Search

Stable neutralization (N) escape variants of Venezuelan equine encephalitis (VEE) virus were selected by anti-E2 glycoprotein monoclonal antibodies (MAbs) that neutralize viral infectivity, block viral hemagglutination, and passively protect mice. The nucleotide sequence of the E1, E2, and E3 genes of four variants revealed a clustering of single mutations in a domain spanning E2-182 to E2-207. The conformation of this short linear sequence affects antigenicity in the N domain because reduction and alkylation of virus disrupted binding of some E2 neutralizing MAbs. Serologic evidence for interaction of E2 epitopes also was obtained. Mutations in the N domain of VEE virus did not alter the kinetics of binding to Vero cells. They did, in some cases, produce attenuation of virulence in mice.

Attenuated mutants of Venezuelan equine encephalitis virus (VEE) were isolated by selection for rapid penetration of cultured cells (R. E. Johnston and J. F. Smith, 1988, Virology 162, 437-443). Sequence analysis of these mutants identified candidate attenuating mutations at four loci in the VEE E2 glycoprotein gene: a double mutation at E2 codons 3 and 4, and single substitutions at E2 76, 120, and 209. Each candidate mutation was reproduced in an isogenic recombinant VEE strain using site-directed mutagenesis of a full-length cDNA clone of VEE. Characterization of these molecularly cloned mutant viruses showed that mutation at each of the four loci in the E2 gene was sufficient to confer both the accelerated penetration and attenuation phenotypes. Inoculation of the molecularly cloned viruses into rodent models that differ in their response to VEE suggested that individual mutations affected different aspects of VEE pathogenesis. Full-length clones containing multiple mutations were produced by combining independently attenuating mutations. Molecularly cloned viruses carrying two or three mutations were more attenuated in sensitive animal models than viruses which contained any single mutation alone. However, these highly attenuated strains still retained the ability to induce an immune response sufficient to protect against a high dose challenge with virulent VEE. These results indicate that production of a molecularly cloned live virus vaccine for VEE is feasible.

A cDNA clone containing all of the 26 S mRNA coding region of the RNA genome of Venezuelan equine encephalitis (VEE) virus, virulent strain Trinidad donkey (TRD), has been constructed and sequenced. The nucleotide and deduced amino acid sequences of the 26 S RNA of VEE virus conform to the general organization of the alphavirus subgenomic mRNA. Excluding the poly(A) tail, the VEE 26 S RNA is 3913 nucleotides long with a protein coding region of 3762 nucleotides. Codon usage in the translated region is nonrandom and correlates well with that reported for Sindbis (SIN), Semliki Forest (SF), and Ross River (RR) alphaviruses. Highly conserved sequences of 19 to 22 nucleotides representing putative replicase recognition sites occur at the 26 S RNA junction region of the 42 S genomic RNA and at the 3' terminus immediately preceding the poly(A) tail. The conserved sequence at the 26 S/42 S junction region of VEE virus differs from that of other alphaviruses in that an ochre termination codon (UAA) is substituted for a GGU (Gly) codon present in the other viruses. The 5' and 3' noncoding regions (30 and 121 nucleotides, respectively) of the VEE 26 S RNA are shorter than has been reported for several other alphaviruses. The approximate transmembrane domains of the VEE E1 and E2 envelope glycoproteins have been identified. VEE E1 contains a single asparagine-linked glycosylation site, whereas E2 has three such sites, all of which are apparently glycosylated. The deduced amino acid sequence of the VEE polyprotein shows an overall homology of 44 to 46% with the precursor polyproteins of SIN, SF, and RR viruses. VEE virus capsid, E1, and E2 structural proteins show 43 to 46%, 50 to 53%, and 36 to 41% homology, respectively, with the cognate proteins of SIN, SF, and RR viruses.

Nucleotide sequence analysis of cDNA clones covering the entire genomes of Trinidad donkey (TRD) Venezuelan equine encephalitis (VEE) virus and its vaccine derivative, TC-83, has revealed 11 differences between the genomes of TC-83 virus and its parent. One nucleotide substitution and a single nucleotide deletion occurred in the 5′- and 3′- noncoding regions of the TC-83 genome, respectively. The deduced amino acid sequences of the nonstructural polypeptides of the two viruses differed only in a conservative Ser (TRD) to Thr (TC-83) substitution in nonstructural protein (nsP) three at amino acid position 260. The two silent mutations (one each in El and E2), one amino acid substitution in the El glycoprotein, and five substitutions in the E2 envelope glycoprotein of TC-83 virus were reported previously ( B. J. B. Johnson, R. M. Kinney, C. L. Kost, and D. W. Trent, 1986 , J. Gen. Viral . 67, 1951–1960). The genome of TRD virus was 11,444 nucleotides long with a 5′-noncoding region of 44 nucleotides. The carboxyl terminal portion of VEE nsP3 contained two peptide segments (7 and 34 amino acids long) that were repeated with high fidelity. The open reading frame of the nonstructural polyprotein was interrupted by an in-frame opal termination codon between nsP3 and nsP4, as has been reported for Sindbis, Ross River, and Middelburg viruses. The deduced amino acid sequences of the VEE TRD nsP1, nsP2, nsP3, and nsP4 polypeptides showed 60–66%, 57–58%, 35–44%, and 73-71% identity with the aligned sequences of the cognate polypeptides of Sindbis and Semliki Forest viruses, respectively. The lack of homology in the nsP3 of the viruses is due to sequence variation in the carboxyl terminal half of this polypeptide.

cDNA clones of the St. Louis encephalitis (SLE) virus genome have been obtained and the nucleotide sequence of 4.7 kb corresponding to the 5′ terminal half of the genome determined. The genome contains a 5′ noncoding region of 98 nucleotides followed by a single continuous open reading frame that encodes three structural proteins in the order capsid (C), membrane precursor (prM)-membrane (M), and envelope (E). Immediately following the C-terminus of E are located nonstructural proteins NS1 through NS3. The SLE amino acid sequence homology with yellow fever (YF), Murray Valley encephalitis (MVE), West Nile (WN), and dengue-2 (DEN) viruses over the sequenced region is 39, 66, 64, and 43%, respectively. The start of each SLE protein has been assigned on the basis of N-terminal sequence data and potential proteolytic cleavage sites homologous with YF and MVE viruses. Flaviviruses have conserved glycosylation sites in prM and NS1 proteins, although only one of the two glycosylation sites in the SLE E protein is conserved in MVE and DEN viruses. An evolutionary tree showing relationships of SLE, MVE, WN, YF, and DEN-2 flaviviruses is proposed on the basis of the amino acid sequences of the C proteins.