Proc. Natl. Acad. Sci. USA Vol. 85, pp. 5997-6001, August 1988 Evolution Western equine encephalitis virus is a recombinant virus (RNA recombination/Alphavirus/evolution of RNA viruses) CHANG S. HAHN, SHLOMO LUSTIG*, ELLEN G. STRAUSS, AND JAMES H. STRAUSSt Division of Biology, 156-29, California Institute of Technology, Pasadena, CA 91125 Communicated by James Bonner, April 14, 1988 ABSTRACT The alphaviruses are a group of 26 mosquito- virus; O'Nyong-nyong virus; and Ross River virus. Sindbis borne viruses that cause a variety of human diseases. Many of and Semliki Forest viruses have been intensively studied as the New World alphaviruses cause encephalitis, whereas the models for alphavirus replication (7). Sindbis virus is widely Old World viruses more typically cause fever, rash, and distributed, being found in Europe, India, southeast Asia, arthralgia. The genome is a single-stranded nonsegmented Australia, and Africa. Close relatives of this virus, such as RNA molecule of + polarity; it is about 11,700 nucleotides in Ockelbo virus in Europe (8) and Babanki virus in Africa, length. Several alphavirus genomes have been sequenced in cause disease in humans characterized by fever, rash, and whole or in part, and these sequences demonstrate that alpha- arthritis. Chikungunya and O'Nyong-nyong viruses have viruses have descended from a common ancestor by divergent caused large epidemics in Africa of a dengue-like disease also evolution. We have now obtained the sequence of the 3'- characterized by fever, rash, and arthralgia. Ross River virus terminal 4288 nucleotides of the RNA of the New World is the causative agent of epidemic polyarthritis in Australia Alphavirus western equine encephalitis virus (WEEV). Com- and the South Pacific. parisons of the nucleotide and amino acid sequences of WEEV Complete or partial RNA sequences have been obtained for with those of other alphaviruses clearly show that WEEV is Sindbis virus (9), Semliki Forest virus (10-12), Ross River recombinant. The sequences of the capsid protein and of the virus (13), EEEV (14), and VEEV (15). Comparison of these (untranslated) 3'-terminal 80 nucleotides of WEEV are closely nucleotide sequences and their encoded amino acid sequences related to the corresponding sequences of the New World has demonstrated that the alphaviruses are related by linear Alphavirus eastern equine encephalitis virus (EEEV), whereas descent from a common ancestor (7). The relationships found the sequences of glycoproteins E2 and El of WEEV are more are compatible, for the most part, with those derived from closely related to those of an Old World virus, Sindbis virus. studies of serological cross-reactivity, which depends only Thus, WEEV appears to have arisen by recombination between upon antigenic epitopes in the structural proteins. In serolog- an EEEV-like virus and a Sindbis-like virus to give rise to a new ical studies, however, WEEV has always been something of a virus with the encephalogenic properties of EEEV but the puzzle. It is a New World virus that often causes encephalitis, antigenic specificity of Sindbis virus. There has been specula- but serologically it is most closely related to Sindbis virus, an tion that recombination might play an important role in the Old World alphavirus not normally associated with encepha- evolution of RNA viruses. The current finding that a wide- litis. To explore the relationship of WEEV to other alphavi- spread and successful RNA virus is recombinant provides ruses, we have obtained the sequence of the 3'-terminal 4288 support for such an hypothesis. nucleotides of the WEEV genomet and found that WEEV appears to have arisen by recombination between an EEEV- The 26 members of the Alphavirus genus of the family like virus and a Sindbis-like virus. Togaviridae are mosquito-borne viruses that form an impor- tant group of disease agents (1-3). The New World alphavi- MATERIALS AND METHODS ruses include western equine encephalitis virus (WEEV) and Virus RNA Preparation. WEEV RNA [strain BFS1703, eastern equine encephalitis virus (EEEV), both of which are isolated from Cx. tarsalis in July 1953 in Kern County, capable ofcausing encephalitis in humans and causing severe California (16)] was obtained from Mark Stanley and James disease in horses. WEEV has a wide geographic distribution, Hardy (University of California, Berkeley). The virus had being found from western Canada to Mexico and, discontin- been passed twice by i.c. inoculation of suckling mice and uously, to Argentina. WEEV is transmitted in the western four times (including three plaque isolations) in VERO cells. United States by the mosquito Culex tarsalis; birds serve as For RNA preparation, virus grown in VERO cells was an important vertebrate reservoir. In the eastern United purified by pelleting onto a 30o sucrose cushion followed by States, WEEV is replaced by Highlands J virus (HJV), whose isopycnic banding in Nycodenz (Nyegaard, Oslo). After primary vector is Culiseta melanura. From serological stud- pelleting and dissociation in NaDodSO4, the RNA was ies (3, 4) and from limited sequencing studies (5, 6), WEEV extracted by phenol/chloroform treatment, precipitated with and HJV are known to be very closely related, and HJV can ethanol, purified on a discontinuous sucrose gradient, and be considered to be a strain of WEEV (2). In the eastern concentrated by ethanol precipitation. United States, the range of HJV overlaps that of EEEV, Cloning and Sequencing. Clones containing the 3'-terminal whose primary vector is also Cs. melanura. Other New 4288 nucleotides of WEEV RNA were obtained by using an World alphaviruses include Venezuelan equine encephalitis oligo(dT)-tailed vector as a primer as described (17). Clones virus (VEEV), found in Central and South America; Fort Morgan virus, found in Colorado; and Aura virus, found in Abbreviations: WEEV, western encephalitis virus; EEEV, eastern South America. encephalitis virus; VEEV, Venezuelan equine encephalitis virus; The Old World alphaviruses include Sindbis virus, the HJV, Highlands J virus. prototype alphavirus; Semliki Forest virus; Chikungunya *Present address: Israel Institute for Biological Research, P.O. Box 19, Ness-Ziona, 70450, Israel. tTo whom reprint requests should be addressed. The publication costs of this article were defrayed in part by page charge MThe sequence reported in this paper is being deposited in the payment. This article must therefore be hereby marked "advertisement" EMBL/GenBank data base (IntelliGenetics, Mountain View, CA, in accordance with 18 U.S.C. §1734 solely to indicate this fact. and Eur. Mol. Biol. Lab., Heidelberg) (accession no. J03854). 5997 Downloaded by guest on October 1, 2021 5998 Evolution: Hahn et al. Proc. Natl. Acad. Sci. USA 85 (1988) were sequenced by using the chemical sequencing method through the 3'-terminal untranslated sequence, which ends in (18, 19). a poly(A) tract. We have previously sequenced the amino termini of the RESULTS three structural proteins of the McMillan strain of WEEV (isolated in 1941 in Canada from the brain of a fatal human Partial Sequence of WEEV RNA. The translated sequence case) and thus established the start points of the structural of the 3-terminal 4170 nucleotides of the WEEV genome is proteins (21). Comparison of the amino acid sequence of the shown in Fig. 1. This sequence begins in the region encoding McMillan strain with that deduced here for the BFS1703 the carboxyl terminus of nonstructural protein 4, continues strain (isolated from mosquitos in 1953 in California) reveals through the junction region between the nonstructural and four amino acid differences in 142 amino acids for which structural proteins containing the start of the subgenomic comparison is possible (one in C, one in E2, and two in El). mRNA that is translated to give the structural proteins (20), However, reevaluation of the original data for the McMillan and progresses through the coding sequence of the three strain suggests that the apparent difference in the capsid structural proteins of the virus (a nucleocapsid protein, C, proteins may result from a misscall in the McMillan sequence and two envelope glycoproteins, E2 and El) and finally and that there are no differences between the capsid proteins I S R Y E I I L A G L I I T S L S T L A E S V K N F K S I R G N P I T L Y G * 39 UCCAGAUACGAGAUCAUACUGGCAGGCCUGAUCAUCACGUCCCUGUCCACGUUAGCCGAFAGCGUUAFGAACUUCAVGAGCAUAPGAGGGAACCCAVUCACCCUCUACGGCUGACCUAA 119 __ __. P -lF P Y _P O _L _N F P_ P_ V MY _P T N P M A Y R D P N P P R 27 120 AUAGGUGACGUAGUAGACACGCACCUACCCACCGCCAA*UG-UUU-CCA-UAC-CCUC-AG-CUG-AAC'UUU-CCA-CCA-GUU-UAC-CCUA-CA-AAU-CCG-AUG-GCUUACCGAGAUCCAAACCCUCCUAGG 239 28 C R W R P F R P P L A A O I E D L R R S I A N L T F K G R S PN P P P G P PP K 87 240 UGCCGCUGGAGGCCGUUUCGGCCCCCGCUGGCUGCUCAAAUCGAAGAUCUUAGGAGGUCGAUAGCCAACUUAACUUUCAAACAACGAUCACCUAAUCCGCCGCCAGGUCCACCGCCAAAG 359 a8 K K K S A P K P K P T O P K KK K OG A K K T K R K PK P G K R 0 R M C M K L E 107 380 AAGAAGAAGAGUGCUCCUAAGCCAAAACCUACUCAGCCUA^AAAGAAGAAGCAGCAAGCCAAGAAGACGAAACGCAAGCCUAAACCAGGGAAACGACAGCGUAUGUGUAUGAAGUUGGAG 479 108 S D K T F P I M L N G O V N G Y A C V V G G R L M K P L H V E G K I D N E 0 L A 147 480 UCGGACAAGACAUUUCCGAUCAUGCUGAACGGCCAAGUGAAUGGAUACGCUUGCGUUGUCGGAGGAAGGCUGAUGAAACCACUCCACGUUGAAGGAAAAAUCGAUAAUGAGCAAUUAGCG 599 148 A V K L K K A S M Y D L E Y G D V P 0 N M K S D T L 0 Y T S D K P P G F Y N W H 187 600 GCCGUG^AAUUGAAGAAGGCUAGCAUGUACGACUUGGAGUAUGGCGACGUUCCCCAGAAUAUGAAAUCAGACACGCUGCAGUACACCAGCGACAAACCACCGGGCUUUUACAACUGGCAC 719 188