Vpu-mediated CD4 down-regulation and degradation is conserved among highly divergent SIV(cpz) strains.

Human immunodeficiency virus type 1 (HIV-1) along with simian immunodeficiency viruses from chimpanzees (SIV(cpz)) and three species of Old World monkeys from the genus Cercopithecus have been shown to encode a Vpu protein. To date, the functional characterization of Vpu has been limited to a small number of subtype B and more recently subtype C Vpu proteins. Using a recently developed VpuEGFP reporter system, we have shown that the subtype B and C Vpus are capable of preventing CD4 from being expressed on the cell surface. Using the same reporter system, we report here on the expression and functional analysis of Vpu protein from four SIV(cpz) isolates (CAM13, ANT, TAN1, and GAB1). All four SIV Vpu fusion proteins were efficiently expressed and prevented CD4 expression on the cell surface and induced CD4 degradation. This was surprising as three of the SIV(cpz) Vpu fusion proteins had only one canonical casein kinase II (CK-II) site (CAM13, ANT, TAN1) while previous studies with laboratory adapted HXB2 had indicated that both CK-II sites were required for CD4 degradation. Both ANT and TAN1 Vpu sequences encoded five consecutive negatively charged amino acids residues following the only CKII site (SAIEEDEE for ANT; SGVEEDEE for TAN1). We thus explored the possibility that this stretch of negatively charged amino acids might substitute for the lack of second CK-II site. Substitution of the aspartic acid at position 61 and glutamic acid at position 63 in the SIV(cpz) ANT Vpu within with lysine residues abolished the ability of this protein to down-modulate cell surface expression of CD4. Similarly, change of a serine to an alanine residue following the single consensus CK-II site of the CAM13 Vpu (SGNESDGGEEE) abolished CD4-down-regulation, suggesting that this serine was phosphorylated in the absence of a canonical CK-II site. Our results indicate that the serine was required, suggesting that this serine was phosphorylated by CK-II or possibly another cellular kinase. Taken together, these results show for the first time that Vpu proteins from SIV(cpz) isolates, although quite diverse in sequence and predicted secondary structure from the HIV-1 subtype B protein, are capable of down-regulating CD4, which is one of the major functions of the HIV-1 protein.

Fusion of the upstream vpu sequences to the env of simian human immunodeficiency virus (SHIV(KU-1bMC33)) results in the synthesis of two envelope precursor proteins, increased numbers of virus particles associated with the cell surface and is pathogenic for pig-tailed macaques.

Previous studies have shown that the gene coding for the Vpu protein of the human immunodeficiency virus type 1 (HIV-1) is 5' to the env gene, is in a different reading frame, and overlaps the env by 90 nucleotides. In this study, we examined the processing of the Env protein as well as the maturation and infectivity of a virus (SHIV(Vpenv)) in which a single nucleotide was removed at the vpu-env junction, fusing the first 162 bases of vpu to the env ORF. Pulse-chase analysis revealed that SHIV(Vpenv)-infected cells gave rise to two precursor glycoprotein species (gp160 and gp175). Immune precipitation results also revealed that an anti-Vpu serum could immune precipitate the gp175 precursor, suggesting that the amino-terminal Vpu sequence was fused to the Env protein. Growth curves revealed that the SHIV(Vpenv)-inoculated cultures released approximately three times more p27 into the culture medium than parental SHIV(KU-1bMC33). Electron microscopy revealed that while both viruses matured at the cell plasma membrane, significantly higher quantities of virus particles were cell associated on SHIV(Vpenv)-infected cells compared to cultures inoculated with parental SHIV(KU-1bMC33). Furthermore, virus was observed maturing into intracellular vesicles of SHIV(Vpenv)-infected cells. To assess the pathogenicity of SHIV(Vpenv), three pig-tailed macaques were inoculated with the SHIV(Vpenv) and monitored for 6 months for CD4(+) T cell levels, viral loads, and the stability of the deletion at the vpu-env junction. Our results indicated that SHIV(Vpenv) caused a severe CD4(+) T cell loss in all three macaques within weeks of inoculation. Sequence analysis of the vpu gene analyzed from sequential PBMC samples derived from macaques revealed that this mutation was stable during the period of rapid CD4(+) T cell loss. Sequence analysis showed that with increasing time of infection, the one base pair deletion was repaired in all three macaques inoculated with SHIV(Vpenv) with the reversion occurring at 10 weeks in macaque CT1G and at 12 weeks in macaque CP3R and CT1R. These results indicate that fusion of the first 54 amino acids of Vpu to Env results in intracellular maturation of virus, and accumulation of virus within intracellular vesicles as well as on the cell plasma membrane. Our results indicate that while fusion of the vpu gene to env results in a virus that is still pathogenic for pig-tailed macaques, there is a selective pressure to maintain the vpu and env genes in separate reading frames.