Border disease of sheep and goats

HAL Id: hal-00902531https://hal.archives-ouvertes.fr/hal-00902531

Submitted on 1 Jan 1998

HAL is a multi-disciplinary open accessarchive for the deposit and dissemination of sci-entific research documents, whether they are pub-lished or not. The documents may come fromteaching and research institutions in France orabroad, or from public or private research centers.

L’archive ouverte pluridisciplinaire HAL, estdestinée au dépôt et à la diffusion de documentsscientifiques de niveau recherche, publiés ou non,émanant des établissements d’enseignement et derecherche français ou étrangers, des laboratoirespublics ou privés.

Border disease of sheep and goatsPeter F. Nettleton, Janine A. Gilray, Pierre Russo, Elyess Dlissi

To cite this version:Peter F. Nettleton, Janine A. Gilray, Pierre Russo, Elyess Dlissi. Border disease of sheep and goats.Veterinary Research, BioMed Central, 1998, 29 (3-4), pp.327-340. �hal-00902531�

Review article

Border disease of sheep and goats

Peter F. Nettletona Janine A. GilrayaPierre Russo’ Elyess Dlissib

a Moredun Research Institute, International Research Centre, Pentiands Science Park,Bush Loan, Penicuik, Midlothian EH26 OPZ, Scotland, UKh Cneva Sophia-Antipolis, BP 1 I 1-06902 Sophia-Antipolis cedex, France

C Institut De La Recherche Vétérinaire, La Rabta - 1006 Tunis, Tunisia

(Received 30 January 1998; accepted 13 March 1998)

Abstract - Border disease (BD) is a congenital virus disease of sheep and goats first reported in n1959 from the border region of England and Wales. BD virus (BDV) is a pestivirus in thegenus Flaviviridae and is closely related to classical swine fever virus and bovine virus diar-rhoea virus (BVDV). Nearly all isolates of BDV are non-cytopathogenic (ncp) in cell culture.There are no defined serotypes but pestiviruses isolated from sheep exhibit considerable antigenicdiversity and three distinct antigenic groups have been identified. Distribution of the virus is world-wide. Prevalence rates vary in sheep from 5 to 50 % between countries and from region-to-region within countries. The disease in goats is rare and characterized by abortion. Clinical signsin sheep include barren ewes, abortions, stillbirths and the birth of small weak lambs. Affectedlambs can show tremor, abnormal body conformation and hairy fleeces (so-called ’hairy-shaker’or ‘fuzzy’ lambs). Vertical transmission plays an important role in the epidemiology of the dis-ease. Infection of fetuses can result in the birth of persistently infected (PI) lambs. These PI lambsare viraemic, antibody negative and constantly excrete virus. The virus spreads from sheep to sheepwith Pl animals being the most potent source of infection. Apparently healthy PI sheep resultingfrom congenital infection can be identified by direct detection of viral antigen or viral RNA inleukocytes or by isolation of ncp virus from blood or serum in laboratory cell cultures. Isolationof virus is unreliable in lambs younger than 2 months old that have received colostral antibody.The isolation of virus from tissues of aborted or stillbom lambs is difficult but tissues from PI sheepcontain easily detectable levels of virus. To detect the growth of virus in cell cultures it is essen-tial to use an immune-labelling method. Acute infection is usually subclinical and viraemia is tran-sient and difficult to detect. Sheep may also be infected following close contact with cattleexcreting the closely related BVDV. © Inra/E!sevier, Paris ,

border disease / border disease virus / pestivirus / sheep / goat

* Correspondence and reprintsTel.: (44) 131 445 5 111; fax: (44) 131 445 611 I; e-mail: nettpC!mri.sari.ac.uk

Résumé - Border disease ou hypomyélinogenèse congénitale chez le mouton et la chèvre. La« border disease » (BD) est une maladie virale congénitale qui sévit chez le mouton et la chèvre.Elle a été décrite pour la première fois en 1959 à la frontière (en anglais, border, d’où le nom dela maladie, NDLR) entre l’Angleterre et le Pays de Galles. Le virus de la BD (V BD) est un pes-tivirus du genre Flaviviridae qui est fortement apparenté au virus de la peste porcine classique etau virus de la diarrhée bovine virale (VDBV). Presque tous les isolats de VBD sont non-cyto-pathogènes (ncp) en culture cellulaire. Il n’y a pas de sérotype défini, mais il existe une diversitéantigénique considérable entre les différents pestivirus isolés du mouton, et trois groupes anti-géniques distincts ont été identifiés. Le virus se retrouve dans le monde entier. Les taux de pré-valence chez le mouton varient de 5 à 50 % selon le pays et même d’une région à l’autre dans unmême pays. La maladie chez la chèvre est rare et se caractérise par un avortement. Les signes cli-niques chez le mouton incluent stérilité, avortements, mortinatalité, et naissance d’agneaux petitset faibles. Les agneaux affectés peuvent avoir des tremblements, une constitution physique anor-male, et un pelage très fourni. La transmission verticale joue un rôle important dans l’épidé-miologie de la maladie. L’infection de foetus peut entraîner la naissance d’agneaux infectés demanière persistante (IP). Ces agneaux IP sont virémiques, séronégatifs, et excrètent constammentle virus. Le virus se répand d’un mouton à l’autre, les animaux IP étant la source d’infection laplus importante. Les moutons IP, apparemment sains, mais infectés in utero, peuvent être iden-tifiés par la détection directe d’antigène viral, ou d’ARN viral dans les leucocytes, ou par l’iso-lement de virus ncp à partir du sang ou du sérum dans les cultures cellulaires de laboratoire.L’isolement du virus n’est pas fiable chez les agneaux de moins de 2 mois qui ont reçu les anti-corps du colostrum. L’isolement du virus à partir de tissus d’agneaux avortés ou mort-nés est dif-ficile, mais les tissus de moutons IP contiennent des niveaux de virus facilement détectables.Afin de détecter la croissance du virus dans les cultures cellulaires, il est essentiel d’utiliser unetechnique d’immuno-marquage. L’infection aiguë est généralement subclinique, et la virémieest transitoire et difficile à détecter. Les moutons peuvent également être infectés après un contactavec des bovins excrétant le virus, apparenté, de la diarrhée bovine virale. © Inra/Elsevier, Paris

border disease / hypomyélinogenèse congénitale / pestivirus / mouton / chèvre

1. INTRODUCTION

Border disease in sheep and goats iscaused by the pestivirus, border diseasevirus (BDV). BDV spreads naturally bythe oro-nasal route and by vertical trans-mission. It is a cause of congenital dis-ease in sheep and goats but can also causeacute and persistent infections. This paperwill principally describe BDV infectionin sheep. Infection in goats is rare and willonly be described where it is known todiffer from infection in sheep. Compre-hensive reviews of the elucidation of thecause and pathology of the disease havebeen published (Barlow and Patterson,1982; Terpstra 1985). Major advancessince then have been in the molecularstructure of the virus and the relationshipsbetween sheep isolates and pestivirusesfrom other species (Nettleton and Entri-can, 1995; Paton, 1995; Thiel et al., 1996).

2. VIRUS PROPERTIES

The genus pestivirus within the familyFlaviviridae has been divided into classi-cal swine fever virus (CSFV), bovine virusdiarrhoea virus (BVDV) and border dis-ease virus (BDV). The viruses were namedafter the diseases from which they werefirst isolated and traditionally pestivirusesisolated from pigs have been termedCSFV, those from cattle BVDV and thosefrom sheep BDV. It is now known thatcross-infection between species occursreadily and viruses have consequentlybeen grouped more according to their reac-

tivity with monoclonal antibodies and totheir nucleotide sequences at selected

genomic regions. Viruses have also beentested by cross-neutralization experimentswith a view to identifying the best virusesfor candidate vaccines.

Pestiviruses are enveloped, sphericalparticles approximately 50 nm in diameter.The pestivirus genome is a positive sin-gle-stranded RNA molecule, approxi-mately 12.5 kb long. There is a single openreading frame (ORF) flanked by a 5’-non-coding region (5’-NC) of 356-385 basesand a 3’-non-coding region (3’-NC) of223-228 bases (figure /). The ORFencodes proteins composed of 4 000amino acids processed by viral and cellu-lar enzymes. Of the proteins within theORF the first protein is a non-structuralautoprotease NPr° followed by the struc-tural C nucleocapsid protein and glyco-proteins, Erns, E I and E2. Of the glyco-proteins, E2 is the immunodominant majorenvelope protein. The remaining proteinsare non-structural of which NS2-3 hasattracted most interest owing to its role incytopathogenity of pestivirus isolates. Vir-tually all pestivirus isolates from sheepand goats are non-cytopathic in cell cul-ture. Two cytopathic sheep isolates havebeen described, however, and in both theseit has been shown that they contain inser-tions of cellular sequence within the NS2-3 encoding region which results in itscleavage to NS2 and NS3 (Becher et al.,1996). This is analogous to the similarprocess in BVDV viruses which is asso-ciated with the development of mucosal

disease in cattle. Such cattle are persis-tently infected with non-cytopathic (NCP)BVDV following in-utero infection. Muta-tion of the persisting virus RNA in theregion coding for NS2-3 can result inoverproduction of NS3 which is correlatedwith the development of mucosal disease.

From studies with monoclonal anti-bodies and phylogenetic analysis ofgenomic sequences from the 5’-NC, NproC and E2 coding regions a consensus isemerging that there are four principalgenotypes of pestiviruses: BVDV-l,BVDV-2, BDV and CSFV (Paton et al.,1994; Becher et al., 1995; Tijssen et al.,1996). Two pestiviruses, one from a deerand one from a giraffe, do not fit into anyof these groups and constitute separatevirus types (Van Rijn et al., 1997).

While CSF viruses are predominantlyrestricted to pigs, examples of the otherthree genotypes have all been recoveredfrom sheep (Vilcek et al., 1997). Exami-nation of three goat isolates has shownthem to be BVDV-1 types (Becher et al.,1997).

The complete genomic sequence of theAmerican BDV reference strain, BD-31,has recently been published (Ridpath andBolin, 1997). When compared with theavailable complete genomic sequences ofother pestiviruses the predicted amino-acid sequence identity varied from 71 1(BVDV-1) to 78 % (CSFV). Phylogeneticanalysis segregated the pestiviruses intotwo branches, one containing BD31 andthe CSFV strains and one containingBVDV-1 and BVDV-2 strains. The

regions of highest sequence identity werein the 5’-NC and the non-structural virus

polypeptide NS3. Comparison of theBD31 sequence with incompletesequences from nine other BDVs showedclose homology with five including theAustralian reference strain X818 and theUK Moredun reference strain. Of the

remaining four viruses, three were moresimilar to viruses from the B VDV-2 geno-

type and one (R2727) was more similarto viruses from the BVDV-1 genotype.

There are relatively few reports ofcross-neutralization experiments involv-ing pestiviruses from sheep and goats.Nevertheless four principal serologicalgroups have been identified (Wensvoortet al., 1989), which correlate with themonoclonal antibody and genotype group-ings (Paton et al., 1995). As with the geno-typing results the giraffe and deer isolatesare serologically distinct from the otherfour groups (Dekker et al., 1995).

Due to the paucity of cross-neutraliza-tion results with sheep pestiviruses weselected ten non-cytopathic isolates fromdifferent regions of the UK and comparedthem with a cattle pestivirus. The source ofthe viruses is shown in table L All viruseswere plaque-purified three times and pairsof pestivirus-naïve lambs were infectedintranasally with 5 x 10! TCIDso of eachvirus. Serum was collected from each lambLO weeks later and tested in micro neu-tralization tests against approximately 100TCIDso of all the viruses. All the lambs

seroconverted, except one infected withR2727, and produced neutralizing anti-body titres between 256 and 2880 againsttheir homologous virus. The geometricmean titres of neutralizing antibody pro-duced by pairs of sheep are shown intable II. It can be seen that the BD virusesare serologically related but there is a spec-trum of antigenic cross-reactivity, withtwo clusters of more strongly relatedstrains. Thus G1480, JH2816, A1870,L991 and G2048 (The Moredun BDVgroup) appear to be closely related to eachother but distinct from the second groupcontaining D861, B 1056 and Weybridge(The Weybridge BDV group). The twostrains G1305 and R2727 do not i mmedi-

ately fall into either group. The coefficientof antigenic similarity (R value) for the11 viruses was determined as described

previously (Howard et al., 1987; Nettle-ton, 1987). The closer the value of R is to

100 the more closely related are the strains.Values of R ;::: 25 depict a less than four-fold difference between strains which isnot significantly different within the con-ditions of the neutralization tests. A valueof R < 5 signifies a > 20-fold reciprocaldifference between the homologous andheterologous titres, and has been used todefine virus serotypes.

The R values shown in table III empha-size further the two groups of serologi-cally distinguishable BDV isolates, theone group related to Moredun BDV andthe other to the Weybridge and BVDviruses. These antigenic differences cor-relate well with genotyping results since allthe Moredun BDV related isolates type astrue BDVs whereas representatives of theother group all fit into what is now knownas the BVDV-1 genogroup (Becher et al., .,

1994; Vilcek et al., 1994). These resultstogether with recent genotyping studieson a total of 38 UK sheep isolates showthat 23 (60 %) are true BD viruses and

10 (26 %) belong to the BVDV-I group. Afurther five isolates (13 %) belong to theBVDV-2 group (Vilcek et al., 1997). ).

The relevance of all these typing resultsto vaccine development requires furtherwork. In particular, there is a dearth ofcross-protection studies. The only cross-protection test published used field brainmaterial to infect and challenge pregnantewes. The viruses recovered from thesefield outbreaks were G 1480 (Moredun ref-erence strain) a true BD virus and B 1056a BVDV- isolate. In that thorough exper-iment 12 pregnant ewes previouslyexposed to BVDV- were challengedintramuscularly with heterologous BDvirus on the 54th day of gestation; I ewes(92 %) had diseased lambs. A further 11 Ipregnant ewes previously exposed to BDVwere similarly challenged with the het-erologous BVDV-1 strain; one eweaborted and five had diseased progeny,i.e. 55 % of the ewes had diseased lambs.In contrast, similar sized groups of

immune ewes all had normal lambs when

challenged with the homologous virus towhich they had been previously exposed(Vantsis et al., 1980).

This result could be expected given theR values of the two viruses in table III andwould imply that any BD vaccine shouldcontain at least one representative fromthe BDV and BVDV-1 groups. The pro-tection between BVDV-2 and these othertwo sheep-infecting groups will need tobe studied.

3. CLINICAL DISEASE

3.1 Acute infections

Healthy newborn and adult sheepexposed to BDV experience only mild orinapparent disease. Slight fever and a mildleucopaenia are associated with a short-lived viraemia detectable between days 4and 11 post-infection after which serumneutralizing antibody appears.

Occasional BDV isolates have beenshown to produce high fever, profoundand prolonged leucopaenia, anorexia, con-junctivitis, nasal discharge, dyspnoea anddiarrhoea, and 50 % mortality in younglambs. One such isolate was recoveredfrom a severe epidemic of Aveyron dis-ease among milk sheep in France in 1984(Chappuis et a]., 1986). A second suchisolate was a BDV contaminant of a liveCSFV vaccine (Wensvoort and Terpstra,1988).

3.2. Fetal infection

The main clinical signs of BD are seenfollowing the infection of pregnant ewes.While the initial maternal infection is sub-clinical or mild the consequences for thefetus are serious. Fetal death may occurat any stage of pregnancy but is commoner

in fetuses infected early in gestation. Smalldead fetuses may be resorbed or their abor-tion pass unnoticed since the ewes con-tinue to feed well and show no sign of dis-comfort. As lambing time approaches, theabortion of larger fetuses, stillbirths andthe premature births of small, weak lambswill be seen. The infection of pregnantgoats produces severe placentitis and highdeath rates among kid fetuses (Barlow andPatterson, 1982).

During lambing, an excessive numberof barren ewes will become apparent but itis the diseased live lambs that present themain clinical features characteristic of BD.The clinical signs exhibited by BD lambsare very variable and depend on the breedof sheep, the virulence of the virus andthe time at which infection was introducedinto the flock (Barlow and Patterson, 1982;Bonniwell et al., 1987; Roeder et al.,1987). Affected lambs are usually smalland weak, many being unable to stand.Nervous signs and fleece changes are oftenapparent. The nervous symptoms of BDare its most characteristic feature. Thetremor can vary from violent rhythmiccontractions of the muscles of the hind

legs and back to barely detectable finetrembling of the head, ears and tail. Fleeceabnormalities are most obvious in smooth-coated breeds which have hairy fleecesespecially on the neck and back. Abnormalbrown or black pigmentation of the fleecemay also be seen in BD-affected lambs.

With careful nursing a proportion ofBD lambs can be reared although deathsmay occur at any age. The nervous symp-toms gradually decline and can have dis-appeared by 3 to 6 months of age. Weak-ness, swaying of the hind-quarters togetherwith fine trembling of the head may reap-pear at times of stress. Affected lambsoften grow slowly and under normal fieldconditions many will die before or around

weaning time. Occasionally this is the firstpresenting sign of disease when losses atlambing time have been low and no lambs

with obvious symptoms of BD have beenborn.

Some fetal infections occurring aroundmid-gestation can result in lambs withsevere nervous signs, locomotor distur-bances and abnormal skeletons. Suchlambs have lesions of cerebellar hypopla-sia and dysplasia, hydranencephaly andporencephaly resulting from necrotizinginflammation. The severe destructivelesions appear to be immune mediated,and lambs with such lesions frequentlyhave high concentrations of serum anti-body to BDV. Most lambs infected in lategestation are normal and healthy and areborn free of virus but with BDV antibody.Some such lambs, however, can be still-born or weak and many die in early life.

3.3. Persistent viraemia

When fetuses are infected before theyhave any immune system and survive theyare born with a persistent viraemia. Theovine fetus can first respond to an anti-genic stimulus between approximately 60and 85 d of its 150-d gestation period. Infetuses infected before the onset ofimmune competence viral replication isuncontrolled and 50 °!o fetal death is com-mon. In lambs surviving infection in earlygestation, virus is widespread in all organs.Such lambs appear to be tolerant to thevirus and have a persistent infection usu-ally for life. A precolostral blood samplewill be virus positive and antibody nega-tive. Typically, there is no inflammatoryreaction and the most characteristic patho-logical changes are in the central nervoussystem (CNS) and skin. At all levels inthe CNS there is a deficiency of myelin.This may be slight in lambs with mild orno symptoms but is severe in lambs withpronounced nervous symptoms. In the skinthere is an increased size of primary woolfollicles and fewer secondary wool folli-cles causing the hairy or coarse fleece.

Persistently viraemic sheep can be diag-nosed by virus isolation/detection in ablood sample. Viraemia is readilydetectable at any time except in the first 2months of life when virus is masked bycolostral antibody and in animals olderthan 4 years old some of which developlow levels of anti-BDV antibody (Nettle-ton et al., 1992). Although virus detectionin blood during an acute infection is dif-ficult, persistent viraemia should be con-firmed by retesting animals after an inter-val of at least 3 weeks.

Some viraemic sheep survive to sex-ual maturity and are used for breeding.Lambs born to these infected dams are

always persistently viraemic. Persistentlyviraemic sheep are a continual source ofinfectious virus to other animals and theiridentification is a major factor in any con-trol programme. Sheep being tradedshould be screened for the absence ofBDV viraemia.

Rams that are persistently infected (PI)usually have poor quality, highly infec-tive semen and reduced fertility. All ramsused for breeding should be screened forpersistent BDV infection on a blood sam-ple. Semen samples can also be screenedfor virus but they are much less satisfac-tory than blood due to their toxicity forcell cultures.

3.4. Late-onset disease in persistentlyviraemic sheep

Some persistently infected sheephoused apart from other animals sponta-neously develop intractable scour, wast-ing, excessive ocular and nasal dischargessometimes with respiratory distress. Atnecropsy such sheep have gross thickeningof the distal ileum, caecum and colonresulting from focal hyperplastic enteropa-thy. Cytopathic BDV can be recoveredfrom the gut of these lambs. With no obvi-ous outside source of cytopathic virus it

is most likely that such virus originatesfrom the lamb’s own virus pool. Otherpersistently infected sheep in the groupdo not develop the disease. This syndrome,which has also been suspected in occa-sional field outbreaks of BD, has severalsimilarities with bovine mucosal disease

(Gardiner et al., 1983; Nettleton et al.,1992; Monies and Simpson, 1997).

4. EPIDEMIOLOGY

Border disease is widespread in Europe,Australasia and North America and hasalso been reported from Israel and NorthAfrica. Antibody prevalence rates amongadult sheep vary from 5 to 50 °!o betweencountries and from region-to-region withincountries. The virus is not stable outsidethe host and its successful transmission isdue to spread by PI sheep. Spread withina flock can take years in sheep rearedextensively at grass but trough-feeding orother husbandry that allows close nose-to-nose contact will hasten spread. Insheep reared intensively indoors spreadfrom PI animals occurs more readily andserious outbreaks of BD at lambing timecan occur in sheep housed together in earlypregnancy.

Pestivirus exchange between sheep andcattle occurs readily, and outbreaks of BDhave been caused by transmission of virusfrom cattle to sheep (Carlsson, 1991). Noreport of natural disease in cattle due to

spread of virus from sheep has been doc-umented, but serial experimental exchangeof virus between PI cattle and sheep hasbeen investigated. There was a high degreeof genetic stability but the expression ofone or more epitopes on the E2 glycopro-tein appeared to depend on the host species(Paton et al., 1997). There are no reports ofBD originating from pigs but an outbreakof CSF-like disease in pigs was shown tohave been caused by BDV (Roehe et al.,1992). There remains the theoretical pos-

sibility of sheep and goats developing BDfrom contact with other species of rumi-nants since at least 52 species of captive orfree-living ruminants are known to beinfected by pestiviruses (Nettleton, 1990).

Pestiviruses are important contaminantsof modified live virus (MLV) vaccines.All MLV vaccines produced in ovine,bovine or porcine cell cultures or in mediasupplemented with serum from thesespecies risk being contaminated with pes-tivirus. Outbreaks of BD have been asso-ciated with the use of such vaccines;sheep-pox and orf virus vaccines in sheepand an orf virus vaccine in goats (Nettle-ton and Entrican, 1995).

5. DIAGNOSIS

5.1. Virus isolation

Pestiviruses are notorious contaminants

of laboratory cell cultures with fetal bovineserum being the commonest way it isintroduced. It is essential that laboratories

undertaking virus isolation have a guar-anteed supply of pestivirus-free suscepti-ble cells and fetal bovine or equivalentserum which contains no antipestivirusactivity and no contaminating virus. Thevirus can be isolated in a number of pri-mary or secondary ovine cell cultures (e.g.kidney, testes, lung). Ovine cell lines forBDV growth are rare and commerciallyunobtainable. Semi-continuous cell linesderived from fetal lamb muscle (FLM) orsheep choroid plexus can be useful butdifferent lines vary considerably in theirsusceptibility to virus.

From live animals the most sensitive

way to confirm pestivirus viraemia is towash leukocytes three times in culturemedium before co-cultivating them withsusceptible cells for 7 d. Cells are frozenand thawed once and an aliquot passagedon to further susceptible cells grown on

flying coverslips. Three days later the cellsare stained for the presence of pestivirususing an immunofluorescence (IF) orimmunoperoxidase (IP) test.

From dead animals, tissues should becollected as 10 % weight by volume invirus transport medium. In the laboratorythey are ground up, centrifuged to removedebris and the supernatant passed through0.45-!m filters. Spleen, thyroid, thymus,kidney, brain, lymph nodes and gut lesionsare the best organs for virus isolation.

Semen can be examined for the pres-ence of BDV but raw semen is stronglycytotoxic and must be diluted usually atleast 1:10 in culture medium. Since the

major threat of BDV-infected semen isfrom PI rams, blood is a more reliable clin-ical sample than semen for identifyingsuch animals.

5.2. ELISA for antigen detection

The first ELISA for pestivirus antigendetection was described for detectingviraemic sheep. This has now been mod-ified into a double monoclonal antibody(mab) capture ELISA for use in sheep andcattle. Two capture mabs are bound towells in microtitre plates and two othermabs conjugated to peroxidase serve asdetector mabs (Entrican et al., 1995). Thetest is most commonly employed to iden-tify persistently infected viraemic sheepusing washed, detergent-lysed bloodleukocytes. The sensitivity is close to thatof virus isolation and it is a practicalmethod for screening high numbers ofbloods. As with virus isolation, high lev-els of colostral antibody can mask persis-tent viraemia. The ELISA test is more

effective than virus isolation in the pres-ence of antibody, but may give false neg-ative results in viraemic lambs youngerthan 2 months old. The ELISA is usuallynot sensitive enough to detect acute BDVinfections on blood samples. As well as

testing leukocytes the antigen ELISA canalso be used on tissue suspensions, espe-cially spleen, from suspected PI sheep and,as an alternative to IF and IP methods, oncell cultures.

Several pestivirus ELISA methods havebeen published and a number of commer-cial kits are now available for detectingBVDV in cattle. While some of these maybe suitable for use in sheep further evalu-ation is required.

5.3 Nucleic acid detection

Ovine pestiviruses can be detected bythe reverse transcriptase-polymerase chainreaction (RT-PCR) using primers whichalso detect pestiviruses from other species(Vilcek et al., 1994). While RT-PCR hasnot yet been evaluated for diagnostic pur-poses in sheep it is likely to be of futurevalue. The detection of viral RNA in fetaltissues may yet be an important applicationsince other methods are insensitive. The

exquisite sensitivity of RT-PCR makes ita valuable tool for detecting low levelvirus contaminations as in cell culture con-stituents or vaccines (Sandvik et al., 1997).

5.4. Serological tests

Antibody to BDV can be detected insheep sera using virus neutralization (VN)or an ELISA. Control positive and nega-tive reference sera must be included in

every test. These should give results withinpredetermined limits for the test to be con-sidered valid. Single sera can be tested todetermine the prevalence of BDV in aflock, region or country. For diagnosis,however, acute and convalescent sera arethe best samples for confirming acuteBDV infection. Bleeds from one animalshould always be tested alongside eachother on the same plate.

The choice of virus for use in the VNtest is difficult due to the antigenic diver-sity among pestiviruses. Reference strainsof cytopathic BD virus, e.g. Moredun orcytopathic BVD viruses, e.g. OregonC24V or NADL can be used. No singlestrain is ideal. Account should be takenof the locally predominant genotype iso-lated from sheep. Local strains should betested to see which gives the highest anti-body titre with a range of positive sheepsera (Brockman et a]., 1988). The VN canalso be performed with an NCP strain withan IP staining system being used for thereadout.

A monoclonal antibody capture ELISAfor measuring BDV antibodies has beendescribed. Two pan pestivirus mabsdetecting different epitopes on theimmunodominant non-structural proteinNS 2-3 are used to capture detergent-lysedcell culture-grown antigen. The resultscorrelate qualitatively with the VN test(Fenton et al., 1991).

6. CONTROL

The control of BDV in a sheep flockhas two essential requirements: the iden-tification of PI sheep and the prevention ofinfection of susceptible pregnant ewesespecially during the first half of gesta-tion.

The control of BDV in infected flocksis difficult and will depend on the require-ments of farmers in relation to their farm-

ing methods. In a flock which has recentlyhad a sporadic outbreak of BD, the entirelamb crop and the sheep that introducedinfection must be removed to slaughterbefore the start of the next breeding sea-son. In endemically infected flocks of highcommercial value antibody testing of dif-ferent aged sheep will identify immuneand susceptible groups and further bloodtesting can be used to identify antibodynegative, virus positive PI sheep. In other

flocks the identification and disposal ofPI sheep may not be practicable. In whichcase the level of flock immunity can beraised by deliberately exposing breedingstock to known PI lambs outwith the

breeding season. The rate of virus spreadwill be increased by close herding indoorsfor at least 3 weeks.

There is currently only one availablecommercial vaccine for the control ofBDV. This is a killed adjuvanted vaccinewhich contains representative strains ofBD and BVD-1 viruses (Brun et al., 1993).It should be administered to young ani-mals before they reach breeding age inorder to maximize their immunity duringearly pregnancy. Annual booster dosesmay be required.

To prevent introduction of BD into aflock, only home-bred replacementfemales should be used. New rams shouldbe blood tested for virus before purchaseor in quarantine after arrival on the farm.

If new ewes have to be bought theyshould also be blood tested to detect any PIvirus carriers. In the absence of blood test-

ing, and as an aid to control all infectionsof breeding, newly purchased ewes shouldalways be mated and kept separate fromthe rest of the flock until lambing time.Because of the risk of infection of sheepfrom PI cattle it is essential that pregnantewes are never mixed with cattle.

7. CONCLUSION

The control of BD remains problem-atical owing to the efficient spread of thevirus by PI carrier-sheep and limited infor-mation on cross-protection afforded bydifferent virus genotypes. Further vaccinedevelopment is required with candidatevaccines being tested for efficacy in preg-nant sheep. This approach has now beenadopted for efficacy testing of BVD vac-cines in cattle (Brownlie et al., 1995). Asour knowledge of pestivirus strains from

ruminants increases it should be possibleto develop pestivirus vaccines which pro-tect both sheep and cattle.

ACKNOWLEDGEMENTS

We thank G.D. Bell, P.S. Bramley and G.Thomson for helpful discussions and S.Edwards for supplying strains of BDV. Fund-ing for this work was provided by the ScottishOffice for Agriculture, Environment and Fish-eries Department and Grampian Pharmaceu-ticals.

REFERENCES

Barlow R.M., Patterson D.S.P., Border disease ofsheep: a virus-induced teratogenic disorder,Adv. Vet. Med. (Suppl. J. Vet. Med.) 90 (1982)87-90.

Bechcr P., Shannon A.D., Tautz N., Thiel H.-J.,Molecular characterization of border diseasevirus, a pestivirus from sheep, Virology 198(1994)542-551. 1 .

Becher P., Konig M., Paton D.J., Thiel H.-J., Fur-ther characterization of border disease virusisolates: Evidence for the presence of morethan three species within the genus pestivirus,Virology 209 ( 1995) 200-206.

Becher P., Meyers G., Shannon A.D., Thiel H.-J.,Cytopathogenicity of Border-disease virus iscorrelated with integration of cellular sequencesinto the viral genome. J. Virol. 70 (1996)2992-2998.

Becher P., Orlich M., Shannon A.D., Horner G.,Konig M., Thiel H.-J., Phylogenetic analysisof pestiviruses from domestic and wild rumi-nants, J. Gen. Virol.78 (1997) 1357-1366.

Bonniwell M.A., Nettleton P.F., Gardiner A.C., Bar-low R.M., Gilmour J.S., Border disease withoutnervous signs or fleece changes, Vet. Rec. 120(1987) 246-9.

Brockman S.J., Wood L., Edwards S., Harkness J.W.,Selection of an appropriate pestivirus strain forborder disease serodiagnosis, Vet. Rec. 122( I 988) 586-587.

Brownlie J., Clarke M.C., Hooper L.B.. Bell G.D.Protection of the bovine fetus from bovine viraldiarrhoea virus by means of a new inactivatedvaccine, Vet. Rec. 137 (1995) 58-62.

Brun A., Lacoste F., Reynaud G., Kato F., Saint-Marc B., Evaluation of the potency of an inac-tivated vaccine against border disease pestivirusinfection in sheep, in: Edwards S. (Ed.), Pro-

ceedings of the Second Symposium on Pes-tiviruses, Fondation Marcel Merieux, Annecy,1993, pp. 257-259.

Carlsson U., Border disease in sheep caused by trans-mission of vims from cattle persistently infectedwith bovine virus diarrhoea virus, Vet. Rec.128(1991)145-147.

Chappuis G., Brun A., Kato F., Dauvergne M., Rey-naud G., Duret C., Études s6rologiques etimmunologiques realisees a la suite de I’isole-ment d’un pestivirus dans un foyer ovin chezdcs moutons de L’Aveyron, in: Espinasse J.,Savey M. (Eds.), Pestiviroses des ovins et desBovins, St6 Francaise de Buiatrie, Paris, 1986,pp. 55-66.

Dekker A., Wensvoort G., Terpstra C., Six antigenicgroups within the genus pestivirus as identi-fied by cross-neutralization assays, Vet. Micro-biol., 47 (1995) 317-329.

Entrican G., Dand A., Nettleton P.F., A double mon-oclonal-antibody ELISA for detecting pestivirusantigen in the blood of viraemic cattle andsheep, Vet. Microbiol. 43 (1995) 65-74.

Fenton A., Sinclair J.A., Entrican G., Herring J.A.,Nettleton P.F., A monoclonal antibody captureELISA to detect antibody to border diseasevirus in sheep sera. Vet. Microbiol. 28 (1991)327-333.

Gardiner A.C., Nettleton P.F., Barlow R.M., Virol-ogy and immunology of a spontaneous andexperimental mucosal disease-like syndromein sheep recovered from clinical border dis-ease, J. Comp. Pathol. 93 ( 1983) 463-469.

Harkness J.W., King A.A., Terlecki S., Sands J.J..Border disease of sheep: isolation of the viruson tissue culture and experimental reproduc-tion of the disease, Vet. Rcc. 100 ( 1977) 71-72.

Howard C.J. Brownlic J., Clarke M.C., Comparisonby the neutralization assay of pairs of non-cytopathogenic and cytopathogenic strains ofbovine virus diarrhoea virus isolated from casesof mucosal disease, Vet. Microbiol. 13 ( 1987)361-369.

Monies R.J., Simpson V.R.. Syndrome in sheepresembling mucosal disease in cattle, Vet. Rec.14I ( 1997) 45 1.

Nettleton P.F., Pathogenesis and epidemiology ofborder disease. Ann. Rech. Vet. 18 (1987)147-155.

Nettleton P.F., Pestiviruses infections in ruminantsother than cattle, Rev. Sci. Tech. Off. Int. Epi-zoot. 9 ( 1990) 13 I-50.

Nettleton P.F, Entrican G.. Ruminant pestiviruses,Br. Vet. )5) (1995) 615-642.

Nettleton P.F., Gilmour J.S., Herring J.A.. SinclairJ.A., The production and survival of lambs per-sistently infected with border disease virus,

Comp. Immunol. Microbiol. Infect. Dis. 15 5(1992) 179-188.

Paton D.J., Pestivirus diversity, J. Comp. Pathol.1 12 ( 1995) 215-236.

Paton D.J., Sands J.J., Edwards S., Border diseasevirus: delineation by monoclonal antibodies,Arch. Virol. 135 (1994) 241-252.

Paton D.J., Sands J.J., Lowings J.P., Smith J.E., IbataG., Edwards S., A proposed division of the pes-tivirus genus into subgroups using monoclonalantibodies, supported by cross-neutralizationassays and genetic sequencing, Vet. Res. 26(1995) 92-109.

Paton D.J., Gunn M., Sands J., Yapp F., Drew T.,Vilcek S., Edwards S., Establishment of serialpersistent infections with bovine viral diarrhoeavirus in cattle and sheep and changes in epi-tope expression related to host species, Arch.Virol. 142 (1997) 929-938.

Ridpath J.F., Bolin S.R., Comparison of the com-plete genomic sequence of the border diseasevirus, BD3 to other pestiviruses. Virus Res. 50(1997)237-243.

Roeder P.L., Jeffrey M., Drew T., Variable natureof border disease on a single farm: the infectionstatus of affected sheep, Res. Vet. Sci. 43(1987) 28-33.

Roche P.M., Woodward M.J., Edwards, Character-isation of p20 gene sequences from a borderdisease-like pestivirus isolated from pigs, Vet.Microbiol. 33 (1992) 231-238.

Sandvik T., Paton D.J., Lowings P.J., Detection andidentification of ruminant and porcine pes-tiviruses by nested amplification of the 5untranslated cDNA region, J. Virol. Methods 64(1997) 43-56.

Terpstra C., Border disease: a congenital infection ofsmall ruminants, Prog. Vet. Microbiol.Immunol. I (1985) 175-198.

Thiel H.-J., Plagemann P.G.W., Moennig V., Pes-tiviruses, in: Fields B.B., Knipe D.M., How-ley P.M. (Eds.), Field’s Virology, 3rd ed., Lip-pincott-Raven, Philadelphia, 1996, pp.1059-1073.

Tijssen P., Pcllerin C., Lecomte J., Van Den Hurk J.,Immunodominant E2 (gp53) sequences ofhighly virulent bovine viral diarrhoea group 11viruses indicate a close resemblance to a sub-

group of border-disease viruses, Virology 217 7(1996)356-361. 1 .

Van Rijn P.A., Van Gennip H.G.P., Leendertse C.H.,Bruscke C.J.M., Paton D.J., Moormann R.J.M.,Van Oirschot J.T., Subdivision of the pestivirusgenus based on envelope glycoprotein E2,Virology 237 (1997) 337-348.

Vantsis J.T., Barlow R.M., Gardiner A.C., LinklaterK.A., The effects of challenge with homolo-gous and heterologous strains of border dis-ease virus on ewes with previous experienceof the disease, J. Coinp. Pathol. 90 (1980)39-45.

Vilcek S., Herring A.J., Herring J.A., Nettleton P.F.,Lowings J.P.L., Paton D.J., Pestiviruses iso-lated from pigs, cattle and sheep can be allo-cated into three genogroups using polymerasechain reaction and restriction endonuclease

analysis, Arch. Virol. 136 (1994) 309-323.

Vilcek S., Nettleton P.F., Paton D.J., Belak S., Molec-ular characterization of ovine pestiviruses, J.Gen. Virol. 78 (1997) 725-735.

Wensvoort G, Terpstra C., Bovine viral diarrhoeavirus infection in piglets born to sows vacci-nated against swine fever with contaminatedvirus, Res. Vet. Sci. 45 (1988) 143-148.

Wensvoort G., Terpstra C., De Kluyver E.P., Char-acterizittion of porcine and some ruminant pes-tiviruses by cross-neutralization, Vet. Micro-hiol. 20 (1989) 291-306.