xCT/SLC7A11 antiporter function inhibits HIV-1 infection

Abstract

Human macrophages are protected by intrinsic antiviral defenses that provide moderate protection against HIV-1 infection. Macrophages that do become infected can serve as long-lived reservoirs, to disseminate HIV-1 to CD4⁺ T cells. Infection of macrophages with HIV-1 and HIV-2 is inhibited by constitutive mobilization of antioxidant response master transcription regulator Nrf2. The downstream mediator of this restriction was not identified. Among the tens of genes controlled directly by Nrf2 in macrophages, we found that xCT/SLC7A11, a 12-transmembrane, cystine-glutamate antiporter promotes antiretroviral activity. We show here that depletion of xCT mRNA increases HIV-1 infection. Reconstitution of xCT knock out cells with wild-type xCT but not a transport-deficient mutant restores anti-HIV-1 activity. Pharmacological inhibitors of xCT amino acid transport also increase infection. The block is independent of known restriction factors and acts against HIV-1 and HIV-2. Like the block triggered through Nrf2, xCT function impedes infection immediately before 2-LTR circle formation.

Keywords: 2-LTR circle; Erastin; HIV-1; HIV-2; Late reverse transcription; Restriction; SLC7A11; Sulfasalazine; Sulforaphane; xCT.

Fig 1:. Constitutive levels of xCT expression inhibit HIV-1 in primary hMDMs and differentiated human monocytic cells.

(A) 2.5×10⁵ hMDMs from three different donors (D1–3) were mock-transfected or transfected with a non-targeting siRNA or xCT-targeting siRNA twice in four days. Forty-eight hours after the second transfection, the cultures were mock infected or infected with VSV-G pseudotyped HIV-1 luciferase reporter virus. Twenty-four hours after infection, cells were lysed, and luciferase activity was measured. (B) Representative samples from (A) were lysed in Laemmli buffer and run on an SDS-PAGE gel for characterization by western blot with antibodies specific for xCT or tubulin. (C) Densitometry was performed on the bands from (B) and signals were normalized to the non-targeting siRNA group. (D) 5×10⁴ PMA-Differentiated THP-1s were transfected with a non-targeting siRNA or xCT-targeting siRNA once and infected three days later with VSV-G pseudotyped HIV-1 luciferase reporter virus. Twenty-four hours after infection, cells were lysed, and luciferase activity was measured. (E) Similar to (D) except cultures were infected with HIV-1 dual tropic 89.6 Env pseudotyped HIV-1 luciferase reporter virus. (F) Representative samples from (D) were lysed in Laemmli buffer and run on an SDS-PAGE gel for characterization by western blot with antibodies specific for xCT or tubulin. (G) 5×10⁴ WT or xCT KO PMA-differentiated THP-1 cells were treated with 10 μM SFN or an equivalent volume of DMSO in complete RPMI-1640 for twenty-four hours before being lysed in Laemmli buffer and run on an SDS-PAGE gel for characterization by western blot with antibodies specific for xCT, Nrf2, or tubulin. (H) 5×10⁴ WT or xCT KO THP-1 cells were treated as in (G) for twenty-four hours before infection with VSV-G pseudotyped HIV-1 luciferase reporter virus. Twenty-four hours later, cells were lysed and luciferase activity was measured. The data represent mean values ± SEM from three independent experiments.

Fig 2:. Reintroduction of wild-type xCT, but not a transport defective xCT mutant, inhibits HIV-1 in xCT KO THP-1 macrophages.

(A) Graphical representation of xCT showing the amino acid residue numbers of key segments and the relative location of the C327L mutation. (B) xCT KO THP-1 cells were transduced with either an empty lentiviral vector or lentiviral vectors coding for WT xCT (xCTi) or xCT_C327L (C327L xCTi) with expression controlled by a doxycycline-inducible promoter. PMA-differentiated THP-1 cells were treated for twenty-four hours with media containing vehicle or 2 μg/mL doxycycline. Lysates from these cells were characterized by western blotting with antibodies specific for xCT or tubulin. (C) 5×10⁴ PMA-differentiated THP-1 cells, as described in (A) were cultured for six days in complete RPMI-1640 supplemented with PMA, in the presence or absence of 2 μg/mL doxycycline before infection with VSV-G pseudotyped HIV-1 luciferase reporter virus. Twenty-four hours after infection cells were lysed and luciferase activity was measured. (D) Radiotracer assay of xCT-mediated cystine transport in PMA-differentiated WT, xCT KO, xCT KO xCTi, and xCT KO C327L xCTi THP-1 cells treated for forty-eight hours with 2 μg/mL doxycycline. xCT activity assays were performed in chemically defined medium containing 12.5 μM L-[¹⁴C] cystine and inhibitors of alternate cystine transporters as specified in Material and Methods. The xCT activity was quantitated by measuring intracellular L-[¹⁴C] cystine accumulation during a twenty-minute incubation at 37° C and further normalized to the specific activity of L-[¹⁴C] cystine and total protein content. The data represent mean values ± SEM from three independent experiments.

Fig 3:. Pharmacological inhibitors of xCT transport activity increase HIV infection efficiency in differentiated human macrophages.

(A) 5×10⁴ PMA-differentiated WT or xCT KO THP-1 cells were treated with 0.625 mM SSZ or an equivalent volume of DMSO in complete RPMI-1640 for twenty-four hours before infection with VSV-G pseudotyped HIV-1 luciferase reporter virus. Twenty-four hours later, cells were lysed and luciferase activity was measured. (B) 5×10⁴ PMA-differentiated THP-1 cells were infected with VSV-G pseudotyped HIV-1 luciferase reporter virus as in panel (A) except the cultures were treated with 20 μM erastin. (C) 5×10⁴ hMDMs were infected with VSV-G pseudotyped HIV-1 luciferase reporter virus as in (A) except cultures were treated with 1.25 or 2.5 mM SSZ. (D) 5×10⁴ PMA-differentiated SAMHD1-KO THP-1 cells were treated and infected as in (A). (E) 5×10⁴ PMA-differentiated WT were treated as in (A) twenty-four hours before infection with VSV-G pseudotyped HIV-2 luciferase reporter virus. (F) 5×10⁴ PMA-differentiated WT were treated as in (A) twenty-four hours before infection with VSV-G pseudotyped HIV-1 vpr(−) luciferase reporter virus. The data represent mean values ± SEM from three independent experiments.

Fig 4:. SSZ increases infection only when it is applied before the virus.

(A) Cultures of 5×10⁴ PMA-differentiated WT THP-1 cells were treated with 0.625 mM SSZ or an equivalent amount of DMSO in complete RPMI-1640 containing PMA at the indicated timepoints. Cell cultures were infected with VSV-G pseudotyped HIV-1 luciferase reporter virus. Seventy-two hours later, cells were lysed, and luciferase activity was measured. The data represent mean values ± SEM from three independent experiments.

Fig 5:. xCT mediates a block to HIV that is active in dividing cells.

(A) WT THP-1 monocytes, xCT KO THP-1 monocytes, HeLa, HEK293T, and SupT1 cells were lysed and characterized by western blot with antibodies for xCT or tubulin. (B) 5×10⁴ undifferentiated WT and xCT KO THP-1 cells were treated with 0.625 mM SSZ or an equivalent amount of DMSO in complete RPMI-1640 for twenty-four hours before infection with VSV-G pseudotyped HIV-1 luciferase reporter virus. Twenty-four hours later, cells were lysed and luciferase activity was measured. (C) 1×10⁴ HeLa cells were treated and infected as in (B) except treated in complete DMEM. (D) As in (C) except with HEK293T. (E) As in (B) except with SupT1 cells. The data represent mean values ± SEM from three independent experiments.

Fig 6:. xCT inhibits HIV-1 following reverse transcription but before 2-LTR circle formation.

(A) 10⁵ WT THP-1 cells were mock treated or treated with 0.625 mM SSZ an equivalent amount of DMSO in complete DMEM or with 10 μM AZT. Forty-eight hours after treatment with SSZ or DMSO, and twenty-four hours after AZT treatment, cells were mock infected or infected with VSV-G pseudotyped HIV-1 GFP reporter virus. Heat inactivated virus served as a control for plasmid carryover. Eight hours later, cells were lysed and total DNA was extracted. Samples were probed with primers targeting sequence markers for late reverse transcription by qPCR. (B) As with (A) except cells were lysed twenty-four hours post infection and probed for 2-LTR circles. (C) As with (B) except integrated HIV-1 gag and host genomic Alu sequences are amplified in the first round of PCR before probing for markers of integrated viral DNA by qPCR. (D) WT THP-1 monocytes were lysed and characterized by western blot with antibodies for MxB or tubulin. The data represent mean values ± SEM from three independent experiments.