Ancient adaptive evolution of tetherin shaped the functions of Vpu and Nef in human immunodeficiency virus and primate lentiviruses

Abstract

Tetherin/BST-2 is a host-encoded protein that restricts a wide diversity of viruses at the stage of virion release. However, viruses have evolved antagonists of Tetherin, including the Vpu and Nef proteins of primate lentiviruses. Like other host genes subject to viral antagonism, primate Tetherin genes have evolved under positive selection. We show here that viral antagonists acting at three independent sites of selection have driven the evolution of Tetherin, with the strongest selective pressure on the cytoplasmic tail domain. Human Tetherin is unique among the Tetherins of simian primates in that it has a 5-amino-acid deletion that results in the loss of the residue under the strongest positive selection. We show that this residue at amino acid 17 is the site of the functional interaction of Tetherin with Nef, since single amino acid substitutions at this single position can determine the susceptibility of Tetherin to Nef antagonism. While the simian immunodeficiency viruses SIVcpz and SIVgor are able to antagonize their hosts' Tetherin with Nef, human immunodeficiency virus type 1 (HIV-1) Vpu has evolved to counteract Tetherin in humans. We mapped the adaptations in the N-terminal transmembrane domain of Vpu that allow it to counteract human Tetherin. Our combined evolutionary and functional studies have allowed us to reconstruct the host-pathogen interactions that have shaped Tetherin and its lentivirus-encoded antagonists.

FIG. 1.

HIV-1 Vpu, SIVcpz Nef, and SIVgor Nef are potent antagonists of their hosts' Tetherins. (A) Viral infectivity assay of an HIV-1 ΔVpu ΔNef reporter virus (HIV1VpuFSLuc2) released into the supernatant, with increasing amounts of untagged human Tetherin (0, 3.125, 6.25, 12.5, 25, and 50 ng) in the presence of either 25 ng Vpu (filled circles) or 25 ng Nef (open circles) from HIV-1 strain Lai (orange) or Q23-17 (blue). The relative infectivity was normalized to the viral infectivity when the respective Vpu/Nef was expressed in the absence of Tetherin. (B) Native chimpanzee Tetherin was titrated in the presence of either 25 ng Vpu (filled circles) or 25 ng Nef (open circles) from strain SIVcpzUS (orange) or SIVcpzTan3.1 (blue). (C) Native gorilla Tetherin was titrated in the presence of either 25 ng Vpu (filled circles) or 25 ng Nef (open circles) from SIVgor. (D) 293T cells were cotransfected with a bicistronic vector expressing human CD4 and enhanced GFP (eGFP) and the indicated Vpu constructs. Cells were stained (with APC) for CD4 expression and were analyzed by flow cytometry. GFP-positive events were gated, and the percentages of CD4-positive event counts in the presence versus the absence of the indicated Vpu constructs were compared.

FIG. 2.

Species-specific antagonism by HIV-1, SIVcpz, and SIVgor viral antagonists. (A) Heat map of Vpu/Nef antagonism of human, chimpanzee, and gorilla Tetherins. The infectivity of an HIV-1 ΔVpu ΔNef reporter virus released into the supernatant (normalized to that in the absence of Tetherin) was assayed in the presence of human (25 ng), chimpanzee (25 ng), or gorilla (25 ng) Tetherin (columns) and of the indicated Vpu/Nef construct (rows). The percentages in each column indicate the infectivity of the virus relative to that in the absence of exogenous Tetherin (first column). (B and C) The results for HIV-1 Q23-17, SIVcpzUS, and SIVgor are plotted on radar charts. (B) Relative infectivity (expressed as a percentage) of an HIV-1 ΔVpu ΔNef reporter virus released into the supernatant in the presence of human, chimpanzee, or gorilla Tetherin, as indicated, and of Vpu from HIV-1 Lai (blue), SIVcpzUS (red), or SIVgor (green). (C) Relative infectivity (expressed as a percentage) of an HIV-1 ΔVpu ΔNef reporter virus released into the supernatant in the presence of human, chimpanzee, or gorilla Tetherin, as indicated, and of Nef from HIV-1 Lai (blue), SIVcpzUS (red), or SIVgor (green).

FIG. 3.

Tetherin was under strong positive selection in the cytoplasmic tail domain. (A) Likelihood ratio tests were used to determine if any codons were associated with dN/dS ratios significantly greater than 1 (hence under positive selection). Neutral models (M7) were compared to selection models (M8) under the F61 model of codon substitution. Similar results were obtained in a comparison of M1 (neutral) versus M2 (selection) (data not shown). (B) An alignment of the cytoplasmic tail and transmembrane domains of Tetherins from the 20 primates used in the analyses is shown, with three positively selected codons (codons 9, 17, and 43) shaded (indicated by stars); these were identified by PAML and REL as being subjected to positive selection with high posterior probabilities (P, >0.95). Codon 52, which is critical for resistance against HIV-1 Vpu but was not identified as evolving under positive selection, is boxed. (C) PAML omits regions that have deletions in the alignment from the analysis. Therefore, PAML analyses were repeated without human tetherin. Codons with a posterior probability of >95% were highlighted in a PAML analysis of tetherin genes from all primates, excluding humans. (D) Summary of the REL analysis of whole tetherin genes from 20 primates. Sites displaying positive selection signals with a significance (Bayes factor) greater than 50 (codons 9, 17, and 43) are shown in boldface. Codon 52, which did not meet the cutoff, is included for reference. (E) Likelihood ratio tests were performed between the M7 (neutral) and M8 (selection) models for the full tetherin gene, without human tetherin, or without human tetherin and with amino acids 9, 17, and 43 omitted from the alignment.

FIG. 4.

An amino acid under selection in the cytoplasmic tail of Tetherin explains the specificity of Nef. Shown are results of assays to determine the infectivity of an HIV-1 ΔVpu ΔNef reporter virus released into the supernatant. The relative infectivity was normalized to the viral infectivity obtained when the indicated Vpu/Nef protein was expressed in the absence of Tetherin. Error bars indicate standard deviations. (A) Infectivity was assayed either in the absence of Tetherin (open bars) or in the presence of AGM (shaded bars) or AGM C17W (filled bars) Tetherin and in the presence of the indicated Vpu or Nef construct. (B) Infectivity was assayed either in the absence of Tetherin (open bars) or in the presence of chimpanzee Tetherin (shaded bars), chimpanzee W17C Tetherin (filled bars), or ChimpΔDDIWK Tetherin (hatched bars). (C) Infectivity was assayed either in the absence of Tetherin (open bars) or in the presence of human Tetherin (shaded bars) or ancestral human Tetherin (filled bars) and in the presence of the indicated Vpu or Nef construct. (D) Infectivity was assayed either in the absence of Tetherin (open bars) or in the presence of human Tetherin (shaded bars), C. cephus Tetherin (filled bars), or Francois' leaf monkey (FLM) Tetherin (hatched bars) and either without Vpu or with SIVmus Vpu. The “ancestral” human Tetherin is the human Tetherin with the amino acids DDIWK restored in place of the deletion, and with the E following this deletion replaced with a K (Fig. 3).

FIG. 5.

The N-terminal transmembrane domain of Vpu confers Tetherin antagonism activity. (A) (Left) Alignment of the N-terminal transmembrane domains of the chimeric Vpu constructs between HIV-1 strain Q23-17 (shaded areas) and strain SIVcpzUS (open areas). (Right) Expression of chimeric Vpu constructs. Western blot analysis was performed to determine the expression levels of C-terminally HA epitope-tagged Vpu and β-actin. (B) Assay of the infectivity of an HIV-1 ΔVpu ΔNef reporter virus (HIV1VpuFSLuc2) in the presence of human Tetherin (25 ng) and native Vpu proteins (untagged) as indicated. (C) Expression of Tetherin on the cell surface as determined by flow cytometry in the presence or absence of different native Vpu proteins (untagged). Forward scatter is shown along the y axis, and Tetherin (detected with an FITC-labeled secondary antibody) is shown along the x axis. The solid line represents gating, determined by using untransfected 293T cells as controls, and the percentage of cells expressing cell surface Tetherin is given on the upper right. (D) Flow cytometric analysis of 293T cells cotransfected with a bicistronic vector (expressing human CD4 and eGFP) and the indicated native Vpu constructs (untagged). GFP-positive events were gated, and the percentages of CD4-positive (APC-labeled) event counts in the presence or absence of the indicated Vpu constructs were compared.

FIG. 6.

The N-terminal transmembrane domain of SIVcpz Vpu is divergent from that of the highly conserved HIV-1 Vpu. Shown is an alignment of the consensus sequence for the N-terminal transmembrane domain (aa 1 to 28) of HIV-1 group M Vpu with the transmembrane domains of Vpu proteins from SIVcpz strains. Sequence logos were plotted from 1,271 sequences of HIV-1 group M Vpu (top) and from the indicated SIVcpz strains (bottom). The two regions important for Tetherin antagonism activity (aa 1 to 8 and aa 14 to 22) are shaded.