Previous Article | Next Article ![]()
Journal of Virology, July 2005, p. 9128-9133, Vol. 79, No. 14
0022-538X/05/$08.00+0 doi:10.1128/JVI.79.14.9128-9133.2005
Copyright © 2005, American Society for Microbiology. All Rights Reserved.
Richard A. Bowen,3 and
Scott C. Weaver1,2*
Center for Biodefense and Emerging Infectious Diseases,1 Department of Pathology, University of Texas Medical Branch, Galveston, Texas 77555-0609,2 Animal Reproduction and Biotechnology Lab, Colorado State University, Ft. Collins, Colorado 805223
Received 14 February 2004/ Accepted 26 March 2005
|
|
|---|
|
|
|---|
A member of the Alphavirus genus in the Togaviridae family, VEEV is an enveloped virus with a nonsegmented, positive-sense RNA genome of approximately 11.4 kb containing a 5'-methylguanylate cap and a 3'-polyadenylate tail (14). Upon arrival in the cytoplasm following receptor-mediated endocytosis, the viral genome is translated by cellular components into four nonstructural proteins (nsP1 to -4) that participate in viral replication. The 3' one-third of the genome is expressed from a 26S subgenomic message that encodes three major structural proteins, the capsid, E2 and E1 envelope glycoproteins, that are involved in packaging of the viral genome and in the production of infectious viral particles via budding from the plasma membrane (23).
Serotype IAB and IC epizootic VEEV strains are believed to arise from enzootic subtype ID VEEV progenitors via mutations in the E2 protein and possibly other genes and genome regions (4). Previously, chimeric VEEV strains containing the genetic backbone of enzootic strains and either the full complement of structural genes (17) or the partial envelope glycoprotein genes from epizootic strains (10), as well as reciprocal constructs, were used to map the equine virulence and epizootic phenotypes. The guinea pig, the small animal model studied to date that most closely resembles equines in its differential response to enzootic versus epizootic VEEV strains (20, 21), was used to assess the epizootic phenotype in these studies. An intermediate guinea pig viremia and virulence phenotype exhibited by all of the chimeric VEEV strains suggested that both envelope and nonenvelope genes and genome regions are determinants of the epizootic phenotype (10). However, unlike equines that almost invariably survive enzootic VEEV infection without detectable disease and with little or no viremia, guinea pigs become viremic and suffer fatal disease after infection by some enzootic strains, including those of the serotype ID lineage believed to have generated all IAB and IC epizootic strains (4). This limitation of the guinea pig model indicates the need to confirm the results of reverse genetic studies with experimental equine infections.
We therefore infected horses with wild-type enzootic and epizootic VEEV strains derived from infectious cDNA clones, as well as with chimeric strains that differ only in seven amino acid residues within the envelope glycoproteins. Our results demonstrated that these envelope gene regions are both necessary for and sufficient to generate the epizootic, equine amplification-competent phenotype from an enzootic VEEV strain.
|
|
|---|
Plasmids. Recombinant plasmids encoding cDNA copies of the RNA genomes of relevant epizootic and enzootic VEEV strains were described previously (1, 7, 10, 13). The epizootic subtype IC strain 3908 (EpA) was isolated from a febrile human during the major 1995 Venezuelan epidemic (36). It was passaged once in C6/36 mosquito cells prior to RNA extraction and cDNA clone (pM1-3908) production (7). The Trinidad donkey (TrD) subtype IAB strain was isolated in 1943 and passaged once in a guinea pig brain, six times in Vero cells, and once in BHK cells before RNA extraction and cloning as described previously (13). This parent virus is highly virulent for horses (R.A.B., unpublished), and virus derived from the clone is identical to the parent in a variety of in vitro assays and in mouse infections (S.C.W., unpublished). Enzootic subtype ID strain ZPC 738 (EnA) was isolated from a sentinel hamster in 1997 in Venezuela (28) and was passaged once in Vero cells prior to RNA extraction and cDNA clone (pM1-738) construction (1). Enzootic strain 66637 (EnB) was isolated from a sentinel hamster in Venezuela in 1981 (24) and was passaged once in baby mice and once in Vero cells. Both of the enzootic ID strains were demonstrated previously to be avirulent for horses (29).
Chimeras (Fig. 1) were generated by swapping cDNA fragments to produce plasmids containing an SP6 promoter, 5'-untranslated region (UTR), nsP1 to -4, subgenomic 26S promoter, and 3'-UTR sequences of the backbone virus; the C-terminal 111 amino acids in the capsid, the complete PE2 and 6k genes, and the N-terminal 108 amino acids of the E1 protein gene were derived from the second virus strain to generate chimeras. The 3908/ZPC738-E2 (Ep/EnA-E2) and ZPC738/3908-E2 (En/EpA-E2) chimeric clones were described previously (10), and the 3908/66637-E2 (Ep/EnB-E2) and TrD/66637-E2 (TrD/EnB-E2) clones were generated using the same methods (7). The amino acid differences between EnA and EpA in the swapped fragments included only one residue in the E3 protein, five in E2, and a single difference in E1 (Table 1) (10). EnB, which is slightly more distantly related to epizootic strains including EpA than is enzootic strain EnA (28), had three additional amino acid differences in the E3, E2, and 6k genes of the swapped fragment, respectively, compared to both of the other VEEV strains (Table 1). The IAB strain TrD is similar in amino sequence to EpA but has several additional, unique amino acid differences in the fragment exchanged, compared to all others used in this study (Table 1).
![]() View larger version (35K): [in a new window] |
FIG. 1. Genome maps of parental infectious clones and reciprocal envelope chimeras showing the DNA fragments swapped. The exchanged fragments between AflII and SnabI restriction sites at genomic positions 8031 and 10298 contained the last 111 codons in the capsid protein gene, the complete PE2 and 6k protein genes, and the first 108 codons of the E1 protein gene of the epizootic (Ep) parent in the backbone of the enzootic (En) parent, and vice versa. Shaded regions correspond to genome regions obtained from epizootic viruses.
|
|
View this table: [in a new window] |
TABLE 1. Genetic differences among VEEV strains in the fragment swapped to generate chimeric viruses
|
Antibody assays. All horses were screened for alphavirus antibodies prior to experimental infection using 80% endpoint plaque reduction neutralization tests (PRNT) with the following viruses that represent those typically administered to North American horses in multivalent inactivated vaccines: VEEV strain TC-83, eastern equine encephalitis virus strain 82V2137, and western equine encephalitis virus Fleming strain. Following infection, seroconversion was assessed using PRNT with the homologous enzootic ID and epizootic IC strains or with the strain contributing the PE2 gene for chimeric infections.
Virus titrations. Virus titers were determined by plaque assay using Vero cell monolayers in six-well plates as described previously (2).
Horses. Mixed-breed horses 1 to 12 years of age, obtained from a local horse seller, were placed into the animal biosafety level 3 (ABSL-3) large animal facility at Colorado State University and acclimated 3 to 7 days prior to infection. For the duration of the study, they were fed a pelleted ration plus a small amount of mixed grain twice daily and provided water ad libitum.
Infection of horses. Four horses each were infected with the wild-type VEEV strains EpA and EnA rescued from the infectious cDNA clones. These VEEV strains were previously shown to be indistinguishable from their parent strains in cell culture replication and mouse virulence (10). Three horses were infected with strain En/EpA-E2 containing the epizootic PE2 gene in the enzootic backbone, and two horses each were infected with the reciprocal chimeras Ep/EnA-E2 and Ep/EnB-E2. One horse was infected with the TrD/EnB-E2 chimera. All horses were inoculated subcutaneously in the left shoulder with 2,000 PFU in a volume of 1.0 ml of phosphate-buffered saline. Horses were bled daily from the jugular vein, and rectal temperatures were recorded twice daily. Horses exhibiting severe encephalitis signs and/or anorexia for 48 h were euthanized by an intravenous overdose of pentobarbital. All surviving horses were bled and sacrificed 14 to 18 days postinfection.
Statistical analysis. One-way analysis of variance with the Tukey-Kramer multiple-comparisons test and the Kruskal-Wallis test with Dunn's multiple-comparisons test were performed for comparisons of peak viremia magnitude and febrile response among virus strains using GraphPad InStat version 3.05 for Windows 95/NT (GraphPad Software, San Diego, CA.; www.graphpad.com).
|
|
|---|
![]() View larger version (27K): [in a new window] |
FIG. 2. Febrile responses of horses infected with epizootic strain EpA (four horses), enzootic strain EnA (four horses), and chimeric VEEV strains En/EpA-E2 (three horses), Ep/EnA-E2 (two horses), and TrD/EnB-E2 (one horse). Vertical bars represent standard errors of the means. Horizontal, dashed lines show the normal rectal temperature range for horses.
|
Viremia. As in previous studies using the parent virus (29), enzootic subtype ID strain EnA rescued from the cDNA clone generated no detectable viremia (<0.6 PFU/ml serum) in the four infected horses (Fig. 3). In contrast, the epizootic subtype IC strain EpA generated a peak mean viremia titer of 3.5 log10 PFU/ml, 2 days after infection, with a viremia duration of 4 days in three horses and 2 days in the other horse. The Ep/EnA-E2, Ep/EnB-E2, and TrD/EnB-E2 chimeras were phenotypically indistinguishable from the enzootic ID strain, with undetectable viremia, while the reciprocal En/EpA-E2 chimera produced a mean peak viremia (3.8 log10 PFU/ml) similar to EpA. Viremia generated by the En/EpA-E2 chimera lasted only 3 days in all three horses, in contrast to 4 days of viremia in 3/4 of the horses infected with strain EpA. The timing of viremia generally correlated with the febrile responses.
![]() View larger version (18K): [in a new window] |
FIG. 3. Mean viremia titers of horses infected with epizootic EpA (four horses) and chimeric VEEV strain En/EpA-E2 (three horses). Strains EnA (four horses), Ep/EnA-E2 (two horses), Ep/EnB-E2 (two horses), and TrD/EnB-E2 (one horse) produced no viremia detectable by plaque assay on Vero cells. Vertical bars represent standard errors of the means. The y-axis scale begins at the detection limit of the plaque assay (0.6 log10 PFU/ml).
|
|
|
|---|
Our data on the epizootic phenotypes of chimeric enzootic/epizootic VEEV strains provide additional support for the hypothesis that epizootic strains evolve periodically from enzootic progenitors via small numbers of mutations in the envelope glycoproteins. By swapping a genetic fragment encoding only seven amino acid differences in the E3 (1), E2 (5), and E1 (1) envelope glycoprotein genes, we generated the epizootic, equine amplification-competent phenotype from the genetic backbone of a Venezuelan subtype ID strain that is closely related to the phylogenetically predicted ancestor of strain EpA and all other epizootic VEEV strains. Although mutations outside this envelope glycoprotein gene region may contribute slightly to equine disease and viremia, such as the viremia duration to day 4 that was exhibited by most horses infected with the epizootic IC strain but not by those infected with the chimera, the envelope mutations included in our chimeric strains appear to be the major determinants of the epizootic phenotype, especially the viremia titers that are responsible for epidemic transmission via mosquito vectors. Envelope glycoprotein mutations have also been shown to mediate changes in the vector host range of VEEV (6, 7), and so determining whether the same mutations affect both mosquito and equine phenotypes will be of epidemiological interest.
Identification of which of these seven amino acid substitutions is critical to the epizootic phenotype awaits further studies using site-directed mutagenesis of the ID genome, which are under way in our laboratory. This information will be critical to not only to understanding the molecular genetic basis of the equine amplification and virulence phenotypes but also to predicting the frequency at which the epizootic mutants are generated in nature via the error-prone replication of RNA viruses (11), including alphaviruses (32). An estimation of this frequency will help to determine if VEE outbreaks are limited more by genetic constraints (an infrequent occurrence of epizootic mutants) or by ecological and epidemiological conditions conducive to efficient transmission among equines by mosquitoes.
Equine pathogenesis of VEEV. In general, VEEV infection of equines can result in (i) inapparent infection, followed by seroconversion; (ii) systemic disease characterized by tachycardia, fever, depression, and anorexia; or (iii) encephalitic disease that follows systemic disease and is often fatal (12, 16). The epizootic strains induce high-level viremia, and VEEV can be isolated from the blood as early as 1 to 2 days after experimental infection, indicating rapid initiation of productive replication. High-level viremia, fever, and lymphopenia correlate with the development of encephalitis (25; R.A.B., unpublished). Horses infected with both epizootic and enzootic strains develop neutralizing antibodies soon after the clearance of virus from the blood, and antibodies usually persist for at least months and probably years in survivors (25, 27). Preexisting VEEV neutralizing antibodies in horses provide protection from fatal VEE and suppress viremia, as shown in subsequent experimental challenge using equine-virulent strains (25).
In the past, only subtype IAB/C strains of VEEV have been implicated in major epizootics (30). These strains can induce high-level viremia and are usually neurovirulent in equines. The pathogenesis of epizootic and enzootic strains in equines is inadequately understood, and we attempted in our experiments to locate the genetic determinants responsible for virulence. Our results underscore the importance of the envelope glycoproteins in the generation of viremia and encephalitic disease. At the moment, we can only speculate about the mechanisms responsible for these phenotypes. However, in previous studies, we found no evidence for reduced replication of the enzootic strain or of chimeric viruses in vitro or in a rodent model (10). In addition, as previously reported the parental viral strains used in this study possess similar levels of resistance/sensitivity to type I murine interferon in vitro, indicating either the unimportance of this antiviral mechanism or the unsuitability of the murine model. Other hypotheses for the difference in enzootic versus epizootic strain pathogenesis, such as more efficient enzootic VEEV clearance from the blood of horses to reduce viremia or the differences in tissue tropism in equines, should be tested. Interestingly, positive-charge amino acid replacements (Arg, Lys) were implicated phylogenetically (4) and were contained in the epizootic fragments shown in this study (Table 1) to transform enzootic VEEV to the equine amplification-competent phenotype. These replacements are similar to E2 mutations known to attenuate alphaviruses, including VEEV, following selection for glycosaminoglycan binding during cell culture passage (3, 4, 8, 9). Understanding their effect on replication and pathogenesis in equines, which respond in a uniquely differential manner to epizootic versus enzootic VEEV strains, will be critical to elucidating the molecular mechanism of VEE emergence.
Present address: Center for Vector-borne Diseases and Department of Pathology, Microbiology and Immunology, School of Veterinary Medicine, University of California, Davis, CA 95616. ![]()
|
|
|---|
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»