Mycobacteria Inhibition of IFN-γ Induced HLA-DR Gene Expression by Up-Regulating Histone Deacetylation at the Promoter Region in Human THP-1 Monocytic Cells¹

Mycobacteria are facultative intracellular pathogens that are able to survive and multiply within macrophages for an extended period of time. Mycobacterium tuberculosis, the causative agent of tuberculosis, is responsible for 3 million deaths annually. Infections by Mycobacterium avium are common in AIDS patients (1, 2). The activity of macrophages is regulated by IFN-γ. IFN-γ increases expression of MHC class II molecules and other IFN-γ-inducible genes in macrophages (3) and activates anti-mycobacterial activity (4, 5, 6). However, macrophages infected with mycobacteria respond poorly to IFN-γ. Studies have shown that cell-mediated immune response of infected macrophages is impaired by decreased IFN-γ-induced expression of MHC class II genes and other IFN-γ-inducible genes (7, 8, 9, 10).

IFN-γ-stimulated signal transduction results in the activation of the JAK-STAT pathway in which STAT1 is tyrosine phosphorylated and translocates to the nucleus where it binds to the IFN-γ activation site (GAS)³ sequence in the promoters of IFN-γ-induced genes (11, 12, 13). Studies from this laboratory (7) have investigated the mechanism involved in the inhibition of IFN-γ-induced gene expression in M. avium-infected mouse macrophages. We observed that M. avium inhibits IFN-γ-inducible genes by interfering with the JAK-STAT1 signal transduction pathway, resulting in reduced phosphorylation of IFN-γRα, JAK1/JAK2, and STAT1. Recognition of M. avium by macrophages has been shown to involve TLR2 (14, 15). IFN-γ-induced gene expression is also inhibited by prior treatment with TLR2 agonists (16, 17, 18). Although TLR2 signaling does not inhibit STAT1α phosphorylation, we showed that TLR2 stimulation increases expression of the transcriptionally inactive STAT1β by stabilizing the STAT1β mRNA and thereby decreasing IFN-γ-induced gene expression by dominant negative inhibition (18). In contrast, M. tuberculosis infection of human macrophages results in more limited inhibition of IFN-γ-induced gene expression with some genes being inhibited such as CD64, HLA-DR, and CIITA, while others were unaffected (7, 8, 9, 10). M. tuberculosis infection of human macrophages also does not appear to inhibit the JAK-STAT1 pathway.

The purpose of the current study was to compare the effects of M. tuberculosis, M. avium, and TLR2 stimulation on IFN-γ induction of HLA-DR gene expression in human THP-1 monocytic cells differentiated with PMA. Class II MHC expression in macrophages results from IFN-γ-induced expression of CIITA (19, 20, 21). Although CIITA does not bind to DNA directly, it functions as a transcriptional coactivator by interacting with MHC class II promoter-bound transcription factors (22) and coordinating histone acetylation modification at the HLA-DRA promoter (23) by interacting with CREB-binding protein (CBP/p300) (24, 25), which has intrinsic histone acetyltransferase (HAT) activity (26, 27, 28). Histone acetylation is reversible and is regulated by HATs, which promote gene activation, and histone deacetylases (HDACs), which promote repression of acetylation-sensitive genes.

We found that induction of HLA-DR expression by IFN-γ was highly inhibited by M. avium, M. tuberculosis, and TLR2 stimulation, whereas induction of CIITA was partially inhibited. In contrast, IFN regulatory factor (IRF)-1 expression was not altered. Because studies (29, 30) have shown that MHC class II expression can be inhibited by HDAC activity, we investigated the role of HDAC activity in the inhibition of HLA-DR expression. Our studies show that mycobacteria and mycobacterial products inhibit IFN-γ-induced HLA-DR expression in human macrophages through a repressor pathway involving chromatin deacetylation in the HLA-DRα promoter.

PMA, sodium butyrate, and RNase A were purchased from Sigma-Aldrich. MS-275 was purchased from Alexis Biochemicals. Human recombinant IFN-γ was obtained from Genentech. Pam₃CSK₄ was acquired from EMC Microcollections. [³²P]dCTP was obtained from Amersham Biosciences. Proteinase K was purchased from Invitrogen Life Technologies.

The human THP-1 monocyte leukemia cell line, obtained from American Type Culture Collection (ATCC TIB-202), was cultured in RPMI 1640 containing 10% heat-inactivated FBS, 4 g/L glucose, 1.0 mM sodium pyruvate, 10 mM HEPES, 0.05 mM 2-ME, 100 U/ml penicillin, and 100 μg/ml streptomycin at 37°C in 5% CO₂. Before each experiment, THP-1 monocytes were plated in six-well culture plate and treated with 10 ng/ml PMA for 48 h to differentiate the cells into macrophage-like cells.

Mycobacterium infection and IFN-γ stimulation M. avium (ATCC 35713) were grown and stored as described previously (7). Bacteria concentration was determined by OD at 600 nm wavelength and confirmed by plate counting. Gamma-irradiated M. tuberculosis H37Rv (Colorado State University; National Institutes of Health, National Institute of Allergy and Infectious Diseases Contract N01 AI-75320) was diluted in Middlebrook 7H9 broth (Difco) and centrifuged 800 rpm for 10 min to eliminate clumped bacteria. The protein concentration in the supernatant was determined by the Bradford protein assay (Bio-Rad). The supernatant was stored in aliquots at −80°C. Frozen M. avium and M. tuberculosis aliquots were thawed and briefly sonicated before use. Differentiated THP-1 cells were treated with M. avium, gamma-irradiated M. tuberculosis, and Pam₃CSK₄ for 16–20 h. The THP-1 cells were then stimulated with 100 U/ml human IFN-γ for the indicated time in each experiment.

Total cytoplasmic RNA was isolated by the acid guanidinium isothiocyanate-phenol-chloroform extraction method (31). Fifteen micrograms of RNA were separated in 1% formaldehyde agarose gels and transferred to Hybond-N⁺ membranes by capillary blotting. Northern blot hybridization was performed as described previously (32). The cDNA probes to HLA-DRα (ATCC clone 57392), DRβ (ATCC clone 57081), human IRF-1 (Genestorm clone RG001570; Invitrogen Life Technologies), murine β-actin, and murine G3PDH (7) were labeled with [³²P]dCTP by the High Prime labeling system (Roche).

One microgram of total RNA was reverse transcribed with avian myeloblastosis virus reverse transcription system (Promega). PCRs were performed in 1× PCR buffer, 3 mM MgCl₂, 0.2 μM dNTP, 0.4 μM of each gene-specific primer, and 2 U of Platinum TaqDNA polymerase (Invitrogen Life Technologies). The amplification was 95°C for 5 min, then 27–35 cycles of 94°C for 45 s, 60°C for 45 s, and 72°C for 1 min.

The following primers were used: human mammalian Sin3A (mSin3A) (617 bp) sense, 5′-TGTTCACCATTCATGCCTACATTGCC-3′, and antisense, 5′-GGCGGTCCTCCGATACATATAGTCC-3′; G3PDH (988 bp), sense, 5′-TGAAGGTCGGTGTGAACGGATTTGGC-3′, and antisense, 5′-CATGTAGGCCATGAGGTCCACCAC-3′; and human β-actin (170 bp) sense, 5′-CCCCGCGAGCACAGA-3′, and antisense, 5′-CACCGATGGAGGGGAAGAC-3′. PCR products were visualized on 1.6% agarose gels containing ethidium bromide.

Whole cell lysates were prepared by extraction in lysis buffer containing 1% Triton X-100, 20 mM Tris-HCl (pH 8.0), 150 mM NaCl, 3 mM sodium pyrophosphate, 10% glycerol, 2 mM sodium orthovanadate, and protease inhibitors pepstatin (6 μg/ml), aprotinin (6 μg/ml), leupeptin (10 μg/ml), and 4-(2-aminoethyl)-benzenesulforyl fluoride (100 μg/ml). Nuclear extracts were prepared from 10⁷ differentiated THP-1 cells as described previously (7). Protein concentration was determined by the Bradford protein assay (Bio-Rad).

Twenty-five to 40 μg of protein were analyzed by SDS-PAGE using 10% Tris-glycine gels (Invitrogen Life Technologies), followed by transfer to Hybond-P membranes (Amersham Biosciences). Membranes were blocked in 5% BLOT-QuickBlocker (Genotech) in TBS containing 0.05% Tween 20 for 1 h and incubated with primary Abs at 4°C overnight. The detection step was performed with HRP-coupled anti-mouse or anti-rabbit IgG Abs (1:2500 and 1:5000; Genotech). Primary Abs were monoclonal mouse anti-tyrosine-phospho-STAT1 Ab (1:4000; Zymed Laboratories), polyclonal rabbit anti-STAT1α p91 Ab (1:1000; Santa Cruz Biotechnology), polyclonal rabbit anti-HDAC1 Ab (1:200; Santa Cruz Biotechnology), polyclonal rabbit anti-HDAC2 Ab (1:200; Santa Cruz Biotechnology), and polyclonal rabbit anti-mSin3A Ab (1:200; Santa Cruz Biotechnology). Blots were developed with the femtoLucent chemiluminescence detection system (Genotech).

EMSA was performed in 20-μl binding reactions containing 5 μg of nuclear extracts, 10 mM Tris-HCl (pH 7.5), 50 mM NaCl, 50 mM DTT, 5 mM MgCl₂, 10% glycerol, 0.2% Nonidet P-40, 1 μg of poly(dI:dC), and 70,000 cpm of [³²P]dCTP-labeled, double-stranded GAS oligonucleotide radiolabeled by fill-in reaction with Klenow DNA polymerase. The sequence of the GAS oligonucleotide (5′-AGCCATTTCCAGGAATCGAAA-3′) contains the optimum GAS sequence (5′-TTCCSGGAA-3′) for STAT1 DNA binding (33). The samples were electrophoresed in 5% polyacrylamide gels in 0.5× Tris-borate EDTA. The gels were then dried and analyzed by autoradiography.

Chromatin immunoprecipitations were done using the Acetyl-Histone H4 ChIP assay kit (Upstate Biotechnology). A total of 4 × 10⁶ cells was fixed in 1% formaldehyde at 37°C for 10 min, then washed with ice-cold PBS containing protease inhibitors, 1 mM PMSF, 1 μg/ml aprotinin, 10 μg/ml 4-(2-aminoethyl)-benzenesulforyl fluoride, and 1 μg/ml pepstatin A. Cells were incubated in 400 μl of lysis buffer containing 1% SDS, 10 mM EDTA, 50 mM Tris-HCl (pH 8.1), and protease inhibitors for 10 min on ice. The chromatin was sheared on ice by sonication to lengths between 300 and 800 bp by six 10-s bursts with a Branson 350 sonifier (Branson Precise Processing Group). Two-hundred microliters of sheared chromatin were diluted 1/10 with ChIP dilution buffer containing 0.01% SDS, 1.1% Triton X-100, 1.2 mM EDTA, 16.7 mM Tris-HCl (pH 8.1), 167 mM NaCl, and protease inhibitors. Chromatin was precleared with 80 μl of salmon sperm DNA-blocked protein A agarose beads for 30 min at 4°C, then incubated with 5 μl of rabbit anti-acetyl-histone H4 Ab (Upstate Biotechnology) at 4°C overnight. In a separate set of experiments, chromatin was incubated with 25 μl of rabbit anti-mSin3A, and rabbit anti-CREB-binding protein (CBP) (Santa Cruz Biotechnology). A no-Ab control reaction was set up as well. The immune complexes were precipitated with 60 μl of salmon sperm DNA-blocked protein A agarose beads for 30 min at 4°C, followed by washing with low salt wash buffer (0.1% SDS, 1% Triton X-100, 2 mM EDTA, 20 mM Tris-HCl (pH 8.1), and 150 mM NaCl), high salt wash buffer (0.1% SDS, 1% Triton X-100, 2 mM EDTA, 20 mM Tris-HCl (pH 8.1), and 500 mM NaCl), LiCl wash buffer (0.25 M LiCl, 1% Nonidet P-40, 1% deoxycholate, 1 mM EDTA, and 10 mM Tris-HCl (pH 8.1)), and TE buffer (1 mM EDTA and 10 mM Tris-HCl (pH 8)). The immune complexes were eluted with 500 μl of 1% SDS, 0.1 M NaHCO₃ and incubated in the presence of 200 mM NaCl at 65°C for 4 h to reverse the histone-DNA cross-links. The samples were then digested with 40 μg/ml proteinase K, 10 mM EDTA, and 40 mM Tris-HCl (pH 6.5) at 45°C for 1 h. The DNA was purified by phenol-chloroform extraction, followed by ethanol precipitation. The recovered DNA was dissolved in 20 μl of molecular grade water. Two microliters of DNA were used for real-time PCR as described below.

Total RNA was extracted with RNeasy Mini kit (Qiagen) following manufacturer’s instructions. One microgram of RNA was reverse transcribed with avian myeloblastosis virus reverse transcription system (Promega) and dissolved in 50 μl of molecular grade water. Two microliters of cDNA were analyzed by real-time PCR using LightCycler-FastStart DNA Master SYBR Green I (Roche), according to manufacturer’s directions. Reactions were run at 95°C for 10 min and 40 cycles of 95°C for 15 s, 60°C for 5 s, and 72°C for 15 s on a RotorGene 2000 Real-Time Quantitative Thermal Cycler System (Corbett Research, Pyrosequencing). The following primers were used for detecting HDAC and CIITA gene expression: HDAC1 217-773 (5′-GAAATCTATCGCCCTCACAAAGCCAATGC-3′, 5′-TAGGACTCGTCATCAATCCCGTCTCGG-3′); HDAC2 487-995 (5′-TTTAATGTTGGAGAAGATTGTCCAGCG-3′, 5′-GCACCACACTGTAATACCACAGCACTAGG-3′); HDAC3 144-795 (5′-CAGACCCATAGCTGGTCCTGCATTACG-3′, 5′-AGAAGTCCACTACCTGGTTGATAACCGGC-3′); and CIITA 3222-3713 (5′-TGACCTGGGTGCCTACAAACTC-3′, 5′-GCAAGATGTGGTTCATTCCGC-3′). Gene expression was normalized with G3PDH primers 66-291, (5′-GAAGGTGAAGGTCGGAGTC-3′, 5′-GAAGATGGTGATGGGATTTC-3′).

Double-strand DNA purified from the ChIP assay was also subject to real-time PCR with previously described HLA-DRA promoter primers (5′-GATTTGTTGTTGTTGTTGTCCTGTTC-3′, 5′-CCCAATTACTCTTTGGCCAATCAGAAAAATATTTTG-3′) (20). The relative values was calculated by the comparative C_T method (34). Statistical analysis was determined using one-way ANOVA and Tukey’s test. Differences were considered to be significant if p < 0.05.

To investigate whether M. avium inhibits IFN-γ-inducible gene expression in human THP-1 cells, we infected PMA-differentiated human THP-1 cells with M. avium at 5:1, 10:1, 20:1, and 40:1 bacteria to cell ratios for 16 h, followed by stimulated with human rIFN-γ at 100 U/ml for 20 h. The expression of HLA-DR and IRF-1 was investigated by Northern blot analysis and expression of CIITA by real-time RT-PCR. Expression of HLA-DRα and HLA-DRβ mRNA was highly induced by IFN-γ. The IFN-γ induction of the MHC class II genes was inhibited by M. avium infection, and the degree of inhibition correlated with the dosage of M. avium (Fig. 1,A). Expression of IRF-1 mRNA was also highly induced by IFN-γ treatment. However, the expression remained unaffected by M. avium infection in human THP-1 cells (Fig. 1,B). Similar induction of CIITA gene expression by IFN-γ was also observed in CIITA gene expression (Fig. 1 C). M. avium infection also decreased the induction of CIITA expression by IFN-γ. However, at bacteria:cell ratios of 20:1 and higher, where IFN-γ induction of HLA-DRα is inhibited completely, substantial induction of CIITA was still observed ∼20-fold.

To determine the effect of M. avium infection of THP-1 cells on STAT1 activation, we examined by EMSA the binding of STAT1 to a consensus GAS element and by Western blotting STAT1 tyrosine phosphorylation. Nuclear extracts prepared from M. avium-infected THP-1 cells stimulated with IFN-γ showed no reduction of STAT1 binding to the GAS element (Fig. 2,A). There was also no reduction in STAT1 tyrosine phosphorylation by Western blot analysis (Fig. 2,B). We did observe a slight increase of total STAT1 protein expression in infected cells, suggesting that M. avium infection increases STAT1 gene expression in THP-1 cells (Fig. 2 B). These results are in contrast to our previous studies of M. avium-infected mouse macrophages in which we observed inhibition of the JAK-STAT1 pathway (7) but in agreement with the recent observations in M. tuberculosis-infected human THP-1 cells (10). Taken together, these results indicate that in human THP-1 cells, inhibition of HLA-DR gene expression by M. avium infection is independent of the JAK-STAT1 signal transduction pathway.

We also investigated whether M. tuberculosis affects IFN-γ-induced gene expression through a similar mechanism as M. avium. We infected PMA-differentiated THP-1 cells with gamma-irradiated M. tuberculosis for 16 h and stimulated the THP-1 cells with IFN-γ for 20 h. Gene expression of HLA-DRα, HLA-DRβ, and IRF-1 was examined by Northern blot analysis. M. tuberculosis also inhibited HLA-DRα and HLA-DRβ gene expression in a dose-dependent manner. In contrast, IRF-1 gene expression was not affected following M. tuberculosis infection (Fig. 3 A).

Activation of the macrophage antimicrobial response involves stimulation of TLRs. M. avium activation of macrophages involves TLR2 (14, 15), whereas M. tuberculosis activation of macrophages involves both TLR2 and TLR4 (35, 36). As both M. tuberculosis and M. avium can activate macrophages through TLR2, we investigated whether TLR2 stimulation of THP-1 cells could also affect IFN-γ-inducible HLA-DR gene expression. PMA-differentiated THP-1 cells were treated with a TLR2 agonist, synthetic bacterial lipopeptide Pam₃CSK₄ (37), for 16 h followed by IFN-γ for 20 h. We found that IFN-γ-induced HLA-DRα and HLA-DRβ gene expression was also inhibited in differentiated THP-1 cells treated with Pam₃CSK₄. IRF-1 gene expression was not inhibited by TLR2 agonist treatment (Fig. 3 B). Taken together, the results indicate that M. avium and M. tuberculosis, acting through TLR2, inhibit HLA-DR gene expression at an unknown step that is independent of JAK-STAT1 activation.

Studies have shown that interactions of HAT and HDAC activity are important in regulating gene expression at nucleosomes (38, 39). Acetylation of lysine residues in the N-terminal tails of the core histone proteins results in the uncoiling of the chromatin, allowing increased accessibility for transcription factors, while tightly bound DNA around a nucleosome core suppresses gene transcription by decreasing the accessibility of transcription factors to the gene promoter (40). Gene-specific repression is associated with the recruitment of multicomponent HDAC complexes to promoters (40). Suppression of HDAC activity in human tumor cell lines, which fail to express HLA-DR expression upon IFN-γ stimulation, has been shown to rescue IFN-γ-inducible HLA-DR gene expression (29). Therefore, we were interested in determining whether modulation of chromatin conformation via enhanced HDAC activity was a possible mechanism involved in the HLA-DR gene repression by mycobacteria in THP-1 cells. To test this hypothesis, we treated PMA-differentiated THP-1 cells with a HDAC inhibitor, sodium butyrate, at the time of mycobacteria infection. A butyrate dose-response experiment was done on HLA-DRα and HLA-DRβ expression with M. avium-infected THP-1 cells. Treatment with butyrate resulted in a recovery of HLA-DRα and HLA-DRβ gene expression in M. avium-infected THP-1 cells in a dose-dependent manner (Fig. 4,A). Similar results were observed in M. tuberculosis-infected and Pam₃CSK₄-treated THP-1 cells (Fig. 4,B). The IRF-1 gene expression was not affected by butyrate (Fig. 4,B). Butyrate alone was not an inducer of HLA-DRα and HLA-DRβ gene expression in THP-1 cells. It also did not alter IFN-γ-induced HLA-DR expression in THP-1 cells that were not treated with mycobacteria; it only rescued mycobacterial suppressed HLA-DR gene expression. Butyrate is a weak inhibitor of HDAC activity, and the high concentrations of butyrate required to inhibit HDAC activity raises concerns that butyrate may have other effects on cells. As an alternative to butyrate, experiments were also done using MS-275, which is a novel HDAC isoform selective inhibitor that preferentially inhibits HDAC1 over HDAC3 and HDAC8 (41). As shown in Fig. 4 C, MS-275 also rescued M. avium inhibition of HLA-DR gene expression. Thus, these results using two different HDAC inhibitors suggest HDAC activity at the HLA-DR gene is involved in repression by M. avium and M. tuberculosis infection, and this effect may be regulated by TLR2 signaling in THP-1 cells.

HLA-DR gene expression is controlled by a complex promoter region containing W/S, X1, X2, Y box, and octamer elements (42, 43). To determine whether mycobacteria infection alters DNA-binding proteins interacting with the promoter region of HLA-DR genes, we examined by EMSA the DNA-binding capacity of nuclear extracts, isolated from M. avium-infected THP-1 cells, to X1, X2, Y, and octamer elements in the promoter and the YY1-binding element in the first exon of the HLA-DRα gene. We did not observe any changes in the DNA-binding capacity of the nuclear proteins from mycobacteria-infected THP-1 cells (data not shown). We next investigated whether chromatin remodeling of the promoter of the HLA-DRα gene by histone acetylation is affected by mycobacteria infection in vivo. PMA-differentiated THP-1 cells were infected with M. avium at 20:1 bacteria to macrophage ratio or gamma-irradiated M. tuberculosis at 50 μg/ml or Pam₃CSK₄ at 50 ng/ml for 16 h, followed by stimulation with human rIFN-γ at 100 U/ml for 24 h. ChIPs were performed using Abs against acetylated-histone H4. PCR specific for the HLA-DR promoter was done with the DNA purified from the chromatin immunoprecipitates (Fig. 5,A). To ensure that the amount of input chromatin was equal, DNA from 1% of chromatin was also amplified by PCR. To compare differences in level of acetylated-histone H4, real-time quantitative PCR was performed (Fig. 5,B). We found that the baseline amount of acetylated HLA-DR promoter was low without treatment, whereas IFN-γ stimulation increased the histone H4 acetylation at the HLA-DR promoter region to more than three times the baseline level. This effect was impaired significantly by M. avium, M. tuberculosis infection, and Pam₃CSK₄ treatment (Fig. 5 A). These results indicate that mycobacteria infection inhibits IFN-γ-induced HLA-DR gene expression by histone deacetylation at the HLA-DR promoter through a mechanism that involves a TLR2 signal transduction pathway.

The decrease in the amount of acetylated histone at HLA-DRα promoter suggests that HDAC activity is elevated at the promoter region by mycobacteria infection. One possibility is an increase in HDAC gene expression. To determine whether mycobacteria infection is able to up-regulate HDAC gene expression in THP-1 cells, we used real-time RT-PCR to examine the HDAC1, HDAC2, and HDAC3 mRNA expression following infection. HDACs were expressed constitutively in PMA-differentiated THP-1 cells, whereas no significant increase in mRNA expression was detected after M. avium infection. Fig. 6,A shows the results for HDAC1 and HDAC2. Identical results were obtained for HDAC3 (data not shown). These results were confirmed by Western blot analysis with Abs to HDAC1 and HDAC2 (Fig. 6 B).

mSin3A, which is associated physically with HDAC1 and HDAC2 in multicomponent complexes, has been recognized as an important corepressor protein in HDAC-regulated transcriptional repression (44, 45, 46). Zika et al. (30) have reported that mSin3A is required for MHC class II repression. Therefore, we hypothesized that repression of the IFN-γ-induced HLA-DR gene expression was related to the regulation of the corepressor mSin3A. PMA-differentiated THP-1 cells were infected in time course experiments with M. avium, M. tuberculosis, or Pam₃CSK₄. By Western blot analysis, the expression of mSin3A was not detectable in the nuclear extracts from untreated THP-1 cells. However, a dramatic increase in mSin3A protein expression was observed following mycobacteria infection and TLR2 stimulation. This increase in protein expression peaked at 8–14 h and dropped to baseline by 24 h after M. avium infection or TLR2 stimulation (Fig. 7,A). These results were confirmed by RT-PCR (Fig. 7,B). mSin3A mRNA levels were increased following infection by mycobacteria and TLR2 stimulation (Fig. 7 C).

The presence of mSin3A and CBP at the HLA-DRα promoter was studied by chromatin immunoprecipitation (Fig. 8). IFN-γ stimulation of PMA-differentiated THP-1 cells induced the association of the coactivator CBP with the HLA-DRα promoter. This association was absent in M. avium-infected cells stimulated with IFN-γ. Instead, in these cells, the corepressor mSin3A was present at the HLA-DRα promoter. The association of the corepressor with the HLA-DRα promoter required M. avium infection and IFN-γ because M. avium infection or IFN-γ alone did not induce association of mSin3A with the promoter. Thus, our results identify a novel mechanism for inhibition of IFN-γ-induced HLA-DRα gene expression by mycobacteria through increased expression of the corepressor mSin3A and association with the HLA-DRα promoter, resulting in histone deacetylation and transcriptional repression.

Intracellular pathogens have been shown to compromise the host immune response by affecting the IFN-γ response of infected macrophages. Compromised MHC class II expression in response to IFN-γ in infected macrophages contributes largely to the inhibition of Ag presentation by intracellular pathogens. Our previous studies with murine macrophages have shown that M. avium inhibits IFN-γ-inducible gene expression by down-regulating IFN-γR expression and IFN-γRα, JAK1, JAK2, and STAT1 phosphorylation (7, 47). Similar research with other pathogens, including Toxoplasma gondii (48), Leshmania donovani (49), and Erlichia chaffeensis (50), has also shown inhibition of IFN-γ-induced JAK-STAT1 signaling. However, studies have shown that IFN-γ-induction of the JAK-STAT1 pathway is not inhibited by M. tuberculosis infection in human macrophages (9, 10). Our observations in this article with M. avium show that M. avium infection also does not inhibit the JAK-STAT1 pathway in human THP-1 monocytic cells. We do show by Northern blot hybridization that infection with M. avium and M. tuberculosis or treatment with a TLR2 agonist profoundly attenuates IFN-γ-induced HLA-DR mRNA expression.

IRF-1 and CIITA expression are essential for MHC class II expression. Several studies have shown that CIITA expression is reduced significantly in macrophages infected with mycobacteria or treated with mycobacteria lipoprotein (7, 10, 51). We also observed a reduction in IFN-γ-induced CIITA mRNA expression. However, at M. avium concentrations that are sufficient to completely repress HLA-DR gene expression in human THP-1 cells, there was still a significant level of CIITA mRNA expression induced by IFN-γ. IRF-1 expression is required for IFN-γ induction of CIITA. We found that IRF-1 mRNA expression is not affected by mycobacteria infection and TLR2 stimulation. These results indicate that, in THP-1 cells, inhibition of IFN-γ induction of HLA-DR by mycobacteria occurs both at the level of CIITA expression and downstream of CIITA expression. HLA-DRα and HLA-DRβ promoters are complex, containing elements that bind constitutive transcription factors that interact with CIITA to induce gene expression. We found no differences by EMSA in the expression of transcription factors binding to the promoter elements of HLA-DR, suggesting that mycobacterial infection is not altering the expression of these constitutive transcription factors.

Because transcriptional regulation in eukaryotes is influenced strongly by posttranslational modification of histones, chromatin remodeling through histone acetylation and deacetylation was examined as a mechanism for inhibition of HLA class II gene expression. In this article, we first demonstrated that inhibition of HDAC activity with butyrate and MS-275 could restore IFN-γ-inducible HLA-DR gene expression that was inhibited by mycobacterial infection or TLR2 stimulation. Inhibition of HDAC activity without mycobacterial infection did not induce HLA-DR gene expression with or without IFN-γ. These results indicate only the repressed the expression of HLA-DR gene by mycobacterial infection could be restored by inhibiting HDAC activity and suggest that mycobacterial infection may up-regulate HDAC activity at the regulatory region of the HLA-DR gene. By ChIP assay, we demonstrated that histone H4 acetylation at the HLA-DRα promoter was blocked by mycobacterial infection or stimulation with a TLR2 agonist. Thus, these results indicate a hypoacetylated state of the HLA-DRα gene promoter exists when mycobacteria-infected THP-1 cells are stimulated with IFN-γ.

One possible explanation for these observations is that mycobacterial infection causes a general effect by up-regulating HDAC gene expression. However, mRNA expression and protein expression of HDAC1, HDAC2, and HDAC3 was not changed by infection with mycobacteria or by TLR2 stimulation. We did observe a substantial increase in the expression of the HDAC corepressor, mSin3A. HDACs are the enzymatic component of multiprotein complexes. HDAC1 and HDAC2 in mammalian cells are complexed with corepressors mSin3A or NurD (nucleosome remodeling HDAC) (52). Overexpression of mSin3A but not NurD by transient transfection has been shown to completely repress CIITA-mediated MHC class II gene expression (30). A model for gene-specific repression has been proposed in which the mSin3A-HDAC complex is tethered to promoters by the interaction of mSin3A with DNA-binding transcriptional factors (53). Thus, it is likely that in the mycobacterial repression of IFN-γ-induced HLA-DR expression, the mSin3A-HDAC complex is recruited to HLA-DRA and HLA-DRB promoters by the interaction of mSin3A with an unidentified transcription factor. Because mSin3A association with the HLA-DRα promoter occurs only in infected cells that are stimulated with IFN-γ, the most likely candidate is CIITA. CIITA promotes HLA-DR transcription by interacting with components of the basal transcription machinery and chromatin remodeling enzymes, including the histone acetyltransferase CBP/p300. Our ChIP data shows that CBP is not associated with the HLA-DRα promoter in mycobacteria-infected cells stimulated with IFN-γ. Thus, mSin3A and CBP may compete for interaction with CIITA and determine up-regulation or down-regulation of HLA-DR expression. A similar activator-repressor switching has been shown to occur in the regulation of the c-fos promoter by the Elk-1 transcription factor (54). Initially following growth factor stimulation, Elk-1 is recruited to the serum-responsive element of the c-fos promoter and associates with coactivators and histone acetyltransferases, resulting in up-regulation of c-fos expression. Following a temporal delay, the mSin3A/HDAC1 complex interacts with Elk-1, resulting in repression of c-fos transcription. Other possible interacting partners for mSin3A include transcriptional repressors that interact with X2 and Y boxes. Transcriptional repressors that interact with the X2 box of the human HLA-DPA promoter and the Y box of the mouse I-A β promoter have been described previously (55, 56).

In these studies, we also observed that stimulation of TLR2 with Pam₃CSK₄ exerted the same effect on HLA-DR expression as infection with M. avium or treatment with irradiated M. tuberculosis. This suggests that TLRs are involved in initiating the inhibition of HLA-DR expression in mycobacteria-infected THP-1 cells. This is consistent with the observations that mycobacteria does not need to be alive to inhibit IFN-γ-induced gene expression (7, 10). We and others (16, 17, 18) have examined the effects of TLR2 stimulation on IFN-γ-induced gene expression in mouse macrophages. These studies showed that TLR2 stimulation of mouse macrophages inhibited IFN-γ-induced expression of several genes, including FcγRI, IRF-1, CIITA, and MHC class II genes. These studies also showed that TLR2 stimulation does not inhibit IFN-γ-induced STAT1α phosphorylation. However, our studies (18) showed that TLR2 stimulation of mouse macrophages increased phosphorylation and protein expression of STAT1β, which is transcriptionally inactive and inhibits IFN-γ-induced gene expression by acting as a dominant negative mutant. The increased expression of STAT1β was shown to result from increased stability of STAT1β mRNA in TLR2-stimulated macrophages. The studies in the present report with PMA-differentiated THP-1 cells differ from these previous mouse studies in that only MHC class II gene expression is inhibited by TLR2 stimulation. However, our PMA-differentiated THP-1 cells do not express the STAT1β protein. Only STAT1α was observed on Western blots (Fig. 2). We do not know why our THP-1 cells did not express STAT1β. Human primary macrophages express both STATα and STAT1β (9). Thus, the lack of STAT1β could be unique to THP-1 cells or due to an unknown effect of PMA treatment on RNA splicing.

The absence of global inhibition of IFN-γ-induced genes in differentiated human THP-1 cells has enabled us to identify a novel mechanism by which mycobacteria inhibit IFN-γ-induced MHC class II expression. Our studies suggest that mycobacteria through TLR2 signaling interferes with chromatin remodeling by up-regulating expression of the corepressor mSin3A, which following IFN-γ stimulation associates with HLA-DRα promoter and recruits HDACs, resulting in inhibition of histone acetylation. Additional investigation needs to be done to elucidate the mechanism by which TLR2 signaling regulates mSin3A expression and how mSin3A associates with the MHC class II promoter. In conclusion, the IFN-γ activation of macrophages is inhibited by mycobacteria through a number of different mechanisms. These include the down-regulation of the expression of the IFN-γ receptor following mycobacteria infection of mouse macrophages (7, 47), differential expression of the transcriptionally inactive STAT1β (18), and up-regulation of HDAC corepressor mSin3A expression.

The authors have no financial conflict of interest.