Abstract
Phagocytic cells provide the first line of defense against mycobacteria. We examined the relative mycobacteriostatic contributions of normal human alveolar macrophages (HAM), peripheral blood monocytes (PBM), and polymorphonuclear leukocytes (PMN) in the early time period after infection with mycobacteria (48 h). Cells were infected with Mycobacterium bovis (BCG) or M. tuberculosis H37Ra and their ability to inhibit growth was determined by mycobacterial incorporation of [3H]uracil. HAM inhibited the growth of both mycobacteria (44.2 ± 7.9 and 37.6 ± 10.5% inhibition, respectively). Two populations of HAM donors were subsequently defined: inhibitors and noninhibitors. The ability to inhibit growth of H37Ra correlated with that of BCG. In contrast to HAM, PBM and PMN did not inhibit mycobacterial growth. Because nitric oxide (NO) has been proposed to mediate growth inhibition in murine models, we examined whether NO was responsible for the early growth inhibition of mycobacteria by HAM. As expected, in murine peritoneal macrophages (MPM) IFN- γ (2,500 U / ml) enhanced growth inhibition of BCG; the effect was abolished by the nitric oxide synthase (NOS) inhibitor NMMA. In contrast, IFN- γ failed to enhance growth inhibition by HAM or PBM and NMMA had no effect. MPM expressed inducible nitric oxide synthase (NOS2) mRNA in response to LPS and IFN- γ and produced NO. Neither NOS2 mRNA nor NO could be detected in HAM stimulated with LPS and IFN- γ or mycobacteria. These data demonstrate that HAM, but not PBM or PMN, have NO-independent mycobacteriostatic activity in the early time period after infection with mycobacteria.
Immunocompetent subjects exposed to Mycobacterium tuberculosis mount an early and late immune response that ultimately destroys most tubercle bacilli and usually prevents development of clinical disease. Whereas the late response is dependent on the acquisition of CD4+ T-cell-mediated immunity and characterized by granuloma formation consisting of epithelioid and multinucleated giant cells (1, 2), the early response is characterized by an influx of phagocytic cells. These phagocytic cells provide the first line of defense against mycobacteria. Their ability to inhibit the growth or kill M. tuberculosis may determine the course of infection. Phagocytic cells involved in the early human host response to M. tuberculosis include neutrophils (PMN), and cells of the monocytic lineage such as monocytes and alveolar macrophages. Zhang and colleagues (3) have recently demonstrated that sites of the lung involved in acute infection with M. tuberculosis are characterized by an abundance of PMN in the bronchoalveolar lavage (BAL). An increase in monocytic phagocytes is demonstrated by the abundance of alveolar macrophages in the BAL (4).
The question arises as to which cells are capable of destroying mycobacteria early in the infection. Various reports have demonstrated antimycobacterial effects by different phagocytic cells, including PMN (5), monocytes (6) and human alveolar macrophages (HAM) (7, 8). However, direct comparison of these studies is difficult because of differences in strains of mycobacteria, infecting ratios, and techniques used to measure mycobacterial viability.
Binding and phagocytosis of mycobacteria is mediated by multiple cell surface receptors, including those for fibronectin (CD29/CD49d), mannose, hyaluronic acid (CD44), surfactant protein A and complement (CR1; CD35, CR3; CD11b/CD18, CR4; CD11c/CD18) (9-11). Redundancy in receptors function has been demonstrated since blocking a specific receptor does not block phagocytosis (12). Different receptors may be used by discrete cell types to stimulate intracellular signals, which may participate in an effect on mycobacterial growth. Phagocytosis is also influenced by serum factors such as complement or opsonizing antibodies (7, 9). Once phagocytosed, the intracellular plight of mycobacteria differs from that of most other bacteria. Lysosomal fusion of mycobacteria-containing phagosomes is abnormal in murine peritoneal, bone-marrow-derived macrophages and human monocyte-derived macrophages. This abnormal fusion, characterized by an absence of the lysosomal proton ATPase, prevents destruction of mycobacteria (13, 14).
The mechanisms by which phagocytic cells can destroy mycobacteria are unknown, but several have been proposed. Reactive oxygen products have been thought to be uninvolved in mycobactericidal function (15). Apoptosis has recently been suggested as a mechanism for killing of intracellular M. avium-M. intracellulare (16) and infection of HAM with M. tuberculosis promotes apoptosis (17). Although purified human neutrophil defensins kill M. tuberculosis, killing of mycobacteria by intracellular defensins has not yet been documented (18). Nitric oxide (NO) has been well demonstrated to be critical for mycobactericidal activity of murine phagocytes (15). The role of NO in the human response to mycobacteria is less well characterized. The presence of nitric oxide synthase (NOS2) has recently been described in alveolar macrophages from patients with tuberculosis (19), but the kinetics of its production and its role in mycobacteriostasis remains to be clarified (20).
To further understand the role of phagocytic cells as the early defense against infection with mycobacteria we evaluated the relative ability of human alveolar macrophages (HAM), peripheral blood monocytes (PBM), and polymorphonuclear leukocytes (PMN) to inhibit the growth of mycobacteria within the early time period after infection. We demonstrated that HAM can inhibit the growth of M. tuberculosis H37Ra as well as BCG in this early phase. In contrast, PBM and PMN failed to inhibit growth. Unlike the murine model, mycobacteriostasis in HAM and PBM was not enhanced by IFN-γ or was it dependent on production of NO. These data demonstrate a different capacity between human phagocytic cells to provide protection from early infection with mycobacteria and provide insight into mechanisms involved in mycobacterial destruction.
HAM were obtained from two sources. Lungs of transplant donors from The National Disease Research Interchange (Philadelphia, PA) were used as one source. These lungs were obtained within 24 h and were lavaged with 50-ml aliquots of phosphate-buffered saline (PBS) containing heparin (5 U/ml). PPD status of these donors was unknown. In addition, HAM were obtained from normal volunteers after bronchoscopy and bronchoalveolar lavage with an Olympus bronchoscope according to the New York University Review Board of Research Associates approved protocol. Normal subjects had normal physical examinations, chest radiographs, EKG, and pulmonary function tests; 38% of these donors were PPD positive.
HAM from both sources were washed three times (PBS) and subsequently purified by adherence to bacterial petri dishes in RPMI supplemented with 10% FBS with cefotaxime (50 μg/ml) (Hoechst-Roussel, Somerville, NJ) and fungizone (2.5 μg/ml) (GIBCO-BRL, Gaithersburg, MD), henceforth referred to as supplemented RPMI. After incubation (37° C, 5% CO2, 18 h), nonadherent cells and antibiotics were removed by extensive washing with PBS. Cells were > 95% pure (Wright's stain, α-naphthyl acetate esterase staining and mAb staining with HAM56; Dako, Carpinteria, CA). Adherent HAM were subsequently removed by vigorous pipetting (0° C) prior to seeding into 96-well plates.
Peripheral blood monocytes (PBM) were prepared from normal subjects. Heparinized (10 U/ml) venous blood was diluted 1:1 with PBS and centrifuged over Ficoll-Paque. PBM were adhered to heat-inactivated human AB serum-coated plates (15 min) in supplemented RPMI (21) and lymphocytes were removed by extensive washing with PBS. Viability was > 95% and cells were identified morphologically and were > 85% pure (Wright's stain and α-naphthyl acetate esterase staining). Adherent PBM were subsequently removed by vigorous pipetting (0° C) prior to seeding into 96-well plates.
Monocyte-derived macrophages were prepared by prolonged incubation of PBM before their exposure to mycobacteria (10). PBM from a single donor were seeded at 1 × 105 cells/well in four 96-well plates. The PBM were cultured in supplemented RPMI (37° C, 5% CO2) for 1 to 9 d. At a defined time point, monolayers were washed twice (PBS) and infected with mycobacteria (48 h).
Peripheral blood PMN were isolated by Ficoll-Paque centrifugation and dextran sedimentation followed by hypotonic lysis of residual erythrocytes. Viability was > 95%.
Murine peritoneal macrophages (MPM) were obtained from BALB/ cAnNCrlBR mice (Charles River Laboratories, Wilmington, MA). Mice were killed by cervical dislocation, and resident MPM were harvested by flushing the peritoneal cavity with a total volume of 30 ml PBS. Yields of MPM were 1 to 2 × 106 per mouse and, typically, cells from three allogeneic mice were combined for each experiment. Cells were seeded in 96-well plates in supplemented RPMI. After 18 h, nonadherent cells were washed away with PBS.
M. bovis (bacillus Calmette-Guérin; BCG Glaxo; ATCC 35741), and M. tuberculosis H37Ra (ATCC 25177) was grown in Middlebrook 7H9 broth (Becton Dickinson, Cockeysville, MD) supplemented with albumin-dextrose complex (ADC) and fungizone (2.5 μg/ml) in 5% CO2 (37° C). Mycobacteria were subcultured every 14 d to ensure continuous logarithmic growth. Before each experiment, log-phase bacteria were pelleted, washed with PBS, and resuspended in RPMI supplemented with 10% FBS and fungizone (2.5 μg/ml). To ensure addition of single bacilli, bacteria were disaggregated in an ultrasonic water bath (2 min) (7, 14). Mycobacteria were centrifuged (2 min, 850 × g) and the supernatant, containing single bacteria, was removed. Mycobacteria were quantified by absorbance at 300 nm. OD was correlated with colony-forming units (CFU) from serial dilutions of mycobacteria on 7H9 agar. An OD of 0.2 corresponded to 1.3 × 107 BCG/ml and 2.6 × 107 H37Ra/ml.
Inhibition of mycobacterial growth was determined by measuring [3H]uracil incorporation into mycobacteria (6, 22, 23). All cells (1 × 105) were seeded in flat-bottomed 96-well plates in 100 μl RPMI supplemented with 10% FBS and fungizone. Wells were prepared in quintuplicate for each condition. These cells are differentiated and do not divide. Whereas HAM and PBM developed adherent monolayers, PMN and lymphocytes did not adhere to the wells. In experiments to evaluate the role of IFN-γ and NO on inhibition of mycobacterial growth, cells were preincubated with either human rIFN-γ (R&D Systems, Minneapolis, MN) or murine rIFN-γ (Genzyme Diagnostics, Cambridge, MA) at a concentration of 2,500 U/ml (24 h) in the presence or absence of NG-monomethyl-l-arginine (NMMA) (0.5 mM, 1 h) (Sigma, St. Louis, MO). The media used during the time of mycobacterial growth was the same as that used to prepare the phagocytes except that the cefotaxime was omitted. Mycobacteria (100 μl in RPMI, 10% FBS, fungizone) were added to the cells at a multiplicity of infection (MOI) of mycobacteria:cells 15:1. This MOI allowed adequate measurement of the incorporation of [3H]uracil with minimal cytotoxicity. Inhibition of growth of BCG and M. tuberculosis H37Ra by HAM was independent of the MOI between 5:1 and 50:1 mycobacteria:HAM. At defined time points, cells were permeabilized (saponin, 0.05%), and [3H]uracil (0.25 μCi) (New England Nuclear, Boston, MA) was added (6, 22, 23). Incorporation of [3H]uracil by viable mycobacteria was allowed (37° C, 16 h) after which sodium azide (0.2%) was added to kill the mycobacteria. Incorporation of [3H]uracil into bacterial RNA was determined using standard cell-harvesting techniques. Briefly, the total contents of each well were adsorbed onto a betaplate (LKB-Wallac, Gaithersburg, MD) and incorporation of [3H]uracil was determined with a betaplate counter (LKB-Wallac). Mycobacteria have been demonstrated to incorporate 80% of the [3H]uracil into RNA and 20% into DNA (24). Incorporation of [3H]uracil by uninfected phagocytic cells accounted for < 1% of total [3H]uracil incorporation and was subsequently subtracted as background from all experiments. This technique would measure [3H]uracil incorporated into intracellular mycobacteria as well as any extracellular mycobacteria; however, in preliminary studies, incorporation of [3H]uracil in supernatants from wells containing cells and mycobacteria was < 3% of total [3H]uracil. Cold uracil release by monocytes does not influence incorporation of [3H]uracil by mycobacteria (6). Incorporation of [3H]uracil by BCG alone was directly proportional to the number of mycobacteria in the wells over the range 1.5 × 104 to 1.5 × 107 BCG/ well. Because growth of mycobacteria was detected in media alone, ability of phagocytic cells to inhibit growth is expressed as a percentage of [3H]uracil uptake by mycobacteria grown in the absence of cells according to the formula: 1-[3H]uracil uptake in mycobacteria + cells/ [3H]uracil uptake in mycobacteria × 100 at the same point in time. Because this assay cannot distinguish between bacteriostatic and bactericidal activity we refer to these activities as inhibition of mycobacterial growth. Cell viability, determined by trypan blue exclusion was > 90% for control and infected cells at 48 h. The viability of the infected cells decreased to 75% by 72 h. Therefore, we used 48 h exposure, when we had > 90% phagocytic cell viability, as our time point for determining mycobacterial growth.
To confirm the results of the [3H]uracil assay, mycobacteria were also quantified by determining CFU. Cells in 96-well plates were lysed with Tween 80 (0.1%) (Sigma) and cell lysates were sonicated (16). Serial dilutions of mycobacteria were spread on petri dishes containing 7H10 agar supplemented with ADC and fungizone. Plates (prepared in triplicate) were incubated (4 wk) in 5% CO2 at 37° C and CFU were counted. Inhibition of growth was determined by comparing CFU from wells containing mycobacteria in the absence of cells. Initial studies of the experimental system using THP-1 monocyte/macrophage cell lines infected with BCG and M. tuberculosis H37Ra showed a strong correlation of [3H]uracil incorporation with counting of CFU.
Primers specific for NOS2 sequences were designed that were complementary for both human and murine NOS2 mRNA (Genbank accession L09210 and M87039) (25). The upstream primer was CTACTCCATCAGCTCCTCCC, and the downstream primer was ACAGCACCGAAGATATCTTC. Primers amplified a 671 bp fragment, which spanned nucleotides 2927 to 3598 in the human and 2958 to 3629 in the murine sequence. Total RNA was isolated using TRIzol reagent (GIBCO-BRL), and first-strand synthesis of cDNA was performed using a Superscript Preamplification Kit (GIBCO-BRL). Template first-strand cDNA was added to a reaction mixture that included dNTPs (0.2 mM), Mg2+ (25 mM), appropriate oligonucleotide primers (1 μm) and Taq polymerase (0.025 U/μm) (Promega, Madison, WI) in a final volume of 50 μl. PCR conditions were 94° C (5 min), 40 cycles of 94° C (1 min), 60° C (1 min), 72° C (1 min 30 s), followed by 72° C terminal extension (10 min). PCR products were resolved in 2% agarose gels and stained with ethidium bromide. The identity of the PCR products was determined by digestion with the 6-bp cutter restriction enzymes ApaLI, NcoI, PstI, or PvuII, which cut the product once and gave bands of the predicted sizes. PCR products of equal intensity derived from primers specific for the housekeeping gene glyceraldehyde-3-phosphate dehydrogenase (GAPDH) were used to demonstrate similar starting amounts of cDNA.
Nitrite (NO2 −), an oxidative end product of NO synthesis, was determined by colorimetric assay using the Griess reagent (26). Briefly, 24-h cell culture supernatants were added to a 96-well enzyme-linked immunosorbent assay (ELISA) plate in triplicate, and an equal volume of Griess reagent (1% sulfanilamide, 0.1% napthylethylenediamine, 2.5% H3PO4) was added. The absorbance at 550 nm was measured on an ELISA plate reader, and the concentration of NO2 − was calculated by comparing optical density values with a standard curve of NaNO2 in supplemented RPMI.
All data are presented as mean ± SEM. Data were analyzed using the paired or unpaired t test when applicable, with significant values considered as p < 0.05. Correlatation was determined using unweighted κ.
To determine the ability of HAM to inhibit the growth of mycobacteria, HAM were seeded in 96-well plates and BCG or M. tuberculosis H37Ra was added at a MOI of 15:1 (mycobacteria:cells). Mycobacteria were also plated at the equivalent number in media alone in the absence of cells. Growth of BCG (Figure 1) and M. tuberculosis H37Ra (data not shown) in media alone displayed a time-dependent increase in uptake of [3H]uracil consistent with a doubling time of 24 h. This rate of growth correlates with that previously published (14). HAM demonstrated a rapid time-dependent ability to inhibit the growth of BCG. Most of the inhibition of growth was reached by 24 h (56 ± 15% inhibition, n = 3, p = NS), although inhibition of growth continued at 48 h (76 ± 9% inhibition, n = 3, p < 0.05). HAM from all subjects inhibited the growth of BCG at 48 h (44.2 ± 7.9% inhibition, n = 12, p < 0.0005) (Figure 2). HAM derived from all subjects inhibited the growth of M. tuberculosis H37Ra (37.6 ± 10.5% inhibition, n = 12, p < 0.005). However, there was clear donor variability in the response of HAM to M. tuberculosis H37Ra, and therefore we also analyzed the data in two groups. These groups were defined as “inhibitors” and “noninhibitors.” An inhibitor was defined as a subject whose HAM had > 1% inhibition of growth of M. tuberculosis, and a noninhibitor was defined as a subject whose HAM were permissive for mycobacterial growth. Inhibition of growth of M. tuberculosis H37Ra by inhibitor and noninhibitor groups was 59.1 ± 7.6% inhibition, n = 8, p < 0.0001 and −5.5 ± 2.9% inhibition, n = 4, p = NS, respectively. Inhibition of growth of M. tuberculosis H37Ra correlated with inhibition of growth of BCG (coefficient of 0.7705, unweighted κ of 0.47). Although of small sample size, differences in ability of HAM to inhibit growth did not correlate with PPD status of the normal subjects.
Fig. 1. Inhibition of growth of BCG by HAM. HAM (1 × 105) were infected with mycobacteria (MOI, 15:1) and viability of mycobacteria was determined by incorporation of [3H]uracil. Data are presented as cpm of mycobacteria grown in the presence of cells compared with cpm of mycobacteria grown in media alone (mean ± SEM, n = 3). After 48 h of exposure, HAM elicited a 64% decrease in cpm.
[More] [Minimize]
Fig. 2. Inhibition of growth of BCG and M. tuberculosis H37Ra by HAM. HAM (1 × 105) were infected with mycobacteria (MOI, 15:1) and the effect on mycobacterial growth was determined at 48 h. Inhibition of growth is expressed for total HAM (n = 12), as well as for two groups defined as “inhibitors” (n = 8) and “noninhibitors” (n = 4) based on the ability of HAM to inhibit growth of M. tuberculosis H37Ra. These groups were compared by unpaired t test.
[More] [Minimize]Because HAM are derived predominantly from monocytes (27), and because monocytes phagocytose mycobacteria (7), we examined whether PBM altered the growth of mycobacteria. Freshly isolated PBM were seeded into 96-well plates and exposed to mycobacteria (MOI of 15:1, 48 h) and the incorporation of [3H]uracil by mycobacteria was determined. In contrast to the effects of HAM, freshly isolated PBM did not significantly inhibit growth of BCG or M. tuberculosis H37Ra (Figure 3).
Fig. 3. Inhibition of growth of BCG and M. tuberculosis H37Ra by HAM and peripheral blood leukocytes. Cells (1 × 105) were infected with mycobacteria (MOI, 15:1) and the effect on growth was determined at 48 h for HAM, PBM, and lymphocytes and at 4 h for PMN. Results are expressed as means ± SEM (n = 3 to 18).
[More] [Minimize]To determine whether the macrophage differentiation of PBM would affect inhibition of mycobacterial growth, monocytes were differentiated into monocyte-derived macrophages by adherence for 1 to 9 d. Monocyte-derived macrophages were exposed to BCG or M. tuberculosis H37Ra (48 h) after various days of maturation. Freshly isolated monocytes or 1-d-old monocyte-derived macrophages failed to inhibit the growth of mycobacteria. As monocyte-derived macrophages were allowed to differentiate further there was a trend for them to inhibit the growth of both BCG and M. tuberculosis H37Ra. Monocyte-derived macrophages showed maximal inhibition of growth of BCG (47.8 ± 2.7% inhibition, n = 4, p = NS) after 7 d of differentiation and maximal inhibition of M. tuberculosis H37Ra (36.7 ± 4.5% inhibition, n = 3, p < 0.05) after 4 d.
PMN are present at sites of active mycobacterial infection and have been suggested to kill mycobacteria (3, 5). To determine whether human PMN inhibited the growth of mycobacteria, PMN were incubated with BCG or M. tuberculosis H37Ra and inhibition of growth was determined by uptake of [3H]uracil. Because of the short life-span of PMN, inhibition of growth of mycobacteria was measured only after 4 and 24 h of exposure. PMN were > 95% viable at 24 h. PMN failed to inhibit the growth of BCG or M. tuberculosis H37Ra at 4 h (Figure 3) or 24 h (not shown). Because of the reported requirement for autologous serum for the phagocytosis of mycobacteria by phagocytes (7, 9), experiments were also performed in the presence of 10% autologous serum. Autologous serum did not alter growth inhibition of either mycobacteria (not shown).
As expected, lymphocytes isolated from peripheral blood failed to inhibit the growth of mycobacteria.
In murine models, IFN-γ is a potent cytokine that is required for host defense against mycobacteria (28, 29). We therefore compared the response of murine peritoneal macrophages (MPM) with that of human phagocytes. MPM displayed a modest, but not statistically significant, inhibition of the early infection by BCG and M. tuberculosis H37Ra at 48 h (35.3 ± 16.6% inhibition of growth, n = 4, p = NS and 47.7 ± 21.6% inhibition, n = 3, p = NS, respectively). As expected, preincubation of MPM with IFN-γ (2,500 U/ml, 24 h) enhanced the ability of MPM to inhibit the growth of BCG and M. tuberculosis H37Ra (193% more inhibition than control, p < 0.05 and 172% of control, p = NS, respectively) (Figure 4, top panel). To determine whether IFN-γ also enhanced mycobactericidal activity in HAM and PBM, HAM or PBM were also preincubated with IFN-γ (2,500 U/ml, 24 h) and then exposed to BCG or M. tuberculosis H37Ra. IFN-γ neither enhanced nor suppressed inhibition of growth of BCG or M. tuberculosis H37Ra by HAM (Figure 4, middle panel). In PBM, IFN-γ actually suppressed the ability to inhibit growth of BCG (81% of control, p < 0.05), but the effect on M. tuberculosis H37Ra did not reach statistical significance (Figure 4, bottom panel).
Fig. 4. Effect of IFN-γ and NMMA on inhibition of growth of BCG and M. tuberculosis H37Ra. Cells (1 × 105) were incubated with IFN-γ (2,500 U/ml, 24 h) in the absence or presence of NMMA (0.5 mM, 1 h) prior to infection with mycobacteria (MOI, 15:1). Inhibition of growth by MPM (top panel ), HAM (middle panel ), and PBM (bottom panel ) was determined after 48 h of infection. Data were normalized and are presented as a percentage of inhibition by untreated cells to allow for comparison of the effects of IFN-γ and NMMA between the various cell types. Results shown are means ± SEM (n = 3 to 8).
NO has been well described to participate in the murine response to mycobacteria (15). To determine the role of NO in inhibition of mycobacterial growth, inhibition by MPM was evaluated in the presence of the nitric oxide synthase (NOS) inhibitor NMMA. As expected, IFN-γ-stimulated growth inhibition by MPM was prevented by NMMA (Figure 4, top panel). The response of human phagocytes differed from that of MPM. NMMA had no effect on the ability of IFN-γ-stimulated HAM or PBM to inhibit growth of either mycobacteria (Figure 4, middle and bottom panels).
To determine whether cells were capable of synthesizing NOS2, we evaluated the presence of mRNA for NOS2 by RT-PCR. After 40 cycles of PCR, amplification products derived from NOS2 mRNA were detected in MPM and in the J774 murine macrophage cell line after stimulation with lipopolysaccharide (LPS) (100 ng/ml) + IFN-γ (2,500 U/ml). In contrast, amplified RT-PCR products were not obtained from HAM that had been stimulated for 2 or 24 h with BCG, M. tuberculosis H37Ra, or LPS + IFN-γ (n = 3) (Figure 5).
Fig. 5. RT-PCR analysis of NOS2 mRNA expression in phagocytic cells after exposure to mycobacteria. HAM (lanes 1 to 4) were cultured with PBS (lane 1), BCG (MOI, 15:1; lane 2), M. tuberculosis (lane 3), or LPS (100 ng/ml) + IFN-γ (2,500 U/ml) (lane 4). Murine peritoneal macrophages (lane 5) and J774 cells (lane 6) were cultured with LPS (100 ng/ml) + IFN-γ (2,500 U/ml). Data are presented as a representative sample (n = 3).
[More] [Minimize]To determine whether NO was produced by the various cells, the release of nitrite was measured using the Griess reagent in a colorimetric assay. MPM produced nitrite after infection with BCG (27.5 nmol/106 cells/24 h) or LPS + IFN-γ (9.6 nmol/106 cells/24 h) (n = 3). Nitrite was not detectable in 24-h culture supernatants from HAM that had been stimulated with BCG or LPS + IFN-γ (n = 9) (Figure 6).
Fig. 6. NO2 − release in 24-h culture supernatants by HAM and MPM. NO2 − release was determined by a colorimetric assay using the Griess reagent. Results are expressed as mean ± SEM (n = 3 to 9).
[More] [Minimize]To understand the critical role of phagocytic cells in the host defense against mycobacteria, it is important to delineate their ability to inhibit the growth of these organisms. Human alveolar macrophages, monocytes, and PMN are all found in inflamed areas of the lung associated with mycobacteria, and they participate in the early response to mycobacterial infection. Thus, we compared the ability of each of these cells to inhibit growth of mycobacteria. HAM, but neither PBM nor PMN, inhibited the growth of BCG or M. tuberculosis H37Ra. These data, using normal human phagocytic cells and a consistent system with a low infectivity rate clearly characterize the differing potential between discrete cell types to affect the growth of mycobacteria. The mechanism by which growth is inhibited remains obscure. Although most likely an intracellular process, inhibition of growth of mycobacteria may occur in both the intracellular and the extracellular environment. Our system would not differentiate between these two mechanisms. Mehta and colleagues (30) have shown that cells types differ in their ability to incorporate mycobacteria and alter cell growth. One possible mechanism by which HAM may have an enhanced ability to inhibit growth of mycobacteria may be via an enhanced ability to incorporate mycobacteria intracellularly. However, HAM, monocytes, and PMN have all been documented to phagocytose mycobacteria, albeit through different receptors. A difference in bacteriostatic properties of each cell type may also be due to differences in intracellular signaling events induced by engagement of these select receptors (7, 9). In addition, extracellular processes, including the release of inflammatory mediators or oxidants, could also potentially account for differences in growth inhibition.
We clearly demonstrate that HAM were bacteriostatic for mycobacteria. Our data on HAM concur with those of Steele and colleagues (8). These investigators examined the ability of mixed BAL cells predominantly from smokers and diseased patients to inhibit the growth of a M. tuberculosis clinical isolate. Hirsch and colleagues (7) also demonstrated inhibition of growth of M. tuberculosis H37Ra by HAM. However, this study used a high MOI (100:1), whereas we used a low infecting ratio consistent with the experiments of others (6, 22, 23) and one that more closely resembles a physiologic response. We have also found cell toxicity at higher rates of infection. We also demonstrated the presence of two types of responses. HAM derived from all subjects inhibited the growth of BCG to a statistically significant amount (44.2 ± 7.9% inhibition). However, there were clearly differences in the response. Thus, we characterized the subjects into two populations: inhibitors and noninhibitors as defined by their response to M. tuberculosis H37Ra. When we evaluated the data in this manner there was a correlation between the ability of cells to inhibit growth of BCG and M. tuberculosis H37Ra. These data suggest that individual host factors may therefore ultimately influence mycobacterial virulence.
In contrast to HAM, PBM did not significantly inhibit the growth of BCG or M. tuberculosis H37Ra. Previous studies have demonstrated abnormal phagosomal maturation in monocytes infected with mycobacteria (14), and the inability of monocytes to inhibit mycobacterial growth in our system may be due to this reported abnormality. Although it did not reach statistical significance, there was a trend for monocytes to inhibit growth as they differentiated. This finding is in agreement with Douvas and colleagues (31) who demonstrated increased inhibition of growth of M. tuberculosis Erdman in monocyte-derived macrophages cultured for 3 and 7 d, compared with freshly isolated PBM.
PMN have been hypothesized to affect mycobacteria because of their early presence at sites of infection, their ability to phagocytose, and their abundance of bactericidal proteins (18). We failed to measure any significant inhibition of growth of BCG or M. tuberculosis H37Ra by PMN in the absence or presence of autologous serum. Our data are in contrast to those of Jones and colleagues (5) who reported rapid inhibition of growth of M. tuberculosis H37Ra by PMN. However, they used evolution of 14CO2 from growing cultures as an index of subsequent mycobacterial growth and their growth inhibition was less apparent when the MOI was increased from 1:10 to 1:1 H37Ra:PMN. Our data are in agreement with Denis (32) who found no evidence of killing of M. tuberculosis by PMN.
IFN-γ has been well documented to be essential in the murine response to mycobacteria (15, 22, 28, 29). Newport and colleagues (33) have shown that humans who have a susceptibility to mycobacterial infections also have a mutation in the IFN-γ receptor gene, suggesting a role for IFN-γ in the human host response to mycobacteria. IFN-γ enhanced inhibition of growth of mycobacteria by murine peritoneal macrophages in our system. However, in contrast to the effect on murine cells, IFN-γ failed to enhance inhibition of growth of BCG or M. tuberculosis H37Ra by human alveolar macrophages. These data demonstrate a clear difference between murine and human responses. Our study does not exclude a role for IFN-γ in the human host response to M. tuberculosis, but it suggests that its importance may be in the late, rather than the early, response to infection.
NO has been demonstrated to be critical for the response to mycobacteria in murine models (15). We too were able to demonstrate a role for NO in murine cells since inhibition of growth of mycobacteria by murine macrophages was prevented by the nitric oxide synthase inhibitor NMMA, and NOS2 mRNA and NO were clearly detected in murine macrophages. In contrast to the responses in murine cells, we did not detect a role for NO in our human system. Inhibition of growth of mycobacteria by HAM or PBM was not mediated by NO since NMMA had no effect on inhibition of growth. We were also unable to demonstrate expression of NOS2 mRNA or production of NO itself in HAM. Interestingly, Nicholson and colleagues (19) recently described the presence of NOS2 by immunohistochemistry in HAM from patients with untreated culture-positive M. tuberculosis infection, whereas HAM from normal control subjects did not stain for this enzyme. Our data differ from those of Nicholson and colleagues since we were unable to detect mRNA for NOS2 even in cells stimulated with mycobacteria. However, their studies were performed in chronically infected patients, and it is possible that, whereas the early response of HAM to mycobacteria does not stimulate and involve the production of NO, the late, or chronic, response may be associated with an induction in mRNA for NOS2.
These results demonstrate clear differences between human and murine models but the question of the mechanism by which human cells inhibit the growth of mycobacteria remains unanswered. Interestingly, M. tuberculosis H37Rv prevents spontaneously occurring apoptosis in PBM (34) yet promotes apoptosis in HAM (17). If apoptosis were the mechanism responsible for inhibition of mycobacterial growth, this may explain the difference we observed in the ability of these cells to inhibit growth of mycobacteria.
We used a model with M. tuberculosis H37Ra and BCG. Although differences have been demonstrated between the mycobacterial strains H37Ra and H37Rv in mice, the growth rate in human monocyte-derived macrophages are similar (35). We cannot extrapolate our results to H37Rv; however, we have elucidated differences in responses between mice and humans.
In summary, our data directly compare the relative ability of human alveolar macrophages, monocytes, and PMN to inhibit growth of mycobacteria. We demonstrate clear differences between the ability of phagocytic cells to alter the growth of mycobacteria. We demonstrate that HAM, but not other phagocytic cells, are capable of inhibition of growth of both BCG and M. tuberculosis H37Ra early after infection, albeit with donor heterogeneity. Although abnormal phagosomal maturation has been demonstrated in monocytes and murine macrophages, this abnormality has yet to be studied in HAM and may differ (13, 14). In addition, in contrast to murine models, inhibition of mycobacterial growth by HAM in the early period after infection is not enhanced by IFN-γ and is not mediated by the generation of NO. Future studies will define alternate cellular properties that are related to antimycobacterial activity.
The writers thank Dr. Koh Nakata for helpful discussion and for assistance with preparing peripheral blood monocytes.
Supported by Grants RO1 HL51631, RR00096, and HL51494, KO7 HL03050 from the National Institutes of Health.
| 1. | Dunn P. L., North R. J.Virulence ranking of some Mycobacterium tuberculosis and Mycobacterium bovis strains according to their ability to multiply in the lungs, induce lung pathology, and cause mortality in mice. Infect. Immun.63199534283437 |
| 2. | Kaufmann S. H. E.Immunity to intracellular bacteria. Annu. Rev. Immunol.111993129163 |
| 3. | Zhang Y., Broser M., Cohen H., Bodkin M., Law K., Reibman J., Rom W. N.Enhanced interleukin-8 release and gene expression in macrophages after exposure to Mycobacterium tuberculosis and its components. J. Clin. Invest.951995586592 |
| 4. | Law K. F., Jagirdar J., Weiden M. D., Bodkin M., Rom W. N.Tuberculosis in HIV positive patients: cellular response and immune activation in the lung. Am. J. Respir. Crit. Care Med.153199613771384 |
| 5. | Jones G. S., Amirault H. J., Andersen B. R.Killing of Mycobacterium tuberculosis by neutrophils: a nonoxidative process. J. Infect. Dis.1621990700704 |
| 6. | Rook, G. A. W., and S. Rainbow. 1981. An isotope incorporation assay for the antimycobacterial effects of human monocytes. Ann. Inst. Pasteur Immunol. 132D:281–289. |
| 7. | Hirsch C. S., Ellner J. J., Russell D. G., Rich E. A.Complement receptor-mediated uptake and tumor necrosis factor-α-mediated growth inhibition of Mycobacterium tuberculosis by human alveolar macrophages. J. Immunol.1521994743753 |
| 8. | Steele J., Flint K. C., Pozniak A. L., Hudspith B., Johnson M. M., Rook G. A. W.Inhibition of virulent Mycobacterium tuberculosis by murine peritoneal macrophages and human alveolar lavage cells: the effects of lymphokines and recombinant gamma interferon. Tubercle671986289294 |
| 9. | Schlesinger L. S., Bellinger-Kawahara C. G., Payne N. R., Horwitz M. A.Phagocytosis of Mycobacterium tuberculosis is mediated by human monocyte complement receptors and complement component C3. J. Immunol.144199027712780 |
| 10. | Schlesinger L. S., Horwitz M. A.Phagocytosis of Mycobacterium leprae by human monocyte-derived macrophages is mediated by complement receptors CR1 (CD35), CR3 (CD11b/CD18), and CR4 (CD11c/CD18) and IFN-gamma activation inhibits complement receptor function and phagocytosis of this bacterium. J. Immunol.147199119831994 |
| 11. | Downing J. F., Pasula R., Wright J. R., Wigg H. L., Martin W. J.Surfactant protein A promotes attachment of Mycobacterium tuberculosis to alveolar macrophages during infection with human immunodeficiency virus. Proc. Natl. Acad. Sci. U.S.A.92199548484852 |
| 12. | Zimmerli S., Edwards S., Ernst J. D.Selective receptor blockade during phagocytosis does not alter the survival or growth of Mycobacterium tuberculosis in human macrophages. Am. J. Respir. Cell Mol. Biol.151996760770 |
| 13. | Sturgill-Koszycki S., Schlesinger P. H., Chakraborty P., Haddix P. L., Collins H. L., Fok A. K., Allen R. D., Gluck S. L., Heuser J., Russell D. G.Lack of acidification in Mycobacterium phagosomes produced by exclusion of vesicular proton-ATPase. Science2631994678681 |
| 14. | Clemens D. L., Horwitz M. A.Characterization of the Mycobacterium tuberculosis phagosome and evidence that phagosomal maturation is inhibited. J. Exp. Med.1811995257270 |
| 15. | Chan J., Xing Y., Magliozzo R. S., Bloom B. R.Killing of virulent Mycobacterium tuberculosis by reactive nitrogen intermediates produced by activated murine macrophages. J. Exp. Med.175199211111122 |
| 16. | Laochumroonvorapong P., Paul S., Elkon K. B., Kaplan G.H2O2 induces monocyte apoptosis and reduces viability of Mycobacterium avium-M. intracellulare within cultured human monocytes. Infect. Immun.641996452459 |
| 17. | Keane J., Katarzyna M., Balcewicz-Sablinska, Remold H. G., Chupp G. L., Meek B. B., Fenton M. J., Kornfeld H.Infection by Mycobacterium tuberculosis promotes human alveolar macrophage apoptosis. Infect. Immun.651997298304 |
| 18. | Miyikawa Y., Ratnakar P., Rao A. G., Costello M. L., Mathieu-String O.`Costello, R. I. Lehrer, and A. CatanzaroIn vitro activity of the antimicrobial peptides human and rabbit defensins and porcine leukocyte protegrin against Mycobacterium tuberculosis. Infect. Immun.641996926932 |
| 19. | Nicholson S., Bonecini-Almeida M. G., Lapa e Silva J. R., Nathan C., Xie Q. W., Mumford R., Weidner J. R., Calaycay J., Geng J., Boechart N., Linhares C., Rom W. N., Ho J. L.Inducible nitric oxide synthase in pulmonary alveolar macrophages from patients with tuberculosis. J. Exp. Med.183199622932302 |
| 20. | Albina J. E.On the expression of nitric oxide by human macrophages: why no NO? J. Leukoc. Biol.581995643649 |
| 21. | Sierra-Madero J. G., Toossi Z., Hom D. L., Finegan C. K., Hoenig E., Rich E. A.Relationship between load of virus in alveolar macrophages from human immunodeficiency virus type 1-infected persons, production of cytokines, and clinical status. J. Infect. Dis.16919941827 |
| 22. | Flesch I., Kaufmann S. H. E.Mycobacterial growth inhibition by interferon-γ-activated bone marrow macrophages and differential susceptibility among strains of Mycobacterium tuberculosis. J. Immunol.138198744084413 |
| 23. | Stach J. L., Gros P., Forget A., Skamene E.Phenotypic expression of genetically controlled natural resistance to Mycobacterium bovis (BCG). J. Immunol.1321984888892 |
| 24. | Somogyi, P. A., and I. Foldes. 1983. Incorporation of thymine, thymidine, adenine and uracil into nucleic acids of Mycobacterium tuberculosis and its phage (abstract). Ann. Inst. Pasteur Microbiol. 134A:19– 28. |
| 25. | Geller D. A., Lowenstein C. J., Shapiro R. A., Nussler A. K., Di Silvio M., Wang S. C., Nakayama D. K., Simmons R. L., Snyder S. H., Billiar T. R.Molecular cloning and expression of inducible nitric oxide synthase from human hepatocytes. Proc. Natl. Acad. Sci. U.S.A.90199334913495 |
| 26. | Ding A. H, Nathan C. F., Stuehr D. J.Release of reactive nitrogen intermediates and reactive oxygen intermediates from mouse peritoneal macrophages: comparison of activating cytokines and evidence for independent production. J. Immunol.141198824072412 |
| 27. | Thomas E. D., Ramberg R. E., Sale G. E., Sparkes R. S., Golde D. W.Direct evidence for a bone marrow origin of the alveolar macrophage in man. Nature192197610161018 |
| 28. | Dalton D. K., Pitts-Meek S., Keshav S., Figari I. S., Bradley A., Stewart T. A.Multiple defects of immune cell function in mice with disrupted interferon-γ genes. Science259199317391742 |
| 29. | Kamijo R., Le J., Shapiro D., Havell E. A., Huang S., Aguet M., Bosland M., Vilcek J.Mice that lack the interferon-γ receptor have profoundly altered responses to infection with Bacillus Calmette-Guérin and subsequent challenge with lipopolysaccharide. J. Exp. Med.178199314351440 |
| 30. | Mehta P. K., King C. H., White E. H., Murtagh J. J., Quinn F. D.Comparison of in vitro models for the study of Mycobacterium tuberculosis invasion and intracellular replication. Infect. Immun.64199626732679 |
| 31. | Douvas G. S., Berger E. M., Repine J. E., Crowle A. J.Natural mycobacteriostatic activity in human monocyte-derived adherent cells. Am. Rev. Respir. Dis.13419864448 |
| 32. | Denis M.Human neutrophils, activated with cytokines or not, do not kill virulent Mycobacterium tuberculosis. J. Infect. Dis.1631991919920 |
| 33. | Newport M. J., Huxley C. M., Huston S., Hawrylowicz C. M., Oostra B. A., Williamson R., Lewin M.A mutation in the interferon-gamma-receptor gene and susceptibility to mycobacterial infection. N. Engl. J. Med.335199619411949 |
| 34. | Durrbaum-Landmann I., Gercken J., Flad H. D., Ernst M.Effect of in vitro infection of human monocytes with low numbers of Mycobacterium tuberculosis bacteria on monocyte apoptosis. Infect. Immun.64199653845389 |
| 35. | Paul S., Laochumroonvorapong P., Kaplan G.Comparable growth of virulent and avirulent Mycobacterium tuberculosis in human macrophages in vitro. J. Infect. Dis.1741996105112 |
