Dst

TaTha Sb S

Co

1.

ofa

20di

Veterinary Microbiology 170 (2014) 224–231

A

Art

Re

Re

Ac

Ke

DIV

Pe

Ch

Cla

*

htt

03

IVA vaccine properties of the live chimeric pestivirusrain CP7_E2gif

nya von Rosen a,b, Desislava Rangelova a, Jens Nielsen a,omas Bruun Rasmussen a,*, Ase Uttenthal a

ection for Virology, DTU National Veterinary Institute, Technical University of Denmark, Lindholm, DK-4771 Kalvehave, Denmark

ection for Molecular Disease Biology, Department of Veterinary Disease Biology, Faculty of Health and Medical Sciences, University of

penhagen, Dyrlaegevej 88, 1, DK-1870 Frederiksberg C, Denmark

Introduction

Classical swine fever (CSF) is a devastating viral disease pigs worldwide. The causative agent, CSF virus (CSFV), issmall enveloped single stranded RNA virus (King et al.,12). Together with the very closely related bovine viral

arrhoea virus (BVDV) and border disease virus (BDV),

CSFV belongs to the genus pestivirus within the virusfamily Flaviviridae. While BVDV and BDV have broad hostranges, including both ruminants and pigs, CSFV appears togive rise to natural infections in pigs only (Meyers andThiel, 1996). Vaccination with live attenuated vaccines,such as the C-strain, protects pigs from clinical CSF disease(van Oirschot, 2003). However, the use of such vaccines,makes it is impossible to distinguish serologically vacci-nated animals from those infected with CSFV.

The live chimeric pestivirus vaccine constructs, e.g.CP7_E2alf (Reimann et al., 2004) or the analogous

R T I C L E I N F O

icle history:

ceived 29 August 2013

ceived in revised form 31 January 2014

cepted 4 February 2014

ywords:

A diagnostics and vaccine

stivirus

imeric vaccine

ssical swine fever virus (CSFV)

A B S T R A C T

Live modified vaccines to protect against classical swine fever virus (CSFV), based on

chimeric pestiviruses, have been developed to enable serological Differentiation of Infected

from Vaccinated Animals (DIVA). In this context, the chimeric virus CP7_E2gif vaccine

candidate is unique as it does not include any CSFV components. In the present study, the

DIVA vaccine properties of CP7_E2gif were evaluated in comparison to the conventional

live attenuated Riemser C-strain vaccine. Sera and tonsil samples obtained from pigs

immunised with these two vaccines were analysed. No viral RNA was found in serum after

vaccination with CP7_E2gif, whereas some serum samples from C-strain vaccinated

animals were positive. In both vaccinated groups, individual viral RNA-positive tonsil

samples were detected in animals euthanised between 7 and 21 days post vaccination.

Furthermore, serum samples from these animals, together with archival samples from pigs

vaccinated with CP7_E2gif and subsequently CSFV challenged, were analysed for specific

antibodies using ELISAs and for homologous neutralising antibodies. In animals vaccinated

with CP7_E2gif, neutralising antibodies were detected from day 10. However, the sera

remained negative for anti-CSFV E2-specific antibodies whereas pigs vaccinated with C-

strain seroconverted against CSFV by 14 days after vaccination, as determined by a CSFV-

E2 specific blocking ELISA. One week after subsequent CSFV challenge, a strong anti-CSFV

E2 reaction was detected in CP7_E2gif vaccinated pigs and anti-Erns antibodies were

detected from 10 days after infection. In conclusion, CP7_E2gif has the potential to be used

as a DIVA vaccine in combination with detection of anti-CSFV E2-specific antibodies.

� 2014 Elsevier B.V. All rights reserved.

Corresponding author. Tel.: +45 35 88 78 50; fax: +45 35 88 79 01.

E-mail address: tbrur@vet.dtu.dk (T.B. Rasmussen).

Contents lists available at ScienceDirect

Veterinary Microbiology

jou r nal h o mep ag e: w ww .e ls evier . co m/lo c ate /vetm i c

p://dx.doi.org/10.1016/j.vetmic.2014.02.018

78-1135/� 2014 Elsevier B.V. All rights reserved.

Cti

caabbfrdppimfogrdCthasapCpdpavu2

focfoIDapnEvituCinaidvfrehv

mthpEvaBoo

T. von Rosen et al. / Veterinary Microbiology 170 (2014) 224–231 225

P7_E2gif (Rasmussen et al., 2007), represent Differentia-

on of Infected from Vaccinated Animals (DIVA) vaccineandidates as promising alternatives to the current livettenuated CSFV vaccines. Thus, CP7_E2gif is noteworthys there is no CSFV sequence present in the genome. Theackbone consists, as for CP7_E2alf, of BVDV strain ‘‘CP7’’,ut the BVDV envelope protein E2 has been replaced by E2om BDV strain ‘‘Gifhorn’’. Pigs infected with pestivirusesevelop protective antibodies against the viral structuralroteins E2, Erns and also against the non-structuralrotein NS3 (Wensvoort et al., 1989). E2 is the most

munogenic and variable protein among the pestivirusesllowed by Erns (Koenig et al., 1995). Pestiviruses are

enetically and structurally related and serological cross-eactivity between all members of the genus can beemonstrated (Schirrmeier et al., 2004). Hence, theP7_E2gif vaccine should in principle be able to inducee same level of protective antibodies as achieved by a live

ttenuated CSFV vaccine or wild-type strain. However, thepecificity of CP7_E2gif induced antibodies will be directedgainst BVDV Erns and BDV E2, respectively, and will notossess specific anti-CSFV capacity. Thus, this allows for aSFV-specific serological based detection test to accom-any a CP7_E2gif based vaccine construct in order toifferentiate infected from vaccinated animals. In princi-le, both E2- and Erns-specific antibodies should be suitables CSFV-specific serological markers for the CP7_E2gifaccine. This is supported by a recent evaluation studysing seven commercial CSFV ELISA kits (Schroeder et al.,012).

Within the study by Schroeder and colleagues it wasund that commercial ELISA kits could be applied as

ompanion diagnostic tests for CP7_E2gif with thellowing contributions; (1) PrioCHECK CSFV 2.0 andEXX CSFV Ab for detection of CSFV E2 antibodies

llowing differentiation of CSFV infected pigs (CSFV E2ositive) from those vaccinated with CP7_E2gif (CSFV E2egative), (2) PrioCHECK CSFV Erns for detection of CSFVrns in CSFV infected pigs (CSFV Erns positive) from thoseaccinated with CP7_E2gif (CSFV Erns negative). Moreover,

was concluded that the Chekit CSFV Marker kit could besed for detection of Erns antibodies generated byP7_E2gif vaccination as well as from natural pestivirusfection. Thus, together with a CSFV-specific E2 or Erns

ntibody test, detection of pestivirus Erns antibodies couldentify a successful immunisation with the CP7_E2gif

accine. However, due to a limited number of samplesom CP7_E2gif vaccinated animals it was decided toxpand the investigation by performing a more compre-ensive serological evaluation of the DIVA potential of thisaccine candidate.

Here we report the serological DIVA potential of the livearker vaccine CP7_E2gif to induce an antibody responseat can be discriminated from the antibody response

roduced after infection with CSFV, using commercialLISA tests. Furthermore, the aim was to analyse theaccine properties of CP7_E2gif in comparison to the livettenuated CSFV C-strain Riemser1 (C-strain) vaccine.esides investigation of neutralising antibodies, the latterbjective also encouraged analyses to detect the presencef vaccine virus, both as live virus and as RNA genome, in

serum samples and tonsil material from the soft palatecollected during post mortem examination. In addition, thepotential side effects in connection with the use of thevaccines were addressed.

For this purpose, a vaccination experiment was carriedout with pigs, which were vaccinated with eitherCP7_E2gif or C-strain vaccine, and followed serologicallyfor up to 4 weeks. In addition to this, archived serumsamples were used to evaluate the DIVA serologicalpotential of CP7_E2gif using commercial ELISA test kits.

2. Materials and methods

2.1. Vaccine virus

The vaccine material consisted of a 6th passage virusstock from sheep foetal thymoid cells (SFT-R) of thechimeric virus CP7_E2gif diluted in Eagle’s minimalessential medium (EMEM) (Rasmussen et al., 2007). TheSFT-R cells were obtained from Friedrich-Loeffler Institute(FLI, Insel Riems, Germany). The live attenuated CSFV C-strain vaccine Riemser1 (AC Riemser Schweinepestvak-zine, Riemser Arzneimittel AG, Germany) was usedaccording to the instructions from the manufacturer.

2.2. Animal study

For the vaccination experiment, 6–7-week-old DanishLandrace pigs, obtained from a conventional Danish swineherd were used. Two groups of 12 pigs were vaccinatedintra muscularly in the neck with either 105 TCID50 (50%tissue culture infective dose) of the CP7_E2gif vaccine or104 TCID50 of the C-strain vaccine. The pigs weremonitored daily for relevant clinical signs and lesionsupon vaccination. Three pigs from each group wereeuthanised at 7, 14, 21 and 28 days post vaccination(dpv), respectively, for collection of spleen and tonsils forvirological examination as well as a general pathologicalexamination. Serum samples were collected at 0 (beforevaccination), 3, 7, 10, 14, 17, 21, 24 and 28 dpv and storedat �40 8C until analysis.

The animal study was carried out in accordance withthe requirements of the Danish Animal ExperimentationInspectorate; license no 2008/561-1541.

2.3. Archival serum samples

Serum samples obtained from a previous study ofCP7_E2gif vaccination with subsequent CSFV Eystrupchallenge 28 days post vaccination (Rasmussen et al.,2007) were further analysed in the present study. Theserum samples were collected at 0, 3, 7, 14, 17, 21, 24, 28,35, 38, 42, 49, 57, 63 and 70 dpv and originated from twoexperimental groups with ‘‘CP7_E2gif vaccinated andchallenged’’ pigs (Pigs 11–15) and ‘‘mock-vaccinated andchallenged’’ pigs (Pigs 16–20).

2.4. Serological examination

All serum samples were tested in duplicate forantibodies directed against pestiviruses, using pelleted

fu(HPrNeCSNethporecomplanneanwfoFr

2.5

sa(U(Rsefroet50

wsaCyne

3.

3.1

shvawshwNoan

3.2

va

strfroVIthCPtovade(dCP

T. von Rosen et al. / Veterinary Microbiology 170 (2014) 224–231226

ll virus as antigen, with an in-house blocking ELISAave, 1984) and two commercial blocking ELISAs;ioCheck CSFV Ab 2.0 (Prionics, Lelystad B.V, Thetherlands) detecting anti-CSFV E2 antibodies and ChekitF-Marker (IDEXX, Laboratories B.V. Schiphol-Rijk, Thetherlands), detecting anti-pestivirus Erns antibodies. For

e blocking ELISA the OD% was related to an internalsitive control and inhibition (100-OD%) of >30% wasgarded as positive and 25–30% as doubtful. Themmercial ELISAs were carried out according to theanufacturer’s instructions. In addition, the serum sam-es from the animal study were tested for neutralisingtibodies against BDV Gifhorn or CSFV C-strain usingutralisation tests (VNT) (Hyera et al., 1987) with SFT-Rd porcine kidney cells (PK15), respectively. BDV Gifhorn

as obtained from the Community Reference Laboratoryr CSFV, Hannover, Germany and the PK15 cells fromiedrich-Loeffler-Institute (FLI), Insel Riems, Germany.

. Virological examination

Virus isolation (VI) from serum, spleen and tonsilmples was performed in PK15 cells for C-strainttenthal et al., 2003) and in SFT-R cells for CP7_E2gifasmussen et al., 2007). Total RNA was extracted fromrum using QIAamp RNA Blood Mini kit (Invitrogen) andm tonsils by Boom-silica RNA extraction (Uttenthal

al., 2003). Quantitative real-time RT-PCRs targeting the-non translated region (50-NTR) of either BVDV or CSFVere used for detection of viral RNA in serum and tonsilsmples as previously described (Uttenthal et al., 2003).cle threshold (Ct) values >50 were considered asgative.

Results

. Clinical and pathological examination

None of the CP7_E2gif or C-strain vaccinated pigsowed any clinical symptoms after vaccination. In bothccination groups, a few of the pigs occasionally had daysith body temperatures slightly above 40.0 8C (data notown). At necropsy, six out of the twelve pigs vaccinatedith C-strain, had whitish depleted borders of the spleen.

tissue reactions were observed at the injection site iny pigs from the vaccine groups.

. Virological examination of CP7_E2gif or C-strain

ccinated pigs

To determine the infection kinetics of CP7_E2gif and C-ain in pigs, serum, spleen and tonsil samples collectedm the present experimental pig study was followed by

and RT-PCR analysis for detection of viral RNA. However,e latter did not include examination of spleen material.7_E2gif virus could not be isolated from serum, spleen ornsil samples (data not shown) in the group of CP7_E2gifccinated pigs. Furthermore, BVDV RNA could not betected in serum from the CP7_E2gif vaccinated pigsata not shown). However, tonsil material obtained from

euthanised at 7 dpv (2 out of 3), 14 dpv (2 out of 3), and inthree out of three pigs euthanised at 21 dpv (Fig. 1). BVDVRNA could not be detected in tonsils from any of the pigseuthanised on 28 dpv. C-strain vaccine virus could not beisolated from serum, spleen and tonsil samples from thegroup of C-strain vaccinated pigs (data not shown). Two ofthe C-strain vaccinated pigs had detectable, but low, levelsof CSFV RNA in the serum (Ct-values of 40 and 43,respectively). CSFV RNA could be detected in the tonsilsfrom all three C-strain vaccinated pigs euthanised at 7 dpvand in two out of three pigs euthanised at 14, 21 and28 dpv, respectively (Fig. 1).

3.3. Serological examination of CP7_E2gif or C-strain

vaccinated pigs

All results from the serological examination of CP7_E2-gif and C-strain vaccinated pigs in the animal experimentcarried out in the present study, including the results fromthe different ELISAs and the neutralisation tests, are shownin Table 1. The ELISA results from the CP7_E2gif vaccinatedand challenged pigs (archival serum samples) are shown inFig. 2.

3.3.1. CP7_E2gif vaccinated pigs

Pestivirus antibodies tested using a pestivirus blockingELISA (based on whole CSFV) were detected in serumsamples from two out of three CP7_E2gif vaccinated pigs at24 dpv. The last pig did not seroconvert during theexperimental period (Table 1). CSFV E2 specific antibodieswere not observed in CP7_E2gif vaccinated pigs using thePrioCHECK CSFV Ab 2.0 ELISA (Table 1). Pestivirus Erns

specific antibodies measured using the Chekit CSF-MarkerELISA, were detected from 10 dpv in one out of nineCP7_E2gif vaccinated pigs and in four out of six animals at21 dpv (Table 1). Neutralising antibodies against BDVGifhorn were detected (titre 14–190) in serum samplescollected from CP7_E2gif vaccinated pigs between 14 and21 dpv. All but one of the pigs tested were positive at21 dpv (Pig 7) (Table 1).

3.3.2. C-strain vaccinated pigs

C-strain vaccinated pigs were positive for pestivirus

25

30

35

40

45

50

0 7 14 21 28

Ct

Days Post Vaccinat ion

CP7_ E2gi f C-str ain

No Ct

Fig. 1. Real-time RT-PCR results achieved using tonsil material obtained

from vaccinated pigs targeting the 50-NTR region of BVDV (for CP7_E2gif)

or CSFV (for C-strain). Results below the line are negative (Ct-values >50).

tibodies at 14 dpv in two out of nine pigs and at 28 dpv

7_E2gif vaccinated pigs was BVDV RNA positive in pigs an

fonapnvsa1(Tcthto(P

3

wath

T

N

V

d

T. von Rosen et al. / Veterinary Microbiology 170 (2014) 224–231 227

r all 3 remaining pigs (Table 1). In comparison, one out ofine C-strain vaccinated pigs was detected as CSFV E2ntibody positive by 10 dpv and all but one pig wasositive by 21 dpv (Table 1). The latter pig remainedegative throughout the whole period. Only one C-strainaccinated pig was tested positive for pestivirus Erns-pecific antibodies at 17 dpv (Table 1). Neutralisingntibodies against CSFV C-strain were detected (titre1–23) in five out of nine C-strain vaccinated pigs at 10 dpvable 1). At 21 dpv neutralising antibodies (titre 14–134)

ould be detected in all remaining pigs. However at day 28,e neutralising titres were lower (titre 14–50) compared

the previous testing day with the exception of one pigig 12).

.3.3. CP7_E2gif vaccinated and challenged pigs

Archival serum samples from pigs that were vaccinatedith CP7_E2gif and subsequently challenged 28 days

fterwards with CSFV Eystrup, were likewise tested usinge in-house blocking ELISA and the two commercial

ELISAs for antibodies against the chimeric virus as well asagainst CSFV.

Pestivirus antibodies (as measured in the pestivirusblocking ELISA) were initially detected in one serumsample from the group of CP7_E2gif vaccinated pigs 28 dpv(Fig. 2a). Ten days after virus challenge (38 dpv), pestivirusantibodies could be detected in four out of five serumsamples from this group. From 21 days after challenge(49 dpv) and throughout the testing period, all of theanalysed samples from CP7_E2gif vaccinated and chal-lenged pigs tested positive. In the group of pigs mock-vaccinated and challenged 28 dpv, the serum samplesremained negative for pestivirus antibodies until 10 daysafter challenge (38 dpv), when two out of five sampleswere positive. Three pigs survived the virus challenge andwere all pestivirus antibody positive at 42 days afterchallenge (70 dpv).

CSFV E2 specific antibodies (PrioCHECK CSFV Ab 2.0ELISA) were not detectable in samples from the group ofpigs that were vaccinated with CP7_E2gif and challenged

able 1

eutralising antibody titre and ELISA results from CP7_E2gif and C-strain vaccinated pigs.

CP7_E2gifa

Pig no.

dpv 1 2 3 4 5 6

7 <10/�/�/� <10/�/�/� <10/�/�/� ND/�/�/� ND/�/�/� ND/�/�/�10 <10/�/�/� <10/�/�/� <10/�/�/+14 <10/�/�/� <10/�/�/+ <10/�/�/+17

21

24

28

Pig no.

dpv 7 8 9 10 11 12

7 ND/�/�/� ND/�/�/� ND/�/�/� ND/�/�/� ND/�/�/� ND/�/�/�10 <10/�/�/� <10/�/�/� <10/�/�/� <10/�/�/� <10/�/�/� <10/�/�/�14 <10/�/�/� 14/�/�/� <10/�/�/� <10/�/�/+ <10/�/�/� <10/�/�/�17 <10/�/�/� 14/�/�/+ 14/�/�/+ <10/�/�/+ <10/�/�/+ <10/�/�/�21 <10/�/�/� 57/�/�/+ 190/�/�/+ 40/�/�/+ 113/�/�/+ 57/�/�/�24 12/�/�/+ 24/+/�/+ 14/+/�/+28 95/�/�/+ 113/+/�/+ 190/+/�/�

C-strainb

Pig no.

dpv 1 2 3 4 5 6

7 <10/�/�/� <10/�/�/� <10/�/�/� <10/�/�/� <10/�/�/� <10/�/�/�10 <10/�/�/� <10/�/+/� 23/�/�/�14 <10/�/�/� 14/�/�/� 16/+/+/�17

21

24

28

Pig no.

dpv 7 8 9 10 11 12

7 <10/�/�/� <10/�/�/� <10/�/�/� <10/�/�/� <10/�/�/� <10/�/�/�10 16/+/�/� <10/�/�/� 14/�/�/� 12/�/�/� 11/�/�/� <10/�/�/�14 23/�/+/� <10/�/�/� <10/+/+/� 31/�/+/� 40/�/�/� 14/�/�/�17 16/�/+/� 14/�/+/+ 28/+/+/� 47/�/�/� 131/�/+/� 23/�/�/�21 28/+/+/� 14/�/+/� 40/+/+/� 49/�/+/� 134/+/+/� 23/�/�/�24 56/+/+/� 226/+/+/� 51/�/�/�28 14/+/+/� 14/+/+/� 50/+/�/�

irus neutralisation test (VNT) titre/pestivirus blocking ELISA/PrioCHECK CSFV Ab 2.0 ELISA (E2)/Chekit CSF-Marker ELISA (Erns). ND, not determined; dpv,

ays post vaccination. Data in bold represent pigs having either a positive VNT titre or ELISA result. VNT titres <10 were considered negative.a VNT against BDV Gifhorn.b VNT against CSFV C-strain.

0

10

20

30

40

50

60

70

80

90

100

0 7 14 21 28 35 42 49 56 63 70

% In

hibi

tion

challenge

(a)Pestivirus

0

10

20

30

40

50

60

70

80

90

100

110

0 7 14 21 28 35 42 49 56 63 70

% In

hibi

tion

challenge

(b)CSFV E2

0

10

20

30

40

50

60

70

80

90

100

110

120

0 7 14 21 28 35 42 49 56 63 70

% In

hibi

tion

Days Post vaccina�on

CP7_E2gif challenged with CSFV Eystrup Mock-vaccinated challenged with CSFV Eystrup

challenge

(c)PestivirusE rns

T. von Rosen et al. / Veterinary Microbiology 170 (2014) 224–231228

utiwlafitecp

wos(Fmdad

4

teC

rCfrCin2naesgsvgrootomlithinvafiscwd2

a

F

C

T

w

T. von Rosen et al. / Veterinary Microbiology 170 (2014) 224–231 229

ntil 7 days after virus challenge (35 dpv) (Fig. 2b). At thisme point, four out of five serum samples from this groupere positive for CSFV E2 specific antibodies. Three dayster, i.e. 10 days post challenge, serum samples from allve pigs that were CP7_E2gif vaccinated and challengedsted positive. For the group of pigs mock-vaccinated and

hallenged the serum samples were CSFV E2 antibodyositive 14 days after challenge (42 dpv).

Pestivirus Erns antibodies (Chekit CSF-Marker ELISA)ere detected from 14 dpv in serum samples from one out

f five CP7_E2gif vaccinated and challenged pigs, and in allamples from this group at 7 days post challenge (35 dpv)ig. 2c). Serum samples from the group of pigs which wereock vaccinated and challenged were only analysed from

ay 0, 28, 35, 42 and 70 post vaccination. Pestivirus Erns

ntibodies could be detected in these samples from 14ays after challenge (42 dpv).

. Discussion

The presented study includes characterisation andsting of the chimeric modified live CSFV marker vaccine

P7_E2gif and accompanying DIVA diagnostics.The absence of clinical signs of infection or tissue

eactions at the application site after vaccination withP7_E2gif and C-strain is in agreement with observationsom other studies using intramuscular administration ofP7_E2gif (Rasmussen et al., 2007) and oral/oronasal ortramuscular administration of C-strain (Koenig et al.,

007; Belak et al., 2008; Tignon et al., 2010). Furthermore,ecropsy of the CP7_E2gif vaccinated pigs did not revealny pathological findings which are in accordance witharlier observations (Rasmussen et al., 2007). The demon-tration of changes in the spleen in half of the pigs in theroup vaccinated with C-strain is in contrast with othertudies with post mortem examination of pigs afteraccination with C-strain (Kaden et al., 2010), thougheneral swelling of lymph nodes as well as mild haemor-hages, both typical signs of CSFV infection, has beenbserved previously (Belak et al., 2008). Isolation of eitherf the vaccine viruses from spleen as well as serum andnsil samples in cell cultures was negative, indicating thatultiplication and spread of both vaccine viruses was very

mited or not present and hence, not the obvious cause ofe changes observed in the spleens. However, thiscapacity to isolate both C-strain and CP7_E2gif vaccine

irus could be explained by the fact that live virusppeared earlier in the organ material than 7 dpv; therst day of testing in the present study and with only ahort-termed viraemia. In a study using a similar chimericonstruct, CP7_E2alf (E2 is from CSFV instead of BDV) asell as C-strain for vaccination, VI from the tonsils was

emonstrated from 4 and 3 dpv, respectively (Tignon et al.,010).

BVDV RNA originating from the chimeric vaccine couldlso not be detected in serum samples from CP7_E2gif

vaccinated pigs, which is in accordance with a previousstudy where serum samples from CP7_7E2gif vaccinatedand subsequent challenged pigs also tested negative forviral RNA (Rasmussen et al., 2007). However, CP7_E2gifvaccine virus was detected in the pigs; tonsils from sevenout of twelve vaccinated pigs were viral RNA positive, butwith Ct-values between 39 and 50 (Fig. 1), indicating a lowamount of genomic viral material. For comparison, nineout of twelve C-strain vaccinated pigs had detectable levelsof CSFV RNA in their tonsils (Ct-values between 37 and 43),reflecting a slightly higher amount of genomic viralmaterial than in the CP7_E2gif vaccinated pigs. Thisdifference could explain the detection of CSFV RNA intwo serum samples from C-strain vaccinated pigs. Limiteddetection of genomic material in blood, purified leucocytesor serum samples have also been reported in other studiesperformed with CP7_E2alf and C-strain vaccination(Koenig et al., 2007; Tignon et al., 2010). Thus it can beconcluded that in spite of the lack of viraemia CP7_E2gifreplicates quite similarly to C-strain according to the RNAdetection.

CSFV E2-specific antibodies were not detected in serumsamples from CP7_E2gif vaccinated pigs until 7 days afterCSFV challenge (Table 1 and Fig. 2b), whereas, as expected,all but one of the samples from C-strain vaccinated pigstested positive starting from 10 dpv (Table 1). Thisconfirms the absence of cross-reactivity between theantibodies induced by CP7_E2gif and the specific detectioncapability of the PrioCheck CSFV Ab 2.0 ELISA test. Theobservation that one C-strain vaccinated pig did notseroconvert at all during the test period according to thePrioCHECK CSFV Ab 2.0 ELISA test, whereas pestivirusantibodies could be detected in serum samples from thispig at post vaccination day 28 when using the in-housepestivirus blocking ELISA (Table 1) demonstrated theimportance of using several tests for serological diagnos-tics. Another study has shown seroconversion in all 18 C-strain vaccinated animals tested at post vaccination day 18using a different commercial ELISA for detection of E2-specific antibodies (HerdCheck CSFV, IDEXX) (Tignon et al.,2010). This may indicate, that the PrioCHECK CSFV Ab 2.0ELISA that we have used is less sensitive, which has alsobeen observed previously (Schroeder et al., 2012). TheELISA was chosen due to the best specificity andreproducibility among the kits tested, which is also ofvery high importance in an outbreak situation.

Pestivirus antibodies directed against the whole viruswere detected poorly, and later than pestivirus Erns-specific antibodies in serum samples from CP7_E2gifvaccinated pigs using the in-house blocking ELISA(Table 1) and the Chekit CSF-Marker ELISA, respectively(Table 1). The same observations were made for theantibodies detected before challenge in the archivalsamples from CP7_E2gif vaccinated and subsequentlychallenged pigs (Fig. 2a and c). These observations showsthat antibodies are induced by the chimeric vaccine and

ig. 2. ELISA results for vaccinated and CSFV challenged pigs (archival samples). (a) In-house blocking ELISA (anti-pestivirus), (b) PrioCHECK CSFV Ab (anti-

SFV E2) and (c) Chekit CSFV Marker (anti-pestivirus Erns). Expressed as mean � standard deviation for serum from the residual pigs alive at the sampling time.

he threshold of the 3 ELISA tests used is indicated by the dotted line. Results below the line are negative. In some of the ELISAs only samples from selected days

ere analysed.

setethC-HorafroAnMinlolevAespascofinseteetELstrbosuin

PrCPatinMinpohoantewapou

refoonCPtioanin20anfroandeCPfin

hadianadinva

T. von Rosen et al. / Veterinary Microbiology 170 (2014) 224–231230

roconversion can be detected between 10 and 24 dpv ifsting for pestivirus Erns-specific antibodies. In contrast tois, pestivirus antibodies were detected in samples fromstrain vaccinated pigs already at 14 dpv (Table 1).wever, pestivirus Erns antibodies seem to appear more

pidly in CP7_E2gif vaccinated pigs as only one samplem the C-strain vaccinated pigs was detected as positive.

explanation for this could be that the Chekit CSF-arker ELISA used had insufficient sensitivity for C-strainduced CSFV Erns antibodies, as previously found (Range-va et al., unpublished) or that C-strain only causes a lowel of anti-Erns antibody production as suggested bybischer et al. (2013). This delayed production of Erns-ecific antibodies may be a unique feature of the C-strainsociated with the attenuation of the virus strain and thusntribute to its avirulent phenotype. In comparison to thedings from the C-strain vaccinated animals, several

rum samples from mock-vaccinated and challenged pigssted positive at 14 dpv (Fig. 2c). Previously, Kortekaas al. (2011), has demonstrated that the Chekit CSF-MarkerISA could be used successfully after vaccination with a C-ain derived deletion mutant with detection of anti-dies occurring at 28–35 dpv. These observations furtherpport the idea that Erns antibodies appear more rapidly

CP7_E2gif vaccinated pigs than in C-strain vaccinated.In conclusion the CSFV E2 specific antibody test,

ioCHECK CSFV Ab 2.0, was able to differentiate between7_E2gif vaccinated pigs and pigs vaccinated with a live

tenuated CSFV vaccine or CP7_E2gif vaccinated andfected (challenged) pigs. Moreover, the Chekit CSF-arker ELISA, was able to detect pestivirus Erns antibodiesduced by the chimeric vaccine and at an earlier timeint than the in-house blocking ELISA. However, the in-use blocking ELISA could on the other hand detecttibodies that could not be detected using the two othersts. Hence, a combination of these three ELISA testsould strengthen the serological diagnostic capability ifplying CP7_E2gif for emergency vaccination during antbreak.Neutralising antibodies against pestiviruses are

garded as an important defence against disease. There-re, we also included a comparison of the strength andset of the neutralising response induced by the7_E2gif vaccine with that induced by C-strain vaccina-n. CP7_E2gif has been shown to induce BDV neutralisingtibodies between 10 and 14 dpv followed by a 10-foldcrease shortly after CSFV challenge (Rasmussen et al.,07). In this study, similar levels of BDV neutralisingtibody titres developed after 14–21 days within serumm CP7_E2gif vaccinated pigs (Table 1). Neutralisingtibodies to CSFV in pigs vaccinated with C-strain weretected 10 dpv and reached comparable titres as seen for7_E2gif (Table 1), which correlates with previousdings for C-strain (Renson et al., 2013).In summary, our results demonstrate that CP7_E2gif

s marker properties as it allows for DIVA serologicalagnostics based on discrimination of CSFV specific E2tibodies using a commercial ELISA as companion test. Indition, our immunisation study shows that CP7_E2gifduces a strong immune response as early as 10 days post

compared to vaccination with C-strain. Furthermore, noimmediate side effects could be observed after vaccinationwith CP7_E2gif and the vaccine virus seems to replicate asefficiently as the C-strain vaccine in the tonsils of pigs.

Conflict of interest

The authors declare they have no conflict of interest.

Acknowledgments

The authors thank all animal caretakers and laboratorytechnicians involved for their excellent assistance. Furtherthanks to Carolina Stenfeldt for critically reading thismanuscript. The work was financed in part by the EUNetwork of Excellence, EPIZONE (Contract no. Food CT-2006-016236), CSFV_goDIVA (Contract no. 227003) andThe Research School for Biotechnology (FOBI), Denmark.

References

Aebischer, A., Mueller, M., Hofmann, M.A., 2013. Two newly developedE(rns)-based ELISAs allow the differentiation of Classical Swine Fevervirus-infected from marker-vaccinated animals and the discrimina-tion of pestivirus antibodies. Vet. Microbiol. 161, 274–285.

Belak, K., Koenen, F., Vanderhallen, H., Mittelholzer, C., Feliziani, F., DeMia, G.M., Belak, S., 2008. Comparative studies on the pathogenicityand tissue distribution of three virulence variants of classical swinefever virus, two field isolates and one vaccine strain, with specialregard to immunohistochemical investigations. Acta Vet. Scand. 50,34.

Have, P., 1984. Detection of antibodies against swine fever virus byenzyme-linked immunosorbent assay (ELISA). Acta Vet. Scand. 25,462–465.

Hyera, J.M., Liess, B., Frey, H.R., 1987. A direct neutralizing peroxidase-linked antibody assay for detection and titration of antibodies tobovine viral diarrhoea virus. Zentralbl. Veterinarmed. B 34, 227–239.

Kaden, V., Lange, E., Kuster, H., Mueller, T., Lange, B., 2010. An update onsafety studies on the attenuated ‘‘RIEMSER Schweinepestoralvakzine’’for vaccination of wild boar against classical swine fever. Vet. Micro-biol. 143, 133–138.

King, A.M.Q., Adams, M.J., Carstens, E.B., Lefkowitz, E.J., 2012. VirusTaxonomy: Classification and Nomenclature of Viruses: Ninth Reportof the International Committee on Taxonomy of Viruses. ElsevierAcademic Press, San Diego, CA, USA.

Koenig, P., Hoffmann, B., Depner, K.R., Reimann, I., Teifke, J.P., Beer, M.,2007. Detection of classical swine fever vaccine virus in blood andtissue samples of pigs vaccinated either with a conventional C-strainvaccine or a modified live marker vaccine. Vet. Microbiol. 120, 343–351.

Koenig, M., Lengsfeld, T., Pauly, T., Stark, R., Thiel, H.J., 1995. Classicalswine fever virus: independent induction of protective immunity bytwo structural glycoproteins. J. Virol. 69, 6479–6486.

Kortekaas, J., Ketelaar, J., Vloet, R.P., Loeffen, W.L., 2011. Protective efficacyof a classical swine fever virus C-strain deletion mutant and ability todifferentiate infected from vaccinated animals. Vet. Microbiol. 147,11–18.

Meyers, G., Thiel, H.J., 1996. Molecular characterization of pestiviruses.Adv. Virus Res. 47, 53–118.

Rasmussen, T.B., Uttenthal, A., Reimann, I., Nielsen, J., Depner, K., Beer, M.,2007. Virulence, immunogenicity and vaccine properties of a novelchimeric pestivirus. J. Gen. Virol. 88, 481–486.

Reimann, I., Depner, K., Trapp, S., Beer, M., 2004. An avirulent chimericpestivirus with altered cell tropism protects pigs against lethal infec-tion with classical swine fever virus. Virology 322, 143–157.

Renson, P., Le, D.M., Keranflech, A., Cariolet, R., Koenen, F., Le Potier, M.F.,2013. CP7_E2alf oral vaccination confers partial protection againstearly classical swine fever virus challenge and interferes with patho-geny-related cytokine responses. Vet. Res. 44, 9.

Schirrmeier, H., Strebelow, G., Depner, K., Hoffmann, B., Beer, M., 2004.Genetic and antigenic characterization of an atypical pestivirus iso-

late, a putative member of a novel pestivirus species. J. Gen. Virol. 85,3647–3652. ccination with a level comparable to the C-strain vaccine

S

T

T. von Rosen et al. / Veterinary Microbiology 170 (2014) 224–231 231

chroeder, S., von Rosen, T., Blome, S., Loeffen, W., Haegeman, A.,Koenen, F., Uttenthal, A., 2012. Evaluation of classical swine fevervirus antibody detection assays with an emphasis on the differ-entiation of infected from vaccinated animals. Rev. Sci. Technol. 31,997–1010.

ignon, M., Kulcsar, G., Haegeman, A., Barna, T., Fabian, K., Levai, R., Vander, S.Y., Farsang, A., Vrancken, R., Belak, K., Koenen, F., 2010. Classicalswine fever: comparison of oronasal immunisation with CP7E2alfmarker and C-strain vaccines in domestic pigs. Vet. Microbiol. 142,59–68.

Uttenthal, A., Storgaard, T., Oleksiewicz, M.B., de Striker, K., 2003. Experi-mental infection with the Paderborn isolate of classical swine fevervirus in 10-week-old pigs: determination of viral replication kineticsby quantitative RT-PCR, virus isolation and antigen ELISA. Vet. Micro-biol. 92, 197–212.

van Oirschot, J.T., 2003. Vaccinology of classical swine fever: from lab tofield. Vet. Microbiol. 96, 367–384.

Wensvoort, G., Terpstra, C., de Kluijver, E.P., Kragten, C., Warnaar, J.C.,1989. Antigenic differentiation of pestivirus strains with monoclonalantibodies against hog cholera virus. Vet. Microbiol. 21, 9–20.