Veterinary Microbiology, 32 (1992) 101-115 Elsevier Science Publishers B.V., Amsterdam

101

Genomic variability among globally distributed isolates of equine arteritis virus T . W . M u r p h y , W . H . M c C o l l u m a n d P.J. T i m o n e y

Gluck Equine Research Center, University of Kentucky, Lexington, KY USA B.W. K l i n g e b o r n

National Veterinary Institute, Uppsala, Sweden B. H y l l s e t h

Norwegian college of Veterinary Medicine, Oslo, Norway W. G o l n i k

Faculty Veterinary Medicine, Agricultural University, Wroclaw, Poland B. E r a s m u s

Veterinary Research Institute, Onderstepoort, South Africa (Accepted 28 January 1992)

ABSTRACT Murphy, T.W., McCollum, W.H., Timoney, P.J., Klingeborn, B.W., Hyllseth, B., Golnik, W. and Erasmus, B., 1992. Genomic variability among globally distributed isolates of equine arteritis virus. Vet. Microbiol., 32:101-115. Equine arteritis virus (EAV), a non-arthropod borne togavirus, has been shown to have a global distribution. To date, no major antigenic variation has been demonstrated between EAV isolates from different geographic origins. In this study, the genomic RNA of EAV isolates obtained from horses of different breeds in various countries around the world was oligonucleotide fingerprinted. Comparisons of these fingerprints were used to determine the extent of genomic variation among such isolates. Comparisons among isolates from North American horses revealed, for the most part, oligonucleotide homologies of less than 60%. Only 29 of the 98 comparisons revealed greater than 60% oligonucleotide homology. Nonetheless, several comparisons indicated a close epidemiologic relationship between isolates from horses of different breeds located in different states. Though all European isolates were of Standardbred origin and were from horses located in northern European countries, the majority had oligonucleotide homologies of less than 60%. Where oligonucleotide homology was apparent, it was, with one exception, greater than 70%. The two isolates from New Zealand had 93.2% oligonucleotide homology. This is indicative of an extremely close epidemiologic relationship. Comparisons between EAV isolates from around the world revealed oligonucleotide homologies between viruses from North America, Europe and New Zealand. In several instances, this homology was greater

Correspondence to: T.W. Murphy, Gluck Equine Research Center, University of Kentucky, Lexington, KY 40546, USA. This work is published as paper 91-4-44 by permission of the Dean and Director, College of Agriculture, University of Kentucky and Kentucky Agriculture Experiment Station.

0378-1135/92/$05.00 © 1992 Elsevier Science Publishers B.V. All rights reserved.

102

T.W. MURPHY ETAL.

than 70% and in one case greater than 80%. No oligonucleotide homology was evident in comparisons involving the virus from South Africa. The high level of genomic conservation between certain EAV isolates of disparate geographic origins may reflect dissemination of the virus associated with the international movement of horses. The extent of genomic variation demonstrated between most of the EAV isolates used in this study confirms the need for further investigation of genomic heterogeneity among strains of this virus before techniques that rely upon nucleic acid hybridization can be effectively applied as diagnostic procedures.

INTRODUCTION

Equine arteritis virus (EAV), an RNA virus of the family Togaviridae (Porterfield et al, 1978 ), was initially identified in 1953 as the etiologic agent of an influenza-abortion syndrome that was subsequently termed equine viral arteritis (EVA) (Doll et al., 1957). McCollum and co-workers ( 1971 ) demonstrated transmission of the virus from acutely infected to susceptible incontact horses via the respiratory route. This appears to be a major route of transmission of infection during an epizootic. The virus appears to be restricted to Equidae and no inter-epizootic reservoir of EAV was known until demonstration that stallions could become inapparent carriers of the virus. Carrier stallions persistently shed EAV in semen (Timoney et al., 1986 ) and are capable of infecting mares at time of breeding (Timoney et al., 1987). Mares thus venereally infected are in turn capable of infecting susceptible incontact horses via the respiratory route. Serologic studies have provided evidence of global distribution of EAV (McCollum and Bryans, 1973: McGuire et al., 1974; Moraillon and Moraillon, 1978; Timoney and McCollum, 1991 ). To date, epizootics or outbreaks of EVA have been reported in North America and Europe while carrier stallions have been identified in North America, Europe, Africa, Australia and New Zealand (Timoney and McCollum, 1991 ). There has been no evidence so far of serologic sub-groups or major antigenic variation between EAV isolates of disparate chronologic and geographic origins. Furthermore, the attenuated modified live vaccine developed from the prototype Bucyrus strain of EAV (Doll et al., 1968: McCollum, 1969) has been shown to provide protective immunity against challenge with strains of virus in current circulation (Timoney and McCollum, unpublished data). Preliminary studies of genomic relatedness based on oligonucleotide fingerprint comparisons between a limited number of North American EAV isolates (Murphy et al., 1988) revealed that most of these isolates have a high degree of oligonucleotide homology with one another. For the most part, the isolates used in these preliminary comparisons were from thoroughbred stallions residing in Kentucky and New York state. Notwithstanding limitations in the number of isolates examined, their breed and geographic distribution, one isolate was shown to have a markedly divergent oligonucleotide migra-

GENOMIC VARIABILITYOF EQUINE ARTERITIS VIRUS

103

tion pattern. Furthermore, oligonucleotide fingerprint comparisons of EAV isolates from different horses involved in the 1984 epizootic in Kentucky and between virus isolates from the same carrier stallion over an extended period of time indicate an ongoing process of genomic variation of this virus (Murphy et al., 1988; Murphy, unpublished data). This study was undertaken to determine occurrence, extent and geographic distribution of genomic variation among epidemiologically unrelated EAV isolates made from horses of various breeds from around the world. Twentynine EAV isolates from North America, Europe, New Zealand and South Africa were oligonucleotide fingerprinted and the resultant autoradiographs compared. Oligonucleotide fingerprint comparisons were made between isolates from horses in the same general geographic region and between selected isolates from horses in different continents to determine if particular oligonucleotide fingerprint "types" were limited in their breed and geographic distribution. MATERIALS AND METHODS

Virus isolates. Designation, date of specimen collection, specimen type, breed of horse, geographic location of the horse at time of sampling and passage history of EAV isolates used for comparison in this study are given in Table 1. All of the viruses used in this study were confirmed to be EAV in one-way plaque reduction tests using rabbit antiserum prepared against the Bucyrus strain of the virus. European and South African virus isolates were obtained either as wet frozen or lyophilized tissue culture fluids (TCF). Infective semen from two carrier stallions in New Zealand was kindly provided by Dr. Robin Hopkirk, Rotorua, New Zealand. North American isolates used in this study were made directly either from semen or tissue homogenates. Further propagation of all isolates was in the RK-13 rabbit kidney cell line (ATCC CCL 57). Cell culture. Cultures of RK-13 cells were provided by Mr. Wayne Roberts, Livestock Disease Diagnostic Center, Department of Veterinary Science, University of Kentucky, at passage level 340. Cell culture procedures and virus production. Propagation of RK- 13 cells was as previously described (McCollum, 1969). The methods of Timoney et al. (1986) were utilized for primary isolation of virus from semen samples. Virus isolates were passaged twice in 150 cm 2 monolayer cultures of RK-13 cells to produce seed virus stocks with greater than 107 pfu/ml. Virus seed stocks were inoculated onto 12 × 150 cm 2 monolayer cultures of RK-13 cells, allowed to adsorb for 2 h at 37°C then overlaid with 25 ml of Eagle's minimum essential medium with 2% (v/v) fetal bovine serum

104

T.W. MURPHYET AL.

TABLE 1 Designation, isolation data and passage history of equine arteritis virus isolates used in oligonucleotide fingerprint comparisons Isolate Designation

Date

Location

Breed b

Sample Type

Passage History d

87-CA-AI 87-KY-F1 87-PA-A 1 88-KY-A1 88-KY-B1 88-KY-GI 88-KY-H 1 88-KY-K1 88-MI-AI 88-NJ-AI 88-OK-A1 88-PA-B1 88-PA-C1 88-PA-D1 89-MN-A 1 89-MN-BI 89-NZ-A1 89-NZ-B 1 VTR 22 VTR 41 VTR 26 VTR 37 VTR 42 VTR 8 SVH 18 SVH 23 VTR 300 NEAV 1 88-SA-A1

9/10/87 4/24/87 12 / 8 / 87 2/10/88 2/24/88 4-6/88 2/8/88 4/19/88 1/ 14/88 12/22/88 1/20/88 2/23/88 5/19/88 5/19/88 1/24/89 1/24/89 3/8/89 3/8/89 7/25/88 7/28/88 10/22/88 9/8/88 8 / 8 / 88 9/2/88 10/2/88 8/14/88 -/-/85 a 3/7/88 1/25/88

California Kentucky Pennsylvania Kentucky Kentucky Kentucky Kentucky Kentucky Michigan New Jersey Oklahoma Pennsylvania Pennsylvania Pennsylvania Minnesota Minnesota New Zealand New Zealand Sweden Sweden Sweden Sweden Sweden Sweden Sweden Sweden Poland Norway South Africa

DWB TB STB TB TB TB STB TWH STB STB ST STB STB STB ARAB ARAB STB STB STB STB STB STB STB STB STB STB STB STB LP

Semen Semen Semen Semen Semen T.H. c Semen Semen Semen Semen T.H. c Semen Semen Semen Semen Semen Semen Semen Semen Semen Semen Semen Semen Semen Semen Semen T.H. c Semen Semen

RK-13 p4 RK-13 p4 RK-13 p3 RK-13 p4 RK-13 p4 RK-13 p5 RK-13 p4 RK-13 p4 RK-13 p4 RK-I 3 p4 RK-13 p5 RK-13 p4 RK-13 p4 RK-13 p4 RK-13 p4 RK-13 p4 RK-13 p4 RK-13 p4 RK-13 p4 RK-13 p4 RK-13 p5 RK-I 3 p5 RK-13 p4 RK-13 p4 RK-13 p5 RK-13 p4 RK-13 p5 E.K.pl/RK-14 p4 RK-13 p5

aComplete data on exact date of sample collection unavailable for this isolate. bARAB-Arabian; DWB-Dutch Warm Blood; LP-Lippizaner; STB-Standardbred; TB-Thoroughbred; TWH-Tennessee Walking Horse. 'T.H.-Tissue homogenate of aborted fetus. ~RK-13-Rabbit kidney cell line (ATCC CCL 57 ); E.K.-primary equine kidney.

(MEM:2%FBS) and incubated at 37 ° C. Infective TCF was harvested when cytopathic changes were evident in 80-100% of the cells, usually after 3648h.

Virus purification. Equine arteritis virus was purified from infective TCF by polyethylene glycol precipitation and ultracentrifugation as described by Trent and Grant (1980).

GENOMIC VARIABILITY OF EQUINE ARTERITIS VIRUS

105

Viral RNA extraction. Virions were disrupted with Proteinase K and sodium dodecyl sulfate. The RNA containing aqueous phase was extracted twice with phenol: chloroform: isoamyl alcohol (25 : 24: 1 v/v/v).

RNA digestion and 5'-end labeling. RNA was ethanol precipitated and resuspended in TE buffer (20mM Tris: 2mM EDTA, pH 8.0) containing 5 units of Ribonuclease T1 (Calbiochem, San Diego, CA). The mixture was incubated at 37°C for 1 h. The oligonucleotides were 5'-end labeled with 5 units of T4 polynucleotide kinase (Pharmacia Biotechnologies, Piscataway, N J) and gamma-32p ATP (New England Nuclear Products. Boston, MA) as described by Trent et al. (1983).

Two dimensional electrophoresis. Two dimensional electrophoretic separation of labeled oligonucleotides was carried out using a modification of the technique of Lee et al. ( 1979 ) based on the original method of De Wachter and Fiers (1972). Modifications included use of citrate buffer (1 M) o f p H 2.1 in formulation of the first dimension gels and allowing the bromophenol blue dye marker to migrate 24 cm in both dimensions. These modifications resulted in reproducible oligonucleotide migration patterns that were more easily compared. Comparison ofautoradiographs. The short oligonucleotides (less than 12 nucleotide residues) found in the upper region of oligonucleotide fingerprints are distributed in a pattern that is common to all isolates used in this study. This region was used in alignment of the oligonucleotide fingerprints being compared, it being more useful for this purpose than the position of the tracking dyes. Oligonucleotide fingerprints with highly divergent patterns in this region were repeated. A variable number (44-58) of the larger oligonucleotides migrated to a region located below the regular pattern of short oligonucleotides. These larger oligonucleotides were numerically designated and used for comparison purposes. Results were reported as percentages of oligonucleotide homology and reflected the number of larger oligonucleotides shared by the isolates under comparison. Even using the common upper region as an alignment guide in fingerprint comparisons, confidence in the fidelity of comparisons displaying less than 60% oligonucleotide homology was low and resuits of these comparisons were therefore reported as having "low homology" (LH). To ascertain the reproducibility of the oligonucleotide patterns used in this comparison study, the entire procedure from virus propagation through two dimensional electrophoresis was repeated for several randomly selected isolates and for any isolate with a highly divergent oligonucleotide fingerprint pattern. The divergent nature of the oligonucleotide fingerprint of such isolates was confirmed upon repeated oligonucleotide fingerprint analysis. This study involved oligonucleotide fingerprint comparisons between iso-

106

T.W. MURPHY ET AL.

lates from the same geographic region including: ( 1 ) 16 North American isolates made since 1987; (2) 10 European isolates; (3) two isolates from New Zealand; (4) one isolate from South Africa. Seven isolates of North American origin were selected as reference strains on the basis of breed and geographic location of the stallions from which the viruses were isolated. The oligonucleotide fingerprints of these isolates were used for comparison with other North American isolates and with stallion isolates of other geographic origins. Five European isolates were selected, on the basis of geographic location of the stallions from which the viruses were isolated, as reference strains for comparisons involving European isolates. One of the two isolates from New Zealand was selected as a reference strain for comparison with the other New Zealand isolate and for comparison with European isolates. To assess genomic relatedness of isolates from different geographic regions, reference isolates from throughout North America and one of the New Zealand isolates were compared to the European and South African isolates. RESULTS Twenty-nine isolates of EAV from various geographic regions and equine breeds having no known epidemiologic relationship with one another were oligonucleotide fingerprinted. Representative oligonucleotide fingerprints of isolates of North American (88-KY-A1), Swedish (VTR 41) and NewZealand (89-NZ-A1) origin are presented in Figs la-3a. Comparisons involving fingerprints of these isolates demonstrated high levels of oligonucleotide homology with fingerprints of certain isolates from the same geographic region. Graphic representations of these oligonucleotide fingerprints and numerical designation of the oligonucleotides used in the determination of oligonucleotide homology are provided in Figs lb-3b. Comparisons among North American E A V isolates. Isolates 87-CA-A1, 87PA-A1, 88-KY-A1, 88-KY-K1, 88-OK-A1, 88-PA-B1 and 89-MN-A1 were the isolates selected as reference strains for comparisons between North American isolates. Results of these comparisons are presented as percentages of oligonucleotide homology (Table 2 ). Comparisons of North American isolates indicated that most oligonucleotide fingerprints were too divergent to allow levels of oligonucleotide homology to be accurately assessed. With the exception of 88-KY-G 1 and 88-KYK1, however, every isolate demonstrated levels of oligonucleotide homology greater than 60% with at least one other isolate. Comparisons among all North American isolates indicated that 88-OK-A1 demonstrates greater than 60% oligonucleotide homology with 9 of 15 isolates while 88-KY-A1 and 87-PAA 1 demonstrated greater than 60% with eight and seven isolates, respectively. However in the majority of cases in which it could be assessed, oligonucleo-

GENOMIC VARIABILITYOF EQUINE ARTERITIS VIRUS

~

107

~--,--®~ .....-~_.._.~_..~~~~ .....

@

@ @

@ Fig. 1 a, Oligonucleotide fingerprint of 88-KY-A1. b, Graphic representation of fingerprint and numeric designation of larger oligonucleotides used for determination of oligonucleotide homology. Interrupted line indicates demarcation between smaller oligonucleotides used only for fingerprint alignment and the larger oligonucleotides used in comparisons.

108

T.W. MURPHY ET AL.

@ @ @

@

Fig. 2 a, Oligonucleotide fingerprint of VTR 41. b, Graphic representation of fingerprint and numeric designation of larger oligonucleotides used for determination of oligonucleotide homology. Interrupted line indicates demarcation between smaller oligonucleotides used only for fingerprint alignment and the larger oligonucleotides used in comparisons.

GENOMICVARIABILITYOF EQUINEARTERITISVIRUS

109

O 0

®®

® @

@ @ @

@ Fig. 3 a, Oligonucleotide fingerprint of 89-NZ A 1. b, Graphic representation of fingerprint and numeric designation of larger oligonucleotides used for determination of oligonucleotide homology. Interrupted line indicates demarcation between smaller oligonucleotides used only for fingerprint alignment and the larger oligonucleotides used in comparisons.

110

T.W. MURPHY ET AL.

TABLE 2 Percentage o f oligonucleotide h o m o l o g y in fingerprints c o m p a r i s o n s o f recent N o r t h A m e r i c a n isolates o f e q u i n e arteritis virus. 87-CA-A1

87-PA-A1

88-KY-A1

88-KY-K1

88-OK-A1

88-PA-BI

89-MN-A1

87-CA-A1 87-KY-F1 87-PA-Al 88-KY-A1 88-KY-BI 88-KY-G 1 88-KY-H 1 88-KY-K1

67.4 61.2 61.7 LH LH LH LH

61.2 64.3 69.7 67.9 LH 60.7 LH

61.7 73.2 69.7 73.2 LH LH LH

LH a LH LH LH LH LH LH LH

LH 61.2 67.0 63.9 65.3 LH 65.4 LH

LH 64.3 63.4 61.3 LH LH 66.1 LH

LH 62.5 LH LH 66.7 LH 68.8 LH

88-MI-AI 88-N J-A1 88-OK-A1 88-PA-B1 88-PA-C1 88-PA-D1 89-MN-A1 89-MN-BI

LH LH LH LH LH LH LH LH

LH LH 67.0 63.4 LH LH LH LH

67.9 62.5 63.9 61.3 LH LH LH LH

LH LH LH LH LH LH LH LH

LH 83.7 LH 61.2 63.3 LH 85.7

LH LH LH LH LH LH LH

LH LH LH LH LH 62.5 66.7

aLH-low ( < 60%) oligonucleotide h o m o l o g y TABLE 3 Percentage o f oligonucleotide h o m o l o g y in fingerprints c o m p a r i s o n s o f recent E u r o p e a n isolates o f e q u i n e arteritis virus.

VTR 8 V T R 22 V T R 26 V T R 37 V T R 41 V T R 42 V T R 300 NEAV 1 SVH 18 SVH 23

VTR 8

V T R 41

SVH 18

V T R 300

NEAV 1

87.2 LH a LH .69.0 81.4 LH LH LH LH

69.0 79.2 81.5 72.2 LH LH LH LH LH

LH LH LH LH LH LH LH LH LH

LH LH LH LH LH LH LH LH LH

LH LH LH LH LH LH LH LH LH

aLH-Iow ( < 60%) oligonucleotide h o m o l o g y

tide homology was less than 70%. Higher levels of homology ( > 70%) were demonstrated in a limited number of cases with the highest levels evident in comparisons of 88-OK-A1 with 89-MN-B1 (85.7%) and with 88-NJ-A1 ( 83.7% ). Comparison of 88-KY-A 1 with 87-KY-F 1 and 88-KY-B 1 revealed, in both cases, an oligonucleotide homology of 73.2%.

GENOM1C VARIABILITY OF EQUINE ARTERITIS VIRUS

111

TABLE 4 Percentage ofoligonucleotide homology in fingerprints comparisons involving Norrth American, European, New Zealand and South African isolates of equine arteritis virus. 87-CA-A1 87-PA-A1 88-KY-A1 88-KY-KI 88-OK-A1 88-PA-B1 89-MN-A1 89-NZ-Al VTR 8 VTR 22 V T R 42 89-NZ-A1 89-NZ-B 1 88-SA-A1

LH a LH LH LH LH LH

80.4 75.0 73.2 73.0 66.1 LH

66.1 66.1 67.9 71.6 66.1 LH

LH LH LH LH LH LH

71.4 71.4 67.3 LH LH LH

62.5 64.3 64.3 62.2 LH LH

66.7 LH LH 68.8 LH LH

72.7 70.5 70.5 93.2 LH

aLH-low ( < 60%) oligonucleotide homology

Comparisons among European EA V isolates. Isolates VTR 8, VTR 41, VTR 300, NEAV 1 and SVH 18 were selected as reference strains for comparisons of European EAV isolates. Results of these comparisons are presented as percentages of oligonucleotide homology in Table 3. As observed in comparisons among North American isolates of EAV, the majority of oligonucleotide fingerprints of the European isolates were too divergent to allow accurate assessment of oligonucleotide homologies. However, comparisons of VTR 8 with VTR 22 and VTR 42 demonstrated 87.2 and 81.4% oligonucleotide homology, respectively. Similarly, comparisons of VTR 41 with VTR 26 and VTR 22 revealed oligonucleotide homologies of 81.5 and 79.2%, respectively. Comparison of the oligonucleotide fingerprints of VTR 300, NEAV 1 and SVH 18 with fingerprints of other European EAV isolates did not reveal any strains with greater than 60% oligonucleotide homology. Comparisons among New Zealand isolates. Comparisons between the isolates from two New Zealand carrier stallions, 89-NZ-A 1 with 89-NZ-B 1, revealed oligonucleotide homology of 93.2% (Table 4 ). Comparisons among EAV isolates of different geographic origin. Oligonucleotide fingerprints of North American EAV isolates 87-(,A-A1, 87-PA-A1, 88KY-A 1, 88-KY-K 1, 88-OK-A 1, 88-PA-B 1 and 89-MN-A 1 along with one isolate, 89-NZ-A1, from a New Zealand carrier stallion were used as reference strains in comparisons with one South African and eleven European virus isolates. Results of comparisons involving the South African, New Zealand and those European isolates having greater than 60% oligonucleotide homology with North American or New Zealand reference isolates are presented in Table 4. As in the case of earlier comparisons between isolates from the same geographic region, the extent of divergence in oligonucleotide migration patterns was such as to preclude accurate assessment of the genomic relationship

112

T.W. MURPHY ET AL.

between those isolates used as reference strains, the majority of European isolates and the South African virus isolate. Only in some comparisons involving isolates VTR 8, VTR 22 and VTR 42 was oligonucleotide homology greater than 60% evident. North American EAV isolate 87-PA-A1 demonstrated the highest overall level of oligonucleotide homology in comparisons with European isolates VTR 8, VTR 22 and VTR 42, with respective homologies of 80.4%, 75.0% and 73.2%. Isolates 88-KY-A1, 88-OK-A1, 88-PA-B 1 and 89-NZ-A1 were all found to have oligonucleotide homologies of greater than 60% with the three European, isolates while 89-MN-A 1 demonstrated greater than 60% oligonucleotide homology only with isolate VTR 8. New Zealand isolate 89-NZ-A 1 demonstrated greater than 70% oligonucleotide homology with 87-PA-A 1 and 88KY-A1 and greater than 60% homology with 88-PA-B1 and 89-MN-A1. The other New Zealand isolate, 89-NZ-B 1, had greater than 60% homology with 87-PA-A1 and 88-KY-A1. DISCUSSION

Based on the computer simulated model ofAaronson et al. ( 1982 ) and the experimentally derived data of Young et al. ( 1981 ) and Kew et al. (1984), conservation of 85, 50 and 25% of the large oligonucleotides in comparisons between oligonucleotide fingerprints of different viral isolates reflect 1, 5 and 10% nucleotide sequence divergence between their respective genomes. We found that determination of oligonucleotide homologies by visual comparison of fingerprints sharing less than 60% of the larger oligonucleotides was very difficult and would have required coelectrophoresis of the genomic digests to resolve ambiguities in these interpretations. Since the majority of nearly 200 comparisons involved in this present study displayed such low levels of oligonucleotide homology, co-electrophoresis was considered logistically unfeasible. Consequently, oligonucleotide homology was quantified only for isolates with estimated genomic sequence homologies of 96% or greater. In nearly two thirds of the comparisons between North American isolates, the extent of nucleotide sequence divergence and thus oligonucleotide heterogeneity was such as to preclude quantification of the percentage of oligonucleotide homology. In relation to the North American EAV isolates examined, the findings of this study are in marked contrast with those of a previous similar study in which a limited number of North American EAV isolates from horses involved in epizootics of EVA and from inapparent carrier stallions were compared. Results of the earlier study indicated that most of the epidemiologically unrelated viruses were genomically very homologous (Murphy et a!, 1988). The higher levels of oligonucleotide homology recorded may be a reflection of the limited breed diversity and geographic distribution of the horses from which isolates of EAV were derived.

GENOMIC VARIABILITY OF EQUINE ARTERITIS VIRUS

113

Interestingly, the highest level of oligonucleotide homology (85.7%) among North American isolates included in this study was observed between a viral isolate from a Thoroughbred fetus aborted during an outbreak of EVA in Oklahoma and that from an Arabian carrier stallion located in Minnesota. The second highest level (83.7%) was between the same Oklahoma fetal isolate and an isolate from a standardbred stallion residing in New Jersey. In both cases, the levels of oligonucleotide homology observed in these comparisons were similar to those previously observed between isolates from horse involved in the same epizootic (Murphy eta!, 1988 ). High levels of homology between horses of disparate geographic origin should not be altogether unexpected in view of the frequency of interstate and international movement of horses for racing, show and breeding purposes. A surprising feature however, was that high level of homology evident between virus isolates from horses of different breeds. Current management practices do not promote significant inter-breed contact of equines, thus restricting opportunities for occurrence of viral transmission via the respiratory route and even more so with respect to transmission via the venereal route. In spite of this, isolates of EAV with very similar genomic RNA have been obtained from horses of several different breeds. For the most part, comparisons between the European isolates of EAV revealed a lack of oligonucleotide homology similar to that observed among North American isolates. However, several of the comparisons indicate that the relationship between certain isolates may be extremely close. Isolates VTR 8, V T R 22 and V T R 42 yielded oligonucleotide fingerprints with greater than 80% oligonucleotide homology and similar levels are evident in comparisons of V T R 41 with V T R 22 and V T R 26. As mentioned previously (Murphy et al., 1988 ), levels of oligonucleotide homology similar to these were demonstrated in comparisons of isolates involved in the same epizootic. Results of comparisons between North American, New Zealand, South African and European isolates indicated that isolates VTR 8, V T R 22, VTR 42, 89-NZ-A1 and 89-NZ-B 1 may have some epidemiological link with viruses of North American origin. Isolate 87-PA-A 1 demonstrated more oligonucleotide homology with the fingerprints of these three European viruses than with any of the other North American isolates. That oligonucleotide fingerprints of 89-NZ-AI and 89-NZ-B1 were nearly identical indicates that the horses from which these viruses were isolated, in all probability were infected from the same source. The horses from which viruses VTR 42 and 89-NZ-A1 were isolated were both imported into their respective countries from North America as fully mature stallions. These stallions may, in fact, have been infected at some unknown point before export and thus may have introduced that particular strain of EAV into their respective countries of destination. Though the stallion from which the South African isolate was obtained originally came from eastern Europe, no oligonucleotide fingerprint homology could be dem-

114

y.w.MURPHYETAL.

onstrated either with the European isolates or, for that matter, with any other virus isolate included in this study. It is obvious from the lack of demonstrable oligonucleotide fingerprint homology between the majority of isolates involved in this study that genomic variation is a c o m m o n occurrence among strains of EAV in current circulation. The biological significance of this variability however, remains to be determined. This apparent genomic diversity will need to be further addressed before tests utilizing the polymerase chain reaction (Chirnside et al., 1990) or other virus detection methods that involve nucleic acid hybridization can be effectively applied as diagnostic procedures. CONCLUSIONS The results of this study demonstrated higher levels of genomic heterogeneity among isolates of EAV than were indicated in earlier oligonucleotide fingerprint studies. Variation in the genomic R N A of most of the viruses included in this study was such as to preclude reliable oligonucleotide fingerprint comparisons. In a limited n u m b e r of comparisons however, levels of oligonucleotide homology were similar to those observed previously between isolates from horses involved in the same epizootic. Furthermore, data from this study m a y indicate some epidemiological relationship between isolates from horses of different breeds and of divergent geographic origins. REFERENCES Aaronson, R.P., Young, J.P. and Palese, P., 1982. Oligonucleotidemapping: evaluation of its sensitivity by computer-simulation.Nucleic Acids Res., 10: 237-246. Chirnside, E.D. and Spaan, J.M., 1990. Reverse transcription and cDNA amplification by polymerase chain reaction of equine arteritis virus (EAV). J. Virol. Methods, 30:133-140. De Wachter, R. and Fiers, W, 1972. Preparative two-dimensional polyacrylamidegel electrophoresis of 32P-labeled RNA. Anal. Biochem., 49:184-197. Doll, E.R., Bryans, J.T., McCollum, W.H. and Crowe, M.E.W., 1957. Isolation of a filterable agent causing arteritis of horses and abortion by mares. Its differentiation from the equine abortion (influenza) virus. Cornell Vet., 47: 3-41. Doll, E.R., Bryans.J.T., Wilson, J.C. and McCollum,W.H., 1968. Immunization against equine viral arteritis using modified live virus propagated in cell cultures of rabbit kidney. Cornell Vet., 58: 497-524. Holland, J.J., Kennedy, S.I.T., Semler, B.L., Jorles, C.L., Roux, L. and Grabau, E.A., 1980. Defective interfering RNA viruses and the host cell response. In: Fraenkel-Conrat, H. and Wagner, R.R. (Editors), Comprehensive Virology, Vol. 16, Plenum, New York, pp. 137192. Kew, O.M., Nottay, B.K. and Obijeski, J.F., 1984. Applications of oligonucleotidefingerprinting to the identificationof viruses. In: Methods in Virology,AcademicPress, Inc., New York., pp. 41-83. Lee, Y.F., Kitamura, N., Nomoto, A. and Wimmer, E., 1979. Sequence studies of polivirus RNA. IV. Nucleotide sequence complexitiesof polivirus type 1, type 2 and type 1 defective interfering particles RNAs, and fingerprint of the poliovirus type 3 genome. J. Gen. Virol., 44:311-322.

GENOMICVARIABILITYOF EQUINEARTERITISVIRUS

115

McCollum, W.H., 1969. Development of a modified virus strain and vaccine for equine viral arteritis. J. Am. Vet. Med. Assoc., 155: 318-322. McCollum, W.H. and Bryans, J.T., 1973. Serological identification of infection by equine arteritis virus in horses of several countries. In: Bryans, J.T. and Gerber, H. (Editors) Proc. Third Int. Conf. Equine Infectious Diseases. Karger, Basel, pp. 250-262. McCollum, W.H., Pickett, M.E. and Bryans, J.T., 1971. Temporal distribution of equine arteritis virus in respiratory mucosa, tissues and body fluids of horses infected by inhalation. Res. Vet. Sci., 2: 459-464. McGuire, T.C., Crawford, T.B. and Henson, J.B., 1974. Prevalence of antibodies to herpesvirus type 1 and 2, arteritis and infectious anemia viral antigens in equine serum. Am. J. Vet. Res., 35: 181-185. Moraillon, A. and Moraillon, R., 1978. Results of an epidemiological investigation on viral arteritis in France and some European and African countries. Ann. Rech. Vet., 9: 43-45. Murphy, T.W., Timoney, P.J. and McCollum, W.H., 1988. Analysis of genetic variation among strains of equine arteritis virus. In: Powell, D.G (Editor) Equine Infectious Diseases V: Proc. Fifth Int. Conf. University Press of Kentucky, Lexington, Kentucky, pp.3-12. Porterfield, J.S., Casals, J., Chumakov, M.P., Gaidamovich, S.Ya., Hannoun, C., Holmes, I.H., Horzinek, M.C., Mussgay, M., Oker-Blom, N., Russell, P.K. and Trent, D.W., 1978. Togaviridae. Intervirology, 9: 129-148. Timoney, P.J. and McCollum, W.H., 1991. Equine viral arteritis: Current clinical and economic significance. Proceedings of the American Association of Equine Practitioners, 403-409. Timoney, P.J., McCollum, W.H., Murphy, T.W., Roberts, A.W., Willard, J.G. and Carswell, G.D., 1987. The carrier state in equine arteritis virus infection in the stallion with specific emphasis on the venereal mode of virus transmission. J. Reprod. Fertil., Suppl., 35: 95-102. Timoney, P.J., McCollum, W.H., Roberts, A.W. and Murphy, T.W., 1986. Demonstration of the carrier state in naturally acquired equine arteritis virus infection in the stallion. Res. Vet. Sci., 41: 279-280. Trent, D.W. and Grant, J.A., 1980. A comparison of new world alphaviruses in the western equine encephalomyelitis complex by immunochemical and oligonucleotide fingerprint techniques. J. Gen. Virol., 47:261-282 Trent, D.W., Grant, J.A., Rosen, L. and Monath, T.P., 1983. Genetic variation among dengue 2 viruses of different geographic origin. Virology, 128:271-284. Young, J.F., Taussig, R., Aaronson, R.P. and Palese, P., 1981. Advantages and limitations of the oligonucleotide mapping technique for the analysis of viral RNAs. In: Bishop, D. L. and Compans, R.W. (Editors) The replication of negative strand viruses, edited by: Elsevier New York pp. 209-215.

Genomic variability among globally distributed isolates of equine arteritis virus.

Equine arteritis virus (EAV), a non-arthropod borne togavirus, has been shown to have a global distribution. To date, no major antigenic variation has...
1MB Sizes 0 Downloads 0 Views