Resistance of Transmitted Founder HIV-1 to IFITM-Mediated Restriction

Abstract

Interferon-induced transmembrane proteins (IFITMs) restrict the entry of diverse enveloped viruses through incompletely understood mechanisms. While IFITMs are reported to inhibit HIV-1, their in vivo relevance is unclear. We show that IFITM sensitivity of HIV-1 strains is determined by the co-receptor usage of the viral envelope glycoproteins as well as IFITM subcellular localization within the target cell. Importantly, we find that transmitted founder HIV-1, which establishes de novo infections, is uniquely resistant to the antiviral activity of IFITMs. However, viral sensitivity to IFITMs, particularly IFITM2 and IFITM3, increases over the first 6 months of infection, primarily as a result of neutralizing antibody escape mutations. Additionally, the ability to evade IFITM restriction contributes to the different interferon sensitivities of transmitted founder and chronic viruses. Together, these data indicate that IFITMs constitute an important barrier to HIV-1 transmission and that escape from adaptive immune responses exposes the virus to antiviral restriction.

Keywords: CD4; HIV-1; IFITM; antiviral restriction; co-receptor; escape; neutralization; transmitted founder virus; type 1 interferon.

Graphical abstract

Figure 1

Sensitivity of HIV-1 to IFITM Inhibition Varies with Co-receptor Usage (A) GFP-encoding HIV-1 vectors pseudotyped with the indicated envelope glycoproteins were used to infect U87/CD4/CCR5 or U87/CD4/CXCR4 cells stably expressing IFITMs 1, 2, or 3 or empty vector (control). The percentage of GFP⁺ cells compared to control was determined by flow cytometry, and results shown represent a mean of independent experiments. Statistics were performed using an unpaired two-tailed t test (^∗∗∗p < 0.001; ^∗∗p < 0.01; ^∗p < 0.05; ns, p > 0.05). (B) Cumulative viral replication at 96 hr in the presence of IFITMs differs for CXCR4-using or CCR5-using HIV-1 virus isolates. U87/CD4/CCR5 or U87/CD4/CXCR4 cells were infected at an MOI of 0.05, and replication was monitored by viral output on HeLa-TZMbl indicator cells (see also Figures S1D and S1E). Each virus point represents a mean of three independent experiments. Inset red dashed boxes: CCR5-using CH105 and RHPA are founder virus isolates that are resistant to IFITM inhibition. Removal of these outliers reveals data significance between CXCR4 and CCR5 users in the presence of IFITM1 (^∗∗∗p < 0.001; ^∗∗p < 0.01; ^∗p < 0.05; ns, p > 0.05, based on an unpaired two-tailed t test). (C) Pseudotyped HIV vectors were produced with wild-type YU2 and Hxb2 envelopes or V3 loop swap variants and analyzed as in (A). All error bars represent ± SEM (n = 3). See also Figure S1.

Figure 2

A Conserved N-Terminal Tyrosine Residue Is Crucial for IFITM Localization and Affects the IFITM Inhibition Phenotypes of HIV-1 (A) Schematic representation of human IFITM1, IFITM2, and IFITM3 proteins, indicating the localization of a conserved tyrosine at the N terminus of IFITMs 2 (Y19) and 3 (Y20). The localization of the IFITM proteins was assessed using U87 cells transduced with the indicated HA-tagged IFITM. Cells were stained with anti-HA (green) antibody and co-stained with early or late endosomal markers anti-EEA1 (red) or anti-CD63 (red). (B) Top: localization of HA-tagged IFITM2-Y19F or IFITM3-Y20F in U87 cells as in (A). Bottom: localization of HA-tagged IFITM2 or IFITM3 in U87, with or without small interfering RNA (siRNA)-mediated knockdown of AP-2, is shown. (C) As in Figure 1C, pseudotyped viruses were produced with wild-type YU2 and Hxb2 envelopes as well as V3 loop swaps. The percentage of infected cells, in the presence of either wild-type or mutant IFITM protein, was determined by flow cytometry (^∗∗∗p < 0.001; ^∗∗p < 0.01; ^∗p < 0.05; ns, p > 0.05, unpaired two-tailed t test). (D) U87/CD4/CXCR4 cells expressing IFITM2 or IFITM2-Y19F were infected with 89.6 env-pseudotyped HIV-1 vector (left) or with full-length HIV-1 89.6 at an MOI of 0.05 (right), and they were analyzed as in Figures 1A and S1D, respectively (^∗∗∗p < 0.001, unpaired two-tailed t test). (E) The effect of AP2 siRNA knockdown in U87/CD4/CXCR4 on the inhibition of 89.6 proviral replication. Cells were treated with control or AP2-specific SMARTpool siRNA and infected with 89.6. Then 48 hr post-infection, supernatants were assessed for viral production, and lysates were examined for the expression of AP2 and loading control HSP90 by western blot. Statistical significance was determined by using an unpaired two-tailed t test (^∗∗∗p < 0.001; ^∗∗p < 0.01; ^∗p < 0.05; ns, p > 0.05). (F) Effect of endocytosis inhibitors dynasore and Pitstop2 on IFITM restriction of pseudotyped virus entry. U87/CD4/CoR cells expressing the indicated IFITM were exposed to dynasore (80 μM), Pitstop 2 (30 μM), or DMSO as a control, for 30 min prior to infection with the indicated env-pseudotyped viruses. The percentage of GFP⁺ cells was determined by flow cytometry (^∗∗∗p < 0.001; ^∗∗p < 0.01; ^∗p < 0.05; ns, p > 0.05, unpaired two-tailed test). All error bars represent ± SEM (n = 3). See also Figure S2.

Figure 3

TF Virus Replication Is Resistant to IFITMs, with Sensitivity Arising at 6 Months (A) U87/CD4/CCR5 IFITM-expressing cells were infected with the indicated TF molecular clones or cognate 6-month variant at an MOI of 0.05, and infection was assayed with supernatants harvested every 24 hr for 7 days. Viral production was measured on HeLa-TZMbl indicator cells. Data represent a summary of three independent experiments. (B) Comparison of the relative cumulative replication in 3A at the 96-hr time point was analyzed by a paired Mann-Whitney test (^∗∗∗p < 0.001; ^∗∗p < 0.01; ^∗p < 0.05; ns, p > 0.05). (C) HIV-1 vectors pseudotyped with envelopes of CH058, CH077, and CH470 TF and 6-month viruses were used to infect U87/CD4/CCR5 IFITM cells, analyzed as in Figure 1A. The percentage of infected cells was determined by flow cytometry, and results shown represent three independent infection experiments (^∗∗∗p < 0.001; ^∗∗p < 0.01; ^∗p < 0.05; ns, p > 0.05, unpaired two-tailed test). (D) Sequential envelopes derived from the CAP256 patient (CAP256.1MO.C7J [1 month], CAP256.3MO.C9 [3 months], CAP256.8MO.31 [8 months], CAP256.12MO.1 [12 months], CAP256.14MO.5b [14 months], CAP256.21MO.A1 [21 months], and CAP256.39MO.10 [39 months]) were used to produce pseudotyped HIV-1 vectors and infect of U87/CD4/CCR5 IFITM cells (^∗∗p < 0.01; ^∗p < 0.05; ns, p > 0.05, unpaired two-tailed test). All error bars represent ± SEM (n = 3). See also Figure S3 and Table S1.

Figure 4

Reversion of Neutralization Escape Mutants in 6-Month Viruses Restores IFITM Resistance (A) The gp120:gp41 trimer structures show the NAb escape mutations of CH040, CH058, and CH077 in red. Gp120, green; gp41, blue. Images were drawn using PDB: 5ACO in Pymol. (B) U87/CD4/CCR5 IFITM cells were infected with the indicated 6-month chronic viruses or those containing NAb escape reversion mutations at an MOI of 0.05. Cumulative replication over a period of 96 hr was assayed for viral production on HeLa-TZMbl indicator cells as in Figures 3A and 3B. Statistics were performed using a paired Mann-Whitney test (^∗∗∗p < 0.001; ^∗∗p < 0.01; ^∗p < 0.05; ns, p > 0.05). See also Table S2.

Figure 5

Effect of Endocytosis Inhibitors and Surface CD4 Levels on IFITM Restriction of TF and 6-Month Viral Entry (A) U87/CD4/CCR5 IFITM cells were treated with control or AP-2-specific siRNAs for 48 hr and then infected with the indicated HIV-1-pseudotyped vector. GFP⁺ cells were analyzed as in Figure 1A (^∗∗∗p < 0.001; ^∗∗p < 0.01; ^∗p < 0.05; ns, p > 0.05, unpaired two-tailed test). Knockdown was assessed by western blot for AP-2μ. (B) As in (A) but the cells were pretreated with DMSO, dynasore, or Pitstop2. (C) U87/CD4/CCR5 IFITM-expressing cells were infected with TF viruses and their matched chronic pairs at an MOI of 0.5 in the presence of 0, 10, or 100 ng/mL of CD4-blocking antibody SK3 for 6 hr. Then 48 hr post-infection, virus production was measured by infection of HeLa-TZMbl indicator cells (^∗∗p < 0.01; ^∗p < 0.05; ns, p > 0.05, unpaired two-tailed t test). All error bars represent ± SEM (n = 3). See also Figure S4.

Figure 6

IFITM2 and IFITM3, in Particular, Contribute to the Inhibition of HIV-1 Replication in Primary CD4⁺ T Cells (A) HIV-1 vectors pseudotyped with TF and 6-month envelope variants were used to challenge U87/CD4/CCR5 cells that had been pretreated overnight with 1,000 U/mL universal IFN-1. Infected cells were analyzed by flow cytometry 48 hr later (^∗p < 0.05; unpaired; unpaired two-tailed t test). (B) U87/CD4/CCR5 cells were transduced with IFITM-specific or control lentiCRISPR-p2A-GFP vectors sufficiently to achieve 80%–98% GFP⁺. The efficiency of CRISPR knockout of IFITM expression, in the absence or presence of IFN-1 (1,000 U/mL), was determined by western blotting. HSP90 served as a loading control. (C) Cells from (B) were challenged with CH077 at an MOI of 0.05 in the presence of absence of 1,000 U/mL IFN-1, and cumulative replication at 96 hr was determined as previously described. (D) Primary CD4⁺ T cells were transduced with shRNAs targeting IFITM1, 2, or 3 or a control shRNA and cultured for 72 hr. Transduced cells were cultured with or without 500 U/mL IFN-1 for 24 hr before determining the efficiency of IFITM knockdown by western blot analysis. HSP90 served as a loading control. (E) CD4⁺ T cells, expressing control or IFITM-specific shRNAs, were infected with the indicated TF and chronic matched-pair viruses at an MOI of 0.05. A time course of replication, in the presence or absence of 500 U/mL universal IFN-1, was assayed with supernatants harvested every 24 hr for a total of 5 days. Infectious virus was determined by infection of HeLa-TZMbl indicator cells. Data shown are representative of the replication values at 96 hr, and IFITM shRNA knockdown data are expressed relative to the virus-specific TF control sample. Data shown are representative of three independent donors and independent experiments. Statistical significance between indicated pairs was determined using an unpaired two-tailed t test (^∗∗p < 0.01; ^∗p < 0.05; ns, p > 0.05). All error bars represent ± SEM (n = 3). See also Figure S5.