Alveolar Macrophages from HIV-Infected Subjects are Resistant to Mycobacterium tuberculosis In Vitro

Alveolar Macrophages from HIV-Infected Subjects are Resistant to Mycobacterium tuberculosis In Vitro Richard B. Day, Yana Wang, Kenneth S. Knox, Rajam...
Author: Bryan Johns
2 downloads 2 Views 237KB Size
Alveolar Macrophages from HIV-Infected Subjects are Resistant to Mycobacterium tuberculosis In Vitro Richard B. Day, Yana Wang, Kenneth S. Knox, Rajamouli Pasula, William J. Martin II, and Homer L. Twigg III Division of Pulmonary/Critical Care Medicine, Department of Medicine, Indiana University Medical Center, Indianapolis, Indiana

HIV-infected individuals frequently develop Mycobacterium tuberculosis (MTB) infection. Alveolar macrophages (AM) are the initial host defense against this organism. We measured MTB growth in AM from normal and HIV-infected subjects after in vitro exposure. Intracellular growth of MTB was reduced in AM from HIV-infected subjects compared with normal macrophages. This was confined to subjects with CD4 counts greater than 200/␮l. Growth of avirulent mycobacteria in HIV macrophages was significantly less than virulent MTB. Because avirulent MTB is more sensitive to tumor necrosis factor-␣ (TNF-␣), we examined the relationship between cytokine secretion and mycobacterial growth. Higher AM spontaneous TNF-␣ secretion was associated with reduced MTB growth in normal AM. This relationship was not seen in HIV-infected subjects, suggesting that other factors contributed to mycobacteria resistance. Mycobacteria-induced TNF-␣ secretion was inversely associated with growth in normal AM but not in HIV-infected subjects. Finally, binding and internalization of MTB was augmented in HIV macrophages compared with normal, demonstrating that reduced intracellular MTB growth was not due to impaired phagocytosis. In conclusion, the increased incidence of MTB infection in HIV-infected subjects does not appear to be due to a defect in macrophage innate immunity.

Most individuals infected with HIV will have pulmonary complications during the course of their disease. Especially prominent is the development of infections typically controlled by the cellular immune system, such as pneumonia caused by Mycobacteria tuberculosis (MTB). HIV-infected subjects exposed to MTB are significantly more likely to develop active disease compared with normal subjects (1). In fact, the increase in MTB cases in the United States since the mid-1980s has been attributed in part to the AIDS epidemic (2). In the lung, MTB infection is normally controlled by innate and acquired immunity. First, alveolar macrophages (AM) phagocytize MTB into phagolysosomes where it can be killed by lysosomal enzymes, reactive oxygen species, and reactive nitrogen species (1, 3). Subsequently, AM elicit

(Received in original form February 20, 2003 and in revised form August 1, 2003) Address correspondence to: Homer L. Twigg III, M.D., Indiana University Medical Center, 1481 West 10th St., 111P-IU, Indianapolis, IN 46202. E-mail: [email protected] Abbreviations: alveolar macrophages, AM; bronchoalveolar lavage, BAL; enzyme-linked immunosorbent assay, ELISA; fluorescein isothiocyanate, FITC; monocytes, Mo; Mycobacterium tuberculosis, MTB; peripheral blood mononuclear cells, PBMCs; phosphate-buffered saline, PBS; tumor necrosis factor-␣, TNF-␣. Am. J. Respir. Cell Mol. Biol. Vol. 30, pp. 403–410, 2004 Originally Published in Press as DOI: 10.1165/rcmb.2003-0059OC on September 11, 2003 Internet address: www.atsjournals.org

a specific T cell immune response in which AM play a key role as accessory cells (1). The resulting T cell response leads to the production of macrophage-activating cytokines, which enhance intracellular killing of MTB (4, 5). The increased incidence of active MTB disease in HIV-infected subjects could be due to defects at either of these two stages. Phagocytosis of MTB is a complicated process that begins with the binding and ingestion of organisms by macrophages. This is followed by the formation of intracellular phagosomes, fusion of phagosomes with lysosomes, and ultimately killing of ingested organisms via several potential mechanisms (i.e., oxidative response). Failure to kill intracellular mycobacteria leads to the formation of a latent state characterized by persistence of viable organisms within macrophages. The increased incidence of mycobacterial disease in HIV infection could be due to impaired macrophage function at any step in the phagocytic process. Some investigators have shown that AM from HIV-infected subjects are defective in their ability to phagocytize yeast (6). However, impaired AM phagocytosis is not a uniform finding in HIVinfected subjects (7, 8). Furthermore, HIV infection is associated with generalized macrophage activation in the lung, a state wherein macrophage phagocytic function might be expected to be enhanced. In this regard, work in our lab (9–11) and others (12, 13) has shown that both macrophage cytokine secretion and secretion of macrophage-activating cytokines by lung lymphocytes is enhanced in HIV infection. Thus the ability of HIV AM to phagocytize and kill MTB is unknown. In this work we compared intracellular MTB growth in AM from normal volunteers and HIV-infected subjects after MTB infection in vitro, and examined the ability of AM to ingest the organism. Given that macrophage-activating cytokines such as tumor necrosis factor-␣ (TNF-␣) are important for intracellular killing of MTB (14), we also measured spontaneous and mycobacterial-induced TNF-␣ secretion by AM. Our data demonstrate that AM from HIV-infected patients with early stage disease are intrinsically resistant to intracellular MTB growth and that this resistance is unrelated to TNF-␣ secretion. Furthermore, this decrease in intracellular MTB growth occurs despite an augmented ability to ingest the organism, suggesting that killing of intracellular MTB is not impaired in HIV infection.

Materials and Methods Subjects Twenty HIV-positive male patients (mean age 38.1 ⫾ 6.6 yr) served as the study population. Five were nonsmokers and fifteen were

404

AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL. 30 2004

current or ex-smokers. All HIV-positive individuals had no current pulmonary symptoms and underwent bronchoscopy only for the purposes of this study. Chest radiographs were obtained on the day of bronchoscopy if one had not been performed in the previous 6 mo and were normal in all cases. The mean CD4 lymphocyte count of the population was 393.5 ⫾ 249.7 cells/␮l (range 2–850 cells/␮l). HIV-positive patients were further divided into two groups based on peripheral blood CD4 counts: early-stage disease, ⭓ 200/␮l (n ⫽ 15); and late-stage disease, ⬍ 200 ␮l (n ⫽ 5). Fifteen out of 20 HIV-infected subjects were on Highly Active Antiretroviral Therapy (HAART). Twelve normal adults (mean age 26.8 ⫾ 3.2 yr), five of whom were smokers, served as the control group. All normal subjects were healthy, free of pulmonary symptoms, and had not had a respiratory tract infection in the past 4 wk. This work was reviewed and approved by the institutional review board at Indiana University. All subjects signed an informed consent statement before participation in this study.

Bronchoalveolar Lavage After the upper airways were anesthetized with 2% topical lidocaine, bronchoalveolar lavage (BAL) was performed through a fiber-optic bronchoscope wedged in subsegmental bronchi in the right middle and right lower lobes. Room temperature normal saline was instilled in 50-ml aliquots into two separate bronchi. Typically, 250–300 ml were instilled to obtain a return of 125–200 ml. Recovered lavage fluid was kept on ice until processed. HIVpositive patients yielded a mean of 21.0 ⫾ 17.1 ⫻ 106 bronchoalveolar cells, of which 16 ⫾ 3.6% were lymphocytes by morphologic criteria. Normal volunteers yielded a mean of 15.3 ⫾ 12.6 ⫻ 106 cells, of which 20.7 ⫾ 3.8% were lymphocytes.

MTB Isolation H37Rv and H37Ra strains of MTB (American Type Culture Collection, Rockville, MD) were cultured at 37⬚C in a 5% CO2 incubator in dispersed form in Middlebrook 7H9 broth (Difco Laboratories, Detroit, MI) containing albumin, dextrose, and catalase as enrichments. Mycobacterial cultures harvested during log growth rate (10–14 d) were centrifuged, washed once with saline, and adjusted to a concentration of 10 ⫻ 106 organisms/ml. Bacterial concentration was determined using a Spectronic 20D spectrophotometer (Milton Roy Co., Rochester, NY). To achieve a single cell suspension, the mycobacterial suspension was briefly sonicated at low intensity (20 W for 5–10 s). Then the suspension was gently agitated and allowed to stand for 5 min. The top portion of the suspension containing bacilli was used in the assay. Each batch of the bacterial suspension was stained with Kinyoun stain (Midatlantic Biomedical Inc., Paulsboro, NJ) and observed under a microscope to check the purity of the suspension. Routinely, samples of bacteria were also grown on 7H11 Middlebrook medium (Difco Laboratories) and maintained as stock.

Preparation of Lung and Peripheral Blood Mononuclear Cells Lavage fluid was filtered through a 100-␮m nylon cell strainer (Becton Dickinson, Franklin Lakes, NJ) to remove debris and centrifuged at 400 ⫻ g for 10 min. The cell pellet was washed twice in Hanks’ balanced salt solution and resuspended in RPMI 1640 medium with 25 mM HEPES (GIBCO, Grand Island, NY) supplemented with 5% heat-inactivated fetal bovine serum (GIBCO). Cell viability was determined by 1% trypan blue exclusion and was routinely ⬎ 95%. Immediately after bronchoscopy, peripheral

blood was collected in a heparinized syringe for isolation of peripheral blood monocytes (Mo). Peripheral blood mononuclear cells (PBMCs) were separated by centrifugation through a FicollHypaque gradient. PBMCs were washed and resuspended in the same manner as the bronchoalveolar cells.

Measurement of Intracellular MTB Growth AM and monocytes were plated on 96-well microtiter plates at a concentration of 100,000 cells/well and allowed to adhere overnight. Prior studies from our lab have shown that there is no difference in the ability of AM from normal volunteers and HIVinfected subjects to adhere to plastic (15). Nonadherent cells were removed and the resulting monolayers were incubated with either virulent (HR37Rv) or avirulent (HR37Ra) MTB at 10:1 and 1:1 organism:cell ratios. Cell monolayers were incubated with mycobacteria for 2 h, washed with phosphate-buffered saline (PBS), and cultured for 2 d to allow time for establishment of infection. After 2 d of culture, intracellular mycobacteria growth was measured using the standard radiometric (BACTEC) method (16). Briefly, cells were lysed in 0.25% SDS lysis buffer for 15 min, neutralized with 20% BSA, and the entire cell lysate was inoculated into BACTEC 12B vials (Becton Dickinson, San Jose, CA) containing 7H12 Middlebrook broth, an antimicrobial PANTA supplement used to prevent growth of any contaminating organisms, and C14 labeled palmitic acid. Each vial was filled with CO2 and incubated at 37⬚C. Mycobacterial growth was measured by the ability of viable organisms to metabolize radiolabeled palmitic acid and release radioactivity into the BACTEC bottle head space, which is analyzed by the BACTEC instrument every 48 h for a total of 31 d or until maximum growth has been reached. Data are expressed as a growth index ranging from 0–999, with an index of ⬎ 10 indicating growth (manufacturer’s instructions, Becton Dickinson, Sparks, MD). Time to maximal growth was the number of days from initial inoculation of the sample into BACTEC bottles to a growth index of 999. Time to initial growth was the number of days from initial inoculation to the time two successive measurements showed a growth index of greater than 10.

Measurement of TNF-␣ by Enzyme-Linked Immunosorbent Assay AM and monocytes were plated on 96-well microtiter plate at concentration of 100,000 cells/well and allowed to adhere overnight. Nonadherent cells were removed and the resulting monolayers were incubated with either virulent or avirulent MTB at a 10:1 organism:cell ratio. AM and monocytes were incubated with MTB for 2 h, washed, and cultured for 2 d. Supernatants were collected and frozen at ⫺70⬚C until further use. TNF-␣ secretion was measured using a commercially available enzyme-linked immunosorbent assay (ELISA; Coulter-Immunotech, Miami, FL) with a sensitivity of 10 pg/ml.

Attachment and Phagocytosis Assay MTB were labeled with fluorescein isothiocyanate (FITC) according to the method of Ezekowitz and coworkers (17). MTB were incubated in 0.1 mg/ml FITC in PBS at a concentration of 107 MTB/ml for 30 min at 37⬚C. The MTB suspension was centrifuged and washed twice with PBS at room temperature to sufficiently remove any free or unincorporated FITC label. AM and monocytes were plated on 96-well microtiter plates at concentration of 250,000 cells/well and allowed to adhere overnight. Nonadherent cells were removed and FITC-labeled MTB organisms were

Day, Wang, Knox, et al.: Macrophage Resistance to Tuberculosis in HIV Infection

added at a 10:1 organism/cell ratio and incubated with AM and monocytes at 37⬚C for 1 and 2 h. After incubation, nonadherent mycobacteria were removed by washing with PBS. The number of attached and ingested MTB was measured in a cytoflour machine (Millipore Model #2350; Bedford, MA). 0.1% trypan blue in PBS was then added to quench the surface-bound FITC-labeled MTB signal and measurements repeated to determine internalized organisms (18). The percentage of bound MTB that was internalized was determined by dividing the fluorescence after quenching with trypan blue by total fluorescence before quenching.

Statistics Comparisons between group means was performed using the Mann-Whitney Rank Sum test for nonparametric data. Paired comparisons between virulent and avirluent strains and AM versus Mo within individual subjects was performed using the Wilcoxon paired-sample test. Correlation between TNF-␣ secretion and MTB growth was analyzed using Spearman’s Rank Order correlation coefficient. P values ⭐ 0.05 were considered significant.

Results Measurement of Intracellular MTB Growth in AM and Mo To examine intracellular MTB growth in AM and Mo, cells were isolated from eighteen HIV-positive and eight normal volunteers. Cells were infected with virulent and avirulent MTB in vitro and cultured for 2 d. After 2 d, cells were lysed and the cell lysate containing viable mycobacteria was inoculated into BACTEC bottles. Time to initial evidence of growth and maximum growth was measured. Thus this assay measures the amount of intracellular growth that occurred in Mo and AM during the 2 d of culture, with greater growth during this period resulting in a larger inoculum into BACTEC bottles. Intracellular growth of virulent and avirulent MTB in AM from HIV-infected subjects was significantly inhibited compared with normal AM. (Figure 1). There was a dose-dependent response for both MTB strains, with time to maximum growth longer when cells were infected at a 1:1 organism:cell ratio compared with a 10:1 ratio. When time to initial growth was examined, there was

405

a delay in growth in HIV AM compared with normal AM for the avirulent MTB strain only (HIV AM versus normal AM: 11.6 ⫾ 1.6 d versus 6.6 ⫾ 1.2 d at 10:1 infection ratio, P ⫽ 0.037; 16.0 ⫾ 2.0 d versus 11.6 ⫾ 2.8 d at 1:1 infection ratio, P ⫽ 0.15). There was no difference in MTB growth between AM and Mo in either HIV-infected subjects or normal volunteers (data not shown). Because MTB infection typically affects HIV-infected subjects later in the course of their disease, we next examined whether intracellular MTB growth was greater in phagocytes from HIV-infected individuals with lower CD4 counts (below 200/␮l) compared with subjects with CD4 counts above 200/␮l. AM from HIV-infected subjects with CD4 counts above 200/␮l supported less intracellular MTB growth than normal AM (Figures 2A and 2B). AM from HIV-infected subjects with CD4 counts below 200/␮l did not significantly differ from normal AM (Figures 2C and 2D). Thus, although MTB grows poorly in AM from HIVinfected subjects as a group, this appears to be confined to individuals with earlier stage HIV infection. In HIV-infected subjects, avirulent MTB consistently grew more slowly than virulent MTB in mononuclear phagocytes. Data for initial and maximum growth in AM are shown in Figure 3. This difference in growth rate between avirulent and virulent strains was not observed in normal AM and Mo (data not shown). TNF-␣ Secretion Avirulent MTB is known to be more susceptible to the effects of TNF-␣ on human AM than virulent MTB (19). This suggests that TNF-␣ secretion by activated HIVinfected AM could be influencing MTB growth in AM. To explore the relationship between levels of TNF-␣ secretion and susceptibility to MTB infection, we measured spontaneous TNF-␣ secretion by AM and correlated it with intracellular MTB growth. There was no difference in spontaneous TNF-␣ secretion between AM from normal volunteers and HIV-infected subjects (data not shown). However, in normal subjects greater spontaneous TNF-␣ secretion was

Figure 1. Intracellular growth of virulent (A ) and avirulent (B ) MTB in AM. AM (105) were isolated from HIV⫹ (filled circles) or normal (open circles) volunteers and exposed to MTB at 10:1 or 1:1 mycobacteria/cell ratios. MTB growth was measured via the BACTEC method and days to maximal growth was determined. Intracellular growth of virulent and avirulent MTB was significantly impaired in AM from HIV-infected subjects compared with normal AM. HIV, n ⫽ 18; normal, n ⫽ 9.

406

AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL. 30 2004

Figure 2. Intracellular growth of virulent and avirulent MTB in AM from HIV-infected subjects (filled circles) with CD4 counts ⭓ 200/␮l (A and B ) or ⬍ 200/␮l (C and D) and from normal volunteers (open circles). AM (105) were exposed to MTB at 10:1 or 1:1 mycobacteria/ cell ratios. MTB growth was measured via the BACTEC method and days to maximal growth was determined. Intracellular MTB growth was significantly impaired in AM from HIV-infected subjects with CD4 counts ⭓ 200/␮l compared with normal AM. Intracellular MTB growth was identical in normal volunteers and HIV-infected subjects with CD4 count ⬍ 200/␮l. HIV CD4 ⭓ 200/␮l, n ⫽ 13; HIV CD4 ⬍ 200/␮l, n ⫽ 5; normal, n ⫽ 9.

associated with slowed intracellular MTB growth by both virulent and avirulent MTB (Figure 4A). No such correlation was seen in HIV-infected subjects (Figure 4B). We next examined whether there were differences in AM TNF-␣ secretion after exposure to virulent and avirulent MTB and whether this correlated with MTB growth. As with spontaneous TNF-␣ secretion, there was no difference in MTB-induced macrophage TNF-␣ secretion between the HIV-positive and normal group or between virulent and avirulent strains (data not shown). Interestingly, in normal AM, increased MTB-induced TNF-␣ secre-

tion was associated with increased intracellular growth of both virulent and avirulent MTB (Figure 5A). Again, no such correlation was seen in HIV-infected subjects (Figure 5B). Attachment and Phagocytosis The initial step in MTB infection is uptake of the mycobacterium by AM. To explore whether differences in intracellular MTB growth in AM was mediated by differences in phagocytosis, assays were performed to determine the ability of normal and HIV-infected AM to bind and internalize MTB. HIV-infected AM bound FITC-labeled

Figure 3. Intracellular growth of virulent and avirulent MTB in HIV AM. HIV AM (105) were exposed to virulent (filled bars) or avirulent (open bars) MTB at 10:1 or 1:1 mycobacteria/cell ratios for 2 h, and subsequent time to maximal growth was determined. Growth of avirulent MTB in HIV AM was significantly slower than virulent MTB at all MTB:cell ratios tested (n ⫽ 18).

Day, Wang, Knox, et al.: Macrophage Resistance to Tuberculosis in HIV Infection

407

Figure 4. Correlation between spontaneous TNF-␣ secretion by normal (A ) and HIV-infected subjects (B ) and days to maximal growth with virulent (black circles) and avirulent (shaded circles) MTB. AM (105) were cultured in 5% fetal bovine serum/RPMI for 2 d, and TNF-␣ concentrations were determined in supernatants by ELISA. In normal subjects (A ), greater spontaneous secretion was associated with impaired intracellular growth of both virulent and avirulent MTB (P ⫽ 0.297 for both virulent and avirulent MTB). No correlation between spontaneous TNF-␣ secretion and MTB growth was observed in HIV-infected subjects. Normal, n ⫽ 7; HIV, n ⫽ 16.

virulent and avirulent MTB more avidly than normal AM (Figures 6A and 6B). The percentage of bound MTB subsequently internalized was similar in the two groups (percent virulent MTB internalized at 2 h: normal 33 ⫾ 9%, HIV 56 ⫾ 22%; % avirulent MTB internalized at 2 h: normal 33 ⫾ 11%, HIV 41 ⫾ 10%). Given the increased binding of MTB to the surface of HIV AM, this resulted in a greater internalized MTB burden in HIV AM (Figures 6C and 6D). Thus MTB growth in HIV AM is decreased despite an even greater mycobacterial burden in these cells due to augmented phagocytosis.

Discussion In this study we have demonstrated that (i ) intracellular growth of MTB is decreased in alveolar macrophages from HIV-infected subjects; (ii) this growth inhibition is confined to HIV-infected subjects with CD4 counts above 200/␮l; (iii) intracellular growth of avirulent MTB in AM is lower than growth with virulent MTB; (iv) in normal subjects greater spontaneous TNF-␣ secretion is associated with reduced intracellular MTB growth, whereas greater MTB-

induced TNF-␣ secretion is associated with enhanced growth; (v ) no correlation between HIV AM TNF-␣ secretion and MTB growth is seen; and (vi) reduced intracellular MTB growth in HIV AM occurs despite an enhanced ability of HIV AM to bind and internalize MTB. Thus AM from HIV-infected subjects appear to be resistant to intracellular MTB growth compared with normal AM despite a significantly increased intracellular mycobacterial burden. These in vitro results suggest that the increased incidence of tuberculosis in HIV-infected subjects is not due to an intrinsic defect in macrophage innate immunity. Host defense against MTB requires intact innate and acquired immune responses. Initially the organism is phagocytized by mononuclear phagocytes. If the mycobacterial burden is large and not controlled by phagocytosis alone, a cellular immune response is generated, resulting in the proliferation of TH1 lymphocytes which secrete macrophage activating cytokines. Although prior studies have defined the role of AM in MTB infection, it is unclear how this role is altered by prior infection with HIV. However, because it is well established that mononuclear phagocytes,

Figure 5. Correlation between virulent (black circles) and avirulent (shaded circles) MTBinduced TNF-␣ secretion in normal (A ) and HIV-infected subjects (B ) and days to maximal growth. AM were exposed to MTB, cultured for 2 d, and TNF-␣ concentrations were determined in culture supernatants by ELISA. In normal subjects, greater MTB-induced AM TNF-␣ secretion was associated with increased intracellular MTB growth with both virulent and avirulent MTB (P ⫽ 0.139 for both virulent and avirulent MTB). No such correlation was seen in HIV infected subjects. Normal, n ⫽ 8; HIV, n ⫽ 16.

408

AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL. 30 2004

Figure 6. Attachment (A and B ) and phagocytois (C and D ) of MTB to normal (gray lines) and HIV-infected (black lines) AM. Adherent AM (2.5 ⫻ 105) were exposed to FITC-labeled MTB for 1 or 2 h, washed, and the amount of total MTB attached was determined by measuring fluorescence. Internalized MTB was determined by measuring fluorescence after quenching surface bound organisms with 0.1% trypan blue. HIV-infected AM bind greater amounts of virulent (A ) and avirulent (B ) MTB than AM from normal subjects. Because the percentage of bound organisms that is internalized is similar in normal and HIV AM, this results in a greater amount of phagocytized organisms in HIV AM exposed to both virulent (C ) and avirulent (D ) strains. *P ⬍ 0.05. Normal, n ⫽ 4; HIV, n ⫽ 3.

including AM, from HIV-infected subjects are chronically activated (13), it is reasonable to speculate that these cells may be primed to efficiently phagocytize invading organisms. Because AM are responsible for the initial response to infection in the lung, we chose to focus our study on the ability of this cell to bind and phagocytize mycobacteria, the role TNF-␣ has in host defense against MTB in HIV infection, and ultimately the intrinsic ability of AM from HIV-infected subjects to resist MTB infection. The initial step in phagocytosis is binding and uptake of the organism. Some investigators have suggested that phagocytosis is impaired in HIV infection (6). However, this is not a universal finding (7, 8). In studies that have examined binding and ingestion directly, this function appears to be preserved if not enhanced in HIV-infected phagocytes. Ieong and colleagues have shown that AM infected with HIV in vitro have decreased fungicidal activity but no impairment in the ability to bind and internalize yeast (21). Expression of other surface receptors known to be important in the binding and uptake of MTB, including mannose receptors, complement receptors, and Fc receptors, is also altered in HIV infection. For example, although mannose receptors are downregulated on AM in HIV-infected subjects (22), the expression of complement and Fc receptors is increased on some phagocyte populations (23, 24). Furthermore, we have previously shown that BAL fluid from HIV-infected subjects promotes MTB binding to normal AM due to increased concentrations of surfactant protein A in HIV BAL (25). Thus it is not surprising that we found an enhanced ability of AM from HIV-positive subjects to bind and ingest MTB. Once internalized, MTB is either killed or establishes a latent infection. In many instances innate immunity alone is sufficient to eradicate MTB (26, 27). Our data demonstrate

that, despite a higher initial intracellular mycobacterial load, after 2 d of culture the number of viable intracellular organisms in HIV AM is decreased compared with normal AM. Our data do not allow us to differentiate between whether MTB was killed more efficiently or grew more slowly in HIV AM. However, examination of the different growth characteristics and dose responses of virulent and avirulent MTB allows us to speculate on the mechanisms behind impaired MTB growth in HIV AM. Germane to this discussion is the extensive body of literature demonstrating that MTB, primarily avirulent strains, induce apoptosis in macrophages that have ingested the organism (19, 26, 28, 29), which has been postulated to play an important role in killing of intracellular MTB. In this regard, MTB growth inhibition in HIV AM is most pronounced for the avirulent strain and is most significant when the mycobacterial inoculum is large (i.e., 10:1 MTB:AM ratio), when the apoptosis-inducing effect may be most pronounced. In contrast, macrophage apoptosis is less important in controlling virulent MTB. In this instance, prior macrophage activation with upregulation of lysosomal enzymes, reactive oxygen species, or reactive nitrogen species may be more important in host defense. Thus, control of MTB would be less dependent on the size of the mycobacterial load. In fact, the larger the mycobacterial burden, the more likely you are to overwhelm AM defenses. Thus control of virulent MTB is better at lower mycobacterial inoculums (i.e., 1:1 MTB:AM ratio). MTB also induces the production of cytokines important in host defense. In particular, the role of TNF-␣ in host defense against virulent and avirulent MTB is complicated. Although most investigators have found that virulent and avirulent strains induce similar amounts of TNF-␣ secretion from AM (20, 28, 30), this is not a universal finding (19).

Day, Wang, Knox, et al.: Macrophage Resistance to Tuberculosis in HIV Infection

However, growth of avirulent MTB in macrophages appears to be inhibited by TNF-␣ (19, 31), whereas growth of virulent strains is less affected by this cytokine (19, 20). Because avirulent strains induce more apoptosis in AM than virulent strains, this raises the possibility that TNF-␣ is directly initiating apoptotic pathways in MTB-infected cells (32). In fact, apoptosis induced by avirulent MTB appears to be dependent on TNF-␣ secretion (20, 28). In contrast, virulent MTB strains appear to avoid the toxic effects of TNF-␣ through the secretion of soluble TNF receptors (20) and may even derive a growth benefit from this cytokine (19). Reduced intracellular MTB growth was associated with spontaneous TNF-␣ secretion in normal AM, though this did not reach statistical significance in this small study. This is consistent with work by others demonstrating that TNF-␣ primes macrophages to kill MTB (1, 3, 14). Failure to observe a similar relationship in HIV-infected AM could be due to several factors. TNF-␣ secretion by AM may already be above the threshold needed for optimum activation and further secretion above this level may not be of any benefit. Alternatively, other factors (i.e., interferon-␥) may be required for efficient AM activation (33). This possibility is intriguing given prior work in our lab demonstrating that interferon-␥ production decreases in late-stage HIV infection (11) and our current observation that AM from HIV subjects with late-stage disease are less resistant to infection with MTB. The positive relationship between growth of MTB in normal AM and MTB-induced TNF-␣ secretion adds further support to the importance of this cytokine in normal host defense against tuberculosis and is consistent with the fact that MTB can directly stimulate TNF-␣ secretion in infected cells. The lack of any correlation between MTB-induced TNF-␣ secretion and MTB growth in HIVinfected subjects may reflect the fact that HIV AM TNF-␣ secretion is being regulated by other more prominent mediators. Alternatively, as mentioned above, cosecretion of soluble TNF receptors by AM in response to infection with MTB may be negating the protective effect TNF-␣ has against MTB infection (20). We did not measure intracellular TNF-␣ in our experiments, which could be contributing to our data failing to reach statistical significance. Engele and colleagues demonstrated that up to 20% of MTBinfected AM stained positive for intracellular TNF-␣. Interestingly, however, when co-stained for intracellular MTB, these investigators demonstrated that relatively few MTB infected AM (15%) also contained intracellular TNF-␣, leading them to conclude that a significant portion of TNF-␣ secretion in response to MTB infection is coming from uninfected cells. The ultimate goal of effective phagocytosis is elimination of the offending organism. Not only was time to maximal growth delayed in AM from HIV-infected patients, but at lower inoculums (1:1 organism to cell ratio), more AM from HIV-positive subjects had no evidence of infection at 30 d after exposure to MTB compared with normal AM (3 out of 18 versus 1 out of 9 for virulent MTB, 7 out of 18 versus 1 out of 9 for avirulent MTB). This was confined primarily to individuals with CD4 counts ⬎ 200/␮l. Furthermore, this was more likely to occur with avirulent than virulent MTB strains, consistent with the increased sensitivity of avirulent

409

strains to phagocytosis. Thus it appears that the increased incidence of MTB infection in the HIV population cannot be readily attributed to impaired intrinsic macrophage phagocytic function. Rather, our data suggest that AM from HIV-infected subjects are able to control MTB infection until late-stage HIV infection. We speculate the increased susceptibility in late-stage HIV infection reflects a decline of macrophage-activating cytokines as the TH1 immune response is progressively impaired. It is important to bear in mind that our in vitro experiments may not reflect events in vivo, where macrophage function is occurring in the presence of other potential immunologic modifiers, including surfactant, which has been shown to play an important role in host defense against MTB (25). The coexistence of HIV and MTB remains an important healthcare concern. Besides the recognized increase in MTB infection in HIV-infected subjects (1, 2), there is now mounting evidence that MTB itself can upregulate HIV infection, both directly and through increased expression of chemokine receptors known to be important in HIV infection (34–36). In contrast, very little has been done examining the reciprocal situation, namely direct alterations in MTB growth induced by replicating HIV. In one report, co-infecting normal macrophages with MTB and HIV did not alter intracellular growth of MTB (37), suggesting that HIV itself does not affect MTB replication, but clearly more work needs to be done in this area. Regardless, efforts to control both HIV and MTB infection remain important in these patients. In this regard, recent studies suggest that highly active antiretroviral therapy can restore immune responses to MTB (38). In summary, our study demonstrates that intracellular MTB growth is significantly impaired in AM from HIVinfected individuals in early-stage disease. This reduced intracellular growth is present despite an increased mycobacterial burden in HIV AM by virtue of enhanced ingestion of mycobacterial organisms. Reduced intracellular MTB growth is not correlated with TNF-␣ production. In late-stage HIV infection, AM support greater intracellular MTB growth, which is not any different from normal AM. These results suggest that the increased susceptibility of MTB in HIV infection is not due to impaired macrophage innate immunity. Rather, we speculate that the progressive impairment in T cell immunity that occurs with HIV disease progression is the main contributor to MTB infection in these individuals. Acknowledgments: This work was supported by grants HL-59834 (H.L.T.) and HL 61285 (W.J.M.) from the National Heart, Lung and Blood Institute, MO1 RR750 (General Clinical Research Center award), and AI 25859 (Acquired Immune Deficiency Syndrome Program of the National Institute of Allergy and Infectious Disease).

References 1. Barnes, P. F., and R. L. Modlin. 1996. Human cellular immune responses to Mycobacterium tuberculosis. Curr. Top. Microbiol. Immunol. 215: 197–219. 2. Snider, D. E., Jr., and W. L. Roper. 1992. The new tuberculosis. N. Engl. J. Med. 326:703–705. 3. Schlesinger, L. S. 1996. Role of mononuclear phagocytes in M. tuberculosis pathogenesis. J. Investig. Med. 44:312–323. 4. Barnes, P. F., J. S. Abrams, S. Lu, P. A. Sieling, T. H. Rea, and R. L. Modlin. 1993. Patterns of cytokine production by mycobacterium-reactive human T-cell clones. Infect. Immun. 61:197–203.

410

AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL. 30 2004

5. Zhang, M., Y. Lin, D. V. Iyer, J. Gong, J. S. Abrams, and P. F. Barnes. 1995. T-cell cytokine responses in human infection with Mycobacterium tuberculosis. Infect. Immun. 63:3231–3234. 6. da Bonecini-Almeida, M., E. Werneck-Barroso, P. B. Carvalho, C. P. de Moura, E. F. Andrade, A. Hafner, C. E. Carvalho, J. L. Ho, A. L. Kritski, and M. G. Morgado. 1998. Functional activity of alveolar and peripheral cells in patients with human acquired immunodeficiency syndrome and pulmonary tuberculosis. Cell. Immunol. 190:112–120. 7. Gordon, S. B., M. E. Molyneux, M. J. Boeree, S. Kanyanda, M. Chaponda, S. B. Squire, and R. C. Read. 2001. Opsonic phagocytosis of Streptococcus pneumoniae by alveolar macrophages is not impaired in human immunodeficiency virus-infected Malawian adults. J. Infect. Dis. 184: 1345–1349. 8. Cameron, M. L., D. L. Granger, T. J. Matthews, and J. B. Weinberg. 1994. Human immunodeficiency virus (HIV)-infected human blood monocytes and peritoneal macrophages have reduced anticryptococcal activity whereas HIV-infected alveolar macrophages retain normal activity. J. Infect. Dis. 170:60–67. 9. Twigg, H. L., G. K. Iwamoto, and D. M. Soliman. 1992. Role of cytokines in alveolar macrophage accessory cell function in HIV-infected individuals. J. Immunol. 149:1462–1469. 10. Spain, B. A., D. M. Soliman, R. A. Sidner, and H. L. Twigg. 1995. Enhanced proliferation and IL-2 secretion by lung lymphocytes from HIV-infected subjects. Am. J. Physiol. 269:L498–L506. 11. Twigg, H. L., B. A. Spain, D. M. Soliman, K. Knox, R. A. Sidner, C. Schnizlein-Bick, D. S. Wilkes, and G. K. Iwamoto. 1999. Production of interferon-gamma by lung lymphocytes in HIV-infected individuals. Am. J. Physiol. 276:L256–L262. 12. Sierra-Madero, J. G., Z. Toossi, D. L. Hom, C. K. Finegan, E. Hoenig, and E. A. Rich. 1994. 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. 169:18–27. 13. Buhl, R., H. A. Jaffe, K. J. Holroyd, Z. Borok, J. H. Roum, A. Mastrangeli, F. B. Wells, M. Kirby, C. Saltini, and R. G. Crystal. 1993. Activation of alveolar macrophages in asymptomatic HIV-infected individuals. J. Immunol. 150:1019–1028. 14. Fenton, M. J., and M. W. Vermeulen. 1996. Immunopathology of tuberculosis: roles of macrophages and monocytes. Infect. Immun. 64:683–690. 15. Twigg, H. L., and D. M. Soliman. 1994. Role of alveolar macrophage-T cell adherence in accessory cell function in human immunodeficiency virus–infected individuals. Am. J. Respir. Cell Mol. Biol. 11:138–146. 16. Reddy, M. V., S. Srinivasan, B. Andersen, and P. R. Gangadharam. 1994. Rapid assessment of mycobacterial growth inside macrophages and mice, using the radiometric (BACTEC) method. Tuber. Lung Dis. 75:127–131. 17. Ezekowitz, R. A., D. J. Williams, H. Koziel, M. Y. Armstrong, A. Warner, F. F. Richards, and R. M. Rose. 1991. Uptake of Pneumocystis carinii mediated by the macrophage mannose receptor. Nature 351:155–158. 18. Mariencheck, W. I., J. Savov, Q. Dong, M. J. Tino, and J. R. Wright. 1999. Surfactant protein A enhances alveolar macrophage phagocytosis of a live, mucoid strain of P. aeruginosa. Am. J. Physiol. 277:L777–L786. 19. Engele, M., E. Stobel, K. Castiglione, N. Schwerdtner, M. Wagner, P. Bolcskei, M. Rollinghoff, and S. Stenger. 2002. Induction of TNF in human alveolar macrophages as a potential evasion mechanism of virulent Mycobacterium tuberculosis. J. Immunol. 168:1328–1337. 20. Balcewicz-Sablinska, M. K., J. Keane, H. Kornfeld, and H. G. Remold. 1998. Pathogenic Mycobacterium tuberculosis evades apoptosis of host macrophages by release of TNF-R2, resulting in inactivation of TNFalpha. J. Immunol. 161:2636–2641. 21. Ieong, M. H., C. C. Reardon, S. M. Levitz, and H. Kornfeld. 2000. Human immunodeficiency virus type 1 infection of alveolar macrophages impairs their innate fungicidal activity. Am. J. Respir. Crit. Care Med. 162:966–970.

22. Koziel, H., Q. Eichbaum, B. A. Kruskal, P. Pinkston, R. A. Rogers, M. Y. Armstrong, F. F. Richards, R. M. Rose, and R. A. Ezekowitz. 1998. Reduced binding and phagocytosis of Pneumocystis carinii by alveolar macrophages from persons infected with HIV-1 correlates with mannose receptor downregulation. J. Clin. Invest. 102:1332–1344. 23. Capsoni, F., F. Minonzio, A. M. Ongari, G. Colombo, G. P. Rizzardi, P. Bonara, A. D’Arminio-Monforte, and C. Zanussi. 1992. Increased expression of IcG Fc receptor type I on neutrophils and monocytes from HIVinfected subjects. Clin. Exp. Immunol. 90:175–180. 24. Palmer, S., and A. S. Hamblin. 1993. Increased CD11/CD18 expression on the peripheral blood leucocytes of patients with HIV disease: relationship to disease severity. Clin. Exp. Immunol. 93:344–349. 25. Downing, J. F., R. Pasula, J. R. Wright, H. L. Twigg, III, and W. J. Martin, II. 1995. Surfactant protein A promotes attachment of Mycobacterium tuberculosis to alveolar macropahges during infection with human immunodeficiency virus. Proc. Natl. Acad. Sci. USA 92:4848–4852. 26. Klingler, K., K.-M. Tchou-Wong, O. Brandli, C. Aston, R. Kim, C. Chi, and W. N. Rom. 1997. Effects of mycobacteria on regulation of apoptosis in mononuclear phagocytes. Infect. Immun. 65:5272–5278. 27. Schluger, N. W., and W. N. Rom. 1998. The host immune response to tuberculosis. Am. J. Respir. Crit. Care Med. 157:679–691. 28. Keane, J., M. K. Balcewicz-Sablinska, H. G. Remold, G. L. Chupp, B. B. Meek, M. J. Fenton, and H. Kornfeld. 1997. Infection by Mycobacterium tuberculosis promotes human alveolar macrophage apoptosis. Infect. Immun. 65:298–304. 29. Keane, J., H. G. Remold, and H. Kornfeld. 2000. Virulent Mycobacterium tuberculosis strains evade apoptosis of infected alveolar macrophages. J. Immunol. 164:2016–2020. 30. Saukkonen, J. J., B. Bazydlo, M. Thomas, R. M. Strieter, J. Keane, and H. Kornfeld. 2002. ␤-chemokines are induced by Mycobacterium tuberculosis and inhibit its growth. Infect. Immun. 70:1684–1693. 31. Hirsch, C. S., J. J. Ellner, D. G. Russell, and E. A. Rich. 1994. Complement receptor-mediated uptake and tumor necrosis factor-␣-mediated growth inhibition of Mycobacterium tuberculosis by human alveolar macrophages. J. Immunol. 152:743–753. 32. Zheng-gang, L., H. Hsu, D. V. Goeddel, and M. Karin. 1996. Dissection of TNF receptor 1 effector functions: JNK activation is not linked to apoptosis while NF-␬␤ activation prevents cell death. Cell 67:565–576. 33. Fenton, M. J., M. W. Vermeulen, S. Kim, M. Burdick, R. M. Strieter, and H. Kornfeld. 1997. Induction of gamma interferon production in human alveolar macrophages by Mycobacterium tuberculosis. Infect. Immun. 65:5149–5156. 34. Nakata, K., W. N. Rom, Y. Honda, R. Condos, S. Kanegasaki, Y. Cao, and M. Weiden. 1997. Mycobacterium tuberculosis enhances human immunodeficiency virus-1 replication in the lung. Am. J. Respir. Crit. Care Med. 155:996–1003. 35. Toosi, Z., K. Nicolacakis, L. Xia, N. A. Ferrari, and E. A. Rich. 1997. Activation of latent HIV-1 by Mycobacterium tuberculosis and its purified protein derivative in alveolar macrophages from HIV-infected individuals in vitro. J. Acquir. Immune Defic. Syndr. 15:325–331. 36. Fraziano, M., G. Cappelli, M. Santucci, F. Mariani, M. Amicosante, M. Casarini, S. Giosue, A. Bisetti, and V. Colizzi. 1999. Expression of CCR5 is increased in human monocyte-derived macrophages and alveolar macrophages in the cours of in vivo and in vitro Mycobacterium tuberculosis infection. AIDS Res. Hum. Retroviruses 15:869–874. 37. Meylan, P. R., J. R. Munis, D. D. Richman, and R. S. Kornbluth. 1992. Concurrent human immunodeficiency virus and mycobacterial infection of macrophages in vitro does not reveal any reciprocal effect. J. Infect. Dis. 165:80–86. 38. Schluger, N. W., D. Perez, and Y. M. Liu. 2002. Reconstitution of immune respoonses to tuberculosis in patients with HIV infection who receive antiretroviral therapy. Chest 122:597–602.

Suggest Documents