Single-cell suspension cultures of MTB H37Ra (ATCC 25177) and MTB H37Rv (kindly provided by Dr. Richard L. Friedman, University of Arizona, Tucson, AZ) were prepared as described previously (12). The tuberculin PPD used in the in vitro assays was purchased from Statens Serum Institut (Copenhagen, Denmark) and was used at a final concentration of 5.0 µg/ml.

LPS, 1,25-D3, N^G-monomethyl-L-arginine (L-NMMA, a nonselective NOS inhibitor), and N^G-nitro-L-arginine methyl ester (L-NAME) were purchased from Sigma (St Louis, MO). Mouse anti-human TLR2 mAb (clone TL2.1, IgG2a), mouse anti-human TLR4 mAb (clone HTA125, IgG2a), and isotype control (IC) mAb (IgG2a) were purchased from eBioscience (San Diego, CA). Specific antibodies against extracellular-regulating kinase (ERK) 1/2, phospho-(Thr202/Tyr204)-ERK1/2, p38, phospho-(Thr180/Tyr182)-p38, and phospho-IKKα/β were purchased from Cell Signaling Technology. Antibody against iNOS Ab (IgG), β-actin, IκB-α (IgG1), and MEK-1 mAb (H-8, IgG2_b) were obtained from Santa Cruz Biotechnology (Santa Cruz, CA). The specific inhibitors U0126 for MEK, caffeic acid phenethyl ester for NF-κB (CAPE), BAY 11-7082 for IκB-α phosphorylation, and SB203580 for p38 mitogen-activated protein kinase (MAPK) were obtained from Calbiochem (San Diego, CA). Recombinant human (rh) IFN-γ and rhtumor necrosis factor (TNF)-α were obtained from R&D systems (Minneapolis, MN). The psiRNA-h7SKGFPzeo plasmids for generating hTLR2 or hTLR4 short hairpin RNA (shRNA) were obtained from Invitrogen (Carlsbad, CA). Transfection of these plasmids into human primary MDMs was performed using amaxa Nucleofector technology (Gaithersburg, MD). Protein A-Sepharose for the MEK-1 assay was obtained from Amersham Bioscience (Freiburg, Germany), and [γ-³²P]-ATP was obtained from PerkinElmer (San Jose, CA).

This study was approved by the local Ethical Committee of Konyang University Hospital's review board, which oversees studies on samples from human subjects. Healthy volunteers provided informed written consent to take part in this study. Adherent monocytes were collected from peripheral blood mononuclear cells donated by healthy subjects, as previously described (12). Human MDMs were prepared by culturing peripheral blood monocytes for 4~5 days in the presence of 0.2 ng/ml human macrophage colony-stimulating factor (Sigma). In order to show that the stimulatory capacity of mycobacteria was not the result of contamination with LPS, experiments were performed that by addition of the specific LPS-inhibiting oligopeptide polymyxin B (10 µg/ml) before mycobacterial stimulation. To assess their sensitivity to rhIFN-γ and 1,25-D3, human MDMs were incubated in the presence of these factors then stimulated with PPD or MTB.

The detection of NO was detected with minor modifications using the NO-sensitive fluorescent probe DAF-2 DA (Calbiochem) and laser-scanning confocal microscopy (model LSM 510, Zeiss), as described previously (13). NO production was also measured indirectly as nitrite accumulation in cell culture medium by using the Griess reaction (14).

Human MDMs were treated as indicated and processed for analysis by reverase transcriptase-polymerase chain reaction (RT-PCR) and western blot, as described previously (12). RT-PCR for human iNOS mRNA was performed after total RNA was extracted from the cells using TRIzol (Invitrogen). For Western blot analysis, antibodies to phosphorylated and total p38, JNK, ERK1/2, and β-actin were used at 1:1,000 dilutions. Specific bands were developed by ECL (Amersham Biosciences).

Confluent MDMs were infected with MTB in the presence or absence of pre-treatment of either IFN-γ or 1,25-D3 alone, or their combination. After incubation, the cells were lysed with ice-cold lysis buffer and in vitro MEK assay was conducted, as previously described (12). The gel was dried and autoradiography was performed to visualize the ³²P-labeled MBP. Densitometry was performed on films and the fold increase was calculated as experimental sample/control sample.

For the quantification of intracellular Mtb, we employed two independent methods (15). For colony-forming units (CFUs) assay, the cells were lysed with 0.3% saponin (Sigma Chemical) to release the intracellular bacteria, after various periods of incubation. The lysates of the infected cells were sonicated in a preheated 37℃ water bath sonicator (Elma, Singen, Germany) for 5 min. Aliquots of the sonicates were then diluted tenfold in 7H9 medium. Four dilutions of each sample were plated separately on 7H10 agar plates and incubated at 37℃ with 5% CO₂ for 21 days. For [³H] uracil incorporation assays, the cells were lysed then transferred to 96-well round-bottom plates and incubated in the presence of 1 µCi [³H] uracil (Amersham-Pharmacia). After 24 h, the mycobacteria were killed by treatment with paraformaldehyde (final concentration, 4%) for 30 min. [³H]uracil incorporation was measured using a beta counter (Berthold, München, Germany).

Data obtained from independent experiments are presented as the mean±SD and were analyzed using either paired t-tests with Bonferroni adjustment or ANOVA for multiple comparisons. Differences were considered significant at p<0.05.

To address the role of NO during mycobacterial infection in human cells, we first characterized the pattern of in vitro NO production and iNOS expression in primary human MDMs (Fig. 1). Upon stimulation with the PPD antigen of MTB H37Ra, human primary MDMs exhibited very low levels of nitrite production as measured by Griess nitrite assays, with accumulated nitrite levels below 10 µM after at 96-h incubation (Fig. 1A). There was no cytotoxicity in cells cultured for 168 h (data not shown). Human iNOS gene induction by PPD in MDMs reached a peak at 6 h, based on RT-PCR analysis (Fig. 1B). Given that rhIFN-γ and rhTNF-α stimulate macrophages to upregulate the expression of iNOS in mice (16), the effects of these stimuli were examined on nitrite production in human MDMs. As clearly shown in Fig. 1C, the addition of IFN-γ (10 ng/ml), TNF-α (10 ng/ml), or a combination of both failed to increase the nitrite levels in human MDMs after PPD treatment. In addition, there was no significant increase in nitrite levels after combined treatment of IFN-γ and TNF-α, when compared with those cultured with media control (data not shown). These results suggest that neither IFN-γ nor TNF-α influence the up-regulation of nitrite production in human MDMs, as has been reported previously (17).

Several lines of evidence indicate that 1,25-D3 regulates host resistance to MTB. In addition to its function in the expression of the antimicrobial peptide cathelicidin (10), the vitamin D receptor triggered by 1,25-D3 is also involved in the production of substantial amounts of NO in the human promyelocytic cell line HL-60, when the cells are differentiated to a macrophage-like phenotype (9). To confirm the sensitivity of NO production to IFN-γ and 1,25-D3, we incubated human primary MDMs in the presence of these two factors. As shown in Fig. 2A, 1,25-D3 or IFN-γ alone scarcely induced an increase in nitrite production after 10 days in culture with PPD. However, the combined treatment with IFN-γ and 1,25-D3 led to a significant increase in nitrite production by human MDMs (p<0.001 vs. IFN-γ or 1,25-D3 alone on day 10 in culture with PPD; Fig. 2A). In addition, combined stimulation with IFN-γ and 1,25-D3 induced a significant and synergistic increase in NO production, as measured with the NO-sensitive fluorescent probe DAF-2 DA (p<0.001 vs. IFN-γ or 1,25-D3 alone; Fig. 2B). Furthermore, we tried to examine whether similar effects were detected in cells cultured with MTB H37Ra. As shown in Fig. 2C, combined stimulation with IFN-γ and 1,25-D3 induced a significant increase in iNOS protein expression in human MDMs treated with MTB H37Ra (Fig. 2C). The nitrite production, and expression of iNOS and NO were significantly attenuated by pretreatment with either L-NMMA or L-NAME, both of them commonly used competitive inhibitors of NO synthase (Fig. 2A~C). In all experiments, we found that the effects of combined stimulation with IFN-γ and 1,25-D3 without PPD or MTB H37Ra were negligible for induction of iNOS expression or nitrite production (data not shown). Therefore, this combined stimulation efficiently up-regulated human iNOS and NO in human macrophages.

We next investigated the molecular mechanism by which IFN-γ and 1,25-D3 induce iNOS activation. Previous studies reported that MAPKs and NF-κB regulate the expression of iNOS genes (18,19). Thus, we first tested whether incubation with IFN-γ, 1,25-D3, or both affected the activation of NF-κB in MTB H37Ra -treated human macrophages. As shown in Fig. 3A, the expression level of IκB-α was dramatically attenuated in MDMs following combined treatment for 30 min. In contrast, there was a remarkable increase in the MTB H37Ra-induced phosphorylation of IKKα/β in cells primed with IFN-γ and 1,25-D3, compared with untreated control cells or cells treated with IFN-γ or 1,25-D3 alone (Fig. 3A). Exposure to MTB H37Ra led to strong phosphorylation of ERK1/2 and p38 MAPK in human MDMs at 30 min after stimulation. Although the addition of IFN-γ or 1,25-D3 alone had no effect, their combined treatment up-regulated phospho-ERK1/2 but not phospho-p38 (Fig. 3B). In addition, a similar up-regulation of MEK1 activity was observed in MDMs primed with IFN-γ plus 1,25-D3 (Fig. 3C).

Notably, these separate responses (NF-κB, ERK1/2, and p38) were correspondingly blocked by the NOS inhibitors L-NAME and L-NMMA (Fig. 3A~C), indicating that iNOS induction activates the NF-κB and MAPK pathways in human MDMs. Further, inhibition of MEK1 with U0126 or NF-κB inhibition with specific inhibitors (CAPE and BAY11-7082) (20) reduced MTB-induced iNOS expression in human MDMs primed with IFN-γ and 1,25-D3, in a dose-dependent manner (Fig. 3D). The inhibition of p38 MAPK with SB203580 or JNK1/2 inhibition with SP600125 did not significantly diminish iNOS mRNA expression (Fig. 3D and data not shown). The results of these experiments indicate that the activation of ERK1/2 and NF-κB is necessary for the modulation of iNOS by combined IFN-γ and 1,25-D3 treatment.

It has been demonstrated that TLR2 plays an essential role in iNOS induction in vivo (21). However, there have been no reports about its role in iNOS mRNA expression in human monocytes/macrophages during MTB infection. We evaluated the roles of TLR2 and TLR4 in MTB H37Ra- or PPD-induced iNOS expression after combined treatment with IFN-γ and 1,25-D3 in human monocytes or in MDMs. In the representative experiments shown in Fig. 4A (and in data not shown), MTB H37Ra- or PPD-induced iNOS protein expression was significantly inhibited by a specific mAb against TLR2, whereas it was not affected by anti-TLR4 mAb. In addition, siRNA-mediated knockdown of TLR2, but not TLR4, resulted in diminished MTB H37Ra-induced iNOS protein expression (Fig. 4B). The knockdown of TLR2, but not TLR4, also resulted in decreased IκB-α degradation and IKKα/β phosphorylation in human MDMs (Fig. 4B). These data indicate an essential role for TLR2 in MTB H37Ra-stimulated iNOS induction by human macrophages primed with IFN-γ and 1,25-D3.

Recent studies reported that the addition of 1,25-D3 to primary human MDMs infected with virulent MTB reduced the number of viable bacilli through the induction of the antimicrobial peptide cathelicidin (10). We examined whether combined IFN-γ and 1,25-D3 treatment affected the induction of cathelicidin. As shown in Fig. 5A, PPD stimulation strongly induced the expression of cathelicidin mRNA in MDMs treated with 1,25-D3. However, the expression of cathelicidin was virtually identical between cells treated with 1,25-D3 and cells treated with 1,25-D3 plus IFN-γ (Fig. 5A). In addition, cathelicidin expression was not influenced by pretreatment with either L-NMMA or L-NAME (data not shown). These data demonstrate that the combined treatment with IFN-γ and 1,25-D3 does not act by enhancing the expression of cathelicidin.

As a physiological endpoint for gauging the importance of iNOS in antimycobacterial defenses, we quantitated intracellular mycobacterial growth using both [³H]-uracil uptake and CFU assays. To estimate the intracellular inoculum of MTB, MDMs were lysed after 1, 4, and 7 days of incubation with bacteria. Upon treatment with 1,25-D3 alone, virulent MTB H37Rv showed reduced survival in MDMs (Fig. 5B and C). In CFU assays, combined treatment with 1,25-D3 and IFN-γ resulted in a 2-fold decrease in the intracellular survival of virulent MTB H37Rv, relative to that seen with 1,25-D3 alone (Fig. 5B). The antimycobacterial effect induced by this combination was significantly attenuated by pretreatment with either L-NMMA or L-NAME (data not shown). Collectively, these results suggest that iNOS induction by the combined treatment with 1,25-D3 and IFN-γ is sufficient to influence intracellular mycobacterial growth during MTB infection.