Nucleotide sequence of the genes coding for the membrane glycoproteins E1 and E2 of rubella virus.

The complete nucleotide sequence of the genes coding for the two membrane glycoproteins E1 and E2 of rubella virus has been determined from cloned cDNA derived from the 40S genomic RNA. A sequence of 2451 nucleotides extending from the poly(A) tract towards the 5' end is presented. Within one continuous open reading frame E2 is located amino-terminally followed by E1 and a 58 residue long untranslated 3' region preceding the poly(A) tract. The coding regions of E2 and E1 are unusually G/C rich, 71.4% and 66.4% respectively. At the carboxy-terminal end of the coding region of E1, there is an inverted complementary nucleotide sequence, which could form a 13 base pair hairpin structure. Mature E2 and E1 are both preceded by a stretch of uncharged mainly non-polar amino acids, 21 and 20 residues in length, respectively. These could serve as signal peptides that mediate the membrane translocation of the proteins. At the carboxy termini of both proteins there are stretches of hydrophobic amino acids, 19 and 27 residues in length, which probably represent the transmembrane anchors of the proteins. The size of mature E1 is 481 amino acids (mol. wt. 51,502), whereas the exact size of mature E2 could not be established as its carboxy-terminal end could not be located in the sequence. A maximum size of 262 amino acids (mol. wt. 28,277) is, however, suggested. Between the E2 and E1 genes, there is a stretch of seven amino acids, five of which are arginines, which may serve as cleavage sites for a trypsin-like protease. E1 contains three and E2 four potential sites for asparagine-linked glycosylation. Both proteins are cysteine-rich (5%). Comparison of the rubella virus amino acid sequence to those of several alphaviruses indicated no sequence homology.