

ERIG preparations iron? various marufactdrers have been shown to induce serum sickness in some recipients In several prospective studies of recipients the incidence of serum sickness ranged from 082'0 to 6 19% depending on the prote1n content of the ERG 19 10) Most cases were relatively minor and did not require alteration of the post expos-ire treatment regimen Nevertheless the purification proceduies used by different manufacturers appear to influence the incidence of adverse reactions ( 1 0 ) Although [he reported rates are much lower than those associated with the first available preparations of ERIG (46%) ( 1 1 ) further stud~es are reqoired ro improve the safety of this product

References

1 Lepinc P Atanasiu P Prodclction of antirabies serum of animal origin In Kaplan rvlM Koprowsk H eds Laboralory techniques in rabies 3rd ed


Geneva, World Health Organizat~on, 1973 (WHO Monograph Series, No. 23): 299 303.

2 Habel K Public health reports (Washington) 1945. 60' 545

3 Cohn EJ et al A system for the separation into fractions of the protein and lipoprotein components of biological tissues and fluids Journal of the American Chemical Societj/ 1946 68 459-475

4. Hao YL, Wickerhauser M A simplified method for preparation of imn~une- serum globulin. Vox sanguin~s. 1981. 40. 278-285.

5. Requirernents for rabies vaccine for human use. WHO Expeit Committee on Biological Standa~dization. Thirty-fiist report. Geneva, World Health Organ- zaton. 1981 (WHO Technical Report Series. No. 658), Annex 2

6 Requirements for rabies vaccine for human use (amendment 1992) WHO Expert Cornrnittee 011 Biological Standardizat~on Forty-third report Geneva World Health Organization, 1994 (WHO Technical Report Series, No 840), Annex 4

7 Phanuphak P et at lmmunoenhancement with combined rabies atid a u m n - ium adjuvarited vaccines Vaccine 1989 7 249-252

8. WHO Expert Corninittee on Biolog~cal Slandardization. Thirty-fifth report. Geneva, World Health Organization, 1985 (WHO Techn~cal Report Series, No 725).

9. Wilde H et al. Purified equine rabies immune globuiin: a safe and affordable alternative to human rabies Immune globuiin. Builettn of the World Health Organization, 1989, 67: 731 -736

10 Wiide H Chutivongse S Equine rabies immune globulin a product with an cindeserved poor reputation. American journal of tropical medic~ne and hygiene, 1990. 42: 1751 78.

1 1 . Karliner JS. Belaval GS Incidence of reactions following administration of antirabies serum: a study of 526 cases Journal of the American Med~cal Association. 1965, 193: 359 362.

CHAPTER 45

Purification techniques for hetersiogous rabies antiserum R. Gluck' & D, iahert2

In view of the high cosis of rabies immunoglobcilin of hurnari origin (HRIG) heterologous (manly equine) immunoglobulins are still frequently prescribed for the prevention of rabies in persons who have been severely exposed (category Ill) to the virus even thoiigii they inay cause senstization and are eliminated more rapidly than HRIG

Puri fcat~on techniques can be used to reduce the risk of sensit~iation to ERIG Their objective is to maximize the specific activity and to minimize the allergenic substances In the product When these techniques are ~mplemented i t is advisable to adhere to the recomniendatons of the WHO Expert Committee on Biological Standard~zaton ( 1 )

The purification of imn~unoglobulins from human plasma is carried out according to the technique of Cohn et a1 (2) based on the selective precipitation of proteins by chilled ethano This technique has been adapted for purifying heterologous immunoglobulins and was described in the previous edition (3) This chapter describes two other techniques for the puritication of ERlG which may be applied to clarified equine serum or plasma

Preservation and storage of serum or plasma

After coagulation of the collected blood and retraction of the clot for 48 hours at 2-8 "C, it is poss~ble to ob ta~n serum by siphoning. Plasma can be obtained by decanting the collected blood into an appropriate volume of anticoagulant solution, e.g. ctr lc acid buffer, pH 4.8 (see Aniiex: 1 litre per 10 litres of blood) and siphoning after 12 and 24 hours at 2-8'C.

Plasma and serum are usually stored froiel i at - 20°C. Alternatively, an antiseptic may be added and the blood derivatives kept at 2 8'C until purification. The antiseptics most corninonly used are phenol and methyl derivatives (cresol or metacresol), e.g. 0.15-0.20% m-cresyl acetate The latter is preferable to phenol. which can yield coloured derivatives duriiig purification. Phenol or its derivatives are initially diluted to 10 times the required concentration with distilled water and added very slowly, while shaking, to the serum or plasma to avoid prec~pitation of the protein.

A I of these procedures should be carried out uiider conditions close to asepsis to avoid microbial containination and proliferation, which could lead to pyrogeris being present in Ihe final product.

' Head. V~roiogy SWISS Serun? and Vaccne I r i s l i t ~~ te . Berne, Sw~tzef land Director, Pasiei lr MBr ie~ ix nstitdte. V a - d e - R ~ L I I I France


During refrigerated storage, seruin and plasma generate a sediment or coaguluni which should be eliminated by continuous centrifugation at 5600 g and 15°C in a centr~fuge eqilipped with a 50-litre retention bowl. Between 500 and 600 litres of serum or plasma Inay be centrifuged per hour. Starting from these clarified sera, two purification techniques may be applied.

Purification by enzyme treatment and heat denaturation

This technique (d-7) consists of:

- cleavage of the inimunoglobulins by a proteolytc enzyme, pepsin, followed by separation of the F(ab')2 fragments, responsible for their protective activity, from the Fc fragment;

- selective denaturation by heat; - fractionated precipitation by neutral salts, such as ammonium sulfate.

Nimination of albumin and pigments

1 . Collect the effluent of centrifugation from the retention bowl in a stainless steel or glass tank equipped with a stirrer Dilute by half with distilled water

2. Add pure crystalline ammonium sulfate slowly, while stirring, to give a final concentration of 1.75 rnolVl. Check the pH and adjust to close to neutrality if necessary.

3 After at least 6 hours, collect the precipitate by filtering the mixture on a filter press equipped with nylon gauze (500 litres hour per m* of gauze under 196 Pa) The filtrate, which must be clear, sliould be discarded It contains most of the albiirnin, other non-precipitabe proteins, pigments and free antiseptic, w i i ch could react with the immunoglobulins during subsequent procedures.

4. Resuspend the precipitate in a volume of distilled water equal to twice the volume of serum or plasma used.

Pepsin digestion

Place the protein suspension obta~ned in a thermoregulated double-walled reactor equipped with a stirrer Adjust the pH to 3 2-3 3 with dilute hydrochloric acid (HCI) and add an appropriate amount of pepsln (about 1-2 g per litre of the serum or plasma to be purified) The temperature shoiild be iricreased to 32 C with stirring and maintained for 30 niinutes

Pepsin digestion is essential and must be carefully controlled The pepsin IS

tested to ensure its constant qiiality from one batch to the next and thereby the reproducibility of each digestion The technique described by Anson (7) using proteolysis carr~ed out with bovine haemoglobin is suitable for this purpose

Moreover it is necessary to control the purity of pepsin and to check whether it is free of blooa group A factor which could conta~ninate the purified immunoglobulins For this purpose the absence of blood group A factor should be verified with appropriate tests which should include an anti-A serum of human or lgn as coritrol If this is not possible the pepsin can be purified by adsorption on calcium phosphate gel at pH 6 0-6 1

PURIFICATION TECHNIQUES FOR HETEROLOGOUS ANTISERUM

Selective denaturation by heat

1 At the end of digestion adjust the pH of the purified suspension to 4 3 with dilute ammonium solution (NH,)

2 Add an appropriate amount of ammonium sulfate (NH,SO,) while stirring to achieve a final concentration of NH,SO, of 0 75 mol/l

3 Add 1 rnl of toluene and 5 g of activated charcoal per litre of suspension 4 Bring the temperature as rapldy as possible to 55'C and when this tempera-

ture has been reached filter immediately on a filter press equipped with type K2 Seitz filter d~scs

5 Rinse the filter with a 0 75 moll1 solutiorl of ammonium sulfate and conserve the filtrate

Adjustment of the pH of a concentrated suspension requires certain precautions. The suspension should be diluted to a lower concentration of 0.3 mol/l with distilled water before determining the pH.

Concentration of the protective fractions

1 . After adjusting the pH of the filtrate to neutrality, increase the concentration of ammonium sulfate in the flitrate to 1.75 mol/l.

2 Leave the suspension to stand for several hours, and then filter on a filter press equipped with nylon gauze with a very fine mesh (500 litres/hour per m' of gauze under 196 Pa). Caref~llly collect the preciptaie, which contains the fractonated irnmunoglob~~lins.

Dialysis

All of the above steps are conducted in the presence of a concentration of ammonium sulfate sufficiently high to be bacteriostatic. From this stage of dialysis onwards, precautions should be taken to avoid bacterial contamination as the medium is very favourable to microbial proliferation.

Dalysis, to eliminate all traces of ammonium sulfate, is controlled with the aid of Nessler's reagent on the dialysate, and ultimately on the retained material. Dialysis is carried out by diafiltration using filters with a relative molecular mass cut-off of 10000, However, if this type of equipment is not available, a haemo- dialyser or dialysis tubes sterilized with a 0.5Oh phenol soluton can be used.

If necessary, a preliminary ultrafiltration of the dialysate may be carried out to ensure that no pyrogenic substances are present.

Clarification

The technique most commonly used consrsis of adsorption onto alumrn~urn hydroxide gel The gel can be activated with glutamic acld (A Hansen personal communication) The required quantity of gel 1s approximately 3 5 g (expressed in aluminium oxide) per litre of solution The protein coricentration of the solutron is adjusted to 8-10% After several hours the suspension is centrifuged at 5600 g then filtered on a filter press equipped with type EK Setz filter discs Approximately 150 litres of suspension can be centrrfuged per hour


Purification by precipitation using ethacridine lactate and ethanol

Ethacridine lactate precipitation

This technique was introduced in 1956 (8) for the purification of HRlG and has since been adapted to the purficatiori of ERIG Ethacridine lactate is added to the seriim or plasma to g ve a final concer~tration of 5 2 g per litre Approximately 90% of the added amount is e l~mnated with the precipitated proteins while the remainder which ensures the complete precipitation of unwanted proteins, is rernoved by adsorption on ctharcoal

1 Add the calculated volume of aqueous 1 2% ethacridne lactate solution io the clarified serum or plasma during 15 30 minutes stirring well

2 Stir the mixture for a further 30-40 minutes at 20LC and then hold at the same temperature for 1-2 hours withoiit stirring

3 Carefully decant the supernatant and centrifuge the precipitate at 5600 g and 20 C for 30 minutes

4 Determine the volume of the resulting supernatant combined with that obtained in step 2 For each litre of this s o l ~ ~ t i o n add 6 g of activated charcoal to adsorb the dissolved ethacridne lactate Stir the mixture for at least 2 hours at room temperature

5 Fliter the solution on a 40-V Seitz Orion-type filter press equipped with filter sheets at a pressure of 70-100 kPa Rinse the filter with 40 litres of distilled water and add the rinsing water to the filtrate

Ethanol precipitation

1 Add the calculated volume of ice-cold 95O"' ethanol to the filtrate during 4-8 hours to give a final concentration of 27% stirring well During this step the temperature should be gradually lowered froin 0 C to - 4 C

2 When all the alcohol has been added lower the temperature to - 10'C and hold the solution at this temperature for at least 16 hours (or overnight) with stirring

3 Filter the solution on a 40-V filter press measuring 40 x 4 0 cm containing carrier sheets covered on the smooth side with filter paper of the same size During this step the pressure should be gradually increased from 100 kPa to 200 kPa while maintaining a constant temperature The filtration takes about 3 0-3 5 hours The filtrate is collected for recuperation of the ethanol

4 Remove the filter cake containing the precipitated irnrnunoglobulins (IgG) by pumping compressed air through the filter for 10 minutes Collect the filter cake in a suitable stainless steel vessel equipped with a stirrer and resuspend in about 6 litres of phosphate-buffered saline-sod~um chloride buffer (PBS-NaCI) pH 7 4 (see Annex) at room temperature

5 Clarify the suspension by filtering on a 20 X 20-cm filter containiny filter sheets of the same size at a pressure of 50-100 kPa Rinse the filter with 4 litres of PBS-NaCl pumped through the filter for 5-10 minutes

6 Combine the filtrate and rinsing solvent Freeze-dry the solution containing the precipitated IgG in a freeze-drier reserved for the lyophilization of animal moteins

PURIFICATION TECHNIQUES FOR HETEROLOGOUS ANTISERUM

Stabilization and preservation of purified HRIG

Regardless of the purification technique used, the purified HRIG suspension is stabilized using a concentrated buffer containing glycine (10 gll) and sodium chloride (5 gll) to be isotonic (pH 6.1-7.1). and an antiseptic such as thiomersal' (1.5-2.0 g//) or m-cresyl acetate (0.1 is added as preservative.

Standardization of the final product

After sterile filtration, the activity of the preparation IS determined by appropriate tests (see Chapter 47) to permit standardization of the final bulk product. The standard dose of ERlG for post-exposure treatment of humans is 40 IU per kg of body weight.

References

1. Requirements for the collection, processing and quality control of blood, blood components and plasma derrvatives. WHO Expert Committee on Biological Standardization. Forty-third report. Geneva, World Health Organization, 1994 (WHO Technical Report Series, No. 840), Annex 2.

2. Cohn EJ et al. Preparation and properties of serum and plasma proteins. IV. A system for the separation into fractions of the protein and lipoprotein components of biological tissues and fluids. Journalof the American ChernicalSociety, 1946, 68 : 459-475.

3. Selimov M, Gordienko E. Preparation of antirabies immunoglobulin of animal origin: method used in the USSR. In: Kaplan MM, Koprowski H, eds. Laboratory techniques in fables, 3rd ed. Geneva, World Health Organization. 1973 (WHO Monograph Series, No. 23): 304-306.

4. Pope CG. The action of proteolytic enzymes on the toxins and proteins in immune sera. I. True digestion of the proteins. Bn'tish journal of experimental pathology, 1938, 19: 245-251.

5. Pope CG. The action of proteolytic enzymes on the anti-toxins and proteins in immune sera. II. Heat denaturation after partial enzyme action. Britishjournalof experimental pathology, 1939, 20: 132-201

6. Harms AJ. Purification of antitoxic plasmas by enzyme treatment and heat denaturation. Biochemical journal, 1948, 42: 390-396.

7. Anson ML. The estimation of pepsin, trypsin, papain and cathepsin with haemoglobin. Journal of general physiology, 1938, 22: 79-89.

8. Horejsi J, Smetana R. The isolation of gamma globulin from blood serum by Rivanol. Acta medica Scaridinavica, 1956, 155 (Fasc. l ) : 65-70.

' Also known as rnercuro!hiolate and th~merosal


Annex Preparation of buffers

Citric acid buffer, pH 4.8 C~tr ic a c d , rnoriohydrate Dextrose Trisodiiim cilrate, dhydrate Distilled water to make

PBS-sodium chloride buffer, pH 7.4 Sodium phosphate, dibasic, dihydrate (Na2HPO,.2H,O) Sodium chloride (NaCI) Potassium phosphate. monobasic (KH,PO,) D~stilled water to make

CHAPTER 46

Production of human rabies imrnunoglobulinl P Fournler2 & R. K Sikes3

introduction

The combination of local treaiment of the wound, passive inirnuiiization with rabies immunoglobulins and vaccination is recommended for all severe (category Ill) exposures to rabies ( I ) . The use of homologous immunoglobulins for human post-exposure treatment virtually eliminates the risk of anaphylaxis and serum sickness associated with heteroogous serum products. In 1965, approximately 16% of persons treated with antirabies serum of equine origin were reported to have developed serum sickness; among persons over 15 years of age, the incidence was 46% (2). In the past few years, purified equine imrnunoylobulins have become available, and in recent studies, the incidence of serum sickness among recipients was reported to be < 1- 6.2% (3-6).

To avoid such reactions, human rabies imrnunoglobulin (HRIG) preparations have been developed and used for post-exposure treatment in most industrialized countries. HRIG is well tolerated, but it is expensive and available in only limited quantities. Since 1975, this product has been administered to more than 250000 people in the United States and no cases of serum sickness have been reported

Formula

HRIG for intramuscular adm~nistration is a 10-189.6 solution o f ~mmunoglobul~n in 0.3 mol/l glycine, and preserved with a 1 : 10000 thiomersa14 solution. The rabies virus-neutralizing antibody content is standardized to contain 150 International Units ([U) per rnl. It is supplied in 2-m1 (300 IU) and 10-m1 (1500 IU) vials for paediatric and adult use respectively.

Source and shipment of blood

Donors of plasma for the production of HRIG should have demonstrated high levels of virus-neutralizing antibody (at least 5 IUIml) following pre-exposure or post-exposure immunization with a potent cell-culture rabies vaccine preferably human diploid cell (HDC) vaccine One or more booster immunizations with HDC vaccine should be given 1-2 weeks before the first collection of blood In order to prepare HRIG of suiiicient potency and assuming a 15-20-fold concentration ot antibody during preparation it IS necessary to start with a rabies immune plasma

' Based on the chapter by R K. Stkes in the previous edition Former Production Manager Sera and Vacciries, Pasteur-Merieux ins:itute, Marcy i 'Eto~ie, France ' Former Veterinary Director Ofilce 0: Veterinary Public Health Services Cer7ters for rliseasi: Control and

Prevent~on. A!an!a. GA, USA Also known as tbimcrosal an0 mercurotiiioate


pool containing 15 IU per m (as measured by the RFFIT). Studies have shown that, following a booster dose of antirabies vaccine, about 10% of people who have received pre-exposure prophylaxis and about 40% of those with a history of post- exposure treatment will develop atltibody titres of sufficient levels.

The collecting centre may obtain a unit of plasma by plasmapheresis or by separation of a unit of whole blood aseptically. The plasma is shipped refrigerated to the laboratory and is stored frozen until it is ready to be pooled for fractionation.

Reagents

The preparation and storage of [he reagents required are described in the Annex.

Technique

The immunoglobulins are separated from the plasma by the cold ethanol fractionation technique described below (7).

Precipitation of fraction I

1. Pool the plasnia; pour individual units througti cheesecloth into a tared vat large enough to contain 1.2 litres per litre of plasma.

2. Determine the pH; i f necessary, adjust to 7.0-7.4 with acetate buffer (80-fold concentrate) or 0.5 mol/l sodium phosphate solution.

3. Freeze three 15-ml samples of pooled plasma 4. Determine the weight of plasma. 5. Keeping the plasnia at O J C , add 163 g of cold 95% ethanol per kg of plasma.

Adjust the rate so that the addition of ethanol takes about 1 hour. Reduce the temperature to - 2°C during the addition. Stir for 30 minutes after the addition of ethanol is completed,

6. Centrifuge at l 0 000 g for 30 minutes. collecting the centrifugate in a tared vat large enough to contain 1.7 l~tres per litre of original plasma. Maintain the temperature at - 2'C to - 3°C d u i ~ n g centrifugation Freeze a 15-ml sample of sopernatant I (SI) and weigh it. Keep the sample at - 2'C, and begin precipitation of fraction 11-Ill.

Precipitation of fraction 11-111 (P2)

1 For every kg of supernatant l add 552 g of 95% ethanol to which has been added 1 33 m1 of acetate buffer 80-fold concentrate Adjust the rate so that the addition of the ethanol takes about 1 hour Cool to - 9 C during the add~tion

2 Stir at - 9°C fol 30 minutes after the addition of ethanol is completed 3 Centrifuge at l 0 000 g fo r 30 minutes collecting the centrifugate In a tared bowl

Maintain the temperature at - 6 - C to - 9 - C during centrifugation 4 Determine the weight of the precipitate which is fraction 1 1 - 1 1 1 (P2) Freeze in the

bowl at - 20°C and hold for precipitation of fraction I-lllw

Precipitation of fraction 11-lllw (P2w)

1. Remove frozen P2 from the centrifuge bowl and rapidly homogenize to a

PRODUCTION OF HRlG

uniform suspension in a mixture of water and crushed ice. Use 2 g of water- and-ice rn~xture per gram of P2. Avoid excessive foarr~~ng.

2. For every gram of P2 rnmediately add a mixture containing 0.107 ml of 0.5 mol/l sod~um phosphate, dibasic (Na,HPO,) In 2.89 g of water and crushed ice.

3. Stir in the cold until all the solid is in suspension. 4. Pour into a vat containing 20 g of cold water per grani of P2. and stir at 1 'C for

30 minutes or until no ice remains in the mixture. The vat should be large enough to contain 4 litres per l 0 0 g of P2.

5. Remove 5 ml of the mixture from the vat, add 5 m1 of 0.15 mol. I sodium chlor~de (NaCI) solution, and determine the pH: i f necessary. adjust the pH of the mixture to 7.0-7.4 by adding acetate buffer (80-fold concentrate) or 0.5 rnol//l sod~um phosphate, dibasic, so l~~ t ion .

6. Freeze a 15-m1 sample. 7 Add 14 1 g of 95'.0 erlianol per gram of P2; adjiist the rate so that the addition of

[tie ethanol requires anout 1 hoiir. Cool to - 6 C during this step Stir at - 6 C for 2-4 tioiirs after the addition of ethanol is completed.

8. Centrifuge at 10 000 g for 30 minutes and collect the centrifugate in a tared bowl. Maintain the temperature at - 5 ' C to - 7 -'C during centrifugation.

9. Freeze a 15-m1 sample of supernatant 11-lllw (SZw). 10. Determine the weight of precipitate, which is fraction l1 l l l w (P2w). Freeze in the

bowl at - 20'C and hold for precipitation of fraction Ill.

Precipitation of fraction 111

1. Remove the frozen P2w from the bowl, and rapidly homogenize to a uniform suspension in a mixture of crushed ice and water. Use 2 g of water-and-ice mixture per gram of P2w. Avoid excessive foaning.

2, Immediately add 2 m1 of cold 0.175 mol,/l sodiunl acetate solution per gram of P2w and stir in the cold until all the solid is in suspension.

3. Adjust the pH to 5.2 + 0.1 by adding acetate buffer (80-fold concentrate). diluted 1 :25 w ~ t h cold water, and then add more cold water until the total amount added is 1 ml per grani of P2w.

4. Stir in the cold for 1 hour or until no ice remains in the mixture. Determine the pH and adjust, if necessary. to the range 5.1-5.3

5. Pour into a vat containing 13.5 g of cold water per gram of P2w. The vat should be large enough to contain 2.9 litres per 100 g of P2w.

6. Add 8.1 g of 9594 ethanol per gram of P2w: adjust the rate so that the addition of ethanol requires about 1 hour. Cool to - 6"C during this step. Stir at - 6°C for 30 minutes after the addition of ethanol is completed.

7. Centrifuge at 10000 g for 30 minutes, collecting the centrifugate in a cold pressure tank and filter into a tared vat. The vat should be large enough to contain 3.1 litres per 100 g of P2w. Use a 1.2-pm epoxy reinforced glass filter over a 0.8-pm membrane. Maintain the temperature at - 5°C to - 6'C during centrifugation and filtration.

8. Freeze a 15-ml sample of supernatant Ill (S3). 9. Determine the weight of 53. maintain at - 6 ° C and proceed with the

precipitat~on of fract~on 11.


Precipitation of fraction I1

1 Stir S3 vigorously at - 6 ° C and for every kg of this solution slowly add 2 g of sodium bicarbonate (NaHCO,)

2 Add 2 ml of the vat mixtcire to 8 m1 of 0 15 mol I sodium chloride (NaCI) solution and determine the pH if necessaiy add additional sodium bicarbonate to adjust the pH to 7 2-7 6

3 Add 94 7 g of 95% ethanol per kg of solution Adjust the rate so that the addition of ethano requires about 1 hour Cool to 9 C during this step

4 Centrifuge at 10000gfor 30 minutes collecting the centrifugate in a tared bowl Maintain the temperature at - 6'C to - 9 C during centrifugation

5 Determine the weight of fract~on ll freeze in the bowl at - 20 C and hold for lyophilization

Fraction I1 lyophilization and final preparation

1. Remove the frozen fraction ll from the bowl, and rapidly homogenize to a uniform suspension in cold 0.3 mol/l g l ycne using 1 ml per gram of frozen paste.

2. Add 4 g of cold water' per gram of paste; stir for 2-4 hours in the cold, and leave to stand overnight at 0°C without stirring.

3. Decant the skipernatant, lyophrlize and determine the weight of the lyophilized product.

4. For preparatiori of the final solution of the product, the quantities are calculated as follows:

(a) Dry weight of powder = weight of Iyophilized powder (g) X 0.98. (b ) Volume of powder (ml) = a X 0.75. (c) Volume of water to be added (ml)

= ml of glycne added before lyophiiization - b. (d) Dry weight of globulin (g)

= a - (0.0225 X m1 of glycine added before lyophilization). (e) Final volume (ml) = d10.17. ( f ) Volume of 0.3 mol/l glycine to be added (ml) = e - ( b - c). (g ) Volume of 10?& thiomersal to be added (ml) = e x 10-3.

5. Put the required volume of water' and 0.3 mol j l glycne (see step 4) in a beaker with a magnetic stirring bar. and add the lyophilized globulin Stir until the globulin is completely d~ssolved: avoid excessive foaming.

6. Add the required volume of 10% thiomersal (see step 4). A second person should check this calculation and observe the addition. Determine the pH and adjust, if necessary, to 6.8 with acetale buffer. 80-fold concentrate.

7. Filter the globulin solution through depth f l ters down to 1.2pm and then through membrane filters down to 0.45 pm.

8. In the sterile room, sterilize the globulin solution by filtration through a 0.2-/1m filter into the sterilized bulk product container,

9. Remove a 2-ml sample from the bulk container for sterility testing with liquid thioglycolate medium.

' Use Water for Inlection see reference 8

PRODUCTION OF HRlG

10. Place the bulk container material n a refrigerator at 1-3"C, and remove samples for testing (see below). When all the tests are complete, distribute into final containers.

Testing

Tests for safety and sterility, pyrogenic substances, heat stability pH and turbidity are needed to ensure that the product satisfies the minimum requirements recommended by the WHO Expert Committee on Biological Standardization (9). For potency requirements, see Chapter 47, page 420.

Disadvantages

HRlG is expensive, because of the cost of HDC vaccine and the time required for rabies virus-neutralizing antibodies to be produced in donors following immunization. Since the manufacturing process requires a supply of donors whose antibody titres are at least 15 IU per ml, the product is also limited in terms of quantity. In spite of these disadvantages. the Immunization Practices Advisory Committee (ACIP) of the United States recommends that HRIG should be used in preference to ERIG (10).

References

1. WHO Expert Committee on Rabies. Eighth report. Geneva. World Health Organization, 1992 (WHO Technical Report Series, No. 824), Annex 2.

2. Karliner J S Belaval GS. lrcidence of reactions following administration of antirabies serum. a study of 526 cases. Journal of the American Medical Association, 1965, 193: 359- 362.

3. Wilde H et al. Safety of equine rabies immune globulin. Lancet, 1987. 2: 1275

4. Wilde H et al. Adverse effects of equine rabies immune globulin. Vaccine. 1989, 7: 10-11.

5. Wilde H, Chutvongse S. Equine rabies immune globulin: a product with an undeserved poor reputation. American journal of tropical medicine and hygiene, 1990, 42: 175-1 79.

6. Wilde H et al. Purified equine rabies immune globulin: a safe and affordable alternative to human rabies immune globulin. Bulletin of the World Health Organization, 1989, 67. 731 7 3 6 .

7. Cohn EJ et al. Preparation and properties of serum and plasma proteins. A system for the separation into fractions of the protein and lipoprotein components of biological tissues and fluids. Journalofthe American Chemical Society, 1946, 68: 459- 475.

8. The international pharmacopoeia, 3rd ed. Volume 2; Volume 3. quality specifications. Voiuine 4: tests, methods, and general requiremenls: qualiiy


specifications for pharmaceut~cal siibstances, exopients, and dosage forms. Geneva, World Health Organization. 1981; 1988; 1994.

9. Requirements for the collection, processing and quality control of blood, blood components and plasma derivatives. WHO Expert Committee on Biological Standard~zation. Forty-third report Geneva, World Health Organization, 1994 (WHO Technical Report Series, No. 840), Annex 2.

10. Recommendations of the Inimunization Practices Advisory Committee (ACIP). Morbid~ty and mortality weekly report, 1991, 40 (No. 28 RR-3).


All reagents must fulfil the specifications described in The internationaipharmaco- poeia. '

Distilled water Water is collected from the still directly into a pressure-filtration tank and filtered through a sterile 2-pm membrane into a sterile dispensing vessel. Except when pyrogen-free water for injection' is required, this water is used in all procedures and for making ice and reagents.

95% (v/v) ethanol 95% ethanol is filtered through a 0.2-pm membrane (not sterile) into dispensing vessels and stored under refrigeration (cold room).

Acetate buffer, 80-fold concentrate Sodium acetate Glacial acetic acid Distilled water to make

When this concentrate is diluted with 80 parts of water, the pH of the buffer should be 3.98-4.02.

' See reference 8.

CHAPTER 47

Potency test for antirabies serum and immunoglobulinl E A F/tzgera/d2

Principle

The method described below consists in neutralizing a constant dose of the previously ttrated challenge virus with a series of different dilutions of antirabies serum The method is used mainly for the assay and poterlcy testing of therapeutic antirabies serum and immunoglobulin but i t is also applicable to any serum containing rabies antibody Thus it can be used to determine the antibody titres of human sera collected during therapeutic trials of different vaccines

The method comprises the following three stages

1 Preparation and titration of the challenge virus 2 Serum-vir~is neutralization-preparation of the serum and of the serum virus

mixtures, inoculation of the indicator system (I e mice or cell cultures) 3 Interpretation of the results

Preparation and titration of challenge virus

Challenge virus

The strain normally used is the Challenge Virus Standard (CVS) as used in the National Institutes of Health (NIH) test for rabies vaccine potency (see Chapter 37) Any laboratory strain of fixed virus may also be used provided that its LD,, for the mouse is known and remains constant

Titration of stock virus

The challenge virus is stored as a 20°4 suspension or as the supernatant following low-speed centrifugation of such a suspension This material is dispensed into ampoules and frozen at - 80°C or below Shortly before the test is carried out an ampoule is taken from the stock and thawed rapidly under the tap Serial tenfold dilutions are then prepared (2 X 10-2 up to 2 X 10-') The diluent used is double- distilled water to which has been added a protein stabil~ier such as 0 2 % gelatin or 2% normal horse serurn inactivated for 30 minutes a i 56 'C

' Based on the ~twpter by ltie late P Atanasiu in the previous ed~tion D~rector Division of Proddct Quallf) Control Center for B~ologics Evaluation and Resaarch Food and Drug Admin~stration Bethesda MD USA See Chapter 15

41 7


1. Add 0.5 m1 o i each virus dilution from 2 x 10-3 to 2 X 10.~' to each of five dilution tubes (13 X l 0 0 mm or equivalent).

2 Add 0.5 mi of inactivated normal horse serum diluted 1 ' 5 to each tube, so that the final concentrations of virus in the tubes are I O - ~ , 10-5, 10-6 and 1 0 - ~

3 After shaking the tubes, inciibale them for 1.0-1.5 hours at 37 "C in order to ensure that the conditions are the same as those of the serum-virus neuiraliza- tioi? step.

4. Remove the tubes froin the incubator and chill in an ice-water bath 5. Inoculate groups of mice weighing 14-16 g each inlracerebrally with 0.03 ml of

each virus dilution, using at least 10 mice for each dilution. Provided that the highest dilution is used first, all the inoculations may be made with the same syringe.

6. Observe the mice for 14 days and record the number that die after the first 5 days. Include any mice showing signs of rabies on the 14th day.

For the calculation o i the virus titre, see Appendix 3, Table A3.4 (page 452), Examples 1-3.

Calculation of challenge doses

As an example of how challenge doses are calculated, assume that a given CVS stock preparation contains to5 84 LD5, per 0 03 m1 I e 1 LD,, is contained in 0 03 m of a 10-5 8 4 dilution of the stock virus For sera of high t~t re between 100 and 500 LD,, are used in the serum-virus neutralization assay To find the CVS that will contain 500 LD,, in 0 03 m1 the logarithm of 500 is subtracted from the logarithm of the assumed titre of the stock preparation

log dilution of challenge preparation containing 500 LD,, of virus per 0.03 m1 = log 105.84 - log 500 = 5.84 - 2.70 = 3.14

Since 3.14 = log 1380, this means that the CVS stock preparation will have to be diluted 1 : 1380. For this purpose. first make serial teniold dilutions of the stock CVS until a l o M 3 dilution is obtained then add 0.38 ml of diluent for every 1 m1 of the 10-3 dilution. The final dilution will then contain 500 LD,, of virus per 0.03 m1

For sera of low titre (e.g. postvaccinai human sera), only 50 LD,, may be needed. The calculation then becomes.

Since 4.14 = log 13800, the CVS stock preparation will have to be diluted 1 : l 4000 by making serial tenfold dilutions down to a dilution of I O - ~ and then adding a further 0.4 ml of diluent for every 1 m1 of this dilution, T h ~ s gives a final dilution containing 50 LD,, of virus per 0.03 ml.

Serum-virus neutralization

Inactivation of sera

The sera to be tested are inactivated for 30 minutes at 56°C

41 8

POTENCY TEST FOR ANTIRABIES SERUM AND IMMUNOGLOBULIN

Neutralization

1 . Prepare serial twofold dilutions of the serum or imrnunogiobulin under test, starting with 1 500 The final diliition should show no evidence of neutralization.'

2. Transfer 0.5 ml of each of these dilutions to a test tube. 3. Add 0.5 m1 of the vlrus d~lut ion corresponding to 500 LD,, (in the example glven

above, a dlut ion of 1 :1380) to each tube. This results in a twofold dilution of both the virus and the serum, so that the final virus is 1 :2760 or 1 0 - ~ . ~ ~ , and the final serum dilutions are 1 : 1000, 1.2000, 1 :4000. 1 :8000, etc. Incubate the serum virus mixtures at 37°C for 1 . 0 1 5 hours.

It is essential to include in the test a reference serum, which is titrated at the same time as the unknown sera. This reference serum is a serum of known titre that has previously been calibrated against the International Standard for Rabies lmmunoglobulin2 ( 1 , 2), and then stored in 1-2-ml aliquots at - 20°C or below.

Titration of virus control

It is essential to determine the actual quantity of virus used in the test. For this purpose, mix 0.5 m1 of the challenge vrrus (dilution 1 : 1380 = 500 LD,, per 0.03 ml) with 0.5 ml of diluent containing 0.2% gelatin or 2% normal horse serum. This yields a dilution of challenge virus that contains approximately 250 LD,, per 0 03 ml, Label this tube "0" and prepare three serial tenfold dilutions and label the tubes " - l " , " - 2" and " - 3". After shaking, incubate the tubes containing the sera under test and the control tubes at 37°C for 1.0-1.5 hours.

Inoculation of mice

1 Remove the tubes from the incubator and place in a vessel filled wlth Ice water 2 Inoculate groups of mice weigh~ng 14-16 g each intracerebrally with 0 03 m1 of

each d l u t ~ o n using at least 10 mice per dilution This should be done both for the sera under test and for the virus control When a large number of sera are being tested, it is important to make the injections of the virus control after having injected half the sera this ensures that the dilutons of the vlrus control and the sera under test are kept for the same average length of time before making the injections

3 Place each group of mice in a labelled cage and keep under observation for 14 days Record the number of mice that die after the flrsl 5 days


The following example is based on the titration of an experimental batch of therapeutic serum with dilutions of the challenge virus referred to in the example on page 418

' If the levels of antlbody are expected to be low. as in vaccinated persons, serum dilutions starting with 1 : l0 or 1 20 are employed and Ihe final virLis dilution s h o ~ ~ i d be approximately 50 LD,, Available to national laboiatories on request, from ti le international Laboratory for Biological Stan- dards. State Serum Institute. 80 Amager Boulevard, DK-2300 Copenhagen S Denmark


Determination of LDS, actually used in the test

Number of mice

Tube Died Survived

Applying the same calculatioii as in Exarnple 2 in Apperdix 3 (Table A34 page 452 and Table A3 3 page 450) the 50% end-point dilution of the birus is found to be 1 0 - ~ The antilog of 2 3 ( = 200) is the actual number of LD,, used in the reaction mixtures for neutralization

Calculation of the of the serum under test

This type of calculation 1s illustrated in Appendix 3 by Example 6, Table A3.5, page 453: dilution factor = 1 :2, In this case, the total nurnber of survivors is determined and the 509'6 end-point dilution is estimated by using Table A3.1 (page 448). For the serum iinder test, the ED,, is found to be 10-3 or 1 :l620 The ED,, of the reference serum is similarly estimated from the tables

Example 7 in Appendix 3, page 453, illustrates the calculation of the ED,, for a low-titre serum diluted 1.5, 1 :25, 1 ,125 and 1 .625 This range of serum dilutions is frequently used for the postvaccina antibody assay of human sera. However, use of a fivefold dilution scheme increases the variabil~ty of the test, It !s therefore recommended that a twofold dilution scheme be used wherever possible.

The number of International Units (IU) contained in the current International Standard for Rabies Immunoglobulin, Human, is fixed at 59 IU per ampoule ( 1 ) . To express the potency of a serum under test in IU, its neutralizing power must be compared with that of the International Standard or preferably, as the stock of international reference preparation is lim~ted. w ~ t h that of a reference serum that has been calibrated against the International Standard. This is done by calculating the difference between the logarithms of the 5OC% end-point dilutions of the two sera. In example 6 referred to above, the 50% end-poinl d~lution of the serum under test is 10 3.2'. Assuming the 5096 end-point dilution of the reference serum to be 10-~.'~, the calculation becomes:

The serum under test is therefore 10°,02 = 1.05 times as potent as the reference serum, so that its potency I S 1.05 X 59 = 62 \U per rnl.

Currently used potency tests

The mouse neutralization test (MNT) and the rapid fluorescent focus inhibition

test (RFFIT) (see C h a ~ t e r 15) are the two rriost widely iised potency tests for

POTENCY TEST FOR ANTIRABIES SERUM AND IMMUNOGLOBULIN

antirabies serum and immunoglobiilin. Both neth hods are based on the neutralization of rabies virus as described above followed by the use of mice (MNT) or cell cultures (RFFIT) as the indicator system.

Three laboratories (3-5j reported significant differences in the potency of antirabies serum and immunoglobulin preparations when tested by the MNT and RFFIT. AS a result, a collaborative study was ii it iated by WHO to investigate this problerri (6; see also Chapte~ 2, page 15). which indicated that the more variable

of the two tests (MNT) was in some way responsible for the discrepancy (7). However, the MNT continues to be favoured by niany national control laboratories for testing the potency of antirabies serum and immunoglobulin preparations because of fears that use of the RFFIT w ~ l l underestimate the potency of the

antibody and result in administration o i a therapeutic dose significantly greater than the recommended dose of 20 IU per kg of body weight (8). This could result in the inhibition of the antibody response in persons receiving post-exposure vaccination. i f rabies serum or immunoglobulin is given with the rabies vaccine.

These problems can be reduced by using a homologous reference preparation. For example, although significant differences were reported between the MNT and RFFIT when human immunoglobulin preparations were tested against an equine reference serum (3, 5), no such difierences were found when equine antisera were tested (3). Therefore, laboratories intending lo replace the MNT by the RFFIT must first deterinne, using appropriate (homologous) reference preparations. whether the two tests give eq~iivalent results. This is especially important i f they will be testing human rabies imrni~noglobulin preparations

References

1. WHO Expert Committee on Bioiogical Standardization. Thirty-f,Wh repot-t. Geneva, World Health Organization, 1985 (WHO Technical Report Series, No. 725).

2 Fitzgerald €A, Rastogi SC. A collaborative study to establish an International Standard Rabies Immunoglobulin of human origin. Journal of biological standardization, 1985, 13: 327-333.

3. Haase M, Seinsche D, Schneder W. The mouse neutralization test in comparison with rapid fluorescent focus inhibition test: differences in the results in rabies antibody determination. Journal o f biological standardization, 1985, 13: 123-1 28.

4. Kurz J et al. Comparative studies of two potency tests for antirabies serum: neutralization test in mice (MNT) and rapid fluorescent focus inhibition test (RFFIT). Developments in biological standardization, 1986, 64: 99-1 07.

5. Glijck R et al. Human rabies immur~oglobulin assayed by the rapid fluorescent focus inhibition test suppresses active rabies imnlunization. Journal of biological sfandardizatio,7, 1987, 15. 177-1 83.

6. Potential problems with unitage of the International Standard of Rabies lmmunoglobulin. Weekly epidemiological record, 1986, 61: 375.


7. Lyng J, Weis Bentoi? M, F~tzgerald EA. Potency assay of antibodies against rabies. A report on a collaborat~ve study Journal of biological standardization. 1989. 17: 267-280.

8. WHO Expert Coinmittee on Rahles. Eighth reporl. Geneva, World Health Organl- zaiion, 1992 (WHO Techn~cal Report Series, No. 824). Anriex 2.

Appendices

APPENDIX 1

Simple technique for the collection and shipment of brain specimens for rabies diagnosis J. Barra t '

Introduction

Laboratory diagnosis of rabies is usually performed on brain samples For the diagnosis to be accurate however the sample must be well preserved during shipment to the laboratory This requires that the an~mal s head or entire body should be rapidly cooled and kept cold during shipment On arrival at the laboratory, the animals skull is opened and b r a n samples are taken and tested by various diagnostic techniques as described in Part II (Chapters 4-10) However sometimes it is difficult to meet all the above requirements for example in tropical countries or when epidemiological studies are conducted in the field This appeqdix therefore describes several techniques for collecting preserving ship- ping and testing b r a n speciniens under difficult conditions

Collection of samples

Collecting brain samples by opening the skull is a safe procedure when performed by trained technicians with appropriate equipment in a room specially designed for the purpose, It is, however, more hazardous when practised outside the laboratory. A technique for collecting brain samples without opening the skull has therefore been developed. The samples are removed via the occipital foramen (foramen magnum) using a drinking-straw (5 mm in diameter) as follows ( 1 ) :

1. If the carcass of the suspect animal is intact, remove the skin from the top of the animal's head (Fig. A l . l ) . Bend the head downwards and cut the cervical muscles (Fig. A1.2) transversely as far as the vertebrae betiind the external occipital protuberance. Leaving the animal's head in this position, open up the atlanto-occipital joint by cutting the dorsal atlanto-occipital membrane and then the meninges (Fig. A1.3). If the head has already been removed from the body, expose the occipital foramen in the same way.

2. To take the samples. insert the straw into the occipital foramen (A1.4) with a slight twisting movement towards one of the eyes. This will allow samples to be taken from the medulla oblongata, the base of the cerebellum, the hppocam- pus (Arnmon's horn) and the cerebral cortex (Fig. A1.5). Do not use a mouth pipette or suction apparatus (which could become contaminated).

3. Before withdrawing the straw, pincl? it between the fingers in order to prevent brain tissue escaping from the straw.

'Head, Rabies D I ~ Q ~ O S I S Section, Laboratore d'Etudcs sur la Rage et a Palhoioy~e des Anirnaux sauvages. Centre Natonal d E i u d e s v6t6inaires e! ai~nierrtaires (CNEVA). MalzBv~lle. France.

425


Fig. A l , l Removal of the skin from the top of a slog's head exposing the muscles covering the joint between the occipital bone and first cervical vertebra

Fig. A 9.2 Cutting the cervical muscles

Fig. A 1.3 Exposure of the external occipital protuberance

APPENDIX 1

ci,#p.laI prnti~1ieinni.r~

By courtesy of J. Barrat.

Fig. A 1.4 Exposure of the occipital foramen 7 occ

Ey courtesy of J. Barrat

After withdrawing the straw, place i t In a 7-ml polystyrene tub diameter) containing a preservative, as described below.

Preservation and shipment of specimens

Two solutions are available for the preservation and shipment of brait

427


Fig. A1.5 Saggital section of a dog's head, indicating the angle of insertion of the straw

4- . . . - - Da!ec;~on n whrh thp straw should bc n s c l c d n t 3 the b-an

B.v courtesy of J. Bairat.

A choice has to be made according to the diagnostic techniques that will be used in the laboratory.

Formaldehyde inactivates the rabies virus and may be used when the sample is to be tested by the fluorescent antibody (FA) technique (after treatment with trypsti) or histologically. Glycerol does not inactivate the virus and may also be used when the sample is to be tested by the FA technique, or inoculated into animals or cell cultures.

Formaldehyde solution

The formaldehyde solution may be prepared from paraformaldehyde (4% (w/v) solution in 0.01 mol/l phosphate-buffered saline or PBS, pH 7 4'); alternatively, the solution may be prepared using a commercially available 37% formaldehyde solution (10% (vlv) in deionized water). When formaldehyde is used, the brain sample is squeezed out of the straw into the tube containing the solution with the aid of forceps and a glass rod (Fig. A1.6). The tube is then sealed with a stopper.

The preservative is a 50% solution of double-distilled glycerol in 0.01 mol/l PBS, pH 7.4, supplemented with 0.1 gll t h i ~ m e r s a l . ~ When this preservative is used. the brain

'See Chapter 8 Anriex 1 'AISO knowp as rnercurotl i~olate and t h ~ m e r o s a l

428

APPENDIX 1

Fig. A1.6 Transferring a specimen from the straw into formaldehyde solution

By courtesy of J Barrai

sample is not extracted from the straw. The straw is cut according to the length of the sample (Fig. A1 7). The sample is then Immersed in the tube containing the preservatve. The straw should be cut so that the stopper car1 be pushed in firmly.

Examination of the samples

Samples preserved in formaldehyde

Histopathological diagnosis by Mann's technque for example, I S performed after embedding the sample in paraffin. Before the FA test can be performed, however, the sample should be treated w t h trypsn (2, 3). as follows:

1 Grind a 3-mm lengtii of the sample In 5 rnl of 001 mol I PBS at pH 7 4 2 Centrifuge the sample at 1000 g for 10 mnii tes and dlscard the supernatant


Fig. A1.7 Preserving a specimen coNected by the straw technique in glycerol solution

By courtesy of J Barrat

3. Reconstitute the pellet in 5 rnl of 0.25?/0 trypsln solution in 0.01 rnol/l PBS, pH 7.4, without calcium or magnesium. Store the tube at 4'C overnight.

4. Centrifuge the tube again at 1000 g fo r 10 minutes and discard the supernatant. Wash the pellet twice in 0.01 mol:/l PBS pH 7.4, without calc~um or magnesium The FA test is then performed on a dried and fixed smear from the pellet.

The above technique was tested in the following way on day 0 fox brains received for diagriosis were processed and examined by the FA test as fresh unfixed specimens E~ght samples were taken from each posit~ve brain and fixed In formaldehyde solution and kept at room temperature (20°C) After 2 5 8 and 20 days a sample was examined histologically by Mann's techn~qbe and another was treated with trypsin as described above and then examined by the FA test

APPENDIX l

Table Al . 1 Results of FA test performed on rabid brain samples atler 2,5,8 and 20 days in formaldehyde solution at room temperature

Test Positive results

2 days 5 days 8 days 20 days

FA with grinding 100% (98198) 97% (95198) 98% (96198) 91 % (89/98) FA w ~ l h o u l grinding 92% (48152) 75% (53171) 850/0 (60171) 62% (44171)

Table A1 . l shows that the sensitivity of the FA test was increased by grinding the specimens before testing (higher percentage of positive results). The fluorescence was also brighter and more abundant when samples were ground.

Fifty of the 98 ground samples were also tested histologically (by Mann's technique) before treatment with trypsin. When both tests were used, none of these samples tested negative after 20 days in formaldehyde solution.

Comparison of coilect~on techntques The straw technique was evaluated as follows On day 0 fox brains received for d~agnosis were processed and examined by the FA test as fresh unfixed specimens Samples were taken from the positive brains before and after opening the skull and stored in formaldehyde solution at room temperature for 8 days They were then examined by the FA test 95% (1021107) of the samples collected by the straw techn~que tested positive, compared with 98% (96198) o i those taken after opening the skull

Comparison of laboratory techn~ques After 8 days of storage in formaldehyde, all of the positive specimens tested positive when examined by the FA test, whereas only 98% tested positive by histological examination (Mann's technique). The percentage of positive results was reduced to 96% for the FA test after 20 days of storage, but remained constant for histology.

Samples preserved in glycerol

Glycerol does not inactivate rabies virus; immunochemical techniques may therefore be used in ~n vitro (4, 5) as well as in vivo inoculation tests. Before these tests are carried out, however, the brain specimen should be washed in 0.01 moli! PBS, pH 7.4, to eliminate as much glycerol as possible. The FA test or inoculation tests are then performed as usual.

Samples were collected from 130 rabid foxes and stored in glycerol solution for 8 days at either 20°C or 37°C. A I the samples were found to be positive by the FA test and neuroblastorna cell inoculation. regardless of the storage temperature. Brain samples should not be removed irom the straw when kept in glycerol. If they are left unprotected in glycerol, it is more difficult to handle the specimens and the percentage of positive results is reduced (18% of positive samples tested negative after 8 days of storage at 37'C).


References

1 . Barrat J, Blancoii J. Simplified technique for the collection, storage and shipment of brain specimens for rabies diagnosis. In: Thraenhart 0 et al., eds Progress in rabies contro/. (Proceedings of the Second Internai~onai lMV1 Essen /WHO Symposium 017 "New Developments in Rabies Control". Essen, 5 7 July 1988 and Report of the WHO Consultation on Rabies, Essen. 8 July 1988.) Royal Tunbridge Wells, Wells Medical. 1989: 185-191.

2. Barnard BJH, Vosges SF. A simple technique for the rapid diagnosis of rabies in formalin-preserved brain Onderstepoort journal of veterinary research. 1982, 49: 193-194

3. Umoh JH, Blenden DC Immunofluorescent staining of rabies virus antigen in formalin-fixed tissue after treatment with trypsin. Bulletin of the World Health Organization, 1981. 59: 737-744.

4. Barrat J et al. Diagnostic de la rage sur culture cellulaire, comparaison des resultats de I'inoculation au ~ieuroblastome murin et de l'inoculation a la sours. [Diagnosis of rabies in cell culture, comparison of results of inoculation of mouse neuroblastorna c,ells and mouse inoculation.] Comparative immuriol- ogy, microbiology, and infectioiis disease, 1986, 1 1 207-214.

5. Portnoi D, Favre S, Sureau P. Use of neuroblastoma cells (NMB) for the isolation of street rabies virus from field specimens. Rabies information exchange. 1982. 6: 35-36.

APPENDIX 2

Techniques for the preparation of rabies conjugates P. Perrin '

General considerations

This appendix describes techniques for the production of purified rabies r b o - nucleoproten (RNP or nucleocapsid) and glycoprotein (G protein) the produc- t ~ o n of sera against these antigens the purification of immunoglobulin G (IgG) and the conlugation of IgG with fluorescein isothiocyanate (FITC) peroxidase or biotin

These lgGs and conjugates can be used in

- the fluorescent antibody (FA) test for diagnosis (see Chapter 7); - the enzyme-linked immunosorbent assay (ELISA) for the diagnosis of serotype

1 lyssaviruses and, to a lesser extent, rabies-related viruses (rapid rabies enzyme immunodiagnosis or RUED and RREID-blot), and for the diagnosis of lyssaviruses, including European bat lyssaviruses (EBL) and Mokola viruses (RREID-lyssa) (see Chapter 9);

- the ELISA for testing the potency of rabies vaccines (see Chapters 41 and 42).

The identification of rabies antigens in the tissue of infected animals by the FA test was first reported by Goldwasser & Kissling in 1958 ( 1 ) and subsequently described by Dean & Abelseth (2) Dean & Abelseth used inactivated, infected mouse-brain suspension as the source of antigen and obtained antibody preparations after treatment with ammonium sulfate. This method was improved by using purified RNP for immunization of rabbits, and purifying the resulting anti-RNP antibodies. Rabbit anti-RNP lgGs have been used in both the FA test (3) and RREID (4, 5). The RREID has recently been amplified by using biotin conjugate and extended to all lyssaviri~ses by using a mixture of polyclonal antibodies against various lyssaviruses (6).

Rabies G protein is the major antigen capable of inducing rabies virus- neutralizing antibodies (7) and conferring protection against an intracerebral challenge (8, 9). Furthermore, it can be ttrated by the ELISA for testing the potency of rabies vaccines (70, 1 1 ) .

Production and purification of polyclonal antibodies

Production and purification of rabies virus

1. Seed approximately 5 X 106 BHK-21 C13 cells in a 150-cm2 culture flask and add Dulbecco's minimum essential medium (DMEM; see Chapter 11, Annex)

'Lyssav~rus Labora!ory Pasteur i r s t t t ~ t e . Paris Frarice

433


containing 300 pg of glutamine (G) and 30 pg of gentamicin (genta) per ml and supplemented with 5% fetal calf serum (FCS) and 5% newborn bovine serum. Incubate at 37'C for 3-4 days or until a monolayer is formed.

2, Tryps~nlze the cells by adding 0.1% trypsin solution. Stop the reaction by adding DMEM-G-genta supplemented with 10% FCS.

3. Centrifuge the cells for 8 miniites at 500 g and resuspend in the same medium supplemented with 5% FCS.

4. Pooi the resuspended cells lo give approximately 2.4 X l o 9 cells (about thirty 150-cm2 flasks are required for the infection of six roller bottles), place in a sterile glass or plastic bottle, and adlust the cell concentration to 5.3 X 1O6 cells per ml.

5. Inoculate the cells in suspension with the Paris Pasteur virus (BHK strain) at an input multiplicity of infection of 0.1 plaque-forming units (PFU) per cell or 0.3 MICLD,, (median lethal dose for mice inoculated by the intracerebral route) per cell. incubate for 1 hour at 34°C with gentle agitation.

6. Adjust the ceil concentration to 2.7 X 106 cells per ml and the FCS concentration to 10% by diluting the culture medium twofold with DMEM-G-genta supplemented with 15% FCS.

7. Seed each 850-cm2 roller bottle w ~ t h 150 m1 of the infected cell suspension (4 X 10' cells or 26.5 ml per 150-cm2 flask) and incubate at 37°C for 24 hours.

8. Discard the cell-culture medium and replace with fresh medium supplemented with 0.3% bovine serum albuniin (BSA) instead of FCS (70 ml per roller bottle). Incubate the roller cultures at 34°C for 24 4 8 hours to produce RNP, or for 48 hours to produce rabies virus.

9. Harvest the cell supernatant and clarify the harvested virus by centrifuging at 900 g fo r 10 minutes, Inactivate the virus by adding P-propioactone (BPL) to a i inal concentration oi 1 :4000.

10. Concentrate the vlral particles by centrifuging the clarified supernatant at 80000 g for 90 minutes on a 50% sucrose "cushion" (dissolved in NT buffer; see Annex 1). Collect the band corresponding to the concentrated virus with a capillary tube connected to a peristaltic pump, dilute threefold with NT buffer and layer onto a linear 20-50% sucrose gradient. Centrifuge for a further 2 hours at 80000 g collect the band corresponding to the whole viral particles (the lowest band in the middle part of the gradient) as above and store at -20'C.

For the production of rabies-related viruses such as EBL or Mokola virus, the above procedure should be modified as follows Instead of BHK-21 C13 cells, BSR celis (a cione of BHK-21 cells (12)) are nfected in suspenson at a multiplicity of

infection of 0.3 MICLD,, and, after 1 hour of incubation, are grown in flat- bottomed cell-culture flasks containing DMEM-G-genta supplemented with 1096 FCS, and adjusted to pH 7 with 0.5 ml/l sod~urn b~carbonate (NaHCO,) (see Annex 1) . EBL and Mokola virus have a relatively slow multiplication cycle and should be harvested later than Pasteur viruses.

Production and purification of rabies ribonucleoprofei~p (RNP)

The rabies RNP generated by virus infeciion IS extracted from the infected cells and purified according to the technique of Compans et al (13). as modified by Sokol

( 7 4 )

APPENDIX 2

1 . Inoculate about 2.4 X log BHK-21 C13 cells with Paris Pasteur virus (BHK strain), as above. lncubate in roller cultures at 37-C for 24 hours and at 34°C for 24-48 hours, and then collect the infected cells by scraping.

2. Wash the cells twice in 40 ml of NT buffer, pH 7.6, and centrifuge at 900 g and 4°C for 10 minutes.

3. Resuspend the cell pellet (about 5 ml) in 5 ml of cool distilled water (4-6 "C) and incubate for 1 hour at 4 "C.

4. Disrupt the cells by 8 strokes in a glass Dounce-type homogenizer and centrifuge at 900 g for 10 minutes. Collect the supernatant and store it at 4°C.

5. Resuspend the cell pellet in 5 ml of NT buffer. Disrupt the cells again (8 strokes) and centrfuge as above. Repeat this step twice and mix tile supernatants together.

6. Transfer the pooled supernatants into a Corex-type tube and centrifuge at 9000 g for 10 minutes.

7. Distribute the supernatant in 3-4-m aliquots into two 5-ml polycarbonate tubes each containing 2 g of caesium chloride (CsCI). When all the caesium chloride has completely dissolved, increase the volume of the tube contents to 5 ml by adding NT buffer and centrifuge at 125 000 g and 4°C for 22 hours. The RNP will form a band in the middle part of the gradient, which can be collected with a capillary tube connected to a peristaltc pump. I f necessary, the RNP can be purified further by another centrifugation in caesium chloride: 1.8 ml of the RNP band IS added to 3 ml of NT buffer containing 1.3 g of caesium chloride and centrifuged as above.

8. Dialyse the band corresponding to RNP against NT buffer for 48 hours at 4OC (see Annex 2). After determining the total protein concentration, dilute the RNP in NT buffer to a final concentration of 800 pg per ml. Total protein content is measured with the Lowry technique (15) and the RNP antigen content is determined by the RREID (see Chapter 9).

9. Distribute the diluted RNP into 500-p1 aliquots, and store at - 20°C.

Production and purification of rabies glycoprotein

The rabies G protein is prepared with the Paris Pasteur virus (BHK strain) as previously described (9, 16), using purified virus obtained as described above.

1. Solubilize the G protein from the intact purified virus by adding an equal V O I U ~ ~ of 4% octyl-p-D-glucopyranoside (OGP) dissolved in NT buffer, pH 7.6, containing 0.45 m o j l sodium chloride (NaCI). lncubate for 1 hour at room temperature.

2. Layer the mixture onto a 25% sucrose "cushio~i" and centrifuge at 200000 g fo r 70 minutes to sediment the viral cores.

3. Layer the supernatant containing G protein onto a sucrose gradient consisting of threezones: on the bottom, 1 m1 of 60% sucrose; in the middle, 8 ml of 5-25% linear sucrose gradient; and on top. 1 ml of 3% sucrose. The sucrose solutions contain 0.5 mol/l sodium chloride and 2% OGP. Centrifuge at 150000 g for 36 hours.

4. Collect the resulting fractions in 0.45-ml volumes from the bottom of the tube and determine the G protein content of each fraction using the ELISA (see Chapters 41 and 42).


5. Pool the fractions containing G protein and dialyse against NT buffer for 48 hours at 4'C, and then sterilize the purified G protein by filtration on a 0.20-pm Milex-type filter (see Annex 2).

6. Aiter measuring the protein content of the purified preparation with the Lowry technique (15), store at 4°C until use.

Production and purification of rabbit anti-RNP IgG

New Zealand rabb~ts (male 3 kg) are inimunized at weekly intervals with 1 m1 of an emulsion containing 400 pg of ribonucleoprotein in complete Freund s adjuvant (equal volumes) for 1 month The injections are given intramuscularly The rabbits are then given a series of booster injections (without adjuvant or with incomplete Freund S adjuvant) at monthly intervals, and bled ior control 1 2 weeks after the last booster injection The blood samples are taken using an 18-gauge needle inserted into the central ear artery Prior to the bleedings the dorsal surface of the ear is brlskly massaged over the artery and cleaned with 70% ethanol

Serial twofold dilutions of immunized and non-immunized rabbit sera are tested by indirect immunofluorescence on mouse brain smears or cells infected with the Challenge Virus Standard (CVS) or by ELSA in plates sensitized with purified RNP The titre of the seruln is expressed by the reciprocal of the highest dilution that gives the most intense fiiorescence For a serum to be suitable its titre should be no less than 1 1500

The rabbit anti-RNP IgG is then chromatographically purified on quaternary aminoethyl (QAE) Sephadex A 50 (cross-linked dextran) as described by Joustra & Lundgren (17) Briefly the procedure is as follows

1 Dilute the sera twofold in EDAA buffer (see Annex l ) , pH 7.0, and dialyse it against the same buffer overnight at 4 'C

2. Fill a glass column with QAE-Sephadex (80 m1 of equilibrated gel per 50 m1 of diluted and dialysed serum) equilibrated in the same buifer. Layer the serum onto the column and collect the resulting fractions in 2-m1 volumes. The fractions corresponding to the peak of elution of the IgG are determined by their absorbency at 280 nm. In these conditions. all the serum proteins are bound to the beads except IgG, which is eluted with the loading buffer.

3. Pool the fractions with an optical density greater than 1.0. Add an equal volume of saturated ammonium suiate solution, pH 7.0 (see Annex l ) , with agitation, to precipitate the IgG.

4. Leave the suspension to stand for 30 minutes at 4"C, and then centrifuge for 30

minutes at 1500 g. 5. Dissolve the pellet in a minimum volume of distilled water, and dialyse against

0.01 rnolll phosphate-buffered saline (PBS), pH 7.0 (see Chapter 8, Annex).' 6. Test for the absence of animonium sulfate in the dialysate using Nessler's

reagent (see Annex 2), and determine the protein concentration of the purified preparation by the Lowry method (15). The latter should be at least 25 mglml for conjugation with fluorescein.

'The pH s h o ~ ~ i d he adjusted to 7 0 , usng diI~i!e hydrochiorc acid ( i iC l ]

APPENDIX 2

7. Filter the IgG preparation on a 0.22-pm Millex-type filter (see Annex 2) and store at 4'C, or dilute twofold in 50% buffered glycerol, pH 7, and store at -20°C.

Production and purification of rabbit anti-G protein IgG

New Zealand rabbits are immunized with purified G protein (150 pg per injection), as above. The sera are tested by the RFFIT (see Chapter 15); sera exhibiting a virus-neutraliz~ng antibody titre higher than 100 International Un~ts (IU)/ml are selected for IgG purification. as described above.

Production and purification of mouse monoclonal antibodies

Anti-rabies rnonoclonal antibodies (MAbs) are produced in mice as ascitic fluid, as described in Chapter 11. The method described here is used to obtain semi- purified IgG which can be conjugated with fluorescein or peroxidase. All the purification steps are carried out at 4'C.

l . Clarify the ascitic fluid by centrifugation at 5000 g for 15 minutes. 2. Add an equal volume of saturated ammonium sulfate solution, pH 7.0 (see

Annex l), with stirring, to precipitate the MAbs Leave to stand for 1 hour: and centrifuge again at 5000 g for 45 minutes to sediment the precipitate.

3. Resuspend the pellet in 0.01 rnol/l PBS, pH 7.0 arid dialyse against the same buffer for 48 hours (see Annex 2).

4. After dialysis is complete, centrifuge the dalysate at 5000 g for 45 minutes to eliminate the precipitate, which does not contain more than 2% of the required antibodies. Filter the supernatant on a 0.22-pm Millex-type filter (see Annex 2) and store at 4"C, or dilute twofold in 50% buffered glycerol, pH 7, and store at - 20 "C.

Production of antirabies IgG conjugates

Production of IgG-fluorescein isothiocyanate

1 . Before conjugation, equilibrate the IgG preparation by adding 0.1 volume of 0.5 mol/l carbonate buffer, pH 9.0 (see Annex 1).

2. Dissolve the fluorescein isothiocyanate (FITC) in a minimum volume of 0.05 moll1 carbonate buffer, pH 9.0 (see Annex l ) , and add to the IgG preparation (1 mg of FITC per 100 mg of IgG).

3. After mixing. adjust the pH to 9.0 with 0.5 molll sodium carbonate (Na,CO,), and adjust the protein concentration to 20 mg/ml with 0.05 moll1 carbonate buffer. The mixture should be stirred at 4'C overnight.

4. Filter the mixture through a G-50Sephadex column (previously equilibrated w ~ t h 0.01 rnolll PBS, pH 7.0) to eliminate unconjugated FITC (see Annex 2).

5. Pool the first coloured fractions (which contain the conjugated IgG molecules) eluted from the column, sterilize by filtration on a 0.22-pm Millex-type filter (see Annex 21, and store at 4'C (or dilute twofold in 50% buffered glycerol, pH 7 . and store at -20°C).

6. Titrate the fluorescent conjugate by direct immunofluorescence on mouse- brain smears infected with CVS or on infected cells. The working dilution chosen


should correspond to the highest dilution that gives a strong fluorescence on infected cells and a very low background fluorescence on uninfected cells. In these conditions, the FlTC conjugate must be diluted between 100-fold and 300- fold. Between 1.3 pg and 3 0 pg of FITC should be bound per mg of IgG (see Annex 2).

Absorption of anti-RNP NTC conjugate on mouse brain suspension To avoid nonspecific reactions when anti-RNP FlTC conjugate is used for diagnosis on brain smears, the conjugate should be adsorbed on brain suspension as described below.

1. Place a tube containing 10 g of uninfected mouse brains in an ice bath, add 30 m1 of cool sterile 0.01 mol/l PBS, pH 7.0, containing gentamicin (60 glml), and grind until a suspension is formed.

2. Adjust the volume to 50 m with PBS; then add the lgG-FlTC conjugate, with stirring. (The volume of conjugate to be added is calculated for 100 ml of ready- for-use conjugate adsorbed on a 10% brain suspension: i.e. 1 m1 i f the titre is 1 : loo. ) The mixture should be stored for 1 hour at room temperature.

3. Centrifuge the brain-conjugate suspension at 50009 for 1 hour. Carefully harvest the supernatant and store at 4°C. Resuspend the pellet in 25 m1 of PBS less the volume of the conlugate previously added. Stir for 45 minutes at room temperature.

4. Centrifuge the suspension again, and carefully harvest the supernatant and store at 4°C. Resuspend the pellet in 25 rnl of PBS, stir for 45 minutes at room temperature, and centrifuge again.

5. Pool the three supernatanis, and adjust the volume to 100 m1 with PBS. Inactivate the mixture with P-propiolactone (see Chapter 20, page 235), and then store at 4°C until use. (The mixture can also be freeze-dried for longer storage.)

Production of IgG-peroxidase conjugate

The IgG-peroxidase conjugate is prepared by the two-step conjugation technique ( 18) as follows.

1. Dissolve 10 mg of horseradish peroxidase in 0.2 m1 of 1 % glutaral diluted in phosphate buffer, pH 6.8 (see Annex l ) . Leave the preparation to stand for 18 hours at room temperature.

2. Filter the mixture through a G-25 Sephadex column (0.9 cm X 60 cm) equili- braied with a 0.15 molil sodium chloride (NaCI) solution (see Annex l ) . Pool the brown coloured fractions, which contain the activated peroxidase, and adjust the volume to 1 ml with the 0.15 rnol/l sodium chloride solution.

3. Dialyse the IgG preparation against the0.15 moll1 sodum chloride solution, and then adjust the concentration to 5 mg of protein per ml.

4. Add 1 m1 of the dialysed IgG to l ml of peroxidase solution and mix. Adjust the pH of the mixture to 9.5 by adding 0.5 molll carbonate buffer, pH 9.0 (see Annex l ) , and incubate for 24 hours at 4'C.

5. Dialyse the IgG-peroxidase conjugate against 0.01 molil PBS, pH 7.0, and then sterilize by filtration on a 0.22-pm Millex-type filter (see Annex 2). Store at 4"C, or dilute twofold in 50% buffered glycerol, pH 7, and store at -20°C.

APPENDIX 2

6. Titrate the peroxidase conjugate by the ELlSA as for the RRED test (see Chapter 9) using purified RNP or mouse-bran suspension infected with CVS. The working dilution chosen should correspond to the highest dilution that gives a high (2.0-3.0) optical density (OD) response with infected b r a n suspension (or RNP) and a very low background (OD=0.01-0 05). If the nonspecificispecific response ratio is higher than 0.10, the conjugate must be diluted between 500- fold and 5000-fold.

Production of IgG-biotin conjugate

In order to obtain a higher sensitiv~ty than with usual peroxidase conjugate IgG can be conjugated with b o t ~ n in order to use the biotin streptavdin-ampl~fied system The bot in is revealed by streptavidn conjugated with peroxidase (see Chapter 9)

Blotin-lgG conjugate IS prepared as described by Ternynck (19)

1 . Adjust the IgG concentration to 5 mglrnl, and dialyse the conjugate against 0.1 molil carbonate buffer, pH 8.5.

2. After dialysis, adjust the IgG concentratio.n to 2 rnglml with 0.1 molll carbonate buffer. Add 75 P I of b io tn N-hydroxysuccinirnidester (BNHS; 1.5 rng dissolved in 300 pi of anhydrous dimethylformarride) to 2.5 rnl of dialysed IgG. Mix the two solutions and keep, with stirring, at room temperature for 1 hour.

3. Filter the mixture through a G-25 Sephadex column (1.6 cm X 60 cm) equilibrated with 0.01 molil PBS, pH 7 0 to eliminate unconjugated BNHS.

4. Determine the protein content of the eluted fractions by measuring their optical density at 280 nm.

5. Pool the fractions containing gG--b iotn conjugate (OD greater than 0.8). 6. Add 1 mg of bovine serum albumin to 1 rnl of IgG-biotin conjugate. Dilute the

stabilized conjugate in the same volume of glycerol and store at 4°C until use.

The titration method and the criteria retained for biotin conjugate are the same as for peroxidase conjugates, except that the titre is generally 3-5-fold higher for biotin conjugates.

References

1. Goldwasser R , Kissling R. Fluorescent antibody staining of street and fixed rabies virus antigens. Proceedings of the Sooety of Experimentai Biology and Med~c~ne, 1959. 98: 21 9-223.

2. Dean DJ, Abelseth MK. The fluorescent antibody test. In: Kaplan MM, Koprowski H, eds. Laboratory techniques in rabies. 3rd ed. Geneva, World Health Organization, 1973 (WHO Monograph Series, No. 23): 73-84.

3. Atanasiu P et al. lmmunofluorescence and immunoperoxidase in the diagnosis of rabies. In: Kurstak E, Morisset R, eds. Viral irnrnunodiagnosis. New York, Academic Press, 1974: 141 155.

4. Perrin P, Rollin PE, Sureau P. A rapid rabies enzyme imrnunodiagnosis


(RREID): a useful and simple technique for the routine diagnosis of rabies. Jou1~7al of biological standardization, 1986. 14: 2 17-222.

5. Perrin P. Sureau P. A collaborative study of an experimental kit for rapid rabies enzyme imrnunodiagnosis (RREID) Buliefin of the World Health Organization, 1987, 65: 489-493.

6. Perrin P et al. A modified rapid enzyme immunoassay for the detection of rabies and rabies-related virus: RREID-lyssa. Biologicais. 1992. 20. 51-58

7. Wiktor T et al. Antigenic properties of rabies virus components. Journal of immi~t?ology, 1973, 110: 269-276.

8. Cox JH, Detzschold B. Schneder LG. Rabies virus glycoprotein 11. Biological and serological characterization, Infect~on and immunity, 1977, 16: 757-759.

9. Perrin P, Thibodeau i, Sureau P. Rabies irnmunosome (sub-unit vaccine) structure and iinrnunogencity. Pre- and post-exposure protection studies. Vaccine. 1985, 3 . 325-331

10. Thraenhart 0. Rarnakrist-inan K. Standardizat~on of an enzyme immunoassay for the in vitro potency assay of inactivated tissue culture rabies vaccines: determination of the rabies vlrus glycoprotein with polyclonal antisera Journal of biological standard~zation, 1989, 17: 291 -309.

11. Perrin P, Morgeaux S. Sureacr P lri vitro rabies vaccine potency appraisal by ELISA: advantages of the immunocapture method with a neutralizing anti- glycoprotein monocIor?al antrhody. B~olog~cals, 1990 18. 321-330.

12. Sato M et al. Plaqiie formation of herpes virus homnis type 2 and rubella virus in variants isolated from colonies of BHK-21iWI-2 cells formed in soft agar. Archives of virology, 1977, 53: 269.

13. Conipans R , Choppn P, The length of the helical nucleocapsid of Newcastle disease virus. Virologica 1967. 33 344-346.

14. Sokol F. Purification of rabies virus and isolation of its components. In: Kaplan MM. Koprowski H eds Laboratory techniquesin rabies. 3rd ed. Geneva. World Health Organization, 1973 (WHO Monograph Series, No. 23) 165-178.

15. iowry D et al. Protein measuremerit w!tli Folin phenol reagent. Journal of biological chemistry, 1951. 195: 265-275

16. Perrin P, Portnoi D Sureau P. Etude de l'absorption et de la penetration du virus rabique: interactions avec les cellules BHK-21 et des membranes artificielles. [Study of the absorption and penetration of rabies vlrus. interactions with BHK-21 cells and artificial membranes] Annaies de 1'lr;stitut

Pasteur: Virology, 1982, 133E: 403-422.

APPENDIX 2

17. Joustra M, Lundgren H. Preparation of freeze-dried, monomeric and immuno- chemically pure IgG by a rapid and reproducible chromatographic technique. In: Peeters H, ed. Proiides of the bioiogical fiuids, (Proceedings of the 17th Colloquium.) Bruges, Arrtchap, 1969: 51 1-51 5.

18. Avrarneas S, Ternynck T. Peroxidase-labelled antibody and FAb conjugates. lmmunochemistry, 1971, 8: 11 75-1 179.

19. Ternynck T. Avrameas S. Techniques mmuno-enzymatiques. [lmmuno- enzymatic techniques.] In: Techniques en immunologic. [Techniques in immunology.] Paris, INSERM, 1987. 87-92.

Annex 1 Preparation of buffers and reagents

Acetic acid, 1 mol/l Acetic acid solution, l .05 g l m l or 17.3 mol l l Distilled water

Saturated ammonium sulfate solution, pH 7.0 Ammonium sulfate D~stilled water

50.00 m1 815.00 rnl

Carbonate buffer, 0.05 mol/l, pH 9.0 Sodium bicarbonate solution. 0.5 mo!. 10-fold dilution 900.00 rnl

Adjust the pH to 9.0 with a 10-fold dilution of sodrum carbonate solution, 0.5 mol;'l, prepared as below.

Carbonate buffer, 0.1 mol/l, pH 8.5 Sodium bicarbonate solution, 0.1 mol/l, prepared as below 900.00 ml

Adji~st the pH to 8.5 with sodium carbonate solution, 0.1 mol/l, prepared as below.

Carbonate buffer, 0.5 mol/l, pH 9.0 Sodium bicarbonate solution, 0.5 mol l l , prepared as below 900.00 ml

Adjust the pH to 9.0 with sodium carbonate solution, 0.5 mol/l, prepared as below.

Ethylenediamine (EDAA) buffer, pH 7.0 Acetic acid solution, 1 rnol/l, prepared as above 73.00 m1 Distilled water 900.00 ml

Adjust the pH to 7.0 with ethylei-ted~amine, and then add distilled water to make 1000.00 ml.


50% Buffered glycerol, pH 7 Mix one volume of glycerol with one volume of 0.01 mol/l PBS, pH 7.0 (see Chapter 8, ~ n n e x ) . '

Nessler's reagent Nessler's reagent Sodium hydroxide (NaOH), 30% solut~on

NT buffer, pH 7.6 Sodium chloride (NaCl) 7.60 g Trometamol 6.06 g Distilled water 900.00 m1

Adjust the pH to 7.6 with hydrochloric acid (HCI) solution, 2 mol/l, and then make up to 1000.00 ml with distilled water.

Phosphate buffer, pH 6.8 Sodium phosphate, monobasic (NaH,PO,.H,O) solution, prepared as below 900.00 m1

Sodium bicarbonate solution, 0. l mol/l Sodium bicarbonate (NaHCO,) Distilled water to make

Sodium bicarbonate solution, 0.5 mol/l Sodium bicarbonate (NaHCO,) Distilled water to make

Sodium carbonate solution, 0. 1 mol/l Sodium carbonate (Na2C0,,10 H,O) Distilled water to make

Sodium carbonate solution, 0.5 mol/l Sodium carbonate (Na2C0,,10 H20) Distilled water to make

Sodium chloride solution, 0.15 mol/l Sodium chloride (NaCI) Distilled water to make

Sodium phosphate, monobasic solution, 0.1 mol/l Sodium phosphate, monobasic (NaH2P04,H,0) 13.80 g Distilled water to make 1000.00 ml

'The pH should be adjusted to 7 0 using dilute hydrochloric acid ( H C )

442

APPENDIX 2

Annex 2 Additional techniques

Visking dialysis tube treatment

1 Place the Visking dialysis tube in boiled distilled water for 3 minutes 2 Wash with distilled water at room temperature 3 Store at 4 'C in 0 15 mol I edetic acid ' 4. Before using the tube for dialysis, treat with 20?& glycerol in NT buffer (see

Annex 1 ) for 1 hour at 4 'C, and then wash with distilled water

Millex filter treatment for separation of glycoprotein

1. Filter with 10 m1 of 0.01 molll PBS, pH 7.0. containing 0.1% poysorbate 80 2 Filter with 10 m1 of 0.01 molll PBS. Repeat.

Filtration in Sephadex gel column

As an alternative to dialysis, columns made from Sephadex (cross-linked dextran) G-25 or G-50 gel may be used to remove ammonium sulfate. gutaral or unreacted fluorescein, as follows:

1. Stir approximately 100 g of Sephadex into 500 ml of 0.01 mol j l PBS, pH 7.0, contained in a l-litre flask. Leave to stand for at least 2 hours (to allow the gel to swell) and then remove the supernatant by suction.

2. Mix the gel thoroughly with approximately 900 ml of 0.01 mol:' PBS. allow to settle for exactly 5 minutes, then remove the supernatant by suction. Repeat the process several times until the supernatant fluid clears within 5 minutes. This procedure removes small particles of Sephadex that would otherwise reduce the flow rate of the Sephadex coliirnn.

3. Store the gel at room temperature until use.

Removai of ammonium suifate 1. For each 10 ml of globulin solution. use a column of Sephadex gel of 200 mm

X 15 mm, equilibrated w ~ t h 0.01 mol/ l PBS (see above). Collect fraction when the optical density (OD) exceeds 0.5 at 280 nm.

2. Record the volume and determine the total protein content of the fraction.

Removal of unbound FITC FlTC not bound to protein is rapidly and conveniently separated from globulins by use of a Sephadex column. For each 10 ml of conjcigated IgG a column of Sephadex gel of at least 150 mm X 15 mm is required.

1. Fix a glass column of appropriate length (approx. 300 mm X 15 mm) in an upright position.

2. Fill the column with Sephadex gel eqiliibrated with 0.01 mol j l PBS, pH 7.0 (see above).

3. Allow the gel to settle for several hours until the gel bed is evenly packed

' Also known as e!hyIenediarrline !etraacetate or EDTA

443


4. Rinse the column with 0.01 mol/l PBS, taking care that the Sephadex gel never becomes dry.

5. Before use, drain the buffer until the level has reached the top of the gel bed 6. Carefully layer the FITC-labelled globulins on top of the Sephadex gel. Avoid

stirring up the gel. 7. After the globulins have entered the gel, add 0.01 mol/l PBS on top of the

column. Exert enough pressure to ensure an even flow. 8. FITC-labelled globulins separate into two visible bands of orange colour. The

first fraction, which rapidly approaches the lower end of the column, represents the FITC-labelled globulins. The second, slower-moving fraction is unreacted dye. Collect all the first fraction.

9. Stop collecting when the colour intensity of the eluting fraction weakens, dispense in l -m l amounts, and store at 4°C (alter diluting twofold in 50% buffered glycerol, pH 7) or freeze-dried.

10. Columns may be re-used after the Sephadex has been freed from residual, unreacted dye.

Detection of ammoniuin sulfate For each m1 of dialysis sample, add 50 p1 of Nessler's reagent. The absence of precipitate indicates that all the ammonium sulfate has been eliminated.

Estimation of bound FlTC

Prepare a standard FITC solution and measure its optical density (OD) at 490 nm (maximum absorption of free FITC) and 496 nm (maximum absorption of bound FITC).

The concentration of bound FlTC (pg/ml) is determined using the following formula:

a X OD of the IgG-FITC conjugate at 496 nm

OD of the standard FlTC solution at 490 nm X 0.75

where:

a = protein concentration (pgiml) of the standard FlTC solution

The protein concentration of the IgG-FlTC conjugate (pg lml) is then determined using the formula ( I ) :

OD of the conjugate at 280 nm - (OD of the conjugate at 496 nm X 0.36)

Reference

1. Pillot J, Gisler R. La specificite des reactions d'immunofluorescence. I- Controle de differents facteurs in f luen~ant la specificite des conjuges fluo- rescents destines a I'etude des systemes cellulaires. [Specificity of immunofluorescence tests. l-Control of different factors influencing the specificity of fluorescent conjugates used in the study of cell systems.] Annales de I'lnstitut Pasteur, 1968, 115: 416 434.

APPENDIX 3

Methods for the calculation of titres M. F. A. Aubert'

The calculation of 50y0 end-point dilutions can be done using several methods of varylng complexity and accuracy, Among these methods. the method of Reed &

Muench, which is described in Chapter 38 (page 371), is not as simple as commonly stated though it is still widely used. Errors are often made, for example, in calculating the cumulative totals of surviving or dead mice, the percentages of non-decimal totals and the differences of logaritlinis. In addition, the calculation of cumulative totals means that any errors made on one dilution are extended to all others. It is therefore recommended that the method of Reed & Muench should be replaced by the Spearman-Karber method or the graphic method described here.

Spearman-Karber method2

This is the best known of the simple methods It is applicable only when the dilution factor is constant and when the range of dilutions IS sufficiently wide to include both the dilution at and below which 100% of the animals will usually be positive and the dilution at and above which 100% of the animals will usually be negative

Positive animals may be either those that die or those that survive depending on the type of test as follows

1 . Aflimais that die are cour7ted as positive in the case of - titration of live virus vaccine (Chapters 22, 23 and 39); - titration of challenge virus in the NIH test (Chapter 37); - titration of challenge virus in the quantitative assay and potency test of

antirabies serum (Chapter 47); - the in vitro virus neutralization test (Chapter 16); - the Habel test and the modified Habel test for potency (Chapter 38).

2. Animals that survive are counted as positive in the case of. - titration of sera by the "serum dilution--constant virus" technique (Chapter

47); - determination of the antigenic value of a vaccine by the "vaccine

dilution-constant virus" technique (NIH test, Chapter 37).

Procedure using the Spearman-Karber formula

The Spearrnan-Karber formula is as follows:

log,, end-point dilution = - ( X , - d/2 + d E i / n , )

' D~rector, Laboratore d'Etudes sur la Rage e l la Patl?oiog~e des Anlrnaux sauvages, Centre Nat~onal d'Etudes veterinalres et al~rnentaires (CNEVA) WHO Collaborating Centre for Researcti and Maiidys- rnent In Zoonoses Control, Malzev~lie, France Based on the chapter by R, J Loreni and K Boge In the prevlous editlon


where:

X, = log;, of the reciprocal of the lowest d~lution at which all animals are positive: d = log,, of the dilution factor; n, = number of animals used at each individual dilution (after discounting

nonspecific deaths); r, - number of positive animals (out of n, ) .

Summation is started at dilution X,.

For a material with a symmetrical dose-response curve, the value calculated by the Spearman-Karber formula is the dilution that would cause a positive response in exactly 50% of the animals in a very large population, i.e. the 50% end-point dilution. Sera and vaccines usually have symmetrical dose-response curves. On the other hand, the dose-response curves of virus suspensions are known to exhibit a certain degree of asymmetry. The result of this is that the end-point dilution of virus suspensions as calculated by the Spearman-Karber method is the dilution that would cause the death of exactly 43% of the animals, i.e. the 43% end- point dilution However, this value s just as suitable for measuring the infectivity of a virus suspension as is the 50% end-point dilution. In this chapter, the term "50% end-point dilution" or simply "end-point dilution" refers to the true 50% end-point dilutioti in the case of sera and vaccines and to the 43% end-point dilution in the case of virus suspensions. To avoid confusion, some authors prefer to use the term "mean effective dose" in this sense.

Example I . End-point dilutlon of a virus suspension dilution 10-3.6 10-4.6 10-5.6 10-6.6

r l 5 5 0 1 "i 5 5 5 4

loglo of the end-point dilution = - [4.6 - 1.012 + l .O (1 + 0.25)] = - (4.6 - 0.5 + 1.25) = - 5.35

End-point dilution = 1 0 - ~ . ~ ~

Example 2. End-po~nf dilution of a vaccine, NIH potency test dilution 10-0.7 10-1.4 10-2.1 10-2.8

r, 10 8 1 0 " 1 15 14 16 15

I f it is assumed that at dilution loO-' 16 animals would have been inoculated and that all of them would have been positive, then X, = 0.0 and d = 0.7, and:

log,, of the end-point dilution = - CO.0 - 0.712 + 0.7 (1 + 0.667 + 0.571 + 0.0625)] = - ( - 0.35 + 0.7 X 2.30) = - 1.26

However, this is only a rough estimate based on a possibly false assumption. It is equally possible that not all the an~mals inoculated with undiluted vaccine would have been positive, which would have given a lower end-point dilution. Such an

APPENDIX 3

assumption is without any theoretical basis, but is reasonable if applied with suitable caution. Nevertheless, it is preferable to repeat the titration with a more appropriate range of dilutions, and this is essential when there are serious shortcomings in the data, as in Example 2.

One of the main advantages of the Spearman-Karber method is that the formula can be applied using a pocket calculator. When no accidental deaths occur among the experimental animals, the calculation can also be made using tables. as described below,

Procedure using tables

The calculation of the 50% end-point dilution comprises the following steps:

1. Note the log,, of the reciprocal of the lowest dilution (of virus, serum or vaccine) at which all the animals are positive (i.e. i f the dilution is I O - ~ , take log,, 103 = 3).

2. Determine the total number of positive animals at the dilution defined in step 1 and at all higher dilutions.

3. Read from the table for the appropriate dilution factor (Tables A3.1-A3.3) the numerical value attributed to the number of positive animals determined in step 2.

4. Add up the values determined in steps 1 and 3. This total represents log,, of the reciprocal of the 5096 end-point dilution.

Tables A3.4 and A3.5 show several examples of titration and use of the Spearman-Karber method with tables.

Graphic method'

The graphic method is simpler and more direct than the Spearman-Karber method and provides more information about the distribution of the results of an experiment including how well they fit to the classical model of a dose response curve Another advantage is that ~t can be used even when the dilution factor is not constant and when the number of animals is not the same for all dilutions such as when nonspecific deaths occur

When the percentage mortality is plotted on the y-axis using a usual scale the resulting dose-response curve IS sigmoid Considerable skill is required to draw such a curve and to adjust it to all experimental po~nts However i f an appropriate scale (neoprobit units) is used the dose-response curve becomes a straight line and is therefore very easy to draw

' For fur!her Information, see reference l.


Table A3.1 Numerical values for calculation of titres: dilution factor = 2

Total no. No. of inoculated animals per dilution of positive atlimals n = 4 n = 5 n = 6

for each further positve a n m a l add 0 075

for eaci i further posjt.ve arirrial add 0 060

for each fdrther p o s ~ t ve anmal add 0.050

APPENDIX 3


Total no. No, of inoculated animals per dilution of positive animals n = 4 n = 5 n = 6 n = 1 6

for each f~~r ther postlve anmal add 0175

0 35 0 49 0 63 0 77 0 91 1 05

119 1 33 1 47 161 1 75 1 89 2 03 2 17 2 31 2 45

2 59 2 73 2 87 3 01 3 15 3 29 3 42 3 56 3 70 3 84

fob each f ~ r t h e r posltive anmal add 0 140

0 35

0 47 0 58 0 70 0 82

0 93 1 05 116 1 28 1 40 151 0 35 l 63 0 39 1 75 0 44 1 86 0 48 1 98 0 52

2 10 0 57 2 21 0 61 2 33 0 66 2 45 0 70 2 56 0 74 2 68 0 79 2 80 0 83 2 91 0 87 3 03 0 92 3 15 0 96

3 26 1 00 3 38 1 05 3 49 1 09 3 61 114 3 73 118 3 84 1 22

1 27 for each further 3, positiv? anmal

1 35 add01165 1 4 0

l 44 1 49 l 53 1 57 1 62

for each further pos tve an mal add 0 0437


Total no. No. of inoculated animals per dilution of positive animals n = 4 n - 5 n - 6 n - 8 n = l 0

0.50 0 75 1 .oo 1 25 1.50 1.75 2.00

2.25 2.50 2.75 3.00 3.25 3.50 3.75 4.00 4 25 4.50

4.75 5.00 5.25 5.50 5.75

for each further positve animal add 0.25

for each further postive anmal add 0.2

for each 4.00 3.1

further 4.13 3.2

positive 4.25 3 3

animal 4.38 3.4

add 0.167 4'50 3.5 for each 3 6 further 3 7 postive 3.8 anmal 3.9 add 0.125 4.0

for each further postive anmal add 0 1

APPENDIX 3

Procedure for determining the 50% end-point dilution

Using the data of Example 1 of Table A3.4 (page 452), the percentage mortality for each dilution IS determined as follows:

Dilution l~g,~(l /di lution) No. of mice Percentage mortality

pos.8 neg.

'Posltive animals are lhose t h a t die

The dilutions are plotted on the x-axis (Fig. A3.1 ) . The percentage mortality is plotted on the y-axis. This gives flve points that are

used in plotting the straight line. The aim is not to join all the points with a broken line, but only to judge approximately which is the line nearest ail the points. When more than one line may be drawn, it should be borne in mind that the most important po~nts are those nearest the 50% mortality level.

The 50% end-point dilution (D,,) corresponds to the dilution at which 50% of the inoculated animals die. In this example log,, (I/D,,) = 5.6 and D,, = 10 5.6

(see Fig. A3.1).

Fig. A3.1 Determination of 50% end-point dilution by the graphic method

3 4 5 6 7 8 logt0 (lidilution) WHO 94 794

r D m

Table A3.4 Examples of titration of virus suspensions (positive animals are those that die) o - - 3

Example 1 Example 2 Example 3 Example 4 Example 5 5 0

n = 5 n = 5 n = 5 n = l0 n- l 0 n <

-- -- -l m

Dilution 0 I

(d~lution POS neg POS neQ POS neg POS neQ PoS neg E factor = 10) 4!

Calculation

step: 1 2 3 C

4

3The lowest dilutlon at whlch all ar~lmdls are usually positlve IS assumed to be to0' ( I cl uridlluteti vlrus suspension) 1 hls may be justlfled In thls erample since Ihe range of vlrus dllutlons with both positives and nrqatives usualv covers 2 3 t e ~ f o l d steps only

1s assurnrd that no posltlvrr would have occurred at higher dilutions than 10 had thpy been Vsted C See Tahlr A3 ?

APPENDIX 3


Fig. A3.2 Neoprobif paper for determining the 50% end-point dilution by the graphic method

3 4 5 6 7 8 logl0 (lidilution) WHO 9490 1

Estimating the confidence interval

The graphic method allows a rough estimate to be made of the 95% confidence interval (Cl,,) of the D,, according to the iormula:

where:

Cl,, = 95% conidence interval: D,,, = log,, of the reciprocal of the iowest dilution at which 100% o i the

animals are positive; D, = log,, of the reciprocal of the highest dilution at which 0 % of the animals

are positive; n i = the number of anitnas used at each individual dilution (after discounting

accidental deaths): W , = the weighting of the corresponding percentage mortality.

D,,, and D, are directly read from the graph, In this example D, = 7.3 and D,,, = 3.9, Individual values of W , are read on the right-hand scale of the neoprobit paper.' on the level of the points of the plotted line corresponding to each individual dilution In this example W , = 0.04 for the 10 dilution. 0.52 for the 10-5 dilution. etc.

' Fig A3 2 is provded as a model for copylng

APPENDIX 3

Hence Cl,,

This method also allows a rough comparison to be made between the 50% end-point dilutions of two preparations titrated at the same time,

IDSO - Dkol When - - - - > 1 0,

J(c/,,)' + the difference between the preparations can be considered as significant

Considerations

Determining the 50% end-point dilution from a regression line plotted according to subjective judgement could be regarded as inaccurate. However, in a study involving several scientists with no previous experience in biological assays, it was shown that plotting regression lines "by eye" usually provides very satisfactory approximations (2). Moreover, a graph perniits inaccurate or paradoxical results to be readily identified, and leads to more cautious statements about the conclusions of an experiment. In conclusion, the graphic method has proved to be easy to use and very convenient for immediafe estimation of data.

More complex methods

The Spearman-Karber method and the graphic method are considered to be adequate for estimating 50% end-point dilutions when only an order of magnitude is required on a single titration. However, estimates obtained by these methods should not be used for statistical interpretation, e.g. precis~on of the estimates, comparisons between titrations within the same experiment or comparisons between experiments. In particular, it is almost impossible to test the parallelism of the dose-response lines of two preparations by these methods.

Computers are now widely available, and should help laboratories to shift to relatively more complex methods as a common practice.

As already stated, the dose-response curve is a sigmoid, which is not easy to adjust from experimental data. The methods described here therefore aim to transform this sigmoid into a straight line through an appropriate transformation of the percentage mortality (or survival) obtained for each dilution. Several such methods have been used, including angular, probit (see below) and logit transformation, which yield results that differ only at the second or third decimal.

Angular transformation

In angular transformation, the observed percentage mortality (or survival) ( p ) is transformed according to the following formula:

$ = 2 arcsin ,,I;


The va~.iance (S:,) is calculated from the formula l l n , where:

S$ =the variance of 4 at each dilution, n, = the r-iumber of animals used at each dilution

The main advantage of this transformation is that the variance of 6 is independent of p.

The calculation procedure is therefore simplified: this is basically a regression analysis with 4 for the-y-axisvalues and log,, (1 ;dilution) for thex-axisvalues. The D,, is deduced from the regression lirie

4 = b log,, ( l /dilution) + a, 7

with 4 = 2arcsin ,/l12 = n/2.

The standard deviation of D,, is calculated using the following formula:

where:

k = the number of dilutions used in the test; 17 = the number of animals used at each dilution; b = the slope of the regression line.

The number of dilutions used (k ) has to be considered only for the "useful" dilutions: dilutions yielding mortality rates of 100% or 0% among inoculated animals should not be considered. The number of animals used at each dilution (n) should be the same, although occasional nonspecific deaths at one or two dilutions can be tolerated.

Dilution log,, (l/dilution) pos. total 7 I$ = 2 arcsin J ri/ni

(ri 1 (nil

Values of q can be read from Table A3.6. Theoretically, for p = 0% and p = 100%, cp should be 0 and respectively.

However, the corrected values 2 arcsin J1/4n and n - 2 arcsin Jwn, which are given in Table A3.6, should be used.

Calculation of the regression Ihne gives: y = - 0.53~ + 4.54 log 0 (l /'D5,) = 110.53 (4.54 - 1.57) = 5.60 and Cl,, = 5.60 + 0.74

In order to compare several results obtained with the same scale of dilutions

APPENDIX 3

and with an equal number of animals in all dilutions, it is possible to use an analysis of variance (3 ) .

Probit transformation

Probit transformatioii is the most commonly used method of transformation. It was developed by Bliss ( 4 , 9 and subsequently modified by statisticians such as Finney

(2, 6). The first difficulty of probit transformation is that the variance of probt depends

not only on the number of animals used, but also on the percentage of positive results: then each value will contribute to the plotting of the regression line in proportion to its precision. The iterative process requires a calculator. If a computer is available, special software can be acquired for this purpose. WHO collaborating centres involved in vaccine potency testing may be able to provide technical assistance, and this method has been described in the literature

Bayesian analysis

This technique was originally developed by Racine et al. for the classification of toxic substances (7) and subsequently adapted to the analysis of NIH test results by Fluhler & Thraenhart (8). Bayesian analysis uses the results obtained during previous experiments ("prior information") to design and interpret further tests ("posterior information"). This enables a more precise estimate of the D,, to be obtained using fewer animals than with the logit or probit methods.

Another advantage of Bayesian analysis is that it provides the total range of probability of the D,, w~thin its fiducial limits, in contrast to the classicai methods, which provide mean and fiducial limit values only.

Data from an initial NIH test lead to the determination of five constants that contain information about the outcome of the test, including the D,, value and the variance. If the result is not satisfactory, a second test should be performed, but fewer animals are required. The new results will then be combined with the five constants obtained in the first test in Bayesian analysis, which increases the accuracy of the estimate of the 4,.

Table A3.6 Table of values of yl = 2 arcsin /p

No. of No. of inoculated animals per dilution (n,) positive animals ( r , ) 5 6 7 8 9 10

0 0.457 0.41 1 0.380 0.355 0.334 0.31 7 1 0.927 0.841 0.775 0.723 0.680 0.644 2 1.369 1.231 1.128 1.047 0.982 0.927 3 1.772 1.571 1.427 1.31 8 1.231 1.159 4 2.214 1.911 1714 1.571 1.459 1.369 5 2.690 2.301 2.014 1823 1.682 1.571 6 P 2.730 2 366 2.094 1.91 1 1 772

7 - - 2.761 2.419 2.1 60 1.982 8 - P 2.786 2.462 2.214 9 - - 2 806 2.498

10 - - - 2.824


Fig. A3.3 Probability density of NIH test results on four vaccines, as determined by Bayesian analysis

Potency (loglo (IU) + 1)

Because the normal (Gaussian) distribution for the D,, values is calculated, it is possible to deterniine the level of significance with which the potency values (in IU) of two or more vaccines d~ffer from each other. Differences between vaccines can be easily visualized (Fig. A3.3).

Bayesian analysis appears to be especially useful for determining the potency of standard vaccines such as product-specific reference vaccines (PSRV). A computer program for performing Bayesian analysis is available on request from the WHO Collaborating Centre for Reference and Research on Neurological Zoonoses, Essen, Germany.

Conclusion

The choice of method for calculating the 50% end-point depends on the facilities, equipment and personnel available in the laboratory. The graphic method is the easiest and quickest technique, requiring only a sheet of paper and a pencil.

The use of tables of numerical values facilitates the practice of the Spearman-Karber method, which is still widely used in many laboratories. Angular transformation may be employed i f a pocket calculator equipped with a suitable program for regression analysis IS available. However, i f a computer is available, it

is recommended that pioblt analysis and Bayesian analysis should be used.

References

1. Aubert M. Une methode simple de calcul des titres des suspensions virales, vaccinales ou seroneutral~santes: la methode graphique [A simple method of calculation of titres of viral, vaccinal or seroneutral~zing suspensions, the graphic method.] Revue scientifique et technique de /'Office international des Epizooties, 1982, 1 : 823-833.

APPENDIX 3

2. Finney DJ. Subjective judgement in statistical analysis. An experimental study. Journal of the Royal Statisti'cal Society, 1951, B13 284-297.

3. Beranger G et al. Utilisation de la transformation angulaire 2 arcsln 6 et de I'analyse de la variance dans les titrages des serurns et vaccins sur I'animal.

r [Use of angular transformation 2 arcsin J p and analysis of variance in the titration of sera and vaccines in animals.] Sciences techniques et pharmaceu- tiques. 1981, 10: 229-244

4. Bliss CL. The calculation of the dosage-mortality curve. Annais in applied b io iog~, 1935, 2 2 : 134&167.

5. Bliss CL. The comparison of dosage mortality data. Annals ~n applied biology, 1935, 22: 307-333.

6. Finney DJ. Statistical method in biological assay, 3rd ed. London, Griffin & Co. Lid, 1978: 508.

7. RacineA Grieve AP, Fluhier H. Bayesian methods in practice: experiences in the pharmaceutical industry. Applied statist~cs, 1986, 3 5 : 93-150.

8. Fluhler H, Thraenhart 0. Bayesiari analysis of qualitative dose-response curves. In: Thraenhart 0 et al., eds. Progress in rabies conirol. (Proceedings of the Second lnternat~onal iMV1 Essen//WHO Symposium on "New Developments in Rabies Controi': Essen, 5-7 July 1988 and Report of the WHO Consuitation on Rabies, Essen, 8 July 1988.) Royal Tunbridge Wells, Wells Medical, 1989 : 297-304

APPENDIX 4

Addresses of international institutions for technical cooperation in rabies control

The following WHO Collaborating Centres and other organizat~ons and inst~tu- tions are prepared to collaborate with national services on request:

Zoonoses centres

PAHO/WHO Pan American lnsttute for Food Tel.. (01) 792008714047 Protection and Zoonoses (INPPAZ) Fax: (01 ) 793 0927

Calle Talcahuano 1660 Casillo de Correo 44 1640 Martinez Provincia de Buenos Aires Argentina

Mediterranean Zoonoses Control Centre Tel.: (01 ) 6399367 PO Box 66074 Fax: (01 ) 638 01 63 GR-15510 Holargos Greece

International centres for biological standards, reference preparations and reference reagents

Department of Biological Standardization Tel.: 32 683 268 State Serum Institute F ax: 32 683 868

5 Artillerivej Telex: 31 316 Serum DK

DK-2300 Copenhagen S Denmark

Collaborating and related reference centres

Rabies

WHO Collaborating Centre for Control, Tel.: (0613) 9989320 Ext. 4831

Pathogenesis and Epidemiology of Rabies Fax: (0613) 9528072 in Carnivores

Rabies Unit, Pathology Section Animal D~seases Research Institute (ADRI) Agriculture Canada 801 Fallowfield Road, PB 11300, Station H Nepean, Ontario K2H 8P9 Canada

APPENDIX 4

WHO Collaborating Centre for Reference Tel.: (01 j 45 688 754 and Research on Rabies Fax. (01) 40613 015

Pasteur Institute Telex: PASTEUR 250609 F 28 rue du Docteur Roux F-75724 Paris CBdex 15 France

WHO Collaborating Centre for Rabies Surveillance and Research

Rabies Laboratory Federal Research lnstitute for Animal

Virus D~seases Postfach 1149 D-72076 Tubingen Germany

WHO Collaborating Centre for Rabies Epidemiology

National lnstitute of Communicable Diseases 22 Shamnath Marg Post Box 1492 Delhi 110054 India

WHO Collaboratng Centre for Traning in Rabies Vaccine Production and Quality Control

Rabies Division Pasteur lnstitute of lndia Coonoor 643 103 (Nilgirisj India

Tel.. (07071 ) 6031/32/33 Fax: (07071 ) 603 201 Telex: 17 707 131

Tel.: 21 250121 846 Telex: 853203 PllC IN Cables: LYSSA, COONOOR

WHO Collaborating Centre for Reference Tel.: (021) 6698714 and Research on Rabies Telex: 214265 IPIN

Rabies Department (Research and Production) Cables: Institute Pasteur, Pasteur Institute of Iran Rabies 69 Pasteur Avenue 13164 Teheran Islamic Republic of lran

WHO Collaborating Centre for Research Tel.: (02) 252 61 17 on Rabies Pathogenesis and Prevention Fax: (02) 254 021 2

Queen Saovabha Memor~al Institute Telex: 82 535 TRCS BKKTH Thai Red Cross Society Div~sion of Science 1871 Rama IV Road 10330 Bangkok Thailand


WHO Collaborating Centre for Neurovrology Centre for Virology, Department of

Imm~inology,'Microb~ology Thomas Jefferson Universty Jefferson Medical College 1020 Locust Street Philadelphia, PA 19107 6799 USA

WHO Collaborating Centre for Reference and Research on Rabies

Rabies Laboratory Ceriters for Disease Control and Prevention 1600 Clifton Road Atlanta, GA 30333 U SA

WHO Collaborating Centre for Reference and Research on Rabies

The Wistar Institute oi Anatomy and B~ology

36th Street at Spruce Philadelphia, PA 19104 USA

Veterinary pubiic health

WHO Collaborating Centre for Research and Training in Veter~nary Public Health

School of Veterinary Medcine Bischofsholer Damm 15 0-30173 Hanover Germany

WHOIFAO Collaborating Cent1.e ior Research and Training in Veternary Public Health

Indian Veterinary Research Institute Modular Laboratory Bui ldng zatnagar 243 122 Bareilly (U P.) l n d ~ a

WHO Collaborating Centre for Research and Training in Veterinary Epidemiology and Management

lstituto Zooprof~lattico Sperimentale dell'Abruzzo e del Mol~se "G. Caporale"

Via Campo Boario Casella Postale 137 1-641 00 Terarno Italy

Tel.: (0215) 955 6967 Fax: (021 5) 955 2365

Tel.: (0404) 639 1050 Fax: (0404) 639 31 63 Telex: 549 571 CDC ATL

Tel.: (0215) 898 370314 Fax (021 5) 898 3995 Telex: 71 0 670 0328

Tel.: (051 1 ) 856 8768169 Fax: (051 1 ) 856 7685 Telex: 922 034 TlHO D

Tel.: 72 965 Telex: 577 205 IVRI IN Cables: VETEX

Tel.: (0861) 3321 Fax: (0861) 332251

APPENDIX 4

WHOIFAO Collaborat~ng Centre for Research Tel (06) 4440097 and Tra~ning in Veter~nary Public Health Fax (06) 444 0097

Laboratorio di Parassitologa Telex 610071 ISTISAN l Laboratorio di Medicna Veterinaria lstituto Superiore d Sanita Viale Regina Elena 299 1-00161 Rome Italy

WHO Collaborating Centre for Graduate Tel.. (0919) 829 4210 Residency and Programme on International veterinary Public Health

North Carolina State University College of Veterinary Medicine 4700 Hillsborough Street at

William Moore Drive Raleigh, NC 27606 USA

WHO Collaborating Centre for Training and Reference in Clinical Pathology of Neotropical Primates

Battelle Pacific Northwest Laboratories 902 Battelle Boulevard Richland. WA 99352 USA

WHO Collaborating Centre for Tropical Veterinary Public Health Programme arid Training

School of Veterinary Medicine Tuskegee University Tuskegee, AL 36088 USA

Tel.. (0509) 376 31 93 Fax. (0509) 376 9449

Tel.: (0205) 727 81 74 Fax: (0205) 727 81 77

WHO Collaborating Centre for Veterinary Tel.: (0703) 231 7666 Education in Management and Publ~c Fax: (0703) 231 7367 Health

Department of veterinary Medical Sciences Virginia-Maryland Regional College of

Veterinary Medicine Virginia Polytechnic lnstituie and State

University Blacksburg, VA 24061-0442 USA


WHO Collaborating Centre for Veternary Public Health Systems Research and Analysis

Tufts University School of Veterinary Medicine Department of Medicine, Section of

International Veterinary Medicine 200 Westboro Road North Grafton, MA 01536 USA

Comparative virology

WHO Collaborating Centre for Collection and Evaluation of Data on Comparative Virology

Institute of Medical Microbiology, Infectious and Epidemic Diseases

Veterinary Faculty University of Munich Veterinarstrasse 13 D-80539 Munich Germanv

Zoonoses

WHO Collaborating Centre for the Molecular Epidemiology of Parasitic Infections

School of Veterinary Studies Murdoch University Murdoch, WA 6150 Australia

WHO Collaborating Centre for ELISA and Molecular Techniques for Zoonoses Diagnosis

Animal Production Unit FAO/IAEA Central Laboratory for ELISA and

Molecular Techniques in Diagnosis of Animal Diseases

International Atomic Energy Agency PO Box 100 A-1400 Vienna Austria

Tel.: (0508) 839 5302 Fax: (0508) 839 2953

Tel.: (089) 21 802 527152% Fax: (089) 21 802 597

Tel.: (09) 332 221 1 Fax: (09) 332 2507 Telex: AA 92 71 1

Tel.: (02254) 73 951 Fax: (02254) 73 951 222

APPENDIX 4

WHO Collaborating Centre for Traritng and Tel.: (01 1 ) 2992842/2909755 Research in Urban Zoonoses Control Fax: (01 1) 299 9823

Centre for the Control of Zoonoses Rua Santa EulAlia No. 86, CEP 02031-020 Santana SZo Paulo Brazil

WHO Collaborating Centre for Research Tel.: (01) 444 267 and Training in Zoonoses Telex: Beijing 9083

C h ~ n a National Centre for Preventive Medicine PO Box 5 Changping, Beijing 102206 China

WHO Collaborating Centre for Research Tel.. (083) 292 608 and Management in Zoonoses Control Fax: (083) 293 31 3

Laboratoire dlEtudes sur la Rage et la Cables: CNEVA 54220 Pathologie des Animaux sauvages Malzev~lle

Centre National dlEtudes veterinaires et alimentaires (CNEVA)

B.P. 9 F-54220 Malzeville France

WHO Collaborating Centre for Reference Tel.: (0201) 793 414 and Research on Neurological Zoonoses Fax: (0201) 7235929

Institute of Virology Telex: 857 9573 KLlES D University of Essen PO Box 1021 41 Hufelandstrasse 55 D-45122 Essen Germany

WHO Collaborating Centre for Prevention Tel.: (095) 377 8492 and Control of Zoonoses Telex: 41 1 258 Zerno SU

The Kovlenko Institute for Experimental Veterinary Medicine of the Russian Federation

Kuzminky Moscow 109472 Russian Federation

WHO Collaborating Centre for Zoonoses Tel.: (095) 176 795311 76 021 9 Central Research lnstitute of Epidemiology

of the Russian Federation Ministry of Health Novogireevskaya 3a Moscow 1 11 123 Russian Federation


WHO Collaborating Centre for Reference Tel.: (01) 365 1380 and Research on Parasitic Zoonoses Fax: (01) 363 0478

lnstitut fijr Parasitologie der Telex: 55 575 UNlZl

Veterinarmedizinischen und der Mediznischen Fakultat

University of Zurich Winterthurerstrasse 266a CH-8057 Zurich Switzerland

WHO Collaborating Centre for Reference Tel.: (031) 631 2500 and Research on Viral Zoonoses Fax: (031) 631 2534

Institute of Veterinary Virology University of Berne PO Box 2735 CH-3012 Berne Switzerland

WHO Collaborating Centre for Reference and Trainng on Enteric Zoonoses

Department of Microbiology College of Veterinary Medicine Unversity of Missouri Columbia, MI 6521 1 USA

WHO Collaborating Centre for Reference and Training in Remote Sensing and Geographic Information Systems for Veterinary Medicine

School of Veterinary Medicine Louisiana State University Baton Rouge, LA 70803-8402 USA

Tel.: (0314) 8823083 Fax: (031 4) 882 2950

Tel.: (0504) 346 3335 Fax: (0504) 346 3331

WHO Collaborating Centre for Tel.: (04) 705 885 Research and Management on Fax: (04) 794 980 Zoonotic Disease Control

Central Veterinary Research Laboratory P.O. Box 8101 Causeway, Harare Zimbabwe

International organizations and services

Division of Emerging and other Communicable Diseases Surveillance and Control

World Health Organization 121 1 Geneva 27 Switzerland

Tel.: (022) 791 2575 Fax: (022) 791 0746 Telex: 41 5 41 6 Cables: UNISANTE GENEVE

APPENDIX 4

WHO Regional Off~ce for Africa PO Box No. 6 Brazzaville Congo

WHO Regional Office for the Americas/Pan American Sanitary Bureau

525, 23rd Street NW Washington, DC 20037 USA

WHO Regional Office for the Eastern Mediterranean

PO Box 1517 Alexandria 2151 1

Egypt

WHO Regional Office for Europe 8 Scherfigsvej DK-2100 Copenhagen 0 Denmark

WHO Regional Office for South-East Asia

World Health House lndraprastha Estate Mahatma Gandhi Road New Delhi 110002 India

WHO Regional Office for the Western Pacific

PO Box 2932 Manila 1099 Phil~ppines

Animal Production and Health Division Food and Agriculture Organization of

the United Nations (FAO) Via delle Terme di Caracalla 1-00100 Rome Italy

International Oifice o i Epizootics (OIE) 12 rue de Prony F-75017 Paris France

Tel.: 833860 Fax: 831 879 Telex: 521 715364 Cables. UNISANTE

BRAZZAVILLE

Tel.: (0202) 861 3200 Fax: (0202) 223 5971 Telex: 248 3381440 057 Cables: OFSANPAN

WASHINGTON

Tel.: (03) 482 0223 Fax: (03) 483 891 6 Telex: 54 028 Cables UNISANTE

ALEXANDRIA

Tel.: 39 171 717 Fax: 31 181 120 Telex: 15 348 Cables UNISANTE

COPENHAGEN

Tel.: (01 1) 331 7804 Fax: (01 1 ) 331 8607 Telex: 31 6 5095 Cables: WHO NEW DELHI

Tel.: (02) 521 8421 Fax: (02) 521 1036 Telex: 27 652 Cables: UNISANTE MANILA

Tel.: (06) 57 971 Fax: (06) 57 973 152 Telex: 61 0 181 Cables: FOODAGRI ROME

Tel.: (01 ) 42 274 574 Telex: 642 285 Cables: INTEREPIZOOTIES

PARIS


Arab Organzat~on for Agr~cultural Development

Sharia El Gamaa Khartoum Sudan

Commiss~on of the European Communities (CEC)

200 rue de la Loi 8-1049 Brussels Belgium

Organization of African Unity (OAU) PO Box 3243 Addis-Ababa E t h i o ~ i a

Nongovernmental organizations

International Council for Laboratory Animal Scence (ICLAS)

Department of Physiology University of Kuopio 70211 Kuopio F~nland

International Union for the Conservation of Nature and Natural Resources (IUCN)

Avenue du Mont-Blanc 1196 Gland Switzerland

World Society for the Protection of Animals (WSPA)

Park Place, 10 Lawn Lane London SW8 1UD England

World Wide Fund for Nature (WWF) Avenue du Mont-Blanc 1196 Gland Sw~tzerland

Tel.: (02) 235 11 11 Telex: 21 877 Cables: COMEUR BRUXELLES

Tel.: (01) 157 700 Telex: 21 406 Cables. OAU ADDIS ABABA

Tel.: (071) 163 080 Fax: (071) 163 410

Tel.: (022) 649 1 14 Fax: (022) 642 926 Telex: 41 9 605

Tel.: (0171) 7930208 Fax: (0171) 163410

Tel.: (022) 649 11 1 Fax: (022) 644 238

Index

Note: Page numbers in italics refer to fig- AvOl (avirulent Orsay I ) vaccine 325 ures and tables. 2,2'-Azino-bis(3-ethylbenzthiazoline-6- "Vaccine" refers to rabies vaccine. sulfonic acid) (ABTS) 206, 393

ABT - see Antibody-binding test Acetic acid solution 441 Acetylcholine receptor 121 Acetylethyleneimine 317 Acrylamide solution 170 Adenovirus expression systems 349 Ammonium sulfate 441

detection 444 removal from Sephadex gel column

443 Ammon's horn (see also Specimens,

brain) 55, 66-67, 68, 76 electron microscopy 209-211, 212

Anamnestic response, following immunization 331

Angular transformation 455-457 Animal tissues, disposal of 5 Antibody - see Virus-neutralizing anti-

body Antibody-binding test, modified 357, 394-

396 Antigen

detection in diagnostic procedures 10-1 3, 88-95, 105-1 11, 11 8-1 19

quantification tests for rabies vaccines 378-381, 383-387, 389- 393,394-395

Antimalarial prophylaxis, following rabies vaccination 6

Antirabies immunoglobulins - see lmmunoglobulin

Antirabies serum 15 of equine origin

production 401 -403 purification 405-409

of human origin 411-415 potency test 41 7-421

Ascites product~on 139, 140 Autointerference 11 7 Avidin-biotin

in RRElD 105 in RTClT 100, 103-1 04 staining of neuroblastorna cells 100

Babes' tubercles 66 Baby hamster kidney cells - see BHK-

21 cells Baculovirus expression systems 349 Bayesian analysis 457-458 Beijing virus strain 114. 272, 306 BHK-21 cells 11, 96, 98, 114, 116, 117-

121, 186, 212-213, 316, 325 growthlvirus media 190

Biotin-lgG conjugate 106, 11 0, 433 production 439

Bouin-Dubosq-Brazil fixation mixture 77 Brain specimens - see Specimens, brain Brain-tissue rabies vaccines (see also

Sheep brain vaccine, P-propiolactone-inactivated; Suckling- mouse brain (SMB) vaccine) 223-228

adverse effects 224-225 developments 226-228 efficacy 226 fatality rate 225, 226 local reactions 226-227 neurological complications 224-225 post-exposure treatment 223

Buffers and reagents acetic acid solution 441 acetylethyleneimine 317 acrylamide solution 170 ammonium sulfate solution, satu-

rated 441 annealing buffer 171 2,2'-azino-bis(3-ethylbenzthiazoline-

6-suifonic acid) (ABTS) 206, 393 Bouin-Dubosq-Brazil fixation mixture

77 carbonate buffer 112, 441 Challenge Virus Standard diluent 368 citrate buffer 113, 410 colour solution, polymerase chain

reaction 171 detection buffer, polymerase chain

reaction 171


Dulbecco's phosphate-buffered saline solution A (PBSA) 382

ethylenediamine (EDAA) buffer 441 extraction buffer, polymerase chain

reaction 171 Fuchsin-safranine blue stain 78-79 glycerol, 50% buffered 442 Hanks' balanced salt solution 309 hybridization buffer, polymerase

chain reaction 171 labelling solution, polymerase chain

reaction 172 NT buffer 442 PBS (phosphate-buffered saline) 103,

113, 242 PBS-polysorbate 113 PBS-polysorbate-BSA 11 3 PBS-sodium chloride 410 PCR buffer 172 phosphate buffer 442 reverse transcriptase (RT) buffer,

polymerase chain reaction 172 Sellers' stain 62-64 sequencing solutions, polymerase

chain reaction 172-173 sodium bicarbonate solution 442 sodium carbonate solution 442 sodium chloride solution 442 sodium phosphate solution 442 sodium-potassium phosphate buffer

241 sodium-sodium citrate (SSC) buffer

173 stopping solution, polymerase chain

reaction 173 trometamol-acetate-edetic acid

(TAE) buffer 174 trometamol-borate-edetic acid (TBE)

buffer l74 trometamol-edetic acid (TE) buffer

174

Canine rabies control (see also Cell- culture vaccines for veterinary use; Chicken-embryo vaccine; Modified live-virus vaccines) 155, 31 9

Carbonate buffer 112, 441 Carcasses, disposal of 5 C-ELISA - see ELISA, competitive Cell culture (see also Cell-culture vac-

c i n e ~ ) 114-115, 116, 117-122 applications 11, 11 9-1 22, 139-1 40 cytopathology 11 7 methods of virus propagation 115,

11 7 pathogenesis 121-1 22 persistent infection 11 7-1 18 susceptible cells, cell lines and

strains 11 4-1 15 Cell-culture vaccines

for human use (see also Dog kidney cell vaccine; Fetal rhesus monkey diploid cell vaccine; Human diploid cell vaccine; PHKC vaccine; Purified chick-embryo cell vaccine; Vero cell vaccine; Vnukovo-32 primary hamster kidney cell vaccines) 271-276

adverse reactions 272-273 cost 275-276 efficacy 274-275 hypersensitivity reactions 273 neurological reactions 273-274 post-exposure treatment 274

failures 274-275 regimens 275-276

pre-exposure immunization 276 safety 272-274

for veterinary use 314-322 cell cultures 31 4-31 6 control tests 31 9 inactivated 31 7-318, 31 9 live attenuated 317, 318-319 manufacturing requirements 31 8-

31 9 preparation 31 6-31 8 production facility 320, 321-322 stabilizer 378 virus strains 314, 315

CER cells 11, 96, 114, 120-122 Challenge Virus Standard (CVS) 4, 181,

194, 244, 292, 315, 317, 325, 341, 357, 360-361, 41 7

diluent 368 Chick embryo-related cells - see CER

cells Chicken-embryo vaccine (see also Puri-

fied chick-embryo cell vaccine) 226, 260-267, 275

control tests 265-266 preparation 260-264 stabilizing solution 266-267

Citrate buffer 11 3, 410 Competitive ELISA - see ELISA, com-

petitive Complement fixation test 121 Conjugates -see Immunoglobulin, con-

jugates Cornea1 impressions, use in diagnosis 13,

93 CVS - see Challenge Virus Standard

D5, - see End-point dilution, 5ooh Defective interfering (DI) particles 28, 118 Diagnosis (see also Diagnostic proce-

dures) 10-13, 16, 93, 142 brain samples, technique for collec-

tion and shipment 425, 426, 427- 431

histopathological 66-69, 70, 71, 72- 73, 74-75, 76, 77-79

INDEX

Diagnostic procedures (see also Diagno- sis; Negri bodies)

competitive ELISA 200-204 fluorescent antibody test 88-91, 92,

93-95 in vitro virus neiltralization test 193-

199 mouse inoculation test 80-86, 87 PCR technique 157-1 62 RFFlT 181 -1 87 RRElD 105-106, 107, 108-111 virus isolation in cell culture 96-101

Disinfectants 5 Dog kidney cell vaccine 292, 301-303,

303-305 administration 303 control tests 302 preparation 301-302

Dogs (see also Cell-culture vaccines for veterinary use; Chicken-embryo vaccine; Modified live-virus vac- c i n e ~ )

oral vaccination, precautions 330- 331

quarantine of suspect rabid 60 Dot-blot analysis 159, 161, 162-1 63 Dot-immunobinding assays 14 Duck-embryo vaccine (see also Purified

duck-embryo vaccine) 225-226 Dulbecco's phosphate-buffered saline

solution A (PBSA) 382 Duvenhage virus 28, 38, 107, 117, 149,

154, 348

Eagle's basal medium 21 6 Eagle's growth medium (EGM) 194 Eagle's medium, Dulbecco's modified

(DMEM) 144 Eagle's minimum essential medium

(EMEM) 102-1 03 modified (EMEM-10) 97, 103

Early death phenomenon 122, 331 EBL - see European bat lyssaviruses Electron microscopy 209-213, 214-215,

216-217 ELlSA (see also Essen-ELISA) 11-14,

357 competitive 200-204, 206-208 for determination of glycoprotein con-

tent of vaccines 383-387 End-point dilution, 50% (D,,) 447, 451,

452, 455, 457-458 Enzyme-linked immilnosorbent assay -

see ELISA Equine antlrabies immunoglobulin (ERIG) - see Immunoglobulin, equine

ERA vaccine 317, 324, 325-326 ERA virus strain 4, 315, 317, 325, 341,

349, 389 ERIG -see Imrnunoglobulin, equine

Escherichia coli expression system 347 Essen-ELISA 389-391, 392, 393 Ethacridine lactate precipitation 408 Ethanol precipitation 408-409 Ethylenediamine buffer 441 Eukaryotic expression systems 347-349

adenoviruses 349 baculoviruses 349 poxviruses 347-348

European bat lyssaviruses (EBL) (see also Lyssavirus) 28, 154

identification of isolates using monoclonal antibodies 149, 151, 152- 153, 154

Evelyn Rokitniki Abelseth vaccine - see ERA vaccine

Evelyn Rokitniki Abelseth virus strain - see ERA virus strain

Expression systems eukaryotic 347-349 prokaryotic 347

FBKC vaccine 272, 274 Feline immunodeficiency virus (FIV) 342 Feline infectious peritonitis (FIP) 342 Feline leukaemia virus (FeLV) 342 Fermi-type vaccine 4, 223 Fetal bovine kidney cell - see FBKC

vaccine Fetal rhesus monkey diploid cell vaccine

297-299 administration 299 control tests 298-299 expiry date 299 laboratory tests 299 preparation 297

Fluorescein isothiocyanate (FITC)-lgG conjugate 88, 91, 197

production/purification 437-438, 443-444

Fluorescent antibody (FA) test 10-11, 13, 88-91, 92, 93-95, 119, 195, 197- 198, 394, 430-431, 433

direct method 91, 92, 93 indirect method 137-138

Flury strain vaccines - see High egg passage (HEP) Flury strain vaccine; Low egg passage (LEP) Flury strain vaccine

Focus-forming dose (FFD,,) 183, 184, 187

calculation 191 Formaldehyde, for preservation of brain

specimens 428 Formalin, primary hamster kidney cell

vaccine inactivation 306, 307 Fuchsin-safranine-blue staining method

78-79 Fuenzalida-Palacios technique 227, 243


Glassware, laboratory precautions 5 Glycerol, for preservation of brain speci-

mens 428-429 Glycoprotein - see G protein G protein (see also ELISA, for determi-

nat~on of glycoprotein content of vaccines; Essen-ELISA: Single radial immunodiffusion test; VRG vaccine; VRG virus) 35-36

expression systems 341, 347-348 fluoresce~it antibody staining 11 9, 137 monoclonal antibodies against 137,

146, 147-148, 149, 150, 383- 384, 386-387

production 433-434,435-436 purification 176-1 77, 435-436 strircture 40-41

Graphic method of titre calculation 447, 45 1, 454-455

Grinders 81 GSC vaccine 325 Guinea-pig potency test 357

for chicken-embryo vaccine 374-376. 3 77

Habel test for potency 308. 357,369-373 modified 370

Haemagglutination test 121 Hanks' balanced salt solution 309 HEL cell line 115, 116 Hempt-type vaccine 223 Heterologous rabies antiserum - see

Antirabies serum, equine origin High egg passage (HEP) Flury strain vac-

cine 4, 226, 260, 272, 315, 325 HRlG - see Immunoglobulin, human Human diploid cell line 115, 116, 280, 290 Human diploid cell vaccine 6. 253, 271-

272, 280-283, 284. 411 adverse reactions 272-273 control tests 281-283 efficacy 274-275 hypersensitivity reactions 273 neurological reactions 274 post-exposure treatment 275, 283

cost 271 pre-exposure immunization 276 preparation 280-281. 284

Human embryonic lung (HEL) cell line 115, 116

Human rabies ~mmunoglobcilin - see Immunoglobulin, human

Hybridoma (see also Monoclonal antibodies)

characterization 141-1 42 cloning by limiting dilution 139 freezinglthawing 140-141 production 133-1 36, 139-140 screening 136-139

Identity test for sheep brain vaccine, 0-propiolactone-inactivated 236. 237

IgG conjugates - see Immunoglobulin, conjugates

Immunization (see also individual vaccine types)

oral, animals 324, 341-344 pre-exposure 4-7

lmmunoblotting 142 l m m u n ~ f l u ~ r e ~ c e n c e

staining of neuroblastoma cells 99, 100

tests for ribonucleoprotein 147 lmmunoglobulin 7, 15

assay of 41 7-421 conjugates 89-90, 91, 93, 94

IgG-biotin 106, 110, 433, 439 IgG-fluorescein isothiocyanate

395. 437-438

equine 142 potency testing 41 7-421 production 401-403 purification 405-409

human 142, 274. 405 cost 405, 415 formula 41 1 plasma donors 411-412, 41 5 post-exposure treatment 274, 411 potency testing 41 7-420 production 411-41 5, 416

lmmunoprecipitat~on 141-142 Inclusion bodies (see also Negri bodies)

60, 118 Infection, functional analysis of

host species 32 penetration into host cell 32 routes of infection 32 transcription, replication and budding

33-35 viral proteins involved in host immune

response 35-36 Innocuity test

purified chick-embryo cell vaccine 294

purified duck-embryo vaccine 257 sheep brain vaccine, p-propiolactone-

inactivated 236, 238 Instruments, laboratory precautions 5 Interference phenomena 83, 86 In vitro virus neutralization test (see also

RFFIT) 193-1 99 lsoelectric focusing 176-1 77 lsopycnic centrifugation 176 lsotyping 141

Kelev chicken-embryo vaccine 260, 261, 31 5

Kissling virus strain 297

INDEX

Laboratory precautions 4-6 Lagos bat virus (see also Lyssavirus) 28,

154 identification of isolates using mono-

clonal antibodies 149, 152-753, 154

Low egg passage (LEP) Flury strain vaccine 4, 118, 226, 260, 272, 290, 315, 376, 389

Lowry technique 435, 436 Lyssa virus 3

classification 28 diagnosis 433 evolution 36, 43. 44, 45 identification of isolates using mono-

clonal antibodies 145-1 49, 150. 151, 152-153, 154-155

McCoy cells 96, 11 5, 117 Mann's staining method 78, 429-431 Media

Eagle's basal 21 6 Dulbecco's modified (DMEM) 144 growth (EGM) 194 minimum essential (EMEM) 102-

103 modified (EMEM-1 0) 97, 103

growth, BHK-21IMNA cells 190 medium 199 303-305

Median tissue-culture infective dose (TCID,,) 194, 195, 198. 327

MLV vaccines - see Modified live-virus vaccines

MNA cells - see Neuroblastoma cells, murine

Modified live-virus vaccines 324-331 in vitro antigen quantification tests

358 potency tests 357 SAD-B1 9 (BHK-Tu) vaccine 326-327 SAD-Bern (SAD-BHK) vaccine 326 SAD-P5188 vaccine 327-328 safety assessment 330-331, 355 SAG, vaccine 329 SAG, vaccine 329-330 Vnukovo-32 vaccine 328-329

Mokola virus (see also Lyssavirus) 28, 7 54

identificatron of isolates using monoclonal antibodies 149, 152-153, 154

Molecular biology 36, 37-38, 39-45 Monoclonal antibodies 119, 121, 133,

201, 202 characterization 141 -1 42 conjugation 207 d~rected against

G protein (MAb-Gs) 137, 146, 147-148, 149, 150,383-

384, 386-387 N protein (MAb-Ns) 137 RNP (MAb-RNPs) 145-146, 147,

148. 149 for identification of lyssaviruses 12,

16.145-149, 150,151, 152-153, 154-1 55

production 133-142, 433, 437 screening 136-1 39

Mouse inoculation test 10, 11, 60-62, 80- 86, 8 7

Mouse neuroblastoma cells - see Neuroblastoma cells, murine

Mouse neutralization test (MNT) 13, 15, 187,401,403, 420-421

Mycoplasma test in human diploid cell vaccine 282

Negri bodies 10, 118, 120 differential diagnosis 59-60, 70 embedding technique 77 fixation 77 fluorescent antibody test 92, 93 histopathological diagnosis 66, 68,

69, 70, 74, 77-79 microscopic examination 55, 59-61,

62-65 staining techniques 62-65, 77-79,

100 Nessler's reagent 442 Neuroblastoma cells

avidin-biotin staining 100. 103-1 04 immunofluorescence staining 99, 100 rnurine 115, 117-118, 120-121, 181,

182-1 83, 195-1 96 growth media 190

for virus isolation 96-1 04 NIH potency test 240,356,357,360-368,

396 modified 365-367

Nil-2 cell lines 114, 118, 316 Northern blot analysis 161, 162 N protein 35, 36

expression systems 16, 347, 349 fluorescent antibody staining 119 monoclonal antibodies against 137 structure 39

NT buffer 180, 442 NTE buffer 179 Nucleocapsid - see RNP Nucleoprotein - see N protein

Orthopoxviruses (see also Vaccinia virus; VRG virus) 341, 343-344

Parallel line bioassay technique 391 Paris Pasteur virus (see also Pasteur

virus) 31 5, 435 Pasteur vaccine 4 Pasteur virus 149, 234, 272, 315, 317,


389 PBS (phosphate-buffered saline) 103,

113, 242 PBS-polysorbate-BSA 11 3 PBS-sodium chloride buffer 410 Peroxidase-antiperoxidase staining 100 Peroxidase-lgG conjugate 433

production 438-439 PHKC vaccine (see also Vnukovo-32 pri-

mary hamster kidney cell vac- c i n e ~ ) 272, 306-309

administration 308 control tests 307-308 expiry date 308 post-exposure "treatment failures"

275 preparation 306-307

Phosphate buffer 442 Phosphate-buffered saline 103, 113, 242 Pitman-Moore (PM) virus strain 253, 280,

286, 301, 31 7 Plaque assay 13, 117. 120, 121 Plasticware, laboratory precautions 5 Polyclonal antibody production/purifica-

tion 433-434 Polymerase chain reaction (PCR) tech-

nique 16, 36, 157-169, 170-174 Potency test (see aiso Guinea-pig po-

tency test; Habel test for potency; NIH potency test) 14-15

antirabies serum 417-421 cell-culture vaccines for veterinary

use 31 7-31 8, 31 9 dog kidney cell vaccine 302 equine antirabies immunoglobulin

403, 41 7-421 fetal rhesus monkey diploid cell vac-

cine 298-299 human antirabies immunoglobulin

41 7-421 modified live-virus vaccines for vet-

erinary use 356 primary hamster kidney cell vaccine

307-308 purified chick-embryo cell vaccine

292-293 purified duck-embryo vaccine 255,

257 sheep brain vaccine, P-propiolactone-

inactivated 236, 238, 240 suckling-mouse brain vaccine 247,

249 Vero cell vaccine 288 Vnukovo-32 primary hamster kidney

cell vaccines 312 Poxvirus expression systems 347-348 Primary hamster kidney cell vaccine -

see PHKC vaccine Probit transformation 457 Prokaryotic expression systems 347

P-Propiolactone hypersensitivity reactions 273 inactivation of

cell-culture vaccines for veterinary use 31 7-31 8

dog kidney cell vaccine 301, 302 fetal rhesus monkey diploid cell

vaccine 298 human diploid cell vaccine 271 purified chick-embryo celi vaccine

290 purified duck-embryo vaccine

253, 255-256 sheep brain vaccine 234-240,

241 -242 suckling-mouse brain vaccine

245, 246 Vero cell vaccine 287

iitration of 239-240 Protective clothing 4-5 Purification techniques

for antirabies serum1immunoglobulin 405-409

for monoclonal antibodies 437 for polyclonal antibodies 435-436 for rabies virus 175-179, 179-1 80,

433-436 G protein 176-177, 177-178,

435-436 recombinant proteins 177-1 78 RNP 177, 434-435

Purified chick-embryo cell vaccine 226, 272, 290-294

administration 293 adverse reactions 272-273 control tests 292-293 efficacy 274-275 expiry date 293 hypersensitivity reactions 273 laboratory tests 294 post-exposure "treatment failures"

274 preparation 290-292

Purified duck-embryo vaccine (see aiso Duck-embryo vaccine) 253-259

control tests 257-258 expiry date 258 potency 253, 255 preparation 253-257

Purified Vero cell rabies vaccine - see Vero cell vaccine

PVR vaccine - see Vero cell vaccine Pyrogenicity tests

human rabies immunoglobulin 415 purified chick-embryo cell vaccine

294

Rabies tissue-culture infection test - see RTClT

Rapid fluorescent focus inhibition test -

INDEX

see RFFlT Rapid rabies enzyme immunodiagnosis - see RRElD

Recombinant vaccines for oral immunization of wildlife 341-344, 358

Reed & Muench method for calculation of 50% end-point dilutions 266, 371- 373

Restriction enzyme analysis 165-1 67 Reverse transcriptase buffer 172 RFFlT 13, 15, 181-187, 401. 403, 420-

421, 437 RNP

monoclonal antibodies against 145- 146, 147, 148

production 433, 434-435 purification 177

RREID 11-12, 105-106, 107, 108-111, 11 2-1 13, 163-1 64, 435

RREID-biot - see RRElD RREID-lyssa - see RRElD RTCIT 96-1 01

avidin-biotin staining method 103- 104

media 102-103

SAD virus strain 148, 166-1 68, 31 0, 31 5, 317,325-330

SAD-B1 9 (BHK-Tu) vaccine 326-327 SAD-Bern (SAD-BHK) vaccine 326 SAD-P5188 vaccine 327-328 Safety precautions 3-7

in vitro virus neutralization test 193 mouse inoculation test 84 PCR technique 158-159

Safety tests 355-358 chicken-embryo vaccine 265 human rabies immunoglobulin 415 modified live-virus vaccines 330-331,

355 purified ducic-embryo vaccine 257

SAG, vaccine 329 SAG, vacclne 329-330 Salivary gland - see Specimens, sali-

vary gland Sellers' staining method 59, 60, 62-65,

77-78 Semple-type vaccines 5. 223-224, 227-

228, 234 Sheep brarn vaccine, b-propiolactone-

inactivated 234-240, 241-242 biochemical tests 238-239 buffers 241-242 composition 234 control tests 236, 237-238 dosage schedule 240 expiry date 236 preparation 235-237 stabilizer 242 standard vacclne 240

Single radial immunodiffusion (SRD) test 357, 378-382

Slide preparation for histopathological examination 55, 57, 58, 59

impression method 57, 58 rolling technique 59 smear method 59

Sodium bicarbonate solution 442 Sodium carbonate solution 442 Sodilrm chloride solution 442 Sodium dodecyl sulfate-poiyacrylamide

gel electrophoresis (SDS-PAGE) 178-1 79

Sodium phosphate solution 442 Sodium-potassium phosphate buffer 241 Sodium-sodium citrate (SSC) buffer 173 Southern blot analysis 161, 163 Spearman-Karber method of titre calcu-

lation 237, 266, 371, 445-447, 448- 450, 452-453

Specimens brain

dissection 55, 56-57, 66, 69, 71 preservation and shipment 427-

431 removal 68-69. 70, 71. 72-73.

74, 76, 77 sampling technique 425, 426-

428 sterility 61-62, 82 tissue preparation

for FA test 90 for histological examination

55, 56, 57, 58, 59, 71. 77-79

for mouse inoculation test 81-83

for RTClT 97, 98-99 salivary gland, tissue preparation

for FA test 90 for mouse inoculation test 81-83 for RTCIT 98

Spodoptera frugiperda cells 349 Stability tests

human rabies immunoglobul~n 475 primary hamster kidney cell vaccine

308 pirrif~ed chick-embryo cell vaccine

293.294 purified duck-embryo vaccine 258

Staining methods Fuchsin-safranine-blue 78-79 Mann's 78 Sellers' 77-78

Sterility tests chicken-embryo vaccine 265 human diploid cell vaccine 282 human rabies immunoglobulin 415 purified duck-embryo vaccine 255,

257


sheep brain vaccine, P-propiolactone- inactivated 236, 238

suckling-mouse brain vaccine 247, 249

Street-Alabama-Dufferin (SAD) virus strain - see SAD virus strain

Suckling-animal brain-tissue vaccines (see also Suckling-mouse brain (SMB) vaccine) 224-225

Suckling-mouse brain (SMB) vaccine 224-225,227,243-250

control tests 244, 245-247, 249 expiry date 247 formula 243 inactivation of virus

by p-propiolactone 245, 246 by ultraviolet light 246, 248-250

minimum potency requirements 365 post-exposure schedule 243

cost 275 pre-exposure schedule 243 preparation 243, 247 stabilizer solutions 250

TAG vaccine 325 Thermosensitive Gif 1 (TSG) vaccine 325

avirulent (TAG) 325 Tissue-culture vaccines - see Cell-

culture vaccines Titre calculation methods

angular transformation 455-456 Bayesian analysis 457-458 graphic method 447, 451, 454-455 probit transformation 457 Reed & Muench method 371-373 Spearman-Karber method 445-447,

448-450, 452-453 Trometamol buffers 174 TSG vaccine 325 Tyler & Beesing technique for P-propio-

lactone titration 239-240

Ultraviolet light inactivation of cell-culture vaccines for veterinary

use 317-318 suckling-mouse brain vaccine 246,

248-250 Vnukovo-32 primary hamster kidney

cell vaccines 311

Vaccines (see also individual v a c c i ~ ~ e types)

antigen quantification tests 378-396

ELlSA 383-396 SRD 378-382

brain-tissue 223-250 cell-culture 271-337 embryonating egg 253-267 genetically engineered 341-351 potency tests 355-377

guinea-pig potency test for chicken-embryo vaccine 374-377

Habel test 369-373 NIH test 360-368

safety testing 355-358 Vaccinia virus (see also VRG vaccine)

341, 343, 347-348 van Gehuchten and Nelis lesions 66, 67 Vero cell vaccine (PVRV) 272, 285-288,

401 adverse reactions 272-273 cell cultures 285-286 control tests 287-288 efficacy 274-275 expiry date 288 post-exposure treatment regimens

275-276 preparation 286-287

Vesicular stomatitis virus 31, 42, 117-118 Virion 14

electron microscopy 209-213, 214- 215

morphology 28, 29-30 structure 30, 31. 209

Virus neutralization test, in vitro (see also RFFIT) 193-1 99

Virus-neutralizing antibody tests competitive ELlSA 200-204 in vitro virus neutralization test 193-

199 RFFIT 181-1 87

Vnukovo-32 primary hamster kidney cell vaccines (see also PHKC vaccine)

for human use 272, 310-312 administration 312 control tests 31 2 post-exposure "treatment

failures" 275 preparation 31 0-31 1

for veterinary use 328-329 Vnukovo-32 virus strain 114, 272, 310,

328 VRG vaccine 341-343 VRG virus 341. 348

laboratory techniques in rabies

Documents