Molecular epizootiology and evolution of vesicular stomatitis virus New Jersey

Management Strategies for Reducing the Risk of Equines Contracting Vesicular Stomatitis Virus (VSV) in the Western United States

Vesicular stomatitis viruses (VSVs) cause a condition known as vesicular stomatitis (VS), which results in painful lesions in equines, cattle, swine, and camelids, and when transmitted to humans, can cause flu-like symptoms. When animal premises are affected by VS, they are subject to a quarantine. The equine industry more broadly may incur economic losses due to interruptions of animal trade and transportation to shows, competitions, and other events. Equine owners, barn managers, and veterinarians can take proactive measures to reduce the risk of equines contracting VS. To identify appropriate risk management strategies, it helps to understand which biting insects are capable of transmitting the virus to animals, and to identify these insect vectors' preferred habitats and behaviors. We make this area of science more accessible to equine owners, barn managers, and veterinarians, by (1) translating the most relevant scientific information about biting insect vectors of VSV and (2) identifying practical management strategies that might reduce the risk of equines contracting VSV from infectious biting insects or from other equines already infected with VSV. We address transmission risk at four different spatial scales-the animal, the barn/shelter, the barnyard/premises, and the surrounding environment/neighborhood-noting that a multiscale and spatially collaborative strategy may be needed to reduce the risk of VS.

Field Evaluation of a Multiplex Real-Time Reverse Transcription Polymerase Chain Reaction Assay for Detection of Vesicular Stomatitis Virus

Sporadic outbreaks of vesicular stomatitis (VS) in the United States result in significant economic losses for the U.S. livestock industries because VS is a reportable disease that clinically mimics foot-and-mouth disease. Rapid and accurate differentiation of these 2 diseases is critical because their consequences and control strategies differ radically. The objective of the current study was to field validate a 1-tube multiplexed real-time reverse transcription polymerase chain reaction (real-time RT-PCR) assay for the rapid detection of Vesicular stomatitis New Jersey virus and Vesicular stomatitis Indiana virus strains occurring in Mexico and North and Central America. A comprehensive collection of 622 vesicular lesion samples obtained from cattle, horses, and swine from throughout Mexico and Central America was tested by the real-time RT-PCR assay and virus isolation. Overall, clinical sensitivity and specificity of the real-time RT-PCR were 83% and 99%, respectively. Interestingly, VS virus isolates originating from a specific region of Costa Rica were not detected by real-time RT-PCR. Sequence comparisons of these viruses with the real-time RT-PCR probe and primers showed mismatches in the probe and forward and reverse primer regions. Additional lineage-specific primers and a probe corrected the lack of detection of the missing genetic lineage. Thus, this assay reliably identified existing Mexican and Central American VS viruses and proved readily adaptable as new VS viruses were encountered. An important secondary result of this research was the collection of hundreds of new VS virus isolates that provide a foundation from which many additional studies can arise.

Genetic analysis of vesicular stomatitis virus-New Jersey from the 1995 outbreak in the western United States

OBJECTIVE

To compare molecular associations between the vesicular stomatitis virus (VSV)-New Jersey isolates of the 1995 outbreak with those from previous outbreaks between 1982 and 1985 in the western United States.

SAMPLE POPULATION

23 virus isolates considered representative of the 1995 outbreak of vesicular stomatitis.

PROCEDURE

Viral gene coding for surface-envelope protein G was evaluated by use of nucleotide sequencing and phylogenetic analysis.

RESULTS

Changes in up to 0.77% of the nucleotide bases and 1.35% of the amino acids were detected among the 1995 viral isolates, whereas changes in up to 3.2 and 2.9% of the nucleotides and amino acids, respectively, were found, compared with the 1982 to 1985 viruses. Insertions or deletions were not found in the entire gene, which spanned 1,554 nucleotide bases.

CONCLUSIONS AND CLINICAL RELEVANCE

Phylogenetic analysis indicated that the 1995 VSV-New Jersey belongs to a lineage distinct from that of the 1982 to 1985 viruses that caused previous outbreaks in the western United States. Furthermore, it also is distinct from strains from Central America and from the Georgian Hazelhurst strain.

Re-emergence of vesicular stomatitis in the western United States is associated with distinct viral genetic lineages

Phylogenetic analysis of partial phosphoprotein and glycoprotein gene sequences showed that a single genetic lineage of vesicular stomatitis virus (VSV) serotype New Jersey (NJ) caused the 1995 and 1997 outbreaks of vesicular stomatitis (VS) in the western United States. While distinct from VSV-NJ strains causing previous outbreaks in the western United States and those circulating in feral swine in the southeastern United States, this lineage was closely related to viral lineages circulating in the Mexican states of Guerrero, Veracruz, and Oaxaca in 1996, 1989, and 1984 respectively. In 1997 and 1998, VSV serotype Indiana 1 (IN1) re-emerged in the western United States after 30 years. Viruses causing these outbreaks grouped within a single genetic lineage distinct from VSV-IN1 isolates causing outbreaks in the western United States in 1929 and 1956 but closely related to a strain circulating in the state of Colima in central Mexico in 1997. Our data showed that sporadic VS outbreaks in the western United States are caused by genetically distinct viral lineages closer to those circulating in enzootic areas of central and southern Mexico than to those causing previous outbreaks in the United States. The genetic evidence and temporal distribution of outbreaks are not consistent with a pattern of long-term maintenance of VSV in the western United States.

Effect of population patchiness and migration rates on the adaptation and divergence of vesicular stomatitis virus quasispecies populations

The effect of migration among different isolated virus quasispecies populations on their adaptation and diversity was analysed through experimental evolution. An in vitro cell system was employed to simulate migration of vesicular stomatitis virus between isolated homogeneous host cell populations. The results clearly demonstrated a positive correlation between the migration rate and the magnitude of the mean fitness reached by the virus quasispecies populations. The results also showed, although less clearly, that fitness differences among quasispecies decreased with the magnitude of migration. These results are in close agreement with predictions of standard population genetics theory. These results can be explained in terms of the spread of beneficial mutations, originating in a single isolated quasispecies, through the entire system formed by the different quasispecies populations contained in different host cell populations.

The genetic diversity and epizootiology of infectious hematopoietic necrosis virus

Infectious hematopoietic necrosis virus (IHNV) is a rhabdovirus which causes a serious disease in salmonid fish. The T1 ribonuclease fingerprinting method was used to compare the RNA genomes of 26 isolates of IHNV recovered from sockeye salmon (Oncorhynchus nerka), chinook salmon (O. tshawytscha), and steelhead trout (O. mykiss) throughout the enzootic portion of western North America. Most of the isolates analyzed in this study were from a single year (1987) to limit time of isolation as a source of genetic variation. In addition, isolates from different years collected at three sites were analyzed to investigate genetic drift or evolution of IHNV within specific locations. All of the isolates examined by T1 fingerprint analysis contained less than a 50% variation in spot location and were represented by a single fingerprint group. The observed variation was estimated to correspond to less than 5% variation in the nucleic acid sequence. However, sufficient variation was detected to separate the isolates into four subgroups which appeared to correlate to different geographic regions. Host species appeared not to be a significant source of variation. The evolutionary and epizootiologic significance of these findings and their relationship to other evidence of genetic variation in IHNV isolates are discussed.

Application of indirect ELISA for detection of bovine antibodies against vesicular stomatitis viruses

Two indirect enzyme-linked immunosorbent assays (I-ELISAs) are described for the detection of bovine serum antibody to the New Jersey (NJ) and Indiana (IN) vesicular stomatitis viruses (VSV). Serum samples at a dilution of 1:200 were incubated with binary ethylenimine-inactivated VSV-NJ and VSV-IN type-specific antigens preadsorbed to microtiter plates. Bound antibodies were detected by a murine monoclonal antibody to bovine IgG1 conjugated with horseradish peroxidase. The performance of each I-ELISA in detecting homotypic and heterotypic antibodies to VSV-NJ and VSV-IN in sequential serum samples from calves experimentally infected with VSV-NJ or VSV-IN was evaluated. The I-ELISAs detected serotype-specific antibodies to either VSV-NJ or VSV-IN in calves infected with the homologous serotype. Homotypic but not heterotypic anti-VSV-NJ antibodies were first demonstrable by the VSV-NJ I-ELISA during the second week postinfection and remained at an elevated level for a period of 11 weeks, with a gradual decrease thereafter. Similar homotypic antibody profiles measured by the VSV-IN I-ELISA in calves inoculated with VSV-IN were observed. The performances of the I-ELISAs were compared using 1,495 microtiter serum neutralization (MTSN) test-negative bovine field sera collected from cattle in Canada (VS free) and 429 samples collected from cattle in the USA and Mexico (VS-epidemic and VS-endemic areas). The diagnostic specificities of the VSV-NJ and VSV-IN I-ELISAs for the Canadian samples relative to the MTSN test results were in the range of 99.8% and 99.7%, respectively.(ABSTRACT TRUNCATED AT 250 WORDS)

Evolution of the G and P genes of human respiratory syncytial virus (subgroup A) studied by the RNase A mismatch cleavage method

The G and P genes of human respiratory syncytial viruses (subgroup A), isolated between 1961 and 1989, were analyzed by RNase A one-dimensional fingerprinting, using the Long strain as the reference. Total RNA extracted from cells infected with the different isolates was hybridized to radiolabeled antisense G or P RNA probes of the Long virus. The RNA:RNA heteroduplexes were digested with RNase A and the resistant products analyzed by gel electrophoresis. Comparative analysis of the cleavage patterns revealed extensive genetic heterogeneity in both genes among viruses isolated in different epidemics. In contrast, 13 viruses isolated in Montevideo during a 3-month period showed much more restricted heterogeneity; thus, 11 viruses represented the predominant type of this outbreak and only 2 other viruses generated different RNA cleavage patterns distantly related to the major type. Statistical analysis of the results obtained indicated progressive accumulation of genetic changes with time along cocirculating evolutionary lineages within the same antigenic subgroup of RS virus. The results are discussed in terms of a model for RS virus evolution.

Polymerase errors accumulating during natural evolution of the glycoprotein gene of vesicular stomatitis virus Indiana serotype isolates

We report the entire glycoprotein (G) gene nucleotide sequences of 26 vesicular stomatitis virus Indiana serotype (VSV IND) type 1 isolates from North and Central America. These sequences are also compared with partial G gene sequences of VSV IND type 2 (Cocal) and type 3 (Alagoas) viruses and the complete G gene sequences of the more distantly related VSV New Jersey (NJ) and Chandipura viruses. Phylogenetic analysis of the G gene sequences by maximum parsimony revealed four major lineages or subtypes within the classical VSV IND (type 1) viruses, each with a distinct geographic distribution. A high degree of VSV genetic diversity was found in Central America, with several virus subtypes of both VSV IND and NJ serotypes existing in this mainly enzootic disease region. Nineteen percent sequence variation but no deletions or insertions were evident within the 5' noncoding and the coding regions of the VSV IND type 1 G genes. In addition to numerous base substitutions, the 3' noncoding regions of these viruses also contained numerous base insertions and deletions. This resulted in striking variation in G gene sizes, with gene lengths ranging from 1,652 to 1,868 nucleotides. As the VSV IND type 1 subtypes have diverged from the common ancestor with the NJ subtypes, their G mRNAs have accumulated more 3' noncoding sequence inserts, ranging up to 303 nucleotides in length. These primarily consist of an imprecise reiteration of the sequence UUUUUAA, apparently generated by a unique polymerase stuttering error. Analysis of the deduced amino acid sequence differences among VSV IND type 1 viruses revealed numerous substitutions within defined antigenic epitopes, suggesting that immune selection may play a role in the evolution of these viruses.

Vesicular stomatitis New Jersey virus glycoprotein gene sequence and neutralizing epitope stability in an enzootic focus

Vesicular stomatitis New Jersey (VS NJ) virus is capable of undergoing rapid evolution in nature and therefore has the potential for antigenic variation. We selected an area of Costa Rica where VS NJ virus is enzootic to study whether this virus used the mechanism of antigenic variation to persist in nature. Three sentinel herds and three nonsentinel herds were observed from 1986 to 1988. Eleven VS NJ virus isolates were collected from naturally infected cattle. Remarkably, nine animals that were bled prior to reinfection with VS NJ virus had neutralizing antibody titers up to 1: 102,400 yet virus was isolated from, and disease was observed in, these animals. Sequence analysis of the portion of the glycoprotein gene coding for the neutralizing epitopes demonstrated that all virus isolates were 98-100% similar with no indication of specific genetic variation. The 3' end of the glycoprotein gene also remained stable in that all isolates were again 98-100% similar in nucleotide sequence. Each isolate was neutralized to equivalent titers with monoclonal antibodies directed against four neutralizing epitopes on the glycoprotein. Additionally, preisolation sera from each animal were able to neutralize the virus that caused the subsequent disease. These results provide evidence that antigenic variation is not a mechanism used by VS NJ virus to persist in an enzootic focus of Costa Rica.

Phosphoprotein and nucleocapsid protein evolution of vesicular stomatitis virus New Jersey

The entire phosphoprotein (P) and nucleocapsid (N) protein gene sequences and deduced amino acid sequences for 18 selected vesicular stomatitis virus isolates representative of the natural genetic diversity within the New Jersey serotype are reported. Phylogenetic analysis of the data using maximum parsimony allowed construction of evolutionary trees for the individual genes and the combined N, P, and glycoprotein (G) genes of these viruses. Virtually identical rates of nucleotide substitutions were found for each gene, indicating that evolution of these genes occurs at essentially the same rate. Although up to 19 and 17% sequence differences were evident in the P and N genes, respectively, no variation in gene length or evidence of recombinational rearrangements was found. However, striking evolutionary differences were observed among the amino acid sequences of vesicular stomatitis virus New Jersey N, P, and G proteins. The N protein amino acid sequence was the most highly conserved among the different isolates, indicating strong functional and structural constraints. Conversely, the P protein amino acid sequences were highly variable, indicating considerably fewer constraints or greater evolutionary pressure on the P protein. Much of the remarkable amino acid variability of the P protein resided in a hypervariable domain located between amino acids 153 and 205. The variability within this region would be consistent with it playing a structural role as a spacer to maintain correct conformational presentation of the separate active domains of this multifunctional protein. In marked contrast, the adjacent domain I of the P protein (previously thought to be under little evolutionary constraint) contained a highly conserved region. The colocalization of a short, potentially functional overlapping open reading frame to this region may explain this apparent anomaly.

Trajectory analysis of winds and vesicular stomatitis in North America, 1982-5

Outbreaks of vesicular stomatitis, serotype New Jersey, during epidemics in the United States and northern Mexico, 1982-5, were examined by backward trajectories of winds to investigate spread and possible sources. The outbreaks selected for analysis did not involve introduction of disease by infected animals. The findings indicate that wind could have been responsible for carrying infection from northern Mexico to Arizona and New Mexico and thence to Colorado and Utah and on to Wyoming, Idaho and Montana. The results of these analyses are consistent with the findings from T1 RNAse fingerprinting of virus isolates from outbreaks during the epidemics. The arrival of the trajectories was associated with the passage of a front and rain or passage of a front alone or rain alone. At the time of the trajectories temperatures of 10 degrees C and higher were recorded at heights up to 2500-3500 m. Introduction by airborne particles would appear unlikely as it would have required a source of at least 10(5) infectious units per minute per animal. Vesicular stomatitis virus had been isolated from Simulium and Culicoides during the epidemic with amounts of virus from Simulium sufficient to suggest biological transmission. The possibility of Simulium infected with vesicular stomatitis virus being carried downwind to introduce disease is discussed in relation to the behaviour of Simulium and the pathogenesis of vesicular stomatitis in large animals.

Spontaneous mutations leading to antigenic variations in the glycoproteins of vesicular stomatitis virus field isolates

Strains of vesicular stomatitis virus, New Jersey serotype (VSV-NJ), isolated from diseased cattle or swine were examined by genomic RNA sequencing for genetic diversity potentially leading to antigenic variations in their type-specific glycoproteins as determined by reactivity with epitope-specific monoclonal antibodies (MAbs). Seven field isolates recovered in Colorado, New Mexico, Georgia, and Mexico during the widespread 1982-1985 epizootic in the western United States resembled the prototypic 1952 Hazelhurst subtype by partial sequence homology, but amino acid reversions to the 1949 Ogden subtype occurred frequently. When studies were performed with MAbs directed to the Ogden subtype glycoprotein, relatively limited antigenic variation, and only in neutralization epitope VIII, was noted among two of five epizootic isolates from Colorado and New Mexico. However, amino acid differences in the glycoprotein of a 1983 isolate from an enzootic region of Georgia resulted in major antigenic deficiencies in epitopes V, VI, and VII as determined by Western blotting and neutralization of infectivity with epitope-specific MAbs. Quite a few genetic but no antigenic differences were noted in an enzootic 1984 isolate from Mexico, a potential origin of the United States epizootic. Marked or complete loss of epitopes VII, VI, VIII, and V can be traced to spontaneous mutations leading to amino acid substitutions at glycoprotein positions 199, 263, 275, and 317, respectively, in the enzootic Georgia isolate 07/83-GA-P and the epizootic New Mexico isolate 06/85-NM-B. By comparison, closely adjacent amino acid substitutions at glycoprotein positions 210, 268, 277, and 364 occurred in epitope-deficient mutants selected for resistance to neutralization by MAbs specific for epitopes VII, VI, VIII, and V, respectively. Two neutralization epitopes designated X and XI were found to be unique for the G protein of the 1952 Hazelhurst isolate..../52-GA-P. The epitope X-specific MAb H21, in particular, failed to neutralize the infectivity not only of the Ogden subtype..../49-UT-B but also was ineffective against all the 1982-1985 field isolates. The classical 1952 Hazelhurst strain of VSV-NJ is genetically and antigenically quite different from those viruses isolated during the 1982-1985 epizootic.

Variability and evolution of the plant RNA virus pepper mild mottle virus

The RNA genomes of 26 isolates of pepper mild mottle virus were compared by their RNase T1 fingerprints. Twenty-three isolates came from epidemic outbreaks in greenhouse-grown peppers in Almería (southeastern Spain) from 1983 to 1987; three other isolates, from 1980, came from Sicily (Italy) and Zaragoza (central Spain). The 26 fingerprints can be classified into 10 different types; nucleotide substitution rates show them to be very similar. Cluster and cladistic analyses group types corresponding to the Almería isolates separate from those of 1980. Intraannual and interannual nucleotide differences were estimated. An evolutionary model for pepper mild mottle virus built on these data indicates a highly stable population, maintaining its diversity through time, with a main prevailing haplotype from which closely related variants arise that do not replace it. This high stability could be due to strong functional constraints on variation, as suggested by the high proportion of invariant versus polymorphic sites in fingerprints.

Extreme heterogeneity in populations of vesicular stomatitis virus

Vesicular stomatitis virus (VSV) sequence evolution and population heterogeneity were examined by T1 oligonucleotide mapping. Individual clones isolated from clonal pools of wild-type Indiana serotype VSV displayed identical T1 maps. This was observed even after one passage at high concentrations of the potent viral mutagen 5-fluorouracil. Under low-multiplicity passage conditions, the consensus T1 fingerprint of this virus remained unchanged after 523 passages. Interestingly, however, individual clones from this population (passage 523) differed significantly from each other and from consensus sequence. When virus population equilibria were disrupted by high-multiplicity passage (in which defective interfering particle interference is maximized) or passage in the presence of mutagenic levels of 5-fluorouracil, rapid consensus sequence evolution occurred and extreme population heterogeneity was observed (with some members of these population differing from others at hundreds of genome positions). A limited sampling of clones at one stage during high-multiplicity passages suggested the presence of at least several distinct master sequences, the related subpopulations of which exhibit at least transient competitive fitness within the total virus population (M. Eigen and C.K. Biebricher, p. 211-245, in E. Domingo, J.J. Holland, P. Ahlquist, ed., RNA Genetics, vol. 3, 1988). These studies further demonstrate the important role of selective pressure in determining the genetic composition of RNA virus populations. This is true under equilibrium conditions in which little consensus sequence evolution is observed owing to stabilizing selection as well as under conditions in which selective pressure is driving rapid RNA virus genome evolution.

Glycoprotein evolution of vesicular stomatitis virus New Jersey

A T1 ribonuclease fingerprinting study of a large number of virus isolates had previously demonstrated that considerable genetic variability existed among natural isolates of the vesicular stomatitis virus (VSV) New Jersey (NJ) serotype [S.T. Nichol (1988) J. Virol. 62, 572-579]. Based on these results, 34 virus isolates were chosen as representing the extent of genetic diversity within the VSV NJ serotype. We report the entire glycoprotein (G) gene nucleotide sequence and the deduced amino acid sequence for each of these viruses. Up to 19.8% G gene sequence differences could be seen among NJ serotype isolates. Analysis of the distribution of nucleotide substitutions relative to nucleotide codon position revealed that third position changes were distributed randomly throughout the gene. Third base changes constituted 84% of the observed nucleotide substitutions and affected 89% of the third base positions located in the G gene. Only three short oligonucleotide stretches of complete sequence conservation were observed. The remaining nucleotide changes located in the first and second positions were not distributed randomly, indicating that most of the amino acids coded by the G gene cannot be altered without reducing the fitness of the VSV NJ serotype viruses. Despite these constraints, up to 8.5% amino acid differences were observed between virus isolates. These differences were located throughout the G protein including regions adjacent to defined major antibody neutralization epitopes. Apparent clusters of amino acid substitutions were present in the hydrophobic signal sequence, transmembrane domain, and within the cytoplasmic domain of the G protein. A maximum parsimony analysis of the G gene nucleotide sequences allowed construction of a phylogram indicating the evolutionary relationship of these viruses. The VSV NJ serotype appears to contain at least three distinct lineages or subtypes. All recent virus isolates from the United States and Mexico are within subtype I and appear to have evolved from an ancestor more closely related to the Hazelhurst historic strain than other older strains. The implications of these findings for the evolution, epizootiology, and classification of these viruses are discussed.

RNA virus evolution and the control of viral disease

RNA viruses and other RNA genetic elements must be viewed as organized distributions of sequences termed quasi-species. This means that the viral genome is statistically defined but individually indeterminate. Stable distributions may be maintained for extremely long time periods under conditions of population equilibrium. Perturbation of equilibrium results in rapid distribution shifts. This genomic organization has many implications for viral pathogenesis and disease control. This review has emphasized the problem of selection of viral mutants resistant to antiviral drugs and the current difficulties encountered in the design of novel synthetic vaccines. Possible strategies for antiviral therapy and vaccine development have been discussed.

RNA genome stability of Toscana virus during serial transovarial transmission in the sandfly Phlebotomus perniciosus

We have carried out a T1 ribonuclease fingerprinting analysis of the RNA genomes of Toscana virus isolates from successive generations of an experimentally virus-infected laboratory colony of Phlebotomus perniciosus sandflies. This analysis detected no virus RNA genome changes during transovarial transmission of the virus over 12 sandfly generations (a period of almost 2 years). These results demonstrate that although RNA viruses can exhibit high rates of mutational change under a variety of conditions, Toscana virus RNA genomes can be maintained in a stable manner during repeated transovarial virus transmission in the natural insect host. The implications of these results for insect RNA virus evolution are discussed.

Genetic diversity of enzootic isolates of vesicular stomatitis virus New Jersey

The RNA genomes of 43 vesicular stomatitis virus (VSV) isolates of the New Jersey (NJ) serotype were T1-ribonuclease fingerprinted to compare the extent of genetic diversity of virus from regions of epizootic and enzootic disease activity. Forty of these viruses were obtained from Central America during 1982 to 1985. The other three were older isolates, including a 1970 isolate from Culex nigripalpus mosquitos in Guatemala, a 1960 bovine isolate from Panama, and a 1976 isolate from mosquitos (Mansonia indubitans) in Ecuador. The data indicate that extensive genetic diversity exists among virus isolates from this predominantly enzootic disease zone. Six distinct T1 fingerprint groups were identified for the Central American VSV NJ isolates from 1982 to 1985. The 1960 VSV NJ isolate from Panama and the 1976 isolate from Ecuador formed two additional distinct fingerprint groups. This finding is in sharp contrast to the relatively close genetic relationship existing among VSV NJ isolates obtained from predominantly epizootic disease areas of the United States and Mexico during the same period (S. T. Nichol, J. Virol. 61:1029-1036, 1987). In this previous study, RNA genome T1 fingerprint differences were observed among isolates from different epizootics; however, the isolates were all clearly members of one large T1 fingerprint group. The eight T1 fingerprint groups described here for Central American and Ecuadorian viruses are distinct from those characterized earlier for virus isolates from the United States and Mexico and for the common laboratory virus strains Ogden and Hazelhurst. Despite being isolated 14 years earlier, the 1970 insect isolate from Guatemala is clearly a member of one of the 1982 to 1985 Central American virus fingerprint groups. This indicates that although virus genetic diversity in the region is extensive, under certain natural conditions particular virus genotypes can be relatively stably maintained for an extended period. The implications of these findings for the evolution of VSV NJ and epizootiology of the disease are discussed.