Logo of clinexpimmunolLink to Publisher's site
Clin Exp Immunol. Nov 2010; 162(2): 289–297.
PMCID: PMC2996596

CCL20 is overexpressed in Mycobacterium tuberculosis-infected monocytes and inhibits the production of reactive oxygen species (ROS)


CCL20 is a chemokine that attracts immature dendritic cells. We show that monocytes, cells characteristic of the innate immune response, infected with Mycobacterium tuberculosis express the CCL20 gene at a much higher level than the same cells infected with non-tuberculous mycobacteria. Interferon (IFN)-γ, a fundamental cytokine in the immune response to tuberculosis, strongly inhibits both the transcription and the translation of CCL20. We have also confirmed that dendritic cells are a suitable host for mycobacteria proliferation, although CCL20 does not seem to influence their intracellular multiplication rate. The chemokine, however, down-regulates the characteristic production of reactive oxygen species (ROS) induced by M. tuberculosis in monocytes, which may affect the activity of the cells. Apoptosis mediated by the mycobacteria, possibly ROS-dependent, was also inhibited by CCL20.

Keywords: apoptosis, chemotaxis, dendritic cells, innate immunity, non-tuberculous mycobacteria


Tuberculosis remains a leading infectious disease and a major cause of death worldwide [1]. The main route of entry of Mycobacterium tuberculosis is via the respiratory tract, until it reaches the pulmonary alveoli. Initially innate immunity may abort the infection through the activities of alveolar macrophages and other cells which are recruited, such as neutrophils and natural killer (NK) cells. When innate immunity fails, the bacteria multiply intracellularly and adaptive immunity determines the formation of the tuberculous granuloma. The arrival of macrophages and lymphocytes controls the bacterial proliferation, although some bacilli will survive in a latent form. Early in the primary infection of a ‘naive’ host, bacteria are transported to regional lymph nodes, causing an intense reaction. The granulomatous reaction and necrosis in the lymph nodes are known as the Ranke complex, characteristic of tuberculosis in childhood [2]. The main candidates to carry the pathogen to the lymph nodes are macrophages and dendritic cells. Dendritic cells are specialized for the presentation of antigen to T cells and have been observed in the tuberculous granuloma, apparently migrated from the peripheral blood [3]. Once infected, they may be attracted by chemokines to the lymph nodes and become a reservoir for mycobacteria [4]. Chemokines are small chemotactic cytokines produced by many cellular types, including M. tuberculosis-infected macrophages, which express members of the chemokine ligand family such as CCL2, CCL3, CCL4 and CCL5 [5].

Förstch et al. found that dendritic cells are appropriate hosts for M. tuberculosis, in which bacilli multiply [6]. At the beginning of infection bacteria reach the pulmonary alveoli, where dendritic cells may find and phagocyte them. The epithelial transmigration of dendritic cells into the alveoli seems to be determined by CCR6, as suggested by experiments conducted in CCR6−/− mice [7]. The only known ligand for this receptor is CCL20, which is expressed constitutively in many cell types and plays an important role in the immunity of mucosal-associated tissues, including lung and gastric mucosa [8]. When dendritic cells respond to a pathogen they mature, and CCR6 expression is down-regulated. Instead, mature dendritic cells express CCR7, and its ligands CCL19 and CCL21 promote their migration to the lymph nodes [9]. A recent report has shown that CCL20 is expressed during human tuberculosis, and macrophages from patients activated with an antigen from M. tuberculosis produce a higher amount of the chemokine than macrophages from healthy volunteers [10]. Furthermore, several studies have shown that other bacteria regulate the transcription of the CCL20 gene [1113].

Besides their chemotactic role chemokines exhibit several other functions, including leucocyte degranulation, NK cell proliferation, dendritic cell maturation, B and T cell development, angiogenesis or tumour growth [9,14]. Pervushina et al. have reported recently that the chemokine PF4 (CXCL4) induces the generation of reactive oxygen species (ROS) metabolites in monocytes [15]. Although ROS production is not associated with the killing of M. tuberculosis[16], it may have a large influence in the macrophage signalling [17] and function. Thus, it has been known for a long time that ROS induces apoptosis [18]; a relevant example is the programmed cell death mediated by the strong generation of ROS in neutrophils infected with M. tuberculosis[19]. Apoptosis has been regarded as a host mechanism of defence because H37Rv, a virulent strain of M. tuberculosis, induces less apoptosis in macrophages than H37Ra, which is avirulent. It has also been argued that apoptotic-infected cells slow the dissemination of the bacteria, and that these cells are phagocytized by fresh macrophages which may be active against the bacteria [20].

In the present study we show that M. tuberculosis increases dramatically the expression of CCL20 in human monocytes, even at a higher degree than other non-tuberculous mycobacteria. We have confirmed that the bacteria survive in dendritic cells, although CCL20 does not seem to promote an anti-mycobacterial activity. We have found, however, that CCL20 inhibits the generation of ROS, which may affect the activity of infected macrophages. Additionally, CCL20 inhibits M. tuberculosis-mediated apoptosis, possibly as a consequence of the down-regulation of ROS production.

Materials and methods

Bacterial strains

Mycobacterium tuberculosis HL186T, M. kansasii HL228K and M. avium HL70A were isolated at the Hospital de León (Microbiology Service), kindly provided by Julio Blanco and Manuela Caño. They were grown on 7H11 agar supplemented with 0·2% glycerol and 10% Middlebrook enrichment oleic acid, albumin, dextrose and catalase (OADC) (Becton Dickinson Microbiology Systems, San Agustín de Guadalix, Madrid, Spain). Legionella pneumophila Philadelphia, ATCC 13151, generously provided by Carmen Pelaz, was grown on buffered charcoal yeast extract (BCYE) agar plates. Bacteria from fresh culture in agar plates were suspended in the serum free medium Macrophage-SFM (Gibco, Invitrogen, Prat de Llobregat, Barcelona, Spain). To obtain isolated mycobacteria, they were sonicated using an S-450 digital ultrasonic cell disruptor (Branson Ultrasonics, Danbury, CT, USA). Pulses of 10 s were applied with a microtip at an amplitude of 10% (2 W), and sonicated bacteria were centrifuged at 100 g for 1 min at room temperature. After recovering the supernatants, sonications were repeated as many times as necessary to obtain individualized bacteria, usually three or four rounds. At the end most bacteria were alive and very few groups remained, with ≤ 5 bacteria per group, as determined by the LIVE/DEAD Baclight bacterial kit (Molecular Probes, Invitrogen, Prat de Llobregat, Barcelona, Spain). This treatment was not necessary for Legionella. After addition of glycerol to 20%, single-use aliquots were frozen at −80°C.

Monocytes, human monocyte-derived macrophages and dendritic cells

Peripheral blood was obtained from healthy volunteers following informed consent and approval of the protocol by the Hospital of León Clinical Research Ethics Board, and each experiment was performed with cells from a different volunteer. Peripheral blood mononuclear cells were isolated by Ficoll-Paque Plus density gradient sedimentation (GE Healthcare, Life Sciences, Uppsala, Sweden), and CD14+ cells (monocytes) were purified by magnetic cell separation (StemCell Technologies, Grenoble, France). We ascertained the purity of cells by flow cytometry with appropriate labelled antibodies (Becton Dickinson) and > 94% of cells in the monocyte preparation were CD14+. Excluded CD14- cells were used for studies of ROS formation. Cell numbers were calculated by counting in a Neubauer chamber and within 4 h from blood collection were cultivated in the serum-free medium Macrophage-SFM. For anti-microbial activity assays monocytes were differentiated to macrophages for 5 days. Dendritic cells were also obtained by incubation of monocytes for 5 days in the presence of 20 ng/ml interleukin (IL)-4 (≥ 5 × 106 units/mg), 10 ng/ml granulocyte–macrophage colony-stimulating factor (GM-CSF) (≥ 1 × 107 units/mg) and 5 ng/ml transforming growth factor TGF-β1 (≥ 2 × 107 units/mg). All cytokines were from Peprotech (London, UK). We checked the differentiation process by CD209 (DC-SIGN) expression. Cells changed from CD14+ CD209- to > 98% CD14- CD209+ (labelled antibodies from Becton Dickinson). All cells were incubated at 37°C in 95% air/5% CO2.

Cellular infection

For anti-microbial activity determination, macrophages and dendritic cells were differentiated as indicated above for 5 days in 96-well plates, always in a total volume of 100 µl; 105 cells were infected with 103 bacteria (multiplicity of infection, MOI = 0·01) in Macrophage-SFM. For dendritic cells, at the moment of infection fresh IL-4, GM-CSF and TGF-β1 were added. When indicated, cells were incubated in either 20 ng/ml of CCL20 or 100 ng/ml of interferon IFN-γ (> 2 × 107 units/mg; Peprotech). Cells were lysed after 4 days by sonication with a microtip at an amplitude of 10% (2 W) for 3 s to release bacteria. At this setting, ultrasounds were able to lyse cells without affecting the bacterial viability. Decimal dilutions of the sonicates were inoculated and incubated at 37°C in either BCYE agar plates (Legionella) for 4 days or 7H9 broth supplemented with 0·2% glycerol and 10% Middlebrook enrichment albumin–dextrose complex (ADC) (mycobacteria) for 10 days or less. CFU were determined for mycobacteria under an inverted microscope at ×100 magnification [21].

For total RNA purification, and supernatant recovery for chemotaxis experiments and CCL20 quantification, immediately after purification 5 × 105 monocytes were infected with 5 × 105 bacteria (MOI = 1) for 18 h in a volume of 1100 µl (24-well plates). When indicated, 100 ng/ml of IFN-γ was added.

Quantitative polymerase chain reaction (qPCR)

Total RNA from infected cells was prepared using the Ultraclean Tissue RNA Isolation Kit (Mo Bio Laboratories, Carlsbad, CA, USA), and reverse-transcribed into cDNA by qScript cDNA synthesis kit (Quanta Biosciences, Gaithersburg, MD, USA). Real-time PCR was performed on a Bio-Rad iCycler system (El Prat de Llobregat, Barcelona, Spain) using SYBR-Green (Molecular Probes, Invitrogen, Prat de Llobregat, Barcelona, Spain). The threshold cycle (Ct) values for each of the target genes were normalized to the Ct of the reference gene EF-1α (elongation factor 1α). The efficiency (E) of the PCR reaction for each gene was calculated using the slope of the standard curve obtained from the Ct of 1/8 dilutions of each amplicon (E = 10−1/slope). Gene expression in infected cells (test samples) relative to non-infected cells (control sample), was calculated at the following ratio [22]: [(Eref)Ct test/(Etarget)Ct test]/ [(Eref)Ct control/(Etarget)Ct control]. Although the data did not follow a normal distribution, log-transformation allowed statistical parametric testing. The primers used for EF-1α and CCL20 were as follows: EF-1α forward 5′-TGTTCCTGTTGGCCGAGTG-3′; reverse 5′-ATTGAAGCCCACATTGTCCC-3′; CCL20 forward 5′-GGCTGCTTTGATGTCAGTGC-3′; reverse 5′-GATGTCACAGCCTTCATTGGC-3′.

CCL20 and CCL2 quantification

To remove bacteria from supernatants, samples were centrifuged for 3 min at 8000 g at room temperature in ultrafree-MC filter units (Millipore Iberica, Madrid, Spain) of 0·45 µm and frozen at –80°C. CCL20 was quantified by the human CCL20/macrophage inflammatory protein (MIP)-3α DuoSet enzyme-linked immunosorbent assay (ELISA) development system (R&D Systems, Minneapolis, MN, USA) and CCL2 by the BD OptEIA human monocyte chemoattractant protein MCP-1 ELISA set (Becton Dickinson).


Monocyte-derived dendritic cells (5 × 104), differentiated as indicated above, were suspended in Macrophage-SFM and placed in BD Falcon Cell Culture Inserts (pore size 8·0 µm) in a volume of 200 µl (24-well plates; Becton Dickinson). The lower wells were supplied with 750 µl of either supernatants from infected cells or 10 ng/ml of CCL20. As controls we included supernatants of non-infected cells or medium without CCL20, respectively. When indicated, 6 µg of anti-CCL20 neutralizing antibody or 2 µg of anti-CCL2 neutralizing antibody (both rabbit polyclonal; Peprotech) were added. As a mock control 6 µg of a purified rabbit polyclonal anti-glutathione S-transferase (GST) antibody (obtained from our laboratory) was used. Plates were incubated for 16 h at 37°C. Migrated cells to the lower wells were counted under an inverted microscope. For statistical analysis, the migration index was calculated by dividing the number of migrated dendritic cells upon treatment by the number of migrated cells in controls.

Analysis of ROS formation

Intracellular production of ROS was measured as luminol-enhanced chemiluminescence. Either 105 or 5 × 105 bacteria (MOI = 1 or 5, respectively), luminol 500 µM (Sigma-Aldrich Spain, Tres Cantos, Madrid, Spain) and Hanks' balanced salt solution were added to 105 monocytes in a total volume of 100 µl, and incubated for 30 min at 37°C. For some treatments 20 ng/ml CCL20, 150 nM phorbol-12-myristate-13-acetate (PMA; Calbiochem, Merck Spain, Madrid, Spain), 3 mg/ml Zymosan (Sigma-Aldrich Spain) or 105 CD14- mononuclear cells were added. Emitted light was measured at 21°C at 30 min in a TopCount-NXT microplate scintillation and luminescence counter (Packard, Perkin Elmer España, Tres Cantos, Madrid, Spain) and ROS formation was expressed as counted photons per second (cps).

Apoptosis quantification

Apoptosis was detected with the fluorescein FragEL DNA fragmentation kit (Calbiochem), which is based on the fluorescent labelling of DNA ends cleaved during apoptosis; 2 × 105 cells were seeded on sterile 11-mm diameter cover glasses, infected with 106 bacteria and incubated in 24 well-plates. When indicated 20 ng/ml CCL20, 5 µM diphenyleneiodonium chloride (DPI; Sigma Aldrich, Spain), 6 µg anti-CCL20 antibody or 6 µg anti-GST mock antibody were added. Infection was stopped at 20 h by removal of the culture medium and fixing in 4% formaldehyde/phosphate buffered saline for 15 min. Preparations were conserved in 80% ethanol at 4°C. The next day cells were labelled according to the manufacturer's instructions, and the total number of cells was determined by DNA fluorescent staining with 4′, 6-diamidino-2-phenylindole, dihydrochloride (DAPI). The proportion of apoptotic cells from DAPI-stained cells was determined by fluorescent microscopy and expressed as a percentage.

Statistical analysis

CFU and relative gene expression data followed a normal distribution when log-transformed. Normally distributed data were analysed by paired Student's t-test or analysis of variance (anova). Pairwise comparisons for anova were performed by Tukey's honestly significant difference (HSD) tests or Dunnett's T3 tests when variances were not homogeneous. Data distributed non-normally were analysed by Kruskal–Wallis test and the corresponding pairwise comparisons by Dunn's test. In all cases, a P-value < 0·05 was considered significant. Analysis was performed with spss version 14·0 (SPSS Ibérica, Madrid, Spain) and G-Stat version 2·0 (Dunn's test; GlaxoSmithKline, Brentford, Middlesex, UK).


CCL20 gene is expressed differentially in monocytes infected with tuberculous and non-tuberculous mycobacteria

In the initial steps of an infection the first response corresponds to innate immunity, represented at the cellular level by monocytes/macrophages, neutrophils and NK cells. The main M. tuberculosis cellular targets are macrophages, which upon infection produce many cytokines that influence the immune response. CCL19, CCL20 and CCL21 may be very important at the beginning of the development of the disease. We analysed them in monocytes infected with MOI = 1 and found no expression of CCL21, and no differences in expression between resting monocytes and infected monocytes for CCL19 (data not shown). On the contrary, the small basal level of expression of CCL20 in resting monocytes increased dramatically when they were infected with M. tuberculosis; the increase ranged from four to 39-fold (Fig. 1a). To investigate whether the observed increase was specific of M. tuberculosis we analysed CCL20 expression in monocytes infected with other non-tuberculous mycobacteria, namely M. kansasii and M. avium, and observed that in both cases this gene was expressed at a much lower level. Although gene expression was higher in M. kansasii than in M. avium infected monocytes, no statistically significant differences were found between any of them and non-infected macrophages (Fig. 1a).

Fig. 1
Measurement of CCL20 expression. (a) Relative gene expression measured by quantitative polymerase chain reaction (qPCR). Data represent the mean ± standard deviation (s.d.) of the relative gene expression log in infected groups in comparison with ...

The importance of IFN-γ in the immunity to tuberculosis is acknowledged widely [23]. We used this cytokine to activate monocytes in our preliminary experiments and observed that it inhibited the constitutive expression of CCL20. In fact, this effect was complete for one of the samples in which the gene expression was abolished totally in non-infected macrophages (data not shown). In other samples, inhibition of CCL20 expression in the presence of IFN-γ also occurred, although it did not reach the low level of expression in non-infected cells (Fig. 1a).

The NK cell may also contribute to the innate response in tuberculosis. Esin et al. have shown recently that NK cells (CD56+CD3-) bind directly to mycobacteria [24]. Indeed, we observed that M. tuberculosis also induced the activation of CCL20 transcription in CD56+ cells (data not shown).

CCL20 is produced by M. tuberculosis-infected monocytes

Although the expression of CCL20 is constitutive in monocytes, we were not able to detect the protein in many of the culture supernatants. CCL20 was, however, present in all supernatants of M. tuberculosis infected monocytes and in three and two of five supernatants of M. kansasii- and M. avium-infected monocytes, respectively. In the presence of IFN-γ the median of CCL20 production by infected monocytes was smaller. Significant differences in the amount of CCL20 were observed only between cells infected with M. tuberculosis and both non-infected cells and infected cells in the presence of IFN-γ (Fig. 1b). These results mirrored closely the observed CCL20 expression: higher gene expression corresponded to higher protein production.

CCL20 influences the migration of immature dendritic cells

CCL20 represents a link between innate and adaptive immunity, because it attracts immature dendritic cells which present antigens to lymphocytes. Supernatants of infected monocytes were used to test the chemotactic activity of secreted CCL20 in monocyte-derived dendritic cells. We did not find significant differences in the migration index of dendritic cells exposed to supernatants of monocytes infected with either M. tuberculosis or M. kansasii compared with non-infected cells. Nevertheless, the index value for both mycobacteria was higher than for non-infected cells, following the pattern of CCL20 production. Anti-CCL20 antibodies seemed to block the activity present in supernatants from M. tuberculosis-infected monocytes, which suggests that in cell supernatants CCL20 is the most abundant chemokine which is active for dendritic cells. This observation provides appropriate control of bioactivity of the CCL20 protein detected by ELISA. As additional control we used a mock antibody (a non-specific polyclonal rabbit antibody), and observed that it did not inhibit cell migration. Perhaps the best-studied CC-chemokine in tuberculosis is CCL2, also known as MCP-1 [25], which also attracts immature dendritic cells [26]. Using ELISA we verified the presence of CCL2 in supernatants of M. tuberculosis infected monocytes (366·9 SD 274·6 pg/ml, n = 3), but blocking the chemokine with a neutralizing antibody did not produce any change in the chemotactic index (Fig. 2a). These data support evidence for the predominant role of CCL20 in the chemotaxis of dendritic cells in our model. A surprising result was the induction of a similar migration index by supernatants from M. kansasii- and M. tuberculosis-infected monocytes, despite strong differences in the production of CCL20, depending on the infecting bacteria. Because results did not reach statistical significance this observation may not correspond to a true biological effect, but it is also possible that M. kansasii induces expression of an unknown chemokine responsible for this migration index.

Fig. 2
Chemotactic response of dendritic cells. Data represent the migration index mean ± standard deviation (s.d.) from supernatants of infected cells (a, n = 3) or from purified CCL20 (b, n = 4). Controls are the supernatant from non-infected cells ...

When we added 10 ng/ml of purified CCL20 to the same dendritic cells and blocked it with anti-CCL20 antibodies, we obtained significantly different migration indexes (Fig. 2b). We therefore believe that the small migration index observed when using supernatants is a consequence of the low amount of CCL20 present (< 100 pg/ml), enough to appreciate differences but not statistically significant.

CCL20 does not augment anti-microbial activity of either macrophages or dendritic cells

Chemokines are known to have biological activities other than chemotaxis. We asked whether the increase of CCL20 production was a defence mechanism of infected cells that might promote increased anti-mycobacterial activity. Alveolar macrophages are critical for the innate immune response, and their production of CCL20 may be important for the recruitment of immature dendritic cells, which would be exposed increasingly to the chemokine. To test whether CCL20 have any influence in the intracellular multiplication of mycobacteria, macrophages and dendritic cells were obtained in vitro from monocytes and their killing activities in the presence of 20 ng/ml of CCL20 were measured. To mimic the natural course of infection as closely as possible, a very low MOI (0·01) was used. The macrophage anti-microbial activity was assessed with L. pneumophila infections, which allowed us to confirm that IFN-γ-activated macrophages were able to restrict bacterial growth. We already knew that the CCL20 used was biologically active, because it promoted migration of dendritic cells (Fig. 2b). All three M. tuberculosis, M. kansasii and M. avium multiplied intracellularly in both macrophages and dendritic cells, reaching statistical differences for growth of M. tuberculosis in both macrophages and dendritic cells, and for non-tuberculous mycobacteria in dendritic cells. However, addition of CCL20 did not decrease the intracellular multiplication rate of the bacteria. On the contrary, CCL20 seemed to favour bacteria proliferation slightly, although differences were only marginal and statistically not significant when compared with cells infected in the absence of the chemokine (Table 1). Therefore, CCL20 does not seem to play any role in the anti-mycobacterial activity of human macrophages and dendritic cells.

Table 1
Intracellular survival of mycobacteria in macrophages and dendritic cells activated with CCL20.

CCL20 inhibits M. tuberculosis-mediated ROS generation

ROS generation is among the cellular functions that may be influenced by chemokines such as PF4, which has been found to promote ROS production in monocytes. We tested whether the well-characterized induction of ROS production by M. tuberculosis could be increased in the presence of CCL20. Surprisingly, we observed that the effect of CCL20 was in fact inhibitory (Table 2). At MOI = 1 there was an increase in the production of ROS above basal levels. Although CCL20 inhibited the total amount of ROS generated, differences did not reach statistical significance. Only when MOI = 5, with a higher ROS production, was the inhibition of CCL20 more noticeable and statistically significant. This and much larger MOI will occur in vivo when the innate response has failed, and the bacterial load increases progressively. At this point we hypothesized that other strong inducers of ROS would suffer similar levels of inhibitions and tested the phorbol ester PMA and the particulate stimulus zymosan. Indeed, the amount of ROS detected was much larger, but the level of inhibition by CCL20 was relatively smaller compared with inhibition in infected monocytes, and statistically not significant. These results suggest that the level of inhibition of CCL20 was dependent on the inducer, and that M. tuberculosis-mediated production was particularly affected. We remained concerned, however, about the purity of the monocyte preparation, because we have indicated that monocytes were < 100% of the total number of cells. Most of the contaminating cells in the purification of monocytes are lymphocytes, which are known to express CCR6, the receptor of CCL20. The addition of 105 CD14- mononuclear cells, however, did not change markedly the amount of both ROS production and inhibition. We conclude that the inhibition of ROS formation was not mediated by CD14- cells. L. pneumophila is a stimulus comparable to M. tuberculosis; the level of ROS production was lower and we again observed inhibition, although it was not statistically significant.

Table 2
Influence of CCL20 in Mycobacterium tuberculosis-mediated reactive oxygen species (ROS) production.

CCL20 inhibits M. tuberculosis-mediated apoptosis

Although highly produced in monocytes infected with M. tuberculosis, ROS does not promote anti-mycobacterial activity. Nevertheless, it may have a large influence in other biological mechanisms such as apoptosis, which in some circumstances has been described to be mediated by ROS. We decided to test the development of M. tuberculosis-mediated apoptosis under the influence of CCL20. We first confirmed that M. tuberculosis induces statistically significant levels of apoptosis in monocytes and, as expected, the addition of CCL20 inhibited apoptosis to levels close to the observed for non-infected cells. The inhibition induced by CCL20 was blocked by a neutralizing antibody but not by a non-specific mock antibody. Furthermore, the nicotinamide adenine dinucleotide phosphate (NADPH) oxidase inhibitor diphenyleneiodonium sulphate (DPI) also decreased the number of apoptotic cells infected with M. tuberculosis (Fig. 3). These results suggest that in monocytes apoptosis is also dependent upon ROS production, and inhibited by CCL20.

Fig. 3
Influence of CCL20 in the apoptosis mediated by Mycobacterium tuberculosis in infected monocytes. Monocytes were infected with a multiplicity of infection of 5, and a negative control (no infection) was included. Data are the proportions (expressed as ...


Approximately 10% of people who become infected with M. tuberculosis will develop the disease. The rest are able to control tuberculosis through an efficient adaptive immunity [27] which, in the first stages, needs the presentation of bacterial antigens by either macrophages or dendritic cells to lymphocytes. There is evidence in the mouse model that infected dendritic cells, rather than macrophages, disseminate to draining lymph nodes [28]. Immature dendritic cells transform into mature cells after phagocytosis of bacteria. They begin to express CCR7, the specific receptor of CCL19 and CCL21 [9]. Infected cells do not seem to present antigen in the lungs, but in the lymph nodes, where they migrate through a CCL19/CCL21-dependent mechanism [29]. We did not detect CCL21 expression and very low levels of CCL19 which, in any case, was not expressed differentially when monocytes were infected with mycobacteria (data not shown). Nevertheless, immature non-infected dendritic cells do not express CCR7. Instead, they express CCR6, the receptor of CCL20 [8], the chemokine responsible for transversing the alveolar epithelium [7]. Lee et al. have shown that an M. tuberculosis antigen activates monocytes to produce the chemokine [10], but they did not test live bacilli. We show that M. tuberculosis also induces a marked increase in CCL20 expression. This induction has already been observed for M. avium[30], but we have found that both M. avium and M. kansasii, another non-tuberculous mycobacteria, induce much lower expression of the gene. Although CCL20 is expressed constitutively in many cell types, M. tuberculosis seems to favour a strong increment of its transcription levels. Furthermore, we have found evidence that the attraction of immature dendritic cells by supernatants of M. tuberculosis-infected cells was abolished fully by anti-CCL20, suggesting that it is the main chemokine produced for this purpose.

We do not know whether CCL20 represents a successful mechanism in the immune response to tuberculosis, although it participates in the first steps of an adaptive response [8], which is thought to be responsible for the appropriate control of the disease in most infected individuals [27]. When tuberculosis may not be controlled, the ability of M. tuberculosis to multiply in dendritic cells attracted by CCL20 may contribute to the survival and dissemination of the bacilli. It has been shown in the murine model that exposure to M. tuberculosis determines the accumulation and infection of dendritic cells in the lungs [28]. Nevertheless, there is some controversy regarding the ability of mycobacteria to multiply inside these cells. Förtsch et al. [6] and Buettner et al. [31] have reported the intracellular proliferation of M. tuberculosis. Conversely, these findings were not corroborated by Tailleux et al., who concluded that intracellular survival of M. tuberculosis was reduced in these cells. They argued that the discrepancy with Förtsch et al. might be the consequence of removing the dendritic differentiating agents (IL-4 and GM-CSF) at the moment of infection, which induced a reversion of cells to a macrophage-like phenotype [32]. However, we kept the cytokines during the infections with no adverse effect in the intracellular multiplication of the microorganism. Moreover, we have also determined that non-tuberculous mycobacteria also proliferate, confirming the results obtained for M. avium by Mohagheghpour et al. [33]. Consequently, the infection of dendritic cells may, in some circumstances, be advantageous for M. tuberculosis.

The inhibition of CCL20 expression by IFN-γ is interesting, although difficult to interpret. Lee et al. found that IFN-γ increased production of CCL20 in monocytes activated with the M. tuberculosis 30-kDa antigen [10]. Our results with live bacilli differ, because we observed inhibition of its expression. In fibroblasts, a different cellular model, the production of CCL20 induced by IL-1β is also inhibited by IFN-γ[34]. It is possible that although the production of CCL20 may be important in the onset of the infection, it plays a less significant role afterwards and is down-regulated by IFN-γ, a critical cytokine in the response to tuberculosis [23].

We have also found that CCL20 affects other biological activities of infected cells, including the inhibition of ROS production. ROS may modulate macrophages by affecting many signalling molecules, as has been shown for transcription factors, protein tyrosine kinases, mirogen-activated protein (MAP) kinases or small guanosine triphosphatases (GTPases) [35]. In fact, experiments with murine macrophages deficient in phagocyte oxidase revealed that the NADPH oxidase activity has a major influence in the transcriptome resulting from M. tuberculosis infection [17]. Although the participation of ROS in mycobacteria killing is uncertain, CCL20 inhibition of ROS generation may be very important for other monocyte/macrophage activities. The inhibitory effect of CCL20 is noteworthy considering that other chemokines, including platelet factor 4 (PF-4) (CXCL4), MCP-1 (CCL2), MIP-1α (CCL3) and regulated upon activation normal T cell expressed and secreted (RANTES) (CCL5), enhance ROS production in monocytes [15,36].

Possibly as a consequence of the inhibition of ROS production, CCL20 also inhibits apoptosis in infected monocytes. Although infected murine macrophages have been described to undergo apoptosis through a ROS-independent mechanism [37], we present evidence that this may not be the case in human monocytes, because we show that the NADPH oxidase inhibitor DPI decreases the number of apoptotic cells. A similar situation has been described for M. tuberculosis-infected neutrophils [19]. Apoptosis has been considered a mechanism of host defence because it denies the bacilli a favourable environment, and because infected apoptotic cells in contact with fresh macrophages reduce the viability of mycobacteria. In this context, inhibition of apoptotic activity by CCL20 may be viewed as detrimental to the host, because it allows prolonged survival of the bacillus in its target cell, and evasion from the phagocytosis by fresh activated macrophages. In contrast, at high MOI, cell death helps M. tuberculosis to exit the macrophage to infect new cells [20]. Therefore, although inhibition of apoptotic activity by CCL20 may favour the pathogen, it is difficult to state that this phenomenon is advantageous only to the bacteria.

Other groups have reported that CCL20 is produced in patients who do not control the disease. Lee et al. have detected the presence of CCL20 in bronchoalveolar lavage of tuberculosis patients. Moreover, human monocyte-derived macrophages from tuberculosis patients produce more CCL20 when activated with the 30-kDa antigen of M. tuberculosis than cells from healthy controls [10], which might reflect the expression of a chemokine from patients in whom the immune response has failed. This idea is in line with the results obtained by Thuong et al., who analysed the gene expression profile of monocyte-derived macrophages stimulated with whole cell lysates of M. tuberculosis H37Rv. They compared the profiles of macrophages obtained from three groups: volunteers heavily exposed to M. tuberculosis with no active disease, patients with pulmonary tuberculosis and patients with meningitis tuberculosis, and found that, among other genes, CCL20 was expressed more in the stimulated macrophages from patients with any of the two forms of the disease compared with individuals with no active tuberculosis [38]. The correlation between CCL20 production and severity of the disease may not have a cause–effect relationship, but it should not be ruled out that in a chronic disease such as tuberculosis CCL20 may have some negative influence.

Although CCL20 may be essential in the development of adaptive immunity to tuberculosis, which in most cases is successful, there is a possibility that M. tuberculosis may benefit from some of the chemokine biological activities. Understanding of the fine regulation of CCL20 production in tuberculosis patients may help to underscore new pathogenic mechanisms of the bacilli.


This work was supported by Junta de Castilla y León [LE07/04], Fondo de Investigaciones Sanitarias [PI05/1288] and Fondo Caja de Burgos a la Investigación Clínica. We thank the nurses that helped us with the blood collection. Dr Rivero-Lezcano is a member of the Fundación Instituto de Estudios de Ciencias de la Salud de Castilla y León and participates in the SACYL research program. Reyes-Ruvalcaba is supported by Universidad Autónoma de Ciudad Juárez and fellowship UACDJ-139 from Public Education Secretary (México). González-Cortés is supported by the Instituto de Salud Carlos III program for national health system research support.




1. World Health Organization. WHO report 2009: Global tuberculosis control: epidemiology, strategy, financing. Geneva: WHO; 2009. WHO/HTM/TB/2009·411.
2. Milburn HJ. Primary tuberculosis. Curr Opin Pulm Med. 2001;7:133–41. [PubMed]
3. Uehira K, Amakawa R, Ito T, et al. Dendritic cells are decreased in blood and accumulated in granuloma in tuberculosis. Clin Immunol. 2002;105:296–303. [PubMed]
4. Martino A. Mycobacteria and innate cells: critical encounter for immunogenicity. J Biosci. 2008;33:137–44. [PubMed]
5. Algood HM, Chan J, Flynn JL. Chemokines and tuberculosis. Cytokine Growth Factor Rev. 2003;14:467–77. [PubMed]
6. Förtsch D, Röllinghoff M, Stenger S. IL-10 converts human dendritic cells into macrophage-like cells with increased antibacterial activity against virulent Mycobacterium tuberculosis. J Immunol. 2000;165:978–87. [PubMed]
7. Osterholzer JJ, Ames T, Polak T, et al. CCR2 and CCR6, but not endothelial selectins, mediate the accumulation of immature dendritic cells within the lungs of mice in response to particulate antigen. J Immunol. 2005;175:874–83. [PMC free article] [PubMed]
8. Schutyser E, Struyf S, Damme JV. The CC chemokine CCL20 and its receptor CCR6. Cytokine Growth Factor Rev. 2003;14:409–26. [PubMed]
9. Esche C, Stellato C, Beck LA. Chemokines: key players in innate and adaptive immunity. J Invest Dermatol. 2005;125:615–28. [PubMed]
10. Lee JS, Lee JY, Son JW, et al. Expression and regulation of the CC-chemokine ligand 20 during human tuberculosis. Scand J Immunol. 2008;67:77–85. [PubMed]
11. Tomimori K, Uema E, Teruya H, et al. Helicobacter pylori induces CCL20 expression. Infect Immun. 2007;75:5223–32. [PMC free article] [PubMed]
12. Sperandio B, Regnault B, Guo J, et al. Shigella flexneri subverts the host innate immune response through manipulation of antimicrobial peptide gene expression. J Exp Med. 2008;205:1121–32. [PMC free article] [PubMed]
13. Latvala S, Pietilä TE, Veckman V, et al. Potentially probiotic bacteria induce efficient maturation but differential cytokine production in human monocyte-derived dendritic cells. World J Gastroenterol. 2008;14:5570–83. [PMC free article] [PubMed]
14. Rossi D, Zlotnik A. The biology of chemokines and their receptors. Annu Rev Immunol. 2000;18:217–42. [PubMed]
15. Pervushina O, Scheuerer B, Reiling N, et al. Platelet factor 4/CXCL4 induces phagocytosis and the generation of reactive oxygen metabolites in mononuclear phagocytes independently of Gi protein activation or intracellular calcium transients. J Immunol. 2004;173:2060–7. [PubMed]
16. Ehrt S, Schnappinger D. Mycobacterial survival strategies in the phagosome: defence against host stresses. Cell Microbiol. 2009;11:1170–8. [PMC free article] [PubMed]
17. Ehrt S, Schnappinger D, Bekiranov S, et al. Reprogramming of the macrophage transcriptome in response to interferon-γ and Mycobacterium tuberculosis: signaling roles of nitric oxide synthase-2 and phagocyte oxidase. J Exp Med. 2001;194:1123–40. [PMC free article] [PubMed]
18. Simon HU, Haj-Yehia A, Levi-Schaffer F. Role of reactive oxygen species (ROS) in apoptosis induction. Apoptosis. 2000;5:415–8. [PubMed]
19. Perskvist N, Long M, Stendahl O, Zheng L. Mycobacterium tuberculosis promotes apoptosis in human neutrophils by activating caspase-3 and altering expression of Bax/Bcl-xL via an oxygen-dependent pathway. J Immunol. 2002;168:6358–65. [PubMed]
20. Lee J, Hartman M, Kornfeld H. Macrophage apoptosis in tuberculosis. Yonsei Med J. 2009;50:1–11. [PMC free article] [PubMed]
21. Fazal N, Bartlett R, Lammas DA, Kumararatne DS. A comparison of the different methods available for determining BCG-macrophage interactions in vitro, including a new method of colony counting in broth. FEMS Microbiol Immunol. 1992;5:355–62. [PubMed]
22. Pfaffl MW. A new mathematical model for relative quantification in real-time RT-PCR. Nucleic Acids Res. 2001;29:e45. [PMC free article] [PubMed]
23. Ottenhoff TH, Verreck FA, Hoeve MA, Vosse DV. Control of human host immunity to mycobacteria. Tuberculosis (Edinb) 2005;85:53–64. [PubMed]
24. Esin S, Batoni G, Counoupas C, et al. Direct binding of human NK cell natural cytotoxicity receptor NKp44 to the surfaces of mycobacteria and other bacteria. Infect Immun. 2008;76:1719–27. [PMC free article] [PubMed]
25. Flores-Villanueva PO, Ruiz-Morales JA, Song C-H, et al. A functional promoter polymorphism in monocyte chemoattractant protein-1 is associated with increased susceptibility to pulmonary tuberculosis. J Exp Med. 2005;202:1649–58. [PMC free article] [PubMed]
26. Zhu K, Shen Q, Ulrich M, Zheng M. Human monocyte-derived dendritic cells expressing both chemotactic cytokines IL-8, MCP-1, RANTES and their receptors, and their selective migration to these chemokines. Chin Med J. 2000;113:1124–8. [PubMed]
27. Bhatt K, Salgame P. Host innate immune response to Mycobacterium tuberculosis. J Clin Immunol. 2007;27:347–62. [PubMed]
28. Humphreys IR, Stewart GR, Turner DJ, et al. A role for dendritic cells in the dissemination of mycobacterial infection. Microbes Infect. 2006;8:1339–46. [PubMed]
29. Wolf AJ, Linas B, Trevejo-Nuñez GJ, et al. Mycobacterium tuberculosis infects dendritic cells with high frequency and impairs their function in vivo. J Immunol. 2007;179:2509–19. [PubMed]
30. Blumenthal A, Lauber J, Hoffmann R, et al. Common and unique gene expression signatures of human macrophages in response to four strains of Mycobacterium avium that differ in their growth and persistence characteristics. Infect Immun. 2005;73:3330–41. [PMC free article] [PubMed]
31. Buettner M, Meinken C, Bastian M, et al. Inverse correlation of maturity and antibacterial activity in human dendritic cells. J Immunol. 2005;174:4203–9. [PubMed]
32. Tailleux L, Neyrolles O, Honoré-Bouakline S, et al. Constrained intracellular survival of Mycobacterium tuberculosis in human dendritic cells. J Immunol. 2003;170:1939–48. [PubMed]
33. Mohagheghpour N, Vollenhoven VA, Goodman J, Bermudez LE. Interaction of Mycobacterium avium with human monocyte-derived dendritic cells. Infect Immun. 2000;68:5824–9. [PMC free article] [PubMed]
34. Hosokawa Y, Hosokawa I, Ozaki K, Nakae H, Matsuo T. Increase of CCL20 expression by human gingival fibroblasts upon stimulation with cytokines and bacterial endotoxin. Clin Exp Immunol. 2005;142:285–91. [PMC free article] [PubMed]
35. Forman HJ, Torres M. Signaling by the respiratory burst in macrophages. IUBMB Life. 2001;51:365–71. [PubMed]
36. Vaddi K, Newton RC. Comparison of biological responses of human monocytes and THP-1 cells to chemokines of the intercrine-β family. J Leukoc Biol. 1994;55:756–62. [PubMed]
37. Rojas M, Barrera LF, García LF. Induction of apoptosis in murine macrophages by Mycobacterium tuberculosis is reactive oxygen intermediates-independent. Biochem Biophys Res Commun. 1998;247:436–42. [PubMed]
38. Thuong NT, Dunstan SJ, Hong Chau TT, et al. Identification of tuberculosis susceptibility genes with human macrophage gene expression profiles. PLoS Pathog. 2008;4:e1000229. [PMC free article] [PubMed]

Articles from Clinical and Experimental Immunology are provided here courtesy of British Society for Immunology
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...