Polarity changes in the transmembrane domain core of HIV-1 Vpu inhibits its anti-tetherin activity

Abstract

Tetherin (BST-2/CD317) is an interferon-inducible antiviral protein that restricts the release of enveloped viruses from infected cells. The HIV-1 accessory protein Vpu can efficiently antagonize this restriction. In this study, we analyzed mutations of the transmembrane (TM) domain of Vpu, including deletions and substitutions, to delineate amino acids important for HIV-1 viral particle release and in interactions with tetherin. The mutants had similar subcellular localization patterns with that of wild-type Vpu and were functional with respect to CD4 downregulation. We showed that the hydrophobic binding surface for tetherin lies in the core of the Vpu TM domain. Three consecutive hydrophobic isoleucine residues in the middle region of the Vpu TM domain, I15, I16 and I17, were important for stabilizing the tetherin binding interface and determining its sensitivity to tetherin. Changing the polarity of the amino acids at these positions resulted in severe impairment of Vpu-induced tetherin targeting and antagonism. Taken together, these data reveal a model of specific hydrophobic interactions between Vpu and tetherin, which can be potentially targeted in the development of novel anti-HIV-1 drugs.

Figure 1. Schematic representation and basic characteristics of Vpu TM mutants.

(A) Deletion and substitution mutagenesis of the Vpu TM domain. The numbering of amino acid residues used is based on the original Vpu sequence of NL4-3. The deleted amino acids are indicated by the dashes, and the substituted amino acids within the TM region are represented in bold. (B) PAGE analysis of Vpu TM mutants under native and denaturing conditions. Lysates from the Vpu variants transfected 293T cells were mixed with SDS sample buffer or native sample buffer and subjected to standard SDS-PAGE or native PAGE, respectively. The gels were transferred to a nitrocellulose membrane and further analyzed by immunoblotting using a mouse anti-myc antibody to detect myc-tagged Vpu protein. The upper panel shows the native condition and the lower panel shows denaturing condition. (C) Subcellular localization of Vpu TM mutants. HeLa cells transfected with Vpu variant expression plasmids were fixed and double stained with a mouse anti-myc antibody (green) and a rabbit anti-TGN46 antibody (red). Images were taken under a fluorescence microscope. At least 30 independent cells were examined in each sample, and the most representative cells are shown.

Figure 2. Defective Vpu TM mutants fail to enhance HIV-1 virus release.

(A) VR1012 control vector or VR1012 encoding Vpu TM variants (500 ng each) was co-transfected with 1 µg proviral plasmids of pNL4-3 WT or pNL4-3ΔVpu in HeLa cells. At 48 h post-transfection, the cultured supernatants were ultracentrifuged to concentrate the virus particles. The virions and the cell lysates were analyzed by immunoblotting using an anti-p24 antibody to detect virion p24 capsid and intracellular Pr55Gag proteins, anti-myc antibody to detect myc-tagged Vpu and an antibody against tubulin to assess sample loading. The viral Pr55Gag protein was examined to exclude variations of transfection efficiency. (B) The relative infectivities of virus released from the cells transfected in (A) were assayed by infecting MAGI cells with equal volumes of supernatant samples. The cells were then fixed and stained for β-galactosidase activity. Virus release of NL4-3 WT was set to 100%. The graph was generated from two independent experiments each performed in duplicate.

Figure 3. Roles of degradation and surface downregulation in Vpu-induced tetherin dysfunction.

(A) 293T cells were co-transfected with 1 µg pNL4-3 WT or pNL4-3ΔVpu and 50 ng HA-tetherin or empty vector. At 48 h, the cells were examined for tetherin expression, and the pelleted virions were analyzed for p24 content. Pr55Gag was examined to exclude variations in transfection efficiencies. Tubulin was detected as a loading control. (B) The relative infectivity of virus released in (A) was assayed by infecting MAGI cells. Virus release of NL4-3 WT in the absence of tetherin was set to 100%. (C) 293T cells were co-transfected with 1 µg pNL4-3ΔVpu, 50 ng HA-tetherin and increasing doses of Vpu. HA-tetherin and Vpu-cmyc were detected in the cells. (D) The cellular tetherin levels and virus released were plotted in a line graph. The tetherin level in the absence of Vpu was set to 100%. Viral output was scored by titration of the supernatants on MAGI cells, and that without tetherin was set to 100%. (E) HeLa cells were co-transfected with 1 µg pNL4-3ΔVpu or WT, along with 500 ng pEGFP-N3, and increasing doses of Vpu. Vpu was detected in the cells with Vpu antiserum, and the pelleted virions were analyzed for p24 content. (F) Surface tetherin of cells in (E) were stained with tetherin antibodies and analyzed by flow cytometry. Cells only transfected pEGFP-N3 was used as a negative control. The samples were gated on EGFP+ cells, and surface tetherin levels are shown in histograms with median values at the top right corner. (G) The levels of surface tetherin and virus released are shown in a line graph. The tetherin level in the negative control was set to 100%. Viral output was scored by titration of the supernatants on MAGI cells, and that of NL4-3 WT was set to 100%. All values are representative of three independent experiments.

Figure 4. Effects of Vpu TM mutations on Vpu-mediated degradation and surface downregulation of tetherin.

(A) 293T cells were co-transfected with 100 ng HA-tetherin expression plasmid along with 200 ng VR1012 control vector or VR1012 encoding Vpu TM variants at a 2∶1 molar ratio. At 48 h post-transfection, the cells were harvested for immunoblotting analysis. Tetherin and Vpu were detected with anti-HA and anti-myc antibody, respectively. Tubulin was detected as a loading control. (B) Tetherin levels were measured using Bandscan software and normalized by tubulin levels. Percentages of degraded tetherin were calculated by subtracting the densitometric intensity values of the indicated Vpu WT or mutant bands from that of the mock band to represent the different abilities of Vpu variants to mediate tetherin degradation. Values are representative of three independent experiments. (C) HeLa cells were co-transfected with 500 ng pEGFP-N3 along with 500 ng VR1012 control vector or VR1012 encoding Vpu TM variants. Cell surface tetherin was stained with BST-2 antibodies, followed by Alexa 633 goat anti-mouse IgG and analyzed by flow cytometry. Samples were gated on EGFP+ cells, and the surface tetherin levels are shown in the histograms with median values at the top right corner.

Figure 5. Defective Vpu TM mutants potently block the physical interaction between Vpu and tetherin.

293T cells were co-transfected with 1 µg VR1012 control vector or VR1012 encoding Vpu TM variants together with 2 µg HA-tetherin expression vector at a 1∶2 molar ratio to minimize the Vpu-mediated tetherin degradation. Cells were harvested 48 h later and subjected to immunoprecipitation using the anti-myc antibody and protein G agarose beads. Cell lysates or co-precipitated proteins were analyzed by immunoblotting to detect HA-tetherin and Vpu-cmyc. Equal loading was controlled by monitoring tubulin. The level of tetherin in each sample was quantified by densitometry, normalized by tubulin levels and shown beside the tetherin blot. The value obtained with the positive control Vpu WT was defined as 100%.

Figure 6. Effects of Vpu TM mutations on Vpu-mediated degradation and surface downregulation of CD4.

(A) 293T cells were co-transfected with 100 ng CD4-HA expression plasmid along with 200 ng VR1012 control vector or VR1012 encoding Vpu TM variants. At 48 h post-transfection, the cells were harvested for immunoblotting analysis. CD4 and Vpu were detected with anti-HA and anti-myc antibodies, respectively. Tubulin was detected as a loading control. (B) CD4 levels were measured using Bandscan software and normalized by tubulin levels. Percentages of degraded CD4 were calculated by subtracting the densitometric intensity values of the indicated Vpu WT or mutant bands from that of the mock band to represent the different abilities of Vpu variants to mediate CD4 degradation. Results shown are the average of two independent experiments. (C) HeLa CD4 cells were co-transfected with 500 ng pEGFP-N3 along with 500 ng VR1012 control vector or VR1012 encoding Vpu TM variants. Cell surface CD4 was stained with CD4 antibodies followed by Alexa 633 goat anti-mouse IgG and analyzed by flow cytometry. Samples were gated on EGFP+ cells, and the surface CD4 levels are shown in the histograms with median values at the top right corner.