INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Tuberculosis (TB) remains a major health problem worldwide. Clinical and pathologic features of TB depend at least in part on the orchestrated secretion of a number of proinflammatory cytokines, such as tumor necrosis factor (TNF)- and interleukin (IL)-1 (1, 2). Cytokines are implicated in the tissue responses to infection with Mycobacterium tuberculosis including granuloma formation, caseation necrosis and delayed-type hypersensitivity (3). TNF- contributes to the granulomatous reaction that may limit further mycobacterial growth (6, 7). TNF- also upregulates adhesion-molecule expression on macrophages, contributing to heterotypic and homotypic cell adhesion, macrophage differentiation, and enhanced phagocytosis (8). Additionally, TNF- is implicated in tissue necrosis, cavity formation, and cachexia in TB. IL-1 is involved in promoting an immune effector-cell response in TB, with a T-helper 1 (Th1) phenotype, which effectively eliminates the TB bacilli (9, 10). Therefore, it is highly important to understand the immunoregulatory mechanisms involved in the control of proinflammatory cytokine production in TB.

Endogenous nitric oxide (NO) has been implicated in the host defense mechanisms in TB, including cytotoxicity (11). Recent studies, including a study by our group, have shown that NO production is enhanced in alveolar macrophages (AM) of patients with active pulmonary TB through an upregulation of inducible NO synthase (iNOS) activity (12, 13). NO was also reported to regulate the synthesis and release of several proinflammatory cytokines including IL-1, TNF-, and IL-8 (14, 15). In the present study, we investigated whether the enhanced endogenous NO production by AM in TB modulates the synthesis and release of IL-1, and TNF- in the cellular response to mycobacterial invasion.

Transcription factors are implicated in the synthesis of several cytokines. Consensus binding sequences for nuclear factor (NF)-B have been identified in the promoter regions of several cytokine genes, including those for TNF- and IL-1 (16). In myocardium, an inhibition of NF-B has been shown to attenuate the synthesis and release of IL-1 and TNF- (17). Recently, M. tuberculosis and its components were reported to cause a constitutive degradation of inhibitor of B(IB)-, the major cytoplasmic inhibitor of NF-B, leading to NF-B activation in monocytes from TB patients (18). The present study was also designed to examine whether NO regulated synthesis of these cytokines at the transcriptional level.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Study Populations

Eleven patients with active pulmonary TB, including six males and five females, aged 46.6 ± 4.2 yr (mean ± SE) were recruited for the study before administration of anti-TB drugs (Table 1). All patients had at least one recent sputum specimen positive for acid-fast bacilli on microscopic examination, and a positive sputum or bronchoalveolar lavage fluid (BALF) culture for M. tuberculosis. None of the patients were current smokers or were human immunodeficiency virus-positive. The nutrition status of the patients was assessed by measurement of their body mass, height, triceps skinfold thickness, midarm circumference, and serum albumin level. Patients with poor nutrition status (body mass < 90th percentile or midarm circumference and triceps skinfold thickness < 25th percentile) were prospectively excluded from the study to avoid the confounding effect of a poor nutrition status on immunity. Patients with asthma, chronic obstructive pulmonary disease, bronchiectasis, systemic or local airway inflammatory diseases (lupus erythematous, sepsis, and pneumonia), diabetes mellitus, and malignancy were also excluded from the study. None of the enrolled patients took corticosteroids or other immunosuppressants.

The control group consisted of 10 healthy, nonsmoking subjects, consisting of six males and four females with an age of 47.2 ± 5.8 yr. Four of the controls presented with minor hemoptysis, but had negative findings on fiberoptic bronchoscopy and chest radiography. The other six control subjects were volunteers. None of these subjects had a history or evidence of lung disease based on physical, chest radiographic, chest computed tomographic, or bronchoscopic examinations. None had had any upper respiratory tract infection within the previous 6 wk, or were receiving antibiotics or other medications at the time of evaluation. The research protocol was approved by the Chang Gung Memorial Hospital Research Committee. Informed consent was obtained from all subjects.

Bronchoalveolar Lavage

Bronchoalveolar lavage (BAL) was performed before treatment in all study subjects, using five aliquots of 50 ml of prewarmed (37° C) 0.9% saline as described previously (19). Briefly, sterile saline solution was instilled into the involved segment of the lung in TB patients and into the right fourth or fifth subsegmental bronchus in normal subjects. The lavage fluid was retrieved by gentle aspiration, pooled, and filtered through two layers of sterile gauze. The BAL fluid (BALF) was centrifuged at 600 × g for 20 min at 4° C. The cell pellet was washed sequentially and resuspended in RPMI-1640 (GIBCO, Grand Island, NY) supplemented with 5% heat-inactivated fetal calf serum (FCS; Flow Laboratories, Paisley, Scotland) at 10⁶ cells/ml. The cell viability was determined by trypan blue exclusion, and in all cases recovered cells were > 90% viable. Differential cell counts were done by counting 500 cells on cytocentrifuge preparations, using Liu's stain. Lavaged alveolar macrophages (AM) from three TB patients (two male and one female, age (41.8 ± 6.7 yr) and three normal subjects (two male and one female, age 43.7 ± 7.3 yr) were randomly selected for Northern blot analysis. The remaining lavaged AM from all three of the TB patients and two of the normal subjects were discarded and not used for other studies.

Culture of AM

AM were placed in plastic culture dishes in RPMI-1640, allowed to adhere for 60 min, and washed three times with warm RPMI-1640 to remove nonadherent cells. Adherent cells were scraped off with a sterile rubber policeman. Such mechanical detachment retrieved > 96% purified AM, but decreased cell viability to 52.8 ± 8.9% (n = 11) for AM of TB patients and to 54.6 ± 11.7% (n = 10) for AM of normal subjects. After adjustment of cell viability, the cells were resuspended (10⁶ viable cells/ml) in RPMI-1640 medium containing 5% FCS, 100 U/ml penicillin, and 100 µg/ml streptomycin. The purified AM were then placed in 12-well Petri dishes at 10⁶ viable cells/ml for 24 h at 37° C under 5% CO₂. After 24 h of culture, 38.6 ± 6.5% (n = 8) and 41.6 ± 7.9% (n = 8) of viable AM from the TB patients and normal subjects, respectively, were nonadherent. The culture supernatant was collected and centrifuged, and the supernatant was frozen at 70° C until assay. To examine the effect of NO production, AM from patients with active pulmonary TB or from normal subjects were cultured in the presence or absence of the NOS inhibitor N^G-monomethyl-L-arginine (L-NMMA; Calbiochem, La Jolla, CA) at a final concentration of 1 mM for 24 h at 37° C, under 5% CO₂. In one set of experiments, pyrrolidine dithiocarbamate (PDTC, 5 µM) (Sigma Chemical Co., St. Louis, MO), an inhibitor of NF-B, was added to cell culture to examine whether the synthesis of IL-1 and TNF- was associated with activation of NF-B. The supernatants were stored at 70° C until assay for cytokines and nitrite.

Measurement of Nitrite in the Supernatant of Culture Media

To measure the concentration of nitrite, 50 µl of culture-medium supernatant were added to a purge vessel containing 5 ml of a reducing solution (1% potassium iodide in acetic acid), which converts nitrite into NO. Quantification of the NO formed from reactive nitrogen intermediates was done by measuring the specific chemiluminescence resulting from the reduction of NO with ozone, using a chemiluminescence analyzer (Model 280; Sievers, Boulder, CO). Ninety four percent of standard mixtures of nitrite and nitrate solutions was converted to NO, using comparison with calibrated standards of NO gas.

Quantification of IL-1 and TNF- Production by Cultured AM

The concentrations of IL-1 and TNF- were measured with a specific enzyme-linked immunosorbent assay (ELISA) kit (R&D System, Minneapolis, MN) using the quantitative immunometric sandwich enzyme immunoassay technique. For the assay, the frozen supernatants prepared from cultured AM were thawed at room temperature and added to the wells of rigid, flat-bottom microtiter plates coated with murine monoclonal antibody to human IL-1 or TNF-. After incubation of the samples and thorough washing of the wells, horseradish peroxidase (HRP)-conjugated antibodies directed against IL-1 or TNF- were added to the test wells. After a second incubation, excess HRP-conjugated antibody was removed by washing. The HRP substrate was then added and the color intensity was measured with a microtiter plate reader. The limit of detection of IL-1 and TNF- with the assay is 3 pg/ml.

Northern Blot Analysis

Total RNA was extracted from cultured AM through the acid guanidinium thiocyanate method as previously described (20). In each case, 20 µg of RNA was fractionated by electrophoresis through 1% agarose-6% formaldehyde denaturing gel, transferred onto nylon filters (Hybond-N; Amersham, Arlington Heights, IL), and crosslinked by UV irradiation. The filters were incubated in 40 ml of prehybridization solution (50% formamide, 0.5% sodium dodecyl sulfate [SDS], 10× Denhardt's solution, 4% calf thymus DNA) (Boehringer Mannheim, Indianapolis, IN) and 4× sodium chloride-sodium phosphate- ethylenediamine tetraacetic acid [EDTA] at 42° C for 6 to 12 h. IL-1 and TNF- were labeled [-³²P]deoxycytosine triphosphate (specific activity: 3,000 Ci/mmol; New England Nuclear, Boston, MA) with a random primer labeling kit (Boehringer Mannheim). The hybridizations were done at 42° C for 10 to 40 h. The filter was then washed once at room temperature for 20 min in solution containing 2× standard saline citrate (SSC) and 1% SDS, followed by two additional washes at 65° C for 60 min with 0.1× SSC plus 0.1% SDS. To ascertain whether equal amounts of RNA were loaded in individual lanes, blots were stripped in 0.1% SDS at 95° C for 5 min and rehybridized with a ³²P-labeled cDNA probe for glyceraldehyde-3-phosphate dehydrogenase (GAPDH) (American Type Culture Collection, Rockville, MD). Blots were then exposed to film (XAR-5; Eastman Kodak, Rochester, NY) with intensifying screens (Fuji-A cassette; Fuji Photo Film, Tokyo) at 70° C for 24 to 48 h. Autoradiographs were scanned (Scanjet Plus; Hewlett-Packard, Palo Alto, CA) to a Macintosh IIsi computer (Apple Computer Inc., Cupertino, CA) and analyzed with National Institutes of Health Image 1.41 software. The IL-1 and TNF- mRNA levels were calculated as ratios of the GAPDH mRNA level.

Immunocytochemistry of NF-B Subunit p65

AM from four TB patients and four normal subjects were purified by the adherence method described earlier, but were cultured in 12-well Petri dishes. After removing nonadherent cells, the adherent cells were cultured under 5% CO₂ for 6 h at 37° C in RPMI-1640 medium containing 5% FCS, 100 U/ml penicillin, and 100 µg/ml streptomycin in the presence or absence of L-NMMA. To examine the effect of L-NMMA (1 mM) on other causes of NF-B activation, we stimulated AM from four additional normal subjects (two males and two females, age 36.8 ± 3.3 yr), with lipopolysaccharide (LPS) (100 µg/ml) for 6 h in the presence or absence of L-NMMA (1 mM). After culture, adherent AM or AM floating in the culture medium were collected and cytospun on poly-L-lysine-coated microscope slides, and were fixed in cold methanol-alcohol (1 to 1, vol:vol) solution for 10 min, followed by washing in 5% bovine serum albumin in PBS. The slides were then incubated at 37° C for 2 h with a mouse antihuman p65 antibody (Transduction Laboratories, Lexington, KY). This antibody recognizes an epitope corresponding to amino acids mapping within the immunogen of p65. After washing in PBS, the slides were incubated with fluorescein isothiocyanate-conjugated goat antimouse IgG (Jackson ImmunoResearch Laboratories, West Grove, PA) as a second antibody. For the negative controls, mouse immunoglobulin G₁ (Dako, Kyoto, Japan) was used at the same protein concentration as the antibody. After being washed twice in PBS, the slides were counterstained with Liu's stain.

Cell Counts

Counts of NF-B-positive cells for each subject were made on 50 AM that were randomly photographed from at least eight fields at a magnification of ×1,000 under a fluorescence microscope (Zeiss, Wetzlaar, Germany). The percentage of cells with nuclear staining was counted from the photographs by an experienced observer unaware of the cells' origin.

Preparation of Cytoplasmic Extracts

Extracts were prepared from 1 × 10⁶ AM cultured for 6 h with or without treatment with L-NMMA (1 mM) or PDTC (5 µM). Cells were pelleted in microfuge tubes, resuspended in 400 µl of Buffer A (10 mM 4-(2-hydroxyethyl)-1-piperazine-N'-2-ethanesulfonic acid, pH 7.9; 10 mM KCl; 0.1 mM EDTA, pH 8.0; 0.1 mM ethylene glycol-bis- (-aminoethyl ether)-N,N,N',N'-tetraacetic acid, pH 7.8; 1 mM dithiothreitol; 0.5 mM phenyylmethylsulfonyl fluoride; 5 mg/ml pepstatin A; 5 mg/ml leupeptin; 5 mg/ml antipain; 100 mg/ml chymostatin; and 5 mg/ ml aprotinin) (Sigma), and left on ice for 15 min. Following this, 25 µl of Nonidet P 40 (octylphenylpolyethylene glycol) (Fluka, Milwaukee, WI) was added and the cells were agitated vigorously. After centrifugation for 30 s, supernatants were aspirated for immunoblot analysis.

Western Blot Analysis

Cell extracts were subjected to SDS-polyacrylamide gel electrophoresis and blotted onto nitrocellulose filters. NF-B subunit p65 was detected with a mouse antihuman p65 antibody (Transduction Laboratories) and alkaline phosphatase-conjugated antimouse antibody. Blots were developed by adding of 5-bromo-4-chloro-3-indole phosphate/nitroblue tetrazolium solution (Sigma) and were then exposed to XAR-5 film.

Statistics

Data were expressed as mean ± SEM. One-way analysis of variance for mixed design was used to compare values for more than two different experimental groups. If variance among groups was noted, Bonferroni's test was used to determine significant differences between specific points within groups. The data were analyzed with Student's t test for paired or unpaired data. For data with uneven variation, the Mann-Whitney U test or Wilcoxon's signed ranks test was used for unpaired or paired data, respectively. A value of p < 0.05 was considered statistically significant.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

BAL

The recovery rate of BAL from patients with TB (46.3 ± 5.8%, n = 11) was significantly lower than that from normal subjects (70.6 ± 2.7%, n = 10) (p< 0.01) (Table 1). There was an increase in the total cell counts and percent lymphocytes and neutrophils in TB patients over those of normal subjects (Table 1).

Nitrite Production by AM

AM retrieved from patients with active pulmonary TB produced a greater magnitude of nitrite after culture for 24 h (2233.0 ± 1024.0 nM/10⁶ cells, n = 8; p < 0.01) than did those of normal subjects (28.5 ± 4.1 nM/10⁶ cells, n = 8) (Figure 1). Incubation with L-NMMA (1 mM) significantly decreased the nitrite generation by AM of patients with active pulmonary TB as well as AM from normal subjects (Figure 1).

View larger version (13K): [in this window] [in a new window]   Figure 1.   Spontaneous production of nitrite from cultured AM of normal subjects (n = 8) and patients with active pulmonary TB (n = 8) in the presence or absence of L-NMMA (LN; 1 mM; n = 8). Data are mean ± SEM. *p < 0.01 compared with normal subjects (Mann-Whitney U test); #p < 0.01 compared with the corresponding L-NMMA-treated group (Wilcoxon's signed ranks test).

Release and Synthesis of IL-1 and TNF- by Cultured AM

After culture for 24 h, AM from patients with active pulmonary TB released a greater amount of IL-1 (499.1 ± 214.7 ng/ml, n = 8; p < 0.04) and TNF- (437.2 ± 209.0 ng/ml, n = 8; p < 0.01) into supernatants, as measured with ELISA, than did those of normal subjects (118.0 ± 45.5 ng/ml and 22.4 ± 8.0 ng/ml, n = 8, respectively) (Figures 2 and 3). Incubation with L-NMMA significantly inhibited the spontaneous release of IL-1 (311.7 ± 174.2 ng/ml, n = 8; p < 0.03) and TNF- (267.0 ± 160.3 ng/ml, n = 8; p < 0.02) by AM of TB patients but not those of normal subjects (Figures 2 and 3). The magnitude of the inhibitory effect of L-NMMA in decreasing IL-1 (by 30.1 ± 11.1%, n = 8; p < 0.02) and TNF- (by 41.5 ± 14.7%, n = 8, p < 0.03) release by AM of TB patients was significantly greater than that for AM of normal subjects (by 5.2 ± 4.8% and 5.9 ± 9.9%, n = 8).

View larger version (14K): [in this window] [in a new window]   Figure 2.   Spontaneous production of IL-1 by cultured AM of normal subjects (n = 8) and patients with active pulmonary TB (n = 8) in the presence or absence of L-NMMA (LN; 1 mM; n = 8). Data are mean ± SEM. *p < 0.04 compared with normal subjects (Mann-Whitney U test); #p < 0.03 compared with the corresponding L-NMMA-treated group (Wilcoxon's signed ranks test).

View larger version (15K): [in this window] [in a new window]   Figure 3.   Spontaneous production of TNF- by cultured AM of normal subjects (n = 8) and patients with active pulmonary TB (n = 8) in the presence or absence of L-NMMA (LN; 1 mM; n = 8). Data are mean ± SEM. *p < 0.01 compared with normal subjects (Mann-Whitney U test); #p < 0.02 compared with the corresponding L-NMMA-treated group (Wilcoxon's signed ranks test).

Expression of mRNA for IL-1 and TNF- by Cultured AM

To establish the presence of mRNA transcripts, we performed Northern blot analysis on AM for expression of mRNAs for IL-1 and TNF-. Northern blot analysis revealed a time- dependent increase in both IL-1 and TNF- mRNA expression in AM from both normal subjects and patients with active pulmonary TB. There was an upregulation of IL-1 and TNF- mRNA in patients with active pulmonary TB as compared with normal subjects, with equal amounts of total RNA in each lane as demonstrated by the concurrent quantity of GAPDH (Figure 4). Adding L-NMMA (1 mM) to the culture medium significantly decreased the levels of both IL-1 and TNF- mRNA expression in AM of patients with active pulmonary TB (Figure 4).

View larger version (26K): [in this window] [in a new window]   View larger version (13K): [in this window] [in a new window]   Figure 4.   Spontaneous IL-1 and TNF- mRNA expression in AM of normal subjects (NC; hatched bars; n = 3) and TB patients (TB; closed bars; n = 3) in the absence or presence of L-NMMA (LN; 1 mM, n = 3). Total RNA was extracted after 6 h of culture and analyzed by Northern blotting. (A) Autoradiogram; (B) Bar graph of densitometry of bands from IL-1 and TNF- normalized to GAPDH signal and expressed as a ratio of the GAPDH mRNA level. Data are representative of three separate experiments.

NF-B Activation in AM

The enhanced release of IL-1 and TNF- from cultured AM of TB patients (n = 4) was significantly inhibited when these patients' AM were cultured in the presence of PDTC (Figure 5). Immunocytochemistry revealed that the active component of NF-B, the p65 subunit, was constitutively expressed in the cytoplasm of AM from both normal subjects (Figure 6A) and TB patients (Figure 6B). There was increased staining in the cytoplasm and a higher percentage of nuclear staining in AM from TB patients (42.8 ± 8.4%, n = 4; p < 0.03) (Figure 6B) than those from normal subjects (19.2 ± 2.1%, n = 4) (Figure 6A). L-NMMA significantly decreased the cytoplasmic staining and percentage of nuclear staining in AM of TB patients (20.4 ± 2.3%, n = 4; p < 0.03) (Figure 6C). LPS (100 µg/ml), used as a control induced increased staining in the cytoplasm and a higher percentage of nuclear staining in AM from four additional normal subjects (36.5 ± 8.4%, n = 4; p < 0.04) (Figure 6E) than in time control cells of these subjects (20.3 ± 1.6%, n = 4) (Figure 6D). In contrast, L-NMMA failed to modify the cytoplasmic staining or percentage of nuclear staining in AM from TB patients (41.9 ± 5.6%, n = 4) (Figure 6F).

View larger version (11K): [in this window] [in a new window]   Figure 5.   Effect of PDTC (5 µM) on spontaneous release (Control) of IL-1 (A) and TNF- (B) from cultured AM of TB patients (n = 4). Data are mean ± SEM. *p < 0.05 compared with corresponding control (Wilcoxon's signed ranks test).

View larger version (131K): [in this window] [in a new window]   View larger version (36K): [in this window] [in a new window]   View larger version (10K): [in this window] [in a new window]   View larger version (5K): [in this window] [in a new window]   View larger version (17K): [in this window] [in a new window]   View larger version (23K): [in this window] [in a new window]   Figure 6.   Immunofluorescence staining, with anti-p65 polyclonal antibody conjugated with FITC, of AM retrieved from normal subjects (A) and TB patients (B). There was strong labeling in the cytoplasm and nuclear staining in AM of TB patients. L-NMMA (1 mM) significantly inhibited the staining in the cytoplasm and nuclei of AM from TB patients (C ). As a control, LPS (100 µg/ml) induced an increase in the cytoplasmic and nuclear staining in AM from additional normal subjects (E ) as compared with a the time control (D). L-NMMA failed to modify LPS-enhanced staining in the cytoplasm and nuclei of AM (F ). A through F, Original magnification: ×1,000.

Western blot analysis of whole-cell extracts prepared from cultured AM of a TB patient revealed a strongly visible p65 subunit reactivity band (Figure 7). By contrast, p65 subunit reactivity was weak in AM from normal subjects an in L-NMMA- treated AM (Figure 7).

View larger version (13K): [in this window] [in a new window]   Figure 7.   Cytoplasmic extracts from AM were assessed for p65 subunit activity by Western blot analysis. Increased p65 reactivity bands were found for TB patients (TB) as compared with normal subjects (NC). The p65 reactivity decreased when AM were cultured with L-NMMA (LN; 1 mM) as compared with culture in medium alone.

Cytotoxicity

After initial adherence and scraping to purify AM, the cell viability before replating was 52.8 ± 8.9% (n = 11) for AM of TB patients and 54.6 ± 11.7% (n = 10) for AM of normal subjects. Incubation with L-NMMA did not significantly influence the cell viability either of AM from TB patients (53.9 ± 7.2%, n = 11) or those from normal subjects (58.8 ± 9.9%, n = 10). There were no significant difference in cell viability of AM from TB patients and those from normal subjects with or without incubation with inhibitors. Incubation with PDTC slightly, but not statistically significantly, decreased the cell viability of AM from TB patients (n = 4), to 39.8 ± 10.5%.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The present study demonstrated that AM from patients with active pulmonary TB synthesize and spontaneously release an increased amount of IL-1 and TNF-. An upregulation of mRNA in macrophages is probably the mechanism of the increased release of these proteins. AM from patients with active pulmonary TB also concomitantly released a large amount of nitrite. Inhibition of endogenous NO production by L-NMMA significantly inhibited the release of IL-1 and TNF-, as well as inhibiting expression of their respective mRNAs in AM from TB patients. These results suggest that the synthesis and release of these proinflammatory cytokines by AM in patients with active pulmonary TB are modulated by the generation of endogenous NO.

Our results in this study, and other recent studies (12, 13), have shown that the enhanced NO production by AM of TB patients is derived from an upregulation of iNOS in these cells. The lipoarabinomannan cell-wall components of mycobacteria have been shown to directly induce NO production by murine and human macrophages (21, 22). Several proinflammatory cytokines, such as interferon-, TNF-, and IL-1 are also implicated in inducing iNOS expression in murine monocytes (23, 24). Thus, human AM might be activated to upregulate their NO synthase activity by both cytokines and by mycobacterial components.

NO may play an important role in resistance to M. tuberculosis infection. NO has been proposed to mediate growth inhibition in murine models (25), but conflicting evidence for such a role exists in humans (21, 26). Nevertheless, our results indicated that NO may play a role in regulating proinflammatory cytokine synthesis and release in pulmonary TB. The generation of TNF- and IL-1 may effectively eliminate the bacilli (8). Thus, the enhanced generation of NO by AM in pulmonary TB may provide a mechanism for self-amplifying signalling in the inflammatory response to infection with M. tuberculosis.

Previous studies of both human and murine phagocytes have indicated that NO plays a regulatory role in IL-1 and TNF- production (27). In the human myeloid leukemia cell line HL-60, NO gas as well as NO donors were shown to increase expression of the mRNAs for both TNF- and IL-1 (31). Endogenous NO was also reported to upregulate TNF- production in transfected phorbol myristate acetate-differentiated U937 cells (27). These findings are consistent with our results, and suggest that NO has an upregulatory effect on TNF- and IL-1 production. However, under certain conditions, NO may downregulate cytokine production. SIN-1 as an NO donor was reported to upregulate TNF- synthesis induced by IL-1, but downregulated that induced by LPS in human mononuclear cells (32). In LPS-stimulated murine RAW264.7 cells, inhibition of NO synthase increased TNF- production (28). Such a potential downregulatory effect of NO on cytokine synthesis was also shown in rat AM (30). It is not clear whether the differences among these studies are related to the source or dose of exogenous NO or to the type of cell or stimuli used. In the present study, L-NMMA failed to inhibit the release of IL-1 and TNF- by AM from two of eight TB patients. The baseline release of IL-1 and TNF- by AM of these two TB patients was as low as that of AM from normal subjects. It is possible that AM from TB patients might not be activated and would therefore be less responsive to the inhibitory effect of L-NMMA as those AM of normal subjects. A difference in activation status may therefore also determine the responsiveness to NO in modulation of cytokine release.

M. tuberculosis or its cell-wall components can directly stimulate mononuclear phagocytes in vitro to release IL-1, IL-6, and TNF-, and can upregulate their respective mRNAs (33). M. tuberculosis and its components are also reported to cause a constitutive degradation of IB-, leading to NF-B activation in monocytes from TB patients (18). In the present study, we assessed NF-B activation by immunostaining and Western blot analysis with an antibody to the p65 subunit, which is responsible for the interaction of NF-B with IB (40). In response to stimuli, IB is phosphorylated and induces dissociation of the IB/NF-B complex, allowing the free NF-B to translocate to the nucleus. Thus, detection of p65 subunits in the cytoplasm and nuclei indicates an induction of transcription and activation of NF-B (41). Our results showed that expression of the p65 subunit in the cytoplasm of AM, and its nuclear translocation, were enhanced in AM of TB patients as compared with those of normal subjects. Inhibition of NF-B activation by PDTC significantly attenuated IL-1 and TNF- release, suggesting that NF-B was activated and that it mediated synthesis of these cytokines in AM of TB patients. L-NMMA treatment of these patients' AM inhibited NF-B activation and translocation to the nucleus, indicating that the modulatory effect of endogenous NO on the synthesis of IL-1 and TNF- in the patients' AM probably occurred through NF-B activation. Our results are consistent with a recent report that NO increases cytokine expression through activation of NF-B in the inflammatory response after hemorrhagic shock (42). In addition, NO was also reported to upregulate TNF- production in peripheral blood mononuclear cells through a change in the binding activity of NF-B (29).

Taken together, the findings in the present and other studies indicate that IL-1, TNF-, and iNOS are concomitantly upregulated in AM in response to mycobacterial stimulation. The enhanced NO generation by AM in TB patients may play an autoregulatory role in amplifying the synthesis of proinflammatory cytokines through NF-B activation.