Genetic Conservation of Highlands J Viruses

Home , Culiseta, Highlands J virus, Robert Shope

VIROLOGY 218, 343–351 (1996) ARTICLE NO. 0203

View metadata, citation and similar papers at core.ac.uk brought to you by CORE ,1 ,2 MARTIN J. CILNIS,* WENLI KANG,† and SCOTT C. WEAVER† provided by Elsevier - Publisher Connector

*Department of Biology, University of California, San Diego, La Jolla, California 92093-0116; and †Center for Tropical Diseases and Department of Pathology, University of Texas Medical Branch, Galveston, Texas 77555-0605 Received November 8, 1995; accepted February 8, 1996

We studied molecular evolution of the mosquito-borne alphavirus Highlands J (HJ) virus by sequencing PCR products generated from 19 strains isolated between 1952 and 1994. Sequences of 1200 nucleotides including portions of the E1 gene and the 3؅ untranslated region revealed a relatively slow evolutionary rate estimated at 0.9–1.6 1 1004 substitutions per nucleotide per year. Phylogenetic trees indicated that all HJ viruses descended from a common ancestor and suggested the presence of one dominant lineage in North America. However, two or more minor lineages probably circulated simultane- ously for periods of years to a few decades. Strains isolated from a horse suffering encephalitis, and implicated in a recent turkey outbreak, were not phylogenetically distinct from strains isolated in other locations during the same time periods. Our findings are remarkably similar to those we obtained previously for another North American alphavirus, eastern equine encephalomyelitis virus, with which Highlands J shares primary mosquito and avian hosts, geographical distribution, and ecology. These results support the hypotheses that the duration of the transmission season affects arboviral evolutionary rates and vertebrate host mobility influences genetic diversity. ᭧ 1996 Academic Press, Inc.

INTRODUCTION in freshwater swamps by the mosquito vector Culiseta melanura (Hayes and Wallis, 1977). Enzootic transmis- Highlands J (HJ) virus is an arthropod-borne virus in the sion foci occur along the Atlantic seaboard of the United family Togaviridae, genus Alphavirus. HJ virus is grouped States, and the Gulf Coast as far west as Texas. Inland within the western equine encephalitis (WEE) antigenic foci also occur in Michigan and upstate New York. HJ complex, which also includes Sindbis, Fort Morgan, Aura, virus is ecologically similar if not identical to EEE virus in and WEE viruses (Calisher and Karabatsos, 1988; Cal- North America, sharing geographic distribution, primary isher et al., 1988). Like other alphaviruses, HJ virus has mosquito vector, and vertebrate amplifying hosts (Morris, a single-stranded plus-sense RNA genome of about 12 1988; Scott and Weaver, 1989). However, unlike EEE, HJ kilobases. Upon cell infection, genomic RNA is released virus is not believed to be pathogenic for humans or into the cytoplasm and serves as a template for the viral equines, with the exception of its implication in a 1964 polymerase to transcribe a minus-strand RNA intermedi- case of encephalitis in a Florida horse (Jennings et al., ate. From the minus-strand template, a full-length geno- 1966; Karabatsos et al., 1988). HJ virus was recently rec- mic RNA and subgenomic 26S RNA are synthesized. Ge- ognized as an important poultry pathogen; widespread nomic RNA is translated as a polyprotein to yield the infection of turkeys was reported in North Carolina during nonstructural proteins nsP1–4, while the subgenomic 1991 (Ficken et al., 1993), and HJ virus has been impli- 26S RNA, identical to the 3؅ one-third of the genome, cated in disease in a variety of other domestic avian encodes only the structural proteins (Strauss and species, including pheasants, chukar partridges, ducks, Strauss, 1994). emus, and whooping cranes (Guy et al., 1993, 1994a,b; The WEE complex and the eastern equine encephalo- Wages et al., 1993). myelitis (EEE) and Venezuelan equine encephalomyelitis Recent studies of New World alphaviruses indicate a (VEE) complexes are the major alphavirus serogroups high degree of genetic conservation and slower rates of found in the New World. HJ virus was originally catego- evolution than many non-arthropod-borne viruses rized as a subtype of WEE virus, but serologic tests and (Weaver, 1995; Weaver et al., 1992b, 1994). EEE virus oligonucleotide mapping later indicated that it is a dis- appears to evolve more slowly in North America than in tinct alphavirus species (Calisher et al., 1988; Trent and the tropics (Weaver et al., 1994). This suggests that the Grant, 1980). The virus has been isolated only in North amount of virus replication, related to the duration of the America, where it is transmitted among passerine birds transmission season and temperature in temperate vs tropical locations, may be one factor regulating arbovirus 1 Present address: Department of Molecular Biology and Biochemis- evolutionary rates. This hypothesis predicts that other try, University of California, Irvine, California 92717. alphaviruses restricted to temperate regions, such as 2 To whom reprint requests should be addressed. HJ virus, should evolve at rates comparable to North

0042-6822/96 $18.00 343 Copyright ᭧ 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.

AID VY 7833 / 6a14$$$$$1 03-29-96 14:34:45 vira AP: Virology 344 CILNIS, KANG, AND WEAVER

TABLE 1 Highlands J Viruses Sequenced

Strain Code Year of isolation Place of isolation Host Passage historya

L2-34A LA52 1952 Louisiana Mosquito ce3 B230 FL60 1960 Florida Bird p5, sm2 WX3-2AP TX63 1963 Texas Bird sm1 64A-1519 FL64 1964 Florida Horse p4, sm1 NCJ5-6X NC65 1965 North Carolina Mosquito de1 MW8-5AD MD68 1968 Maryland Mosquito p2, sm1 SP72-666 CT72 1972 Connecticut Mosquito p2 73V2540 MA73 1973 Massachusetts Mosquito p2, sm1 B-8-74 CT74 1974 Connecticut Bird C6/36-1 78-3331 MA78 1978 Massachusetts Mosquito C6/36-1 WC-431 NJ81 1981 New Jersey Bird C6/36-1 83-230 MA83 1983 Massachusetts Mosquito Unpassaged 111-84 CT84 1984 Connecticut Mosquito sm2 3-81690 RI90 1990 Rhode Island Mosquito v1 R362 NC92 1992 North Carolina Turkey v1 F25 RI92 1992 Rhode Island Mosquito v1 93-504 MA93 1993 Massachusetts Mosquito Unpassaged RV94-204 NJ94 1994 New Jersey Mosquito ce1, de1 M32 RI94 1994 Rhode Island Mosquito v1

a C6/36, mosquito cell; ce, chick cell; de, duck embryo cell; p, undetermined passage; sm, suckling mouse; v, Vero cell.

American EEE virus, estimated at 1.6 1 1004 substitu- RNA–primer mixture was heated to 65Њ for 1 min, fol- tions per nucleotide per year. The similarity of the HJ and lowed by gradual cooling to 23Њ. cDNA was synthesized EEE transmission cycles also predicts that patterns of using 200 units of Superscript reverse transcriptase virus dispersal (gene flow), influencing genetic diversity, (BRL) according to the manufacturer’s protocol, with incu- should be comparable. To test these hypotheses, we bation at 38Њ for 30 min. For PCR amplification, 300 ng sequenced PCR products from representative HJ virus each of the anti-sense primer and a sense primer of strains isolated from 1952 to 1994. Rates and patterns sequence 5؅-TACCCNTTYATGTGGGGW-3؅, designed to of evolution estimated from phylogenetic analyses were anneal to a conserved alphavirus genome region homol- similar those reported previously for EEE virus, support- ogous to nucleotides 10247–10264 of EEE virus (Weaver ing these hypotheses. et al., 1993), were added. The reaction volume was increased to 100 ml with the addition of PCR buffer (10 mM

MATERIALS AND METHODS KCl, 10 mM (NH4)2SO4 ,20mMTris (pH 8.8), 2 mM MgSO , 0.1% Triton X-100, 0.1 mg/ml BSA) and 1.25 U Virus preparation and PCR amplification 4 Taq polymerase (Stratagene). This PCR mix was placed The 19 HJ virus strains we analyzed are listed in Table in a thermal cycler for 30 amplification cycles as follows: 1. Virus stocks were prepared on BHK-21 cell culture heat-denaturation at 95Њ for 30 sec, primer annealing at monolayers at 37Њ with a multiplicity of infection of 0.1 44Њ for 30 sec, and extension at 72Њ for 2 min. to 1.0 PFU per cell. After cytopathic effects were evident, viral RNA was extracted from 100 ml of the cell culture DNA sequencing and analyses supernatant using Trizol (BRL Laboratories), a phenol, guanidine isothiocyanate monophasic solution, ac- Direct dideoxynucleotide sequencing of PCR products cording to the manufacturer’s protocol. Two micrograms was performed using the Cyclist Exo(0) Pfu DNA Se- each of tRNA and glycogen were added as carriers dur- quencing kit (Stratagene) according to the manufacturer’s ing precipitation. Following centrifugation, the precipi- protocol or an Applied Biosystems 377 automated se- tated RNA pellet was washed with 70% ethanol, dried in quencer using the Prism Dyedeoxy chemistry. The 11 vacuum, and resuspended in 1 ml RNasin ribonuclease DNA primers used are listed in Table 2. The nucleotide inhibitor (Promega) and 9 ml5mMTris, 0.1 mM EDTA sequences for 19 strains of HJ virus were aligned with (pH 8.0) buffer. The RNA was mixed with 100 ng of an homologous sequences of o’nyong-nyong (Levinson et -antisense primer of sequence 5؅-GAAATTTTAAAAACA- al., 1990), Ross River (Faragher et al., 1988), Semliki For -AAATA-3؅ designed to anneal to the 3؅ end of the HJ est (Garoff et al., 1980), chikungunya (M. D. Parker, un genome adjacent to the poly(A) tail, according to the published, GenBank Accession No. L37661), Sindbis previously published 3؅ sequence (Ou et al., 1983). The (Strauss et al., 1984), EEE (Weaver et al., 1993), WEE

AID VY 7833 / 6a14$$$$$2 03-29-96 14:34:45 vira AP: Virology GENETIC CONSERVATION OF HIGHLANDS J VIRUSES 345

TABLE 2 among the different alphaviruses examined is presented Oligonucleotide Primers Used for Sequencing in Fig. 2. All HJ viruses formed a monophyletic group supported by a bootstrap value of 100%, indicating that -Sequence (5؅–3؅) Genetic sense they descended from a common ancestor. All other rela tionships among alphaviruses were consistent with our TACCCNTTYATGTGGGGW Plus previous phylogenetic studies (Rico-Hesse et al., 1995; ATCGGTAAAGAGCATTCACG Plus GGAGCATCATCCCTC Plus Weaver et al., 1992a, 1993, 1994). As indicated by sero- CAACGATGTCTGTGGAATC Minus logical relationships (Calisher et al., 1988), the closest AACTCATTTCGAGCACGGT Minus relatives of HJ virus were other members of the WEE AGAGGAGTGCGAATGGACAG Minus antigenic complex; most closely related was WEE, fol- GATAATCATAGAGCTGCAGG Minus lowed by the Old World WEE complex members Sindbis TTATGCACCACGCTTCCT Minus GAAATTTTAAAAACAAAATA Minus and Ockelbo viruses, and the New World Aura virus. A parsimony tree showing relationships among the 19 HJ virus isolates is presented in Fig. 3. Within the HJ (Hahn et al., 1988), Middelburg (Strauss et al., 1983), Aura group, relationships partially but not fully reflected the (Rumenapf et al., 1995), and VEE antigenic subtypes IAB year of isolation of the strains examined. For example, (Kinney et al., 1986), IC, ID (Kinney et al., 1992), IE the basal two strains were isolated in 1963 and 1952, (Sneider et al., 1993), and II (Sneider et al., 1993) viruses respectively, while the most recent isolates were found using the PILEUP program of the Genetics Computer in terminal branches. This topology, as well as the overall Group (Devereux et al., 1984). The complete homologous genetic conservation we observed, indicated that a sin- sequence of WEE virus was included in the phylogenetic gle, predominant lineage of HJ virus existed during the analyses, while only E1 nucleotides homologous to posi- past 4 decades in most regions sampled. However, the tions 10267–11271 of the EEE virus genome (Weaver et basal position of the 1963 Texas isolate, followed by al., 1993) were included for the other alphaviruses be- the 1952 Louisiana strain, suggested that two or more cause of uncertainty in alignments in the C-terminal end independent lineages occurred during the earlier time of E1 and the 3؅ untranslated region. Aligned sequences period. Forcing the TX63 strain into the remaining group underwent phylogenetic analysis using the PAUP parsi- (leaving LA52 in the basal position) increased the tree mony program (Swofford, 1991) and the NEIGHBOR length by two or more steps. Because no other HJ virus neighbor-joining program implemented in the PHYLIP isolates were available from Texas, we were unable to package (Felsenstein, 1993). Parsimony analysis was im- determine if or how long a distinct lineage or genotype plemented using the heuristic algorithm with unordered may have persisted there. Some other terminal group- characters, and sequences were added at random with ings, such as CT84-NC92-NJ94, CT72-MA73-CT74-NJ81, 100 replications. The ACCTRANS option was used for and MA78-MA83-MA93-RI90-RI92-RI94, which excluded parsimony character state optimization. The one-parame- other isolates from the same time period, also suggested ter formula of Jukes and Cantor (1969) was used to gener- simultaneous circulation multiple HJ virus lineages for ate the distance matrix for neighbor-joining analysis. limited time periods. The nearly consistent grouping of Bootstrap analyses with 200 resamplings were used to isolates from Massachusetts and Rhode Island, as well place confidence values on groupings within trees. The as New Jersey and Connecticut (Fig. 3), indicated that MacClade program (Maddison and Maddison, 1992) was some lineages may have been regionally defined and used to trace character changes within parsimony trees. independent for periods of a decade or more. However, our limited collection of viruses precluded determining RESULTS with certainty whether cocirculating lineages were geo- Aligned sequences of the 19 strains of HJ virus, 1200 graphically restricted. These patterns of limited genetic nucleotides in length, are presented in Fig. 1. The se- diversity are very similar to those we reported previously quences were remarkably conserved, with maximum nu- for EEE virus (Weaver et al., 1994). cleotide divergence of 2.2% and E1 amino acid diver- HJ virus strain 64A-1519 (Fig. 3; FL64), isolated from gence of 1.7%. Among alphaviruses, HJ sequences were the brain of a Florida horse suffering from encephalitis most similar to that of WEE virus (ca. 78% nucleotide in 1964 (Jennings et al., 1966; Karabatsos et al., 1988), identity, 89% E1 amino acid identity) and least similar to was genetically similar to other isolates from the 1960s those of o’nyong-nyong and chikungunya viruses (ca. (Fig. 3). Strain R362, isolated during a North Carolina 50% nucleotide identity, 42% E1 amino acid identity). Par- turkey outbreak in 1992, grouped with strains from Con- simony analysis of HJ and homologous alphavirus se- necticut and New Jersey isolated during the past decade. quences yielded one tree of minimum (2671 steps) Based on the genome portion analyzed, our phylogenetic length. The neighbor-joining tree showed identical rela- analysis yielded no evidence that HJ viruses responsible tionships. A parsimony tree depicting relationships for equine or poultry disease belong to a genotype dis-

AID VY 7833 / 6a14$$$$$2 03-29-96 14:34:45 vira AP: Virology 346

03-29-96 14:34:45 vira AP: Virology AID VY 7833 / 6a14$$7833 GENETIC CONSERVATION OF HIGHLANDS J VIRUSES 347

FIG. 1. Aligned nucleotide sequences for 19 strains of Highlands J virus described in Table 1. Amino acid abbreviations are listed above the second position of codons for the E1 coding region.

tinct from that of viruses circulating in other regions of ter reconstructions). Of the 39 changes we analyzed, 28 North America during the same time period. were transitions and 11 were transversions, yielding a Nucleotide substitutions accompanying HJ virus evolu- ratio of 2.5:1. Within the E1 coding region, 26/34 substitution were examined by comparing ancestral sequences, tions were third codon position, 2 were first position, and predicted by the parsimony analysis, with terminal taxa. 6 were second position. The HJ clade defined by the hypothetical ancestor at We also examined nucleotide substitutions maintained node A (Fig. 3), composed of all isolates except the 1963 in the dominant lineage by categorizing changes that Texas strain, was analyzed. This clade was selected be- distinguished the hypothetical ancestral sequence of this cause all predicted ancestral sequences (internal nodes) HJ group from those of all strains isolated during the within this clade, as well as nucleotide substitutions rep- 1990s. Of 16 such substitutions, 13 were transitions (ratio resented in its branches, were unambiguous (the node of 4.3:1) and all were third codon position, synonymous representing the ancestor of the entire HJ virus clade substitutions within the E1 coding region. These data had several ambiguities representing alternative charac- indicate strong selection for conservation of the E1 amino

AID VY 7833 / 6a14$$$$$3 03-29-96 14:34:45 vira AP: Virology 348 CILNIS, KANG, AND WEAVER

FIG. 2. Parsimony tree of alphaviruses. All Highlands J virus strains FIG. 4. Regression analysis of the average rate of evolution (nucleo- fell into a monophyletic group shown in greater detail in Fig. 3. Numbers tide substitution) for Highlands J viruses isolated from 1952 to 1994. indicate bootstrap values of 50% or greater for groups to the right in Branch lengths separating each isolate from the hypothetical ancestor the tree. of the HJ group, expressed as % divergence, were plotted vs year of isolation. The slope, 0.009%/year, is an estimate of the average rate of .acid sequence. Although the 3؅ untranslated region com- evolution prised 13% of the sequence region examined, none of the substitutions that occurred in this portion of the ge- Weaver et al., 1994). Each isolate within the HJ group was nome were maintained in the dominant lineage. This plotted by year of isolation and percent divergence vs the dearth of substitutions within the 3 untranslated region ؅ hypothetical ancestor of the entire HJ clade (Fig. 4). The may reflect selection for conservation of primary or sec- regression yielded an estimated evolutionary rate of ondary RNA structures involved in alphavirus genome 0.009%/year or 9 1 1005 substitutions/nucleotide/year. If replication (Strauss and Strauss, 1994). multiple HJ virus lineages were cocirculating during this We estimated the rate of nucleotide substitution within time period, this rate would represent an average for those the HJ virus genome portion examined using the regression lineages. The data indicated no obvious change in the method described previously (Buonagurio et al., 1986; evolutionary rate during the 1952–1994 time period examined. We also estimated the evolutionary rate by totaling substitutions (branch lengths) between the most recent (RI94, NJ94) isolates and node A (Fig. 3). Node A was selected over the ancestor of the entire HJ group because the branch separating it from the LA52 isolate was relatively short (two steps), suggesting that it occurred not long before 1952. The total branch lengths separating this node from the 1990s isolates were 7 and 8 substitutions, respectively, or 0.58 and 0.67% divergence. Assuming that the ancestor occurred in 1952 (the year of isolation for the basal isolate in this group), we obtained an estimated rate of 0.014–0.016%/year or 1.4–1.6 1 1004 substitutions/nucleotide/year. This rate must be considered a maximum estimate because the hypothetical ancestor may have occurred before 1952.

DISCUSSION Evolutionary rate Our results underscore the genetic stability and se- FIG. 3. Phylogenetic tree showing relationships among 19 strains of quence homogeneity of HJ viruses. Because the non- HJ virus. Numbers indicate bootstrap values of 50% or greater for groups to the right in the tree. Node A indicates the hypothetical ances- structural genes of alphaviruses are more highly con- tor of all of the viruses except the 1963 Texas isolate. Codes for viruses served than the structural genes (Strauss and Strauss, are defined in Table 1. 1994), the genomic evolutionary rate may be lower than

AID VY 7833 / 6a14$$$$$3 03-29-96 14:34:45 vira AP: Virology GENETIC CONSERVATION OF HIGHLANDS J VIRUSES 349

-our estimate for the E1/3؅UTR region. The majority of WEE, and HJ primarily infect birds. Birds are presum substitutions we observed within the E1 gene were syn- ably more efficient than rodents in disseminating vi- onymous, consistent with strong selection for the current ruses among isolated transmission foci. Efficient virus amino acid sequence of the E1 envelope glycoprotein. dissemination may constrain alphavirus evolution by These indications of strong selection support the hypoth- enhancing genome exchange (gene flow), effectively esis that arbovirus evolution is constrained by the dual minimizing founder effects and genetic drift (Weaver, selective pressures imposed by obligate alternation be- 1995; Weaver et al., 1992b). Dissemination may also tween vertebrate and invertebrate hosts (Weaver et al., enhance competition among cocirculating genotypes, 1992b). facilitating competitive exclusion. Relatively slow evolutionary rates in the range 1.6–5 04 1 10 substitutions/nucleotide/year have been esti- HJ virus maintenance mated for several other New World alphaviruses including EEE, WEE, and VEE. Our estimated HJ substitution The source of HJ and EEE viruses that initiate the rate of 0.9–1.6 1 1004 substitutions/nucleotide/year is annual transmission cycle in temperate locations of similar to that of EEE virus in North America (Weaver et North America is unknown. Transovarial transmission al., 1994), but lower than those for EEE, VEE, and WEE (TOT) has been reported for the alphaviruses Ross River viruses that circulate in South and Central America. This (Kay, 1982), WEE (Fulhorst et al., 1994), and Sindbis (Dhi- supports the hypothesis that the abbreviated alphavirus leepan et al., 1996). However, TOT of EEE and HJ viruses transmission season and lower temperatures in temper- has not been demonstrated experimentally (Sprance, ate regions slow annual rates of alphavirus evolution 1981) or supported by consistent field isolation of virus (Weaver, 1995; Weaver et al., 1992b). The lack of appre- from male mosquitoes or larvae (Hayes and Wallis, 1977; ciable change in the rate of HJ virus evolution from 1952 Morris, 1988). Alternatively, these viruses could overwin- to 1994 is inconsistent with the hypothesis that declines ter within resident birds that become persistently in- in passerine bird populations are responsible for the fected in the summer or fall and become or remain vire- changes in EEE virus evolutionary rates we reported pre- mic during the following spring. Although field studies of viously (Weaver et al., 1994). EEE virus provide circumstantial evidence for this hypothesis (Crans et al., 1994), experimental infections are Genetic diversity needed to determine if birds become persistently infected with EEE and/or HJ viruses. The isolation of WEE We obtained evidence that two or more distinct, coc- virus from experimentally infected birds up to 10 months irculating genotypes of HJ virus were maintained for after inoculation suggests that the chronic avian infec- several years or a few decades. However, the virus tions could serve as an alphavirus overwintering mecha- appears to evolve over longer time frames as a single, nism (Reeves et al., 1958). predominant lineage. This genetic conservation and A third hypothesis for transmission initiation, the an- low diversity are consistent with studies of other North nual reintroduction of viruses by migrating birds from American alphaviruses including EEE (Weaver et al., subtropical foci, is not supported by the regional basis 1994) and WEE (Trent and Grant, 1980). In contrast, of some northeastern groupings in our HJ virus trees; EEE (Weaver et al., 1994) and VEE viruses (Weaver these groupings suggest virus maintenance in temperate et al., 1992b) are more diverse in South and Central foci throughout the year, as was also indicated for EEE America. virus (Weaver et al., 1994). However, our limited sampling Possible factors influencing the diversity of alphavi- of viruses precludes ruling out the more complex expla- ruses have been reviewed previously (Weaver, 1995; nation that HJ virus is annually reintroduced from sub- Weaver et al., 1992b). One hypothesis, that the diver- tropical locations in a geographically specific manner. sity of vector taxa influences arbovirus diversity, is not Migratory birds are less likely to be infected with EEE entirely supported by accumulating data. EEE and HJ virus than summer and permanent residents of New Jer- viruses are transmitted by mosquitoes in the genus sey (Crans et al., 1994). This finding supports the hypoth- Culiseta, a species-poor taxon in North America. How- esis that movement of EEE and HJ viruses among geo- ever, WEE virus, which also displays low diversity and graphic regions, allowing for competitive displacement genetic conservation (Trent and Grant, 1980), utilizes of cocirculating genotypes, occurs only occasionally. If the more diverse genera Culex and Aedes (Aviles et HJ virus is maintained in temperate transmission foci al., 1992; Reisen and Monath, 1988). Another hypothe- through the winter, limited dispersal could be provided sis, that host mobility influences arbovirus diversity, by postnesting avian wandering, as was suggested for is better supported. Whereas the most diverse New EEE virus (Weaver et al., 1994). This kind of limited dis- World alphaviruses, VEE and South American variety persal and genetic mixing of HJ viruses may account for EEE viruses, utilize rodent hosts in tropical locations, the small amount of genetic diversity we observed. More less diverse viruses such as North American EEE, detailed phylogenetic studies, examining more isolates

AID VY 7833 / 6a14$$$$$3 03-29-96 14:34:45 vira AP: Virology 350 CILNIS, KANG, AND WEAVER sampled throughout the geographic range of these vi- Hahn, C. S., Lustig, S., Strauss, E. G., and Strauss, J. H. (1988). Western ruses, are needed to further delineate these patterns of equine encephalitis virus is a recombinant virus. Proc. Natl. Acad. Sci. USA 85, 5997–6001. alphavirus dispersal and maintenance. Hayes, C. G., and Wallis, R. C. (1977). Ecology of Western equine encephalomyelitis in the eastern United States. Adv. Virus Res. 21, 37– ACKNOWLEDGMENTS 83. Jennings, W. L., Allen, R. H., and Lewis, A. L. (1966). Western equine We are grateful to Charles Calisher, Yvonne Gonzalez Chase, James encephalomyelitis in a Florida horse. Am. J. Trop. Med. Hyg. 15, 96– Guy, Nick Karabatsos, Robert Shope, Wayne Pizzuti, Barbara Taylor, 97. Robert Tesh, and Dennis Trent for supplying HJ virus isolates. John Jukes, T. H., and Cantor, C. R. (1969). Evolution of protein molecules. Holland provided valuable advice with the research and manuscript, In ‘‘Mammalian Protein Metabolism’’ (H. N. Munro, Eds.), pp. 21–132. and Estelle Bussey provided excellent technical assistance. This work Academic Press, New York. was supported by Grants AI 10984 and AI14627 from the NIH. Karabatsos, N., Lewis, A. L., Calisher, C. H., Hunt, A. R., and Roehrig, J. T. (1988). Identification of Highlands J virus from a Florida horse. REFERENCES Am. J. Trop. Med. Hyg. 19, 603–606. Kay, B. H. (1982). Three modes of transmission of Ross River virus by Aviles, G., Sabattini, M. S., and Mitchell, C. J. (1992). Transmission of Aedes vigilax (Skuse). Aust. J. Exp. Biol. Med. Sci. 60, 339–344. western equine encephalomyelitis virus by Argentine Aedes albifas- Kinney, R. M., Johnson, B. J. B., Brown, V. L., and Trent, D. W. (1986). ciatus (Diptera: Culicidae). J. Med. Entomol. 29, 850–853. Nucleotide sequence of the 26S mRNA of the virulent Trinidad don- Buonagurio, D. A., Nakada, S., Parvin, J. D., Krystal, M., Palese, P., and key strain of Venezuelan equine encephalitis virus and deduced Fitch, W. M. (1986). Evolution of human influenza A viruses over 50 sequence of the encoded proteins. Virology 152, 400–413. years: Rapid, uniform rate of change in NS gene. Science 232, 980– Kinney, R. M., Tsuchiya, K. R., Sneider, J. M., and Trent, D. W. (1992). 982. Genetic evidence that epizootic Venezuelan equine encephalitis Calisher, C. H., and Karabatsos, N. (1988). Arbovirus serogroups: Defini- (VEE) viruses may have evolved from enzootic VEE subtype I-D virus. tion and geographic distribution. In ‘‘The Arboviruses: Epidemiology Virology 191, 569–580. and Ecology’’ (T. P. Monath, Ed.), Vol. I, pp. 19–57. CRC Press, Boca Levinson, R. S., Strauss, J. H., and Strauss, E. G. (1990). Complete Raton, FL. sequence of the genomic RNA of o’nyong-nyong virus and its use Calisher, C. H., Karabatsos, N., Lazuick, J. S., Monath, T. P., and Wolff, in the construction of alphavirus phylogenetic trees. Virology 175, K. L. (1988). Reevaluation of the western equine encephalitis anti- 110–123. genic complex of alphaviruses (family Togaviridae) as determined Maddison, W. P., and Maddison, D. R. (1992). ‘‘MacClade Version 3.’’ by neutralization tests. Am. J. Trop. Med. Hyg. 38, 447–452. Sinauer Associates, Inc., Sunderland, MA. Crans, W. J., Caccamise, D. F., and McNelly, J. R. (1994). Eastern equine Morris, C. D. (1988). Eastern equine encephalomyelitis. In ‘‘The Arbovi- encephalomyelitis virus in relation to the avian community of a ruses: Epidemiology and Ecology’’ (T. P. Monath, Ed.), Vol. III, pp. 1– coastal cedar swamp. J. Med. Entomol. 31, 711–728. 36. CRC Press, Boca Raton, FL. Devereux, J., Haeberli, P., and Smithies, O. (1984). A comprehensive Ou, J.-H., Trent, D. W., and Strauss, J. H. (1983). The 3؅-non-coding set of sequence analysis programs for the VAX. Nucleic Acids Res. regions of alphavirus RNAs contain repeating sequences. J. Mol. 12, 387–395. Biol. 168, 1–15. Dhileepan, K., Azoulas, J. K., and Gibson, C. A. (1996). Evidence of Reeves, W. C., Hutson, G. A., Bellamy, R. E., and Scrivani, R. P. (1958). vertical transmission of Ross River and Sindbis viruses (Togaviridae: Chronic latent infections of birds with western equine encephalomy- Alphavirus) by mosquitoes (Diptera: Culicidae) in southeastern Aus- elitis virus. Proc. Soc. Exp. Med. 97, 733–736. tralia. J. Med. Entomol. 33, 180–182. Reisen, W. K., and Monath, T. P. (1988). Western equine encephalomy- Faragher, S. G., Meek, A. D. J., Rice, C. M., and Dalgamo, L. (1988). elitis. In ‘‘The Arboviruses: Epidemiology and Ecology’’ (T. P. Monath, Genome sequences of a mouse-avirulent and a mouse-virulent strain Ed.), Vol. V, pp. 89–137. CRC Press, Boca Raton, FL. of Ross River virus. Virology 163, 509–526. Rico-Hesse, R., Weaver, S. C., Siger, J. d., Medina, G., and Salas, R. A. Felsenstein, J. (1993). ‘‘PHYLIP (Phylogeny Inference Package) version (1995). Emergence of a new epidemic/epizootic Venezuelan equine 3.5p.’’ Distributed by the author. Department of Genetics, University encephalitis virus in South America. Proc. Natl. Acad. Sci. USA 92, of Washington, Seattle. 5278–5281. Ficken, M. D., Wages, D. P., Guy, J. S., Quinn, J. A., and Emory, W. H. Rumenapf, T., Strauss, E. G., and Strauss, J. H. (1995). Aura virus is a (1993). High mortality of domestic turkeys associated with Highlands New World representative of Sindbis-like viruses. Virology 208, 621– J virus and eastern equine encephalitis virus infections. Avian Dis. 633. 37, 585–590. Fulhorst, C. F., Hardy, J. L., Eldridge, B. F., Presser, S. B., and Reeves, Scott, T. W., and Weaver, S. C. (1989). Eastern equine encephalomyelitis W. C. (1994). Natural vertical transmission of western equine enceph- virus: Epidemiology and evolution of mosquito transmission. Adv. alomyelitis virus in mosquitoes. Science 263, 676–678. Virus Res. 37, 277–328. Garoff, H., Frischauf, A. M., Simons, K., Lehrach, H., and Delius, H. Sneider, J., Kinney, R., Tsuchiya, K., and Trent, D. (1993). Molecular (1980). Nucleotide sequence of cDNA coding for Semlike Forest virus evidence that epizootic Venezuelan equine encephalitis (VEE) I-AB membrane glycoproteins. Nature 288, 236–241. viruses are not evolutionary derivatives of enzootic VEE subtype I-E Guy, J. S., Barnes, H. J., Ficken, M. D., Smith, L. G., Emory, W. H., or II viruses. J. Gen. Virol. 74, 519–523. and Wages, D. P. (1994a). Decreased egg production in turkeys Sprance, H. E. (1981). Experimental evidence against the transovarial experimentally infected with eastern equine encephalitis virus or transmission of eastern equine encephalitis virus in Culiseta mela- Highlands J virus. Avian Dis. 38, 563–571. nura. Mosq. News 41, 168. Guy, J. S., Barnes, H. J., and Smith, J. G. (1994b). Experimental infection Strauss, E. G., Rice, C. M., and Strauss, J. H. (1983). Sequence coding of young broiler chickens with eastern equine encephalitis virus and for the alphavirus nonstructural proteins is interrupted by an opal Highlands J virus. Avian Dis. 38, 572–582. termination codon. Proc. Natl. Acad. Sci. USA 80, 5271–5275. Guy, J. S., Ficken, M. D., Barnes, H. J., Wages, D. P., and Smith, J. G. Strauss, E. G., Rice, C. M., and Strauss, J. H. (1984). Complete nucleotide (1993). Experimental infection of young turkeys with eastern equine sequence of the genomic RNA of Sindbis virus. Virology 133, 92– encephalitis virus and highlands J virus. Avian Dis. 37, 389–395. 110.

AID VY 7833 / 6a14$$$$$4 03-29-96 14:34:45 vira AP: Virology GENETIC CONSERVATION OF HIGHLANDS J VIRUSES 351

Strauss, J. H., and Strauss, E. G. (1994). The alphaviruses: Gene expres- Holland, J. J., and Scott, T. W. (1994). Evolution of alphaviruses in sion, replication, and evolution. Microbiol. Rev. 58, 491–562. the eastern equine encephalomyelitis complex. J. Virol. 68, 158– Swofford, D. L. (1991). ‘‘PAUP: Phylogenetic Analysis Using Parsimony, 169. Version 3.0.’’ Illinois Natural History Survey, Champaign, IL. Weaver, S. C., Bellew, L. A., and Rico-Hesse, R. (1992a). Phylogenetic Trent, D. W., and Grant, J. A. (1980). A comparison of New World alphavi- analysis of alphaviruses in the Venezuelan equine encephalitis com- ruses in the western equine encephalomyelitis complex by immuno- plex and identification of the source of epizootic viruses. Virology chemical and oligonucleotide fingerprint techniques. J. Gen. Virol. 191, 282–290. 47, 261–282. Weaver, S. C., Hagenbaugh, A., Bellew, L. A., Netesov, S. V., Volchkov, Wages, D. P., Ficken, M. D., Guy, J. S., Cummings, T. S., and Jennings, V. E., Chang, G.-J. J., Clarke, D. K., Gousset, L., Scott, T. W., Trent, S. R. (1993). Egg-production drop in turkeys associated with alphavi- D. W., and Holland, J. J. (1993). A comparison of the nucleotide se- ruses: Eastern equine encephalitis virus and Highlands J virus. Avian quences of eastern and western equine encephalomyelitis viruses Dis. 37, 1163–1166. with those of other alphaviruses and related RNA viruses. Virology Weaver, S. C. (1995). Evolution of alphaviruses. In ‘‘Molecular Basis of 197, 375–390. Virus Evolution’’ (A. J. Gibbs, C. H. Calisher, and F. Garcia-Arenal, Weaver, S. C., Rico-Hesse, R., and Scott, T. W. (1992b). Genetic diversity Eds.), pp.501–530. Cambridge Univ. Press, Cambridge, UK. and slow rates of evolution in New World alphaviruses. Curr. Top. Weaver, S. C., Bellew, L. A., Hagenbaugh, A., Mallampalli, V., Microbiol. Immunol. 176, 99–117.

AID VY 7833 / 6a14$$$$$4 03-29-96 14:34:45 vira AP: Virology