Modulation of HIV-1-host interaction: role of the Vpu accessory protein

Abstract

Viral protein U (Vpu) is a type 1 membrane-associated accessory protein that is unique to human immunodeficiency virus type 1 (HIV-1) and a subset of related simian immunodeficiency virus (SIV). The Vpu protein encoded by HIV-1 is associated with two primary functions during the viral life cycle. First, it contributes to HIV-1-induced CD4 receptor downregulation by mediating the proteasomal degradation of newly synthesized CD4 molecules in the endoplasmic reticulum (ER). Second, it enhances the release of progeny virions from infected cells by antagonizing Tetherin, an interferon (IFN)-regulated host restriction factor that directly cross-links virions on host cell-surface. This review will mostly focus on recent advances on the role of Vpu in CD4 downregulation and Tetherin antagonism and will discuss how these two functions may have impacted primate immunodeficiency virus cross-species transmission and the emergence of pandemic strain of HIV-1.

Figure 1

Schematic representations of Vpu. (A) Predicted secondary and tertiary structure of Vpu showing the N-terminal transmembrane domain (TM) and the two α-helices of the cytoplasmic (CYTO) domain. The numbers indicate amino acid positions of the NL4.3 prototypical Vpu allele. In both panels, yellow circles represent phosphorylated serine residues (S52 and S56) sites. The 13° tilt angle of the TM domain is indicated. (B) Vpu topology with the corresponding HIV-1/SIVcpz Ptt Vpu consensus sequences (HIV sequence database, http://www.hiv.lanl.gov). Question marks indicate residues with no consensus available. The red box indicates the conserved sequences recognized by β-TrCP. The blue boxes highlight areas containing putative trafficking signals shown below. X and Φ correspond to variable and hydrophobic amino-acid residues, respectively. αH: α-helix.

Figure 2

Model of Vpu-mediated CD4 degradation. First, Vpu retains CD4 in the ER through TM domains interactions; formation of Env/CD4 complexes could contribute to this retention. In addition, CD4 and Vpu also interact through their cytosolic domains. The minimal region of the CD4 cytoplasmic tail conferring Vpu sensitivity was mapped to the region 414-LSEKKT-419. Recruitment of the SCF^β-TrCP E3 ubiquitin ligase complex by Vpu is mediated by interactions of phosphoserines in Vpu and the WD boxes of β-TrCP. Interactions between Vpu and CD4 result in the trans-ubiquination of the cytosolic tail of CD4 on lysine, serine and threonine residues. These ubiquitination events might further contribute to CD4 retention in the ER but, importantly, target CD4 for degradation by the cytosolic proteasome. This targeting involves a dislocation step mediated by the p97-UFD1L-NPL4 complex, a critical component of ERAD. This complex recognizes K48-linked polyubiquitinated chains on the cytosolic tail of CD4 through the UFD1L co-factor. The p97 protein via its ATPase activity subsequently directs the dislocation of CD4 across the ER membrane where the receptor becomes readily accessible for proteasomal degradation.

Figure 3

Schematic representations of Tetherin. Secondary and tertiary model of human Tetherin. Glycosylation sites at position 65 and 92 are shown as well as the GPI-anchor and the cytoplasmic, transmembrane (TM) and extracellular coiled-coil domains. The functional parallel dimeric state is shown here. (B) Tetherin topology. An amino-acid sequence alignment of human, chimpanzee, rhesus and African green monkey (agm) Tetherin alleles is shown below. Hyphens and bold letters represent respectively deletions and residues in human Tetherin under positive selection. Putative Ub-acceptor residues, cysteine residues involved in dimerization as well as N-glycosylation sites are labelled in orange, pink and red, respectively. Putative trafficking signals, the predicted transmembrane domain and the coiled-coil domain are highlighted in blue, green and yellow. The sites of interaction mapped for SIV Nef and HIV-1 Vpu are boxed in dark blue and dark green, respectively. Note that the SIV Nef-interacting region is deleted in human Tetherin. The site of cleavage prior to addition of the GPI lipid anchor is represented by the dashed line.

Figure 4

Schematic representations of possible direct tethering modes. (A) Tethering by interaction via the ectodomains of Tetherin dimers. One Tetherin molecule is inserted into the virus while the other is anchored into the cellular membrane. (B) Tethering by incorporation of one of the molecule anchors in the virus and the other in the cellular membrane. Different options are shown, including GPI anchors or transmembrane domains of parallel Tetherin homodimers incorporated into a virion and (C) both type of anchors from an antiparallel tetherin homodimer incorporated into virion. The fact that deleting either the GPI anchor or the TM domain prevents the restriction suggests that either configuration A and C are not important contributors of the tethering process or that a single tethering domain is not sufficient to retain virions at the cell surface. Indeed, it is also conceivable that all these potential configurations may contribute to the restrictive activity albeit to different extent.

Figure 5

Unified model of Vpu-mediated cell-surface Tetherin downregulation. Tetherin traffics along the anterograde trafficking pathway and reaches the plasma membrane. The protein is endocytosed in clathrin-coated pits, transported to the TGN and most probably recycles back to the cell surface. Upon expression of Vpu, Tetherin is forming complexes with the viral protein, thus trapping the restriction factor in the TGN, away from sites of viral assembly at the plasma membrane where Tetherin is cross-linking progeny virions. Vpu could intercept endocytosed Tetherin as well as Tetherin arriving from the ER although this remains to be determined. Subsequently, Vpu could induce Tetherin ubiquitination through recruitment of β-TrCP-2, leading to a stronger retention in the TGN. Sequestered ubiquitinated Tetherin conjugates could ultimately be targeted for proteosomal and/or lysosomal degradation. As such, Vpu-mediated Tetherin degradation may represent a complementary mechanism that Vpu could exploit to reach optimal Tetherin antagonism, perhaps, in specific cellular environments.

Figure 6

Adaptations of primate lentiviruses during cross-species transmission and the emergence of pandemic HIV-1 strains. The SIV from chimpanzee is believed to result from recombination events through successive cross-species transmission between the precursors of the SIVgsn/mon/mus and the SIVrcm lineages. The transmembrane domain of Tetherin evolved primarily during transition from the non hominoïd lineage to the hominoïd lineage, explaining why the Vpu protein inherited from the SIVgsn/mon/mus lineage does not exhibit any activity against chimpanzee Tetherin. After transmission from chimpanzees to humans, SIVcpz was unable anymore to use Nef to counteract Tetherin due to a deletion of five amino acids in the cytoplasmic domain of human Tetherin, which usually confers responsiveness to Nef. During evolution/adaptation from SIVcpz to HIV-1, modifications mapped to two regions of the Vpu transmembrane domain have conferred human Tetherin a susceptibility to Vpu, except in the case of the HIV-1 group O. Furthermore, the HIV-1 group N Vpu somehow has lost its ability to mediate CD4 degradation in the process. Only the pandemic HIV-1 group M harbors the two primary Vpu functions. Susceptibility of the transmembrane domain of Tetherin to Vpu is represented by similar colour pairing. The deletion of the five amino acids in human Tetherin cytoplamic tail is represented by the absence of the D/GDIWK sequence. The color gradient indicates co-evolution between SIV/HIV and host.