research and development - department for...

MINISTRY OF AGRICULTURE, FISHERIES AND FOOD CSG 15Research and Development

Final Project Report(Not to be used for LINK projects)

Two hard copies of this form should be returned to:Research Policy and International Division, Final Reports UnitMAFF, Area 6/011A Page Street, London SW1P 4PQ

An electronic version should be e-mailed to [email protected]

Project title Equine viral arteritis: virulence and immunity

MAFF project code SE0750

Contractor organisation and location

Veterinary Laboratories Agency (Weybridge), Woodham Lane, New Haw, ADDLESTONE, Surrey. KT15 3NB

Total MAFF project costs £ 259,835

Project start date 01/04/97 Project end date 31/03/02

Executive summary (maximum 2 sides A4)

There is no information on why some strains of EAV are more virulent than others. Genetic comparisons of field strains of differing virulence is unrewarding, because virulence markers, which may be very subtle, are masked by other differences that are of no relevance. In order to circumvent this difficulty, it is necessary either to obtain or to generate virus pairs that are very closely related, but which have differences in virulence. Genetic comparison of such pairs can identify the determinants of virulence. Once pinpointed in this way, the markers can be readily looked for in other viruses to assess the generality of their impact. Recent work on a related virus of pigs (PRRS) has demonstrated that virulent and non-virulent variants can be identified within a strain on the basis of plaque size in vitro. A better understanding of the immune response to infection could lead to improved methods for distinguishing between vaccinated and non-vaccinated seropositive stallions and between carrier and non-carrier seropositive stallions. If such discrimination could be made on the basis of a blood test this would be a tremendous advantage. Very little is known concerning cell-mediated immune responses, but it is likely that killed vaccines of the type used in Britain would not induce the same type of immunity as wild-type infection and, similarly, virus persistence may be linked to a deficit in the immune response. A first step in investigating cell-mediated immunity is to establish in vitro methods for its quantification and to test the methodology on horses experimentally infected with the equine arteritis virus.The advantages of RT-PCR as a means of detecting viral agents in semen are established. Various assay formats have been developed for EVA and it has been shown that RT-PCR can be more rapid and sensitive than the alternative of virus isolation in cell culture. Our own experience confirms this, in that parallel testing has revealed two stallions from which virus was only detected in semen by RT-PCR. Nevertheless, statutory acceptance of the test has been slow, because of concern that not all genomic variants of EAV may be detected; and because it is difficult to ensure that false positive and false negative results do not occasionally occur. A large and diverse collection of EAV isolates has now been characterised at the genetic level and this offers an opportunity to reassess the specificity of RT-PCR primers used for diagnostic purposes. Additional tools for CSG 15 (Rev. 12/99) 1

Projecttitle

Equine viral arteritis: virulence and immunity

MAFFproject code

SE0750

improving the reliability of RT-PCR have also become available, including novel techniques and an infectious clone of EAV. It is intended to use these to establish a highly reproducible method with a high probability of gaining international acceptance.

This project was supported by matched funding from the European Union (FAIR6-CT98-4123; Key action 5). Specific collaboration was identified between VLA Weybridge and AHT Newmarket, who were also a su-contractor within this ROAME.

The virulent horse passaged “pleural fluid” form of the Bucyrus strain of equine arteritis virus (EAV) was obtained from America and was sub-cloned in vitro to derive a number of homogeneous progeny viruses. Two of these clones which have different plaque sizes (small and large) when grown in cell culture, were inoculated into horses. Both produced clear evidence of disease and of extensive in vivo replication. The most serious illness and profound temperature elevation was seen with the small plaque variant. It is considered that this clone can be used as a representative “virulent” virus. A third clone was derived from the “pleural fluid” virus by extensive passage in a heterologous cell culture system. This “vero adapted” variant is a candidate “avirulent” virus. In the second year, we completed two further experimental horse infection studies. In the first of these, the “vero adapted” virus was inoculated into ponies and shown to be avirulent, causing no illness and replicating to only a very limited extent. In a second experiment, the original “pleural fluid” isolate was inoculated into ponies and its virulence in our challenge model was confirmed. These studies demonstrate that we have successfully generated closely related viruses with striking phenotypic differences. The next stage is to try to relate the phenotypic differences to genotypic markers. To this end the viruses were supplied to partner 1, where the complete genetic sequences of each have been determined.

Determination of the full-length sequence of a virulent derivative of the large plaque (LP) variant, LP3A.1 was initiated and performed by various partners to extend this comparative analysis. Sequencing of products of independent RT-PCR assays were used instead of cloned RT-PCR products (as was done before). P1 prepared and sent the RT-PCR fragments to P2, P3, and P5 for sequencing.

In year one, we compared the plaque sizes of viruses of differing virulence to see if this attribute could correlate to virulence. Similar determinations were carried out for some additional viruses in year two and three, but overall no clear correlation was found. Another possible marker was thought to be rate of in vitro replication, and in year one, one-step growth curves were established for each of our in vitro generated viral variants. In year three again in vitro replication studies were repeated on Bucyrus cloned variants in an attempt to see if there was any difference between low and high virulence cloned viruses. Again, no decisive differences were noted between virulent and avirulent isolates. In year two, we completed these studies by looking at growth efficiency at different temperatures. It was found that all of the viruses analysed grew better at 37ºC than at either 34ºC or 40ºC. There was a suggestion that low virulent viruses grew relatively better at 34ºC than high virulence ones, but more work is needed to confirm a definite association. In conclusion, thus far, a clear in vitro marker for in vivo virulence has not been found.

The different RT-PCR methods described for EAV have been tabulated, and the different methods have been compared in our laboratory. A single, suitable method has been selected for optimisation and the incorporation of a TaqMan fluorogenic probe has given consistent results. The development of a control mimic, followed by a final validation of the TaqMan on different strains of EAV, semen and clinical samples to confirm the chosen method’s reliability has also been completed.

This project has provided DEFRA with excellent added value, since the development of a cytotoxic T cell assay was an opportunistic activity, dependent on horse experiments carried out as part of an EU project. If paid for by DEFRA just for this work, it would have been prohibitively expensive. Additionally, the collaboration of AHT in this work was invaluable and reaped great dividends for both institutes.

CSG 15 (1/00) 2

Projecttitle


MAFFproject code

SE0750

Scientific report (maximum 20 sides A4)

CSG 15 (1/00) 3

The three objectives within the project were as follows:

01 Identification of virulence determinants in the genetic sequence of EAV by selection of closely related viruses with differing virulence, followed by complete genomic comparison.

02 Establish an in vitro assay for measurement of the cell-mediated immunity (CMI) to EAV and use it to examine CMI in experimentally infected horses.

03 Establish and validate a second generation RT-PCR test for detection of EAV in semen, using fluorescent probes to improve specificity, and a mimic system to improve quality control.

Objective 1. Identification of virulence determinants in the genetic sequence of EAV by selection of closely related viruses with differing virulence, followed by complete genomic comparison.

Our strategy towards identification of virulence markers relied on the production of pairs of similar viruses with differing virulence, derived from a virulent precursor virus. We employed two approaches to this end. The first approach utilises a method successfully employed with a related virus, porcine reproductive and respiratory syndrome virus (PRRS), by isolation of large and small plaque variants, so-called because one produces large foci of infected cells in cell culture, whereas the other produces only small foci. The PRRS study showed that large plaque variants are more virulent than small plaque variants. Our second approach to produce related high and low virulence pairs has employed the strategy of passaging a virulent virus several times in a cell type from an abnormal host - African Green Monkey cells, in an attempt to attenuate it.

Methodology

a) Development of method for examination and selection of viral plaques

Plaque purification of virus stocks was required in order to provide a homogeneous viral population for meaningful genotypic and phenotypic comparisons to be made. Secondly it was planned to evaluate differences in plaque morphology and to see if this correlated to viral virulence.

A method of observing EAV plaques has been reported by Hyllseth (1969). Our method was an adaptation of this and of the methods of Park (1996). It involved inoculating 6 well plates of equine embryonic lung cells (EELs) with EAV. After allowing for adsorption, the cells were washed and then overlaid with maintenance medium in a low melting point agarose to facilitate individual plaque growth. After 3 days incubation the plaques were visualised either by neutral red staining (if the plaques were to be picked and a stock grown for further plaquing) or by staining with Crystal violet for permanence prior to measurement and photography.

b) Selection of viral stock from which to derive clones

Most isolates of EAV are not very virulent when inoculated into horses. The most well characterised virulent EAV is the velogenic Bucyrus strain in the form of pleural fluid derived from experimentally infected horses. This is said to cause fatal illness in a proportion of experimentally inoculated animals. A stock of this virus was obtained from Dr McCollum in Kentucky. Our hypothesis is that this will be a mixed population of closely related virus variants from which we will attempt to derive high and low virulence virus pairs. Because such viruses are expected to have few genetic differences, there is a more realistic possibility of determining which of the differences are significant with respect to virulence.

c) Selection of large and small plaque variants

Projecttitle


MAFFproject code SE0750

Because it is expected that cell culture passage may attenuate EAV virulence, a minimum number of passages should be used to derive clonal virus sub-populations, and an equine cell culture system should be used.

Description of cloning and selection procedure:A stock of the velogenic Bucyrus strain of EAV was obtained by inoculating an 800ml flask of EEL cells with the pleural fluid isolate, and after 3 days when a cytopathic effect was observed the virus supernatant was harvested. The virus stock was divided into two.One half of the stock was aliquoted into 1ml tubes and frozen at -800C and the other half aliquoted into 1ml glass vials and freeze dried.

The stock Bucyrus virus was titrated out and inoculated into a 6 well plate of EEL cells. The virus was adsorbed for 1 hour in 5% CO2, washed and then overlaid with MEM 2% FCS in 1% low melting point agarose. After 3 days the cultures were examined for discrete plaques. The small and large plaques were picked separately and each inoculated separately into 25cm flasks of just confluent EEL cells, to allow for growth of separate stocks of increasingly purer small and large plaque viruses. These stocks were in turn titrated and plaques picked twice more as shown in the diagram below, before final stocks of small and large plaques were cultured for horse challenge experiments. The final stocks were centrifuged, filtered to remove cellular debris and then aliquoted in 1ml amounts for storage at -800C.

d) Selection of culture attenuated variant

Description of selection and subsequent cloning procedure:The first passage stock of Bucyrus virus above was passaged 20x in Vero cells, 3x in Baby hamster kidney cells (BHK-21) and then a further 6 passages in Vero cells. A final stock was cloned and plaque picked as above, for use in the equine challenge experiments.

e) Sequence analysis of selected clones

Analysis of the large and small plaque variants and the Vero-adapted variant was performed by sequencing the GL gene. The method used was the Perkin Elmer ABI PRISM Dye Terminator Sequencing Ready reaction Kit, with AmpliTaq DNA Polymerase.

f) Growth curves for supernatant and cell-associated virusesIn vitro growth curves were determined in cell cultures for supernatant and cell-associated virus. Viruses tested were the three clones derived from the velogenic Bucyrus virus as described above, the ARVAC vaccine strain, and the Vienna isolate.For each virus, flasks of EEL cells (25cm2) were inoculated with 1ml of virus (100 TCID50), allowed to adsorb for 1hour, then washed once in MEM without serum and then over-laid with MEM containing 2% FCS. Starting at time 0 and then at mainly 2 hour intervals the flasks were harvested for supernatant and cell associated virus. The cell associated virus was harvested by first collecting the supernatant virus, then the cells were washed twice with MEM and then 2ml of MEM added to the flask before freezing at -80oC.In order to obtain comparable results, all the harvests from both the supernatant and cell associated viruses were thawed and titrated on the same day and on the same stock of EEL cells. After thawing, the cells and

5

Projecttitle



supernatants from the flasks for the cell associated virus were centrifuged before titrating out the supernatants. All virus titrations were prepared in Bijoux and then 25µl was transferred to 96 well microtitre plates containing 75µl of EEL cell suspension at 3x105/ml. All readings were made after 3 days incubation at 37oC in 5% CO2.

g) Antigenic typing.The antigenic similarity of the new Bucyrus variants was assessed by typing with a panel of fifteen monoclonal antibodies previously prepared at VLA. These are virus neutralising mAbs directed at epitopes on the GL protein of EAV. The method used was to grow each of the viruses in microtitre cell cultures and then to apply the mAbs to the fixed cell cultures. Binding of the murine mAbs was assessed by immunostaining with a peroxidase conjugated anti-mouse immunogloulin.

Results and conclusions

Growth characteristicsThe growth characteristics of the variant viruses were studied in vitro, and compared to a number of other laboratory strains of EAV. Their phenotypic characteristics and their effects in ponies, performed in part for the cell-mediated immunity studies (Objective 2) and performed at AHT Newmarket, are given in the table below:Viruses Virulent/non-virulent Plaque size Range of sizes

small/med/ large in mm Vel.Buc.L.P. Virulent Large 4-4.5Vel. Buc.S.P. Non-virulent? Small/med 1.5-2.5Horse pass L.P. Virulent? Large 4-.5Horse pass S.P. Non-virulent Small/med 2-2.5ARVAC Non-virulent Med. 2.5-3.5Vienna Virulent Small/med 0.5- 2.0Bucyrus VLA Unknown Med. 2.5-3.5Vero pass Non-virulent Small/med 2.2-3.2

Additionally, these clones retained their plaque morphology after passage in ponies, indicating that they were relatively stable with respect to this characteristic over the time period of the experiments. These results indicated that, whilst these clones can be distinguished phenotypically in vitro, there does not seem to be any relationship between size of plaque and the pathogenicity of the virus.

Growth curves for supernatant and cell-associated viruses. Similar growth curves were obtained for all phenotypes, both examining for yields of virus in supernatant and cell-associated virus. This indicates that the differences seen in the virulence in vivo is unlikely to be due to increased rates of replication.

Genetic analysis of clonesThe full-length sequences derived from the four variants were compared with one another and with the Leiden clone. These comparisons showed nucleotide differences at various positions among the five sequences. Thirty-six of nucleotide changes were at

6

Projecttitle



a common position in the cell cultured variants (and the Leiden virus), and resulted in amino acid changes in 21 of the occasions (11 non-conservative and 10 conservative changes). Nucleotide substitutions common to two variants were found in the following instances: PF/VC (AG 2243, nsp2); PF/LP (GA 6132, nsp9); PF/EAV (UC 10722, ORF 3 and ORF4); PF/EAV (UC 11044, ORF4); SP/LP (CU 3707, nsp4); LP/SP (10189, ORF2b); SP/EAV (GA 11386, ORF5); VC/EAV (CU 6752, nsp9).

The frequency of nucleotide substitutions was highest in the regions of ORF1ab encoding nsp2 and nsp9, and in ORF2b and ORF5, the latter particularly for the VC variant.

Changes notable for affecting regions of known function or structure were as follows:The LP variant was the one where most of the nucleotide substitutions were conservative, i.e., resulting in functionally compensatory amino acid changes. The five non-conservative amino acid changes found apparently were not located in motifs implicated in virus biology.

In the SP variant there were 9 nucleotide changes resulting in non-conservative amino acid substitutions, of which aGA change from serine (Ser) to asparagine (Asn) at position 2371. This Ser is the residue at the P1´ position in the cleavage site between nsp9 and nsp10 (E2370/S2371).

A number of nucleotide substitutions resulting in amino acid changes in the VC variant were observed. Of these, a UC change in nucleotide position 606 determined a non-conservative change from Phe to Leu at amino acid position 128, within a set of residues considered invariable in the PCP1a protease (Ziebuhr et al., 2000). At nucleotide position 11142, a UA change resulted in a non-conservative change of amino acid 148 in ORF4, from glutamine to leucine. This change is located at a sequence motif with the function of an alternative regulating signal for transcription of subgenomic mRNA5. Compared to the sequence of the Leiden clone of EAV Bucyrus which was used in the experimental studies to demonstrate the function of this motif), the VC variant has an additional change in the last base of this sequence motif (UCA/UACG/A, changed bases in bold). The positions corresponding to EAV nucleotides 11299 to 11304 are missing in the VC strain, determining a deletion of two amino acids, Thr52 and Ala53, in ORF5 encoding the major GP5 glycoprotein. Another change at nucleotide position 11389 (GA) resulted in an aspartate to asparagine shift at amino acid 82 of ORF5, introducing a potential second N glycosylation site. In general, the VC variant had more mutations than other variants in nsp2, nsp9 and ORF5. Many of the changes in ORF5 were clustered within neutralization epitope site C (amino acids 67-90)

The sequence of the Leiden clone of EAV Bucyrus had the most nucleotide substitutions in relation to the pleural fluid sequence. However, most of these changes were either at the third base position and did not change the amino acid sequence or were conservative substitutions. There were only three non-conservative amino acid changes unique to this strain.

Antigenic analysis

Fourteen of the fifteen mAbs reacted with all of the Bucyrus-derived viruses (SP, LP, PF, Leiden, and ARVAC vaccine) except for the VC clone. This virus was recognised by only two of the 15 mAbs (WB10 and WB32), and even these reactions were weak. One mAb, WB38, recognised the ARVAC, SP, LP and PF viruses, but not the VC or Leiden strains.

A comparative analysis of GL, encoded by ORF5, revealed some minor differences in predicted secondary structure among the derivatives, compared to the original PF isolate. The major difference was the loss of a predicted alpha helix domain in amino acid region 70-80 for the VC clone (data not shown). An analysis of predicted antigenic index showed that the most striking change was in the VC clone, compared to the original PF isolate. Four regions of difference were detected. There were two regions of decreased antigenic index, located in regions 36-41 and 204-211. In region 49-57, there was an area of moderately increased antigenicity,

7

Projecttitle



in the area where two amino acids were lost, followed by an area of decreased antigenicity. The largest difference in the antigenic index plots was seen between amino acids 70 and 87. These changes resulted in an area with a high antigenicity, in the first seven amino acids of that region, followed by ten amino acids with lower antigenicity, when compared with the same region of the original PF isolate. This is the same region where a glycosylation site is lost in the VC clone (amino acid 81 of the PF isolate). Few changes were seen when small and large plaque variants were compared, though there was a small area of difference in the LP variant in region 220 to 230, which resulted in a less antigenic/more hydrophilic region in this area.

Equine challenge studies

In each experiment, the clinical signs exhibited by the two ponies were rather similar. All of the ponies inoculated with the LP, SP and PF viruses showed obvious signs of illness including pyrexia and depression. The LP and especially the SP inoculated ponies also showed conjunctivitis, nasal discharge and enlargement of their submandibular lymph nodes. Overall, the most severe signs were seen in the SP inoculated ponies. The two ponies challenged with the VC clone of EAV showed no visibly observable illness, but one pony had mild pyrexia of 38.9ºC for one day. All of the ponies had clinically recovered by 12 dpi.

The virus isolation and serology results for the two ponies in each experiment were also extremely similar to one another. The ponies inoculated with the SP, LP and PF viruses shed virus in their nasal swabs for 9-12 dpi and in rectal swabs for 3-8 dpi. They also had viraemias of 5-11 days duration. In contrast, the level and duration of nasal virus shedding was lower for the VC challenged ponies. These animals had neither detectable viraemia nor rectal virus shedding.

Results and conclusions

The two plaque-purified both replicated extensively and produced obvious illness in the experimentally inoculated ponies. The SP variant produced more marked clinical illness and a longer duration of replication than the LP variant. It also induced a longer lasting antibody response. The results for each pair of ponies were very similar and therefore, the subtle differences between pairs receiving PF, LP and SP viruses may be real. However, the small number of ponies used means that only a crude assessment of differences in viral virulence is possible, and only the VC virus can be considered to be clearly different from the others. The PF variant itself was less virulent in our hands than has been reported by McCollum and Timoney and could not be clearly differentiated from the LP and SP viruses on the basis of induced clinical illness nor extent of viral replication. Therefore, neither the cloning itself, nor the small number of equine cell culture passages needed to produce the SP and LP clones had a marked attenuating effect.

These results suggest that in vitro plaque size is not a reliable indicator of in vivo virulence for EAV. This conclusion is also supported by the analysis of other EAV viruses of defined or relatively defined virulence. In the closely related virus PRRSV, large plaque variants were found to be more pathogenic in pregnant pigs (Park et al.,1996). In foot-and-mouth disease virus, virulence has also been correlated to large plaque size (Borgen et al., 1964) and it has been suggested that large plaque variants may induce a weaker interferon response (Chinsangaram et al., 1999). In Sindbis virus, large plaque variants bind less strongly to heparin sulphate and are more pathogenic in vivo, possibly due to a reduced speed of clearance from the blood (Byrnes and Griffin, 2000). The molecular basis for the plaque size differences of our EAV variants remains to be established. Hyllseth (1969) showed that the plaque size of EAV was increased in the presence of the polycation DEAE-Dextran suggesting that interaction between the viral envelope and negatively charged cell surface receptors could also be a factor in EAV plaque size differences. Meanwhile, EAV is known to bind to heparin (Sano et al., 1998). Comparing our SP and LP variants, it is noteworthy that they have very few differences with respect to the amino acid sequences of their surface glycoproteins. The change from arginine to cysteine at position 52 of GS is in a site of high predicted antigenicity and contributes to an increase in the negative charge of the LP

8

Projecttitle



variant (charge at pH 7 of 4.86 compared to 5.90). This could result in reduced binding to cellular glycosaminoglycans such as heparin sulphate, consistent with the Sindbis and foot-and-mouth disease models for binding to heparin sulphate).

The VC variant, which had been passaged in vitro 30 times, was highly attenuated producing neither obvious illness nor a detectable viraemia. This virus therefore behaved similarly to the vaccine viruses previously reported upon, which had been passaged even more extensively in horse, rabbit and baby hamster kidney cell cultures (Doll et al., 1968; McCollum, 1969; McKinnon et al., 1986; Timoney et al., 1987). These vaccine viruses had to be given intramuscularly to elicit good immune responses, and this could account for the low levels of antibody produced after intranasal inoculation of the VC clone. On the basis of the definitions provided by McCollum and Timoney (1999) the infections elicited by the PF, LP and SP variants would be categorised as moderately severe, whilst that of the VC variant would be considered subclinical.

A herpesvirus was detected in cell cultures inoculated with nasal swab material from the SP inoculated ponies. This virus was characterised as EHV-2 and shown to be present in samples from all of the inoculated ponies. Therefore, it does not seem likely that it had a confounding effect on the determination of the virulence of the EAV variants. Although the small number of ponies used in each experiment reduces the certainty about the true virulence of each of the viral variants, additional experiments have now been conducted with the LP variant as part of vaccine development trials (Castillo-Olivares et al., 2001). In these studies, 18 ponies were intranasally challenged with the same dose of LP variant and the 6 unvaccinated control animals showed very similar responses as the two reported on here.

A striking finding from the genetic comparisons of these Bucyrus virus variants was the large number of nucleotide substitutions common to all of the cell culture grown variants compared to the horse passaged PF isolate. These changes are therefore likely to represent adaptations to growth in cell culture.

Of the many other differences amongst the viral variants, relatively few can be linked to known functional domains. As expected, the low virulent VC variant had more changes from the PF ancestor than the less extensively cultured LP and SP variants. These were particularly prominent in the genomic regions encoding nsp2, nsp9 and GL (ORF5). The changes in ORF5 included a two amino acid deletion at positions 52 and 53, and six other amino acid changes, several between positions 73 and 87 in a previously defined neutralization domain. One of these encoded an extra N-linked glycosylation site at position 82. The VC clone also showed a marked antigenic difference from the other viruses, with an altered antigenic index and hydropathy profile, which was likely the reason why it was no longer recognised by many mAbs that reacted with the other variants. Two escape mutants generated using one of these mAbs (WE20) had changes in GL at positions 71 and 74 (unpublished data). Interestingly, none of the changes in the ORF5 of the VC variant are common to the ARVAC vaccine strain, suggesting that they cannot be considered general markers of attenuation. The same two amino acid deletion found in the VC clone has also been reported in another Bucyrus variant which had also been passaged in Vero cells (EAV-UCD, Balasuriya et al., 1995). EAV-UCD also had identical mutations to the VC clone at positions 73, 84 and 87 of the GL protein. This would be consistent with this part of GL

interacting with a receptor that is altered in Vero cells compared to equine cells. Meanwhile, the additional glycosylation site found in the VC clone but not the PF isolate was reported as being present in an earlier ORF5 sequence of the PF variant passaged twice in RK13 cells (EAV-VIRU, Balasuriya et al., 1995).

In the SP variant a mutation changing a serine residue to asparagine n was found at the cleavage site nsp9/10, This Ser2371 is conserved among the few EAV sequences so far analysed, but in PRRSV and LDV, it is replaced by glycine and lysine, respectively (van Dinten et al., 1999), indicating that the serine at this position may not be a strict requirement for cleavage.

The fewer number and the conservative nature of the substitutions in the large plaque variant seem to indicate either that this variant has become more stable or that it represents a clone that is closer to the original isolate.

9

Projecttitle



The full conservation of gene regions (in general nsp4 to nsp7, nsp11 and nsp12; ORF2a, 3, 6 and 7) show that they are either not prone to changes or reflect conservation imposed by functional constraints. The nsp4 harbours the main protease of EAV, the 3CLSP (Snijder et al., 1996), therefore the high level of conservation is consistent with this feature. Although the roles of nsp 5, 6 and 7 are not known, their strict conservation suggests also functional importance. The number of changes in the nsp9 coding region were unexpected as it would be tempting to assume that its function as the RdRp is likely to impose a high degree of conservation. Those changes may not be detrimental as most of them are of the conservative type.In PRRSV, mutations found in different regions of the genome, particularly nsp1ß, ORF2 and ORF5 have been suggested to contribute to virus attenuation (Allende et al., 2000; Yuan et al., 2001).

In conclusion, we have shown that clonal populations can be established from horse passaged EAV without obvious reduction in virulence, but that in vitro plaque size differences do not correlate to virulence. Nevertheless, even low level cell passage resulted in a considerable number of differences in their genetic sequence from the consensus sequence of the PF ancestor. It would be interesting to know the extent and nature of the quasispecies of viruses that underlie this PF consensus sequence. Many of the changes present in the LP and SP sequences appear to be common to all cell culture passaged Bucyrus variants. More extensive cell culture passage resulted in a highly attenuated virus (VC) with a greater number of differences from the PF ancestor. From these studies, it is apparent that different factors exerting effect on EAV biology (processing efficiency, replicase activity, transcription regulation) among host-related factors, may be involved in determination of virulence of this virus. Whether and which isolated or combined mutations identified on comparison of parental and derived viruses represent virulence markers deserve further investigations.

Objective 2. To establish an in vitro assay for measurement of the cell-mediated immunity (CMI) to EAV and use it to examine CMI in experimentally infected horses.

Methodologya) VirusesA derivative of the EAV strain LP3A (Castillo-Olivares et al., 2001) was used for CTL assay and for the experimental infection of ponies. This virus, designated EAV LP3A+, was obtained by a single passage of EAV LP3A in Equine Embryonic Lung cells (EEL’s). The Bucyrus strain of EAV as prepared by Central Veterinary Laboratory was used in the virus neutralisation test (Edwards et al., 1999). b) Experimental infection:Three EAV seronegative 2-year-old male castrated ponies (7378, 027A and 5D66) were infected by nebulising into the nasopharynx 106 TCID50 of the EAV LP3A+ strain as described previously (Castillo-Olivares et al., 2001). The animals were kept in a contained environment facility during the acute phase of the infection, released when they were no longer excreting virus to an isolated barn for 1 week and then moved to an isolated paddock for another week before they were allowed to have contact with other equines. During the acute phase of the disease the ponies were inspected twice a day to monitor clinical signs. During this period nasopharyngeal swabs and unclotted blood samples were collected for virus isolation and serum samples for assessment of the antibody response. Peripheral blood mononuclear cells were collected pre-infection and at different times during convalescence to analyse the virus specific cytotoxic lymphocyte responses. c) Virus isolation and serologyProcedures to detect and quantify virus load on nasopharyngeal swab extracts and blood and analysis of the antibody responses were performed as described elsewhere (Castillo-Olivares et al., 2001). d) Establishment of primary Equine Dermal Cell Lines Equine dermal primary cell (EDC) lines were established for each pony used in the study. Skin punch biopsies were performed under aseptic conditions, the epidermal portion of the sample removed and the dermal plug inmersed in 20% MEM (Eagle’s minimum essential medium supplemented with antibiotics and 20% foetal calf serum) and transported to the laboratory for immediate processing. The dermal plug was sliced with the help of a sterile scalpel blade and forceps and each slice laid on a well of a 6 well flat bottomed plate. Then fresh 20%

10

Projecttitle



MEM was added to each well and the plates incubated at 37C, 5%CO2 for five days. Once islets of fibroblasts were observed growing, the inoculum was removed and fresh 20% MEM was added and incubation continued at 37C, 5%CO2 until a confluent monolayer was formed. Then the cells were passaged to a 25 cm2 tissue culture flask, expanded and finally resuspended in 10% DMSO in FCS for cryopreservation in liquid N 2 . In the experiments described in this study the cells were used between passages 5 and 14.e) Preparation of targets for 51Cr release cytolytic assayTwenty four hours before the effectors are incubated with the targets, the EDC monolayers were washed twice with PBS (phosphate buffered saline) and cells detached from the vessel surface after incubation with trypsin / EDTA in PBS, 37C, 5 min. Once detached, the cells were resuspended in 10%MEM and centrifuged for 5 min. at 1400 rpm., the supernatant discarded and the cells resuspended in 20% MEM to contain 4 x 105 cells / ml. Half of the cells were inoculated with EAV LP3A+ at approximately an m.o.i. of 0.3; the other half was left uninfected. Each set of cells was then radiolabelled with 15 Ci / ml of Na51CrO4 (Amersham) and 100l / well of each suspension added to half of the wells of a 96 well flat bottomed plate and incubated at 37C, 5% CO2

for 24 hours.f) In vitro stimulation of PBMCWhole venous blood was collected at various times during the course of the study into vacuum tubes containing 10 international units of Sodium Heparin in PBS, to isolate the mononuclear cell fraction by Ficoll-Hypaque density gradient centrifugation. Following fractionation, the PBMC’s were washed 3 times in PBS to remove the platelets as much as possible and resuspended in either 10% DMSO in FCS or in induction medium (1:1 vol/vol mixture of AIM-V / RPMI 1640 suplemmented with 2mM L-glutamine, minimal essential medium non-essential amino acids (0.05 mM each), 0.5 mM sodium pyruvate, 2-mercaptoethanol (55 M), gentamicin (50 gr/ml) and equine serum (7%) collected from the ponies before EAV experimental infection and inactivated at 56C for 40 min).The PBMC were incubated in induction medium for 7 days in upright 75 cm2 tissue culture flasks at 1.1 – 2.0 x 108 cells / flask / 40 mls in the presence or absence of 106.1 TCID50 of EAV LP3A+.g) CTL assayEAV stimulated and unstimulated PBMC cultures were centrifuged at 1400 rpm, 20C, 10 min, without brake, the supernatant discarded and cell pellets resuspended in CTL medium (RPMI 1640 containing 10% heat inactivated equine serum). Both cultures were adjusted to contain the same concentrations of viable cells (determined by Trypan blue exclusion staining) and diluted appropriately in CTL medium to obtain different effector: target ratios when added to the overnight grown fibroblasts. These were washed 3 times with RPMI1640 using 125 ul / well in each wash before the addition of either the effectors, a CTL medium control or total cell lysis solution (2% Triton-X100 in PBS). The plates were incubated for 4 hours at 37C, 5% CO2 after which the contents of each microwell were harvested (Supernatant Harvesting System, Skatron, Newmarket, Suffolk) for quantitation of 51Cr release by gamma counting. The lytic activity of each PBMC culture dilution was assessed against 4 – 6 replicates of EAV infected and uninfected autologous allogeneic radiolabellled targets. The percentage of specific 51Cr release was calculated according to the following mathematical expression: [(e – sp) / (t – sp) ] x100, where e is the experimental 51Cr release, sp is the 51Cr release of the targets incubated with lysis medium only, t represents the total 51Cr release from targets which were incubated with total cell lysis solution.h) Indirect immunofluorescenceEquine dermal fibroblasts or cytospins of PBMC’s were fixed in 4% Formaldehyde, 0.4% Triton-X 100 in PBS for 15 min at room temperature. After washing in PBS, an EAV N specific rabbit polyclonal antiserum (de Vries et al., 1992) diluted 1/100 in 2% Bovine Serum Albumin in PBS was applied to the cells and incubated for 1 hour at 37C. After washing in PBS, the samples were incubated for 1 hour at 37C with an antirabbit IgG FITC conjugated antibody, washed again in PBS and observed on a fluorescence microscope.

Results

a) Clinical, virological and serological response to infectionThe symptomatology presented by the LP3A+ challenged ponies was characteristic of EAV. Although there are natural variations between the response of each individual we can divide the syndrome in three phases: an early

11

Projecttitle



phase, from days 2 to day 4, where a general infection was established and characterised mainly by pyrexia; a second phase from days 5 to 8 where clinical signs (including conjunctivitis, stiff gait, anorexia, depression, ataxia) were observed; and a convalescent phase, from days 9 to 14, characterised by regression of signs of infection and progressive improvement of weakness and general condition of the animal. By day 21 post infection all ponies had recovered completely and remained in good and healthy condition until the end of the study period. Viraemia was detected from day 2 to day 14 post-infection in all three ponies. Ponies 5D66 and 7378 were also viraemic on day 21 (data not shown). Virus nasal excretion began soon after infection reaching maximum levels between days 4 and 6, coinciding with the second clinical phase of the syndrome described above, and stopped completely by day 14 in all three ponies. Virus neutralising antibodies in serum were detectable on day 6 post-infection reaching significant levels on day 8, coinciding with the beginning of the clinical recovery. The antibody titres increased gradually reaching high values on day 14 post infection. VN test results on serum samples collected 6 and 12 months post-infection showed no decrease in titres.

b) Cytotolytic activity of in vitro stimulated PBMCAttempts to detect cell mediated cytotoxic responses against EAV were made using in vitro stimulated (by EAV infection) PBMC recovered from the EAV infected ponies. PBMC (either EAV-stimulated or mock-stimulated) were tested for lysis of either EAV-infected or mock-infected equine dermal cells, recovered from the same or a different pony to the PBMC. The fibroblast-like EDC’s were tested by immunofluorescence microscopy and flow cytometry for the expression of MHC-I and MHC-II molecules in the cell surface using either the anti-equine MCH-I monoclonal antibody H58A or the anti-equine MHC-II monoclonal antibody H42A (VMRD, Inc., Pullman, USA). Only MHC-I was detectable in these cells by immunofluorescence (data not shown) or by flow cytometry. These cells are susceptible to infection with EAV and under the conditions of the assay a high percentage of cells (>80%) express the nucleocapsid gene of EAV 24 hours after infection (i.e. the time when effectors were added to the targets) conserving the original cell morphology. The levels of radioactivity retained and released spontaneously by EAV infected and mock infected EDC’s . The mean 51Cr spontaneous release of all assays performed in this study for EAV infected EDC’s [22.6% (10 %– 33.7%)] and mock infected EDC’s [20% (8 % – 33 %)] indicate that under the conditions of the assay these cells are suitable for measuring cell lysis in a cytotoxicity assay.Various attempts to demonstrate EAV specific cytotoxicity of in vitro EAV-stimulated, in vivo primed PBMC failed when the effectors were derived from cryopreserved stocks. In all these assays, EAV antigen expression was detected in a very high percentage of target cells (>80%). When freshly collected PBMC were employed however, EAV specific cytotoxicity was observed. Pony 7378 PBMC’s , collected 4 months post-infection and incubated in vitro with the virus, lysed autologous EAV infected targets. The percentage lysis of EAV-infected targets incubated with EAV-stimulated PBMC cultures was more than double that of mock-stimulated PBMC’s against EAV-infected targets or EAV-stimulated PBMC’s against uninfected targets. Hematoxylin-eosin staining of 7 day-old PBMC cultures showed an almost complete lack of monocytes in the EAV induced effectors as opposed to the uninfected PBMC’s. Moreover, the lymphocytes of EAV-induced (but not mock-induced) cultures had the appearance of activated lymphoblasts, being generally enlarged and with several mitotic figures apparent in each field. Samples from these same cultures taken 24 hours after inoculation with EAV revealed that approximately 5% of the cells expressed EAV Nucleocapsid antigen. The morphology of the infected cells suggested that they were monocytes.

c) Genetic restriction of cytotoxic responsesTo determine whether the virus specific cytolytic activity detected in EAV stimulated PBMC collected from convalescent ponies was genetically restricted, we performed a series of cross matching cytotoxicity assays . In these experiments EAV stimulated and mock stimulated PBMC’s from one individual were incubated with EAV infected target cells from the same individual or from a different pony. The first of the experiments showed that EAV induced 7378 PBMC collected 6 months post-infection specifically lysing autologous and 027A derived EAV infected targets. All uninfected and 5D66 derived EAV infected cells showed low levels of lysis. Assuming that cytolysis was mediated by class I restricted CTL, the data suggested that pony 7378

12

Projecttitle



shared an MHC-I allele with pony 027A but not with 5D66. To test this hypothesis, a second cross matching cytolysis experiment was carried out in which PBMC’s collected from pony 5D66 six and a half months post-challenge, were incubated with EAV infected or uninfected equine dermal cells from 5D66, 7378 and 027A. Only killing of autologous EAV infected targets was observed, supporting the hypothesis of genetic restriction of the effector cells. In a further experiment, EAV induced and mock induced 027A PBMC, collected 8 months post-infection, were incubated with infected and uninfected targets from the three ponies. Results of this experiment indicated that the 027A effector cells were recognising EAV-infected 027A and 7378 target cells, but not uninfected cells or EAV-infected 5D66 . Mock stimulated effectors did not induce significant specific 51Cr release values in any of these experiments.

DiscussionWe have attempted in this study to determine whether EAV infection in ponies induces a cytotoxic, cell-mediated immune response. For this purpose we have infected three seronegative castrated male ponies using a derivative of a previously tested strain. The symptoms observed in the present study resembled those displayed by ponies infected with the parental virus (Castillo-Olivares et al., 2001). However, anorexia, lethargy, weight loss and pyrexia were slightly more pronounced and lasted longer and ataxia, not a common EAV symptom, was observed in one pony. None of the ponies showed seroconversion to EHV-1, EHV-4, Equine Rhinovirus-1 or Equine Rhinovirus-2. The virological and serological outcome was similar to EAV LP3A infections, yet suggests the virus replicated faster and was cleared from the host later. Of interest is to note that clinical recovery (by day 8) coincided with the development of virus neutralising antibodies in serum. However, cell associated viraemia persisted until day 14 in one pony and day 21 in the other two. Whether the resolution of viraemia was the result of an effector mechanism of cellular immunity or due to the natural senescence of infected leucocytes is not clear but is consistent with cytotoxic lymphocytes eventually clearing virus infected cells in blood and other tissues. The ponies recovered well from the infection and were the source of PBMC’s used as effectors in the cytotoxicity assays.

In this study, PBMC’s isolated from EAV convalescent ponies over a period between 4-9 months post-infection, presented a virus specific, genetically restricted cytolytic activity upon in vitro stimulation with EAV. The relative simplicity with which cytotoxic cells were induced by simply incubating the PBMC’s for 7 days in the presence of virus, has been documented before by Allen et al. (1995) to study EHV-1 cellular immunity in horses. It is worth noting however that, unlike Allen’s work, the cytotoxicity against EAV was not detectable with the use of cryopreserved mononuclear cells. This is probably due to the damage of EAV permissive subpopulations of cells during the freezing and / or thawing processes. In this regard, it should be noted that T lymphocytes, the major population sensitive to EHV-1 infection of PBMCs, survive the freezing/resuscitation process. Monocytes were found to be completely absent from the EAV induced cultures (7 days post-induction) in comparison to the mock induced PBMC’s and EAV N antigen expression was clearly detected in EAV-induced, freshly repared- PBMC’s 24 hours post-induction. Taken together, these observations suggest that EAV-infected monocytes may provide the major stimulus for lymphocyte activation which by day 7 days post-induction are cleared from the cultures as a result of virus infection or via cell-mediated cytolisis. This is consistent with the known tropism in vivo of EAV and other arteriviruses for monocytes.

The use of dermal cells as targets for cytotoxic assays has been documented in other species (Onishi et al., 1999; Sharpe et al., 2001: Korsoy et al. 2001; Flomemberg et al., 1996). In equine immunology studies the use of mitogen stimulated lymphoblasts (Allen et al., 1996: O’Neill et al. 1999) , primary kidney fibroblasts (McGuire et al., 1997) and primary dermal cells (Hannant, personal observations) have been previously used as targets in cytotoxicity assays. We have found the use of equine dermal cells as targets easy and convenient. They are more easily accesible than equine kidney cells and do not require previous incubation under special conditions, in contrast to mitogen stimulated lymphoblasts, they retain Chromium efficiently and are easily infectable with EAV. However, they have the disadvantage of not constitutively expressing MHC-II and therefore CD4+ related cytotoxicity is unlikely to be measurable with this targets.

13

Projecttitle



The cytotoxic activity exhibited by the in vitro stimulated mononuclear cells appeared to be genetically restricted. Determination of the equine leucocyte antigen of the ponies used in the study, which was not possible at the time these experiments were performed, would have helped to definitively corroborate the MHC-I restriction of the cytolytic activity. However, the fact that equine dermal cells only express detectable levels of MHC-I, the genetic restriction and memory of the cytotoxic response and the absence of monocytes from the EAV induced PBMC’s suggests that the cytotoxic effector cells are of the CD8+ phenotype.The cytotoxic responses reported here were assayed between 4 and 9 months post-infection and it is difficult to know if the features of such a response were similar during the first months after infection and what relevance this immune effector mechanism may have in the final clearance of virus following acute infection. In this regard it should be noted that cell associated viraemia was cleared 10 days after the last day of nasal shedding and 12 days after clinical recovery and the development of serum neutralising antibodies, thus the timing of this clearance is consistent with a cell-mediated response. Further immunological studies focused on the early events of EAV infection would help to determine at which point in time cytotoxic, cell-mediated immunity is stimulated and whether the development of this response correlates with the disappearance of viraemia.

Objective 3: To establish and validate a second generation RT-PCR test for detection of EAV in semen, using fluorescent probes to improve specificity, and a mimic system to improve quality control.

Methodology

a) EAV isolates

In order to assess the ability of the TaqMan® RT-PCR assay to detect a wide range of EAV isolates, a panel was assembled, representative of the diversity revealed by published phylogenetic analysis. These viruses which were propagated in RK13 cells using established procedures, varied in virulence, came from different geographical locations and had been isolated over a period of 30 years.

Clinical samples of heparinised blood (n = 9), nasal (n = 27) and faecal swabs (n = 14) for validation of the TaqMan® RT-PCR assay were obtained from eight ponies challenged with different biological clones of the Bucyrus strain of EAV (to be reported in a separate publication). A panel of EAV isolation-positive (n = 18) and negative semen samples (n = 15) were also tested by TaqMan® RT-PCR, selected from semen samples sent in for routine EAV diagnosis by virus isolation.

b) Isolation of RNA

RNA was extracted from 280 l volumes of RK-13 cell culture supernatants of the different virus strains, seminal plasma from centrifuged whole semen or heparinised whole blood using QIAamp viral RNA kits (Qiagen). In addition, nasal and faecal swabs collected from experimentally infected ponies were put into 10 ml and 2 ml of PBS respectively and were squeezed with sterile forceps and centrifuged at 400g. A sample of the resulting suspension (280 l) was taken for RNA extraction (Qiagen). The RNA was stored at –80°C until required, but was freeze–thawed several times in the course of testing the different PCR methods.

c) Evaluation of different RT-PCR and RT-nPCR methods

An initial evaluation of five existing RT-PCR and RT-nPCR methods was made. The relative sensitivity of tests was assessed using 10-fold dilutions of the Bucyrus strain of EAV (Bucyrus-CVL [7]), starting at a concentration of 105.5 TCID50/ml. RT and individual PCR steps were carried out in separate reaction tubes. First strand synthesis was carried out in 20 µl volumes in RT buffer (Promega) containing 2 µl of RNA, 25 pmol of random hexamers (Pharmacia), 0.5 mM of each dNTP, 20 U of RNAsin (Promega), and 100 U of M-MLV reverse transcriptase (Promega). Samples were incubated at 42°C for 30 min. prior to inactivating the RT-enzyme at 95°C for 5 min. Amplification of these cDNA preparations using the respective primers (25

14

Projecttitle



pmol) was undertaken in fresh tubes and 50 µl volumes, using 2.5 µl of the prepared cDNA in a ABgene master mix (Epsom UK) containing 1.25 units of Taq DNA polymerase, 75 mM Tris-HCl (pH 8.8), 20mM (NH4)2SO4, 2.5mM MgCl2, 0.01% (v/v) Tween 20, and 0.2 mM each of dATP, dCTP, dGTP and dTTP. One drop of mineral oil (Promega) was added to the 0.5 ml reaction tubes. For the nested PCRs, 2.5 l of the first round product was added to 45l of the ABgene master mix together with 25 pmol of the appropriate forward and reverse primers and made up to 50 µl with sterile distilled water. After the addition of mineral oil, the PCR assys were performed on a Biometra TRIO-Thermoblock (Whatman). PCR cycling conditions for the published methods [2, 5, 17] were as described. For the ORF-5 RT-nPCR (unpublished) the following cycling conditions were used: initial denaturation, 95C for 5 mins; followed by 32 cycles of template denaturation at 95C for 1 min; primer annealing at 55C for 1 min and extension at 72C for 1 min. The reaction was terminated by a single extension step of 72C for 7 min after which the samples were held at 4C. For the unpublished ORF7 PCR, the following cycling conditions were used: initial denaturation at 95C for 5 min; and then 40 cycles of the following reaction parameters: template denaturation at 95C for 20 s; primer annealing at 58C for 20 s and extension at 72C for 30 s. The reaction was terminated by a single extension step of 72C for 7 min and was finally held at 4C.

d) Construction of an artificial template (Mimic)

A 399 bp fragment of a strain of EAV (isolated from an Eastern European stallion), containing the proposed diagnostic ORF-7 PCR amplicon, was amplified using primers EAV10 and EAV12 modified at the 5' end to provide cleavage sites for EcoR I and Nsi I respectively, and subsequently cloned into pGEM-7f (+) (Promega). Plasmid DNA was purified using QIAprep spin minikit (Qiagen) and the presence of inserts of the appropriate size were confirmed by restriction enzyme digestion and by sequencing (Big Dye, Applied Biosystems).

The strategy for the design of the artificial template (A & B opposite) exploited the presence of a single Pst I site located within the region targeted by the EAV TaqMan® probe (12546-12569). After the linearisation with Pst I, the plasmid was incubated with calf alkaline phosphatase (Promega) to prevent recircularisation. A 295 bp fragment of human beta-actin gene (Promega) generated by PCR using primers with 5’-Pst I linkers: (5’-TTT CTG CAG TCA CCC ACA CTG TGC CCA TCT ACG A-3' and (5’-TTT CTG CAG CAG CGG AAC CGC TCA TTG CCA ATG G-3'), was ligated into the plasmid, and transfected into E.coli DH5. Purified plasmids (QIAprep spin minikit, Qiagen) were linearised with Nsi I and purified (Qiagen gel kit) for use as template for in-vitro transcription. RNA transcripts were prepared using Ambion T7 RNA polymerase kit (Megascript) according to manufacturers instructions, treated with DNAse I and stored at –70°C for use as an artificial template (mimic) in the single tube RT-PCR. Successful amplification of this mimic in the samples was detected using a TaqMan® probe containing a different fluorochrome to that used for detection of the viral amplicon (5'-VIC GCC ACG TCC AGA CAC AGG ATG GCA -3'-TAMRA). The appropriate dilution of mimic RNA used in the RT-PCR TaqMan® assay was determined by serial dilution so that RT-PCR TaqMan® assay sensitivity was not affected.

e) TaqMan® assay method

A TaqMan® probe (5'-FAM-TGG TTC ACT CAC TGC AGA TGC CGG-3'-TAMRA) was designed based on an alignment of 50 EAV sequences available in GenBank for the proposed amplicon in ORF7. This probe spans the single Pst I site exploited during the construction of the mimic. Preliminary experiments indicated that TaqMan® signal would be compromised using 2.5 mM MgCl2. In order to increase the magnesium ion

15

PGEM 7Zf

Pst I

Human -actin (295 bp)

T7Promoter

Nsi IEcoRI

EAV10T7Pst I

EAV12Mimic probe - VIC

MIMIC RNA TRANSCRIPT

WT EAV = 399 bp

Mimic = 694 bp

EAV-TaqMan® probe - FAM EAV10

Pst IEAV12

Pst I Nsi I

EAV RNA TRANSCRIPTA

B

Projecttitle



concentration for optimal TaqMan® assay signal, the Abgene mastermix was replaced with 10 X reaction buffer (Promega).

A single-tube RT-PCR assay was performed containing 50 mM KCl, 10 mM Tris-HCl pH 9.0, 5 mM MgCl2, 0.4 mM of each dNTPs, and 5 U Taq DNA polymerase (Promega). To the master-mix was added 20 U RNAsin, 100 U MMLV-reverse transcriptase (Promega), 25 pmol of EAV10 (5'-GTA CAC CGC AGT TGG TAA CA-3') and EAV12 (5'-ACT TCA ACA TGA CGC CAC AC-3'), 5 pmol EAV probe (FAM reporter), 5 pmol mimic probe (VIC reporter). HPLC water was added to make the final volume up to 50 µl. The mastermix containing the buffers, probes and mimic RNA were aliquoted into a 96 well plate (MicroAmp Perkin-Elmer) and then the sample RNA was added last in a separate work area. In addition to wells containing RNA from test samples, control wells containing only mimic template were also setup between each sample well, substituting HPLC water for the sample RNA. Further control wells without either sample or mimic RNA templates were also set up. After sealing the plate with optical caps (MicroAmp, Perkin Elmer), the plate was placed in a 9700 GeneAMP PCR thermal cycler using the following cycling conditions, (42C, 30 min, inactivation at 95C for 5 min, then 40 cycles of the following reaction parameters: template denaturation at 95C for 20 s, primer annealing at 60C for 60 s and extension at 72C for 30 s. The reaction was terminated by a single extension step of 72C for 7 min and finally held at 4C. The TaqMan® reactions were read on an ABI PRISM 7200 sequence detector (Applied Biosystems).

f) Statistical analysis

Statistical analysis was performed on a scatter graph of the FAM and VIC signals using Statistica (ver 6, Statsoft) in order to define 99.9% confidence ellipses around the water control and no template control wells. These confidence ellipses were used to define EAV positive samples, EAV negative samples and reaction failures. EAV negative samples were grouped with the water control wells, while EAV positive samples were identified as points that were outside of the water control 99% confidence ellipse. Sample failures were grouped with readings from no template controls.

g) Virus detection by virus isolation (VI)

Virus titres within semen and nasal swabs were determined by virus isolation (VI), essentially as described previously [10]. Briefly, in the case of semen samples, an initial 1:10 dilution of sample, starting with 0.5 ml was made in Minimal Essential Medium (MEM) and then doubling dilutions made up to 1:3560 and 0.5 ml inoculated into 25 cm2 flasks of confluent monolayers of RK-13 cells. Flasks were incubated incubated for 1 h at 37C in 5% CO2 and overlaid with 10 ml MEM supplemented with 2% bovine serum. The cells were cultured for a minimum of 7 days at 37°C, before being passaged twice more by trypsinisation (split ratio 1:3). Half the cells from the third passage were also set out in 96-well microtitre plates. The cells were cultured for 3 days at 37°C prior to fixation and antigen detection with either an EAV polyclonal antibody (LS 7005) or EAV-specific monoclonal antibody (WB20 [22]) in conjunction with an indirect immunoperoxidase staining method [10]. In the case of nasal swabs, 150 l of nasal swab eluate was titrated (0.5 log10) in quadruplicate in 96-well plates. Then, 100 l of EMEM containing 3 x 105 RK-13 cells and 10% FBS was added to each well and plates incubated for 3 days at 37 C in 5% CO2. Wells showing CPE were recorded as positive and virus titres were calculated according to the Karber formula and expressed as TCID50/ml of swab extract. Heparinised blood and faecal swabs were only diluted 1:10 for cell culture, but otherwise treated in the same way as for the semen and nasal swabs. Amplification of the RNA isolated from the clinical samples was carried out using the single tube RT-PCR TaqMan® method as described above.

To assess the accuracy of the TaqMan® assay in differentiating EAV from other arteriviruses, a series of 10-fold dilutions of viral RNA were carried out on the isolate H2 (Humberside) of porcine reproductive and respiratory syndrome virus (PRRSV) [6]. In addition, DNA (Qiagen) from equine herpesvirus-1 (RACH strain) and a field isolate of equine herpesvirus-2 were tested by the Taqman® assay.

16

Projecttitle



RESULTS

a) Evaluation of different RT-PCR and RT-nPCR methods

The same buffer system was employed with all 5 different RT-PCR and RT-nPCR methods, in order to determine the set of primers that give optimum sensitivity with a laboratory strain of EAV. Interestingly, an un-nested RT-PCR (using primers EAV 10 and 12) had similar analytical sensitivity as the 4 nested RT-PCRs. This method targeting ORF7, the most homogeneous and stable region of EAV was selected as the diagnostic amplicon. In addition, the ORF-5 (GL ) RT-nPCR which was also included in this comparative exercise also detected all the diverse strains of EAV (data not shown).

b) TaqMan® assay optimisation

Amplification of the mimic template was inhibited when EAV RNA was present (shown below). This was not unexpected, since amplification of the smaller EAV amplicon will be more favourable. At a dilution of 10 -5 of mimic template, there was no detectable competition of EAV template. This concentration of mimic RNA template was added to all future TaqMan® assay reactions as an internal control.

After one step RT-PCR, all 28 RNA samples representing North American and European strains of EAV gave the expected 399-bp fragment in the agarose gels. Preliminary experiments with these primers, in the absence of the TaqMan® probe but using 2.5 mM MgCl2

gave sharp well defined bands upon agarose gel examination, but produced poor results in the TaqMan® assay. It was found that by increasing the concentration of MgCl2 (5mM), the intensity of reporter fluorescence signal was optimised. This protocol was followed for all subsequent TaqMan® assays.

The detection sensitivity of the TaqMan® RT-PCR was compared with gel-based visualisation of PCR product and VI. Samples used were serial log10 dilutions of tissue culture derived EAV (Bucyrus strain) and EAV positive semen titrated in negative semen. Using these samples, there was good agreement between the TaqMan® PCR and the visualisation of PCR product in a gel. Furthermore, EAV detection by VI also had similar detection limits. The TaqMan® assay performed using EAV specific primers gave negative results with template extracted from equine herpesvirus-1 and 2, and PRRSV.

c) Demonstration of EAV in clinical samples

Selected semen samples that had been received at the laboratory for VI in the last 10 years were also tested by the single tube TaqMan® assay. Of the 33 semen samples tested, 18 were positive by the RT-PCR TaqMan® assay (opposite), three more than by VI. All of the nasal swabs, faecal swabs and blood samples (50 samples in all) taken from infected ponies were both positive by VI (mostly on first passage) and by single-tube RT-PCR TaqMan® assay, (data not shown).

DISCUSSION

17

-1000

-800

-600

-400

-200

0

200

400

600

0 2000 4000 6000 8000 10000

399 bp694 bp

VI

PCR

TaqMan®(see graph below)

+ + + - - - -+ + + - - - -

10-1 10-2 10-3 10-4 10-5 10-6 NegDilution of EAV

10 20 40 80 320 640 Neg160Dilution of Semen (reciprocal)

A B

+ + + + + - - -+ + + + + - - -

-50

-45

-40

-35

-30

-25

-20

-15

-10

-5

0

-600

-400

-200

0

200

400

600

800

0 2000 4000 6000 8000

EAV [FAM] EAV [FAM]

Mimic[VIC]

1000 2000 3000 4000 5000

EAV [FAM]

-500

-400

-300

-200

-100

0

100

200

300

Mim

ic [V

IC]

Projecttitle



The method developed here does not require time-consuming electrophoresis and/or hybridization to visualise and confirm specificity. Furthermore, since the PCR plates never had to be opened after adding the reaction components, the risk of contamination was reduced. RNA samples prepared on different occasions from replicate cell cultures and clinical samples and gave very similar TaqMan® results, suggesting good reproducibility of the assay. Negative control samples were routinely interspersed between the samples tested, demonstrating that cross-contamination had been avoided. The method outlined in this report also utilizes an artificial RNA template (mimic) in order to validate negative results. Successful amplification of the mimic was detected by a second reporter dye (VIC). Although there is considerable spectral overlap between the emission spectra of the two reporter dyes used (FAM emax = 527 nm and VIC emax = 558 nm), the use of 2D scatter-plots was able to discriminate the signals due to the respective dyes.

All 28 EAV isolates tested were successfully detected by the TaqMan® assay. The TaqMan® probe recognition site matched all 50 EAV GenBank accessions except 5 sequences which had single nucleotide substitutions. The sense primer (EAV 10) mismatched at one nucleotide position with 10 European strains of EAV and at two and three nucleotide positions with 13 and 3 US strains, respectively. The EAV 12 anti-sense primer was located outside ORF-7, but matched at all the EAV nucleotide positions where sequence has been determined.

A further development of this TaqMan® assay could exploit the use of recently developed robotic extraction protocols to automate the RNA extraction and aliquoting procedures into a 96-well format. This approach would permit a greater throughput to be achieved, and may further reduce the risk of cross-contamination. In addition, the quantification of EAV genome in semen and clinical samples could be achieved by using quantitative RT-PCR methods.

The advantages of the TaqMan PCR assay are that it is often shown to be more sensitive than virus isolation, and can detect virus in degraded or toxic samples. This is a particular advantage for examination of semen, which is renowned for the difficulties it causes in conventional isolation. The single tube format, applied in this assay, decreases carry-over contamination and its ability to be standardized and quantitated overcome difficulties present in current PCR techniques. In addition, TaqMan is very rapid (1 day for results versus up to three weeks for VI). and capable of high throughput which increase efficiency and decrease cost when compared to the standard test of virus isolation. Rapid, sensitive and accurate detection of positive semen and clinical samples can be used to control this economically important disease for the horse industry.

Publications accruing from this work (VLA staff in bold):

Castillo-Olivares, J., de Vries, AAF., Raamsman, MJB., Rottier, PJM., Lakhani, K, Westcott. D., Tearle. JP., Wood, JLN., Mumford, JA., Hannant, D. and Davis-Poynter, NJ. (2001) Vaccination of ponies with the entire GL ectodomain of Equine Arteritis Virus (EAV) provides protection against infection. Journal of General Virology 82; 2425-2435.

18

Projecttitle



Castillo-Olivares, J. Tearle, JP., Montesso, F., Westcott, D., Paton, D., Davis-Poynter NJ., and Hannant, D. (2002) Equine Arteritis Virus Infection in ponies induces a virus specific and genetically restricted Cytotoxic Lymphocyte Response. (In preparation)

Edwards, S., Castillo-Olivares, J., Cullinane, A., Lanle, J., Lenihan, P., Mumford, J. A., Paton, D. J., Pearson, J. E., Sinclair, R., Westcott, D. G. F., Wood, J. L. N., Zientara, S., and Nelly, M. (1999). International harmonisation of laboratory diagnostic tests for equine viral arteritis. Equine Infectious Diseases VIII, 359-362.

Westcott, DG., Hannant, D. Baule, C., Tearle, J., Mittelholzer, C., Drew, TW, Castillo-Olivares, J., Belák, S. & Paton, DJ. Differences in pathogenicity between closely related biological clones of equine arteritis virus (In preparation)

Westcott, DG., King, DP, Drew, TW., Nowotny, N.,Belak, S. & Paton, D. (2002) Use of an internal standard in a closed one tube RT-PCR for the detection of equine arteritis virus RNA with fluorescent probes (In preparation)

Westcott D, Castillo-Olivares J, de Vries AAF, Raamsman MJB, Rottier PJM, Lakhani K, Tearle JP, Wood JLN, Mumford JA, Hannant D, and Davis-Poynter NJ. (2001). Vaccination of Ponies with the entire GL ectodomain of Equine Arteritis Virus (EAV) provides protection against infection. Journal of General Virology 82, 2425-35.

Presentations

Westcott D, Paton D, Hannant D, and Tearle J. (2001) Pathogenesis of Plaque variants of Equine Arteritis Virus (EAV). Association of Veterinary Teachers and Research Workers. April 2001.

Westcott, D. G., Hannant, D., Tearle, J., Mittelholzer, C., Baule, C., Drew, T. W., Belak, S., and Paton, D. J. (2001) Differences in pathogenicity between closely related biological clones of equine arteritis virus. Sixth International Symposium on Positive strand RNA viruses. P1-79. 2001. Paris, France, Pasteur Institute.

Westcott, D. G., King DP, Drew, T. W., Paton, D. J., and Nowotny, N. (2002) Closed one tube reverse transcriptase polymerase chain reaction for the detection of equine arteritis virus RNA with fluorescent probes. Association of Veterinary Teachers and Research Workers. March 2002.

19

Projecttitle



ReferencesAllen G, Yeargan M, Costa LR, Cross R (1995).Major histocompatibility complex class I-restricted cytotoxic T-lymphocyte

responses in horses infected with equine herpesvirus 1. Journal of Virology 69(1):606-12.Allende, R., Kutish, G. F., Laegreid, W., Lu, Z., Lewis, T. L., Rock, D. L., Friesen, J., Galeota, J. A., Doster, A. R. and Osorio, F. A.

(2000) Mutations in the genome of porcine reproductive and respiratory syndrome virus responsible for the attenuation phenotype. Archives of Virology 145, 1149-1161.

Balasuriya, U. B. R., MacLachlan, N. J., de Vries, A. A. F., Rossitto, P. V. & Rottier, P. J. M. (1995). Identification of a neutralization site in the major envelope glycoprotein (GL) of equine arteritis virus. Virology 207, 518-527.

Balasuriya, U. B. R., Patton, J. F., Rossitto, P. V., Timoney, P. J., McCollum, W. H. & MacLachlan, N. J. (1997). Neutralization determinants of laboratory strains and field isolates of equine arteritis virus: identification of four neutralization sites in the amino-terminal ectodomain of the GL envelope glycoprotein. Virology 232, 114-128.

Balasuriya, U. B. R., Rossitto, P. V., DeMaula, C. D. & MacLachlan, N. J. (1993). A 29K glycoprotein of equine arteritis virus expresses neutralization determinants recognized by murine monoclonal antibodies. Journal of General Virology 74, 2525-2529.

Balasuriya, U. B., Timoney, P. J., McCollum, W. H., and MacLachlan, N. J. (1995) Phylogenetic analysis of open reading frame 5 of field isolates of equine arteritis virus and identification of conserved and nonconserved regions in the GL envelope glycoprotein. Virology 214, 690-7.

Byrnes AP and Griffin DE. (2000) Large-plaque mutants of Sindbis virus show reduced binding to heparan sulfate, heightened viremia, and slower clearance from the circulation. Journal of Virology 74, 644-651.

Castillo-Olivares, J., de Vries, AAF., Raamsman, MJB., Rottier, PJM., Lakhani, K, Westcott. D., Tearle. JP., Wood, JLN., Mumford, JA., Hannant, D. and Davis-Poynter, NJ. (2001) Vaccination of ponies with the entire GL ectodomain of Equine Arteritis Virus (EAV) provides protection against infection. Journal of General Virology 82; 2425-2435.

Chinsangaram J, Piccone ME, and Grubman MJ. (1999) Ability of foot-and-mouth disease virus to form plaques in cell culture is associated with suppression of alpha/beta interferon . Journal of Virology 73, 9891-9898.

Chirnside, E. D., de Vries, A. A., Mumford, J. A. & Rottier, P. J. (1995a). Equine arteritis virus-neutralizing antibody in the horse is induced by a determinant on the large envelope glycoprotein GL. Journal of General Virology 76, 1989-1998.

Chirnside, E. D., Francis, P. M. & Mumford, J. A. (1995b). Expression cloning and antigenic analysis of the nucleocapsid protein of equine arteritis virus. Virus Research 39, 277-288.

de Vries, A. A., Chirnside, E. D., Horzinek, M. C. & Rottier, P. J. (1992). Structural proteins of equine arteritis virus. Journal of Virology 66, 6294-6303.

den Boon, J. A., Snijder, E. J., Chirnside, E. D., de Vries, A. A. F., Horzinek, M. C. & Spaan, W. J. M. (1991). Equine arteritis virus is not a togavirus but belongs to the coronaviruslike superfamily. Journal of Virology 65, 2910-2920.

Deregt, D., de Vries, A. A., Raamsman, M. J., Elmgren, L. D. & Rottier, P. J. (1994). Monoclonal antibodies to equine arteritis virus proteins identify the GL protein as a target for virus neutralization. Journal of General Virology 75, 2439-2444.

Doll, E. R., Bryans, J. T., Wilson, J. C. and McCollum, W. H. M. (1968) Immunization against equine viral arteritis using modified live virus propagated in cell cultures of rabbit kidney. Cornell Veterinarian 58, 497-524 .

Edwards, S. C.-O. J. C. A. L. J. L. J. M. J. A. P. D. J. P. J. E. S. R. W. D. G. F. W. J. L. N. Z. S. a. N. M. (1999) International harmonisation of laboratory diagnostic tests for equine viral arteritis. In: Proceedings of the 8th International Conference on Equine Infectious Diseases. Dubai, UAE , Eds: U. Wernery, J. F. Wade. Edited by J. A. Mumford and O. R. Kaaden. Newmarket , R &W Publications pp 359-362.

Edwards, S., Castillo-Olivares, J., Cullinane A., Lable, J., Lenihan, J., Mumford, J. A., Paton, D. J., Pearson, J. E., Sinclair, R., Westcott, D. G. F., Wood, J. L. N., Zientara S. and Nelly, M. International harmonisation of laboratory diagnostic tests for equine viral arteritis. In Proceedings of the 8th International Conference on Equine Infectious Diseases, Dubai, 1998, pp.359- 362. Edited by U. Wernery, J. F. Wade, J. A. Mumford and O.-R. Kaaden. Newmarket 1999: R &W Publications

Flomenberg P, Piaskowski V, Truitt RL, Casper JT (1996). Human adenovirus-specific CD8+ T-cell responses are not inhibited by E3-19K in the presence of gamma interferon. Journal of Virology 70(9):6314-22.Fukunaga, Y., Imagawa, H., Tabuchi, E. & Akiyama, Y. (1981). Clinical and virological findings on experimental equine viral arteritis in horses. Bulletin of Equine Research Institute, Japan, 18, 110-118.

Gerber, H., Steck, F., Hofer, B., Walther, L., Friedli, U. (1978). Serological investigations on equine viral arteritis. In Proceedings of the 4th International Conference on Equine Infectious Diseases, Lyon 1976, pp. 461-465. Edited by B. T. Bryans and H. Gerber. Princeton: Veterinary Publications.

Hedges, J. F., Balasuriya, U. B. & MacLachlan, N. J. (1999). The open reading frame 3 of equine arteritis virus encodes an immunogenic glycosylated, integral membrane protein. Virology 264, 92-8.

Hyllseth, B. (1969) A Plaque Assay of Equine Arteritis Virus in BHK-21 cells. Archiv fur die gesamte virusforschung, 28, 26-33.Iwashita, O. & Harasawa, R. (1987). Translation of equine arteritis virus RNA in rabbit reticulocyte lysates. Nippon Juigaku Zasshi

49, 387-389.Kheyar, A., Martin, S., St.-Laurent, G., Timoney, P. J., McCollum, W. H. & Archambault, D. (1997). Expression cloning and humoral

immune response to the nucleocapsid and membrane proteins of equine arteritis virus. Clinical and Diagnostic Laboratory Immunology 4, 648-652.

20

Projecttitle



Koksoy S, Phipps, A. J., Hayes, K. A. and Mathes L. E. (2001). SV40 Immortalization of feline fibroblasts as targets for MHC-restricted cytotoxic T-cell assays.Vet Immunol Immunopathol 79(3-4):285-95.

MacLachlan, N. J., Balasuriya, U. B. R., Hedges, J. F., Schweidler, T. M., McCollum, W. H., Timoney, P. J., Hullinger, P. J. & Patton, J. F. (1998). Serological response of horses to the structural proteins of equine arteritis virus. Journal of Veterinary Diagnostic Investigation 10, 229-236.

McCollum, W. H. (1969) Development of a modified virus strain and vaccine for equine viral arteritis. Journal of the American Veterinary Medical Association 155, 318-322 .

McCollum, W. H. (1969). Development of a modified virus strain and vaccine for equine viral arteritis. Journal of the American Veterinary Medical Association 155, 318-322.

McCollum, W. H. (1976). Studies of passive immunity in foals to equine viral arteritis. Veterinary Microbiology 1, 45-54.McCollum, W. H. and Timoney, P. J. (1999) Experimental observations on the virulence of isolates of equine arteritis virus. In:

Proc. 8th Int. Conference on Equine Infectious Diseases. Dubai, UAE, pp 558-560.McCollum, W. H., Bryans, J. T., Bryans, J. T. & Gerber, H. (1973). Serological identification of infection by equine arteritis virus in

horses of several countries. In Proceedings of the 3rd International Conference on Equine infectious diseases III, Paris, 1972, pp.256-263. Edited by J. T. Bryans and H. Gerber. Basel: S. Karger

McCue, P. M., Hietala, S. K., Spensley, M. S., Stillian, M., Mihalyi, J. & Hughes, J. P. (1991). Prevalence of equine viral arteritis in California horses. California Veterinarian 45, 24-26.

McGuire TC, Zhang W, Hines MT, Henney PJ, Byrne KM.Frequency of memory cytotoxic T lymphocytes to equine infectious anemia virus proteins in blood from carrier horses.Virology. 238(1):85-93.

McKinnon, AO, Colbern, GT, Collins, JK, Bowen, RA, Voss, JL and Umphenour, NW. (1986) Vaccination of stallions with modified live equine viral arteritis virus. Journal of Equine Veterinary Science 6, 66-69.

Ohishi K, Inui K, Yamanouchi K, Barrett T. (1999).Cell-mediated immune responses in cattle vaccinated with a vaccinia virus recombinant expressing the nucleocapsid protein of rinderpest virus. Journal of General Virology 80, 1627-1634

O'Neill T, Kydd JH, Allen GP, Wattrang E, Mumford JA, Hannant D. (1999). Determination of equid herpesvirus 1-specific, CD8+, cytotoxic T lymphocyte precursor frequencies in ponies.Vet Immunol Immunopathol. 70(1-2):43-54.

Park, B. K., Yoon, I. J., and Joo, H. S. (1996) Pathogenesis of plaque variants of porcine reproductive and respiratory syndrome virus in pregnant sows. Am J Vet Res 57, 320-3.

Paweska, J. T., Aitchison, H., Chirnside, E. D. & Barnard, B. J. (1996). Transmission of the South African asinine strain of equine arteritis virus (EAV) among horses and between donkeys and horses. Onderstepoort Journal of Veterinary Research 63, 189-196.

Sano, Y., Inaba, Y., Uwatoko, K., Kubota, T., Asagoe, T., Kanaya, J., Pan, I. J., Akashi, H., and Fukunaga, Y. (1998) Effect of heparin on hemagglutination by equine arteritis virus. J Vet Med Sci 60, 447-50.

Senne, D. A., Pearson, J. E. & Carbrey, E. A. (1985). Equine viral arteritis: a standard procedure for the virus neutralization test and comparison of results of a proficiency test performed at five laboratories. Proceedings of the United States Animal Health Association 89, 29-34.

Sharpe S, Beer B, Hall G, Dennis M, Norley S and Cranage M.Analysis of SIV-specific CTL in the rhesus macaque model of AIDS: the use of simian fibroblasts as an alternative source of target cells for chromium release assays (2001). J Immunol Methods

Snijder, E. J. & Meulenberg, J. J. M. (1998). The molecular biology of arteritiviruses. Journal of General Virology 79, 961-979.Snijder, E. J., Wassenaar, A. L., van Dinten, L. C., Spaan, W. J., and Gorbalenya, A. E. (1996) The arterivirus nsp4 protease is the

prototype of a novel group of chymotrypsin-like enzymes, the 3C-like serine proteases. J Biol Chem 271, 4864-71.Snijder, E.J., van Tol, H., Pedersen, K.W., Raamsman, M.J. and de Vries, A.A. (1999) Identification of a novel structural protein of

arteriviruses. Journal of Virology 73, 6335-45.Stadejek, T., Bjorklund, H., Bascunana, C. R., Ciabatti, I. M., Scicluna, M. T., Amaddeo, D., McCollum, W. H., Autorino, G. L.,

Timoney, P. J., Paton, D. J., Klingeborn, B. & Belak, S. (1999). Genetic diversity of equine arteritis virus. J Gen Virol 80, 691-9.Steel, R. G. D. and Torrie, J. H. (1980). In Principles and Procedures of Statistics, 2nd Edition. McGraw-Hill. New York.Timoney, P. J. & McCollom, W. H. (1993). Equine viral arteritis in perspective in relation to international trade. Journal of Equine

Veterinary Science 13, 50-52.Timoney, P. J. & McCollum, W. H. (1988). Equine viral arteritis-epidemiology and control. Journal of Equine Veterinary Science 8,

54-59.Timoney, P. J. & McCollum, W. H. (1987) Equine viral arteritis. Canadian Veterinary Journal 28, 693-695.Timoney, P. J. & McCollum, W. H. (1990). Equine viral arteritis: current clinical and economic significance. In Proceedings of the

American Association of Equine Practitioners 36, pp.403-440. Timoney, P. J., Umphenour, N. W., McCollum, W. H. & Powell, D. G. (1988). Safety evaluation of a commercial modified live

equine arteritis virus vaccine for use in stallions. In Proceedings of the 5th International Conference on Equine Infectious Diseases, Lexington, 1987, pp. 19-27. Edited by D. G. Powell. Lexington: The University Press of Kentucky, Lexington, United States of America.

van Dinten, L. C., Rensen, S., Gorbalenya, A. E., and Snijder, E. J. (1999) Proteolytic processing of the open reading frame 1b-

21

Projecttitle



encoded part of arterivirus replicase is mediated by nsp4 serine protease and Is essential for virus replication. J Virol 73, 2027-37.

Wood, J. L., Chirnside, E. D., Mumford, J. A. & Higgins, A. J. (1995). First recorded outbreak of equine viral arteritis in the United Kingdom. Veterinary Record 136, 381-385.

Yuan, S., Mickelson, D., Murtaugh, M. P., and Faaberg, K. S. (2001) Complete genome comparison of porcine reproductive and respiratory syndrome virus parental and attenuated strains. Virus Res 74, 99-110.

Ziebuhr, J., Snijder, E. J. and Gorbalenya, A. E. (2000) Virus-encoded proteinases and proteolytic processing in the Nidovirales. Journal of General Virology 81, 853-879.

Please press enter

22

research and development - department for...

Documents