Complete nucleotide sequence and full-length cDNA clone of S.A.AR86 a South African alphavirus related to Sindbis

Abstract

S.A.AR86 and Girdwood S.A., two South African Sindbis-like arboviruses, are closely related antigenically to the Swedish isolate, Ockelbo82 [Lundström, J. O., Vene, S., Saluzzo, J. F., and Niklasson, B. (1993) Am. J. Trop. Med. Hyg. 49(5), 531-537]. Each of these viruses is associated with human disease, and Girdwood S.A. was isolated from a human case. In addition, S.A.AR86 is unique among Sindbis-like viruses in that adult mice remain sensitive to lethal infection with S.A.AR86. The complete genomic sequences of S.A.AR86 and Girdwood S.A. were determined. The S.A.AR86 RNA genome contained 11,663 nucleotides, excluding the 5' CAP structure and 3' poly(A) tail. In comparison to the consensus sequence of the prototype Egyptian Sindbis strain AR339, S.A.AR86 differed at 5.57% of the nucleotides, including a 54-nucleotide deletion, two insertions of 6 nucleotides each, and a 3-nucleotide insertion in the 3' terminal one-third of the S.A.AR86 nsP3 gene. S.A.AR86 is one of only three alphaviruses sequenced to date that does not have an opal termination codon between the nsP3 and the nsP4 genes. These genes are separated by a cysteine codon in the S.A.AR86 genome. The genome of Girdwood S.A. was 11,717 nucleotides in length, excluding the 5' CAP and 3' poly(A) tail. Girdwood S.A. contained an opal termination codon between nsP3 and nsP4 and did not have the large 54-nucleotide deletion in nsP3, although Girdwood S.A. did contain the remaining insertions and deletions characteristic of S.A.AR86. S.A.AR86 was more closely related to Girdwood S.A. than to the Egyptian isolate, and the South African isolates as a group were more closely related to the Swedish isolate. Comparison of the S.A.AR86 sequence to that of Ockelbo82, Girdwood S.A., and Sindbis virus AR339 revealed several codons where S.A.AR86 differed from the conserved amino acid encoded by the other three viruses. These changes may be related to the ability of S.A.AR86 to initiate a lethal central nervous system infection in adult mice. To fulfill a prerequisite for testing this hypothesis, a full-length cDNA clone of S.A.AR86 was constructed from which infectious genomic RNA replicas could be derived. The sequence of this clone differed from the S.A.AR86 genomic RNA sequence at four translationally silent positions, and virus derived from the clone reproduced the adult mouse neurovirulence phenotype of its biological progenitor.