Exogenous Nef is an inhibitor of Mycobacterium tuberculosis-induced tumor necrosis factor-alpha production and macrophage apoptosis

Abstract

Human immunodeficiency virus-1 (HIV-1) impairs tumor necrosis factor-alpha (TNF-alpha)-mediated macrophage apoptosis induced by Mycobacterium tuberculosis (Mtb). HIV Nef protein plays an important role in the pathogenesis of AIDS. We have tested the hypothesis that exogenous Nef is a factor that inhibits TNF-alpha production/apoptosis in macrophages infected with Mtb. We demonstrate that Mtb and Nef individually trigger TNF-alpha production in macrophages. However, TNF-alpha production is dampened when the two are present simultaneously, probably through cross-regulation of the individual signaling pathways leading to activation of the TNF-alpha promoter. Mtb-induced TNF-alpha production is abrogated upon mutation of the Ets, Egr, Sp1, CRE, or AP1 binding sites on the TNF-alpha promoter, whereas Nef-mediated promoter activation depends only on the CRE and AP1 binding sites, pointing to differences in the mechanisms of activation of the promoter. Mtb-dependent promoter activation depends on the mitogen-activated kinase (MAPK) kinase kinase ASK1 and on MEK/ERK signaling. Nef inhibits ASK1/p38 MAPK-dependent Mtb-induced TNF-alpha production probably by inhibiting binding of ATF2 to the TNF-alpha promoter. It also inhibits MEK/ERK-dependent Mtb-induced binding of FosB to the promoter. Nef-driven TNF-alpha production occurs in an ASK1-independent, Rac1/PAK1/p38 MAPK-dependent, and MEK/ERK-independent manner. The signaling pathways used by Mtb and Nef to trigger TNF-alpha production are therefore distinctly different. In addition to attenuating Mtb-dependent TNF-alpha promoter activation, Nef also reduces Mtb-dependent TNF-alpha mRNA stability probably through its ability to inhibit ASK1/p38 MAPK signaling. These results provide new insight into how HIV Nef probably exacerbates tuberculosis infection by virtue of its ability to dampen Mtb-induced TNF-alpha production.

FIGURE 1.

Nef inhibits M. tuberculosis-mediated induction of TNF-α. A, THP-1 cells (in 96-well plates) were left untreated (−) or treated with either M. tuberculosis (Mtb) or Nef or both (Mtb + Nef) for 24 h. Cells were washed and lysed, and cell death was measured using the cell death ELISA kit according to the manufacturer's instructions. B, THP-1 cells (in 96-well plates) were left untreated (−) or treated with either M. tuberculosis (Mtb) or Nef or both (Mtb + Nef). In a separate set of experiments, Nef was pretreated with polymyxin B (PB-Nef) or heated at 95 °C for 10 min (heat-treated) and then used alone or in combination with M. tuberculosis as described above. The release of TNF-α in the supernatant was quantitated by TNF-α ELISA according to the manufacturer's instructions, 24 h after infection. THP-1 cells were transfected with a TNF-α promoter luciferase reporter construct (wt) (C and E) or with mutants devoid of the indicated transcription factor binding sites (E), along with a β-galactosidase expression vector. Cells were then left untreated (− in C), or incubated with either Mtb or Nef or with both (C and E) as indicated. Cells were lysed, and luciferase activities were determined 14 h after infection. The activities were normalized with β-galactosidase activity. Data represent means ± S.D. for three different experiments. D is a diagrammatic representation of the TNF-α promoter indicating the different transcription factor binding sites (underlined). The sequence in the bottom line indicates the mutation by replacement of the corresponding bases on the upper line.

FIGURE 2.

Nef inhibits M. tuberculosis-mediated activation of p38 MAPK and ASK1. A, THP-1 cells were transfected with vector alone or with a kinase-dead mutant of ASK1 (ASK1(KM)) or dominant-negative ATF2 (ATF2(dn)) or dominant-negative p38 MAPK (p38(agf)). Transfected cells were treated with M. tuberculosis (Mtb) or Nef separately, and release of TNF-α was measured as described under Fig. 1A. Data represent the means ± S.D. of three separate experiments. B–D, THP-1 cells transfected with either empty vector or ASK1 (KM) (B) or Rac1 (dn) (D) or PAK1-KD (D) were left untreated or treated with Mtb or with Nef separately for 90 (B and D) or 60 (C) min. Cell lysates were either immunoblotted with phospho-p38 MAPK antibody and reprobed with p38 MAPK antibody (B and D) or immunoprecipitated with ASK1 antibody, and the immunoprecipitate was used to study the phosphorylation of myelin basic protein using [γ-³²P]ATP followed by autoradiography (C). Actin in the cell lysate was blotted to confirm equal amounts of proteins in cell lysates (bottom panel of C). E, activation of Rac1 in cellular extracts was assessed by affinity precipitation of the Rac1-GTP complex from whole cell lysates using PAK1-PBD followed by Western blotting using anti-Rac1 antibody as described under “Experimental Procedures.” F, cell extracts were prepared, and phosphorylation of PAK1 was assessed by Western blotting using anti-phospho-PAK1 antibody followed by reprobing with PAK1 antibody, respectively. The data in panels B–F are representative of those obtained in three different experiments.

FIGURE 3.

Activation of transcription factor by M. tuberculosis and Nef. A and B, THP-1 cells were left untreated or treated with Mtb (A) or Nef (B), and the activation of c-Jun, Sp1, and ATF2 was quantified using the TransFactor ELISA kit (Clontech) according to the manufacturer's protocol. C, THP-1 cells were treated with Mtb alone or with Mtb and Nef, and the activation of c-Jun, Sp1, and ATF2 was determined using the TransFactor ELISA kit as described above. Data in A–C represent means ± S.D. of three separate determinations. D and E, THP-1 cells were left untreated (−) or preincubated with the inhibitors U0126 (U) or SB203580 (SB) prior to treatment with Mtb. In a separate set of experiments cells were left untreated or treated with either Mtb or Nef or with both. Chromatin was prepared, and ChIP analysis was carried out with primers specific for the AP1- (D) or Sp1- (E) binding site of the TNF-α promoter after immunoprecipitation (IP) with anti-ATF2 (D) or anti-Sp1 (E) antibody. The input panel shows the PCR product obtained when no immunoprecipitation was performed.

FIGURE 4.

Role of MAPKs in the activation of TNF-α. A, THP-1 cells were transfected with empty vector or dominant-negative MEK (MEK(dn)) along with the TNF-α promoter luciferase reporter construct. Transfected cells were left untreated or treated with Mtb or with Nef, and luciferase activity was measured as described previously. B, THP-1 cells were left untreated or treated with Mtb in the absence or presence of Nef for different periods of time. Cell lysates were immunoblotted with phospho-ERK antibody, and blots were reprobed with ERK antibody. C, THP-1 cells were left untreated or treated with Mtb or Nef for the indicated periods of time (in hours), and the activation of FosB was quantified using the TransFactor ELISA kit (Clontech) according to the manufacturer's protocol. D, THP-1 cells were left untreated (−) or preincubated with the inhibitor U0126 prior to treatment with Mtb. In a separate set of experiments, cells were treated with either Mtb or Nef or with both. After treatments, ChIP analysis was carried out with primers specific for the AP1-binding site of the TNF-α promoter after immunoprecipitation (IP) with anti-FosB antibody. The input panel shows the PCR product obtained when no immunoprecipitation was performed.

FIGURE 5.

Role of Nef in TNF-α mRNA stability. A, THP-1 cells were treated with Mtb in the absence or in the presence of Nef for 14 h. Actinomycin D (5 μg/ml) was added, total RNA was harvested, and TNF-α mRNA was estimated by quantitative real-time PCR at 0, 45, and 90 min. Data represents % mRNA remaining at the indicated time points after actinomycin D addition. 100% represents the amount of mRNA at the zero time point. B, schematic diagram of modulation of Mtb-induced activation of TNF-α promoter by Nef.