Increased frequency and cell death of CD16+ monocytes with Mycobacterium tuberculosis infection

Tuberculosis 91 (2011) 348e360

Contents lists available at ScienceDirect

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

IMMUNOLOGICAL ASPECTS

Increased frequency and cell death of CD16þ monocytes with Mycobacterium tuberculosis infection Diana Castaño a, c, Luis F. García a, c, Mauricio Rojas a, b, c, * a

Grupo de Inmunología Celular e Inmunogenética, Instituto de Investigaciones Médicas, Facultad de Medicina, Universidad de Antioquia, Medellín, Colombia Unidad de Citometría de Flujo, Sede de Investigación Universitaria, Universidad de Antioquia, Medellín, Colombia c Centro Colombiano de Investigación en Tuberculosis, Medellín, Colombia b

a r t i c l e i n f o

s u m m a r y

Article history: Received 2 June 2010 Received in revised form 25 March 2011 Accepted 9 April 2011

Monocytes from tuberculosis patients exhibit functional and phenotypical alterations compared with healthy controls. To determine whether these discrepancies can be explained by changes in monocyte subsets, the expression of CD14 and CD16 was evaluated in tuberculosis patients and healthy controls; additionally, some markers related to the mononuclear phagocytes maturation, differentiation and function, such as CD1a, CD1c, CD11b, CD11c, CD13, CD33, CD36, CD40, CD64, CD68, CD80, CD83, CD86, HLA-DR, CCR2, CCR5, and non-specific esterases (NSE) were determined in monocyte subsets. Patients had increased percentage of circulating CD14HiCD16þ and CD14LoCD16þ monocytes. The percentage of monocytes expressing CD11b, CD36, CD64, CD68, CD80, CD86, CCR2 and NSE was lower in CD14HiCD16þ and CD14LoCD16þ cells than in CD14HiCD16 monocytes. M. tuberculosis infected CD16þ monocytes produced more TNF-a and less IL-10 than CD16 cells at 6 h post-infection. Isolated CD16þ monocytes spontaneously underwent apoptosis during differentiation into macrophages; in contrast to CD16 monocytes that became differentiated into monocyte-derived macrophages (MDM) with a minimal induction of cell death. In addition, there were more Annexin V and propidium iodide positive monocytes in the CD16þ subset infected with live M. tuberculosis at 24 h than CD16 monocytes. Under the culture conditions established for this study, the monocyte subsets did not differentiate into dendritic cells. These results show that tuberculosis patients have an augmented frequency of CD16þ circulating monocytes which are more prone to produce TNF-a and to undergo cell death in response to M. tuberculosis infection. Ó 2011 Elsevier Ltd. All rights reserved.

Keywords: Mycobacterium tuberculosis Tuberculosis Mononuclear phagocyte Monocyte Macrophage Monocyte subsets

1. Introduction Macrophages are considered a key component of the defense against Mycobacterium tuberculosis. In naturally resistant individuals and in non-resistant individuals infected with low doses of M. tuberculosis, macrophages may control bacilli replication by activating innate bactericidal mechanisms.1,2,3 However, in susceptible individuals or in those exposed to a high bacterial dose of virulent mycobacteria, mononuclear phagocytes may fail to control the bacillus, resulting in mycobacterial survival and replication.4 M. tuberculosis is able to disturb different processes within the host cells, such as the phagosome and lysosome fusion, to modulate different signaling pathways preventing induction of

* Corresponding author. Sede de Investigación Universitaria, Universidad de Antioquia, Carrera 53 No 61 e 30, Laboratorio 420, Medellín, Colombia. Tel./fax: þ57 4 219 6463. E-mail address: [email protected] (M. Rojas). 1472-9792/$ e see front matter Ó 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.tube.2011.04.002

bactericidal mechanisms, and to induce necrosis, rather than apoptosis of infected cells.5 We have previously reported that in response to mycobacterial infection or stimulation with purified protein derivate (PPD), monocytes from tuberculosis (TB) patients undergo apoptosis and necrosis, while monocytes from healthy controls display mainly apoptosis.6 Apoptosis has been associated with M. tuberculosis control since cell corpses containing the bacilli are recognized and phagocyted by newly recruited macrophages and dendritic cells enhancing the anti-mycobacterial effector mechanisms.7,8 Contrariwise, necrosis allows bacterial dissemination and tissue damage.9,10 However, within the granuloma, infected apoptotic cells may be phagocytozed by cells with lower anti-mycobacterial capacity, resulting in dissemination rather than in control of the infection.11,12 Our group had previously reported that patients with different clinical forms of TB have a higher proportion of circulating CD14þ monocytes and these cells have a decreased expression of HLA-DR and CD36.13 These alterations were reversed by anti-TB

D. Castaño et al. / Tuberculosis 91 (2011) 348e360

treatment,13 suggesting that the differences in monocytes between TB patients and healthy controls could be explained by the systemic effects of active TB.13 Thus, it is possible that differences between monocytes from TB patients and healthy controls can be the result of changes in the subpopulations of circulating monocytes or, alternatively, of the stage of maturation at which these cells enter into circulation. The myelo-monocytic progenitor cells mature in the bone marrow and enter into circulation as CD14þ HLA class IIþ cells, where they remain for about 3 days before migrating into the tissues where they differentiate into macrophages.14 Human monocyte subpopulations have been defined according to the expression of CD14 and CD16,15 in CD14HiCD16, CD14HiCD16þ and CD14LoCD16þ,15e17 which may represent different stages of maturation.18,19,20 The CD14HiCD16þ and CD14LoCD16þ subsets have been found increased in inflammatory conditions such as asthma,21 atherosclerosis,22 cardiovascular events,23 rheumatoid arthritis,24 cancer25 among others. In TB patients, it has been reported that CD14þ monocytes have an increased expression of CD16 compared to healthy controls.26 Previous reports indicated that CD16þ monocytes exhibit characteristics of dendritic cell because the high density of expression of CD11c, CD86 and HLA-DR.27,28 Additionally, monocytes treated with GM-CSF, IL-4 and IL-10 differentiated into dendritic-like cells expressing CD14, CD16, CD1a and CD83.29,30 However, it is not completely understood whether CD16þ and CD16 human monocytes differentiate into macrophages and how these cells respond to M. tuberculosis infection. To further understand the role of the different monocyte subpopulations in TB patients, we compared the expression of markers related with maturation, differentiation and function in different monocyte subpopuplations of TB patients and healthy controls. The results show that patients with pulmonary TB have a higher percentage of circulating CD14HiCD16þ and CD14LoCD16þ monocytes. In addition, CD14HiCD16þ and CD14LoCD16þ monocytes had decreased percentage of CD11b, CD36, CD64, CD68, CD80, CD86, CCR2 and NSE positive cells, compared with CD14HiCD16 monocytes. In vitro, CD16þ monocytes did not differentiate into macrophages. CD16þ monocytes infected with live M. tuberculosis produced more TNF-a and less IL-10 at 6 h compared with CD16 cells. There were more Annexin V and propidium iodide positive monocytes in the CD16þ subset infected with live M. tuberculosis for 24 h than in CD16 monocytes. These observations indicate that TB patients have increased amounts of circulating CD16þ monocytes, which are more prone to produce TNF-a and die in response to M. tuberculosis infection. 2. Materials and methods 2.1. Reagents and antibodies HistopaqueÒe1077, trypan blue, sodium azide, saponin and bovine serum albumin (BSA) were obtained from Sigma Aldrich (St. Louis, MO). Paraformaldehyde (PFA) from Fisher Scientific (Pittsburgh, PA); 2 mm carboxylate-modified microspheres yellowgreen fluorescent (FluoSpheresÒ) and fluorescein diacetate (FDA) were purchased from Invitrogen (Eugene, OR). Milddlebrook 7H10 solid medium, Milddlebrook 7H9 liquid medium and Oleic Acid Albumin Dextrose Catalase Complex (OADC) were obtained from Becton Dickinson (Microbiology Systems, Cockeysville, MD). RPMI1640, phosphate-buffered saline (PBS) and fetal bovine serum (FBS) were purchased from Gibco-BRL (Grand Island, NY). Monoclonal anti-human HLA class II-FITC (DR, DP and DQ, clone Tü39), CD13-PE (clone WM15), CD15-APC (clone HI98), CD33-PE (clone WM53), CD64-PE (clone 10.1), CD40-FITC (clone 5C3),

349

CD80-PECy5 (clone L307.4), CD86-PECy5 (clone 2331), CD36-FITC (clone CB38), CD36-PE (clone CB38), CD40-PECy5 (clone 5C3), CD14-FITC (clone M5E2), CD14-RPE (clone M5E2), CD14-PE (clone M5E2), CD68-PE (clone Y1/82A), CD3-FITC (clone UCHT1), CD16FITC (clone 3G8), CD16-PE (clone 3G8), CD16-PECy5 (clone 3G8), CD16-Alexa Fluor 647 (clone 3G8), CCR2-Alexa Fluor 647 (clone 48607), CD19-PECy5 (clone HIB19), CD56-RPE (clone B159), CD 14-PerCP (clone M4P9), CD62L (clone Dreg-56), CD11c-PE (clone B-ly6), CD11c-PECy5 (clone B-ly6), HLA-DR PECy5 (clone L243), HLA-DR PERCP-Cy5.5 (clone L243), Lineage cocktail 1 (Lin 1 contains FITC-conjugated antibodies to CD3, CD14, CD16, CD19, CD20 and CD56), CD1a-PECy5 (clone HI149), CD83-PE (clone HB15e), isotype controls and Human Inflammatory Cytokine Kit of BDÔ Cytometric Bead Array (CBA) were purchased from Becton Dickinson (BD) Pharmingen (San Diego, CA). Anti-human CD11bFITC (clone VIM12) was obtained from CALTAG Laboratories (Burlingame, CA). Anti-human CCR5-FITC (clone CTC5) was purchased from R&D Systems (Minneapolis, MN). Anti-human CD14-PE (clone RMO52) was obtained from Immunotech, Beckman Coulter (Miami, FL); anti-human CD1c (BDCA-1, clone AD5-8E7) purchased from Miltenyi Biotec Inc (Bergisch Gladbach, Germany) was generously provided by Juan Carlos Hernandez (Grupo de Virología, Universidad de Antioquia, Medellín, Colombia). Annexin V-PE, Annexin V-FITC, Streptavidin-FITC, Annexin V binding buffer 10 and propidium iodide were obtained from BD Pharmingen for flow cytometry analysis. TACSÔ Annexin V kit for fluorescence microscopy was obtained from Trevigen, Inc. (Gaithersburg MD). Limulus Amebocyte Lysate (LAL) from Cambrex (Walkersville, MD). Purified Protein Derivative (PPD) used for tuberculin skin test (TST, RT23) and for in vitro culture (RT50) were purchased from Statens Serum Institute (Copenhagen, Denmark). Human CD2 microbeads and LD columns were purchased from Miltenyi Biotec. Acridine orange was obtained from Merck (San Diego, CA). 2.2. Culture of M. tuberculosis H37Rv Mycobacterium tuberculosis H37Rv was obtained from the Instituto Nacional de Salud (Bogotá, Colombia), and grown in Middlebrook 7H9 liquid media supplemented with OADC. Cultures were harvested after 3 weeks. Mycobacteria were extensibility washed with PBS and labeled or not with 250 ng/ml FDA (FDAM. tuberculosis) for 60 min at 37  C. FDA-M. tuberculosis was incubated with 10% FBS in RPMI-1640 for 1 h and washed again with PBS. A M. tuberculosis unlabeled batch were killed with 1.4% PFA for 18 h at room temperature. Dead bacteria were extensively washed again with PBS. Not labeled mycobacterias were resuspended in RPMI containing 20% glycerol and fluorescent labeled mycobacteria in PBS containing 20% glycerol. Mycobacteria were sonicated to disrupt clumps at 204,8 W, 4  C for 10 cycles of 10 s, and 15 s of rest on ice between each cycle (Sonics Vibra Cell, model CV33. Newtown, CT). Mycobacterial suspension was centrifuged 5 min at 200  g and the supernatant was aliquoted and frozen at 70  C. The mycobacterial concentration was calculated by spectrophotometry at 600 nm and verified by counting the number of colonyforming units (CFU). 2.3. Subjects studied Twenty-six TB patients, bacteriologically confirmed by ZielhleNeelsen staining of the sputum smear, were contacted through the Secretaría de Salud de Medellín and the TB Control Program of the Servicio Seccional de Salud de Antioquia and studied before or within the first 2 weeks of anti-TB treatment. The TB patients studied in different experiments were 31.8  15 years old

350

D. Castaño et al. / Tuberculosis 91 (2011) 348e360

and included 10 (38.5%) females and 16 (61.5%) males. As controls, 24 TST negative (TST) and 22 TSTþ individuals were selected from healthy laboratory personnel at Universidad de Antioquia and recruited for different experiments. The TST controls were 32.2  10 years old; 16 (67%) of them were females and 8 (33%) males. The TSTþ controls were 39.6  12 years old and included 12 (55%) females and 10 (45%) males. Tuberculin skin test was done by intradermal injection of 0.1 ml (2 U) of PPD. Indurations  10 mm measured at 48e72 h were considered positive. All participants were notified about the investigation and they read and signed an informed consent previously approved by the Ethics Committee of the Instituto de Investigaciones Médicas of the Facultad de Medicina, Universidad de Antioquia. HIV-positive individuals, subjects with secondary or primary immunodeficiencies, cancer or autoimmune diseases were excluded from this study. 2.4. Phenotype of mononuclear phagocytes Peripheral blood mononuclear cells (PBMC) were obtained from 10 ml of heparinized venous blood samples, by centrifugation on Histopaque for 30 min at 900  g, room temperature. Cells were washed with PBS and washing buffer (PBS plus 1% BSA and 0.1% NaN3) and resuspended in staining buffer (PBS plus 1% pooled human serum ePHS-). Viability was determined by exclusion of trypan blue (98%). Five hundred thousand cells were stained with antibodies against HLA class II, HLA-DR, CD1a, CD1c, CD11b, CD11c, CD13, CD14, CD15, CD16, CD33, CD36, CD40, CD64, CD80, CD83, CD86, CCR2, CCR5 and isotype control antibodies. Cells were incubated 30 min at room temperature and two additional washes were made with washing buffer. The intracellular staining of CD68 was made in cells fixed with 2% PFA for 20 min, followed by permeabilization with 0.1% saponin plus 1% BSA and anti-CD68 or isotype control, for 30 min at room temperature. Cells were washed twice and 5  104 cells were acquired in a BD FACSort or FACS Canto II (Becton Dickinson Biosciences. San Diego, CA). The percentage of stained cells and the mean fluorescence intensity (MFI) were estimated using the Cell Quest software (version 3.3. BD) or FlowJo 7.6.1 software (Tree Star, Inc. Ashland, OR). Adherent cells were collected at different time points during differentiation as described below. Three hundred thousand cells were stained with anti-HLA class II (HLA-II) and CD14, plus either CD16, CD68, CD86, CD36, CD40, CD62L, CCR2 or the matched isotype antibodies as described above. Cells were fixed with 2% PFA for 20 min, washed and scraped with a rubber policeman. Ten thousand events were acquired in a FACSort and analyzed as described before.

The efficiency in this separation was greater than 80%. Thereafter monocytes were allowed to adhere and differentiate as described below. At 120 h, morphological changes, the expression of different molecules and cell death were determined in vitro. In some cases, monolayers of 24 h of differentiation were infected with M. tuberculosis and the cell death and cytokine production were evaluated after 6 h and 24 h of infection. 2.7. Culture of mononuclear phagocytes and differentiation PBMC were isolated from defibrinated venous blood samples of healthy individuals. PBMC containing 2.5  105 CD14þ cells were plated in 48-well plates (Corning Incorporated Life Science, Lowell, MA) using 1 ml of RPMI-1640 plus 0.5% PHS, for 4 h at 37  C. Then, wells were extensibility washed with pre-warmed PBS plus 0.5% FBS to remove non-adherent cells. Adherent cells were cultured in 1 ml of RPMI-1640 supplemented with 10% PHS for 120 h to allow differentiation into monocyte-derived macrophages (MDM). Before differentiation, more than 85% correspond to monocytes and the remaining cells were CD19þ, CD3þ and CD56þCD3. Adherent cells were enumerated by scraping or lysing to count cells and nuclei, respectively, as described.32 There were not changes in the number of adherent cells during 120 h of culture. In some experiments, sorted CD16 and CD16þ cells were cultured under the described conditions to allow the differentiation into MDM. At the end of the culture there were not Lin DRþ CD11cHi cells, Lin DRþ CD1aþ, Lin DRþ CD1cþ, and Lin, DRþ, CD83þ, indicating that cultured cells did not differentiate into dendritic myeloid cells. 2.8. Phagocytosis assay Sorted CD16 and CD16þ monocytes differentiated for 24 h were infected with FDA-M. tuberculosis at a multiplicity of infection (MOI) of 5:1, or incubated with fluorescent latex beads at a ratio of 5:1. Cells were centrifuged for 5 min at 900  g and incubated for 2 h at 37  C 5% CO2; then the cells were extensively washed with PBS and analyzed by epifluorescence microscopy (Eclipse TS-100. Nikon Corporation. Tokyo, Japan). More than 100 cells were counted in at least five different fields derived from four healthy donors. The results of the phagocytosis assay are shown as the percentage of cells which phagocytosed and as a phagocytic index. The percentage of phagocytosing cells corresponds to the proportion of cells associated (positive) to either FDA-M. tuberculosis or latex beads in each field evaluated. The phagocytic index corresponds to the number of associated beads divided by the number of evaluated cells.

2.5. NSE activity

2.9. M. tuberculosis infection and PPD treatment

FDA hydrolysis by non-specific esterases turned it fluorescent.31 Thus, the NSE activity was determined on monocyte subpopulations based on FDA labeling. Five hundred thousand PBMC were treated with 0.24 mM FDA, for 20 min at room temperature in darkness. Cells were washed three times with staining buffer and stained with anti CD14-PE and CD16-PECy5 during 30 min. After two washes with PBS, cells were analyzed by flow cytometry.

Sorted CD16 and CD16þ monocytes were adhered as described in 2.7 (Culture of mononuclear phagocytes and differentiation). CD16 and CD16þ monocytes were cultured for 24 h in RPMI-1640 plus 10% PHS, and infected with live or dead M. tuberculosis at a MOI of 5 or were treated with 10 mg/ml PPD, centrifuged for 5 min at 900  g and incubated at 37  C 5% CO2 for 6 h. In some experiments CD16 and CD16þ monocytes were infected with M. tuberculosis at a MOI of 5 and 15 for 24 h. Thereafter, the cell death and cytokine production were measured.

2.6. Enrichment of CD16 and CD16þ monocytes One hundred thousand PBMC from healthy controls were magnetically depleted of T lymphocytes by CD2 microbeads on large depletion (LD) columns, according to manufacturer instructions. Collected cells were stained with anti-CD14 and anti-CD16 as described above and separated into CD16þ and CD16 cells by sorting (MoFloÔ XDP Cell Sorter from Beckman Coulter; Miami, FL).

2.10. Determination of cell death Cell death was determined directly on culture plates by simultaneous staining with 4 mg/ml Acridine Orange and 4 mg/ml Propidium Iodide (PI) in PBS for 5 min at room temperature in the dark.33 Cells were washed, resuspended in PBS and analyzed

D. Castaño et al. / Tuberculosis 91 (2011) 348e360

under the epifluorescence microscope. Cells with nuclear condensation and fragmentation were considered apoptotic and those that incorporated PI, necrotic. In parallel experiments, cell death was determined with TACSÔ Annexin V kit with or without PI according to manufacturer instructions. Briefly, adherent cells were washed once with cold PBS, staining in PBS plus Ca2þ for 25 min, at room temperature in the dark, washed twice and analyzed by either epifluorescence microscopy or flow cytometry. Annexin V labeling was considered as apoptosis, PI incorporation as necrosis and positive cells for Annexin V and PI corresponded to late apoptosis. 2.11. Cytokine production The production of IL-12p70, TNF-a, IL-10, IL-1b and IL-8 by CD16 and CD16þ monocytes, was evaluated in 50 ml of culture supernatants using the Human Inflammatory Cytokine Kit of BDÔ CBA. The production of IL-18 was determined by ELISA, according to manufacturer instructions (Medicals & Biological Laboratories Ltd. Naka-ku Nagoya, Japan). 2.12. Statistical analyses Comparisons between two groups were done by Mann Whitney Test and for three or more groups by KruskaleWallis test with the Dunn’s post test. Paired samples were compared using Wilcoxon signed-rank test. Two factor analyses were done by type II analysis of variance (ANOVA). Principal components analysis was performed to obtain the small number possible of linear combinations from the markers evaluated in the monocyte subpopulations. A p value 0.05 was considered statistically significant. For all analyses, GraphPad Prism 5 (GraphPad Software, Inc, La Jolla, CA) and Statistics Plus 4 software (Statpoint Technologies, Inc, Warrenton, VA) were used.

351

3. Results 3.1. TB patients have increased percentage of circulating CD14HiCD16þ and CD14LoCD16þ monocytes Previous reports indicated that TB patients have increased proportion of monocytes compared with healthy controls.13 To further study the differences in monocytes between TB patients and controls, the expression of CD16 was determined on HLA-IIþCD14þ monocytes (Figure 1A). TB patients had increased both the percentage and absolute number of circulating CD16þ monocytes compared with TSTþ and TST controls (data not shown). The expression of CD14 and CD16 has been used to define three monocyte subpopulations15; thus, to determine whether the increase in CD16 expression observed in TB patients is due to a particular subset of CD14þ cells, the percentage of CD14HiCD16, CD14HiCD16þ and CD14LoCD16þ monocytes was determined in freshly isolated PBMC HLA-IIþCD14þ (Figure 1A) from TB patients and healthy controls (Figure 1B and C). TB patients had less CD14HiCD16 monocytes (p < 0.05, Figure 1D), but a significant increase in the percentage of CD14HiCD16þ (p < 0.001) and CD16 MFI (p < 0.001) on these cells compared with TST and TSTþ healthy controls (Figure 1E). Additionally, there was an augmented percentage of CD14LoCD16þ in TB patients compared with TST but not with TSTþ controls (Figure 1F); CD16 MFI was also increased in CD14LoCD16þ monocytes compared with TST and TSTþ healthy controls (Figure 1F). 3.2. CD14HiCD16þ and CD14LoCD16þ monocytes have a decreased percentage of CD68, CD64, CD36, CD11b, CD80, CD86, CCR2 and non-specific esterases activity To further characterize the expression of markers related with maturation, differentiation and function of mononuclear phagocytes in the studied monocyte subsets, the expression of CD1a,

Figure 1. Monocyte subsets studied in TB patients (TBP), TST and TSTþ controls. A. Cytometric analysis was done in the gate of mononuclear cells and the region of HLA class II (HLA-II)þ and CD14þ cells. B. Representative dot plot within this region from a healthy TST control and C. TB patient. D. Percentage of CD14HiCD16 monocytes (region 1), E. percentage and mean fluorescence intensity (MFI) of CD16 on CD14HiCD16þ (region 2) and F. CD14LoCD16þ monocytes (region 3) from TB patients, TST and TSTþ controls. Groups were compared using the KruskaleWallis test.

352

D. Castaño et al. / Tuberculosis 91 (2011) 348e360

Table 1 Expression of different markers related with maturation, differentiation and function in monocytes subsets. Marker

CD11b CD11c CD13 CD33 CD36 CD40 CD64 CD68 CD80 CD86 HLA-DR NSE CCR2 CCR5

CD14HiCD16

CD14HiCD16þ

CD14LoCD16þ

TB Patients

TSTþ Controls

TB Patients

TSTþ Controls

TB Patients

TSTþ Controls

85  11*,c (3685  2172)y 99  1.4 (9996  1861)c,f 99  0.7 (34,667  11,479) 99  0.7 (14,820  3767)f 99  0.7i (35,243  11,693)g 81  24 (1224  1853) 97  3i (1973  513) 79  3c,i (667  61)d 35  10c (500  74) 69  14c (458  98)e 85  3 (951  212)i 97  0.8c,f (15,829  5907) 52  21f (1548  316) 52  25f (525  35)g

91  6g (2885  887) 99  0.3 (9175  1583)d,g 97  3 (19,726  15,401) 99  0.5 (12,721  6070)g 99  0.4f (31,917  2468)h 81  21 (498  190)c 97  4j (1649  424)c 79  5f (623  158) 30  16g (512  45) 62  7d (414  62)f 84  5f (786  279)j 94  3g (12,679  4712) 44  14g (1391  122) 55  17 (473  76)i

69  23 (3739  2438) 98  0.9 (19,217  3932)c 98  0.9 (39,950  17,681) 98  1.0 (11,578  3008) 90  2 (30,314  11,558) 90  13 (1548  1632) 73  7 (2042  559) 65  6c (820  77)d 23  14 (613  41) 58  15 (756  367)e 94  2g (4585  942)i 81  5c (20,229  5037) 29  11 (1537  364) 48  28 (668  98)

75  21 (2687  724) 99  0.7 (14,860  3254)g 97  3 (21,899  18,385) 99  0.6 (10,306  4813) 90  2 (30,300  2451)d 84  20 (826  386) 72  11 (1543  346) 60  10 (754  199) 13  9 (661  190) 46  7 (582  97) 94  3f (3346  1023)j 80  2 (15,585  4839) 23  13 (1686  363) 59  24c (640  140)

43  32c,k (7868  3804) 93  7 (16,967  3340)f 92  9 (19,817  17,381) 90  11.5 (4038  1212)f 72  13i (14,364  8524)g 68  25 (1256  1050) 23  16i (1542  1290) 43  23i (785  110) 3.5  1.6c,l (N.D.)x 43  5c (514  112) 74  11g,l (2520  1210) 69  12f (21,843  5183) 2.3  1.8f (N.D.) 18  11f (1097  743)g

15  25g,k (4345  3845) 98  1.1 (13,733  2189)d 92  10 (11,791  8486) 96  3 (3940  1619)g 84  16f (18,050  2017)d,h 56  35 (970  299)c 22  17j (919  130)c 43  19f (790  158) 1.0  0.4g,l (N.D.) 42  16d (746  250)f 89  3l (1894  600) 56  16g (16,950  3251) 1.2  1.2g (N.D.) 23  20c (1153  640)i

The monocyte subsets were compared using KruskaleWallis test. n ¼ 5e7. c, d, e 0.05; f, g, h 0,01; TB patients and TSTþ controls were compared using Mann Whitney. n ¼ 5e7. k 0.05 and l 0.01. * Percentage of positive cells, mean  SD. y Mean fluorescence intensity (MFI, inside parenthesis)  SD. x The MFI were not determined (N.D.) when the percentage of positive cells was below than 5%.

CD1c, CD11b, CD11c, CD13, CD33, CD36, CD40, CD64, CD68, CD83, CD80, CD86, HLA-DR, NSE activity, CCR2 and CCR5, was evaluated on CD14HiCD16, CD14HiCD16þ and CD14LoCD16þ monocytes from TB patients and healthy individuals (Table 1). The data was analyzed in two ways: first, to establish a characteristic phenotype for every subset, the expression of each molecule evaluated was compared among CD14HiCD16, CD14HiCD16þ and CD14LoCD16þ monocytes (Table 1); second, to clarify whether the phenotype of a subset varies according to the study groups, the expression of each molecule was compared among TB patients, TSTþ (Table 1) and TST controls (data not shown). Because there were not differences in the percentage and the MFI of the evaluated markers between TSTþ and TST controls (data not shown), only the results for TSTþ controls are presented in Table 1. The comparison among monocyte subsets showed that CD14HiCD16 cells exhibited a higher percentage of CD11b, CD64, NSE, CD36, CD68, CD80, CD86, CCR2 and CCR5 positive cells compared to CD14LoCD16þ monocytes (p values are shown in Table 1). The CD14HiCD16þ cells had an intermediate percentage of expression of these markers compared with CD14HiCD16 and CD14LoCD16þ monocytes (Table 1). In contrast, CD14HiCD16 monocytes displayed lower MFI for CD68, CD86 and CCR5 compared to CD14HiCD16þ and CD14LoCD16þ cells. The expression of CD36 in CD14HiCD16 and CD14HiCD16þ subsets was higher than in CD14LoCD16þ monocytes (Table 1). Although the percentage of CD11cþ cells were similar among the three monocyte subpolulations, CD14HiCD16 cells had decreased MFI compared with CD14HiCD16þ and CD14LoCD16þ subpopulations (Table 1). The percentage of CD13 and CD33 positive cells was similar between CD14HiCD16 and CD14HiCD16þ monocytes and slightly decreased in CD14LoCD16þ. The MFI of CD33 was decreased in CD14LoCD16þ subset compared with CD14HiCD16 cells (Table 1).

i, j

0.001.

The CD14HiCD16þ cells had an increased percentage of HLA-DR and higher MFI of this molecule compared with CD14HiCD16 and CD14LoCD16þ monocytes (Table 1). There were less than 3% of CD1a, CD1c and CD83 positive cells in the CD14HiCD16, CD14HiCD16þ and CD14LoCD16þ subsets and there were not differences in these markers among the monocyte subsets (data not shown). Previously, it was reported the presence of lowdensity granulocytes with low CD14 and high CD15 expression in mononuclear cell fraction.34 There were not detected a high expression of CD15 in the monocyte subsets (data not shown); suggesting that CD14HiCD16, CD14HiCD16þ and CD14LoCD16þ cells evaluated in this study are not low-density granulocytes. To establish whether each monocyte subpopulations may be characterized by the expression of a group of markers, a principal component analysis was performed with the percentage of cell expressing CD11b, CD64, NSE, CD36, CD68, HLA-DR, CD40, CD80, CD86, CCR2 and CCR5 in the different monocyte subpopulations from TB patients, TST and TSTþ controls (data not shown). CD1a, CD1c, CD11c, CD13, CD33 and CD83 were not included in the analysis, because there were not significant differences in their percentage among monocyte subpopulations. This analysis clearly clustered the data in three groups that defined the monocyte subpopulations but did not distinguish monocytes from TB patients and healthy controls (data not shown). The CD14HiCD16 cells showed a high proportion of positive cells for CD11b, CD64, NSE, CD36, CD68, CD80, CD86, CCR5 and CCR2; while CD14HiCD16þ showed an intermediate and CD14LoCD16þ a low proportion of positive cells for these markers (Table 1 and data not shown). The CD14HiCD16þ monocytes showed the highest percentages of HLA-DR and CD40 compared with CD14HiCD16 and CD14LoCD16þ cells (Table 1 and data not shown). The principal component analysis of the MFI values did not show a clear separation of monocyte subsets (data not shown).

D. Castaño et al. / Tuberculosis 91 (2011) 348e360

Comparing the study groups, it was observed that the expression of the markers related with maturation, differentiation and function in the monocyte subsets were almost similar between TB patients and TSTþ controls; except for an increased percentage of CD11b in CD14LoCD16þ monocytes of TB patients compared with TSTþ controls (Table 1). In addition, there was a lower percentage of HLA-DRþ cells in CD14LoCD16þ monocytes of TB patients compared with TSTþ controls (Table 1). 3.3. Expression of HLA-II, CD86, CD68 and CD36 increases during differentiation of monocytes To evaluate the differentiation of CD16 and CD16þ monocytes into macrophages, phenotypical changes (morphology and glycoprotein expression) occurring during the differentiation of total monocytes into macrophages were analyzed. As previously

353

reported by other investigators,35,36 mononuclear phagocytes underwent morphological and phenotypical changes during their in vitro differentiation. During the first 24 h of culture, cells were rounded and with few cytoplasmic granules. At 120 h of culture they became elongated, with cytoplasmatic projections, a high number of cytoplasmic granules, and increased cytoplasmic/ nuclear ratio (Figure 2A and B). Importantly, more than 94% of the adherent cells remained viable until 120 h of culture, as demonstrated with acridine orange/PI (Figure 2C) and Annexin V/PI stainings (data not shown). To closely study the changes occurring during monocyte differentiation into macrophages, the expression of CD14, HLA-II, CD16, CD86, CD40, CD68, CD36, CCR2 and CD62L, was also determined in cell monolayers from healthy controls at different time points during in vitro differentiation. After removal of non-adherent cells (time zero), approximately 75  8.0% of adhered cells

Figure 2. Morphological characteristics of mononuclear phagocytes during in vitro differentiation. Cells were analyzed by A. flow cytometry, B. light and C. fluorescent microscopy (40 magnification) at 0 h, 24 h and 120 h of in vitro culture. A. The forward (FSC) and side (SSC) scatter graphs are shown of one representative assay out of five independent experiments. B. The morphological appearance of cells as evaluated by light microscopy and C. fluorescent microscopy of Acridine Orange and Propidium Iodide (IP) stained cells. These pictures are representative of more than 6000 cells evaluated from five different healthy individuals.

354

D. Castaño et al. / Tuberculosis 91 (2011) 348e360

expressed CD14 (Figure 3A). However, there was a decrease in the MFI of this molecule at that time compared with freshly isolated monocytes. At 24 h of culture, there was a reduction in the percentage of CD14þ cells (p < 0.05). At 120 h of culture, the percentage of CD14þ cells was similar to 24 h, but there was an increase in MFI (p < 0.05). Most of adherent cells were HLA-IIþ and this percentage did not change during the culture (Figure 3B); however, HLA-II expression (MFI) steadily increased during the culture time (p < 0.05). The expression of CD16þ decreased with adherence and did not significantly increase until the 120 h of culture (Figure 3C). At that time the percentage of CD16þ cells was higher than the observed at 0 h (p < 0.05). CD40þ cells were barely detectable after adherence (p < 0.05), but they were observed again at 24 and 120 h of culture, accompanied by a non-significant increase in CD40 MFI at 120 h (Figure 3D). The percentage of CD86þ cells slightly decreased at 24 h, but its expression per cell also increased during the culture (p < 0.05, Figure 3E). The percentage of CD36 and its MFI augmented at 120 h of culture compared with 0 h (p < 0.05, Figure 3F). There was a significant increase in the percentage of CD68þ cells and its MFI with the in vitro differentiation (p < 0.05, Figure 3G). The expression of CCR2 and CD62L diminished significantly with monocyte adherence and stayed so until 120 h (data not shown). 3.4. CD16þ monocytes did not differentiate in vitro into MDM The low percentage of CD16þ monocytes expressing markers related with maturation, differentiation and function of monocytes compared with CD14HiCD16 cells (see Table 1), led us to evaluate whether CD16þ and CD16 monocytes differ in their capacity to differentiate into macrophages. CD16 and CD16þ monocytes were enriched by cell sorting (Figure 4A) and differentiated in vitro as described in materials and methods. At 120 h, CD16þ cells were rounded, with few cytoplasmic granules and smaller than CD16 cells (Figure 4B) and the expression of HLA-II (p ¼ 0.014), CD86 (p ¼ 0.05) and CD36 (p ¼ 0.05) was significantly lower on CD16þ than on CD16 mononuclear phagocytes (Figure 4C). Additionally, evaluation of cell death at 120 h of culture revealed that most of the CD16þ cells were late apoptotic (Table 2). These results show that CD16þ monocytes under our experimental conditions do not differentiate into macrophages as it happened for CD16 monocytes. 3.5. Monocytes subsets differ in their phagocytic capabilities and live M. tuberculosis infection induces late apoptosis of CD16þ monocytes The phagocytic capabilities of sorted CD16 and CD16þ monocytes were determined using FDA-labeled M. tuberculosis and latex beads by epifluorescence microscopy. There were lower percentages of CD16þ monocytes able to bind and internalize FDAM. tuberculosis and latex beads compared with CD16 cells (Figure 4D). In contrast, CD16þ monocytes had higher number of beads per cell than CD16 monocytes (Figure 4D, lower left). This assay did not allow to precisely determining the number of FDAM. tuberculosis associated with monocytes. However, CD16þ

Figure 3. Expression of CD14, HLA class II, CD16, CD40, CD86, CD36 and CD68 during differentiation of monocytes at 0, 24 and 120 h of culture as determined by flow cytometry. Percentage (left panels) and MFI (right panels) of A. CD14, B. HLA class II, C. CD16, D. CD40, E. CD86, F. CD36 and G. CD68 positive cells are shown. Results were compared using the KruskaleWallis test with Dunn’s post test (n ¼ 3e4). P, PBMC. M, CD14þ monocytes.

D. Castaño et al. / Tuberculosis 91 (2011) 348e360

355

Figure 4. In vitro differentiation of CD16 and CD16þ monocytes. CD16 and CD16þ monocytes were obtained from PBMC of healthy controls by cell sorting A. Representative dot plots from unsorted (top) and CD16 (below, left) and CD16þ (below, right) sorted monocytes. B. CD16þ and CD16 monocytes analyzed by light microscope (40 magnification) at 120 h of in vitro culture. One representative assay out of four independent experiments. C. Expression of HLA-II, CD86 and CD36 on CD16 and CD16þ monocytes at 120 h of culture determined by flow cytometry. Comparisons were made by the Wilcoxon test. D. Percentage of CD16 and CD16þ monocytes phagocytosing fluorescent latex beads (top, left) and FDA-labeled M. tuberculosis (top, right). Phagocytic index of CD16 and CD16þ monocytes incubated with latex beads (below, left). Representative pictures of CD16 and CD16þ monocytes at 24 h of differentiation that bound and internalized latex beads (40 magnification. Below, right). The percentage of phagocytic cells and the phagocytic index data were obtained of more than 4500 cells evaluated from four different healthy individuals.

monocytes were brighter than CD16 after FDA-M. tuberculosis infection (data not shown). Since, it was not possible to achieve a normalization of M. tuberculosis infection between CD16 and CD16þ monocytes, the same MOI was used with both types of monocytes in the following experiments. Cell death was evaluated by Annexin V/PI (Figure 5A) and acridine orange/PI (data not shown) at 6 h of M. tuberculosis infection or PPD treatment. CD16þ monocytes, compared with CD16 cells, had tendency to increased cell death; however, the difference was not statistically significant (Figure 5B). Cell death was also evaluated in sorted monocytes after infection with live M. tuberculosis at a MOI of 5:1 and 15:1 for 24 h. CD16þ exhibited a higher number of cells positive to Annexin V and PI than CD16 cells in response to the infection (p < 0.0001, Figure 5C). There were a 20% of CD16þ

monocytes Annexin Vþ PIþ in a basal condition, more than 50% with a MOI of 5:1 and more than 80% with a MOI of 15:1 (Figure 5C). In CD16 monocytes, there was less than 6% of Annexin Vþ PIþ in untreated cells, less than 10% with a MOI of 5:1 and a 25% with a MOI of 15:1 (Figure 5C). 3.6. Live M. tuberculosis infection induces high production of TNF-a and low levels of IL-10 in CD16þ monocytes Low levels of IL-12p70 were detected in the supernatants from sorted CD16 and CD16þ cells but they did not change with the infection or PPD treatment (data not shown). CD16þ monocytes produced significantly more TNF-a in response to live M. tuberculosis than CD16 (p  0.001, Figure 6A). In contrast, CD16þ

Table 2 Evaluation of cell death in CD16 and CD16þ monocytes during differentiation into macrophages. CD16*

AP** AO**

CD16þ*

Viable

Apoptosis

Late Apoptosis

Necrosis

Viable

Apoptosis

Late Apoptosis

Necrosis

94.3%y,c  1.4 96.2%c  2.2

2.0%  0.8 1.5%  2.2

2.9%e  2.4 N.D.x

0.8%  1.4 2.3%f  0.3

33.2%c  15.8 45.7%c  12.8

10.1%  3.5 9.6%  6.4

55.7%e  17.0 N.D.x

1.0%  1.9 44.7%f  8.7

CD16 Vs CD16þ monocytes p < 0.001 using Two-way ANOVA. In each technique, more than 1200 cells were analyzed from 4 healthy individuals. Measurements done at 120 h of culture. y Mean  SD. x Late apoptotic cells were not evaluated (N.D.) by this technique. ** AO: Acridin Orange/Propidium Iodide, AP: Annexin V/Propidium Iodide.

c,e,f *

356

D. Castaño et al. / Tuberculosis 91 (2011) 348e360

Figure 5. Cell death of CD16 and CD16þ monocytes after 6 h and 24 h of M. tuberculosis (Mtb) infection or PPD treatment. Sorted CD16 and CD16þ monocytes differentiated for 24 h were infected with live or dead M. tuberculosis at a MOI of 5, or treated with 10 mg/ml of PPD. A. Representative pictures of Annexin V (green) labeling considered as apoptosis, PI incorporation (red) considered as necrosis and the Annexin V plus PI labeling correspond to late apoptosis. Viable monocytes correspond to negative cells for Annexin V and PI staining. B. Percentage of cell death at 6 h of M. tuberculosis infection or PPD treatment. Comparisons were done with Two-way ANOVA. The cell death data are shown as the mean  SD and were obtained of more than 10000 cells evaluated from five different healthy individuals. C. Cell death of CD16 and CD16þ monocytes after 24 h of M. tuberculosis infection at MOI of 5 and 15. Comparisons were done with Two-way ANOVA (n ¼ 3). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

monocytes produced significantly less TNF-a against dead M. tuberculosis (p  0.001) and there were not differences in response to PPD compared to CD16 monocytes (Figure 6A). CD16 monocytes released higher levels of IL-10 after live and dead M. tuberculosis infection than CD16þ cells (p  0.001), which did not show detectable IL-10 production after infection or PPD treatment (Figure 6B). CD16 and CD16þ cells produced similar levels of IL-1b in response to M. tuberculosis (Figure 6C). CD16þ monocytes produced less IL-8 in response to dead M. tuberculosis than CD16 cells (p  0.001, Figure 6D). 4. Discussion It is well known that TB patients have alterations in their immune response, including monocyte phenotypic and functional changes.6,13 Our results show that TB patients have an increased

frequency of circulating CD14HiCD16þ and CD14LoCD16þ monocytes compared with healthy controls and that CD16þ monocytes had decreased expression of markers related with maturation, differentiation and function of mononuclear phagocytes such as CD11b, CD36, CD64, CD68, CD80, CD86, CCR2 and NSE. Additionally, CD16þ cells did not differentiate into macrophages, produced higher levels of TNF-a, lower levels of IL-10 and underwent increased cell death in response to M. tuberculosis infection. There are several inconsistencies about functional and phenotypical studies in monocyte subsets, even in reports from authors working in the same group that are partially explained by different methodological approaches.18 There is not a consensus about the analysis of monocyte subpopulations based on cell surface expression of CD14 and CD16 (FcgRIII). Some authors propose two (CD16 and CD16þ)15 and others three subsets (CD14HiCD16, CD16HiCD16þ and CD14LoCD16þ).15,37,38 In our flow cytometry

D. Castaño et al. / Tuberculosis 91 (2011) 348e360

357

Figure 6. Cytokine production of CD16 and CD16þ monocytes after M. tuberculosis infection or PPD treatment. Sorted CD16 and CD16þ monocytes differentiated for 24 h were infected with live or dead M. tuberculosis at a MOI of 5, or treated with 10 mg/ml of PPD. A. TNF-a, B. IL-10, C. IL-1b and D. IL-8 production tested at 6 h of infection. Comparisons were done with Two-way ANOVA (n ¼ 7). The data are shown as the mean  SD.

analysis, it was possible to clearly identify three subpopulations. The CD14HiCD16 cells, which correspond to the “classic” monocytes and represent approximately 80e90% of peripheral blood monocytes. The CD14LoCD16þ cells, also named “proinflammatory”, that have been reported increased in different infectious and inflammatory conditions.39 The CD14HiCD16þ cells have been also shown to be increased in inflammatory events, such as severe asthma21 and cardiovascular diseases.23 To our knowledge, this is the first report showing an increased percentage of both circulating CD14HiCD16þ and CD14LoCD16þ monocytes in TB patients. A previous report showed increased CD16 expression on CD14þ monocytes from HIV- and HIVþ TB patients, compared with healthy controls, but did not differentiate them according to high or low CD14 expression.26 Other authors did not find differences in CD14LoCD16þ and CD14LoCD16þHLADRþþ monocytes between TB patients and healthy controls40. Another report showed high levels of CD14HiCD16þ monocytes in TSTþ controls compared with TB patients and TST controls.41 These authors proposed that CD16þ monocytes in TSTþ individuals participate in innate protective immune mechanisms against M. tuberculosis. However, we did not find difference in monocyte subpopulations between TST and TSTþ controls. These results highlight the relevance of testing the three monocyte subsets rather than two, since CD14HiCD16 and CD14LoCD16þ subpopulations may mask changes in the CD14HiCD16þ subset. Our group had previously reported that monocyte differences between TB patients and healthy controls disappeared during antiTB treatment,13 hence it is possible to postulate that the increase in CD16þ monocytes is a systemic consequence of the active TB and that bacterial or host factors may be responsible for such increase. We have also reported that monocytes from TB patients produce more IL-10 in response to M. tuberculosis infection and PPD stimulation, compared with monocytes from healthy controls,6 and other authors have observed that treatment of human monocytes with IL-10 induces high levels of CD16,42,43 suggesting a link

between the increase in CD16þ cells and the increment of IL-10þ monocytes observed in TB patients. The increased CD16þ monocyte subpopulations in TB patients compared with healthy controls, prompted us to analyze other markers related with maturation, differentiation and function of mononuclear phagocytes in monocyte subsets. The CD16þ monocytes had decreased percentage of CD11b, CD36, CD64, CD68, CD80, CD86, CCR2 and NSE compared with the CD14HiCD16 subset. Additionally, CD16þ monocytes did not differentiate into macrophages and underwent spontaneous late apoptosis. These results support that CD16þ monocytes are in a stage of less maturation or differentiation (maturation/differentiation), or alternatively, that these monocytes are precursors of dendritic cells rather than macrophages. The evaluation of myeloid maturation and differentiation is complex and there is not an accepted gold standard to accurately establish the maturation/differentiation stage of a monocyte. However, the low percentage of cells expressing CD36, CD64, CD68, CD86 and NSE may be related with a decreased maturation/differentiation stage, as suggesting by the finding that the expression of these molecules increased with the maturation and differentiation of monocytes, as we showed in Figure 3 and reported by other authors.28,44,45,46,47,48 Thus, the differences among circulating monocyte subsets may be partially explained by their different state of maturation.30 In early studies, it was suggested that CD14LoCD16þ monocytes were immature precursors.15,49 This subset showed lower capacity to adhere and phagocytose antibody-coated erythrocytes, compared with CD14HiCD16 monocytes.15 In addition, it has been reported that CD16þ monocytes express lower levels of myeloperoxidase and a-naphthyl acetate esterase compared to CD16 cells, classical markers of the monocytic lineage and expressed early in monocytic differentiation to macrophages.49 However, the studies of human monocyte differentiation in vitro, which demonstrated that CD14HiCD16 monocytes change their phenotype to CD14LoCD16þ in response to M-CSF, LPS and TNF-a19,20; together with the

358

D. Castaño et al. / Tuberculosis 91 (2011) 348e360

transcriptome analysis of monocyte subsets50,51,52 and the high expression of HLA-II per cell and decrease of CD11b and CD33,39,53 led other authors to consider CD16þ monocytes closer to macrophage-like cells than CD14HiCD16 monocytes.18 Our results showed that the in vitro differentiation of total monocytes resulted in a decrease of CD14þ and an increase of CD16þ cells; additionally, we found a higher expression of HLA-DR and decreased of CD11b and CD33 on CD16þ monocytes, as previously reported.53,54 These findings have been interpreted by other authors as evidence that CD16þ monocytes are more mature cells53,54; however, we observed in CD16þ monocytes a low percentage of other markers related with maturation and differentiation such as CD36, CD64, CD68, CD86 and NSE; In addition, when monocyte subsets were separated, the CD16þ cells did not differentiate into macrophages in vitro. Thus, although our results are partially in agreement with previous reports, altogether, our results do not support that CD16þ cells are more mature monocytes (macrophage-like cells), and suggest that the maturation/differentiaion studies of non-separated subsets will require a more careful interpretation. Alternatively, it has been demonstrated that CD16þ monocytes analyzed ex vivo and CD16þ monocytes derived in vitro from PBMC have characteristics of dendritic cells.27,29,30 In addition, CD16þ monocytes become differentiated into dendritic cells easily than CD16 monocytes.55 Under our experimental conditions, the expression of specific markers of dendritic cells (CD1a, CD1c and CD83) was negligible in CD16þ monocytes; however, these cells had high MFI of CD11c and HLA-DR, characteristics of dendritic cells. Although under our culture conditions we did not obtain dendritic cells, and we did not evaluate the differentiation of CD16þ monocytes into dendritic cells in specific culture conditions, the decreased differentiation of CD16þ monocytes into MDM, could be due to the predisposition of these cells to differentiate into dendritic cells as previously reported.55 It has been observed that CD16þ monocytes express more receptors for M-CSF than CD16 cells,51,52 suggesting that CD16þ cells may require additional signals to survive and differentiate in vitro to macrophages, such as M-CSF, or to dendritic cell such as GM-CSF and IL-4. Moreover, it is interesting to note that we found a high constitutive production of IL-12p70 by CD16þ monocytes at 24 h, but not at 6 h, of in vitro culture compared with CD16 monocytes (data not shown), suggesting a dendritic cell like function, as proposed by other authors.55,56 It is important to note that some of markers evaluated in this study are also related with mononuclear phagocyte functions, such as in the phagocytosis of pathogens (CD11b, CD36, CD64), coestimulation of T lymphocytes (CD80, CD86), the lysosomes (CD68, NSE) and in their migratory ability induced by ligands such as the monocyte chemoattractant protein-157 (CCR2). In addition to the phenotype observed in CD16þ monocytes (low percentage of CD11b, CD36, CD64, CD68, CD80, CD86, CCR2 and NSE) compared with CD16 cells, the CD16þ monocytes from TB patients had a decreased percentage of HLA-DR and increased proportion of CD11b positive cells compared with healthy controls. CD11b, a subunit of complement receptor 3, has been described as part of a pathway favoring the intracellular survival of M. tuberculosis58; and the low percentage of HLA-DR, CD80 and CD86 may affect the antigen presentation and the activation of T cells in TB patients favoring the establishment of the infection. The high production of TNF-a and the low of IL-10 by CD16þ cells in response to live M. tuberculosis, has been previously reported in CD16þ monocytes treated with LPS.59 Although, it was not explored in this study, it is possible that the differential cytokine production observed in monocyte subsets could be explained by differences in the expression of surface receptors that bind

M. tuberculosis, such as the mannose receptor that was previously reported to be decreased in CD16þ monocytes.60 In addition, in response to dead M. tuberculosis there was a lower production of TNF-a and IL-8 compared with live mycobacteria by CD16þ monocytes. There was reported changes in the cell wall components from M. tuberculosis between live and dead bacteria, for example, it was noticed that dead mycobacteria had decreased amounts of ManLAM.61 Therefore, the differences in the cytokine production between live and dead mycobacteria could be explained by differences in their cell wall structure. Our finding that in vitro infection with M. tuberculosis induces late apoptosis of CD16þ monocytes after 24 h of infection is in agreement with previous reports showing a basal expression of more pro-apoptotic genes in CD16þ monocytes and their higher susceptibility to undergo apoptosis after activation with specific peptides, compared with CD16 monocytes.50,62 In addition, it was previously proposed that CD16þ monocytes are not efficient removers of apoptotic cells,52 because their low expression of receptors involved in this uptake such as the CD36 (Table 1). Thus, it is possible that after M. tuberculosis infection, the apoptotic CD16þ monocytes may not be removed by bystander CD16þ monocytes, leading to necrosis and mycobacterial release. Additionally, the cell death observed in CD16þ monocytes at 24 h can be associated with the cytokines produced at earlier time points of infection. IL-10 is a cytokine with anti-apoptotic effects that was almost not produced by CD16þ.63,64 It has been determined that TNF-a induced by M. tuberculosis infection is associated with the induction of apoptosis of monocytes.6 In addition to the decreased uptake of apoptotic bodies, the high production of TNFa and low of IL-10 in infected CD16þ monocytes could partially explain the induction of necrosis secondary to apoptosis and may favor the dissemination of the infection. We did not evaluate other anti-mycobacterial responses of CD16þ monocytes, since the intracellular replication of M. tuberculosis or the coestimulation of T cells will take longer periods of time; then, these times will be overlapped with the times of cell death. All together, these results demonstrate that M. tuberculosis infection induces the increase of circulating CD16þ monocytes that are more susceptible to spontaneous or mycobacterial induced late apoptosis. Although the role of CD16þ monocytes in the immunopathogenesis of TB is still poorly understood, it is tempting to speculate that during M. tuberculosis infection, different products secreted by the mycobacteria or the infected macrophages can reach the circulation and promote the increase of circulating CD16þ monocytes, which may migrate into the affected tissues and participate in the granuloma formation. Whether these CD16þ monocytes behave similarly during in vivo and in vitro infection is unknown. However, in this context, the interaction of CD16þ monocytes with M. tuberculosis, may be important for the initial secretion of TNF-a at the site of infection. TNF-a production has been associated with the control of M. tuberculosis infection.28 However, if these monocytes are continually recruit to the lung, the excess of TNF-a plus the late apoptosis of CD16þ monocytes may exacerbate the proinflammatory response that would be detrimental for infected lung tissue and the prognosis of the tuberculosis. Acknowledgments The authors thank the TB patients and control individuals that participated in this study and the personnel and institutions that contributed in the recruitment of the patients. We appreciate the collaboration of Luis F. Barrera for the critical reading of the manuscript, and Camilo Duque for his help in the analysis and interpretation of Acridine Orange/PI staining.

D. Castaño et al. / Tuberculosis 91 (2011) 348e360

Funding: This study was supported by COLCIENCIAS (Bogotá, Colombia) grants: RC431-2004 (“Centro Colombiano de Investigación en Tuberculosis”) and 111540520270 and Programa de Sostenibilidad 2009e2010 grant from the Universidad de Antioquia. Diana Castaño was recipient of a doctoral fellowship from COLCIENCIAS. Conflicts of interest: The authors have no financial or personal conflicts of interest in the present study. Ethical approval: TB patients and healthy controls signed an informed consent previously approved by the Ethics Committee of the Instituto de Investigaciones Médicas of the Facultad de Medicina, Universidad de Antioquia, Medellín, Colombia. References 1. Sundaramurthy V, Pieters J. Interactions of pathogenic mycobacteria with host macrophages. Microbes Infect 2007;9:1671e9. 2. Koul A, Herget T, Klebl B, Ullrich A. Interplay between mycobacteria and host signalling pathways. Nat Rev Microbiol 2004;2:189e202. 3. Liu PT, Modlin RL. Human macrophage host defense against Mycobacterium tuberculosis. Curr Opin Immunol 2008;20:371e6. 4. Bhatt K, Salgame P. Host innate immune response to Mycobacterium tuberculosis. J Clin Immunol 2007;27:347e62. 5. Flannagan RS, Cosio G, Grinstein S. Antimicrobial mechanisms of phagocytes and bacterial evasion strategies. Nat Rev Microbiol 2009;7:355e66. 6. Gil DP, León LG, Correa LI, Maya JR, París SC, García LF, et al. Differential induction of apoptosis and necrosis in monocytes from patients with tuberculosis and healthy control subjects. J Infect Dis 2004;189:2120e8. 7. Schaible UE, Winau F, Sieling PA, Fischer K, Collins HL, Hagens K, et al. Apoptosis facilitates antigen presentation to T lymphocytes through MHC-I and CD1 in tuberculosis. Nat Med 2003;9:1039e46. 8. Lee J, Hartman M, Kornfeld H. Macrophage apoptosis in tuberculosis. Yonsei Med J 2009;50:1e11. 9. Molloy A, Laochumroonvorapong P, Kaplan G. Apoptosis, but not necrosis, of infected monocytes is coupled with killing of intracellular bacillus CalmetteGuerin. J Exp Med 1994;180:1499e509. 10. Pan H, Yan BS, Rojas M, Shebzukhov YV, Zhou H, Kobzik L, et al. Ipr1 gene mediates innate immunity to tuberculosis. Nature 2005;434:767e72. 11. Davis JM, Ramakrishnan L. The role of the granuloma in expansion and dissemination of early tuberculous infection. Cell 2009;136:37e49. 12. Rubin EJ. The granuloma in tuberculosisefriend or foe? N Engl J Med 2009;360:2471e3. 13. Sánchez MD, García Y, Montes C, París SC, Rojas M, Barrera LF, et al. Functional and phenotypic changes in monocytes from patients with tuberculosis are reversed with treatment. Microbes Infect 2006;8:2492e500. 14. Gordon S, Taylor PR. Monocyte and macrophage heterogeneity. Nat Rev Immunol 2005;5:953e64. 15. Passlick B, Flieger D, Ziegler-Heitbrock HW. Identification and characterization of a novel monocyte subpopulation in human peripheral blood. Blood 1989;74:2527e34. 16. Rothe G, Gabriel H, Kovacs E, Klucken J, Stohr J, Kindermann W, et al. Peripheral blood mononuclear phagocyte subpopulations as cellular markers in hypercholesterolemia. Arterioscler Thromb Vasc Biol 1996;16:1437e47. 17. Tanaka M, Honda J, Imamura Y, Shiraishi K, Tanaka K, Oizumi K. Surface phenotype analysis of CD16þ monocytes from leukapheresis collections for peripheral blood progenitors. Clin Exp Immunol 1999;116:57e61. 18. Ziegler-Heitbrock HW. Heterogeneity of human blood monocytes: the CD14þ CD16þ subpopulation. Immunol Today 1996;17:424e8. 19. Munn DH, Bree AG, Beall AC, Kaviani MD, Sabio H, Schaub RG, et al. Recombinant human macrophage colony-stimulating factor in nonhuman primates: selective expansion of a CD16þ monocyte subset with phenotypic similarity to primate natural killer cells. Blood 1996;88:1215e24. 20. Skinner NA, MacIsaac CM, Hamilton JA, Visvanathan K. Regulation of Toll-like receptor (TLR)2 and TLR4 on CD14dimCD16þ monocytes in response to sepsis-related antigens. Clin Exp Immunol 2005;141:270e8. 21. Moniuszko M, Bodzenta-Lukaszyk A, Kowal K, Lenczewska D, Dabrowska M. Enhanced frequencies of CD14þþCD16þ, but not CD14þCD16þ, peripheral blood monocytes in severe asthmatic patients. Clin Immunol 2009;130:338e46. 22. Schlitt A, Heine GH, Blankenberg S, Espinola-Klein C, Dopheide JF, Bickel C, et al. CD14þCD16þ monocytes in coronary artery disease and their relationship to serum TNF-alpha levels. Thromb Haemost 2004;92:419e24. 23. Heine GH, Ulrich C, Seibert E, Seiler S, Marell J, Reichart B, et al. CD14(þþ) CD16þ monocytes but not total monocyte numbers predict cardiovascular events in dialysis patients. Kidney Int 2008;73:622e9. 24. Baeten D, Boots AM, Steenbakkers PG, Elewaut D, Bos E, Verheijden GF, et al. Human cartilage gp-39þ, CD16þ monocytes in peripheral blood and synovium: correlation with joint destruction in rheumatoid arthritis. Arthritis Rheum 2000;43:1233e43.

359

25. Saleh MN, Goldman SJ, LoBuglio AF, Beall AC, Sabio H, McCord MC, et al. CD16þ monocytes in patients with cancer: spontaneous elevation and pharmacologic induction by recombinant human macrophage colony-stimulating factor. Blood 1995;85:2910e7. 26. Vanham G, Edmonds K, Qing L, Hom D, Toossi Z, Jones B, et al. Generalized immune activation in pulmonary tuberculosis: co-activation with HIV infection. Clin Exp Immunol 1996;103:30e4. 27. Grage-Griebenow E, Zawatzky R, Kahlert H, Brade L, Flad H, Ernst M. Identification of a novel dendritic cell-like subset of CD64(þ)/CD16(þ) blood monocytes. Eur J Immunol 2001;31:48e56. 28. 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. 29. Ancuta P, Weiss L, Haeffner-Cavaillon N. CD14þCD16þþ cells derived in vitro from peripheral blood monocytes exhibit phenotypic and functional dendritic cell-like characteristics. Eur J Immunol 2000;30:1872e83. 30. Li G, Hangoc G, Broxmeyer HE. Interleukin-10 in combination with M-CSF and IL-4 contributes to development of the rare population of CD14þCD16þþ cells derived from human monocytes. Biochem Biophys Res Commun 2004;322:637e43. 31. Yourno J, Walsh J, Kornatowski G, O’Connor D, Kumar SA. Nonspecific esterases of leukemia cell lines: evidence for activation of myeloid-associated zymogens in HL-60 by phorbol esters. Blood 1984;63:238e41. 32. Nakagawara A, Nathan CF. A simple method for counting adherent cells: application to cultured human monocytes, macrophages and multinucleated giant cells. J Immunol Methods 1983;56:261e8. 33. Zelenin A. Acridine orange as a probe for cell and molecular biology. In: Manson W, editor. Fluorescent and luminiscent probes for biological activity. 2nd ed. San Diego: Academic Press; 1999. p. 117e35. 34. Denny MF, Yalavarthi S, Zhao W, Thacker SG, Anderson M, Sandy AR, et al. A distinct subset of proinflammatory neutrophils isolated from patients with systemic lupus erythematosus induces vascular damage and synthesizes type I IFNs. J Immunol 2010;184:3284e97. 35. Meisel SR, Xu XP, Edgington TS, Dimayuga P, Kaul S, Lee S, et al. Differentiation of adherent human monocytes into macrophages markedly enhances tissue factor protein expression and procoagulant activity. Atherosclerosis 2002;161:35e43. 36. Netea MG, Lewis EC, Azam T, Joosten LA, Jaekal J, Bae SY, et al. Interleukin-32 induces the differentiation of monocytes into macrophage-like cells. Proc Natl Acad Sci USA 2008;105:3515e20. 37. Skrzeczynska-Moncznik J, Bzowska M, Loseke S, Grage-Griebenow E, Zembala M, Pryjma J. Peripheral blood CD14high CD16þ monocytes are main producers of IL-10. Scand J Immunol 2008;67:152e9. 38. Ziegler-Heitbrock HW, Passlick B, Flieger D. The monoclonal antimonocyte antibody My4 stains B lymphocytes and two distinct monocyte subsets in human peripheral blood. Hybridoma 1988;7:521e7. 39. Ziegler-Heitbrock L. The CD14þ CD16þ blood monocytes: their role in infection and inflammation. J Leukoc Biol 2007;81:584e92. 40. Barcelos W, Martins-Filho OA, Guimaraes TM, Oliveira MH, Spindola-deMiranda S, Carvalho BN, et al. Peripheral blood mononuclear cells immunophenotyping in pulmonary tuberculosis patients before and after treatment. Microbiol Immunol 2006;50:597e605. 41. Barcelos W, Sathler-Avelar R, Martins-Filho OA, Carvalho BN, Guimaraes TM, Miranda SS, et al. Natural killer cell subpopulations in putative resistant individuals and patients with active Mycobacterium tuberculosis infection. Scand J Immunol 2008;68:92e102. 42. Calzada-Wack JC, Frankenberger M, Ziegler-Heitbrock HW. Interleukin-10 drives human monocytes to CD16 p.sitive macrophages. J Inflamm 1996;46:78e85. 43. Allavena P, Piemonti L, Longoni D, Bernasconi S, Stoppacciaro A, Ruco L, et al. IL-10 p.events the differentiation of monocytes to dendritic cells but promotes their maturation to macrophages. Eur J Immunol 1998;28:359e69. 44. Holness CL, Simmons DL. Molecular cloning of CD68, a human macrophage marker related to lysosomal glycoproteins. Blood 1993;81:1607e13. 45. Koeffler HP, Golde DW. Human myeloid leukemia cell lines: a review. Blood 1980;56:344e50. 46. Jayaram Y, Buckle AM, Hogg N. The Fc receptor, FcRI, and other activation molecules on human mononuclear phagocytes after treatment with interferongamma. Clin Exp Immunol 1989;75:414e20. 47. Alessio M, De Monte L, Scirea A, Gruarin P, Tandon NN, Sitia R. Synthesis, processing, and intracellular transport of CD36 during monocytic differentiation. J Biol Chem 1996;271:1770e5. 48. Huh HY, Pearce SF, Yesner LM, Schindler JL, Silverstein RL. Regulated expression of CD36 during monocyte-to-macrophage differentiation: potential role of CD36 in foam cell formation. Blood 1996;87:2020e8. 49. Almeida J, Bueno C, Alguero MC, Sanchez ML, de Santiago M, Escribano L, et al. Comparative analysis of the morphological, cytochemical, immunophenotypical, and functional characteristics of normal human peripheral blood lineage()/CD16(þ)/HLA-DR(þ)/CD14(-/lo) cells, CD14(þ) monocytes, and CD16(-) dendritic cells. Clin Immunol 2001;100:325e38. 50. Zhao C, Zhang H, Wong WC, Sem X, Han H, Ong SM, et al. Identification of novel functional differences in monocyte subsets using proteomic and transcriptomic methods. J Proteome Res 2009;8:4028e38. 51. Ancuta P, Liu KY, Misra V, Wacleche VS, Gosselin A, Zhou X, et al. Transcriptional profiling reveals developmental relationship and distinct biological

360

52. 53.

54.

55.

56.

57.

58.

D. Castaño et al. / Tuberculosis 91 (2011) 348e360 functions of CD16þ and CD16 monocyte subsets. BMC Genomics 2009;10:403. Mobley JL, Leininger M, Madore S, Baginski TJ, Renkiewicz R. Genetic evidence of a functional monocyte dichotomy. Inflammation 2007;30:189e97. Ziegler-Heitbrock HW, Fingerle G, Strobel M, Schraut W, Stelter F, Schutt C, et al. The novel subset of CD14þ/CD16þ blood monocytes exhibits features of tissue macrophages. Eur J Immunol 1993;23:2053e8. Frankenberger M, Sternsdorf T, Pechumer H, Pforte A, Ziegler-Heitbrock HW. Differential cytokine expression in human blood monocyte subpopulations: a polymerase chain reaction analysis. Blood 1996;87:373e7. Randolph GJ, Sanchez-Schmitz G, Liebman RM, Schakel K. The CD16(þ) (FcgammaRIII(þ)) subset of human monocytes preferentially becomes migratory dendritic cells in a model tissue setting. J Exp Med 2002;196: 517e27. Mortellaro A, Wong SC, Fric J, Ricciardi-Castagnoli P. The need to identify myeloid dendritic cell progenitors in human blood. Trends Immunol 2010;31:18e23. Kurihara T, Warr G, Loy J, Bravo R. Defects in macrophage recruitment and host defense in mice lacking the CCR2 chemokine receptor. J Exp Med 1997;186:1757e62. Malik ZA, Denning GM, Kusner DJ. Inhibition of Ca(2þ) signaling by Mycobacterium tuberculosis is associated with reduced phagosome-lysosome fusion

59.

60.

61.

62.

63.

64.

and increased survival within human macrophages. J Exp Med 2000;191:287e302. Belge KU, Dayyani F, Horelt A, Siedlar M, Frankenberger M, Frankenberger B, et al. The proinflammatory CD14þCD16þDRþþ monocytes are a major source of TNF. J Immunol 2002;168:3536e42. Sanchez-Torres C, Garcia-Romo GS, Cornejo-Cortes MA, Rivas-Carvalho A, Sanchez-Schmitz G. CD16þ and CD16- human blood monocyte subsets differentiate in vitro to dendritic cells with different abilities to stimulate CD4þ T cells. Int Immunol 2001;13:1571e81. Shimada K, Takimoto H, Yano I, Kumazawa Y. Involvement of mannose receptor in glycopeptidolipid-mediated inhibition of phagosome-lysosome fusion. Microbiol Immunol 2006;50:243e51. Rouzaut A, Subira ML, de Miguel C, Domingo-de-Miguel E, Gonzalez A, Santiago E, et al. Co-expression of inducible nitric oxide synthase and arginases in different human monocyte subsets. Apoptosis regulated by endogenous NO. Biochim Biophys Acta 1999;1451:319e33. Zhong J, Deaciuc IV, Burikhanov R, de Villiers WJ. Lipopolysaccharide-induced liver apoptosis is increased in interleukin-10 knockout mice. Biochim Biophys Acta 2006;1762:468e77. Ina K, Kusugami K, Kawano Y, Nishiwaki T, Wen Z, Musso A, et al. Intestinal fibroblast-derived IL-10 increases survival of mucosal T cells by inhibiting growth factor deprivation- and Fas-mediated apoptosis. J Immunol 2005;175:2000e9.