key: cord-0004952-hds1j5ln
authors: Mardassi, H.; Mounir, S.; Dea, S.
title: Molecular analysis of the ORFs 3 to 7 of porcine reproductive and respiratory syndrome virus, Québec reference strain
date: 1995
journal: Arch Virol
DOI: 10.1007/bf01322667
sha: e876958c75efe0d52091fb2e744275eb55f0eeb0
doc_id: 4952
cord_uid: hds1j5ln

The cDNA sequence of the 3′-terminal genomic region of the Québec IAF-exp91 strain of porcine reproductive and respiratory syndrome virus (PRRSV) was determined and compared to those of other reference strains from Europe (Lelystad virus) and US (ATCC VR2385, MN-1b). The sequence (2834 nucleotides) which encompassed ORFs 3 to 7 revealed extensive genomic variations between the Québec strain and Lelystad virus (LV), resulting from high number of base substitutions, additions and deletions. The ORFs 5, 3, and 7 seemed to be relatively the most variable; the predicted encoding products of the Québec and LV strains displayed only 52%, 54%, and 59% amino acid identities, respectively. Nevertheless, in vitro translation experiments of the structural genes (ORFs 5, 6, and 7) and radio-immunoprecipitation assays with extracellular virions gave results similar to those previously reported for LV. In contrast, close genomic relationships were demonstrated between Québec and US strains. Taking together, these results indicate that, although structurally similar, North American PRRSV strains belong to a genotype distinct from that of the LV, thus supporting previous findings that allowed to divide PRRSV isolates into two antigenic subgroups (U.S. and European).

Porcine reproductive and respiratory syndrome virus (PRRSV) is a small enveloped RNA virus which has been found to be the causative agent of a new pig disease that occurs in North America and Europe [1, 5, 7, 34] . The disease is characterized by severe reproductive failure in sows and respiratory problems affecting pigs of all ages [12] . The European prototype strain of PRRSV, known as the Lelystad virus (LV), shows close similarities with the Arteriviruses from morphological, * The nucleotide sequence data reported in this paper will appear in the EMBL and GenBank nucleotide sequence databases under accession number L40898. biochemical and molecular aspects [2 t, 33] . This new group of viruses inctude equine arteritis virus (EAV), lactate dehydrogenase-elevating virus (LDV) and simian hemorrhagic fever virus (SHFV) [26] . The viral genome is a positive single-stranded polyadenylated RNA molecule of about 15 kb in length that contains eight open reading frames (ORFs) similarly organized as EAV [6, 8, 21] . The transcription mechanism of the genomic RNA has similarities with that of Coronaviruses, since a 3'-coterminal nested set of six major subgenomic mRNAs has been identified in PRRSV-infected cells [6, 20--22] . The virions contain a nucleocapsid protein of 15 kDa (N), an unglycosylated membrane protein of 18 kDa (M), and a glycosylated membrane protein of 25 kDa (E); it was recently shown that these proteins are encoded by ORFs 7, 6, and 5, respectively [23] .

So far, little is knov~aa about the origin and evolution of PRRSV. Although the clinical syndromes associated with PRRSV infection are similar in North America and Europe [12] , antigenic variations have been reported between US and European isolates and among US isolates [12, 25] . Preliminary sequence analysis of North American isolates revealed high genomic variations with European strains [18, 19] . This is strongly supported by serological findings. To further contribute to the understanding of the genetic basis for antigenic diversity between PRRSV isolates, we determined the cDNA sequence of the 3'-terminal genomic region of the Qurbec reference strain IAF-exp91. Our prime objective was to compare the genomic sequence o f the Qurbec strain to those of other reference strains from Europe and US. Since important amino acid changes were identified between North American strains and LV, we also analysed the proteins specified by a cell culture-adapted Qurbec PRRSV isolate (IAF-Klop) and demonstrated that these viruses remain structurally similar.

The origin and propagation in porcine alveolar macrophages (PAM) of the Qurbec IAF-exp91 and IAF-Klop strains of PRRSV have been previously described [7, 17] . The latter was also adapted to grow in MARC-145 cells, a highly permissive cell clone to PRRSV derived from the MA104 monkey kidney cell fine [14] . The cell line was kindly provided to us by J. Kwang (U.S. Meat Animal Research Center, USDA, ARS, Clay Center, Nebraska). The IAF-Klop strain was plaquepurified twice and yielded titers of 105-106 TCIDs0/ml after five successive passages in MARC-145 cells.

Hyperimmune sera were prepared against IAF-exp91 and IAF-Klop PRRSV strains in New Zealand Albino rabbits and specific-pathogen-free (SPF) piglets, respectively [7, 17] . The porcine hyperimmune sera had only weak neutralizing activity (VN titers of 1:32) towards IAF-Klop strain, but reached titers of 1:2560-1:5120 by indirect immunofluorescence staining. The monoctonal antibodies (MAbs) SDOW 17, VO 17 and EP147, directed against the nucleocapsid protein of the ATCC-VR2332 isolate of PRRSV [25] were kindly provided by D.A. Benfield and E. Nelson (South Dakota State University).

Supernatant fluids of infected cells were clarified and extracellular virions were concentrated by ultracentrifugation, as previously described [18] . Genomic RNA of both Qu6bec PRRSV strains was extracted from 50 ~1 aliquots by the guanidinium isothiocyanate-acid phenol method described by Chomczinsky and Sacchi [4] . Cloning of IAF-exp91 3"end genomic fragments was obtained by setting up a cDNA library, as previously reported [18] .

Sequencing of cDNA clones was performed on both strands by the dideoxynucle0tide chain termination method [28] using T7 DNA polymerase (Pharmacia). The nucleotide and amino acid sequences were analysed with the aid of the Geneworks 2.2 (IntelliGenetics) sequence analysis prosxam. All comparisons were performed with a k-tupple length of one, and costs to open and to lengthen a gap of 5 and 25, respectively.

Enzymatic amplification of ORFs 5 and 6 of the IAF-Klop strain was achieved by RT-PCR [19] . Oligonucleotide primers were designed according to the 3 "end sequence of the IAF-exp91 strain. Two restriction sites for EcoRI (sens primers) and BamHI (antisens primers) were added at the 5' end of each primer for directional cloning. Sequence and position of oligonucleotide primers were the following: The amplified products of ORFs 5 and 6 were agarose-gel purified, digested with EcoRI and BamHI, and finally ligated into a similarly treated pBluescript SK+ plasmid vector (Stratagene) [27] . Enzymatic amplification of ORF 7 was achieved using primer pair 1008PS/1009PR [18] to which EcoRI restriction sites were added at the 5' end of both primers to allow cloning into pBS SK+ vector. The resulting recombinant plasmids pBS5, pBS6 and pUC7 contained the coding sequences of ORFs 5, 6 and 7, respectively, under control of the T7 RNA polymerase promoter. After being linearized by digestion with the appropriate endonuclease (BamHI or EcoRI), recombinant ptasmids were transcribed in vitro using T7 RNA polymerase (Ambion Inc). Capping of transcripts was achieved by adding 0.75 mM of m7Gppp (Pharmacia LKB) to the reaction mixture. Two gl of each transcript were in vitro translated using wheat germ extract (Promega) in the presence of 50 gCi of trans-labeled [35S]methionine (Sp. act.> 1200 Ci/mM, ICN Biochemical). For post-translation modification studies, in vitro translation experiments were also done using lysates of rabbit reticulocytes (Promega) in the presence of canine microsomal membranes (Promega).

Confluent monolayers of MARC-145 cells were infected with IAF-Klop strain at a MOI of 0.1 TCIDs0/cell. At 36h post-infection, cells were washed with PBS, starved for 2h in methionine-free medium, then reincubated for 6 h in culture medium containing 50 ~Cilml of [35S]methionine. Alternatively, the labeling period was extended until maximum cytopathic effect was achieved, and extracellular virus was prepared as previously described [17] . Proteins were immunoprecipitated from infected cell lysates or extracellular virions, then analyzed by electrophoresis in 12 or 15% SDS-polyacrylamide gels, as previously described [17] . In one experiment, immune complexes were treated with 200 mU ofendoglycosidase F/N-glycosidase F (glyco F) (Boehringer Mannheim) as performed by De Vries et al. [9] . In case of the translated products ofORFs 5 to 7, they were either directly analyzed by electrophoresis or after immunoprecipitation.

The sequence of the 3'-terminal portion of the genome of Qu6bec reference strain IAF-exp91 of PRRSV was obtained from 8 viral specific clones having in common their 3' ends. Sequence data obtained from these clones were assembled and a sequence consisting of 2834 nucleotides (nt) was derived. For any given region, at least two distinct clones were sequenced on both strands, except for the 5' most 634 nucleotides which were determined from only the longest clone where no base mismatches between the two strands were observed. The nucleotide and deduced amino acid (aa) sequences determined for the Qu6bec IAF-exp91 strain of PRRSV have been submitted to EMBL Data Library and have been assigned accession no. IA0898.

This genomic region contains five ORFs corresponding to ORFs 3 to 7 of the LV and extends to the poly (A) tail. All the ORFs overlapped, except ORFs 4 and 5. These ORFs each encoded a polypeptide with predicted sizes of29K, 19.6K, 22.4K, 19.1K and 13.6K, respectively (Table 1) . Sequence analyses revealed a high degree of genomic variation between Qu4bec IAF-exp91 and LV strains. A high rate of base substitutions, additions and deletions, randomly distributed along the nucleotide sequence was noted.

The percentages of aa identity between predicted encoding products of ORFs 3 to 7 of the IAF-exp91 strain and those of the LV are presented in Table 2 . The result indicate that ORFs 5, 3, and 7 are the most variable with only 52%, 54%, and 59% aa identities, respectively, when compared to LV. In contrast, ORFs 4 and 6 of the IAF-exp91 strain were relatively less variable, sharing respectively 68 and 81% aa sequence identities with those of the LV (Table 2) . Variations in the length of each The corresponding values for LV are indicated within the parenthesis. In case of ORFs 3, 4, and 5, the lower values correspond to the molecular mass of the proteins after removal of the predicted N-terminal signal sequence, determined using the weight matrix ofVon Heijne [31 ] ORF was also noted making the products of ORFs 7, 5, 4, and 3 of the Qu6bec IAF-exp91 strain shorter than those of the LV (Table 1) , whereas ORF 6 of the Qu6bec strain had an additional aa residue.

Comparison of the aa sequences of the predicted encoding product of each of the ORFs also revealed that most of the variations are located at the amino terminus ( Fig. 1) . This was particularly evident in case ofORF 5 where 32 aa changes were identified within the 35 most N-terminal residues. The same situation was observed at the N-terminus ofORF 3 product, where only 29% identity was found within the 35 most N-terminal residues. Interestingly, despite these extensive aa changes, the potential N-linked glycosylation sites as well as the general hydropathy profiles of the ORFs products were highly conserved (data not shown). Comparative analyses with LV of the IAF-exp91 ORF 7 and the 3'-terminal non coding region have been previously reported [18] . The ORF 7 product showed 59% aa identity with that of LV (Table 2) .

A conserved sequence motif, (U/A)(C/U/A)(A/G)ACC, has been previously identified for LV which might serve as part of the junction site for the leader sequence during transcription ofmRNAs [6, 22] . Except for ORFs 7 and 6 where the sequence motifAACC was present at nearly the same position upstream the ATG start codon, such consensus sequence was not identified upstream the ORFs 5, 4, and 3 of the IAF-exp91 strain (data not shown).

As shown in Fig. 1 , the IAF-exp91 strain appeared to be closely related to the US PRRSV isolate ATCC VR 2385 [20] . Amino acid identities were of 90% for ORF 5 and 96% for both ORFs 6 and 7. No base deletions or additions could be demonstrated between these two North American strains. As is the case for IAF-exp91 and LV, it appears from these comparisons that ORF 5 is relatively highly variable among PRRSV isolates. Interestingly, most aa substitutions identified within ORF5 products of IAF-exp91 and ATCC VR 2385 strains, corresponded to variable sites between LV and IAF-exp91 strains.

So far, only one ORF 4 sequence data is available from the US PRRSV isolates, namely the MN-lb strain [15] . Surprisingly, as much as 17% aa variations exists between this US isolate and IAF-exp91. In addition, the length of this ORF seemed to vary from one strain to the other (178, 171, and 183 aa for IAF-exp91, MN-tb, and LV, respectively). This finding was unexpected, since ORF 4 was less variable than ORFs 5,3, and 7, when IAF-exp91 strain was compared to LV (Table 2) . It has been shown that the 3'-terminal non-coding region (15I nt) of the IAF-exp91 strain is 22 nt longer than that of the LV [18] . On the basis of this dissimilarity, a RT-PCR assay was set up that permitted not only detection but also differentiation between Qurbec and European strains of PRRSV [19] . The homologous non-coding region of the ATCC VR2385 strain has the same length than that of the Qurbec strain and displays 94% nt identity (Fig. 1) .

The Qurbec IAF-Klop strain of PRRSV was used for in vivo labeling experiments and in vitro analysis of structural genes. Contrary to IAF-exp91 strain, IAF-Klop could be serially propagated in the continuous cell line MARC-145, in which labeling experiments could be readily achieved. Sequence analysis showed that IAF-Klop was closely related to IAF-exp91 with amino acid identities higher than 90% for their structural genes (ORF 5: 94%, ORF 6: 98%, ORF 7: 99%). Furthermore, both strains had identical protein profiles as revealed by western immunoblotting (data not shown).

From IAF-Ktop infected cell lysates, five major viral-induced proteins with apparent Ms of 15 K, 19 K, 24.5 K, 29 K, and 42 K could be consistently immunoprecipitated by the homologous porcine hyperimmune serum (Fig. 2A, lane 3) . These proteins were not present in mock-infected cultures (Fig. 2A, lane 1) . Among these proteins, only the 15 K and 19 K species were efficiently immunoprecipitated by the rabbit hyperimmune serum directed against the IAF-exp91 strain (Fig. 2A,  lane 2) . Furthermore an additional minor protein species with apparent Mr of 14.5 K was revealed by the rabbit antiserum, but failed to react with the homologous porcine antiserum. Glyco F treatment ofimmunoprecipitated proteins resulted in loss of the 24.5 K, 29 K and 42 K polypeptide species, thus suggesting their glycosylated nature (data not shown).

To assess the structural nature of the viral proteins identified above, immunoprecipitation experiments were performed using preparations of 3s[S]methioninelabeled and concentrated extracellular virions. Following incubation with homologous porcine hyperimmune serum, only the 15 K, 19 K, and 24.5 K protein species were clearly immunoprecipitated (Fig. 2B, lane 2) . Interestingly, glycoF treatment of the immunoprecipitated proteins resulted in loss of the 24.5K protein; instead a new species with estimated Mr of 16.5K was observed (Fig. 2B, lane 3) .

As expected from the aa sequences, in vitro translation experiments yielded products with estimated Mr of 15 K and 19 K for ORFs 7 and 6, respectively (Fig. 3A, lanes 3 and 4) . Both products were efficiently immunoprecipitated by the anti-IAF Klop porcine hyperimmune serum (Fig. 3B, lanes 3 and 6) and comigrated electrophoretically with the 15 K and 19 K viral proteins. The MAb SDOW 17 reacted toward ORF 7 product (Fig. 3B, lane 5) , as it was also the case for Mabs VO17 and EP147 (data not shown). On the other hand, the in vitro translated product of O R F 5 had an apparent M~ of 18.5 K (Fig. 2A, lane 5) , which was about 4K less than what was expected from its amino acid sequence (Table 1) . This difference could not be due to a defective clone, since all constructions used for in vitro translation experiments were sequenced and no abnormality was detected. Moreover, the O R F 5 product was highly recognized by the anti-IAF Klop porcine hyperimmune serum (Fig 3B, lane 8) . Furthermore, when in vitro translation was carried out in the presence of canine microsomal membranes, a larger protein which migrated approximately to the same distance as the 24.5 K viral protein was observed (Fig. 4, tane 3) . -methionine or translation reaction products of ORF 7, ORF 6, and ORF 5 were incubated with the homologous porcine hyperimrnune serum. Antigen-antibody complexes were analysed by SDS-PAGE on a 12% polyacrylamide gel as shown in 2, 3, 6, and 8, respectively. A normal serum collected from a seronegative SPF pig was used as control for precipitation of ORFs 7, 6, and 5 (4, 7, and 9, respectively). The ORF 7 translation product was also immunoprecipitated by monoclonal antibody SDOWl 7 (5) directed against the 15 K protein of the American reference strain ATCC VR 2332 of PRRSV. Positions of molecular size markers (in kilodaltons) are indicated on the left

In the present study, important genomic variations were demonstrated between a Qu6bec reference strain of P R R S V and LV, the European prototype virus. Since these viruses have a distinct geographic origin, and the fact that the Qu6bec strain seemed to have circulated several years before its initial isolation [7] , genomic heterogeneity a m o n g t h e m was expected. The data obtained from this genomic comparison are in agreement with previous findings by others who demonstrated the existence of antigenic and genomic variations between European and US P R R S V strains, the latters being more heterogeneous [20, 25, 32] . Genomic analyses of the Qu6bec reference strain further substantiate previous findings suggesting that North American and European PRRSV isolates may have diverged into two distinct genotypes [18, 20] . A common feature in the sequences of European PRRSV isolates is the absence of 22 nucleotides in the first half of the 3 '-terminal non-coding region, in comparison with North American isolates studied [18, 20, 30] . It remains to be seen, as more sequence data will be available, whether this property could be considered as a marker trait between European and North American PRRSV isolates. Biological and molecular similarities between LV, LDV, and EAV, have been well documented [2, 6, 10, 21, 26] , with LV and LDV being much more related to each other than to EAV. Despite its high genomic divergence with LV, the Qu6bec PRRSV strain also appeared to be more closely related to LDV than to EAV. The relatedness between PRRSV and LDV in several aspects is intriguing. Indeed, since the latter has been identified many years ago, one can easily speculate that PRRSV may have been derived from LDV. So far, there is no data which suggests an eventual replication of LDV in pigs or PRRSV in mice [13] . Phylogenetic analyses of the polymerase genes ofcoronaviruses, toroviruses, EAV,LDV, and LV, indicated that both LV and LDV may have derived from an EAV-like progenitor, while coronaviruses hypothetically arose from a torovirus progenitor [10] . Genetic recombinations between positive-stranded RNA viruses have been well demonstrated [11, 29] . A copy-choice mechanism, due to a polymerase "jump" from one negativesense R N A template to another, has been proposed for the generation of recombinant coronaviruses [16] . Whether such a mechanism has accounted in the emergence of PRRSV from LDV or from an arterivirus common ancestor, remains questionable. Recently, genomic variations have been also identified within the N and the M genes of several geographically distinct EAV isolates [3, 24] . Despite the distinction of three genomic variants among various EAV strains, their M and N proteins appeared to remain relatively unchanged, which is in agreement with the fact that no distinct antigenic variants of EAV has been identified so far. This contrasts with PRRSV, since the genomic differences between European and North American strains are considerable, and thus, tends to be more heterogeneous.

Recently three structural proteins, N (15 kDa), M (18 kDa), and E (25 kDa), have been identified for LV [23] , in agreement with previous findings with IAF-exp91 and ATCC-2332 strains [17, 25] . The viral structural proteins were shown to be encoded by ORFs 7, 6, and 5, respectively [23] . A comparable protein profile was obtained for the cell culture-adapted Qu6bec IAF-Klop strain, suggesting that although genomically highly divergent, PRRSV isolates are structurally related. Indeed, despite high degrees of aa variations, in vitro translation experiments demonstrated that the primary structure of the ORFs 5 to 7 of the Qu6bec strain remains comparable to that of LV [23] . However, the difference observed in the electrophoretic mobility of the ORF 5 translated product of the Qu6bec strain (18.5 K) with that predicted from the aa sequence (22.4 K) is causing some concerns. Moreover, a 16.5 K protein species was obtained following glycoF treatment of viral proteins, rather than the 19 K protein predicted from the ORF 5 aa sequence after removal of the putative signal peptide. Studies using monospecific antisera should permit to elucidate whether the 16.5 K protein species could represent the primary unglycosylated form of the 24.5 K protein species. Such discrepancy in the size of the ORF 5 product predicted from the aa sequence, and the Mr value estimated from the electrophoretic mobility of the its in vitro translation product, has also been reported in case of LV [23] . A 16.5 K unglycosylated protein was also obtained after glycoF treatment of viral proteins precipitated from lysates of infected cells where the duration of the labeling period was only six hours (data not shown). Therefore, protein degradation could not be considered as an explanation for the discrepancies observed between the in vivo and in vitro molecular weights of the ORF 5 product. An interesting finding that has never been reported before, is the identification of two additional viral specific proteins in lysates of IAF-Klop infected MARC-145 cells. Two bands corresponding to proteins with apparent Mr s of 29 K and 42 K were constantly immunoprecipitated by the homologous porcine hyperimmune serum. Since these proteins were not detected with concentrated preparations of extracellular virions, they were considered as non-structural. In contrast, in case of LV, two proteins with apparently identical M~ s have been reported by western immunoblotting analysis using purified virion preparations, but both proteins could not be immunoprecipitated from lysates of LV-infected cells [23] . Based on the reactivity of anti-peptide sera directed against LV ORFs 5 and 7, it was speculated that the 42 K protein might represent a homo-or heterodimer of the E protein (25 kDa), whereas the 28 K species a dimeric form of the N protein (15 kDa) [23] . It remains to be demonstrated whether these products are distinct in cases of the LV and IAF-Klop strains, or whether the antiserum directed against the Qu6bec strain has a distinct specificity pattern from that used for LV.

In conclusion, the results obtained in the present study are in agreement with previous findings on the existence of antigenic diversity between North American and European strains of PRRSV. High degrees of amino acid variations were identified within the structural genes of the Qu6bec and European prototype strains, which might explain the lack of reactivity of the anti-IAF Klop hyperimmune serum when tested against LV in virus neutralization tests (Dea et al., pers. comm.) . Nonetheless, it remains to be elucidated whether the products of ORFs 3 and 4, which were also found to be highly variable, also contribute to the antigenic diversity of PRRSV isolates. Although antipeptide sera raised against ORFs 3 and 4 reacted positively with LV-infected PAM cells [23] , the exact nature and localization of their products are still unknown.

Characterization of swine infertility and respiratory syndrome (SIRS) virus (isolate ATCC VR-2332)

Sequence of 3' end of genome and of 5'end of open reading frame la of lactate dehydrogenase-elevating virus and common junction motifs between 5' leader and bodies of seven subgenomic mRNAs

Comparison of M and N gene sequences distinguishes variation amongst equine arteritis virus isolates

Single-step method of RNA isolation by acid guanidium thiocyanate phenol-chloroform extraction

Isolation of swine infertility and respiratory syndrome virus (isolate ATCC VR-2332) in North America and experimental reproduction of the disease in gnotobiotic pigs

Molecular characterization of porcine reproductive and respiratory syndrome virus, a member of the arterivirus group

Swine reproductive and respiratory syndrome in Qutbec: isolation of an enveloped virus serologically-related to Lelystad virus

Equine arteritis virus is not a togavirus but belongs to the coronavirusqike superfamily

Structural proteins of equine arteritis virus

Complete genomic sequence and phylogenetic analysis of the lactate dehydrogenase-etevating virus (LDV)

Evolution of plus-strand RNA viruses

Porcine reproductive and respiratory syndrome

Mice and rats (laboratory and feral) are not a reservoir for PRRS virus

Enhanced replication of porcine reproductive and respiratory syndrome (PRRS) virus in a homogeneous subpopulation of MA-104 cell line

Cloning expression, and sequence analysis of the ORF 4 gene of the porcine reproductive and respiratory syndrome virus MN-lb

RNA recombination in animal and plant viruses

Porcine reproductive and respiratory syndrome virus: morphological, biochemical and serological characteristics of Qutbec isolates associated to acute and chronic outbreaks of PRRS

Identification of major differences in the nucleocapsid protein genes of a Qutbec strain and European strains of porcine reproductive and respiratory syndrome virus

Detection of porcine reproductive and respiratory syndrome virus and efficient differentiation between Canadian and European strains by reverse transcription and PCR amplification

Molecular cloning and nucleotide sequencing of the 3'-terminal genomic RNA of the porcine reproductive and respiratory syndrome virus

Lelystad virus, the causative agent of porcine epidemic abortion and respiratory syndrome 0~EARS), is related to LDV and EAV

Subgenomic RNAs of Lelystad virus contain a conserved leader-body junction sequence

Characterization of proteins encoded by ORFs 2 to 7 of Lelystad virus

Genomic variability among globally distributed isolates of equine arteritis virus

Differentiation of US and european isolates of porcine reproductive and respiratory syndrome virus by monoclonal antibodies

Lactate dehydrogenase-elevating virus, equine arteritis virus, and simian hemorrhagic fever virus: a new group of positive-strand RNA viruses

Molecular cloning: A laboratory manual

DNA sequencing with chain-terminating inhibitors

Evolution of RNA viruses

Direct detection of porcine reproductive and respiratory syndrome (PRRS) virus by reverse polymerase chain reaction (RT-PCR)

A new method for predicting signal sequence cleavage sites

Antigenic comparison of Lelystad virus and swine and respiratory syndrome (SIRS) virus

Lelystad virus, the cause of porcine epidemic abortion and respiratory syndrome: a review of mystery swine disease research at Lelystad

Mystery swine disease in the Netherlands: the isolation of Lelystad virus

Centre de recherche en virologie

This report was taken in part from a dissertation to be submitted by H.M. to the Centre de Recherche en Virologie, Institut Armand-Frappier, Universit~ du Qu6bec, in partial fulfillment of the requirements for the Ph.D. degree. This study was partly supported by the Consefl des