Nef proteins of epidemic HIV-1 group O strains antagonize human tetherin

Abstract

Most simian immunodeficiency viruses use their Nef protein to antagonize the host restriction factor tetherin. A deletion in human tetherin confers Nef resistance, representing a hurdle to successful zoonotic transmission. HIV-1 group M evolved to utilize the viral protein U (Vpu) to counteract tetherin. Although HIV-1 group O has spread epidemically in humans, it has not evolved a Vpu-based tetherin antagonism. Here we show that HIV-1 group O Nef targets a region adjacent to this deletion to inhibit transport of human tetherin to the cell surface, enhances virion release, and increases viral resistance to inhibition by interferon-α. The Nef protein of the inferred common ancestor of group O viruses is also active against human tetherin. Thus, Nef-mediated antagonism of human tetherin evolved prior to the spread of HIV-1 group O and likely facilitated secondary virus transmission. Our results may explain the epidemic spread of HIV-1 group O.

Figure 1. Antitetherin Activity of HIV-1 Group O Nef Proteins

(A) Virus release from 293T cells cotransfected with an HIV-1 NL4-3 ΔvpuΔnef construct and vectors expressing the indicated nef alleles or human (HUM) tetherin. Virus yield was determined by p24 ELISA (left) or infection of TZM-bl cells (right). Data show mean percentages (± SEM) relative to those detected in the absence of tetherin (100%) obtained in three independent experiments. The right panel shows the levels of p24 antigen and infectious virus release in the presence of 100 ng of HUM tetherin expression construct. For comparison, results obtained for 18 M- and 9 N-Vpus are shown in the bar diagram. Stars refer to the difference from the EGFP control panel. *p < 0.05, **p < 0.01, ***p < 0.001. (B) Effect of HIV-1 Nefs on surface expression of human tetherin. Shown are the levels of tetherin cell surface expression in the presence of HIV-1 group M Vpus or M-, N-, and O-Nefs relative to those measured in 293T cells transfected with the EGFP control vector and tetherin expression plasmid alone (100%). Each symbol represents one individual nef or vpu allele. Shown are average values derived from four to six independent experiments. (C) PHA-activated PBMCs were transduced with VSV-G-pseudotyped vpu-defective HIV-1 NL4-3 constructs expressing the indicated nef alleles and examined for tetherin surface expression 3 days later. Shown are average levels (± SEM) of surface expression in virally infected (EGFP+) cells relative to uninfected cells (100%). *** indicates that O-Nefs are significantly (p < 0.001) more active than M-, N-, or P-Nefs. (D) Nef-dependent reduction of HUM and GOR tetherin surface expression in 293T cells. Shown are the levels of tetherin cell surface expression in the presence of various Nefs relative to those measured in cells transfected with the control vector expressing only EGFP. The right panel shows the average levels of HUM and GOR tetherin surface expression in the presence of the three SIVgor and HIV-1 O Nefs (see also Figure S1).

Figure 2. Functional Analysis of MRCA HIV-1 Nef Proteins

(A) p24 antigen (left) and infectious virus (right) release from 293T cells cotransfected with an HIV-1 NL4-3 ΔvpuΔnef construct and vectors expressing the NL4-3 Vpu or the O-MRCA Nef and the indicated amounts of a vector expressing HUM tetherin. Shown are percentages relative to those detected in the absence of tetherin (100%). Values are means from three experiments. (B) Infectious virus yield from 293T cells cotransfected with an HIV-1 NL4-3 ΔvpuΔnef construct and vectors expressing the indicated nef alleles in combination with plasmids expressing HUM or GOR tetherin. Shown are average values derived from triplicate infections of TZM-bl indicator cells relative to those obtained in the absence of tetherin expression vector (100%). (C) Levels of tetherin cell surface expression in the presence of HIV-1 group M, N, and O MRCA Nefs relative to those measured in 293T cells transfected with the EGFP control vector (100%). The M-MRCA and M-MRCA P Nefs differ by a single E to P substitution at their N terminus (Figure S2A). Values in all panels are means (± SEM) from at least three experiments. (D) PHA-activated PBMCs were transduced with VSV-G-pseudotyped vpu-defective HIV-1 NL4-3 constructs expressing the indicated nef alleles and examined for tetherin surface expression three days later. The results show average levels (± SEM) in virally infected (EGFP+) cells relative to uninfected cells (100%). (E) Effect of various Nef and Vpu proteins on NF-κB activation. Activation of NF-κB-dependent firefly luciferase expression in 293T cells cotransfected with tetherin or constitutively active IKKβ, NF-κB-dependent (three NF-κB binding sites) firefly luciferase reporter constructs, a plasmid expressing Gaussia luciferase under the control of a minimal promoter, and the indicated nef alleles. Shown are average values derived from three experiments (see also Figure S2).

Figure 3. Mapping of Residues in Human Tetherin Targeted by HIV-1 O Nefs

Effect of alanine substitutions in the cytoplasmic part of HUM-rep tetherin on susceptibility to SIVgor, O-MRCA, and HJ162 nef alleles. Shown are the levels of tetherin cell surface expression in the presence of Nef relative to those measured in cells transfected with the EGFP-only control vector (100%). Values are averages (± SEM) derived from three experiments (see also Figure S3).

Figure 4. Mapping of Residues in HIV-1 O Nefs Involved in Modulation of Human Tetherin

(A and B) Effect of chimeras between HIV-1 O and SIVgor Nefs and mutant O-Nefs on cell surface expression of HUM tetherin. Alterations introduced into the O-MRCA and HJ162 C1 and C2 mutants are shown in Figure S4A. (C) Localization of amino acid residues involved in antitetherin activity in the HIV-1 Nef structure. The alterations in the C1 region of Nef (red) and the NTS-x₄-VS residues (yellow) in NL4-3 Nef critical for the tetherin antagonism, the acidic residue and two leucines (blue) of the dileucine motif involved in the interaction with AP complexes, and two conserved amino acids in between (orange), are highlighted. (D) Effect of changes in and near the ExxxLL motif of Nef on downmodulation of HUM or GOR tetherin and CD4 cell surface expression. Values represent average expression levels (± SEM; n = 3) relative to those obtained in the presence of the vector expressing only EGFP. Color coding in the alignment corresponds to that in the Nef structure shown in (C). See also Figure S4.

Figure 5. Mechanism of Downmodulation of Human Tetherin by HIV-1 O Nefs

(A) Effects of O-Nefs on cell surface expression and internalization of HUM tetherin. 293T cells were transfected with vectors expressing tetherin and the indicated Nef proteins or only EGFP as control. Cells were stained with antitetherin antibodies, incubated at 37°C for different time points, and analyzed by FACS. The left panel shows the total and intracellular (after acidic wash) MFIs of HUM tetherin expression. The right panel shows the percentage of internalized tetherin in the presence of the indicated Nef proteins or the absence of Nef. Values represent averages derived from three independent experiments. (B) Effect of O-Nefs on transport of newly synthesized HUM tetherin to the cell surface. The nuclei of HT1080 cells were comicroinjected with a HUM tetherin expression plasmid together with vectors encoding EGFP, NL4-3 Vpu, or the indicated Nef proteins. About 200 cells were microinjected and analyzed. Microphotographs show representative examples. Cell circumferences are indicated by white dashed lines. Scale bar, 10 μM. (C) Histogram bars depict the relative percentage of cells showing the different categorization (plasma membrane and intracellular versus intracellular only) for each time point from about 200 cells analyzed. (D) O-Nefs trap newly synthesized HUM tetherin at the trans-Golgi network (TGN). Shown are representative confocal microphotographs of HT1080 cells microinjected as described in (B) that were analyzed for the distribution of the TGN marker 46 at 2 hr postinjection. Depicted are overviews and zoom of the merged tetherin (red) and TGN46 (green) channel. Cell circumferences are indicated by white dashed lines. Scale bar, 10 μM (see also Figure S5).

Figure 6. O-Nefs Do Not Antagonize the Short Isoform of Human Tetherin

(A) Amino acid alignment of the long and short isoforms of HUM tetherin. (B) Expression of the short form of human tetherin in the presence of Vpu or the indicated Nef proteins. 293T cells were cotransfected with plasmids expressing the indicated tetherin variants and Vpu or the indicated Nef proteins and examined by western blot 2 days later. (C) Surface levels of tetherin expressionon 293T cells transfected with constructs expressing the long or short isoforms of human tetherin. Results in (C)–(E) show mean values (± SEM) derived from three to six transfections. (D) Downmodulation of the two isoforms of human tetherin by the indicated viral proteins. Shown are the levels of cell surface expression relative to those obtained in the presence of the vector expressing only EGFP. (E) Virus release from 293T cells cotransfected with an HIV-1 NL4-3 ΔvpuΔnef construct and vectors expressing the indicated viral proteins and a plasmid expressing the short isoform of HUM tetherin. Virus yield was determined by p24 ELISA (left) and infection of TZM-bl cells (middle). The right panel shows the correlation between both measurements (see also Figure S6).

Figure 7. Role of O-Nef-Mediated Downmodulation of HUM Tetherin in HIV-1 Infectivity and Replication Fitness

(A) P4-CCR5 indicator cells were infected with vpu-defective HIV-1 NL4-3 IRES-EGFP constructs containing the indicated nef genes or a defective nef allele. Infections were performed in triplicate with virus stocks containing 100 ng of p24 antigen. Shown are average values (± SEM) compared to the infectivity of the nef-defective HIV-1 construct (100%). (B) Levels of tetherin cell surface expression in uninfected PBMC cultures (n = 3) in the absence and presence of IFNα. (C) Efficiency of p24 release in CD4+Tcells infected with HIV-1 NL4-3 constructs expressing the indicated nef alleles. Values present percentages of cell-free p24 antigen out of the total p24 detected in the presence and absence of IFNα and were derived from infections of T cells derived from three PBMC donors, each performed in triplicate. Cell-free and cell-associated p24 antigen was quantified by ELISA at 5 days postinfection. The O nef alleles were examined in the context of a vpu-defective NL4-3 construct. Constructs expressing the WT NL4-3 nef or containing disrupted vpu and/or nef alleles were examined for control. (D) Enhancement of p24 release by HIV-1 group M and N accessory proteins. Data were derived from the experiment shown in (A). Shown are mean values (± SEM; n = 3–9) obtained for the increase of p24 release by the presence of indicated viral genes in the HIV-1 NL4-3 proviral construct compared to the control construct lacking intact vpu and nef genes. (E) Virus replication in CD4+ T cells in the presence and absence of IFNα. The replication kinetics of HIV-1 NL4-3 constructs expressing the indicated Nef proteins are shown in CD4+ T cells from three donors in the absence (upper lines) and presence (lower lines) of 500 U/ml IFNα. Results show median values of p24 antigen production (± SEM; n = 3). Please note that the upper curves overlay one another. (F) Cumulative p24 antigen release in the culture supernatant over 10 days of viral replication in the absence (left) and presence (right) of IFNα. Shown are mean values (± SEM) for NL4-3 constructs expressing the O-MRCA and WT O-Nefs (HJ162, HJ428, HJ256) or the respective NTSVS mutant forms. (G) Ratio of p24 production in the presence and absence of IFNα is plotted for each virus at 5 days postinfection. Symbols are the same as in (B) (see also Figure S7).