Ataxia-telangiectasia-mutated (ATM) protein can enhance human immunodeficiency virus type 1 replication by stimulating Rev function

Abstract

The ataxia-telangiectasia-mutated (ATM) kinase plays a central role in responses to various forms of DNA damage and has been suggested to facilitate human immunodeficiency virus type 1 (HIV-1) integration. Here, we describe a series of experiences that indicate that ATM can enhance HIV-1 replication by stimulating the action of the Rev viral posttranscriptional regulator. The Rev-dependent stimulation of viral late gene expression was observed with ATM-overexpressing cells, a result confirmed with a Rev-dependent reporter construct. Both parameters were also enhanced upon treatment of HeLa cells with caffeine, a xanthine that, in this cellular context, stimulates ATM activity. As well, decreased levels of virions with reduced infectivity were released by ATM knockdown cells. Notably, ATM overexpression did not stimulate the HIV-1 late gene expression within the context of Rev-independent constructs or the Rex-dependent production of capsid from human T-cell leukemia virus type 1 proviral constructs. Altogether, these results indicate that ATM can positively influence HIV-1 Rev function.

FIG. 1.

Overexpression of ATM enhances HIV-1 replication in a kinase-independent manner. (a) 293T cells (2 × 10⁵ cells) were cotransfected with an HIV-1 molecular clone (1 μg) and wild-type (wt) or kinase-defective (kd) ATM-expressing plasmid or the empty vector pcDNA3-FLAG (2 μg). Three days after transfection, p24 levels in the culture supernatant were measured by p24 ELISA. (b) Overexpression of ATM enhances VSV G-pseudotyped HIV-1 production. 293T cells (2 × 10⁵ cells) were cotransfected with HIV-1 packaging construct pCMV R8.91 (2 μg), HIV-GFP (WPTS-GFP) (1 μg), VSV G-envelope-expressing plasmid (pMDG) (1 μg), and wt or kd ATM-expressing plasmids or the empty vector pcDNA3-FLAG (2 μg). Three days after transfection, p24 levels in the culture supernatant were measured by ELISA.

FIG. 2.

Overexpression of ATM enhances HIV-1 replication at a posttranscriptional level. (a) Overexpression of ATM does not affect Tat-mediated transcription from HIV-1 LTR. 293T cells (2 × 10⁴ cells) were cotransfected with the HIV-1-LTR-luciferase (HIV-1-LTR-Luc) reporter gene (100 ng), Tat-expressing plasmid (Tat101-FLAG) (100 ng), and/or ATM-expressing plasmid (200 ng). Twenty-four hours after transfection, luciferase activity in the cellular lysates was measured. Results from three independent experiments are shown, with error bars indicative of the standard deviations from the means. (b) Overexpression of ATM increases production of late viral proteins. 293T cells (2 × 10⁵ cells) were cotransfected with an HIV-1 molecular clone (1 μg) and an ATM-expressing or a control plasmid (2 μg). Three days after transfection, Western blotting of the cellular lysate was performed with anti-ATM, anti-p24 CA, anti-p17 MA, anti-Nef, or anti-Chk2 antibody. (c) Northern blot analysis of cytoplasmic RNA from 293T cells cotransfected with an HIV-1 molecular clone with or without an ATM-expressing plasmid. The loading control was rRNA (18S and 28S rRNA) stained with ethidium bromide.

FIG. 3.

Overexpression of ATM enhances Rev function. 293T cells (2 × 10⁴ cells) were cotransfected with the Rev-dependent luciferase-based reporter gene pDM628 (100 ng), the Rev-expressing plasmid pcRev (100 ng), and/or wild-type (wt) or kinase-defective (kd) ATM-expressing plasmids (200 ng). Twenty-four hours after transfection, luciferase activity in the cellular lysates was measured. Results are from three independent experiments.

FIG. 4.

ATM effect is specific to Rev-dependent gene expression. (a) 293T cells (2 × 10⁵ cells) were cotransfected with the Rev/RRE-dependent gag-pol construct pCMV R8.91, the Rev-independent codon-optimized HIV-1 gag-pol construct pSYNGP, or an HIV-1 gag-pol construct containing four tandem copies of the Mason-Pfizer monkey virus CTE, pGPV-4×CTE (2 μg), as well as ATM-expressing or empty vectors (2 μg). Three days after transfection, Western blotting of the cellular lysate was performed with anti-p24 CA or anti-Chk2 antibody, and p24 levels in the supernatant were measured by ELISA. (b) Overexpression of ATM increases HIV-1 p24 only in the presence of Rev. 293T cells were cotransfected with an RRE-containing HIV-1 gag-pol construct (pGPV-RRE) (2 μg), a Rev-expressing plasmid (1 μg), and/or an ATM-expressing plasmid (2 μg). Western blotting was performed as described for panel a. (c) Overexpression of ATM does not affect HTLV-1 Rex function. 293T cells were cotransfected with the HTLV-1 molecular clone λHTLV-1C or K30p (2 μg) with or without an ATM-expressing plasmid (2 μg). Three days after transfection, Western blotting of the cellular lysate was performed with anti-HTLV-1 p24 or anti-Chk2 antibody. C, control.

FIG. 5.

Reduced HIV replication in ATM knockdown cells. (a) Inhibition of endogenous ATM and ATR protein expression by shRNA-producing lentiviral vectors. Results are shown for Western blotting of cellular lysates with anti-ATM, anti-ATR, and anti-Chk2 antibodies in ATM knockdown (ATMi), ATR knockdown (ATRi), and double-knockdown (DKD) P4.2 cells as well as in P4.2 cells transduced with a control (C) lentiviral vector. (b) Each line was infected with HIV-1 at an MOI of 0.5. HIV-1 replication was assayed by p24 ELISA with the culture supernatants 6 days later. (c) Inhibition of endogenous MRE11 protein expression by shRNA-producing lentiviral vector. MRE11 knockdown (MRE11i) P4.2 cells were infected with HIV-1 at an MOI of 0.5. HIV-1 replication was assayed by p24 ELISA with the culture supernatants 7 days later. (d) HIV-1 Tat-mediated transcription in ATM knockdown HeLa cells. Tat-expressing plasmid (100 ng) and HIV-1-LTR-luciferase (HIV-1-LTR-Luc) reporter plasmid (100 ng) were cotransfected into control P4.2 cells (P4.2:C) or ATM knockdown P4.2 cells (P4.2:ATMi) (2 × 10⁴ cells). A luciferase assay was performed 24 h later. Results are from three independent transfections. (e) Tat-expressing plasmid (100 ng) was transfected into P4.2 cells in triplicate. β-Gal activity of LTR-LacZ-containing cellular lysates was measured at an optical density at 570 nm 24 h later.

FIG. 6.

Decreased production of HIV-1 virions with reduced infectivity from ATM knockdown cells. (a) RT activity in the supernatants of control (C) or ATM knockdown (ATMi) P4.2 cells was measured 2 days after infection with HIV-1 at an MOI of 0.5. (b) Virion particle production in the supernatants of control or ATM knockdown P4.2 cells 55 h after transduction with VSV G-pesudotyped HIV-1 at an MOI of 5. Cells treated with 50 μM zidovudine (AZT) served as a negative control. Results are indicative of duplicate measurements that gave quasi-identical numbers. (c) ATM knockdown P4.2 cells (P4.2:ATMi) or control P4.2 cells (P4.2:C) were infected with wild-type HIV-1 (X4) at an MOI of 0.5. Normalized amounts of virions (as assessed by RT activity) harvested 6 days later from the supernatant of these cells were used to infect P4.2 cells. Virion infectivity was determined by X-Gal (5-bromo-4-chloro-3-indolyl-β-d-galactopyranoside) staining of the LTR-LacZ-containing target cells. Results from two independent experiments are shown. MAGI, multinucleate activation of galactosidase indicator. (d) p24 production in the supernatants of control and ATM knockdown P4.2 cells was measured by ELISA at indicated days after infection with HIV-1 at an MOI of 0.5. Percentages at bottom represent the ratio between ATM knockdown and control cell values. ND, under detection level.

FIG. 7.

Effect of caffeine on the late steps of HIV-1 replication. (a) Caffeine induces hyperphosphorylation of the ATM kinase substrate Chk2, following hydroxyurea treatment. P4.2 cells were pretreated with 4 mM caffeine for 1 h and then treated with 5 mM hydroxyurea (HU) in the presence or absence of 4 mM caffeine for 3 h. Western blotting of the cellular lysates was performed with anti-Chk2 antibody or anti-phospho-Chk2 (Thr68) antibody (P-T68). (b) Caffeine enhances HIV-1 replication in P4.2 cells. P4.2 cells (2 × 10⁵ cells) were pretreated with the indicated concentration of caffeine for 1 h and infected with HIV-1 (X4) at an MOI of 0.5. p24 levels in the supernatants were measured by ELISA 6 days later. (c) ATM knockdown (ATMi) or control (C) P4.2 cells (2 × 10⁵ cells) were pretreated with 1 mM caffeine for 1 h and infected with HIV-1 (X4) at an MOI of 0.5 in the presence or absence of caffeine. p24 levels in the supernatants were measured at 6 days. (d) Caffeine enhances Rev function in P4.2 cells. P4.2 cells (2 × 10⁴ cells) treated with 1 mM caffeine were transfected with the Rev-dependent luciferase-based reporter gene pDM628 (100 ng) with or without the Rev-expressing plasmid pcRev (100 ng). Twenty-four hours later, luciferase activity in the cellular lysates was determined. Results are from three independent experiments.