Fine mapping and characterization of the Rous sarcoma virus Pr76gag late assembly domain

The p2 region of the Rous sarcoma virus (RSV) Gag polyprotein contains an assembly domain, which is required late in replication for efficient budding of virus-like particles from cells (J. W. Wills, C. E. Cameron, C. B. Wilson, Y. Xiang, R. P. Bennett, and J. Leis, J. Virol. 68:6605-6618, 1994). This domain, referred to as the L domain, was previously mapped to the 11 amino acids of p2b. Through the analysis of a series of deletion and substitution mutations, the L domain has now been fine mapped to a highly conserved amino acid sequence, PPPPYV of p2b. Sequences flanking PPPPYV motif can be deleted without any effect on budding. Defects caused by L-domain deletions can be rescued by placing a wild-type copy of the sequence at several other positions in RSV Gag. A proline-rich P(S/T)APP motif is found in many retroviral Gag polyproteins; the motif found in the p6 region of human immunodeficiency virus type 1 has been implicated in late functions of the virus. Substitution of the RSV L domain with this motif in a 10-amino-acid sequence derived from visna leukemia virus results in wild-type release of virus particles from cells. In contrast, the slightly different sequences from Gibbon ape leukemia virus, Moloney leukemia virus, PSAPP alone, or a proline-rich SH3 binding sequence do not efficiently rescue RSV L-domain mutations.