Tuberculosis 94 (2014) 111e122

Contents lists available at ScienceDirect

Tuberculosis journal homepage: http://intl.elsevierhealth.com/journals/tube

IMMUNOLOGICAL ASPECTS

Different responses of human mononuclear phagocyte populations to Mycobacterium tuberculosis Camilo Duque a, Leonar Arroyo a, Héctor Ortega b, c, Franco Montúfar d, Blanca Ortíz a, Mauricio Rojas a, Luis F. Barrera a, * a Grupo de Inmunología Celular e Inmunogenética (GICIG), Instituto de Investigaciones Médicas, Facultad de Medicina, Universidad de Antioquia, UdeA, Calle 70 No. 52-21, Medellín, Colombia b Hospital Pablo Tobón Uribe, Medellín, Colombia c Clínica Cardiovascular Santa María, Medellín, Colombia d IPS Universitaria Clínica León XIII Sede Medellín, Medellín, Colombia

a r t i c l e i n f o

s u m m a r y

Article history: Received 26 June 2013 Received in revised form 29 October 2013 Accepted 2 November 2013

Mycobacterium tuberculosis (Mtb) infects different populations of macrophages. Alveolar macrophages (AMs) are initially infected, and their response may contribute to controlling Mtb infection and dissemination. However, Mtb infection may disseminate to other tissues, infecting a wide variety of macrophages. Given the difficulty in obtaining AMs, monocyte-derived macrophages (MDMs) are used to model macrophageemycobacteria interactions in humans. However, the response of other tissue macrophages to Mtb infection has been poorly explored. We have compared MDMs, AMs and splenic human macrophages (SMs) for their in vitro capacity to control Mtb growth, cytokine production, and induction of cell death in response to Mtb H37Rv, and the Colombian isolate UT205, and to the virulence factor ESAT-6. Significant differences in the magnitude of cell death and cytokine production depending mainly on the Mtb strain were observed; however, no major differences in the mycobacteriostatic/mycobacteriocidal activity were detected among the macrophage populations. Infection with the clinical isolate UT205 was associated with an increased cell death with membrane damage, particularly in IFNg-treated SMs and H37Rv induced a higher production of cytokines compared to UT205. These results are concordant with the interpretation of a differential response to Mtb infection mainly depending upon the strain of Mtb. Ó 2013 Published by Elsevier Ltd.

Keywords: Mycobacterium tuberculosis ESAT-6 Mononuclear phagocytes Cell death Cytokines Interferon gamma

1. Introduction Mycobacterium tuberculosis (Mtb) the etiological agent of tuberculosis (TB) uses both monocytes and macrophages as niches in order to prolong survival, multiplication and transmission in the human population. In opposition to this pathogen, mechanisms portrayed by macrophages have evolved in order to control or to eliminate the disturbing consequences of infection, such as chronic inflammation, pathology and disease. Circulating monocytes give rise to specialized tissue macrophage populations, a process in which microenvironment has proposed to play a directive role [1]. In the current view, Mtb infects alveolar macrophages, which in turn trigger several

* Corresponding author. Grupo de Inmunología Celular e Inmunogenética (GICIG), Instituto de Investigaciones Médicas, Facultad de Medicina, Universidad de Antioquia, Medellín, Colombia. Tel.: þ57 4 219 6448; fax: þ57 4 219 1060. E-mail address: [email protected] (L.F. Barrera). 1472-9792/$ e see front matter Ó 2013 Published by Elsevier Ltd. http://dx.doi.org/10.1016/j.tube.2013.11.001

signaling pathways resulting in a chemokine-dependent recruitment of circulating monocytes and T cells to the site of infection and the production of proinflammatory cytokines, including IFNg, TNFa, IL-1b, IL-6, and IL-18 aimed to the elimination of the invading pathogen. Moreover, anti-inflammatory cytokines, such as IL-10 limits the inflammatory reaction [2]. However, in the majority of occasions, infection may persist for unlimited time, and in a minority of cases progress to an uncontrolled state. This state is characterized by dissemination through the lung tissue, or more rarely, to other tissues, generating a life threatening disease which is most common in infants and immunocompromised individuals [3,4]. Most of the knowledge we already have on the interaction of human macrophages and Mtb has been obtained from the use of monocyte-derived macrophages (MDMs). Blood derived monocytes are cultured in vitro for several days until they differentiate into cells resembling tissue macrophages. However, remarkably few studies have been performed with alveolar macrophages, the natural source of the primary infection, mostly due to ethical issues

112

C. Duque et al. / Tuberculosis 94 (2014) 111e122

related to the invasiveness of the procedure to obtain the samples of bronchoalveolar lavage. In addition, studies using other tissue macrophages are almost inexistent [5]. The genetic characterization and genome sequencing of Mtb strains has shown a greater variability than expected based on the clonal transmission of the bacilli. More interestingly, recent evidence suggests a phylogeographic adaptation of circulating Mtb strains to human populations that in some studies has been associated with different forms of TB disease [6]. Thus, more studies based on the interaction of monocytes and macrophage populations with circulating strains of Mtb are necessary to have a better understanding of the immune mechanisms associated with bacterial control. In this study, we have compared the in vitro response of monocytes, MDMs, alveolar (AMs) and splenic macrophages (SMs) to infection with the laboratory strain of Mtb H37Rv and to the recently isolated Colombian strain of Mtb UT205. To examine the response of these mononuclear cells, the induction of cell death (apoptosis and necrosis) and production of cytokines in response to infection and to the virulence factor ESAT6, and the antimycobacterial activity in the absence or presence of IFNg were determined. Our results are consistent with a variable response of macrophages that mostly depends on the strain of Mtb. 2. Materials and Methods 2.1. Reagents RPMI-1640, Heat inactivated AB Human Serum, Dulbecco’s PBS were purchased from Invitrogen (Carlsbad, CA). Histopaque was purchased from Sigma (St. Louis, MO). Penicillin-streptomycin was purchased from Biowittaker (Walkersville, MD). Glycerol was purchased from Promega (Madison, WI). 2.2. Subjects Healthy volunteers (n ¼ 13) were used as a source of peripheral venous blood for obtaining monocytes and monocytederived macrophages (MDMs) as described below. Alveolar macrophages (AMs) were obtained from bronchoalveolar lavage (BAL) from individuals suspected of respiratory illnesses not associated with HIV-infection and/or malignancies of the myeloid system (n ¼ 8) at the Hospital Universitario Pablo Tobón Uribe, Clínica Cardiovascular La María and IPS Universitaria Clínica León XIII, Sede Medellín (Medellín, Colombia). Splenic macrophages (SMs) were obtained from spleen slices from deceased donors (n ¼ 14) of the Transplantation Programs of the Hospital Universitario Pablo Tobón Uribe, and the IPS Universitaria León XIII Sede Medellín (Medellín, Colombia). The cause of death for the majority of the donors (n ¼ 10, 71.4%) was trauma; the remaining (n ¼ 4, 28.6%), included different cerebrovascular causes. None of the donors were HIVþ.

antibiotics, prewarmed at 37  C. In these conditions, CD14þ cells represented >95% of the adherent cells (results not shown). MDMs were obtained after 5 days of culture of monocyte monolayers in RPMI-1640 supplemented with 10% ABþ inactivated human serum, penicillin and streptomycin (complete medium, CM). Control experiments indicated no significant detachment of cells during the culture period (data not shown). Splenic macrophages were obtained of deceased donors as previously described [5]. Adherent cells were detached by treatment with 0.05% trypsinEDTA for 10 min, washed, counted, and then seeded at 1  105 macrophages/well in 48-well tissue culture plates in CM without antibiotics for 24 h before infection. To prepare AMs, BAL obtained from healthy areas of the lung were centrifuged for 5 min at 650 g and resuspended in CM. One hundred thousand dark granular cells, morphologically corresponding to macrophages, were seeded in 48-well plates and cultured for 4 days in CM. At this point, nonadherent cells were eliminated by extensive washings with warm DPBS supplemented with 0.5% ABþ human serum, and then cultured for additional 24 h in CM without antibiotics before being infected. 2.4. Mycobacteria M. tuberculosis strain H37Rv was obtained from the Instituto Nacional de Salud, Bogotá, Colombia. The Mtb clinical isolates UT205, UT127 and UT379 were obtained from the Centro Colombiano para la Investigación en Tuberculosis (CCITB). Mtb was grown in Middlebrook 7H9 broth supplemented with 10% OADC (BD, NY) and Tween 80 (0.05%), for 2e3 weeks to reach exponential growth phase. Mycobacteria were cultured as previously described [8]. Mycobacterial clumps were disrupted by 6 sonication cycles of 10 s at 4  C, each cycle for 40 W output (CV33 Sonics Vibra Cell, Newtown, CT). The sonicate was gently centrifuged for 5 min at 250 g at 4  C, and the upper bacterial suspension was diluted in freezing medium, adjusted to final absorbance of 0.1 (OD620) and frozen at 70  C until used. The number of colony forming units (CFU) was determined by plating 20 ml of serial dilutions onto petri dishes (Corning, NY), containing Middlebrook 7H10 agar supplemented with glycerol and 10% OADC pH 7.2 and the CFU counted after 3 weeks of culture at 37  C. Upon thawing, mycobacterial viability of FDA stained bacteria (usually more than 90%) was tested by flow cytometry essentially as described [9]. The Colombian strains of Mtb UT205, UT127 and UT379 were collected during a large cohort study conducted by the Centro Colombiano para la Investigación en Tuberculosis (CCITB) during 2005e2009 [10]. All of them belong to the Latinoamerican-Mediterranean family (LAM) of Mtb UT205 was obtained from a household in which an incident case was reported while no incident cases were reported in the household from which the UT127 strain was recovered although the household contacts showed evidence of infection as tested by an in-house IGRA. UT379 was recovered from a household in which no evidence of infection was detected in the household contacts.

2.3. Monocytes and macrophages 2.5. Infection of macrophage populations Monocytes were obtained from peripheral blood of healthy subjects and differentiated into MDMs as previously described [7]. To obtain monocyte monolayers, 2  105 CD14þ cells were seeded in 0.5 ml of RPMI-1640 (Invitrogen, Grand Island, NY) supplemented with 0.5% ABþ inactivated human serum (Invitrogen, Brown Deer, WI), penicillin and streptomycin (Biowittaker, Walkersville, MD), for 4 h at 37  C, 5% CO2, 95% relative humidity, and then extensively washed with DPBS (Invitrogen, Grand Island, NY) supplemented with 0.5% ABþ inactivated human serum and

Macrophage populations (MDMs, SMs and AMs) were infected with Mtb at a multiplicity of infection (MOI) of 5:1 for 6 h, and washed extensively with warm DPBS supplemented with 0.5% ABþ inactivated human serum to eliminate noningested bacteria. Twenty four hours after infection, macrophages were cultured in the presence or absence of recombinant human IFNg (500 ng/ml). The effect of infection and IFNg treatment on cell death and the mycobacteriostatic/mycobactericidal activity was determined

C. Duque et al. / Tuberculosis 94 (2014) 111e122

96 h after infection. As described below, the cytokine/chemokine production of macrophages was determined at 6 and 24 h after infection. For determination of antimycobacterial activity, cells were lysed with 0.1% Triton X-100 to release the intracellular mycobacteria. Serial dilutions of cell lysate were plated into Middlebrook 7H10 supplemented with glycerol and OADC, incubated at 37  C and CFU were counted after 3 weeks. The growth rate of Mtb in the infected macrophages was calculated as the ratio of CFUs recovered at 96 h to CFUs recovered at the 6 h time point. The capacity of the ESAT-6 virulence factor treatment to induce apoptosis and necrosis in macrophages was also tested. Macrophages were cultured in the presence or absence of 10 mg of purified ESAT-6 (obtained from BEI Resources, contract CO0088) for 24 h. In selected experiments, the effect of ESAT-6 on cell death was also determined in monocyte monolayers prepared as previously described [7]. 2.6. Determination of cell death Cell death (apoptosis and necrosis) was determined by morphological and biochemical procedures. To reduce the effects of cell scraping on viability, monocyte and macrophage monolayers were centrifuged for 5 min 650 g, at 20  C and then stained with 200 ml of a solution containing 4 mg/ml of Acridine Orange and Ethidium Bromide (AO/EB) in DPBS. AO is permeable to all cells and emits green fluorescence when intercalated with nucleic acids. The EB is only permeable to cells in which plasma membrane integrity is compromised and emits red fluorescence when intercalated with DNA. Thus, cell death can be determined by the differential uptake of both dyes; the nuclei of viable cells are intact, and emit diffuse green fluorescence. Apoptotic cell nuclei, are smaller (because of the chromatin condensation) and emit a brighter green fluorescence, sometimes chromatin condensation is evident. Necrotic cells display red or orange (combination of both dyes) nuclei. Swollen and anucleated cells were treated as necrotic. The microplates were centrifuged again for 5 min 650 g. The cells were immediately observed on a NIKON TS100 fluorescence inverted microscope (Nikon Corporation, Japan), equipped with a HBO 50 W mercury gas lamp, and then photographed using a DS-fi1 camera coupled with a Digital Sight microphotography system (Nikon Corporation, Japan). Eight to 10 fields per well (corresponding to 500e1000 cells) were photographed at random, from which the percentage of cells with condensed nuclei or membrane damage was estimated. Necrosis was also determined by the detection of mono- and oligonucleosomes in culture supernatants, using a Cell Death Detection ELISAplus (Roche diagnostics, Mannheim, Germany) following manufacturer’s instructions. 2.7. Determination of cytokine/chemokine production by Luminex and ELISA The levels of IL-1b, IL-6, IL-10, IL-12p70, TNFa, and MCP-1 present in the supernatant of uninfected or infected macrophages for 6 and 24 h were determined by MILLIPLEXÒ xMAP Human Cytokine/ Chemokine kit (Billerica, MA) and read in a Bio-plex 200 (Bio-Rad, CA) following manufacturer’s recommendations. Data was analyzed by the BioPLex software (Bio-Rad Laboratories, Hercules, CA). ELISA tests for IL-18 (MBL, Woburn, MA, USA) and IL-27 (BioLegend, San Diego, CA) were performed following manufacturer’s recommendations. Optical density was measured on an ELISA plate reader (Bio-Tek Elx800 NB, Kimpton, UK). Cytokine concentrations were calculated using standard curves generated from recombinant cytokines, and the results were expressed in pg/ml. Detection levels

113

were 2 pg/ml for MILLIPLEXÒ xMAP kit, 12,5 pg/ml for IL-18 ELISA kit and 11 pg/ml for IL-27. 2.8. Statistical analysis Data was analyzed using One or Two way ANOVA unless otherwise stated. For single comparisons, ManneWhitney U test was performed. Data was calculated using GraphPad Prism 5 software (La Jolla, CA). 3. Results 3.1. M. tuberculosis H37Rv and the Colombian clinical isolates of Mtb differ in their ability to induce apoptosis and necrosis in human monocytes Our laboratory [7,11] has shown that in vitro infection of human monocytes with Mtb results in apoptosis. Initially, we infected monocytes from healthy donors with the laboratory strain H37Rv and the recently isolated Colombian strains of Mtb, UT205, UT379, and UT127, at MOIs of 5:1 and 10:1. Twenty four hours after infection, the percentage of cells with nuclear condensation or membrane damage was determined by in situ staining with Acridine Orange and Ethidium Bromide (AO/EB). From high resolution photomicrographs, we counted 500e1000 cells from each well to assure statistical confidence. At the lower MOI of 5:1 infection with UT127 (p < 0.05), UT379 (p < 0.01), and UT205 (p < 0.001), but not H37Rv induced an increase in the percentage of cells with condensed nuclei in comparison to noninfected monocytes. In comparison, infection with UT205 induced an increase in the percentage of cells with condensed nuclei compared to H37Rv (p < 0.05) but comparable percentages to UT127 and UT379 (Figure 1(A)). At this MOI, just UT205 (p < 0.001) increased the percentage of cells with membrane damage compared to noninfected monocytes, and compared to UT379 (p < 0.05), but no significant differences compared to H37Rv and UT127 (Figure 1(B)). Increasing MOI to 10:1, resulted in increased percentage of cells with condensed nuclei in response to infection with H37Rv, UT127, UT379 and UT205 (p < 0.001), and no differences were observed among the Mtb strains (Figure 1(A)). At this MOI, infection with H37Rv (p < 0.01) and UT205 (p < 0.001) but not UT127 and UT379 increased the percentage of cells with membrane damage compared to noninfected monocytes and infection with UT205 increased in the percentage of cells with membrane damage compared to UT379 (p < 0.05) but comparable levels to H37Rv and UT127 (Figure 1(B)). Based on these observations, we selected UT205 and H37Rv to infect monocyte-derived macrophages (MDMs), splenic (SMs) and alveolar (AMs) macrophages, using a MOI of 5:1. 3.2. Cell death induced by infection with Mtb depends on the mycobacterial strain and not on the macrophage population Once we selected H37Rv and UT205 for further experiments, we asked whether SMs, AMs and in vitro differentiated MDMs, may differentially respond to infection with H37Rv or UT205 in terms of cell death. Macrophages were infected with 5:1 MOI, and the percentage of macrophages with condensed nuclei or membrane damage was determined 96 h post-infection. Infection of MDMs with H37Rv or UT205 (p < 0.05) increased the percentage of cells with condensed nuclei compared to noninfected cells; this percentage also increased in SMs infected with H37Rv (p < 0.01) and UT205 (p < 0.001). Infection of AMs with H37Rv or UT205 also showed an increase compared to

114

C. Duque et al. / Tuberculosis 94 (2014) 111e122

Figure 1. Cell death induction in monocytes by M. tuberculosis H37Rv and the Colombian clinical isolates UT379, UT127 and UT205. 2.5  105 CD14þ monocytes were infected with the laboratory strain of Mtb H37Rv, or the Colombian clinical isolates UT379, UT127, and UT205 at MOIs of 5:1 and 10:1 for 24 h. Cell death in situ was determined by staining the cells with Acridine Orange and Ethidium bromide (AO/EB) as described in Materials and Methods. (A) Cells with condensed or fragmented nuclei were considered apoptotic, and (B) cells with loss of membrane integrity were considered necrotic. Statistical differences were calculated by One Way ANOVA using the Bonferroni posttest. Results are displayed as the Mean  SEM (n ¼ 4). Significant increase in the percentage of cell death in infected compared to noninfected cells is indicated to the top of the bars. *, p < 0.05; **, p < 0.01; ***, p < 0.001.

Figure 2. Cell death induction in macrophage populations infected with M. tuberculosis H37Rv or the clinical isolate UT205. Macrophages (2.5  105 MDMs, 1  105 SMs and AMs) were infected with for 6 h with Mtb H37Rv or UT205 at MOI of 5. Macrophage monolayers were extensively washed to eliminate noningested bacteria, and then cultured for additional 90 h. Cell death was determined by in situ staining of the cells with Acridine Orange and Ethidium bromide (AO/EB) as described in Materials and Methods. (A) Cells with condensed or fragmented nuclei were considered apoptotic, and (B) cells with loss of membrane integrity were considered necrotic. (C). Photomicrographs of AO/EB stained noninfected (PHS 10%, subpanels AeC) and Mtb-infected macrophages (subpanels DeF) (SMs, spleen; MDMs, monocyte-derived; AMs, alveolar macrophages) showing morphological and staining characteristics associated with apoptosis and necrosis. A, apoptotic cells; N, necrotic cells. Statistical differences were calculated by Two Way ANOVA using the Bonferroni posttest. Results are displayed as the Mean  SEM (MDMs ¼ 12; SMs ¼ 13; AMs ¼ 4). *, p < 0.05; **, p < 0.01; ***, p < 0.001.

C. Duque et al. / Tuberculosis 94 (2014) 111e122

noninfected cells, but the difference did not reach statistical significance (Figure 2(A)). Infection of MDMs with H37Rv (p < 0.05) and UT205 (p < 0.001) resulted in an increase in the percentage of cells with membrane damage compared to noninfected cells, and a similar result was obtained with SMs (UT205, p < 0.01), and AMs (UT205, p < 0.05) (Figure 2(B)). As compared to macrophages with condensed nuclei, infection resulted in a higher percentage of cells with membrane damage (compare Figure 2(A) and (B)). No significant differences in the percentages of the macrophage populations with condensed nuclei, or membrane damage in response to infection with H37Rv or UT205 were observed (data not shown). 3.3. Stimulation of infected macrophages with IFNg results in an increase of cells with membrane damage and depends on mycobacterial strain and macrophage population We also determined the effects of IFNg on infected cells following the course of the initial natural infection in which macrophages are first infected with Mtb and then activated for antimicrobial activity by IFNg. As shown in Figure 3(A), addition of IFNg to SMs infected with H37Rv (p < 0.001) or UT205

115

(p < 0.001) increased the proportion of cells with membrane damage compared to infected but not IFNg-treated cells. Infection of SMs with the clinical isolate UT205 also induced an increase in the percentage of cells with membrane damage compared to H37Rv, both in absence (p < 0.001) or presence of IFNg (p < 0.001). MDMs and AMs displayed a similar trend to that observed in SMs, but no significant differences were observed, probably due to the high variability of the individual responses. The amount of cells with membrane damage in infected macrophages cultured in the presence or absence of IFNg was also assessed by determining the amounts of mono- or oligonucleosomes released in the supernatant of the cultured cells (Figure 3(B)). Again, the amount of mono- or oligonucleosomes released by SMs infected with UT205 was higher compared to H37Rv, both in absence (p < 0.01) or presence (p < 0.01) of IFNg. In the presence of IFNg, the amount of nucleosomes released was higher in SMs infected with H37Rv and UT205; however, this difference was not statistically significant. No significant differences in the percentages of cells with condensed nuclei were observed in infected macrophages stimulated with IFNg (data not shown).

Figure 3. Effect of the IFNg treatment on necrosis in previously infected macrophages with M. tuberculosis. Macrophages (2.5  105 MDMs, 1  105 SMs and AMs) were infected for 6 h with Mtb H37Rv or UT205 at MOI of 5. Macrophage monolayers were extensively washed to eliminate noningested bacteria, 18 h later stimulated with IFNg and then cultured for additional 72 h. The percentage of cells with membrane damage was obtained from cells stained with AO/EB (MDMs ¼ 12; SMs ¼ 13; AMs ¼ 4) (A), or by the determination of mono- and oligonucleosomes present in the supernatant (MDMs ¼ 5; SMs ¼ 5; AMs ¼ 4) (B) as described in Material and Methods. Statistical differences were calculated by the ManneWhitney U-test. Results are displayed as the Median and IQ range (10e90%) of net values. **p < 0.01, ***p < 0.001.

116

C. Duque et al. / Tuberculosis 94 (2014) 111e122

The antimycobacterial activity of macrophage populations infected with H37Rv or UT205 was also tested at 96 h post-infection. In the absence of IFNg, SMs displayed more CFUs for H37Rv than MDMs (p < 0.05) and for UT205 compared to MDMs (p < 0.001) and AMs (p < 0.05). In the presence of IFNg, a similar trend was observed, but differences were only significant between SMs and MDMs for H37Rv (p < 0.01). These differences could not be explained by the phagocytic uptake at 6 h (Figure 4(C)). We also calculated the ratio of Mtb growth (CFUs at 96 h to CFUs at 6 h) in absence (Figure 4(D)) or presence (Figure 4(E)) of IFNg. UT205 but not H37Rv grew significantly more in SMs compared to MDMs (p < 0.01) and AMs (p < 0.05). This difference was not statistically significant in the presence of IFNg. Overall, the treatment of infected macrophages with IFNg did not result in an appreciable antimycobacterial activity. 3.4. The pattern of cytokines in response to infection depends on the macrophage population and the strain of Mtb To gain more information on the intrinsic capacity of the macrophage populations to respond to H37Rv and UT205, we measured the macrophage-derived production of cytokines known to be important for the antimycobacterial response at 6 and 24 h time points post-infection. As shown in Figure 5, the pattern of production of TNFa, IL-1b, IL-6, IL-18 and MCP-1 by the macrophage populations in response to infection was similar. However, some differences in cytokines production were observed among macrophages. At the basal level (6 h, left panel), SMs produced higher levels of TNFa (p < 0.05), IL-1b (p < 0.01) and MCP-1 (p < 0.001) compared to MDMs, and of MCP-1 (p < 0.01) compared to AMs. In the absence of infection, no significant differences were observed among the macrophages at the 24 h time point. Upon infection with H37Rv for 6 h, AMs (p < 0.001) and SMs (p < 0.05) produced higher levels of TNFa compared to MDMs. AMs but not SMs produced higher levels of IL-6 (p < 0.001) compared MDMs. No significant differences were

observed among the macrophage populations for their production of IL-1b, IL-10 and MCP-1. Twenty-four hours after infection, AMs produced higher levels of TNFa, IL-1b, and IL-6 compared to SMs (p < 0.01, p < 0.001, p < 0.05, respectively) and MDMs (p < 0.01, p < 0.001, p < 0.001, respectively). Moreover, SMs produced higher levels of IL-10 (p < 0.05) and MCP-1 (p < 0.05) compared to MDMs but no significant differences in comparison with AMs. In response to infection with UT205, no significant differences were observed among the macrophage populations at 6 h after infection. Twenty four hours upon infection, AMs produced higher amounts of TNFa (p < 0.01), IL-1b and IL-6 (p < 0.001) compared to MDMs and SMs, respectively. No differences in the amounts of IL-10 and MCP-1 were observed among the macrophage populations. No differences in the levels of IL-18 were observed among the macrophage populations in response to infection with H37Rv or UT205. Macrophages produced very low (IL-12p70) or undetectable (IL-27) levels (data not shown). Since we observed some differences in the amount of cytokines induced by infection with H37Rv or UT205 among the macrophage populations, we also compared the ratios of the net cytokine production between H37Rv and UT205 for each macrophage population. As it can be observed from Table 1, infection with UT205 induces lower levels of cytokines and MCP-1 compared with H37Rv. Major differences in MDMs were found at 24 h after the infection for TNFa (ratio 4), IL-6 (ratio 4.8), IL-10 (ratio 3.1) and MCP-1 (ratio 2.9) and TNFa (ratio 17), IL-6 (ratio 6.9) and IL-10 (ratio 14.6) in SMs while in AMs, considerable differences were observed at 6 h post-infection in TNFa (ratio 11) and IL-6 (ratio 12.2). Interestingly, UT205 induced higher levels of IL-1b compared to H37Rv, except in AMs at 24 h (ratio 4.4). Overall, infection of MDMs, SMs and AMs with H37Rv and UT205 showed a partially overlapping response in the production of cytokines, the infection with UT205 induced a lower response of cytokines compared to H37Rv and AMs produced higher amounts

Figure 4. Effect of IFNg on the ability of macrophages to control infection with H37Rv or UT205. Macrophages (MDMs, 2.5  105; SMs and AMs, 1  105) were infected with M. tuberculosis H37Rv or the UT205 clinical isolate at MOI of 5 for 6 h and then washed to eliminate non-phagocytosed bacilli. Eighteen hours after infection, IFNg was added, and after additional 72 h macrophages were lysed, and serial dilutions of the supernatant were cultured in 7H10 medium supplemented with OADC. (AeB), comparison of the amounts of CFU of H37Rv or UT205 in absence (A), or presence (B) of IFNg. (C) Amount of CFU of H37Rv or UT205 recovered after 6 h of phagocytosis in MDMs, SMs and AMs. The amounts of CFU were estimated from triplicate cultures. (DeE) Growth rate of Mtb H37Rv and UT205 in macrophages (MDMs, SMs and AMs) in absence (D) or presence (E) of IFNg. The amount of CFUs is expressed by 1  105 macrophages seeded (MDMs ¼ 6; SMs ¼ 9; AMs ¼ 3). Statistical differences were calculated by Two way ANOVA using the Bonferroni posttest. Results are displayed as the Mean  SEM. **p < 0.05, **p < 0.01, ***p < 0.001.

C. Duque et al. / Tuberculosis 94 (2014) 111e122

117

Figure 5. Cytokine production by macrophages in response to infection with M. tuberculosis H37Rv and the UT205 clinical isolate. Macrophages were infected as described in Materials and Methods. At 6 and 24 h upon infection, supernatant from noninfected and infected macrophages were collected, filtered and stored at 80  C. Cytokines production in supernatants was determined by MILIPLEXÒ xMAP technology. Differences among macrophage populations in each time point were calculated by Two Way ANOVA (MDMs ¼ 12; SMs ¼ 11; AMs ¼ 4). Asterisks above bars represent statistical differences to noninfected controls; asterisks above brackets represents statistical differences between macrophages. *, p < 0.05; **, p < 0.01; ***, p < 0.001.

of TNFa, IL-1b and IL-6 upon infection with H37Rv and UT205 compared to SMs and AMs, indicating a higher proinflammatory response to Mtb. 3.5. The virulence factor ESAT-6 induces apoptosis in monocytes but not in macrophages It has been shown that the virulence factor ESAT-6 induces apoptosis in type 1 and type 2 pneumocytes and the epithelial

cell line A549 [12,13] and the THP-1 macrophage line [14]. More recently, ESAT-6 has been implicated in necrotic cell death of MDMs [15]. Since it is apparent that tissue macrophages are more resistant to Mtb-induced cell death compared to monocytes (Figures 1 and 2), we determined whether this resistance of macrophages to cell death was observed in the presence of ESAT6. As shown in Figure 6(A) and (B), the treatment of monocytes with ESAT-6 for 24 h induced a significant increase in apoptosis (p < 0.01) and necrosis (p < 0.01) compared to nontreated cells.

118

C. Duque et al. / Tuberculosis 94 (2014) 111e122

Table 1 Comparison of the production of cytokines by MDMs, SMs and AMs in response to infection with Mtb H37Rv or UT205.

TNFa IL-1b IL-6 IL-10 IL-18 MCP-1

6h 24 h 6h 24 h 6h 24 h 6h 24 h 6h 24 h 6h 24 h

MDMs

SMs

AMs

Rv/UT205

Rv/UT205

Rv/UT205

2.1 4.0 0.8 0.3 1.7 4.8 0.9 3.1 1.1 0.7 1.0 2.9

7.7 17.0 0.3 0.4 9.3 6.9 2.6 14.6 1.4 0.8 2.9 0.3

11.0 1.0 0.6 4.4 12.2 0.9 1.2 2.0 1.1 1.3 1.6 0.2

The values represent the ratio of the mean net values per each treatment and were calculated as the mean cytokine of the net values induced by H37Rv divided by the mean cytokine of the net values induced by UT205.

The treatment with ESAT-6 also resulted in a significant increase in apoptosis of MDMs (p < 0.001) and SMs (p < 0.05) but not of AMs (Figure 6(C)), and no significant increase in cells with membrane damage was observed among the mononuclear phagocyte populations compared to nontreated controls (Figure 6(D)). When we compared the capacity of ESAT-6 to induce apoptosis or necrosis in monocytes and macrophage populations, monocytes were more susceptible to apoptosis induction compared to SMs (p < 0.05) and AMs (p < 0.001) but not to MDMs. In addition, MDMs (p < 0.001) and SMs (p < 0.05) were more susceptible to apoptosis induction than AMs, but no

significant difference was observed between MDMs and SMs (Figure 6(C)). We also compared the production of IL-1b, IL-6, IL-10, TNFa and MCP-1 following ESAT-6 treatment for 24 h among the different mononuclear cells populations studied. Macrophages produced significant amounts of TNFa, IL-6 and IL-10 compared to unstimulated controls; MDMs and AMs but not SMs produced significant amounts of IL-1b while SMs but not MDMs and AMs produced significant amount of MCP-1 (Figure 7A). In the presence of ESAT-6, monocytes produced higher amounts of TNFa compared to MDMs (p < 0.01) and AMs (p < 0.001) while SMs produced higher amounts compared to AMs (p < 0.05). No significant differences were observed between MDMs and AMs. Monocytes were also higher producers of IL-1b in comparison with MDMs, SMs and AMs (p < 0.001), and no differences were observed among the macrophage populations. SMs produced higher levels of IL-6 compared to MDMs (p < 0.001) and AMs (p < 0.05), and IL-10 and MCP-1 in comparison to monocytes, MDMs and AMs (p < 0.001). No significant differences in IL-6, IL10 and MCP-1 production were observed between monocytes, MDMs and AMs (Figure 7B).

4. Discussion Functional macrophage heterogeneity may play a crucial role in the innate immune response against Mtb [15e19] Here we have examined the in vitro role of monocytes, MDMs and importantly of tissue macrophages such as AMs and SMs to Mtb in terms of antimycobacterial activity, induction of cell death and cytokine production in response to H37Rv and to the

Figure 6. Effect of ESAT-6 on cell death induction in monocytes, monocyte-derived macrophages, splenic and alveolar macrophages. Monocyte and macrophage monolayers (MDMs, SMs and AMs) were prepared as described in Materials and Methods, and incubated for 24 h in presence of purified ESAT-6 (10 mg/ml). Cells were stained AO/EB as described in Materials and Methods, and the percentage of cells with condensed nuclei or membrane damage was determined from photomicrographs. AeB, percentage of cells with condensed nuclei and CeD, percentage of cells with membrane damage in monocytes (A and B) or the comparison between monocytes, MDMs, SMs and AMs (C and D). Statistical differences in the percentage of apoptotic or necrotic monocytes as a consequence of ESAT-6 treatment were calculated using paired t-Test. Differences among the mononuclear phagocyte populations (Mo, MDMs, SMs and AMs) were calculated by Two Way ANOVA. Results represent the Means  SEM of n ¼ 5 of Mo, n ¼ 12 for MDMs, n ¼ 13 for SMs, and n ¼ 6 for AMs. *p < 0.05, **p < 0.01, ***p < 0.001.

C. Duque et al. / Tuberculosis 94 (2014) 111e122

119

Figure 7. Production of cytokines by monocytes and macrophages in response to ESAT-6. Monolayers of monocytes (Mo), monocyte-derived macrophages (MDMs), splenic (SMs) and alveolar (AMs) macrophages, were incubated for 24 h in presence (10 mg/ml) or absence of the virulence factor ESAT-6. A, production of IL-1b, IL-6, IL-10, TNFa and MCP-1 by MDMs, SMs and AMs. B, comparison of the production of cytokines present in the supernatants of treated and nontreated (NT) cells among the mononuclear phagocyte populations. (MDMs ¼ 9; SMs ¼ 9; AMs ¼ 6). The differences in cytokine and MCP-1 production among the mononuclear phagocyte populations were determined by Two Way ANOVA. Data are presented as the Mean  SEM, and normalized by the amount of cells seeded. *p < 0.05, **p < 0.01, ***p < 0.001.

recently obtained clinical isolate UT205, and to the virulence factor ESAT-6. Our findings indicate that IFNg treatment of previously infected macrophages increases cell death, particularly of necrosis. The development of necrotic lesions is characteristic of active TB. Evidence collected from other mycobacterial infections points for an important role of IFNg in granuloma necrosis [16e 18]. Thus, our results with human tissue macrophages are in line with the previous observations and reinforce the idea that depending on the particular microenvironmental conditions the immune response may be either protective or detrimental. In this scenario, the timing and amount of proinflammatory cytokines such as IFNg might be crucial for infection control. In addition, our observation of the incapacity of IFNg to significantly modulate the in vitro Mtb growth in infected macrophages is consistent with previous reports. Early observations of infected human macrophages with Mtb showed no appreciable antimycobacterial effect of IFNg [19e22]. Since in our model macrophages were first infected with Mtb before the addition of IFNg, mycobacterial products may have attenuated the stimulatory effects of IFNg as firmly established [23e26], leading to the suggestion that it may

constitute one of the main escape mechanisms evolved by Mtb to evade the deleterious consequences of intracellular life in macrophages. The increase in cell death of infected macrophages upon treatment with IFNg may have been a consequence of signaling pathways triggered by other cytokines such as TNFa and IL-1b which are actively produced by macrophages upon infection [27,28]. Interestingly, recent data in the model of zebrafish show that TNFa excess in response to mycobacterial infection induces mitochondrial reactive oxygen species (ROS) leading to necroptosis and release mycobacteria into the growthpermissive extracellular milieu [29]. Monocytes displayed more apoptosis and necrosis in response to infection, and to treatment with the virulence factor ESAT-6 while the tissue macrophages were more refractory to cell death induction. In some circumstances, such as the absence of stimulation, monocytes are reported to be more prone to apoptosis [30]. Also, monocyte differentiation into macrophages was associated with upregulation of Flip, a caspase inhibitor, and a decrease in Fas-mediated apoptosis [31]. Whether the differences observed between monocytes and macrophages stand from monocyte-to macrophage differentiation, is an issue that

120

C. Duque et al. / Tuberculosis 94 (2014) 111e122

deserves more research. Recent studies from our laboratory show that monocytes from pulmonary TB patients are more susceptible to develop necrosis compared to monocytes from healthy individuals when cultured in the presence of PPD or Mtb H37Rv [7,11], suggesting an effect of cellular differentiation on the capacity of mononuclear phagocytes to induce cell death upon Mtb infection. Our results are concordant with previous reports demonstrating apoptosis induction by ESAT-6 in macrophages and epithelial cells [12e15]. Recently, Welin and colleagues [15] reported that infection of MDMs with H37Rv induced a necrotic death, independent of caspase-1 and cathepsin B, but dependent on the ESAT-6 production by Mtb. In fact, ESAT-6promoted NLRP3-dependent necrotic death in THP-1 human macrophages [32]. In response to ESAT-6, macrophages produced proinflammatory cytokines IL-1b, TNFa, and IL-6, as well as the anti-inflammatory IL-10. From those, monocytes produced significantly more IL-1b in response to ESAT-6 compared to macrophages. IL-1b production has been associated with the activation of the inflammasome and pyronecrosis [33]. Interestingly, inflammasome activation, IL-1b production and cell death in monocytes and macrophages in response to bacterial infections have been reported [34]. However, the specific mechanism(s) of cell death in monocytes and macrophages in response to ESAT-6 were not addressed in this study. To compare the response of the different macrophage populations to infection with H37Rv or the UT205 clinical isolate, we determined the cytokine pattern, focusing in those more prominent for the antimycobacterial response [35]. In general, a similar pattern of cytokines production was observed in the different macrophage populations upon Mtb infection. However, some differences were observed in the pattern of cytokine production among the macrophage populations. For example, AMs displayed a more proinflammatory type of response upon infection with H37Rv and UT205 compared to MDMs and SMs. Based on the fact that in vitro alveolar macrophages are less efficient than monocytes and polymorphonuclear phagocytes to kill bacteria in the absence of an inflammatory stimuli [36], and their increased IL-10 and TGFb production, it has been previously proposed that AMs display an alternatively activated phenotype [37], which may preserve the integrity of the alveolar space given the constant influx of particles and bacteria into the lung. Recently, Tomlinson and colleagues [38] presented evidence showing that freshly isolated AMs exhibited a marked proinflammatory signature. Our data show that in the resting condition, AMs showed a trend to produce higher levels of TNFa, IL-1b, and IL-6 although this difference was not statistically significant. However, upon infection with Mtb, AMs displayed a significant production of the three cytokines compared to MDMs and SMs, suggesting that at least under our experimental conditions, AMs showed a proinflammatory phenotype. Initial experiments tested the capacity of 3 Colombian clinical isolates and of the laboratory strain of Mtb to induce cell death in monocytes. Two of them (UT127 and UT379) induced a similar amount of apoptosis compared to H37Rv; however, the clinical isolate UT205 induced higher levels of apoptosis compared to H37Rv, and higher amount of necrosis compared to UT379. UT205 also induced a higher amount of necrosis in IFNg treated splenic macrophages, compared to MDMs and AMs, suggesting a higher virulence compared to the other clinical isolates and H37Rv. Differences in virulence among circulating strains of Mtb have been previously observed. Studies in macrophages infected in vitro with clinical isolates of Mtb have correlated virulence with their capacity to induce cell death, and the induction of cytokines such as TNFa [39,40]. Thus, our data agree with the previous observations

showing a variable cytokine response that depends on the macrophage population being interrogated and the strain of Mtb (H37Rv, UT205); however, it is interesting to note that the recent isolate UT205 induced a lower production of cytokines compared with the laboratory strain H37Rv. Although limited, this evidence tends to sustain the conclusion that strains of Mtb with a higher virulence induce a lower amount of proinflammatory cytokines. The molecular differences between H37Rv and UT205 that may explain the observed differences are presently unknown. We recently published the complete genome sequence of the UT205 clinical isolate [41]. As compared to the H37Rv reference genome, we found a 3.6 kb genomic deletion affecting the dosR encoding genes Rv1996 (conserved hypothetical protein) and Rv1997 (ctpF, metal cation transporting P-type ATPase), but indels or SNPs were found in other six genes belonging to this regulon. Mutations in devR (Rv3133c), encoding the DevR (DosR) response regulator and devS (Rv3132c), encoding a histidine sensor kinase gene have been associated with changes in virulence in animal models [42e44]. However, the dosR gene seems to be not necessary for entry, survival and multiplication in human monocytes in vitro [45], and all other six genes are reported to be not essential [46]. Thus, we favor the interpretation that possible virulence differences between H37Rv and UT205 may be explained by differences in other genes besides those affected in the DosR regulon of UT205 strain. In this study, we have focused on the capacity of Mtb strains to induce cell death in different populations of the mononuclear phagocyte system, including monocytes and tissue macrophages. For this purpose, we have used a morphological and staining criteria (AO/AB) to differentiate viable, apoptotic and necrotic cells. AO/AB staining has been extensively used to characterize cell death in different cell types and conditions [47], and independent studies have shown that AO/EB staining discriminate apoptotic and necrotic cells comparable with other widely used techniques, such as Annexin V/Propidium Iodide staining and TUNEL and caspase 3/7 activity [48e51]. Actually, a plethora of types of cell death are being described, and different molecular mechanisms may explain a similar morphology [52]. In our study, we are not presuming at the mechanism of cell death, and relied mostly in the use of AO/AB and the mono- and oligonucleosomal detection in the supernatant of infected cells to confirm necrotic death. However, at the moment we do not have evidence to conclude that the necrosis we observed is secondary necrosis since we did not measure cell death in infected macrophages early after infection. Another limitation of our study concerns the small sample of AMs used for our comparisons and the fact that BAL samples were obtained from people not presumably healthy, may be compromising macrophage function although care was taken to obtain BAL samples from the healthy areas of the lung and to exclude samples from individuals infected with HIV and Mtb, or with cancer. In addition, macrophages were cultured in vitro for 4 days before stimulation/infection, attenuating any effects of the microenvironment. In summary, we presented novel evidence using human monocytes and different human tissue macrophage populations showing a strain-dependent response to M. tuberculosis infection. Interestingly, in experimental conditions that simulate the initial response to natural infection our data may suggest a possible pathological role of IFNg that by itself or in combination with other cytokines may promote necrotic death of infected macrophages that in a proinflammatory microenvironment may lead to an exacerbation of the pathological response. Acknowledgments We wish to acknowledge to all those who voluntarily donated blood and bronchoalveolar lavage samples, as well as those

C. Duque et al. / Tuberculosis 94 (2014) 111e122

relatives of deceased people who gave their permission to use splenic tissue samples. [17]

Ethical approval: An informed consent was reviewed and approved by the ethics review board from the Facultad de Medicina, Universidad de Antioquia, Hospital Universitario Pablo Tobón Uribe, Clínica Cardiovascular La María, and IPS Universitaria Clínica León XIII Sede Medellín and was read and signed by all participants.

[18]

[19]

[20]

Funding: This study was supported by Colciencias, Bogotá, Colombia, grant 1115-452-21098, and the Programa Estrategia de Sostenibilidad 2013e2014, Universidad de Antioquia, Medellín, Colombia. CD was a recipient of the Program Jóvenes Investigadores e Innovadores “Virginia Gutiérrez de Pineda” award.

[21]

[22]

Authorship: CD, acquisition, analysis and interpretation of data; LA, HO, FM, BO, acquisition of data; MR, analysis and interpretation of data and drafting the article; LFB, conception and design of the study, analysis and interpretation of data, drafting the article and final approval of the version to be submitted.

[23]

[24]

Competing interests: The authors have no financial or personal conflicts of interest in the present study.

[25]

References

[26]

[1] Murray PJ, Wynn TA. Protective and pathogenic functions of macrophage subsets. Nat Rev 2011;11:723e37. [2] Redford PS, Murray PJ, O’Garra A. The role of IL-10 in immune regulation during M. tuberculosis infection. Mucosal Immunol 2011;4:261e70. [3] Sharma SK, Mohan A, Sharma A, Mitra DK. Miliary tuberculosis: new insights into an old disease. Lancet Infect Dis 2005;5:415e30. [4] Newton SM, Brent AJ, Anderson S, Whittaker E, Kampmann B. Paediatric tuberculosis. Lancet Infect Dis 2008;8:498e510. [5] Henao J, Sanchez D, Munoz CH, Mejia N, Arias MA, Garcia LF, Barrera LF. Human splenic macrophages as a model for in vitro infection with Mycobacterium tuberculosis. Tuberculosis (Edinb) 2007;87:509e17. [6] Gagneux S, Small PM. Global phylogeography of Mycobacterium tuberculosis and implications for tuberculosis product development. Lancet Infect Dis 2007;7:328e37. [7] Sanchez MD, Garcia Y, Montes C, Paris SC, Rojas M, Barrera LF, Arias MA, Garcia LF. Functional and phenotypic changes in monocytes from patients with tuberculosis are reversed with treatment. Microbes Infect 2006;8:2492e500. [8] Rojas M, Barrera LF, Puzo G, Garcia LF. Differential induction of apoptosis by virulent Mycobacterium tuberculosis in resistant and susceptible murine macrophages: role of nitric oxide and mycobacterial products. J Immunol 1997;159:1352e61. [9] Norden MA, Kurzynski TA, Bownds SE, Callister SM, Schell RF. Rapid susceptibility testing of Mycobacterium tuberculosis (H37Ra) by flow cytometry. J Clin Microbiol 1995;33:1231e7. [10] del Corral H, Paris SC, Marin ND, Marin DM, Lopez L, Henao HM, Martinez T, Villa L, Barrera LF, Ortiz BL, Ramirez ME, Montes CJ, Oquendo MC, Arango LM, Riano F, Aguirre C, Bustamante A, Belisle JT, Dobos K, Mejia GI, Giraldo MR, Brennan PJ, Robledo J, Arbelaez MP, Rojas CA, Garcia LF. IFNgamma response to Mycobacterium tuberculosis, risk of infection and disease in household contacts of tuberculosis patients in Colombia. PLoS One 2009;4:e8257. [11] Arcila ML, Sanchez MD, Ortiz B, Barrera LF, Garcia LF, Rojas M. Activation of apoptosis, but not necrosis, during Mycobacterium tuberculosis infection correlated with decreased bacterial growth: role of TNF-alpha, IL-10, caspases and phospholipase A2. Cell Immunol 2007;249:80e93. [12] Kinhikar AG, Verma I, Chandra D, Singh KK, Weldingh K, Andersen P, Hsu T, Jacobs Jr WR, Laal S. Potential role for ESAT6 in dissemination of M. tuberculosis via human lung epithelial cells. Mol Microbiol 2010;75:92e106. [13] Choi HH, Shin DM, Kang G, Kim KH, Park JB, Hur GM, Lee HM, Lim YJ, Park JK, Jo EK, Song CH. Endoplasmic reticulum stress response is involved in Mycobacterium tuberculosis protein ESAT-6-mediated apoptosis. FEBS Lett 2010;584:2445e54. [14] Derrick SC, Morris SL. The ESAT6 protein of Mycobacterium tuberculosis induces apoptosis of macrophages by activating caspase expression. Cell Microbiol 2007;9:1547e55. [15] Welin A, Eklund D, Stendahl O, Lerm M. Human macrophages infected with a high burden of ESAT-6-expressing M. tuberculosis undergo caspase-1- and cathepsin B-independent necrosis. PLoS One 2011;6:e20302. [16] Ehlers S, Benini J, Held HD, Roeck C, Alber G, Uhlig S. Alphabeta T cell receptor-positive cells and interferon-gamma, but not inducible nitric oxide

[27]

[28]

[29] [30]

[31]

[32] [33]

[34] [35] [36] [37] [38]

[39]

[40]

[41]

[42]

[43]

121

synthase, are critical for granuloma necrosis in a mouse model of mycobacteria-induced pulmonary immunopathology. J Exp Med 2001;194: 1847e59. Florido M, Cooper AM, Appelberg R. Immunological basis of the development of necrotic lesions following Mycobacterium avium infection. Immunology 2002;106:590e601. Florido M, Appelberg R. Granuloma necrosis during Mycobacterium avium infection does not require tumor necrosis factor. Infect Immun 2004;72: 6139e41. Douvas GS, Looker DL, Vatter AE, Crowle AJ. Gamma interferon activates human macrophages to become tumoricidal and leishmanicidal but enhances replication of macrophage-associated mycobacteria. Infect Immun 1985;50:1e8. Rook GA, Steele J, Ainsworth M, Champion BR. Activation of macrophages to inhibit proliferation of Mycobacterium tuberculosis: comparison of the effects of recombinant gamma-interferon on human monocytes and murine peritoneal macrophages. Immunology 1986;59:333e8. Steele J, Flint KC, Pozniak AL, Hudspith B, Johnson MM, Rook GA. Inhibition of virulent Mycobacterium tuberculosis by murine peritoneal macrophages and human alveolar lavage cells: the effects of lymphokines and recombinant gamma interferon. Tubercle 1986;67:289e94. Denis M, Gregg EO, Ghandirian E. Cytokine modulation of Mycobacterium tuberculosis growth in human macrophages. Int J Immunopharmacol 1990;12:721e7. Hmama Z, Gabathuler R, Jefferies WA, de Jong G, Reiner NE. Attenuation of HLA-DR expression by mononuclear phagocytes infected with Mycobacterium tuberculosis is related to intracellular sequestration of immature class II heterodimers. J Immunol 1998;161:4882e93. Gehring AJ, Rojas RE, Canaday DH, Lakey DL, Harding CV, Boom WH. The Mycobacterium tuberculosis 19-kilodalton lipoprotein inhibits gamma interferon-regulated HLA-DR and Fc gamma R1 on human macrophages through toll-like receptor 2. Infect Immun 2003;71:4487e97. Kincaid EZ, Ernst JD. Mycobacterium tuberculosis exerts gene-selective inhibition of transcriptional responses to IFN-gamma without inhibiting STAT1 function. J Immunol 2003;171:2042e9. Wang Y, Curry HM, Zwilling BS, Lafuse WP. Mycobacteria inhibition of IFN-gamma induced HLA-DR gene expression by up-regulating histone deacetylation at the promoter region in human THP-1 monocytic cells. J Immunol 2005;174:5687e94. Ashida H, Mimuro H, Ogawa M, Kobayashi T, Sanada T, Kim M, Sasakawa C. Cell death and infection: a double-edged sword for host and pathogen survival. J Cell Biol 2011;195:931e42. Abebe M, Kim L, Rook G, Aseffa A, Wassie L, Zewdie M, Zumla A, Engers H, Andersen P, Doherty TM. Modulation of cell death by M. tuberculosis as a strategy for pathogen survival. Clin Dev Immunol 2011;2011:678570. Roca FJ, Ramakrishnan L. TNF dually mediates resistance and susceptibility to mycobacteria via mitochondrial reactive oxygen species. Cell 2013;153:521e34. Mangan DF, Welch GR, Wahl SM. Lipopolysaccharide, tumor necrosis factoralpha, and IL-1 beta prevent programmed cell death (apoptosis) in human peripheral blood monocytes. J Immunol 1991;146:1541e6. Perlman H, Pagliari LJ, Georganas C, Mano T, Walsh K, Pope RM. FLICEinhibitory protein expression during macrophage differentiation confers resistance to fas-mediated apoptosis. J Exp Med 1999;190:1679e88. Wong KW, Jacobs Jr WR. Critical role for NLRP3 in necrotic death triggered by Mycobacterium tuberculosis. Cell Microbiol 2011;13:1371e84. Duncan JA, Gao X, Huang MT, O’Connor BP, Thomas CE, Willingham SB, Bergstralh DT, Jarvis GA, Sparling PF, Ting JP. Neisseria gonorrhoeae activates the proteinase cathepsin B to mediate the signaling activities of the NLRP3 and ASC-containing inflammasome. J Immunol 2009;182:6460e9. von Moltke J, Ayres JS, Kofoed EM, Chavarria-Smith J, Vance RE. Recognition of bacteria by inflammasomes. Annu Rev Immunol 2013;31:73e106. Cooper AM, Mayer-Barber KD, Sher A. Role of innate cytokines in mycobacterial infection. Mucosal Immunol 2011;4:252e60. Fels AO, Cohn ZA. The alveolar macrophage. J Appl Physiol 1986;60:353e69. Valdivia-Arenas M, Amer A, Henning L, Wewers M, Schlesinger L. Lung infections and innate host defense. Drug Discov Today Dis Mech 2007;4:73e81. Tomlinson GS, Booth H, Petit SJ, Potton E, Towers GJ, Miller RF, Chain BM, Noursadeghi M. Adherent human alveolar macrophages exhibit a transient pro-inflammatory profile that confounds responses to innate immune stimulation. PLoS One 2012;7:e40348. Park JS, Tamayo MH, Gonzalez-Juarrero M, Orme IM, Ordway DJ. Virulent clinical isolates of Mycobacterium tuberculosis grow rapidly and induce cellular necrosis but minimal apoptosis in murine macrophages. J Leukoc Biol 2006;79:80e6. Portevin D, Gagneux S, Comas I, Young D. Human macrophage responses to clinical isolates from the Mycobacterium tuberculosis complex discriminate between ancient and modern lineages. PLoS Pathog 2011;7:e1001307. Isaza JP, Duque C, Gomez V, Robledo J, Barrera LF, Alzate JF. Whole genome shotgun sequencing of one Colombian clinical isolate of Mycobacterium tuberculosis reveals DosR regulon gene deletions. FEMS Microbiol Lett 2012;330:113e20. Parish T, Smith DA, Kendall S, Casali N, Bancroft GJ, Stoker NG. Deletion of two-component regulatory systems increases the virulence of Mycobacterium tuberculosis. Infect Immun 2003;71:1134e40. Converse PJ, Karakousis PC, Klinkenberg LG, Kesavan AK, Ly LH, Allen SS, Grosset JH, Jain SK, Lamichhane G, Manabe YC, McMurray DN,

122

[44]

[45]

[46]

[47]

[48]

C. Duque et al. / Tuberculosis 94 (2014) 111e122 Nuermberger EL, Bishai WR. Role of the dosR-dosS two-component regulatory system in Mycobacterium tuberculosis virulence in three animal models. Infect Immun 2009;77:1230e7. Majumdar SD, Vashist A, Dhingra S, Gupta R, Singh A, Challu VK, Ramanathan VD, Kumar P, Tyagi JS. Appropriate DevR (DosR)-mediated signaling determines transcriptional response, hypoxic viability and virulence of Mycobacterium tuberculosis. PLoS One 2012;7:e35847. Malhotra V, Sharma D, Ramanathan VD, Shakila H, Saini DK, Chakravorty S, Das TK, Li Q, Silver RF, Narayanan PR, Tyagi JS. Disruption of response regulator gene, devR, leads to attenuation in virulence of Mycobacterium tuberculosis. FEMS Microbiol Lett 2004;231:237e45. Sassetti CM, Boyd DH, Rubin EJ. Genes required for mycobacterial growth defined by high density mutagenesis. Mol Microbiol 2003;48: 77e84. Zelenin A. Acridine orange as a probe for cell and molecular biology. In: Manson W, editor. Fluorescent and luminiscent probes for biological activity. San Diego: Academic Press; 1999. pp. 117e35. Lecoeur H, Gougeon ML. Comparative analysis of flow cytometric methods for apoptosis quantitation in murine thymocytes and human peripheral

[49]

[50] [51]

[52]

lymphocytes from controls and HIV-infected persons. Evidence for interference by granulocytes and erythrocytes. J Immunol Methods 1996;198:87e99. Leite M, Quinta-Costa M, Leite PS, Guimaraes JE. Critical evaluation of techniques to detect and measure cell death-study in a model of UV radiation of the leukaemic cell line HL60. Anal Cell Pathol 1999;19:139e51. Ribble D, Goldstein NB, Norris DA, Shellman YG. A simple technique for quantifying apoptosis in 96-well plates. BMC Biotechnol 2005;5:12. Baskic D, Popovic S, Ristic P, Arsenijevic NN. Analysis of cycloheximideinduced apoptosis in human leukocytes: fluorescence microscopy using annexin V/propidium iodide versus acridin orange/ethidium bromide. Cell Biol Int 2006;30:924e32. Galluzzi L, Vitale I, Abrams JM, Alnemri ES, Baehrecke EH, Blagosklonny MV, Dawson TM, Dawson VL, El-Deiry WS, Fulda S, Gottlieb E, Green DR, Hengartner MO, Kepp O, Knight RA, Kumar S, Lipton SA, Lu X, Madeo F, Malorni W, Mehlen P, Nunez G, Peter ME, Piacentini M, Rubinsztein DC, Shi Y, Simon HU, Vandenabeele P, White E, Yuan J, Zhivotovsky B, Melino G, Kroemer G. Molecular definitions of cell death subroutines: recommendations of the Nomenclature Committee on Cell Death 2012. Cell Death Differ 2012;109:107e20.