IFN-gamma production by T cells is pivotal for defense against many pathogens, and the proximal promoter of IFN-gamma, -73 to -48 bp upstream of the transcription start site, is essential for its expression. However, transcriptional regulation mechanisms through this promoter in primary human cells remain unclear. We studied the effects of cAMP response element binding protein/activating transcription factor (CREB/ATF) and AP-1 transcription factors on the proximal promoter of IFN-gamma in human T cells stimulated with Mycobacterium tuberculosis. Using EMSA, supershift assays, and promoter pulldown assays, we demonstrated that CREB, ATF-2, and c-Jun, but not cyclic AMP response element modulator, ATF-1, or c-Fos, bind to the proximal promoter of IFN-gamma upon stimulation, and coimmunoprecipitation indicated the possibility of interaction among these transcription factors. Chromatin immunoprecipitation confirmed the recruitment of these transcription factors to the IFN-gamma proximal promoter in live Ag-activated T cells. Inhibition of ATF-2 activity in T cells with a dominant-negative ATF-2 peptide or with small interfering RNA markedly reduced the expression of IFN-gamma and decreased the expression of CREB and c-Jun. These findings suggest that CREB, ATF-2, and c-Jun are recruited to the IFN-gamma proximal promoter and that they up-regulate IFN-gamma transcription in response to microbial Ag. Additionally, ATF-2 controls expression of CREB and c-Jun during T cell activation.

We evaluated the role of regulatory T cells (CD4(+) CD25(+) Foxp3(+) cells, Tregs) inhuman Mycobacterium tuberculosis infection. Tregs were expanded in response to M. tuberculosis in healthy tuberculin reactors, but not in tuberculin-negative individuals. The M. tuberculosis mannose-capped lipoarabinomannan (ManLAM) resulted in regulatory T cell expansion, whereas the M. tuberculosis 19-kDa protein and heat shock protein 65 had no effect. Anti-IL-10 and anti-TGF-beta alone or in combination, did not reduce expansion of Tregs. In contrast, the cyclooxygenase enzyme-2 inhibitor NS398 significantly inhibited expansion of Tregs, indicating that prostaglandin E2 (PGE2) contributes to Treg expansion. Monocytes produced PGE2 upon culturing with heat-killed M. tuberculosis or ManLAM, and T cells from healthy tuberculin reactors enhanced PGE2 production by monocytes. Expanded Tregs produced significant amounts of TGF-beta and IL-10 and depletion of Tregs from PBMC of these individuals increased the frequency of M. tuberculosis-responsive CD4(+) IFN-gamma cells. Culturing M. tuberculosis -expanded Tregs with autologous CD8(+) cells decreased the frequency of IFN-gamma(+) cells. Freshly isolated PBMC from tuberculosis patients had increased percentages of Tregs, compared to healthy tuberculin reactors. These findings demonstrate that Tregs expand in response to M. tuberculosis through mechanisms that depend on ManLAM and PGE2.

Previously we found that regulatory T cells (CD4+CD25+Foxp3+ cells, Tregs) expand in response to M.tuberculosis (M. TB) through mechanism that depend on prostaglandin E2 production. In the current study, we determined the role of NK cells in regulating Tregs expansion in human M. TB infection. M. TB whole cell lysate (TB lysate) resulted in regulatory T cell expansion (15.33±1.02% vs 1.5±0.22%, P<0.001). Addition of monokine-activated NK cells (IL-12, IL-15 and IL-18) markedly inhibited Treg expansion (6.2±1.2% vs 15±1.5%, P<0.001) but freshly isolated NK cells had no effect. NK cells activated with TB lysate-stimulated monocytes also reduced Tregs expansion (13.2±0.3% to 3.4±1.4%, P<0.001) confirming the physiological relevance of this effect. Reduced Tregs expansion was not due to inhibition of PGE2 production or because of IFN γ. Monokine activated NK cells lysed TB lysate-expanded Tregs, but not freashly isolated Tregs (% specific lysis of 21±3% vs 1±1% P<0.001), nor T regs depleted T cells. Anti-NKG2D and anti-NKp46 reduced the % specific lysis of expanded Tregs from 22±3% to 7±3% and 8±3%, respectively (P<0.001) but abs to CD16, DNAM-1, and 2B4 had no effect. NKG2D ligand, ULBP1 expression was markedly upregulated on expanded T regs, compared to T regs depleted cells ( 31±5% vs 2.1 ±0.4%, P=0.002) and freshly isolated Tregs (4.6±1.1%). Anti -ULBP1 significantly inhibited the NK mediated lysis of expanded Tregs (specific lysis of 27±4% vs 9±2%, P<0.001) but abs to ULBP2, ULBP3 and MICA/B had no effect. These finding suggest that NK cells reduce the frequency of M.TB-expanded T regs by direct lysis.