Ubiquitin conjugation to Gag is essential for ESCRT-mediated HIV-1 budding

Abstract

Background: HIV-1 relies on the host ESCRTs for release from cells. HIV-1 Gag engages ESCRTs by directly binding TSG101 or Alix. ESCRTs also sort ubiquitinated membrane proteins through endosomes to facilitate their lysosomal degradation. The ability of ESCRTs to recognize and process ubiquitinated proteins suggests that ESCRT-dependent viral release may also be controlled by ubiquitination. Although both Gag and ESCRTs undergo some level of ubiquitination, definitive demonstration that ubiquitin is required for viral release is lacking. Here we suppress ubiquitination at viral budding sites by fusing the catalytic domain of the Herpes Simplex UL36 deubiquitinating enzyme (DUb) onto TSG101, Alix, or Gag.

Results: Expressing DUb-TSG101 suppressed Alix-independent HIV-1 release and viral particles remained tethered to the cell surface. DUb-TSG101 had no effect on budding of MoMLV or EIAV, two retroviruses that rely on the ESCRT machinery for exit. Alix-dependent virus release such as EIAV's, and HIV-1 lacking access to TSG101, was instead dramatically blocked by co-expressing DUb-Alix. Finally, Gag-DUb was unable to support virus release and dominantly interfered with release of wild type HIV-1. Fusion of UL36 did not effect interactions with Alix, TSG101, or Gag and all of the inhibitory effects of UL36 fusion were abolished when its catalytic activity was ablated. Accordingly, Alix, TSG101 and Gag fused to inactive UL36 functionally replaced their unfused counterparts. Interestingly, coexpression of the Nedd4-2s ubiquitin ligase suppressed the ability of DUb-TSG101 to inhibit HIV-1 release while also restoring detectable Gag ubiquitination at the membrane. Similarly, incorporation of Gag-Ub fusion proteins into virions lifted DUb-ESCRT inhibitory effect. In contrast, Nedd4-2s did not suppress the inhibition mediated by Gag-DUb despite restoring robust ubiquitination of TSG101/ESCRT-I at virus budding sites.

Conclusions: These studies demonstrate a necessary and natural role for ubiquitin in ESCRT-dependent viral release and indicate a critical role for ubiquitination of Gag rather than ubiquitination of ESCRTs themselves.

Figure 1

Fusion of the UL36 catalytic domain DUb to TSG101 inhibits ESCRT-I ubiquitination. (A) Schematic representation of the DUb-TSG101 fusion proteins. DUb and DUb* domains were fused to TSG101 as depicted. (B) Effect of DUb fusion to TSG101 on ESCRT-I ubiquitination. 293T cells were transfected with ESCRT-I members [Flag-tagged VPS28 (800 ng), VPS37 (1.7 μg) and MVB12 (800 ng)] and either strep-TSG101 (2.5 μg) (lanes 1 and 7), DUb-TSG101 (lanes 3 and 9) or DUb*-TSG101 alone (600 ng) (lanes 5 and 11) or with HA-Ub (1.5 μg throughout the study unless otherwise specified) (lanes 2, 4, 6, 8, 10, 12). Immunocomplexes were analyzed by western (WB) blotting (WB) using the indicated antibodies. (C) DUb-TSG101 fusion proteins bind HIV-1 NCp6 region. GST (lanes 1, 3 and 5) and GST-NCp6 (lanes 2, 4 and 6) were captured on beads and then incubated with lysates from 293T cells expressing Flag-ESCRT-I/Tsg101 (lane 2), DUb-TSG101 (lane 4) or DUb*-TSG101 (lane 6). Captured proteins and cell lysates were analyzed by WB using the anti-Flag antibody. GST fusion proteins were visualized by Coomassie blue staining. (D) DUb-TSG101 deubiquitinated Gag assembly complexes. 293T cells were co-transfected with HIV-1 YP- mutant (1 μg throughout the study unless otherwise specified) and HA-Ub alone (lanes 1, 4 and 7), with strep-DUb-TSG-101 (lanes 2, 4 and 6) or strep- DUb*-TSG-101 (lanes 3, 6 and 9). Insoluble Gag-enriched fractions were isolated and solubilized to Strep-Tactin-capture DUb-TSG101 or DUb*-TSG101 containing complexes. Gag proteins associated with these complexes and their ubiquitination status assessed by antibodies to p24 and HA, respectively, and input fractions were probed with the indicated antibodies.

Figure 2

DUb-TSG101 interferes with HIV-1 release. (A) Co-expression of DUb-TSG101 inhibits HIV-1 release. 293T cells were transfected with expression plasmids of HIV-1 YP- (lane 1), or co-expressing Flag-TSG101, Flag-DUb-TSG101 or Flag-DUb*-TSG101 (lanes 2, 3, 4, respectively). (B) DUb-TSG101 failed to replace functionally cellular TSG101. 293T cells were transfected twice with RNAi to TSG101 (lanes 2–5) at 36-h intervals. At the second transfection, cells were co-transfected with expression plasmids of HIV-1 YP- alone (lanes 1 and 2) or either Flag-TSG101^RR (250 ng) (RNAi Resistant form), Flag-DUB-TSG101 ^RR or Flag-DUb*-TSG101^RR (15 ng) (lanes 3, 4, 5, respectively). Cells and viruses were collected 24 hours post-transfection and their protein content was analyzed by WB using the indicated antibodies. Virus release efficiency was also quantified using HeLa TZM-bl assays from 3 independent experiments and expressed relative to WT HIV (A) or WT TSG101 ^RR(B). (C) DUb-TSG101 inhibits late steps of HIV-1 budding. Shown are EM images of thin-sectioned 293T cells co-transfected with HIV-1 and DUb-TSG101 (a and b) or with DUb*-TSG101 (c). A high-magnification image of budding virus particles from panel (a) (rectangle) is shown (b) and black arrows indicate particles tethered to the plasma membrane or to each other. Quantification of budding defects was performed and approximately >250 virus particles from 2 independent experiments were examined and categorized as immature budding particles, or mature released particles (±SD).

Figure 3

Fusion with DUb had no effect on Alix known protein-protein interactions. (A) Schematic representation of the DUb-Alix fusion proteins. The active or inactive UL36 DUb catalytic domain was fused to Alix N-terminal region as depicted. (B) Effect of DUb fusion on Alix ubiquitination. 293T cells were transfected with Flag-Alix (500 ng), Flag-DUb-Alix (500 ng) or Flag-DUb*-Alix (500 ng) (lanes 1, 3, 5, 7, 9 and 11; respectively) or in combination with HA- Ub (lines 2, 4, 6, 8, 10 and 12; respectively). Immunocomplexes were analyzed by WB blot using the indicated antibodies. (C) DUb-Alix fusion proteins bind HIV-1 NCp6 and EIAV NCp9 proteins. GST, GST-NCp6 (right panel) or GST-NCp9 (left panel) fusion proteins were purified on glutathione beads and then incubated with lysates from 293T cells expressing 1.5 μg of Flag-Alix (lanes 2 and 8), Flag-DUb-Alix (lanes 4 and 10) or Flag-DUb*-Alix (lanes 6 and 12). Captured proteins and cell lysates were analyzed by WB blot using an anti-Flag antibody and GST fusion proteins visualized by Coomassie blue staining. (D) DUb-Alix fusion proteins retain binding to CHMP4B. 293T cells were co-transfected with HA-CHMP4B alone (2 μg) (control), or in combination with 1 μg of Flag-Alix (lane 2), Flag-AlixI212D (lane 3), Flag-DUb-Alix (lane 4) or Flag-DUb*-Alix (lane 5). Cell lysates were incubated with anti-Flag antibody-conjugated beads and both input and immunocomplexes analyzed by WB blot using the indicated antibodies.

Figure 4

DUb-Alix interferes with Alix mediated virus release. (A) Co-expression of DUb-Alix inhibits EIAV release. 293T cells were transfected with EIAV proviral DNA alone (500 ng throughout the study unless otherwise specified) (lane 1), with Flag-Alix, Flag-DUb-Alix, Flag-DUb-AlixF676D or Flag-DUb*-Alix (lanes 2, 3, 4, 5; respectively). (B) The active DUb-Alix fusion protein failed to replace cellular Alix to promote EIAV release. 293T cells were transfected twice with Alix RNAi oligonucleotides at 36-h intervals. At the second transfection, cells were transfected with EIAV provirus alone (lanes 1 and 2) or with 100 ng of RR versions of Flag-Alix, Flag-DUb-Alix or Flag-DUb*-Alix (lanes 3, 4, 5; respectively). (C) DUb-Alix fails to rescue HIV-1 PTAP- budding. 293T cells were transfected with expression plasmids of HIV-1 PTAP- alone (1 μg) (lane 1), or in combination with Flag-Alix (500 ng), Flag-DUb-Alix (100 ng) or Flag-DUb*-Alix (100 ng) (lanes 2, 3, 4; respectively). (D) Co-expression of DUb-Alix has no effect on MoMLV release. 293T cells were transfected with MoMLV provirus alone (1 μg) (lane 1), with Flag-DUb-Alix (100 ng) (lane 2) or Flag-DUb*-Alix (100 ng) (lane 3). Protein content of pelleted virions and cell lysates was analyzed 24 h later by WB blotting with the indicated antibodies. Release ratio (%) was calculated using the formula described in supplemental material. Error bars represent standard deviations (SD).

Figure 5

Fusion to DUb suppresses Gag ubiquitination and ability to release virus. (A) Schematic representation of DUb-Gag fusion proteins. DUb catalytic domain was fused to Gag C-terminal end. (B) DUb fusion to Gag suppresses ubiquitination. 293T cells were transfected with strep-tagged Gag, Gag-DUb or Gag-DUb* fusion proteins (lanes 1, 3, 5, 7, 9, 11; respectively) or in combination with HA-Ub (lanes 2, 4, 6, 8, 10, 12; respectively). Cells and virus pellets were collected 24h post-transfection and incubated with Strep-Tactin beads and captured complexes were probed with the indicated antibodies. (C) Gag-DUb* fusion protein co-assembled with WT Gag and restored the release HIV-1 PTAP-/YP-. 293T cells were transfected with HIV-1 PTAP-/YP- alone (lane 1), or Gag-DUb*Strep (lane 2), Gag-DUbStrep (lane 3), or with HIV-1 PTAP-/YP- and increasing amounts of either Gag-DUbStrep (500 ng, 1 μg or 2 μg) (lane 4, 5, 6) or Gag-DUb*Strep (500 ng, 1 μg or 2 μg) (lane 7, 8, 9). (D) Gag-DUb fusion failed to release virus particles and inhibited HIV-1 release in trans. 293T cells were transfected with HIV-1 alone (1 μg) (lane 1), in combination with either Gag-DUbStrep (1.5 μg) (lane 4), or Gag-DUb*Strep (1.5 μg) (lane 5), or with Gag-DUbStrep or Gag-DUB*Strep alone (1.5 μg) (lanes 2, 3; respectively). Cells and virions were harvested as above and their protein content were analyzed by WB blot using an anti-p24 antibody.

Figure 6

Nedd4-2s mediated Gag ubiquitination correlates with virus release. (A) Gag fusion with DUb inhibits virus budding. EM images of thin-sectioned cells co-transfected with HIV-1 and Gag-DUb (a) or Gag-DUb* (b). Insets show high-magnification images of a budding or released virus particle and black arrows indicate arrested assembly sites (a) or virions (b). Quantification of budding defects was performed as described in Figure  2. (B) Nedd4-2s-mediated Gag-ubiquitination relieves DUb-TSG101 inhibitory effect on HIV-1 release. Cells were transfected with HA-Ub (lane 1) or HIV-1 YP- alone (lane 2), or with both alone (lane 3), in combination with Strep-DUb-TSG101 (200 ng) (lanes 4) or with the 3 precedent plasmids and Nedd4-2s (150 ng) (lane 5). Cells and virus were harvested after 24 h; cell lysates were incubated with Strep-Tactin beads and the complexes (upper panel), virus pellets (middle panel) and input fractions (lower panels) were probed with the indicated antibodies. (C) Nedd4-2s ubiquitinates Gag at the membrane. Cells expressing HIV YP- were co-transfected with either HA-Ub alone (lane 1), with DUb-TSG101 alone (lane 2) or in combination with Nedd4-2s (lane 3). The P100 fractions of each sample were either analyzed by WB using anti-p24CA antibody (second panel from top) or incubated with anti-HA antibody coated beads. Eluates were analyzed by WB with anti-p24CA antibody to detect ubiquitinated Gag molecules. (D) Nedd4-2s failed to relieve the inhibitory effect of DUb-Gag. 293T cells were transfected with HIV-1 (1 μg) (lane 1), Gag-DUb*Strep (1.5 μg) (lane 2) or with both (lane 3), with Gag-DUbStrep alone (1.5 μg) (lane 4) or with increasing amounts of Nedd4.2s (100 and 200 ng) (lanes 5 and 6). Cells were also transfected with Gag-DUbStrep in combination with HIV-1 (lane 7) or with increasing amount of Nedd4-2s (lanes 8 and 9). Triangles indicate lanes where increasing amounts of Nedd4-2s are expressed.

Figure 7

Ubiquitination of ESCRT-I is not sufficient for HIV-1 release. (A and B) 293T cells were co-transfected with HA-Ub plasmid and Flag-tagged version of ESCRT-I components [TSG101 (2.5 μg), VPS28 (800 ng), VPS37B (1.7 μg), MVB12B (800 ng); these amounts express comparable levels of ESCRT-I proteins] alone, or in addition to the following plasmids: either Gag-Strep (1.5 μg) (lanes 2, 7, 12) or Gag-DUbStrep alone (1.5 μg) (lanes 3, 8, 13) or in combination with the Nedd4-2s expression plasmid (150 ng) (lanes 4, 9, 14) and with Gag-DUb*Strep alone (1.5 μg) (lanes 5, 10, 15). Sequential centrifugations were performed to separate membrane-enriched P100 fractions from which Gag molecules were immunoprecipitated using Strep-Tactin beads. The protein content of captured complexes (A), input (lower left and center panels) and virus (lower right panel) fractions were analyzed by WB with the indicated antibodies. (B) shows a darker exposure of samples analyzed in lanes 6–10).

Figure 8

Incorporation of Gag-Ub into assembly sites alleviates DUb-ESCRT inhibitory effects. A) 293T cells expressing EIAV (lane 1),were also transfected with increasing amounts of EIAV Gag-Ub expression vector (500 ng and 1μg) (lanes 2 and 3), with Flag-DUb-Alix alone (100 ng) (lane 4), in combination with increasingamounts of Gag-Ub (500 ng and 1μg) (lanes 5 and 6), or with inactive DUb*-Alix alone (100 ng) (lane 7) whereas lane 8 shows expression of Gag-Ub alone. Virions and cells were harvested 24 h post-transfection and their protein contents analyzed by WB using an anti-EIAV antibody. Alix and DUb/DUb*-Alix fusion proteins were detected by an anti-Alix antibody. B) 293T cells expressing HIV-1 YP- (lane 1), were also transfected with increasing amounts of HIV Gag-Ub expression vector (1 and 2 μg) (lanes 2 and 3), with DUb-TSG101 alone (200 ng) (lane 4), or in combination with increasing amounts of HIV Gag-Ub expression vector(1 and 2 μg) (lanes 5 and 6) whereas lane 7 shows expression of HIV Gag-Ub alone (1 μg). Virions and cells were harvested 24 h post-transfection and their protein contents analyzed by WB using an anti-p24CA antibody. TSG101 and DUb-TSG101 fusion protein were detected with an anti-TSG101 antibody.