IFITM Proteins Restrict HIV-1 Infection by Antagonizing the Envelope Glycoprotein - PubMed (original) (raw)

IFITM Proteins Restrict HIV-1 Infection by Antagonizing the Envelope Glycoprotein

Jingyou Yu et al. Cell Rep. 2015.

Abstract

The interferon-induced transmembrane (IFITM) proteins have been recently shown to restrict HIV-1 and other viruses. Here, we provide evidence that IFITM proteins, particularly IFITM2 and IFITM3, specifically antagonize the HIV-1 envelope glycoprotein (Env), thereby inhibiting viral infection. IFITM proteins interact with HIV-1 Env in viral producer cells, leading to impaired Env processing and virion incorporation. Notably, the level of IFITM incorporation into HIV-1 virions does not strictly correlate with the extent of inhibition. Prolonged passage of HIV-1 in IFITM-expressing T lymphocytes leads to emergence of Env mutants that overcome IFITM restriction. The ability of IFITMs to inhibit cell-to-cell infection can be extended to HIV-1 primary isolates, HIV-2 and SIVs; however, the extent of inhibition appears to be virus-strain dependent. Overall, our study uncovers a mechanism by which IFITM proteins specifically antagonize HIV-1 Env to restrict HIV-1 infection and provides insight into the specialized role of IFITMs in HIV infection.

PubMed Disclaimer

Figures

Graphical abstract

Figure 1

Expression of IFITMs Differentially Inhibits HIV-1 Cell-to-Cell Infection (A) Donor Jurkat/inGLuc cells expressing or not expressing IFITMs were infected with HIV-1NL43 pseudotypes bearing VSV-G and cocultured with target Jurkat or Jurkat cells stably expressing IFITMs. (B) HEK293T cells were transiently transfected with HIV-1 NL43/inGLuc plus Env in the presence or absence of IFITMs and were cocultured with Jurkat cells stably expressing or not expressing IFITMs. (C) Effect of IFITMs on cell-free HIV-1 infection. HIV-1NL43 virions were produced from 293T cells expressing or not expressing IFITMs, and cell-free infectivity was determined by infecting HeLa-TZM cells. (D) Expressions of IFITM proteins in Jurkat/inGLuc, Jurkat and 293T cells determined by western blotting. β-Actin serves as loading control. (E) Effect of AMD3100 (4 μg/m), LAT-B (1 μg/ml), and VRC01 (5 μg/ml) on HIV-1 transfer from 293T to Jurkat cells. All reagents were added immediately upon coculturing and maintained throughout the entire assay. (F) Quantification of HIV-1 transmission/fusion from 293T to Jurkat/tdTomato cells using live-cell imaging. HEK293T cells were transfected with HIV-1NL43 Gag-iGFP and cocultured with Jurkat/tdTomato cells for 20–24 hr. The numbers of GFP- and tdTomato-double-positive cells were scored and plotted relative to that of empty vector pQCXIP control. For all figures applicable, the values are means and SDs of three to five independent experiments. ∗p < 0.05; ∗∗p < 0.01; ∗∗∗p < 0.001. See also Figure S1.

Figure 2

Knockdown of IFITM Expression in CD4+ PM1 Cells or PBMCs Promotes HIV-1 Transmission to Jurkat Cells (A) Examination of IFITM protein expression in PM1 cells stably expressing shRNA IFITM1, 2, and 3. Cells were treated with or without 500 U IFNα2b for 18–24 hr, and lysates were subjected to western blotting using anti-IFITM1, anti-IFITM2, or anti-IFITM3 antibodies. Cross-knockdown and antibody cross-reactions among IFITM1, 2, and 3 are noticed. GAPDH serves as loading control due to weak signals for β-actin in PM-1 cells. (B) Summary of flow cytometric analysis of HIV-1 Gag-GFP transfer from PM1 to Jurkat cells. PM-1 cells stably expressing shRNA IFITM1, 2, or 3 or scramble control were infected with NL43-Gag-iGFP; cells were cocutured for 4–8 hr with Jurkat cells that were prelabeled with CMTMR. Cells were gated for GFP (Green) and CMTMR (Red) by flow cytometry (see Figure S2). For 8 hr coculture, cells were also treated with AZT to ensure that productive HIV-1 replication had occurred. Data were from four independent experiments. All values shown are normalized against shRNA control either at 4 or 8 hr after coculture, ∗p < 0.05; ∗∗p < 0.01. (C–E) Knockdown of IFITM expression in PBMCs enhances HIV-1 cell-to-cell infection. Activated PBMCs were treated with 500 U IFNα2b and infected with VSV-G lentiviral pseudotypes encoding shRNA IFITM1, 2, or 3 for 48 hr. The knockdown efficiency was examined by intracellular staining using pooled antibodies against IFITM1, 2, and 3 (1:1:1) by flow cytometry shown in (C). Note that shRNA against IFITM1 knocks down both IFITM1 and IFITM3, shRNA against IFITM2 only knocks down IFITM2, and shRNA against IFITM3 knocks down all three IFITMs. Next, the cells were infected with HIV-1NL43-iGFP bearing VSV-G, which served as donor cells, and cocultured with target Jurkat cells stably expressing tdTomato (Jurkat/tdTomato). The cell-to-cell infection efficiency was determined by measuring GFP+ Jurkat cells 48 hr after coculture using flow cytometry (D) and summarized in (E). ∗p < 0.05. Results are from four independent experiments performed in duplicates. See also Figure S2.

Figure 3

IFITM Proteins Interact with HIV-1 Env and Impair its Processing and Incorporation into Viral Particles 293T cells were transiently transfected with HIV-1NL43 in the presence or absence of IFITM plasmids. The cells were lysed with a membrane-disrupting RIPA buffer containing 0.1% SDS and 1% NP40, and lysates were immunoprecipitated with anti-FLAG antibody. (A) The expression of HIV-1 Env, Gag, and IFITMs in transfected cells was examined by using anti-gp120, gp41, anti-p24, and anti-FLAG, respectively. β-Actin serves as a loading control. The ratios of gp120 versus gp160 and gp41 versus gp160 in the cell lysates were determined by quantifying the band intensities of gp120, gp41, and gp160 using Quantity One (Bio-Rad). (B) Detection of the interaction between IFITMs and HIV-1 Env in viral producer cells. Western blotting was performed by using anti-gp120 (top panel) or anti-gp41 (middle panel) to detect HIV-1 Env. (C) Quantification of HIV-1 Env processing. The results represent average ± SD of four independent experiments. ∗∗p < 0.01; ∗∗∗p < 0.001. (D) Endogenous IFITMs interact with HIV-1 Env in PM1 cells. PM1 cells were infected with HIV-1NL43 or mock-infected, and cells were lysed by RIPA buffer 48 hr after infection. Lysates were subjected to immunoprecipitation using pooled antibodies against IFITM1, 2, or 3, or with a normal rabbit Ig that serves as negative control; western blots were performed using anti-gp41. (E) Detection of IFITMs in supernatants of 293T cells not transfected with HIV-1 plasmid. Supernatants were concentrated by ultracentrifugation using 20% sucrose, and pellets were analyzed by western blotting using anti-FLAG or anti-p24. (F and G) IFITMs are incorporated into HIV-1 virions with different efficiencies, which do not correlate with their phenotypes in inhibiting HIV-1 cell-to-cell infection. Viral supernatants were concentrated from cells used for Figures 3A and 3B by ultracentrifugation (20% sucrose), followed by immunoprecipitation (IP) using HIV-Ig. Western blotting was performed by using anti-FLAG (to detect IFITMs), anti-gp120, or anti-p24. The gp120 band intensities were quantified by Quantity One, with relative values to the vector control indicated. No IFITM protein can be pulled down by HIV-Ig from concentrated supernatant of cells not expressing IFITMs (data not shown). Results represent three independent experiments. See also Figure S3.

Figure 4

IFITM3 Protein Promotes HIV-1 gp120 Shedding and Increased gp160 Turnover HEK293T cells were transfected with plasmids encoding HIV-1 NL4-3 provirus in the presence or absence of IFITM3. Cells were pulse-labeled with 35S-Met/Cys at 37°C for 1 hr and chased in unlabeled medium for the times indicated. The cell lysates and supernatants were immunoprecipitated with HIV-Ig at 4°C overnight and separated by SDS-PAGE. (A) Representative phosphorimages of cell lysates (top) and supernatants (bottom) from three independent experiments are shown. (B–D) Quantifications of the phosphorimaging intensity of HIV-1 gp160 (B), Gag Pr55 (C), and shed gp120 (D). Note that in (B) and (C), we set the signals of HIV-1 gp160 or Gag Pr55 in the absence and presence of IFITM3 to 1.0, respectively. In (D), we set the signal of gp120 in the absence of IFITM3 at 6 as 1.0 for comparison.

Figure 5

The IFITM C Terminus Regulates the Inhibition of HIV-1 Cell-to-Cell Infection and Membrane Fusion (A and B) Effect of IFITM1/2 chimeras and IFITM3 N-terminal mutants on HIV-1 cell-to-cell infection. The experiments were performed using 293T as donor cells and Jurkat as target cells. (C) Effect of parental IFITMs and IFITM1/2 chimeras on HIV-1 virion fusion during cell-to-cell infection. Note that the HIV-1 BlaM-Vpr-based fusion assay was implemented into cell-to-cell infection experiments (see details in Experimental Procedures). (D and E) Effect of IFITMs and chimeras on HIV-1 Env-mediated cell-cell fusion. 293T cells were transfected with HIV-1NL43 along with LXSN that encodes HIV-1 Tat, in the presence or absence of IFITMs. The transfected 293T cells (effector cells) were cocultured with parental HeLa-TZM or HeLa-TZM stably expressing IFITMs (target cells) for 12–24 hr. Cells were lysed, proteins were quantified, and firefly activity was measured. All samples were done in duplicate. Results shown represent three to four independent experiments. (F) Effect of IFITMs in effector cell, target cells, or both on IAV HA-mediated cell-cell fusion. Assays were carried out as described in (D) and (E), except that IAV HA was tested and cocultured cells were treated with pH 5.0 for 1 min following coculture. Results represent three to four independent experiments. For all figures, relative numbers are presented. ∗p < 0.05; ∗∗p < 0.01; ∗∗∗p < 0.001.

Figure 6

Mutations in the HIV-1 env Gene Provide Escape from IFITM-Mediated Inhibition of Viral Replication and Cell-to-Cell Infection (A) Replication kinetics of HIV-1NL43 in SupT1 cells expressing IFITM1 WT or its two mutants, IFITM1Δ(117–125) and IFITM1Δ(108–125). The expression of IFITM1 and its mutants in the cells was induced by doxycycline (Dox) (500 ng/ml), viral replication was determined by measuring the levels of viral RT activity in culture supernatants. (B) Replication kinetics of HIV-1NL43 WT and mutants Vpu40, EnvA539V, and Vpu40/EnvA539V in SupT1 cells expressing (with Dox) or not expressing IFITM1Δ(117–125) (without Dox). (C) Effect of IFITMs on cell-to-cell infection of HIV-1 escape mutants. Experiments were performed as described in the legend of Figures 1A–1C, except HIV-1NL43 mutants Vpu40, EnvA539V, and Vpu40/EnvA539V bearing VSV-G were used for infection of Jurkat/inGLuc cells, which served as donors. Because of the difference between WT and HIV-1 mutants in infectivity in the absence of IFITMs, we plotted the absolute Gluc readouts (RLU) in this and other related figures. (D) Effect of IFITMs on Jurkat-to-Jurkat transfer mediated by the Env cytoplasmic tail deletion mutant (NL43 EnvΔCT). (E) Effect of IFITMs on 293T-to-Jurkat transfer of NL43 EnvΔCT. (F) Effect of IFITMs on cell-cell fusion (between 293T expressing Env and HeLa-TZM) mediated by the Env protein of NL43 EnvΔCT. For all data shown in (A)–(F), results shown represent at least three independent experiments. ∗p < 0.05; ∗∗p < 0.01; ∗∗∗p < 0.001. See also Figure S4.

Figure 7

The Ability of IFITMs to Inhibit Cell-to-Cell Infection/Fusion by HIV-1 BH10, YU2, AD8, HIV-2, and SIVs Are IFITM-Species and Virus-Strain Dependent 293T cells were transfected HIV-1 proviral DNAs encoding BH10, YU2, NL43(AD8) (abbreviated AD8), NL43 along with a VSV-G-expressing plasmid (A) and (B), or the molecular clones of HIV-2 or SIVs (C) and (D). For HIV-1 (A) and (B), the produced viruses were used to infect Jurkat/inGLuc cells overnight, and after thorough washing with PBS, the infected Jurkat/inGLuc cells were cocultured with Jurkat cells (for BH10 and NL43) or CD4+ PBMCs (for YU2 and AD8). For HIV-2 and SIVs (C and D), the transfected 293T cells were cocultured with HeLa-TZM. Cell-to-cell infection/fusion efficiency was determined by measuring GLuc (A) or firefly luciferase activity (D) 24–48 hr after coculture. Cell-free viral infection was performed in parallel, with exactly the same numbers of normal donor plus target cells as described in the text and Experimental Procedures (B and D). ∗∗p < 0.01; ∗∗∗p < 0.001. Because different HIV and SIV isolates have different infectivities in the absence of IFITMs, we also plotted the data using absolute Gluc readouts shown in Figures S5A–S5D. For comparisons between cell-to-cell and cell-free viral infection, please refer to Figures S5E and S5F.

References

1. Agosto L.M., Zhong P., Munro J., Mothes W. Highly active antiretroviral therapies are effective against HIV-1 cell-to-cell transmission. PLoS Pathog. 2014;10:e1003982. - PMC - PubMed
1. Amini-Bavil-Olyaee S., Choi Y.J., Lee J.H., Shi M., Huang I.C., Farzan M., Jung J.U. The antiviral effector IFITM3 disrupts intracellular cholesterol homeostasis to block viral entry. Cell Host Microbe. 2013;13:452–464. - PMC - PubMed
1. Brass A.L., Huang I.C., Benita Y., John S.P., Krishnan M.N., Feeley E.M., Ryan B.J., Weyer J.L., van der Weyden L., Fikrig E. The IFITM proteins mediate cellular resistance to influenza A H1N1 virus, West Nile virus, and dengue virus. Cell. 2009;139:1243–1254. - PMC - PubMed
1. Chesarino N.M., McMichael T.M., Hach J.C., Yount J.S. Phosphorylation of the antiviral protein interferon-inducible transmembrane protein 3 (IFITM3) dually regulates its endocytosis and ubiquitination. J. Biol. Chem. 2014;289:11986–11992. - PMC - PubMed
1. Chutiwitoonchai N., Hiyoshi M., Hiyoshi-Yoshidomi Y., Hashimoto M., Tokunaga K., Suzu S. Characteristics of IFITM, the newly identified IFN-inducible anti-HIV-1 family proteins. Microbes Infect. 2013;15:280–290. - PMC - PubMed

IFITM Proteins Restrict HIV-1 Infection by Antagonizing the Envelope Glycoprotein - PubMed (original) (raw)