Skip to main page content

• HOME
• Current Issue
• Archive
• Alerts
• About ASM
• Contact us
• Tech Support
• Journals.ASM.org

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Reprints and Permissions Copyright Information Books from ASM Press MicrobeWorld Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Brault, A. C. Articles by Weaver, S. C. Search for Related Content PubMed PubMed Citation Articles by Brault, A. C. Articles by Weaver, S. C.

 Previous Article  |  Next Article 

Journal of Virology, February 2002, p. 1718-1730, Vol. 76, No. 4
0022-538X/01/$04.00+0     DOI: 10.1128/JVI.76.4.1718-1730.2002
Copyright © 2002, American Society for Microbiology. All Rights Reserved.

Positively Charged Amino Acid Substitutions in the E2 Envelope Glycoprotein Are Associated with the Emergence of Venezuelan Equine Encephalitis Virus

Aaron C. Brault,^1, Ann M. Powers,^1, Edward C. Holmes,² C. H. Woelk,² and Scott C. Weaver^1*

Center for Tropical Diseases and Department of Pathology, University of Texas Medical Branch, Galveston, Texas 77555-0609 ,¹ Department of Zoology, University of Oxford, Oxford OX1 3PS, United Kingdom²

Received 21 August 2001/ Accepted 7 November 2001

Top Abstract Introduction Materials and Methods Results Discussion References

 
Epidemic-epizootic Venezuelan equine encephalitis (VEE) viruses (VEEV) have emerged repeatedly via convergent evolution from enzootic predecessors. However, previous sequence analyses have failed to identify common sets of nucleotide or amino acid substitutions associated with all emaergence events. During 1993 and 1996, VEEV subtype IE epizootics occurred on the Pacific Coast of the states of Chiapas and Oaxaca in southern Mexico. Like other epizootic VEEV strains, when inoculated into guinea pigs and mice, the Mexican isolates were no more virulent than closely related enzootic strains, complicating genetic studies of VEE emergence. Complete genomic sequences of 4 of the Mexican strains were determined and compared to those of closely related enzootic subtype IE isolates from Guatemala. The epizootic viruses were less than 2% different at the nucleotide sequence level, and phylogenetic relationships confirmed that the equine-virulent Mexican strains probably evolved from enzootic progenitors on the Pacific Coast of Mexico or Guatemala. Of 35 amino acids that varied among the Guatemalan and Mexican isolates, only 8 were predicted phylogenetically to have accompanied the phenotypic change. One mutation at position 117 of the E2 envelope glycoprotein, involving replacement of Glu by Lys, resulted in a small-plaque phenotype characteristic of epizootic VEEV strains. Analysis of additional E2 sequences from representative enzootic and epizootic VEEV isolates implicated similar surface charge changes in the emergence of previous South American epizootic phenotypes, indicating that E2 mutations are probably important determinants of the equine-virulent phenotype and of VEE emergence. Maximum-likelihood analysis indicated that one change at E2 position 213 has been influenced by positive selection and convergent evolution of the epizootic phenotype.

Top Abstract Introduction Materials and Methods Results Discussion References

 
Venezuelan equine encephalitis virus (VEEV) is a member of the family Togaviridae in the genus Alphavirus (14, 48). The virus is enveloped and includes a nonsegmented, positive-sense RNA genome of approximately 11.5 kb. The 5' two-thirds of the genome encodes four nonstructural proteins (nsP1 to 4) that are involved in viral replication. After virus entry into the cytoplasm of cells, a nonstructural polyprotein is translated and utilized in the production of full-length negative-sense RNA. The negative-sense RNA is used for the generation of genomic RNA as well as a subgenomic mRNA (26S) that is homologous to the 3' one-third of the genome. The subgenomic RNA is translated directly into a structural polyprotein that is proteolytically cleaved into the capsid, E2, and E1 envelope glycoproteins (40).

The VEEV antigenic complex of alphaviruses is comprised of six antigenic subtypes (I to VI) (5). Subtype I is comprised of five varieties: AB, C, D, E, and F. Only viruses from subtypes IAB and IC have been implicated in large outbreaks of equine and human encephalitis that have occurred from northern South America to southern Texas (43, 47). These viruses cause large outbreaks by exploiting equines as highly efficient amplification hosts. The remaining, enzootic subtypes--including ID, IE, and II--circulate in continuous sylvatic foci and do not cause epidemic disease because they replicate poorly in equines. However, two recent outbreaks in Chiapas and Oaxaca, Mexico, in which four isolates of VEEV subtype IE were made from horses, demonstrated for the first time the potential for this subtype to emerge and cause equine disease. These strains provide an opportunity to investigate further the mutations associated with epizootic emergence (27).

Subtype IE VEEV occurs from western Panama through much of Central America and as far north as Tampico, Mexico. The first isolate of VEEV subtype IE was made from a pool of Culex (Melanoconion) taeniopus mosquitoes in Almirante, Panama, in 1961 (31). The following year, a human isolate was obtained in Panama. Extensive ecological investigations conducted by Cupp and colleagues indicated that these viruses are transmitted in discrete enzootic foci between small mammals and the enzootic vector C. taeniopus (6, 7). Although humans can develop severe and sometime fatal disease from infection with VEE subtype IE viruses when they contact sylvatic transmission foci, they are believed to be only tangential hosts that do not serve as a reservoir to infect mosquitoes. Experimental infections of previous enzootic IE strains indicated that they are generally equine avirulent and produce little or no viremia in equines (9, 11, 42, 43). The two Mexican outbreaks of equine encephalitis mark the first isolations of subtype IE VEEV from equines and the first confirmed cases of equine disease as a result of VEEV subtype IE infection (27).

Epizootic-epidemic subtype IAB and IC VEEVs appear to emerge when mutations of enzootic ID strains from Colombia or Venezuela allow highly efficient amplification in rural habitats (45, 47). Two amino acid substitutions in the E2 glycoprotein, one in the capsid, and four in the nonstructural region have been associated with the epidemiological and phenotypic switch to equine virulence that accompanied a 1992 epizootic-epidemic in western Venezuela (44, 45). However, when compared to genomic sequences from other enzootic and epizootic VEEVs, no consistent amino acid or nucleotide substitutions were associated with epizootic emergence. This indicates that different sets of mutations in the structural and/or nonstructural proteins could be responsible for the epizootic phenotype.

Previous sequencing studies of the N terminus of the PE2 envelope glycoprotein precursor gene of many enzootic and epizootic VEE subtype IE viruses indicated a high degree of genetic similarity (less than 1% nucleotide sequence divergence) between isolates obtained from the two recent Mexican outbreaks and enzootic isolates taken from sylvatic habitats on the Pacific Coast of Guatemala (29). Based on this high genetic similarity between enzootic and epizootic subtype IE VEEV, we hypothesized that the emergence of epizootic subtype IE viruses could have resulted in a similar manner as epizootic IAB or IC viruses, from enzootic progenitors. To determine the nucleotide substitution(s) that could be involved in the evolution of the subtype IE epizootic phenotype, genomic sequences of all four of the available epizootic Mexican viruses were determined and compared to those of closely related enzootic IE strains. Analysis of these sequences and of complete E2 glycoprotein sequences of a number of other enzootic and epizootic subtypes implicated positively charged amino acid residues with epizootic VEE emergence. Plaque size differences that correlate with the epizootic phenotype appear to be mediated by mutations resulting in increased positive charge on the surface of the E2 protein. Unfortunately, like other closely related enzootic and epizootic VEEV strains, laboratory rodents did not respond differently to the epizootic and enzootic IE strains, complicating reverse genetic studies of VEE emergence.

Top Abstract Introduction Materials and Methods Results Discussion References

 
Subtype IE VEEV studied. To determine the potential role of specific nucleotide and/or amino acid substitutions in the emergence of epizootic subtype IE VEEV, complete genomic sequences were determined for a Guatemalan enzootic strain (80U76) and four epizootic Mexican strains (CPA201, CPA152, OAX131, and OAX142) (Table 1). In addition, complete envelope glycoprotein (E2) sequences of multiple enzootic and epizootic VEEV strains described previously (17, 19, 20, 34, 37, 49) (Table 2) were determined and analyzed.

View this table: [in this window] [in a new window]

TABLE 1. Subtype IE VEEV isolates used for complete genomic sequence analysis

View this table: [in this window] [in a new window]

TABLE 2. Subtype I VEEV isolates utilized for E2 glycoprotein sequence analysis

Virus preparation. For RNA preparation, viruses were diluted and used to infect either baby hamster (BHK-21) cells or rabbit kidney (RK) cells at a multiplicity of infection of <0.1. After approximately 75% of the cells exhibited cytopathic effects, RNA was extracted from 250 µl of the culture supernatant using Trizol LS (Gibco-BRL Laboratories, Rockville, Md.) following the manufacturer's protocol. One- to three-day-old mice were also inoculated intracranially with approximately 1,000 PFU of virus, and brains were triturated in Eagle's minimal essential medium for RNA extraction using Trizol.

Plaque size determination. The plaque phenotypes of the enzootic subtype IE VEEV (strains 80U76, MenaII and 68U201) and epizootic subtype IE viruses (strains CPA201, CPA152, OAX131, and OAX142) were compared as described by Martin et al. (25). In addition, a number of previously characterized enzootic and epizootic VEEVs were used as controls. Briefly, Vero cells were seeded into 10-cm-diameter tissue culture plates and allowed to grow to confluency. Approximately 150 PFU of each virus was diluted in 10 ml of minimal essential medium supplemented with 5% fetal bovine serum and adsorbed to the Vero monolayers for 1 h at 37°C. A 25-ml 0.4% Noble agar (Sigma, St. Louis, Mo.) minimal essential medium overlay was added, and the cells were incubated at 37°C for 72 h. Agar plugs were removed, and the cells were stained with a 20% methanol, 0.25% crystal violet solution. Approximately 30 to 40 well-isolated plaques were measured for each virus. An analysis of variance was performed from the average mean plaque diameters of each virus tested.

Infection of guinea pigs and mice. In order to determine the potential effects of cell culture passage on the in vivo phenotype of VEE subtype IE viruses assayed, virulence of these viruses was determined in guinea pigs as well as suckling and adult mice. Stocks of parental enzootic (68U201) IE and epizootic (CPA201) viruses were inoculated into four (per virus) 6- to 8-week-old (300- to 500-g) strain 13 inbred, English shorthaired guinea pigs. A 200-µl aliquot of each virus (1,000 Vero cell PFU) in phosphate-buffered saline was injected subcutaneously. Animals were observed twice daily for signs of infection. Blood (20 µl) was collected daily from the saphenous vein for 5 days, and the titer of virus in the blood was determined by plaque assay on Vero cells as described elsewhere (32). In addition, 1,000 PFU of strains 68U201, CPA152, and CPA201 were inoculated subcutaneously into 12-week-old Swiss NIH mice (n = 6). To determine the potential role of suckling mice for the selection of the Glu-117 residue, strain CPA201 was intracerebrally inoculated into six suckling mice, and the E2 glycoprotein gene of the viruses recovered from the brains was sequenced following reverse transcription (RT)-PCR amplification.

RT-PCR. Synthesis of cDNA was performed from the RNA preparation using a poly(T) oligonucleotide primer (T25V-Mlu; 5"-TTACGAATTCACGCGT₂₅V-3") for amplification of the structural region of the genome and primer V-IE-6707(-) (5"-GCACCAATTCTCTATGAATCCCAC-3") for amplification of the nonstructural region as described previously (44). Amplicons were produced using the high-fidelity polymerase Pfu Turbo (Stratagene, La Jolla, Calif.). The following conditions were used for synthesis of IE amplicons: 30 cycles of denaturation at 95°C for 30 s, primer annealing for 30 s at a temperature 5°C below the lowest predicted melting temperature for the respective primer pair (Table 3), and extension at 72°C for 2 min per kb amplified. A final 10-min extension was performed in order to ensure complete amplicon synthesis. Amplification of the IE genomes was performed in five overlapping amplicons ranging from 1.5 to 3.1 kb. Primers used for amplification and sequencing are listed in Table 3. Amplicons were synthesized with the following primer pairs: V-IE-XbaI(+)-V-IE-3141(-), V-IE-2908(+)-V-IE-4447(-), V-IE-4252(+)-V-IE-6707(-), V-IE-6509(+)-V-IE-9284(-), and V-IE-9101(+)-T25VMluI(-).

View this table: [in this window] [in a new window]

TABLE 3. Primers utilized for amplification and sequencing of VEE IE genomes

Sequencing of PCR amplicons. Sequencing of amplicons was performed directly using an Applied Biosystems (Foster City, Calif.) Prism 377 sequencer. Samples were processed by the cycle-sequencing dye termination method using the Big Dye Terminator Cycle Sequencing Reaction kit (Applied Biosystems). Approximately 1 µg of amplicon DNA was added to 3 pmol of each primer with the recommended amount of terminator dyes. Samples were cycled under the following conditions: 96°C for 1 min followed by 25 cycles of 96°C for 15 s, 46°C for 10 s, and 60°C for 2 min. Samples were purified using Bio-Rad (Hercules, Calif.) Micro BioSpin P-30 columns, dehydrated and resuspended in formamide-loading dye (4:1).

Amplification and sequencing of the 5" terminus. The 5" untranslated region (UTR) of strains CPA201, CPA152, OAX131, OAX142, and 80U76 was amplified with the 5" RACE System (version 2.0) from Gibco-BRL. Briefly, cDNA was synthesized using the negative-sense primer V-IE-538(-) (Table 3) and tailed with dCTP using terminal deoxynucleotide transferase. Amplification of the 5" end of the modified cDNA was performed using poly(G) forward sense primer in conjunction with V-IE-332(-) to generate an amplicon containing the 5" terminus. Sequence was determined using the V-IE-332(-) primer as described above.

Sequence and phylogenetic analyses. Deduced amino acid sequences were aligned using the PILEUP program in the Wisconsin Package (8) using default parameters, and the nucleotide sequences were aligned manually based on codon positional homology. Phylogenetic analyses were performed using maximum-parsimony, neighbor-joining, and maximum-likelihood (ML) programs implemented in the PAUP 4.0 software package (41). Distance analyses used the Kimura two-parameter and HKY85 formulas to correct for multiple substitutions of the same nucleotides. Unordered and ordered characters (transition/transversion ratio = 4.5:1 based on previous estimates [2, 44]) were used in the parsimony analysis. Homologous sequences of other VEEV subtypes were used as an outgroup. Bootstrap analysis was performed with 1,000 replicates to determine confidence values on the nodes within trees (10). For ML analyses, the general time-reversible model of nucleotide substitution was used (relative substitution rates: A C = 1.512, A G = 4.952, A T = 1.728, C G = 0.2439, C T = 14.110, G T =1), with a proportion of 0.472 nucleotide sites being invariable and a gamma distribution of among-site rate variation (alpha shape parameter) of 1.821. Finally, the estimated base composition was as follows: A = 0.259, C = 0.278, G = 0.249, T = 0.213. The starting tree in the analysis was found using neighbor-joining which was followed by successive rounds of tree bisection reconstruction branch-swapping, identifying the ML substitution parameters at each stage, until the tree of highest likelihood was found.

Identification of positive selection using ML. An ML approach, using various models of codon substitution which differ in how they treat positive selection (51), was used to examine selection pressures acting on the E2 glycoprotein of VEEVs. Seven models of codon substitution were used which either fixed or estimated nonsynonymous/synonymous substitution (d_N/d_S) ratios, denoted as the parameter, for different categories of sites (p) along the sequences. The simplest model, M0, estimates a single value for all sites. In contrast, M1 has parameters that describe solely neutral evolution, by allowing sites to be either deleterious (p₀, ₀ = 0) or neutral (p₁, ₁ = 1). M2 adds a third category of sites (p₂) that may be positively selected if ₂ is >1. A more-sensitive test for positive selection was provided by M3, which estimates individually for three classes of site. Again, cases in which was >1 were allowed. M7 and M8 both use a discrete beta distribution (with 10 categories) to model among sites, although M8 also estimates an 11th category of sites at which can be >1. Because some of these models are nested, they can be compared using a likelihood ratio test, in which twice the difference in log likelihood between models follows a ² distribution (degrees of freedom equal to the difference in the number of parameters between the models). Evidence for positive selection is inferred when models that allow positive selection (M2, M3, and M8) are significantly favored over those that do not (M0, M1, and M7). Sites that were positively selected could also be identified individually using a Bayesian approach that calculates the probability that a site falls into the selected site class. Finally, the free-ratio model estimates for the entire gene for each branch of the tree separately. The likelihood of this model was compared to that of M0, in which a single value applies to the whole tree. All these methods were implemented using the CODEML program of the PAML package (41).

Nucleotide sequence accession numbers. Nucleotide sequence accession numbers AF448535 through AF448539AF448539 were deposited in GenBank.

Top Abstract Introduction Materials and Methods Results Discussion References

 
Attempts to identify a small animal model for the epizootic phenotype. Guinea pig virulence has been partially successful as a marker for epizootic potential of VEEV strains and has been used to successfully differentiate equine-virulent subtype IAB viruses from enzootic VEE subtype IE viruses (32). However, many equine-avirulent, VEE subtype ID viruses have demonstrated guinea pig virulence (35, 45). We evaluated guinea pig virulence as a marker to distinguish IE viruses with an equine-virulent phenotype from their enzootic, equine-avirulent counterparts. Inbred strain 13 English shorthaired guinea pigs developed high-titer viremias with both enzootic strain 68U201 and Mexican epizootic strain CPA201. Viremia titers were not significantly different at any of the time points assayed (Fig. 1), nor were mortality rates, which were 50% for both the epizootic and the enzootic strain groups. There was no difference in average survival time for the enzootic (5.8 ± 1.5 days) or epizootic (5.4 ± 1.2 days) viruses. These results indicate that guinea pigs are not useful as an in vivo marker for the determination of genetic differences associated with epizootic potential of VEE IE viruses. Furthermore, the fact that the CPA201 strain had a similar in vivo phenotype in guinea pigs to that of strain 68U201 provides further support that the Lys-117 substitution by itself was not the result of heparan sulfate adaptation in limited cell culture passages, which has been shown to attenuate Sindbis virus and VEEV (1, 4, 21). However, because these strains differ in several different amino acids in several viral proteins, the effect of the Lys-117 substitution cannot be determined without mutagenesis experiments involving common infectious clone backgrounds.

View larger version (15K): [in this window] [in a new window]

FIG. 1. Guinea pig viremia following infection with enzootic (68U201) and epizootic (CPA201) VEE subtype IE viruses. Guinea pigs were inoculated with 1,000 PFU, and 20 µl of blood was drawn at the postinoculation time points listed. Viremia was determined by plaque assay on Vero cells with a detection limit of 1.7 log PFU/ml. Error bars indicate standard errors. , CPA201; •, 68U201.

Unlike inbred strain 13 English shorthaired guinea pig infections, infection of Swiss NIH mice with VEEV strains 68U201, CPA201, and CPA152 demonstrated marked differences in virulence. Only one of the six mice inoculated with strain CPA201 and three of the five mice inoculated with epizootic strain CPA152 died from subcutaneous inoculation of 1,000 PFU, compared with six of six mice inoculated with enzootic strain 68U201. In contrast, intracerebral inoculation of CPA201 was lethal in 100% (six of six) of suckling mice inoculated. These data indicate that the epizootic Mexican strains were less rather than more virulent than the enzootic strain 68U201, inconsistent with their presumed equine virulence characteristics. These data add to previously published evidence indicating that murine virulence is generally not correlated with the equine virulence of naturally occurring VEEV variants (16, 45).

Plaque size phenotypes. Using Noble agar, analysis of 20 plaques from each of the enzootic and epizootic VEE subtype IE viruses examined indicated that three of the four epizootic, Mexican strains produced plaques which were significantly (analysis of variance: P < 0.001) smaller than those of the enzootic IE viruses examined (Fig. 2). Smaller but still significant differences in plaque size were observed using agarose (data not shown). These results agree with past studies demonstrating a strong, inverse correlation between plaque size and equine virulence (25, 45).

View larger version (12K): [in this window] [in a new window]

FIG. 2. Plaque diameter analysis of enzootic and epizootic subtype IE VEEVs. All enzootic (MenaII, IE.AA [68U201], and 80U76) and epizootic (CPA201, CPA152, OAX131, and OAX142) viruses were incubated on Vero cells for 3 days under a 0.4% Noble agar overlay. Monolayers were fixed with 20% methanol stained with crystal violet, and plaque diameters (n = 15) were measured.

Complete genomic sequence comparisons among subtype IE VEEV isolates. Complete genomic nucleotide sequences of four VEEV strains isolated during the 1993 and 1996 equine epizootics were determined, as well as that of enzootic isolate 80U76 from nearby coastal Guatemala. These sequences were compared to previously determined sequences for subtype IE strains from Guatemala (68U201) (28) and Panama (MenaII) (18) (Table 1). One discrepancy was found between our sequences and partial E2 sequences reported previously for the Mexican strains CPA201 (previously called 93-42124) and OAX131. Oberste et al. (27) reported Glu residues at position 117, while our sequences encoded Lys at this position for three of the four Mexican strains. Because Oberste et al. sequenced the same isolates after one additional passage in suckling mice, we investigated the possibility that the suckling mice selected for the Glu residue and altered the wild-type sequence. Following intracranial suckling-mouse passage of strains 68U201, CPA201, and CPA152, RNA was extracted and sequence analysis was performed with viruses isolated from brain homogenates. Sequence analysis of the E2 envelope glycoprotein indicated that the consensus Lys at E2 position 117 changed to a Glu following this single passage for CPA201 and CPA152 viruses. This result indicated that the suckling-mouse passages done by Oberste et al. (27) probably changed the consensus sequence for some of the Mexican virus strains and that the lower-passage Lys-117 residue probably represents the wild-type consensus sequence. A less-parsimonious but possible explanation would be that the additional mouse passage resulted in reversion to the wild-type Glu-117 amino acid. Unfortunately, the original horse brain samples are no longer available for confirmation of the wild-type sequence of these virus strains, so the natural sequence may never be known with certainty. A strong possibility is that both amino acids are present in the natural viral quasispecies, and the consensus changes depending on the host that is infected. Experiments with infection of natural hosts by viral mixtures are needed to evaluate this hypothesis.

Previous data from partial PE2 sequences suggested that the enzootic isolate, 80U76, was the virus most closely related to the four epizootic IE virus strains (29). In contrast, our complete genomic sequences indicate that, in a pairwise comparison with the four epizootic Mexican IE viruses, enzootic strain 68U201 shared a greater nucleotide and amino acid sequence identity than the 1980 isolate that was made from the same sylvatic focus at La Avellana, Guatemala (Table 1). The 1993 epizootic strain CPA201 shared the closest sequence identity to the Guatemalan enzootic IE viruses, having 181 nucleotide differences (1.6%) versus enzootic strain 68U201, compared to 186, 187, and 189 differences for the 1996 strains OAX142, CPA152, and OAX131, respectively. The Mexican epizootic IE viruses were quite conserved with a total of only 39 nucleotide (99.7% sequence identity) and eight amino acid differences (99.8% identity) between the most divergent strains, CPA201 and OAX131 (Table 4). Interestingly, there were no more than four nucleotide differences between any of the 1996 Mexican epizootic isolates, but all differences were nonsynonymous. The OAX142 and CPA152 isolates were different at a single amino acid position within the E2 glycoprotein (E2-117). As in previous analyses, the MenaII isolate from Panama was more distantly related to the Guatemalan and Mexican isolates, with 7.4 to 7.6% nucleotide sequence divergence.

View this table: [in this window] [in a new window]

TABLE 4. Percent amino acid and nucleotide identity of complete genomic enzootic and epizootic subtype IE VEEV sequences

Sequence comparisons among the closely related enzootic Guatemalan and epizootic Mexican isolates revealed variable nucleotide positions within all major genes, as well as the 5' and 3' UTRs. The only variable 5' UTR nucleotide was position 28, where enzootic strain 80U76 had a U and all others had a C. Within the 3' UTR, strain 68U201 had a unique deletion at position 11436 and unique nucleotides at positions 11353 (U versus A in all other strains) and 11400 (C versus U).

Thirty-five amino acid differences resulted from the 182 nucleotide differences found in the coding regions of the Mexican and Guatemalan subtype IE VEEV genomes (Tables 4 and 5). A total of 31 amino acid differences was found in the nonstructural portion of the genome with only four differences in the structural genes. The greatest concentration of amino acid differences was in the C terminus of nsP3, in which a total of nine amino acid differences were found between genomic nucleotide positions 5046 and 5675. This region is known from previous studies to be variable in sequence and length among VEEV strains (28). A deletion of three consecutive nucleotides (out of the reading frame) was found in this region of nsP3, resulting in a single alanine for the CPA201 strain compared to an aspartic acid and a threonine residue for the other strains.

View this table: [in this window] [in a new window]

TABLE 5. Deduced amino acid differences among enzootic and epizootic subtype IE VEEV

Analysis of the 24 amino acid differences identified between strains CPA201 and 68U201 revealed a total of four residues in epizootic strain CPA201 that were not shared with all of the other three Mexican viruses analyzed. These amino acids were at nucleotide positions 5058 (nsP3-339), 5154 to 5156 (nsP3-375 to -376) and 8923 (E2-117). Another site within the nsP2 gene at nucleotide position number 3497 (amino acid 617) was found in all four epizootic strains and enzootic strain 80U76 but was not found in the other two enzootic viruses, 68U201 and MenaII. The deletion found in CPA201 from nucleotide positions 5154 to 5156 was absent in all three of the epizootic viruses sequenced as well as the three enzootic viruses. The enzootic strain 68U201 was found to have a potential N-linked glycosylation site in the E2 glycoprotein at nucleotide position 9793 (amino acid 407). This enzootic virus encoded a Ser, which could serve as a recognition sequence for N-linked glycosylation (Asn-X-Ser/Thr); the remaining isolates, including two enzootic isolates, encoded an Arg to ablate the potential glycosylation site. However, this part of the E2 glycoprotein near the C terminus probably lies within the cytoplasmic tail and is therefore probably not glycosylated.

Our results indicated that seven deduced amino acid residues were unique to the four epizootic Mexican VEEV genomes. These differences were found in the nsP1 (enzootic Asp versus epizootic Glu at position 106), nsP2 (Asn versus Ser at position 669), nsP3 (Ile versus Asn at position 476, and Leu versus Phe at position 518), nsP4 (Ser versus Ala at position 604), capsid (Asn versus His at position 100), and E2 (Ser versus Asn at position 218) proteins. Synonymous nucleotide substitutions were found throughout the coding regions (nsP1 to 4, capsid, E2, E1, 6K, and E3 genes), with the greatest number (51) in nsP2 (Tables 4 and 5). There was a total of 162 transitions and 29 transversions, yielding a transition-to-transversion ratio of 5.6:1.

Charge differences on the E2 glycoprotein correlated with these plaque size phenotypes. The three epizootic VEE subtype IE viruses that contained a consensus Lys residue at amino acid position 117 of the E2 glycoprotein all formed smaller plaques than did strains OAX142, 80U76, and 68U201 that had a Glu at that residue. Because strains CPA152 and OAX142 differ by only one nucleotide and deduced amino acid (E2-117) in the entire genome, these results implicated E2 charge interaction with polyanions in the unpurified Noble agar as a determinant of plaque size, as first suggested by Martin and Johnston (25). Increased positive charge on the surface of the glycoprotein and its binding to polyanions in agar probably interferes with the ability of the virus to spread during plaque formation, resulting in a smaller plaque phenotype.

Phylogenetic relationships between enzootic and epizootic subtype IE viruses. The genomic sequences of the Guatemalan and Mexican subtype IE viruses, as well as that of strain MenaII from Panama, were aligned with those of other subtype I and II VEEV strains determined previously, and extremely robust phylogenetic trees were constructed. As in trees published previously from partial PE2 sequences, the IE viruses formed a monophyletic group (Fig. 3). All epizootic Mexican viruses were found on the most-terminal branches of the Pacific Coast lineage and were most closely related to the Guatemalan enzootic strains 68U201 and 80U76. All other relationships among VEE serotypes were also consistent with previous analyses.

View larger version (32K): [in this window] [in a new window]

FIG. 3. Maximum-parsimony tree generated from complete genomic sequences of enzootic and epizootic VEE subtype I and II viruses. Strains are designated by subtype, location (PA, Panama; GU, Guatemala; MX, Mexico; FL, Florida; TX, Texas; TR, Trinidad; VE, Venezuela; CO, Columbia), year of isolation (last two digits of year only), and strain name. All nodes in the tree have bootstrap values of 100% when nucleotide sequences are analyzed, except for the 68U201-80U76 grouping, with a bootstrap value of 54%. The table shows nucleotide and amino acid changes assigned to the branch representing the emergence of the epizootic subtype IE Mexican epizootic viruses. Trees generated using the neighbor-joining and ML methods had identical topologies.

Maximum parsimony analysis was used to predict the nucleotide and amino acid changes that accompanied the phenotypic changes associated with the Mexican VEE emergence. The branch leading to the Mexican clade included 54 nucleotide substitutions, 8 of which encoded amino acid changes. In addition to the seven nonsynonymous nucleotide differences unique to the Mexican strains (see above), a mutation encoding a Glu-to-Lys change at E2 position 117 was found in this branch. This amino acid substitution was found in three (CPA201, CPA152, and OAX131) of the four Mexican epizootic VEE IE viruses. The most parsimonious reconstruction for this nucleotide and deduced amino acid is that the OAX142 strain, which encodes Glu in its consensus sequence like the Guatemalan enzootic strains, reverted to this residue after the 1996 emergence or after laboratory passage in suckling mice.

E2 glycoprotein sequence analyses. Comparison of the amino acid changes predicted to have accompanied the Mexican VEE emergence in 1993 with those identified previously for subtype IAB and IC epizootic VEEV strains yielded no common substitutions. However, both the Mexican (Fig. 3) and 1992 Venezuelan subtype IC (44) emergences were accompanied by the replacement of negatively charged or uncharged E2 envelope glycoprotein residues by lysine (Mexican) or arginine (1992 Venezuelan). To determine if this trend extended to other epizootic VEEV emergences, complete E2 sequences reported previously were aligned with those of additional, representative strains that we sequenced (Table 2) and maximum-parsimony trees were constructed from amino acid sequences based on the more robust topology generated from the nucleotide sequences. The amino acid changes accompanying each epizootic emergence were predicted (Fig. 4). For the IAB viruses that probably emerged during the 1920s, the parsimony analysis predicted 7 substitutions, two of which involve the addition of positive charged amino acids (Glu-75Lys and Thr-213Lys) for a net charge change of +3. The 1962 subtype IC emergence was accompanied by 5 E2 amino acid changes, with 3 involving charge changes (Asp-117Gly, Val-179Ile, Ala-192Val, Glu-201Lys, Thr-213Lys) for a net change of +4. As described previously, the 1992 IC emergence involved the replacement of Gly-193 and Thr-213 by Arg residues for a net change of +2.

View larger version (43K): [in this window] [in a new window]

FIG. 4. Neighbor-joining tree generated from complete PE2 envelope glycoprotein gene nucleotide sequences of representative VEE subtype I viruses. Strains are designated by subtype, location (PA, Panama; GU, Guatemala; MX, Mexico; FL, Florida; TX, Texas; TR, Trinidad; VE, Venezuela; CO, Columbia; PE, Peru), year of isolation (last two digits of year only), and strain name. Boxes indicate amino acid substitutions in the E2 envelope glycoprotein associated with emergence of epizootic strains from predicted enzootic progenitors (internal nodes) and accompanying attenuation of the TC-83 vaccine strain from its TRD parent strain. Numbers indicate bootstrap values of adjacent nodes. The tree was rooted using an outgroup consisting of all other VEEV subtypes. Trees generated using the maximum-parsimony and ML methods had identical topologies except for some groupings within the IAB clade.

ML analyses of positive selection in E2 mutations. ML analysis of the evolutionary processes acting on the VEEV E2 glycoprotein provided some support for the action of positive selection. Under M0, which assumes a constant d_N/d_S across the whole sequence, strong purifying selection seems to be the dominant mode of evolution ( = 0.0775). A similar picture is apparent from M1, where 91% of sites are conserved and 9% evolve neutrally; in M2, where most sites are conserved and the extra class of sites (p₂) also depicts relatively strong constraints (₂ = 0.149); and in the more complex M8 model. However, the M3 model, which has the highest likelihood of any model, provides some evidence for highly localized positive selection. Under M3, 93% of sites are strongly conserved (₀ = 0.025), 7% show relatively high variability (₁ = 0.507), and 0.04% are subject to moderate positive selection (₂= 3.320). Comparing these models using a series of likelihood ratio tests, we found that M3 is significantly favored over the M0 and M1 models (p = 5.053 x 10^-33 and 5.117 x 10^-40, respectively), which do not allow positive selection, but not over M2 (P = 0.943). Finally, although the free-ratio model is significantly favored over M0 (p = 1.355 x 10^-6), no branch on the tree has an anomalously high value indicative of positive selection. We therefore conclude that adaptive evolution may have acted on these sequences, but is difficult to show this statistically because it has clearly affected such a small number of codons.

Using a Bayesian approach we determined that just a single site, E2 amino acid position 213, fell into the positively selected class (95% probability) in M3. The E2-213 residue is adjacent to amino acids implicated in antigenicity of different VEEV subtypes (13) and with mouse attenuation of the vaccine strain TC-83 (15). Interestingly, a positively charged (Lys or Arg) amino acid residue was found for all IAB and IC epizootic VEEVs at position 213. All enzootic VEE subtype ID viruses had a Thr residue at this position; however, VEEV subtype II (Everglades virus), which can be considered a ID genetic variant (48), had a Lys residue at this position. The experimentally demonstrated equine avirulence of the VEE subtype II enzootic viruses (12) implicates at least one additional genetic determinant for the development of an epizootic phenotype.

In contrast to the E2-213 of VEE subtype IAB and IC viruses that appear to evolve in response to positive selection, the ML method was incapable of identifying the Lys (E2-117) of the epizootic VEE subtype IE viruses as being under positive selective pressure. This was a result of the mutation being found only in one clade of epizootic IE viruses and not in the other epizootic subtypes (IAB or IC). VEE IE viruses are approximately 20 to 25% divergent in their nucleotide sequences from subtype IAB and the most closely related IC viruses (37). Therefore, it is not surprising that the development of an epizootic phenotype could evolve through the incorporation of several different combinations of amino acid changes.

Assessment of the possible role of cell culture passage on positive-charge mutations in the E2 envelope glycoprotein. Recent studies with another alphavirus, Sindbis virus, demonstrated that passage in vertebrate cells such as BHK-21 results in selection for binding to heparan sulfate, which is mediated by the replacement of uncharged or negatively charged E2 residues by Arg and Lys (4, 21). Many of the enzootic VEEV strains we sequenced had more-extensive passage histories in cell culture than the epizootic strains yet did not show evidence of the positive-charge E2 mutations predicted to be involved in epizootic emergence. The earliest VEEV isolates, epizootic strains, were generally maintained by animal passage before modern cell culture techniques became available, whereas enzootic strains were first isolated in the 1960s and generally maintained in cell culture. If the overall trends in the charge of our E2 sequences had been influenced by laboratory passage, it would be expected that enzootic strains might be influenced more by artificial adaptation to heparan sulfate binding via positive-charge mutations; this prediction is at odds with our data showing positive-charge trends in epizootic rather than enzootic strains. Furthermore, the possibility that the same combinations of epizootic charge mutations occurred independently from laboratory passage selection seems remote, considering the stochastic nature of mutation. Finally, one of the lowest passage strains we examined, epizootic strain 3908 with only one mosquito cell passage, is highly virulent for horses (R. A. Bowen [Colorado State University], personal communication) and has all of the positive charge amino acids implicated in the subtype IC emergence (Fig. 4).

Despite the data cited above indicating the remote possibility that the positive charge changes implicated in epizootic emergence resulted from artificial passage selection, we investigated experimentally the possibility that selection for heparan sulfate binding affected our prediction of E2 mutations mediating VEE emergence by passaging enzootic ID and IE viruses in BHK-21 cells. Twelve unpassaged strains of enzootic subtype ID VEEV isolated during 1997 to 1998 in western Venezuela and central Colombia (26), including several that are extremely closely related to the predicted progenitor of the 1992 VEEV subtype IC epizootic (44), were passaged 10 times each with dilutions to maintain a multiplicity of infection of approximately 0.1. Virus rescued from an infectious cDNA clone of enzootic Guatemalan strain 68U201 (32) was also passaged in triplicate, parallel series in RK cells to duplicate the history of the Mexican subtype IE epizootic strains prior to our acquisition. Following these passages, RNA was extracted from the cell culture supernatants and the complete PE2 gene was amplified in two overlapping RT-PCR amplicons as described previously (34). Amplicons were sequenced directly to determine consensus sequences of the virus populations. In 3 of the 12 subtype ID passages, positive-charge mutations appeared that were similar to those identified previously in Sindbis virus (4, 21). However, only one of these mutations was identical to one of the two predicted to be associated with VEE emergence (strain ZPC-912, replacement of Gly-193 by Arg) (Table 6). The passage of strain 68U201 in RK cells never resulted in a change at E2 position 117, as was predicted to have accompanied the Mexican VEE emergence (Fig. 3). These results underscore the stochastic nature of E2 mutations selected during cell culture passage and indicate that it is extremely unlikely that selection for convergent, artifactual adaptation to cell culture passage is an explanation for the E2 amino acid changes that our phylogenetic analyses predict are associated with epizootic emergence.

View this table: [in this window] [in a new window]

TABLE 6. Amino acid changes in the E2 envelope glycoprotein of enzootic subtype ID VEEV isolates following 10 passages in BHK-21 cells

Top Abstract Introduction Materials and Methods Results Discussion References

 
Emergence of epizootic VEEV from enzootic progenitors. VEEV has been responsible for many outbreaks of human and equine disease in the Americas since the early 20th century. The most important outbreaks have occurred in northern South America and Central America. In 1969 an outbreak of a VEE subtype IAB virus began near the Guatemala-El Salvador border and spread north to southern Texas (23). In 1995 an outbreak of a VEE subtype IC virus spread from northern Venezuela across Lake Maracaibo and into the Guajira Peninsula of Venezuela and neighboring Colombia. An estimated 100,000 human infections occurred, with an unrecorded number of equine cases (50). Although they involved relatively small numbers of equine cases, the two recent (1993 and 1996) Mexican outbreaks have renewed interest in the subtype IE viruses, which were not previously known to have epizootic potential (27).

Previous work based on sequences from the N terminus of the E2 glycoprotein demonstrated that enzootic VEE subtype IAB and IC viruses have evolved or emerged convergently from closely related enzootic ID viruses on at least three separate occasions. Preliminary sequence analysis of the same genetic region of VEE subtype IE viruses, including four isolates made during the two Mexican outbreaks, suggested a similar emergence from an enzootic subtype IE predecessor. To determine the nucleotide substitutions that could be responsible for the acquisition of equine virulence and emergence of a subtype IE epizootic phenotype from an enzootic precursor, complete genomic sequences were determined for all available isolates from the two Mexican outbreaks and compared to those of the most closely related enzootic VEE subtype IE viruses. Our sequence analysis and subsequent phylogenetic analysis confirmed the close genetic identity of these Mexican epizootic and Guatemalan enzootic viruses.

E2 charge differences associated with epizootic VEE emergence. When amino acid changes in the E2 envelope protein were examined, a similar pattern of substitution was found to be associated with all epizootic emergence events (Fig. 4). In all four branches leading to subtype IAB, IC, or IE epizootic clades, amino acid substitutions resulting in increased charge were observed. The evidence for amino acid 213 involvement is especially strong because both the IAB and IC emergences are associated with changes from threonine to either Lys or Arg, respectively (44), and our methods identified this site as affected by positive selection and convergent evolution. Furthermore, the attenuation of the TC-83 vaccine strain from its IAB Trinidad donkey (TRD) parent is accompanied by a decrease in the E2 charge, although the major attenuating mutation is Lys-120, presumably the result of cell culture adaptation (17). Although the mutations involving charge are not believed to be the most important determinants of murine attenuation (15), previous studies indicate that murine virulence does not necessarily correlate with equine virulence (16, 45). Thus, our sequence data provide strong circumstantial evidence that E2 amino acid substitutions leading to increased charge, especially the T-213Lys/Arg substitution, are involved in increased equine virulence that accompanies VEE emergence. Interestingly, the enzootic VEE subtype II virus, Everglades, also has a lysine residue at amino acid 213 of the E2 glycoprotein. This suggests that Everglades virus could possess a genotype capable of emerging into an equine-virulent phenotype with minor amino acid alterations. However, Everglades, although a subtype ID virus in the genetic sense, represents a lineage that has not produced epizootic strains (33). A possible explanation is that minor differences within the enzootic genetic backbone could strongly influence the propensity for epizootic emergence by imposing different requirements for the number of mutations that must occur simultaneously. The use of infectious clones produced in our laboratory from several enzootic subtype ID, IE, and II strains is under way to isolate and evaluate the effect of these positively charged E2 amino acids on the epizootic phenotype and to evaluate the enzootic strain specificity of VEE emergence.

Markers of the epizootic phenotype. A number of criteria have been utilized for the classification of VEEVs as epizootic. The ability to develop a high-titer viremia in equines, resulting in the infection of mosquito vectors, has traditionally been viewed as the most important phenotype for epizootic potential. Animals such as English shorthaired guinea pigs have also been used as virulence models for enzootic IE and epizootic IAB and IC viruses; however, infection of these guinea pigs with some enzootic subtype ID viruses result in lethal infections (35, 36), limiting their usefulness for studying emergence mechanisms (45). Another possible epizootic marker is an increased infectivity for mosquito vectors, such as Aedes taeniorhynchus, that have been implicated in epizootic transmission during many VEE outbreaks (22).

Several in vitro markers of the epizootic phenotype, such as elution from hydroxylapatite chromatography columns at low pH, have also been used to distinguish enzootic and epizootic VEEVs. Hydroxylapatite binding profiles, like plaque sizes, are believed to be related to charge differences on the surface of the E2 glycoprotein (39). Another marker of the epizootic phenotype that has been proposed has been sensitivity to alpha/beta interferon. Spotts et al. (38) reported that epizootic VEEVs, including, Mexican strain CPA201, were more resistant to the effects of exogenously administered murine alpha/beta interferon. Epizootic viruses produced cytopathic effects in the presence of higher concentrations of interferon than did enzootic strains. Mutagenesis of an epizootic subtype IAB virus demonstrated that determinants of interferon sensitivity are encoded by the 5' untranslated genome region as well as the E2 glycoprotein (32, 38).

Plaque size analysis has been used most extensively as a marker for epizootic potential of VEEVs. With the exception of subtypes III and VI and the VEEV subtype IAB vaccine derivative, TC-83, a small-plaque phenotype has been correlated strongly with the equine-virulent phenotype (25). In our studies, enzootic subtype IE isolates 68U201, 80U76, and MenaII produced large plaques while three of the four epizootic Mexican IE viruses (CPA201, CPA152, and OAX131) produced significantly smaller plaques. However, one Mexican subtype IE strain isolated from an equine during the 1996 outbreak, OAX142, had a large-plaque phenotype similar to that of the enzootic strains. These results underscore the limitations of plaque size as a marker of the epizootic phenotype; this marker appears to be useful only for naturally occurring subtype IABCD VEEV strains.

The consensus genomic sequence of strain OAX142 differed from one of the small-plaque isolates, CPA152, by a single nucleotide that encodes an amino acid difference at position 117 of the E2 glycoprotein. The OAX142 consensus sequence encodes a Glu residue that results in a charge difference on the E2 glycoprotein compared with the Lys found in the other Mexican epizootic strains. This charge difference could reduce the electrostatic interaction between the E2 glycoprotein and polyanions present in partially purified agar (Noble) used in our plaque assay. The smaller difference observed with agarose probably reflects a smaller amount of anionic contaminants.

Although the plaque size phenotype was described nearly 20 years ago and was believed to reflect virion charge differences (25), the role of virion charge in expression of the epizootic phenotype remains unknown. Recent studies demonstrated the propensity of another alphavirus, Sindbis virus, to incorporate positive charge on the surface of the E2 glycoprotein in response to selection for heparan sulfate binding (4, 21). Heparan sulfate-adapted viruses produce smaller plaques than wild-type Sindbis viruses, but these strains also exhibit reduced virulence for mice. This contrasts with the association between the small plaque size and equine virulence of most VEEV variants (25, 44, 45). Because equines are unique in their differential response to epizootic and enzootic VEEV, the plaque size and charge differences suggest a fundamental difference in cell binding properties of equine cells important in the pathogenesis process. Studies of VEE pathogenesis in a murine model suggest that dendritic cells deserve study in this regard (24).

The OAX142 isolate from the brain of an encephalitic horse during a VEE outbreak represents an exception to the generally strong correlation between plaque size and equine virulence. If the Glu-117 amino acid in the consensus sequence of this strain represents the virus population present in this horse, this indicates that, at least for subtype IE VEEV, plaque size is imperfect for the prediction of equine virulence and epizootic potential. Although the Lys-117 residue may be selected during cell culture passage, our results suggest that this mutation in the Mexican epizootic VEE IE viruses did not arise independently in all three strains during RK cell passage, since a genetically similar, cloned enzootic strain 68U201 did not acquire this mutation during even more extensive RK cell passages. A mixture of Glu and Lys in the original, natural quasispecies population is a likely explanation. Results from suckling-mouse passages and adult-mouse infections indicate that the Glu residue is selected in both adult and suckling mice. This finding could be used to explain previous reports of a Glu residue in strain CPA201 that was passaged previously in suckling mice (26). The rapid development of a mixed genotype (Glu and Lys) at E2 position 117 following a single suckling-mouse passage indicates an original quasispecies that consisted of both genotypes. Due to the highly selective nature of this substitution in vivo, determination of its role in pathogenesis and epidemiology presents unique obstacles. Use of an infectious cDNA clone of strain 68U201 to assess the role of the Glu and Lys substitutions in homogenous virus populations in vitro and in vivo is under way in our laboratory.

 
We thank Robert Tesh, Robert Shope, Hilda Guzman, David Alstad, and James Pearson for providing some of the alphaviruses used in our analyses.

A.C.B. was supported by the James L. McLaughlin Infection and Immunity Fellowship Fund and by NIH Emerging Tropical Diseases T32 Training Grant AI-107526. A.M.P. was supported by the James W. McLaughlin Fellowship Fund and NIH T32 Training Grant on Emerging and Reemerging Infectious Diseases AI-07536. This research was supported by National Institutes of Health grants AI-39800 and AI-48807 and by the National Aeronautics and Space Administration (NASA).

 
* Corresponding author. Mailing address: Department of Pathology, University of Texas Medical Branch, Galveston, TX 77555-0609. Phone: (409) 747-0758. Fax: (409) 747-2415. E-mail: sweaver{at}utmb.edu.

Present address: Division of Vector-Borne Infectious Diseases, Centers for Disease Control and Prevention, Fort Collins, CO 80522

Top Abstract Introduction Materials and Methods Results Discussion References

 

1
1. Bernard, K. A., W. B. Klimstra, and R. E. Johnston. 2000. Mutations in the E2 glycoprotein of Venezuelan equine encephalitis virus confer heparan sulfate interaction, low morbidity, and rapid clearance from blood of mice. Virology 276:93-103.[CrossRef][Medline]
2
2. Brault, A. C., A. M. Powers, C. L. Chavez, R. N. Lopez, M. F. Cachon, L. F. Gutierrez, W. Kang, R. B. Tesh, R. E. Shope, and S. C. Weaver. 1999. Genetic and antigenic diversity among eastern equine encephalitis viruses from North, Central, and South America. Am. J. Trop. Med. Hyg. 61:579-586.[Abstract]
3
3. Brault, A. C., A. M. Powers, G. Medina, E. Wang, W. Kang, R. A. Salas, J. De Siger, and S. C. Weaver. 2001. Potential sources of the 1995 Venezuelan equine encephalitis subtype ic epidemic. J. Virol. 75:5823-5832.[Abstract/Free Full Text]
4
4. Byrnes, A. P., and D. E. Griffin. 1998. Binding of Sindbis virus to cell surface heparan sulfate. J. Virol. 72:7349-7356.[Abstract/Free Full Text]
5
5. Calisher, C. H., and N. Karabatsos. 1988. Arbovirus serogroups: definition and geographic distribution, p. 19-57. In T. P. Monath (ed.), The arboviruses: epidemiology and ecology, vol. I. CRC Press, Boca Raton, Fla.
6
6. Cupp, E. W., W. F. Scherer, J. B. Lok, R. J. Brenner, G. M. Dziem, and J. V. Ordonez. 1986. Entomological studies at an enzootic Venezuelan equine encephalitis virus focus in Guatemala, 1977-1980. Am. J. Trop. Med. Hyg. 35:851-859.
7
7. Cupp, E. W., W. F. Scherer, and J. V. Ordonez. 1979. Transmission of Venezuelan encephalitis virus by naturally infected Culex (Melanoconion) opisthopus. Am. J. Trop. Med. Hyg. 28:1060-1063.
8
8. Devereux, J., P. Haeberli, and O. Smithies. 1984. A comprehensive set of sequence analysis programs for the VAX. Nucleic Acids Res. 12:387-395.
9
9. Dietz, W. H., Jr., O. Alvarez, Jr., D. H. Martin, T. E. Walton, L. J. Ackerman, and K. M. Johnson. 1978. Enzootic and epizootic Venezuelan equine encephalomyelitis virus in horses infected by peripheral and intrathecal routes. J. Infect. Dis. 137:227-237.[Medline]
10
10. Felsenstein, J. 1985. Confidence limits on phylogenies: an approach using the bootstrap. Evolution 39:783-791.[CrossRef]
11
11. Garman, J. L., W. F. Scherer, and R. W. Dickerman. 1968. A study of equine virulence of naturally occurring Venezuelan encephalitis virus in Veracruz with description of antibody responses. Bull. Pan-Am. Health Org. 65:238-252.
12
12. Henderson, B. E., W. A. Chappell, J. G. Johnston, Jr., and W. D. Sudia. 1971. Experimental infection of horses with three strains of Venezuelan equine encephalomyelitis virus. I. Clinical and virological studies. Am. J. Epidemiol. 93:194-205.[Abstract/Free Full Text]
13
13. Hunt, A. R., A. J. Johnson, and J. T. Roehrig. 1990. Synthetic peptides of Venezuelan equine encephalomyelitis virus E2 glycoprotein. I. Immunogenic analysis and identification of a protective peptide. Virology 179:701-711.[CrossRef][Medline]
14
14. Johnston, R. E., and C. J. Peters. 1996. Alphaviruses, p. 843-898. In B. N. Fields, D. M. Knipe, and P. M. Howley (ed.), Virology, 3rd ed. Lippincott-Raven, New York, N.Y.
15
15. Kinney, R. M., G. J. Chang, K. R. Tsuchiya, J. M. Sneider, J. T. Roehrig, T. M. Woodward, and D. W. Trent. 1993. Attenuation of Venezuelan equine encephalitis virus strain TC-83 is encoded by the 5'-noncoding region and the E2 envelope glycoprotein. J. Virol. 67:1269-1277.[Abstract/Free Full Text]
16
16. Kinney, R. M., J. J. Esposito, J. H. Mathews, B. J. Johnson, J. T. Roehrig, A. D. Barrett, and D. W. Trent. 1988. Recombinant vaccinia virus/Venezuelan equine encephalitis (VEE) virus protects mice from peripheral VEE virus challenge. J. Virol. 62:4697-4702.[Abstract/Free Full Text]
17
17. Kinney, R. M., B. J. Johnson, J. B. Welch, K. R. Tsuchiya, and D. W. Trent. 1989. The full-length nucleotide sequences of the virulent Trinidad donkey strain of Venezuelan equine encephalitis virus and its attenuated vaccine derivative, strain TC-83. Virology 170:19-30.[CrossRef][Medline]
18
18. Kinney, R. M., M. Pfeffer, K. R. Tsuchiya, G. J. Chang, and J. T. Roehrig. 1998. Nucleotide sequences of the 26S mRNAs of the viruses defining the Venezuelan equine encephalitis antigenic complex. Am. J. Trop. Med. Hyg. 59:952-964.[Abstract]
19
19. Kinney, R. M., K. R. Tsuchiya, J. M. Sneider, and D. W. Trent. 1992. Genetic evidence that epizootic Venezuelan equine encephalitis (VEE) viruses may have evolved from enzootic VEE subtype I-D virus. Virology 191:569-580.[CrossRef][Medline]
20
20. Kinney, R. M., K. R. Tsuchiya, J. M. Sneider, and D. W. Trent. 1992. Molecular evidence for the origin of the widespread Venezuelan equine encephalitis epizootic of 1969 to 1972. J. Gen. Virol. 73:3301-3305.[Abstract/Free Full Text]
21
21. Klimstra, W. B., K. D. Ryman, and R. E. Johnston. 1998. Adaptation of Sindbis virus to BHK cells selects for use of heparan sulfate as an attachment receptor. J. Virol. 72:7357-7366.[Abstract/Free Full Text]
22
22. Kramer, L. D., and W. F. Scherer. 1976. Vector competence of mosquitoes as a marker to distinguish Central American and Mexican epizootic from enzootic strains of Venezuelan encephalitis virus. Am. J. Trop. Med. Hyg. 25:336-346.
23
23. Lord, R. D. 1974. History and geographic distribution of Venezuelan equine encephalitis. Bull. Pan-Am. Health Org. 8:100-110.
24
24. MacDonald, G. H., and R. E. Johnston. 2000. Role of dendritic cell targeting in Venezuelan equine encephalitis virus pathogenesis. J. Virol. 74:914-922.[Abstract/Free Full Text]
25
25. Martin, D. H., W. H. Dietz, O. Alvaerez, Jr., and K. M. Johnson. 1982. Epidemiological significance of Venezuelan equine encephalomyelitis virus in vitro markers. Am. J. Trop. Med. Hyg. 31:561-568.
26
26. Moncayo, A. C., G. M. Medina, Z. Kalvatchev, A. C. Brault, R. Barrera, J. Boshell, C. Ferro, J. E. Freier, J. C. Navarro, R. Salas, J. de Siger, C. Vasquez, R. Walder, and S. C. Weaver. Genetic diversity and relationships among enzootic Venezuelan equine encephalitis virus field isolates from Colombia and Venezuela. Am. J. Trop. Med. Hyg., in press.
27
27. Oberste, M. S., M. Fraire, R. Navarro, C. Zepeda, M. L. Zarate, G. V. Ludwig, J. F. Kondig, S. C. Weaver, J. F. Smith, and R. Rico-Hesse. 1998. Association of Venezuelan equine encephalitis virus subtype IE with two equine epizootics in Mexico. Am. J. Trop. Med. Hyg. 59:100-107.[Abstract]
28
28. Oberste, M. S., M. D. Parker, and J. F. Smith. 1996. Complete sequence of Venezuelan equine encephalitis virus subtype IE reveals conserved and hypervariable domains within the C terminus of nsP3. Virology 219:314-320.[CrossRef][Medline]
29
29. Oberste, M. S., S. M. Schmura, S. C. Weaver, and J. F. Smith. 1999. Geographic distribution of Venezuelan equine encephalitis virus subtype IE genotypes in Central America and Mexico. Am. J. Trop. Med. Hyg. 60:630-634.[Abstract]
30
30. Oberste, M. S., S. C. Weaver, D. M. Watts, and J. F. Smith. 1998. Identification and genetic analysis of Panama-genotype Venezuelan equine encephalitis virus subtype ID in Peru. Am. J. Trop. Med. Hyg. 58:41-46.[Abstract]
31
31. Peralta, P. H., and A. Shelokov. 1966. Isolation and characterization of arboviruses from Almirante, Republic of Panama. Am. J. Trop. Med. Hyg. 15:369-378.
32
32. Powers, A. M., A. C. Brault, R. M. Kinney, and S. C. Weaver. 2000. The use of chimeric Venezuelan equine encephalitis viruses as an approach for the molecular identification of natural virulence determinants. J. Virol. 74:4258-4263.[Abstract/Free Full Text]
33
33. Powers, A. M., M. S. Oberste, A. C. Brault, R. Rico-Hesse, S. M. Schmura, J. F. Smith, W. Kang, W. P. Sweeney, and S. C. Weaver. 1997. Repeated emergence of epidemic/epizootic Venezuelan equine encephalitis from a single genotype of enzootic subtype ID virus. J. Virol. 71:6697-6705.[Abstract]
34
34. Salas, R. A., C. Z. Garcia, J. Liria, R. Barrera, J. C. Navarro, G. Medina, C. Vasquez, Z. Fernandez, and S. C. Weaver. 2001. Ecological studies of enzootic Venezuelan equine encephalitis in north-central Venezuela, 1997-1998. Am. J. Trop. Med. Hyg. 64:84-92.[Abstract]
35
35. Scherer, W. F., and J. Chin. 1977. Responses of guinea pigs to infections with strains of Venezuelan encephalitis virus, and correlations with equine virulence. Am. J. Trop. Med. Hyg. 26:307-312.
36
36. Scherer, W. F., J. Chin, and J. V. Ordonez. 1979. Further observations on infections of guinea pigs with Venezuelan encephalitis virus strains. Am. J. Trop. Med. Hyg. 28:725-728.
37
37. Sneider, J. M., R. M. Kinney, K. R. Tsuchiya, and D. W. Trent. 1993. Molecular evidence that epizootic Venezuelan equine encephalitis (VEE) I-AB viruses are not evolutionary derivatives of enzootic VEE subtype I-E or II viruses. J. Gen. Virol. 74:519-523.[Abstract/Free Full Text]
38
38. Spotts, D. R., R. M. Reich, M. A. Kalkhan, R. M. Kinney, and J. T. Roehrig. 1998. Resistance to alpha/beta interferons correlates with the epizootic and virulence potential of Venezuelan equine encephalitis viruses and is determined by the 5' noncoding region and glycoproteins. J. Virol. 72:10286-10291.[Abstract/Free Full Text]
39
39. Stanick, D. R., M. E. Wiebe, and W. F. Scherer. 1985. Markers of Venezuelan encephalitis virus which distinguish enzootic strains of subtype I-D from those of I-E. Am. J. Epidemiol. 122:234-244.[Abstract/Free Full Text]
40
40. Strauss, J. H., and E. G. Strauss. 1994. The alphaviruses: gene expression, replication, and evolution. Microbiol. Rev. 58:491-562.[Abstract/Free Full Text]
41
41. Swofford, D. L. 1998. PAUP*. phylogenetic analysis using parsimony (*and other methods), version 4. Sinauer Associates, Sunderland, Mass.
42
42. Walton, T. E., O. Alvarez, R. M. Buckwalter, and K. M. Johnson. 1973. Experimental infection of horses with enzootic and epizootic strains of Venezuelan equine encephalomyelitis virus. J. Infect. Dis. 128:271-282.[Medline]
43
43. Walton, T. E., and M. A. Grayson. 1988. Venezuelan equine encephalomyelitis, p. 203-231. In T. P. Monath (ed.), The arboviruses: epidemiology and ecology, vol. IV. CRC Press, Boca Raton, Fla.
44
44. Wang, E., R. Barrera, J. Boshell, C. Ferro, J. E. Freier, J. C. Navarro, R. Salas, C. Vasquez, and S. C. Weaver. 1999. Genetic and phenotypic changes accompanying the emergence of epizootic subtype IC Venezuelan equine encephalitis viruses from an enzootic subtype ID progenitor. J. Virol. 73:4266-4271.[Abstract/Free Full Text]
45
45. Wang, E., R. A. Bowen, G. Medina, A. M. Powers, W. Kang, L. M. Chandler, R. E. Shope, and S. C. Weaver. 2001. Virulence and viremia characteristics of 1992 epizootic subtype IC Venezuelan equine encephalitis viruses and closely related enzootic subtype ID strains. Am. J. Trop. Med. Hyg. 65:64-69.[Abstract]
46
46. Watts, D. M., V. Lavera, J. Callahan, C. Rossi, M. S. Oberste, J. T. Roehrig, C. B. Cropp, N. Karabatsos, J. F. Smith, D. J. Gubler, M. T. Wooster, W. M. Nelson, and C. G. Hayes. 1997. Venezuelan equine encephalitis and Oropouche virus infections among Peruvian army troops in the Amazon region of Peru. Am. J. Trop. Med. Hyg. 56:661-667.
47
47. Weaver, S. C. 1998. Recurrent emergence of Venezuelan equine encephalomyelitis, p. 27-42. In W. M. Scheld and J. Hughes (ed.), Emerging infections I. ASM Press, Washington, D.C.
48
48. Weaver, S. C., L. Dalgarno, T. K. Frey, H. V. Huang, R. M. Kinney, C. M. Rice, J. T. Roehrig, R. E. Shope, and E. G. Strauss. 2000. Family Togaviridae, p. 879-889. In M. H. V. van Regenmortel, C. M. Fauquet, D. H. L. Bishop, E. B. Carstens, M. K. Estes, S. M. Lemon, J. Maniloff, M. A. Mayo, D. J. McGeogh, C. R. Pringle, and R. B. Wickner (ed.), Virus taxonomy. Classification and nomenclature of viruses. Seventh report of the International Committee on Taxonomy of Viruses. Academic Press, San Diego, Calif.
49
49. Weaver, S. C., M. Pfeffer, K. Marriott, W. Kang, and R. M. Kinney. 1999. Genetic evidence for the origins of Venezuelan equine encephalitis virus subtype IAB outbreaks. Am. J. Trop. Med. Hyg. 60:441-448.[Abstract]
50
50. Weaver, S. C., R. Salas, R. Rico-Hesse, G. V. Ludwig, M. S. Oberste, J. Boshell, and R. B. Tesh. 1996. Re-emergence of epidemic Venezuelan equine encephalomyelitis in South America. Lancet 348:436-440.[CrossRef][Medline]
51
51. Yang, Z., R. Nielsen, N. Goldman, and A. M. Pedersen. 2000. Codon-substitution models for heterogeneous selection pressure at amino acid sites. Genetics 155:431-449.[Abstract/Free Full Text]

---

Journal of Virology, February 2002, p. 1718-1730, Vol. 76, No. 4
0022-538X/01/$04.00+0     DOI: 10.1128/JVI.76.4.1718-1730.2002
Copyright © 2002, American Society for Microbiology. All Rights Reserved.

This article has been cited by other articles:

• Aguilar, P. V., Adams, A. P., Wang, E., Kang, W., Carrara, A.-S., Anishchenko, M., Frolov, I., Weaver, S. C. (2008). Structural and Nonstructural Protein Genome Regions of Eastern Equine Encephalitis Virus Are Determinants of Interferon Sensitivity and Murine Virulence. J. Virol. 82: 4920-4930 [Abstract] [Full Text]  
• Anishchenko, M., Bowen, R. A., Paessler, S., Austgen, L., Greene, I. P., Weaver, S. C. (2006). From the Cover: Venezuelan encephalitis emergence mediated by a phylogenetically predicted viral mutation. Proc. Natl. Acad. Sci. USA 103: 4994-4999 [Abstract] [Full Text]  
• Holmes, E. C. (2006). The evolution of viral emergence.. Proc. Natl. Acad. Sci. USA 103: 4803-4804 [Full Text]  
• Greene, I. P., Paessler, S., Austgen, L., Anishchenko, M., Brault, A. C., Bowen, R. A., Weaver, S. C. (2005). Envelope Glycoprotein Mutations Mediate Equine Amplification and Virulence of Epizootic Venezuelan Equine Encephalitis Virus. J. Virol. 79: 9128-9133 [Abstract] [Full Text]  
• COFFEY, L. L., WEAVER, S. C. (2005). SUSCEPTIBILITY OF OCHLEROTATUS TAENIORHYNCHUS AND CULEX NIGRIPALPUS FOR EVERGLADES VIRUS. Am J Trop Med Hyg 73: 11-16 [Abstract] [Full Text]  
• GREENE, I. P., PAESSLER, S., ANISHCHENKO, M., SMITH, D. R., BRAULT, A. C., FROLOV, I., WEAVER, S. C. (2005). VENEZUELAN EQUINE ENCEPHALITIS VIRUS IN THE GUINEA PIG MODEL: EVIDENCE FOR EPIZOOTIC VIRULENCE DETERMINANTS OUTSIDE THE E2 ENVELOPE GLYCOPROTEIN GENE. Am J Trop Med Hyg 72: 330-338 [Abstract] [Full Text]  
• EBEL, G. D., CARRICABURU, J., YOUNG, D., BERNARD, K. A., KRAMER, L. D. (2004). GENETIC AND PHENOTYPIC VARIATION OF WEST NILE VIRUS IN NEW YORK, 2000-2003. Am J Trop Med Hyg 71: 493-500 [Abstract] [Full Text]  
• Brault, A. C., Powers, A. M., Ortiz, D., Estrada-Franco, J. G., Navarro-Lopez, R., Weaver, S. C. (2004). Venezuelan equine encephalitis emergence: Enhanced vector infection from a single amino acid substitution in the envelope glycoprotein. Proc. Natl. Acad. Sci. USA 101: 11344-11349 [Abstract] [Full Text]  
• Anishchenko, M., Paessler, S., Greene, I. P., Aguilar, P. V., Carrara, A.-S., Weaver, S. C. (2004). Generation and Characterization of Closely Related Epizootic and Enzootic Infectious cDNA Clones for Studying Interferon Sensitivity and Emergence Mechanisms of Venezuelan Equine Encephalitis Virus. J. Virol. 78: 1-8 [Abstract] [Full Text]  
• Wang, E., Brault, A. C., Powers, A. M., Kang, W., Weaver, S. C. (2002). Glycosaminoglycan Binding Properties of Natural Venezuelan Equine Encephalitis Virus Isolates. J. Virol. 77: 1204-1210 [Abstract] [Full Text]  
• Woelk, C. H., Holmes, E. C. (2002). Reduced Positive Selection in Vector-Borne RNA Viruses. Mol Biol Evol 19: 2333-2336 [Full Text]  
• Brault, A. C., Powers, A. M., Weaver, S. C. (2002). Vector Infection Determinants of Venezuelan Equine Encephalitis Virus Reside within the E2 Envelope Glycoprotein. J. Virol. 76: 6387-6392 [Abstract] [Full Text]  

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Reprints and Permissions Copyright Information Books from ASM Press MicrobeWorld Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Brault, A. C. Articles by Weaver, S. C. Search for Related Content PubMed PubMed Citation Articles by Brault, A. C. Articles by Weaver, S. C.