MAJOR ARTICLE
Down-Regulation of T Helper 1 Responses to Mycobacterial Antigens Due to Maturation of Dendritic Cells by 10-kDa Mycobacterium tuberculosis Secretory Antigen Krishnamurthy Natarajan,1 Vinoth K. Latchumanan,1 Balwan Singh,1 Sarman Singh,2 and Pawan Sharma1 1
Immunology Group, International Centre For Genetic Engineering and Biotechnology, Aruna Asaf Ali Marg, and 2Department of Laboratory Medicine, All India Institute of Medical Sciences, Ansari Nagar, New Delhi, India
Interactions of 10-kDa Mycobacterium tuberculosis secretory antigen (MTSA) with dendritic cells (DCs) were investigated to elucidate the role of secretory antigens in regulating immune responses to M. tuberculosis early in the course of infection. MTSA induced the maturation of different DC subsets. The cytokine profiles of these DCs were characteristic to each DC subset. Of interest, coculture of M. tuberculosis whole-cell extract (CE)–pulsed, MTSA-matured DCs with CE-specific T cells led to a marked reduction in interleukin (IL)–2 and interferon (IFN)–g production, thereby down-regulating proinflammatory responses to mycobacterial antigens. Attenuation of IL-2 and IFN-g levels of CE-specific T cells also was obtained when M. tuberculosis culture filtrate protein–activated DCs were employed as antigen-presenting cells, which suggests that MTSAs induce maturation of DCs at sites of infection, probably to down-regulate proinflammatory immune responses to mycobacteria that may subsequently be released from infected macrophages. Infection with Mycobacterium tuberculosis continues to be a major cause of mortality and morbidity throughout the world, resulting in 3 million deaths and 18 million new cases of tuberculosis each year [1–3]. Although immunization with M. bovis bacille CalmetteGue´rin (BCG) is still considered to be the reference standard against which all other vaccines are measured, its efficacy varies from 0% to 85% in different studies
Received 30 August 2002; accepted 25 November 2002; electronically published 6 March 2003. Animal studies were approved by the Institutional Animal Ethics Committee of India. Financial support: Defence Research and Development Organization, Government of India (grant DALS/48222/LSRB/22/ID/RD/-81 to K.N. and P.S.); Department of Biotechnology, Government of India (grant BT/PR2423/Med/13/087/2001 to P.S.). Reprints or correspondence: Dr. Krishnamurthy Natarajan, Immunology Group, International Centre for Genetic Engineering and Biotechnology, Aruna Asaf Ali Marg, 110-067 New Delhi, India (
[email protected]). The Journal of Infectious Diseases 2003; 187:914–28 2003 by the Infectious Diseases Society of America. All rights reserved. 0022-1899/2003/18706-0006$15.00
914 • JID 2003:187 (15 March) • Natarajan et al.
and in different geographic regions [4–7]. Therefore, any step toward the development of vaccines or vaccine candidates requires a thorough understanding of the protective immune responses generated against this pathogen [8, 9]. Mycobacteria persist in macrophages within the granuloma in the organs of infected hosts [10]. From the phagosomes, where they reside, they are believed to secrete proteins, also called “secretory antigens” [11]. Numerous proteins secreted by the bacteria have been described, and many have been postulated to contribute to the development of protective immunity by serving as targets for the immune system early in the infection where they are likely to be taken up by antigen-presenting cells (APCs) [12–15]. Among the most potent of these APCs are the different subsets of DCs that have the ability to stimulate quiescent, naive, and memory T lymphocytes [16]. DCs exist at various states of activation and maturation that are defined by distinct phenotypic and functional modalities [17]. For example, immature DCs are programmed for antigen capture and display very low levels of T
cell–stimulatory properties (i.e., they express low levels of surface major histocompatibility complex [MHC] and costimulatory molecules). After contact with various stimuli—such as lipopolysaccharide (LPS), tumor necrosis factor (TNF)–a, CD40 ligand (via cognate interactions with T cells), and certain antigens [16, 17]—they undergo a process of maturation. During maturation, they up-regulate their MHC (class I and II) and costimulatory molecules (CD80, CD86, CD40, and CD54) and are very efficient T cell stimulators [18]. Therefore, antigens that are able to induce maturation of DCs play a major role in defining the character of primary immune responses against the pathogen and, thereby, have an important role in determining the course of an infection [19]. A recent report identified 10-kDa M. tuberculosis secretory antigen (MTSA), which was able to prime delayed-type hypersensitivity responses in M. tuberculosis–infected guinea pigs but not in animals infected with M. bovis BCG [20]. MTSA is a product of the Rv3874 gene in the mycobacterial genome and is not expressed by other members of the mycobacterial complex (e.g., M. avium and M. bovis BCG). Recent reports document the importance of MTSA (also known as culture filtrate protein [CFP]–10) in generating protective immune responses against M. tuberculosis [21]. CFP-10–pulsed, monocyte-derived DCs were used to isolate CD8⫹ T cell clones that interacted with M. tuberculosis– but not M. bovis BCG–infected targets [22]. Furthermore, owing to its absence in M. bovis BCG strains used for vaccinations, CFP-10 has been proposed as an important candidate in the diagnosis of M. tuberculosis [23, 24]. We recently showed that MTSA induces the secretion of TNF-a in macrophages and synergizes with IFN-g to induce nitric oxide secretion in macrophages [25]. We also recently showed that MTSA and other mycobacterial antigens induce the differentiation of DCs from bone marrow (BM) precursors [26]. To further characterize the interactions of MTSA with DCs, we examined the interactions of MTSA with DC subsets, to understand the role of mycobacterial antigens in regulating immune responses early in the course of infection. Our results indicate that MTSA induces the full maturation of DCs. These DCs, in effect, down-regulate proinflammatory responses against M. tuberculosis whole-cell extract (CE) and, along with other such secretory proteins, likely plays an important role in regulation of immune responses to mycobacteria. The functional implications of DC maturation by MTSA are discussed.
MATERIALS AND METHODS Animals. Female BALB/c mice (4–6 weeks old) were used in the study for all experiments involving DCs. For enrichment of T cells, either BALB/c or C57BL/6 mice were used. All animals were maintained under pathogen-free, environment-controlled conditions in the small animal facility of the Interna-
tional Centre for Genetic Engineering and Biotechnology (New Delhi). Materials. Fluorescein isothiocyanate (FITC)–tagged monoclonal antibodies against mouse cell-surface molecules (CD80 [clone 1G10], CD86 [clone GL-1], CD54 [clone 3E2], I-Ad [clone AMS-32.1], H-2Dd [clone 3-25.4], CD40 [clone 3/23], CD4 [clone GK 1.1], and CD8a [clone Ly-2]); biotin-conjugated antibodies to CD11c (clone HL3) and CD90 (Thy 1.2; clone 52-2.1); purified anti-CD40 (clone HM-40) and CD16/CD32 (FCgR, clone 2.4G2); and isotype-matched control antibodies were purchased from BD Pharmingen. FITC-conjugated antibody to F4/80 (clone CI: A3-1) was obtained from Serotec. AntiCD4, anti-CD8, anti-CD90, anti-B220, anti-CD11b, anti-CD11c, anti–I-A, and anti-CD19–coated magnetic beads were obtained from Miltenyi Biotec. Mouse recombinant granulocyte-macrophage colony-stimulating factor (GM-CSF) and ELISA kits for the estimation of mouse cytokines were purchased from R&D Systems. Recombinant TNF-a, LPS, polymixin B sulfate, hen egg lysozyme, ovalbumin, polystyrene carboxylate–modified fluorescent beads (0.5 mm), and an E-toxate endotoxin detection kit were obtained from Sigma. M. tuberculosis CE and M. tuberculosis CFPs were obtained from Tuberculosis Research Materials and Vaccine Testing (Colorado State University, Fort Collins). The details of their preparation and composition can be viewed at http://www.cvmbs.colostate.edu/microbiology/tb. Expression and purification of MTSA. Expression and purification of MTSA was done as described elsewhere [25]. In brief, the open-reading frame Rv3874 of M. tuberculosis–encoding MTSA was polymerase chain reaction–amplified from the genomic DNA of a local clinical isolate using the primers 5-GCGGATCCCATGGCAGAGATGAAGACCG-3 (forward) and 5-CCCAAGCTTGTCAGAAGCCATTTGCGAG3 (reverse), with BamHI and HindIII as flanking enzyme sites (underlined). The polymerase chain reaction product (GenBank accession no. AF419854) was first directly cloned into an intermediate vector (pGEM-T Easy; Promega) and sequenced to ascertain the identity. After verification of the sequence, the full-length gene was subcloned into the BamHI and HindIII sites of the bacterial expression vector (pQE31; Qiagen). The recombinant protein, expressed as a polyhistidine-tagged protein, was purified by nickel-nitrilotriacetic acid (Ni-NTA) metal affinity chromatography, according to the manufacturer’s instructions (Qiagen), for purification of recombinant protein in its native form. The purity of MTSA was further confirmed by ion-exclusion high-performance liquid chromatography (HPLC), using a POROSHQ column (PerSeptive Biosystems) with a matrix of crosslinked polystyrene-divinylbenzene flow-through particles coated with fully quarternized polyethyleneimine that is completely ionized over a pH range of 1–14. One milligram of the protein was injected, and the profile was recorded. Maturation of Dendritic Cells by MTSA • JID 2003:187 (15 March) • 915
Generation of DCs from BM. Generation of DCs from BM was done as described elsewhere [26]. In brief, 3 ⫻ 10 6 lymphoid and I-A⫹-depleted BM cells from the tibias and femurs of BALB/c mice were cultured in 6-well plates in RPMI 1640 medium containing 10% fetal calf serum (FCS), 0.05 M 2-mercaptoethanol, 1 mM sodium pyruvate, and 15 ng/mL GM-CSF, with periodic changes of the culture medium. On day 3, loosely-adherent cells were cultured in fresh medium containing GM-CSF and, when required, stimulated with various antigens for 24 h. For some experiments, 25 mg of MTSA was digested with varying amounts of pancreatic trypsin (Gibco BRL) in 200 mM Tris-Cl buffer (pH 7.7) for 10 min at 25C. Enzyme reaction was stopped by the addition of 20% FCS. One unit of trypsin was defined by the manufacturer as 1 mg of enzyme required to completely digest 250 mg of target protein. Trypsin-treated MTSA then was used for the stimulation of BM-derived DCs (BMDCs), as described above. To check for specificity, MTSA was incubated with the F(ab)2 fragment of anti-MTSA polyclonal antibody (obtained by digestion with pepsin and subsequent purification over a protein G column) for 60 min and later was added to cultures. The C-terminal fragment of Plasmodium falciparum merozoite surface protein (MSP)–119 kDa, expressed and purified in the same way as MTSA, also was used as a negative control. Cells at the end of incubation in all sets were either analyzed for the levels of surface molecules by flow cytometry, as described elsewhere [27], or were used in T cell stimulation experiments, as described below. Enrichment of splenic DCs. Spleens from 6–8 mice were pooled and cut into small fragments. These were then digested in RPMI 1640 medium containing 10% FCS, 1 mg/mL type III collagenase, and 325 kU/mL DNase I, with periodic pipetting for 25 min at room temperature. EDTA (0.1 M [pH 7.2]) was added for an additional 5 min to allow disruption of the DC–T cell complexes. Cells were washed and resuspended in Nycodenz (1.077 g/mL; Sigma), overlaid on an additional layer of Nycodenz, and centrifuged at 1700 g for 20 min. The lowdensity fraction then was collected. B and T cells were removed from the low-density fraction by incubation with titrated levels of anti-CD19, anti-CD45R, and anti-CD90 microbeads and by separation through magnetic cell sorter (MACS) columns (Miltenyi Biotec). The percentage of CD11c⫹ cells in the resulting population was found to be 85%–90%, as monitored by fluorescence-activated cell sorter (FACS; FACSCalibur; Becton Dickinson). The cells then were fractionated further into CD11c⫹/ CD8a⫹ and CD11c⫹/CD8a⫺ populations after MACS. Enriched splenic DCs (90%–95%) were cultured with different antigens for 48 h and either were analyzed for changes in the surface expression of markers or were cocultured with allogeneic or syngeneic T cells. Phagocytosis by DCs. BMDCs were stimulated with 25 mg/mL either MTSA or MSP-119 kDa or 20 mg/mL LPS for 24 h. 916 • JID 2003:187 (15 March) • Natarajan et al.
Cells were washed in Hanks’ balanced salt solution (HBSS) and resuspended in PBS containing 5% bovine serum albumin and fluorescent carboxylate latex beads for 4 h at 37C. At the end of the incubation, cells were extensively washed in HBSS and fixed in PBS containing 0.1% paraformaldehyde. Cells then were analyzed by flow cytometry. Enrichment of T lymphocytes. Enrichment of T lymphocytes was done as described elsewhere [28]. In brief, either inguinal lymph nodes or splenocytes from 4–6-week-old BALB/ c and C57BL/6 mice, respectively, were first depleted of adherent cells by panning over plastic plates. B lymphocytes, residual DCs, and macrophages then were removed by 2 rounds of incubation with anti-CD19–, anti-CD45R–, anti-CD11c–, anti-CD11b–, and anti-I-A–coated magnetic beads, followed by separation through MACS columns. The purity of the resulting population of T cells obtained in this fashion was 95%–98%, as determined by anti-CD90 stained cells by flow cytometry. For some experiments, CD4⫹ or CD8⫹ populations of enriched T cells from the lymph nodes of BALB/c mice were further negatively enriched using anti-CD4– or anti-CD8–coated magnetic beads and purification over MACS columns. The resulting CD4⫹ or CD8⫹ T cells were 90%–95% pure, as determined by flow cytometry. The percentage of I-A⫹ cells in all the fractions was found to be !0.5%. Allogeneic mixed leukocyte reactions. For measuring T cell proliferation, 3 ⫻ 10 5 enriched allogeneic C57BL/6 T cells were cultured with 0.6 ⫻ 10 5 irradiated (3000 rads) antigenactivated BM cells or splenic DCs in 96-well plates for 72 h. [3H]-thymidine (1.0 mCi) was added 16 h before harvesting and counting. For measuring cytokine levels, 3 ⫻ 10 6 T cells were cocultured with 0.6 ⫻ 10 6 irradiated DCs in 24-well plates for 48 h, and culture supernatants then were screened for the presence of cytokines, as described below. In vitro syngeneic T cell stimulation. BALB/c mice were immunized subcutaneously (sc) at the base of the tail with various antigens (50 mg/mouse) in incomplete Freund’s adjuvant for 7 days and boosted with a repeat immunization for an additional 7 days. Inguinal lymph nodes from these mice were removed, and T cells (both CD4⫹ and CD8⫹) were enriched as described above. Enriched T cells (3 ⫻ 10 6 cells) were cocultured with 0.6 ⫻ 10 6 antigen-stimulated irradiated DCs for 48 h, and culture supernatants were analyzed for cytokines. In vivo syngeneic T cell stimulation. MTSA- or CE-matured DC subsets (2 ⫻ 10 6 cells) were separately injected sc into naive BALB/c mice at the base of the tail. This was followed by sc transfer of an equal number of either MTSA- or CEspecific enriched T cells at the base of the tail 24 h later. Inguinal lymph nodes were excised 3 days later and cultured in vitro for an additional period of 72 h, and culture supernatants were analyzed for cytokines. Estimation of cytokines. Culture supernatants of DCs or
DC–T cell cocultures at the end of each incubation period were analyzed for the levels of TNF-a, IL-12p40, IFN-g, or IL-10 by use of a sandwich ELISA protocol, as recommended by the manufacturer (R&D Systems). The sensitivity ranges of the cytokines were 31.2–2000 pg/mL for IL-12p40, 15.6–1000 pg/mL for TNF-a, 31.2–2000 pg/mL for IFN-g, and 31.2–2000 pg/mL for IL-10. The samples were correspondingly diluted to obtain values within the linear range of the standards. Quantitation was made against a standard curve obtained for individual cytokine standards provided by the manufacturer.
RESULTS MTSA induces the maturation of BMDCs. We first examined the ability of MTSA to induce the maturation of DCs. It is now well established that culturing BM cells with GM-CSF differentiates them into DC-like APCs that display a predominantly immature phenotype [29, 30]. We first ascertained the purity of MTSA on a silver-stained SDS-PAGE gel and by HPLC analyses. As shown in figure 1A, the fraction eluted from the NiNTA column contained a single band at 10 kDa, the size corresponding to MTSA. No other protein band was present at detectable levels, even in the lane loaded with 50 mg of the protein, which indicates that the fraction contained only a single species of the size corresponding to MTSA. Furthermore, the HPLC profile with 1 mg of MTSA also showed a single peak, with no apparent detectable levels of any other species, again indicating the homogeneity of the fraction containing MTSA. We next stimulated GM-CSF–differentiated BMDCs with MTSA and monitored the changes in the levels of various molecules on these cells. Figure 1B gives the FACS profiles of CD11c⫹ cells stained for the indicated markers. MTSA further up-regulated the levels of both MHC and costimulatory molecules on BMDCs. In particular, CD80 levels were up-regulated by 10-fold, whereas CD86 levels were enhanced by 4-fold. Furthermore, the expression of myeloid marker F4/80 antigen, which is known to be expressed on matured DCs [31], also was enhanced on BMDCs after addition of MTSA. Dose and time response profiles (data not shown) revealed that stimulation with 25 mg/mL MTSA for 24 h gave the best response; hence, all subsequent experiments with BMDCs were performed with that concentration and time of exposure. Because stimulation of allogeneic T cells at a low stimulator-to-responder ratio is a well established marker for mature DCs and myeloid cells [16, 17], we stimulated BMDCs with MTSA and examined the subsequent allogeneic T cell responses thus generated. As shown in figure 1C, the proliferation of allogeneic T cells showed a 2-fold increase in both [3H]-thymidine incorporation and IL-2 production after stimulation with MTSA. Furthermore, the level of IFN-g on allogeneic T cells also increased by nearly 5-fold when BMDCs incubated with MTSA
were used in cocultures (figure 1C). IL-10 levels were largely unaffected. Consistent with earlier reports [16], stimulation with keyhole limpet hemocyanin antigen did not have any effect on DCs, either at the level of surface markers or on the extent of allogeneic T cell stimulation (data not shown). Because the MTSA used in the study was recombinant expressed, it was necessary to rule out the possibility of the observed effects being mediated by endotoxins and/or other low– molecular-weight contaminants in the purified protein. We used various approaches to accomplish this. First, we estimated the endotoxins levels in all the batches of MTSA that were used in the study by use of the E-Toxate kit from Sigma. Endotoxin levels in every batch of MTSA used in the study were tested at 4 different doses, from 10–100 mg/mL. In all the batches, the endotoxin level was !0.03 endotoxin units/mL, as determined by the absence of a hard-gel formation in the 0.03 endotoxin unit standard provided in the kit. Since the concentration of MTSA used in the study was 25 mg/mL, the corresponding levels of endotoxins would be 5-fold lower, indicating that the observed effects in figure 1 were attributable to MTSA. We further ruled out the effects of endotoxins by incubating MTSA with polymixin B sulfate, which is known to inactivate LPS and related endotoxins [32]. BMDCs were then stimulated with polymixin B sulfate–treated MTSA, and the levels of various markers on the cell surface were monitored by FACS. As shown in table 1, although treatment of LPS with polymixin B sulfate completely abolished all its effects, treatment of MTSA with polymixin B sulfate had no effect on the increase in the surface levels of various MHC and costimulatory molecules, compared with untreated MTSA (table 1); this result indicates that endotoxins were not responsible for the observed effects of MTSA. Furthermore, because certain low-molecular-weight compounds, such as polypeptides and other bacterial components [33], may also be present in the recombinant preparation of MTSA and have been shown to activate DCs, we treated MTSA with increasing doses of trypsin and examined the retention of its activity. As shown in table 1, digestion of MTSA with trypsin led to a complete loss of activity (as measured by the mean fluorescence intensity [MFI] of various markers), even at a level of 0.01 U of trypsin, which indicates that intact MTSA in its native form was required for inducing the maturation of DCs. Incubation of MTSA with the F(ab)2 fragment of anti-MTSA did not result in any increase in the surface levels of various markers on DCs, which indicates that the observed effects were specific to MTSA. Incubation of MTSA with preimmune serum had no effect on its activity (data not shown). Furthermore, stimulation of DCs with similarly expressed and purified MSP-119 kDa had no effect on the levels of various markers. The results in table 1 thus indicate that the observed effects in figure 1 were specific to MTSA and not obtained by any contaminant (such as endotoxins, low-molecularweight compounds, or polypeptides) in the recombinant protein. Maturation of Dendritic Cells by MTSA • JID 2003:187 (15 March) • 917
Figure 1. Mycobacterium tuberculosis secretory antigen (MTSA) induces the maturation of immature dendritic cells (DCs). A, Silver-stained 12.5% SDS-PAGE gel loaded with fraction containing MTSA. Lanes 1–5, 1, 5, 10, 25, and 50 mg of MTSA, respectively. M, Molecular weight markers. A high-performance liquid chromatography profile of MTSA corresponding to 1 mg of protein is depicted on the right. B, Granulocyte-macrophage colonystimulating factor–differentiated day-3 DCs from the bone marrow (BM) were stimulated with 25 mg/mL of MTSA for 24 h. At the end of incubation, aliquots of cells were stained for the various markers and analyzed by fluorescence-activated cell sorter (see Materials and Methods). Cell-surface levels of indicated markers on CD11c⫹-gated cells in the presence (thick line) or absence (dashed line) of MTSA from 1 of 4 experiments are shown. The thin line in all the histograms indicates staining with corresponding isotype-matched control antibodies. C, MTSA-stimulated DCs were cocultured with C57BL/6-enriched T cells in 96-well plates for 72 h for measuring [3H]-thymidine incorporation; 1.0 mCi/well of [3H]-thymidine was added 16 h before harvesting and counting (see Materials and Methods). For measuring cytokine levels, DCs were cocultured with T cells in 24-well plates for 48 h, and culture supernatants were screened for the levels of indicated cytokines. Differences in the levels of interferon (IFN)–g between immature and MTSA-matured DCs in panel C were significant at P ! .05 (Student’s t test). Data from 1 of 4 experiments are shown. BMDC, BM-derived DCs.
Because DCs at various states of activation and maturation perform definite functions, with respect to antigen capture versus T cell stimulation [16], we also investigated whether MTSA induced either partial or full maturation of BMDCs. MTSAmatured BMDCs were incubated with stimulants that induce full maturation of DCs, such as LPS, anti-CD40, and TNF-a [16, 17]. The responses of these DCs were evaluated for both cell-surface phenotype and extent of allogeneic T cell stimu-
918 • JID 2003:187 (15 March) • Natarajan et al.
lation. We observed that the addition of these stimuli to MTSAmatured BMDCs had no effect on either the levels of MHC and costimulatory molecules or the extent of allogeneic T cell stimulation (data not shown). Together with the ability to boost the allogeneic T cell stimulation, the collective results in figure 1 indicate that MTSA induced the full maturation of BMDCs, which is consistent with the higher levels of MHC and costimulatory marker expression.
Table 1. Maturation of bone marrow–derived dendritic cells (BMDCs) is specific to Mycobacterium tuberculosis secretory antigen (MTSA). Cell marker Group
CD80
CD86
I-A
H-2D
CD40
CD54 150 18
80 0.9
50 4.5
35 3.2
55 14
30 3.1
BMDCs ⫹ MTSA
780 65
400 39
500 65
650 60
250 28
800 75
BMDCs ⫹ LPS
850 75
585 60
780 81
720 77
280 30
755 80 145 15
BMDCs
BMDCs ⫹ LPS ⫹ PBa BMDCs ⫹ MTSA ⫹ PBa MTSA ⫹ Trypsin (0.01 U) MTSA⫹ Trypsin (10 U)b MTSA⫹ a-MTSA MSP-119 kDad
c
b
75 7.2
52 4.5
38 4.0
57 6.1
25 1.8
750 65
395 40
520 60
625 65
275 29
825 85
79 7.9
49 4.8
32 2.9
54 5.6
28 2.8
140 14
75 7.2
45 4.3
34 3.5
55 5.7
32 2.5
145 15
78 7.0
48 4.2
32 2.8
48 5.2
32 2.5
148 15
75 6.8
45 3.8
30 2.4
51 4.8
24 2.2
135 12
NOTE. Data are mean SD fluorescence intensities of 3 independent experiments. Day-5 BMDCs were stimulated with 25 mg/mL of MTSA alone or treated with various agents for 24 h, stained for the surface levels of various markers (see Materials and Methods), and analyzed by fluorescence-activated cell sorter. a-MTSA, anti-MTSA; LPS, lipopolysaccharide; MSP-119 kDA, merozoite surface protein–119 kDA; PB, polymixin B sulfate. a
LPS (20 mg/mL) or MTSA (25 mg/mL) was incubated with 25 mg of PB for 60 min and then was added to BMDCs; 24 h later, the cultures were analyzed for the levels of indicated markers. b MTSA (25 mg) was digested with 0.01 or 10 U of pancreatic trypsin and added to cultures. Cell-surface densities of various markers were measured as described above. c MTSA was incubated with the F(ab)2 fragment of a-MTSA antibody for 1 h, added to BMDCs for 24 h, and analyzed as described above. d BMDCs were stimulated with 25 mg/mL recombinant MSP-119 kDa of Plasmodium falciparum expressed in Escherichia coli as a His-tagged protein for 24 h, and surface levels of various markers were measured.
MTSA induces the maturation of splenic DCs. It is well recognized that different subsets of DCs present at various locations in the organism and respond in various ways to antigenic challenges [18]. Among the DC subsets reported in the mouse spleen, the 2 most well characterized are CD8a⫹/CD11c⫹ and CD8a⫺/CD11c⫹ DCs [34]. The CD8a⫹/CD11c⫹ DCs are thought to be of lymphoid origin and primarily induce Th1 responses, whereas the CD8a⫺/CD11c⫹ subset of DCs induce Th2 responses [18]. To determine whether MTSA would also induce the maturation of splenic DCs, we carried out parallel experiments, as were done with BMDCs. Enriched CD8a⫺ or CD8a⫹ DCs from spleens were stimulated with MTSA for different periods of time, and the extent of modulation of cell surface levels of costimulatory and MHC molecules was monitored by flow cytometry. MTSA induced a time- and dosedependent increase in the levels of various molecules (data not shown), with maximum up-regulation obtained 48 h after stimulation with 25 mg/mL MTSA; hence, these conditions were used for all subsequent experiments. Figure 2A shows the fold increase over unstimulated controls in the MFI of various molecules of splenic DCs stimulated with MTSA for 48 h. MTSA up-regulated all the molecules analyzed in both DC subsets. In particular, the levels of MHC class II (I-Ad) and MHC class I (H-2Dd) were found to be enhanced by 110–12-fold in CD8a⫺ DCs and by 4–10-fold in CD8a⫹ DCs. The levels of CD40 and CD54 were also up-regulated by 2–4-fold in both DC subsets (figure 2A). Consistent with the increase in MFI of costimu-
latory and MHC molecules, the extent of allogeneic T cell stimulation, as measured by the levels of IL-2, was enhanced by 2fold; levels of IFN-g and IL-10 were also enhanced by several fold when either MTSA-stimulated CD8a⫹ or CD8a⫺ DCs were used as APCs. (figure 2B). The results in figure 2 thus indicate that MTSA induced the maturation of splenic DC subsets as well. MTSA also induced full maturation of both CD8a⫺ and CD8a⫹ DCs, but various stimuli, such as TNF-a, anti-CD40, and LPS, did not have any significant effect on the levels of MHC and costimulatory molecules on MTSA-matured DCs (data not shown). MTSA-matured DC subsets secrete distinct cytokine patterns. We next examined the nature of cytokine profiles secreted by different DC subsets after MTSA stimulation. Table 2 gives the profiles of cytokines secreted by DCs 24 h after stimulation. Consistent with published reports, unstimulated BMDCs and splenic DCs secreted TNF-a and low levels of IL10 [16, 17]. IFN-g and IL-12p40 were below the level of detection in CD8a⫺ DCs and BMDCs. MTSA stimulation resulted in enhanced secretion of TNF-a by 19-fold in BMDCs and by 2-fold in splenic DCs. IL-12p40 levels were maximally induced in CD8a⫹ DCs after MTSA stimulation, compared with either BMDCs or CD8a⫺ DCs. The low levels of IL-12p40 noted in CD8a⫺ DCs after antigen stimulation are consistent results from a prior study that documented low-to-undetectable levels of IL-12p40 production by this DC subset [18]. CD8a⫹ DCs secreted low levels of IFN-g, which was enhanced by several-
Maturation of Dendritic Cells by MTSA • JID 2003:187 (15 March) • 919
Figure 2. Splenic dendritic cell (DC) subsets are matured by Mycobacterium tuberculosis secretory antigen (MTSA). A, Fold increase in the relative mean fluorescence intensity (MFI) of various cell-surface molecules on CD8a⫺ (a) or CD8a⫹ (b) splenic DCs stimulated with 25 mg/mL of MTSA for 48 h, over unstimulated DCs. Solid bars, MFIs of unstimulated DCs; hatched bars, MFIs of MTSA-stimulated DCs. Groups 1–6 in both panels represent MFIs of CD80, CD86, I-A, H-2D, CD40, and CD54, respectively. All bars represent staining on propidium iodide–excluded CD11c⫹ cells. B, MTSAstimulated CD8a⫺ or CD8a⫹ DCs were cocultured with enriched C57BL/6 T cells for 48 h, and culture supernatants were monitored for the levels of indicated cytokines at the end of the incubation period. Data from 1 of 5 experiments are shown. IFN, interferon; IL, interleukin.
fold after MTSA stimulation. IFN-g secretion was observed in BMDCs and CD8a⫺ DCs after MTSA stimulation. IL-10 levels were marginally altered in BMDCs and CD8a⫺ DCs and, in fact, were not detected in CD8a⫹ DCs after stimulation with MTSA. The results in table 2 thus indicate that MTSA induced distinct patterns of cytokine secretion by different DC subsets that might differentially govern the nature of subsequently elicited T cell responses. MTSA-induced T cell responses are MHC class II restricted. We next explored the kind of antigen-specific T cell responses regulated by different DC subsets stimulated by MTSA. MTSA-matured various DC subsets were cocultured with either unfractionated or CD4⫹ or CD8⫹ MTSA-specific T cell subsets, and the profiles of IL-2, IFN-g, and IL-10 production from the interacting T cells were analyzed. MTSAmatured DC subsets induced the proliferation of MTSA-specific T cells, as reflected in the levels of IL-2 (figure 3). A dominance of IFN-g secretion over IL-10 levels was observed in antigenspecific T cells along with all 3 DC subsets, which is consistent with allogeneic T cell responses. However, despite the fact that CD8a⫹ DCs secreted higher levels of IFN-g and IL-12p40 than did CD8a⫺ DCs and BMDCs, no significant differences were observed in the absolute levels of either IL-2 or IFN-g in T cell cocultures with different DC subsets; however, IL-10 levels 920 • JID 2003:187 (15 March) • Natarajan et al.
were found to be higher in T cells cocultured with splenic DCs, compared with T cells cocultured with BMDCs. Nevertheless, since IFN-g levels were far higher than IL-10 levels in all the subsets, the differences in IL-10 levels may not be of much significance with respect to the overall T cell response. Major contribution toward T cell stimulation (more specifically, IL-2 and IFN-g production) in all the sets was obtained by the CD4⫹ T cell subset, indicating that MTSA was primarily MHC class II restricted. CD8⫹ T cell–mediated responses observed were marginal (only 10%–20% of the total). These results are in agreement with studies that document the dominance of CD4⫹ over CD8⫹ T cell responses during early infection by M. tuberculosis [10]. MTSA-matured DCs do not respond to secondary antigenic challenge. Maturation of DCs by antigens, particularly those of infectious organisms, has important consequences for the outcome of an immune response, since the functionality of DCs can take a dramatic turn with respect to the loss or attenuation of their ability to process a second challenge of any antigen. Because the data in figures 1 and 2 revealed that MTSA induced the full maturation of DC subsets, we, therefore, investigated whether MTSA-matured DC subsets still retained the ability to respond to a challenge with various antigens, including the CE of M. tuberculosis. Furthermore, it has been reported
Table 2. Cytokine profiles of Mycobacterium tuberculosis secretory antigen (MTSA)–stimulated dendritic cell (DC) subsets. Cytokine Group
TNF-a
IFN-g
BMDCs
192.0 15.2
ND
BMDCs ⫹ MTSAa
1766 165
145.5 20
CD8a⫺ DCs
21.8 2.5
ND
CD8a ⫹ MTSA
43.7 3.5
71.75 5.2
CD8a⫹ DCs
72.9 7.2
30.7 3.2
140.9 15
1305.8 95
⫺
a
a
CD8a⫹ ⫹ MTSA
IL-12p40 ND 55.2 2.1 ND 43.8 3.5 ND 225.75 25
IL-10 35 1.2 59.6 6.5 38 2.5 43.5 2.4 32 2.8 ND
NOTE. Data are mean SD results (pg/mL/106 cells) of 1 of 3 experiments. Culture supernatants from MTSA-stimulated DC subsets were analyzed for the presence of the cytokines, as described in Materials and Methods. BMDCs, bone marrow–derived DCs; IFN, interferon; IL, interleukin; ND, not detected; TNF, tumor necrosis factor. a
DC subsets were stimulated with 25 mg/mL MTSA for 24 h, and supernatants were analyzed for the indicated cytokines.
elsewhere that the cell-wall skeleton of M. bovis BCG induces the maturation of human DCs [35]. As expected, MTSA-specific T cells cocultured with MTSAmatured BMDCs secreted high levels of IFN-g in culture supernatants, compared with IL-10 levels (figure 4A). Of interest, when MTSA-matured DCs were pulsed with various antigens, the IL-2 and IFN-g T cell responses of the challenging antigens were attenuated by up to 2–3-fold, compared with their respective controls (figure 4A), which indicates that MTSA-matured DCs are less proficient at processing a second challenge with any other antigen. The reduction of IFN-g levels of CE-specific T cells, after MTSA-matured DCs are pulsed with CE, is of particular significance, because the CE components may well constitute parts of whole bacteria that might be released from infected macrophages at sites of infection. Similar observations were obtained when MTSA-matured splenic CD8a⫺ and CD8a⫹ DCs were pulsed with different antigens. Again, the proliferative and proinflammatory responses of CE-specific T cells (as reflected by IL-2 and IFN-g levels) and other antigens were down-regulated by 2-fold, compared with their respective controls (figure 4B and 4C). The IFN-g responses of MTSA-specific T cells in cocultures of MTSA-matured BMDCs pulsed with CE were also marginally down-regulated, but not with those cocultured with similarly treated splenic DCs (data not shown). However, no effect of ovalbumin or hen egg lysozyme on MTSA-specific T cell responses was observed (data not shown). MTSA-matured DCs down-regulate T cell responses to CE antigens in vivo. In an attempt to mimic the early events that would occur after an infection by mycobacteria, whereby their release from macrophages would follow that of secretory proteins from the phagosomal complex, we separately transferred MTSA-matured DC subsets that were pulsed with CE into naive mice. This was followed by a challenge with either MTSA- or CE-specific enriched T cells. Lymph nodes were later cultured, and cytokines levels in their supernatants were esti-
mated. As shown in figure 5, MTSA-specific T cell responses induced by all DC subsets were as expected, with IFN-g levels dominantly expressed over IL-10. However, a challenge with CE-specific T cells in mice that had received MTSA-matured BMDCs led to a significant down-regulation of IL-2 and IFNg levels by 4- and 10-fold, respectively, compared with results obtained when CE-matured DCs were used (figure 5A). A similar trend was evident with splenic DC subsets, insofar as the down-regulation of IFN-g levels were concerned after a challenge with CE-specific T cells, when a 2-fold reduction in IFNg levels was observed (figure 5B and 5C). IL-10 levels also showed an increase by 3-fold in CD8a⫺ DCs, compared with CE-matured DCs, which indicates that there was down-regulation of Th1 responses, together with an increase in regulatory responses, at least in this DC subset. These results indicate that proliferative and proinflammatory T cell responses to M. tuberculosis CE are down-regulated at sites where DC-APCs expressing or loaded with MTSA predominate. MTSA-matured DCs retain phagocytic ability. It is known that matured or antigen-activated DCs show reduced phagocytosis of particulate matter [16, 17]. The results presented above demonstrate that MTSA-matured DCs do not respond to a second challenge with antigen. To investigate whether MTSA-matured DCs display phagocytosis, we incubated these cells with 0.5-mm fluorescent carboxylate–modified latex beads and examined their internalization by flow cytometry. As shown in figure 6A, immature BMDCs readily internalized the beads. Internalization of the beads was, however, only marginally reduced in MTSA-matured BMDCs (figure 6B). As expected, LPS-matured DCs showed a significant reduction (12-fold) in their phagocytic ability (figure 6C), whereas activation of DCs with MSP-119 kDa had no effect on the phagocytic ability of DCs (figure 6D). Similar results were obtained with splenic DCs, where no significant differences were observed between unstimulated and MTSA-matured DCs (data not shown). These Maturation of Dendritic Cells by MTSA • JID 2003:187 (15 March) • 921
Figure 3. T cells responses induced by Mycobacterium tuberculosis secretory antigen (MTSA)–matured dendritic cell (DC) subsets are CD4⫹ restricted Either bone marrow–derived DCs (BMDCs; A) or CD8a⫺ (B) or CD8a⫹ (C) DC subsets were matured with MTSA and cocultured with either unfractionated or CD4⫹ or CD8⫹ MTSA-specific T cells for 48 h, and cytokine levels were estimated in the culture supernatants at end of incubation period (see Materials and Methods). Groups 2–4 in all the panels represent cocultures of MTSA-matured DC subsets with unfractionated, CD4⫹, and CD8⫹ T cells, respectively. Group 1 shows the cytokine levels of unfractionated T cells cocultured with unstimulated DC subsets. The differences in levels of cytokines between CD4⫹ and CD8⫹ T cell groups were statistically significant at P ! .05 . Data from 1 of 3 experiments are shown. IFN, interferon; IL, interleukin.
results are in agreement with previous studies that report reduced, but not abolished, phagocytic ability of DCs when infected with live mycobacteria [36]. DCs activated by M. tuberculosis CFPs down-regulate proinflammatory responses to CE antigens. The results shown in figures 4 and 5 indicate that maturation of DC subsets by MTSA leads to down-regulation of proliferative and proinflammatory responses to CE antigens, as reflected by the decrease in the levels of IL-2 and IFN-g of the interacting T cells, thereby suggesting a possible role of secretory antigens in governing immune responses during an M. tuberculosis infection. However, because the above observations were made with a single antigen, it was prudent to carry out similar studies with various other secretory antigens to add support to the above proposition. To this end, we stimulated various DC subsets with the CFPs of M. tuberculosis, because they would represent an array of antigens that are likely to be released from the phagosomes of infected macrophages at sites of infection. We first depleted MTSA from CFPs by incubation with antibody to MTSA, followed by immunoprecipitation with protein G–conjugated agarose beads, to rule 922 • JID 2003:187 (15 March) • Natarajan et al.
out any effects of MTSA that might be present in the CFP. CFPactivated DCs were pulsed with CE and cocultured with either CE- or CFP-specific T cells for 48 h, and IFN-g and IL-10 levels were then scored in supernatants. As shown in figure 7, like MTSA, CFP-activated DC subsets were also unable to stimulate the proliferation of CE-specific T cells, as indicated by the reduced levels of IL-2 and IFN-g in these cells, suggesting that MTSAs induce maturation of DCs at sites of infection that might lead to an attenuation of proinflammatory responses to CE or possibly whole bacteria once they are released from cytolyzed or infected macrophages. DISCUSSION The recognition that immunization with M. bovis BCG has a variable impact on the transmission of M. tuberculosis has renewed interest in developing more effective vaccines for tuberculosis [4–7]. However, a prerequisite for achieving this goal is a thorough understanding of the pathogenesis of M. tuberculosis, with respect to the kind and degree of immune
Figure 4. Mycobacterium tuberculosis secretory antigen (MTSA)–matured dendritic cells (DCs) are nonresponsive to secondary antigenic challenge. MTSA-matured bone marrow–derived DCs (BMDCs; A), CD8a⫺ (B), and CD8a⫹ (C) DC subsets were pulsed with 10 mg/mL of various antigens for 24 h and cocultured for 48 h with enriched T cells primed with the challenging antigen. Culture supernatants were then screened for cytokine levels. “CONTROL” depicts average values of control groups of either unstimulated DCs or T cells only. Differences in levels of interleukin (IL)–2 and interferon (IFN)–g in all the panels between second antigen–challenged MTSA-matured DCs and their respective controls were statistically significant at P ! .05. Data from 1 of 4 independent experiments are shown. CE-T, cell extract–specific T cells; HE-T, hen egg lysozyme (HEL)–specific T cells; M-T, MTSA-specific T cells; OVA-T, ovalbumin (OVA)–specific T cells.
responses generated early in the course of infection. It is well known that, after infection, cell types recruited foremost at sites of infection are the various DC subsets and their precursors, which are at various stages of their differentiation and maturation [16, 19]. Consequently, interactions of DCs with mycobacteria or parts thereof likely play a determinant role in regulating immune responses that are generated early in the infection and that would essentially permit a framework for the host to tailor its defense strategies for clearing the pathogen. A number of studies have been conducted on the interaction of mycobacteria with DCs. For example, Henderson et al. [36] showed that infection of DCs with mycobacteria causes their activation, as reflected by increased surface densities of various costimulatory and MHC molecules. In addition, infected DCs secreted elevated levels of inflammatory cytokines, including TNF-a, IL-1, and IL-12. DCs were further shown to phago-
cytose mycobacteria. Bodnar et al. [37] further showed that mycobacteria could replicate inside murine BMDCs and that, although DCs were able to restrict their growth, they were nevertheless less efficient than infected macrophages at eliminating the infection. These results further suggested the importance of DCs in priming immune responses to mycobacteria. Furthermore, stimulation of M. tuberculosis–infected DCs via CD40 increased the ability of DCs to mount T cell responses; this was later shown to be primarily attributed to increased expression of costimulatory and MHC molecules on their cell surface [38]. DCs also have been shown to induce protective immunity against M. tuberculosis in a murine model and also against aerosol-mediated infection [39]. A number of microbial lipopeptides and proteins also have been shown to activate and mature DCs [40]. Although a number of studies have been conducted on secretory antigens, most have focused on CFPs Maturation of Dendritic Cells by MTSA • JID 2003:187 (15 March) • 923
Figure 5. Mycobacterium tuberculosis secretory antigen (MTSA)–matured dendritic cell (DC) subsets down-regulate T cell inflammatory responses to cell extract (CE) in vivo. MTSA-matured bone marrow–derived DCs (BMDCs; 2 ⫻ 106 ; A), CD8a⫺ (B), or CD8a⫹ (C) DC subsets were pulsed with CE for 24 h and injected (subcutaneously [sc] at the base of tail) into naive mice, followed by a challenge with an equal number of either MTSA- or CE-specific T cells 24 h later (sc at the base of tail; see Materials and Methods). Three days later, lymph nodes were cultured for 48 h, and cytokines were measured in the culture supernatants. “CONTROL” depicts average values of groups that received either unstimulated or unstimulated DCs followed by antigen-specific T cells. Differences in levels of interleukin (IL)–2 and interferon (IFN)–g in all the panels between CE-pulsed, MTSAmatured and CE-matured DCs were statistically significant at P ! .05. Data from 1 of 3 independent experiments are shown.
and their potential role as vaccine candidates. For example, early secreted antigen target (ESAT)–6 has been designated as an important T cell antigen recognized by protective T cells in animal models of infection with M. tuberculosis [41]. Furthermore, MPT64 and ESAT-6 have been shown to have potential in the diagnosis of M. tuberculosis because they are recognized by T cells in animal models [42, 43] of M. tuberculosis. Like MTSA, both antigens have been found primarily in M. tuberculosis but not in most environmental mycobacteria or BCG [44, 45]. MPT64 has been evaluated as a skin test reagent in guinea pig models of tuberculosis [46] and in humans [47]. These antigens also have been shown to elicit delayed-type hypersensitivity responses in guinea pig models of tuberculosis. ESAT-6 also has been considered as a potential candidate for subunit-based vaccines [48]. Ag85b, a member of the antigen 85 complex (a family of fibronectin-binding proteins involved in mycobacterial cell-wall biosynthesis) [49] and other secretory 924 • JID 2003:187 (15 March) • Natarajan et al.
antigens have been used as potential DNA vaccine candidates [50, 51]. However, despite these studies, information regarding their actual roles at sites of infection in influencing the early immune responses to this pathogen is still lacking. Only very recently has their in vivo presence been demonstrated [11]. In addition, an intriguing question that emerges is the physiological relevance of these secretory antigens, more so in the light of the fact that mycobacteria have been highly successful in devising strategies for immune evasion [10, 52]. To identify possible interactions of secretory antigens with DCs, we have recently shown that mycobacterial secretory antigens, including MTSA, induce the differentiation of DCs from BM [26]. These DCs expressed various cell-surface markers characteristic of DCs, such as CD11c and CD11b; costimulatory molecules CD80, CD86, CD40, and CD54; and MHC class I and class II molecules and also displayed DC-like morphology. Although immature in nature, MTSA-differentiated DCs were equally
Figure 6. Mycobacterium tuberculosis secretory antigen (MTSA)– matured dendritic cells (DCs) display phagocytosis. Bone marrow–derived DCs (BMDCs) were stimulated with either 25 mg/mL of MTSA or merozoite surface protein (MSP)–119 kDa or 20 mg/mL of lipopolysaccharide (LPS) for 24 h. Cells were washed and resuspended in buffer containing fluorescent carboxylated latex beads (see Material and Methods) for 4 h at 37C. Cells were thoroughly washed and analyzed by flow cytometry. A–D, Unstimulated, MTSA-matured, LPS-matured, and MSP-119 kDa–stimulated BMDCs, respectively. Values within the marker (M1) represent percentage of phagocytic cells. Data from 1 of 4 independent experiments are shown.
proficient at stimulating allogeneic T cell responses, compared with GM-CSF–differentiated DCs. To investigate whether MTSA would also induce the maturation of various DC subsets and whether MTSA-matured DCs would in any manner alter the nature of subsequent T cell responses to mycobacterial antigens, the present study was done. We demonstrated that incubation of various DC subsets with MTSA induces their maturation. Consistent with various parameters considered to be the hallmark of DC maturation by various agents, such as soluble proteins like gp96 [53] or CpG-containing DNA motifs [54], increases in the levels of various markers is one aspect of DC maturation, and the addition of MTSA to various DC subsets indeed up-regulated the levels of CD80, CD86, CD40, CD54, and MHC class I and class II molecules. MTSA also increased the expression of the DC maturation marker F4/80 antigen. MTSA-matured DCs were also found to boost allogeneic T cell responses, compared with those of immature DCs—another feature that is attributable to DC maturation [16, 17] and is considered to be a direct translation of upregulated levels of costimulatory and MHC molecules [53, 54]. The addition of various terminal maturation–inducing stimuli had no effect on either the surface levels of various markers or the extent of allogeneic T cell stimulation, which indicates that MTSA induced the full maturation of DCs.
That the observed effects were caused by MTSA and not any contaminant in the recombinant expressed protein was ascertained in the various control experiments that showed the absence of any detectable or significant levels of endotoxins, such as LPS, or other low-molecular-weight compounds that are known to copurify with proteins expressed in Escherichia coli (table 1). Infection of DCs by mycobacteria has been shown to induce differential cytokine production, and the profiles were also shown to differ from those induced by infection of macrophages [55]. Although infected DCs secreted IL-12, IFN-a, and TNF-a, infected macrophages secreted IFN-g, IL-6, and IL-18. Low levels of IL-12 observed in macrophages were attributed to higher levels of IL-10, a known inhibitor of IL-12 secretion [56]. Infection of human DCs by M. tuberculosis has further been shown to lead to the secretion of type 1 IFN genes [56]. Recently, it was also shown that differential cytokine secretion by M. tuberculosis–infected DCs and macrophages resulted in differential effects on naive T cell polarization that were explained on the basis of differential levels of IL-12 secretion that was predominantly expressed by DCs, compared with macrophages [57]. In fact, incubation of mycobacteria infected DCs with IL-10 converted them into macrophages with increased antibacterial activity [58]. It was thus of interest to examine the cytokine profiles of various DC subsets after incubation with MTSA. It has been proposed recently that the nature of T cell responses generated after cognate interactions with APCs is also influenced by the profiles of cytokine milieu, both before and during the interaction [59]. For example, a predominance of IL-12 secreted by DCs skews the response toward Th1, whereas low levels of IL-12 and enhanced levels of IL-10 induce a Th2/Th0 response [60]. Further characterization revealed that MTSA-stimulated DC subsets secreted varying levels of IFN-g and IL-12p40. CD8a⫹ DCs, which are known to skew responses toward Th1 [18], secreted high levels of proinflammatory cytokines TNF-a, IL-12p40, and IFN-g, whereas CD8a⫺ DCs known to trigger Th2 responses displayed low levels of Il-12p40 and moderate levels of IFN-g after MTSA stimulation. Because IL-12p40 (or IL-12p70) directly regulates IFN-g production [60], the low levels of IFN-g in BMDCs and CD8a⫺DCs could probably result from low levels of IL-12p40 induction by MTSA in these DCs, whereas the higher levels of IFN-g in the CD8a⫹ DCs could result from enhanced secretion of IL-12p40 by this subset, indicating that MTSA induced the differential secretion of various cytokines from different DC subsets. That MTSA induced the dominant secretion of proinflammatory cytokines is consistent with related studies that report a similar profile of cytokine secretion after infection of DCs by live mycobacteria [61, 62]. To investigate whether maturation of DCs by MTSA also results in its internalization, processing, and presentation and to Maturation of Dendritic Cells by MTSA • JID 2003:187 (15 March) • 925
Figure 7. Mycobacterium tuberculosis culture filtrate protein (CFP)–activated dendritic cells (DCs) down-regulate interferon (IFN)–g responses of cell extract (CE) antigens. CFP-stimulated DC subsets were pulsed with CE for 24 h and cocultured with either CFP-specific T cells (CFP-T) or CEspecific T cells (CE-T) for 48 h. Culture supernatants were scored for the levels of various cytokines at the end of the incubation period. “CONTROL” depicts average values of unstimulated DCs cocultured with T cells. Differences in levels of interleukin (IL)–2 and IFN-g between CE-pulsed, M. tuberculosis secretory antigen (MTSA)–matured DCs and CE-matured DCs were statistically significant at P ! .05 . Data from 1 of 4 experiments are shown. BMDCs, bone marrow–derived DCs.
investigate whether the differences in cytokine secretory profiles would result in differential T cell responses, we next characterized the antigen-specific T cell responses mediated by MTSA-matured DC subsets. Our results showed that MTSA was indeed presented on MHC molecules during maturation and that matured DCs readily stimulated antigen-specific T cells to secrete IL-2 and IFNg. However, regardless of the differences in cytokine-secretion profiles from various MTSA-matured DC subsets, the overall differences in the relative levels of IFN-g and IL-10 from MTSAspecific T cells were more or less similar. These results, therefore, indicate that factors other than cytokine secretory profiles of DCs may also influence the nature of T helper responses and may possibly be antigen dependent. Furthermore, these results are in agreement with those of Pulendran et al. [63], who showed that antigen-pulsed CD8a⫹ and CD8a⫺ DCs were equally efficient in inducing T cells into secreting IFN-g and IL-2, despite the fact that CD8a⫹ DCs secreted more IL-12, compared with CD8a⫺ DCs. A more detailed analysis showed that the major contribution to IFN-g secretion came from the CD4⫹ T cell subset, which indicates that MTSA was essentially MHC class II restricted. Although MTSA was added exogenously would normally be processed and presented on MHC class II molecules, DCs are often known to cross-present antigens. For example, the outer membrane protein A (ompA) from Klebsiella pneumoniae was shown to induce the maturation of DCs via Toll-like receptor 2 and was, indeed, MHC class I restricted [32]. Our results, however, showed that MTSA was largely MHC class II restricted and, in a way, exemplified the importance of CD4⫹ T cells in mediating immune responses to mycobacterial infection [10]. The role of CD8⫹ T cells in governing immune responses to mycobacteria recently has been highlighted with the identification of a number of CD8⫹ T cell clones mediating immune responses to mycobacterial antigens [64] and by studies showing greater 926 • JID 2003:187 (15 March) • Natarajan et al.
susceptibility of b-2 microglobulin knockout mice over wild-type control mice [65]. Because fully mature DCs are less efficient at processing a second challenge of antigen [16, 17], pulsing of MTSA-matured DC subsets with various antigens did not result in stimulation of T cells specific to the challenging antigen, which indicates that MTSA-matured DCs were attenuated in their capacity to present secondary antigens. The components of the cell envelope of mycobacteria (which would constitute part of CE that was used in the current study) are some of the early antigens that would be recognized by APCs and have been shown to induce the maturation of DCs [35]. We, therefore, characterized the effects of CE-pulsed, MTSA-matured DCs in regulating immune responses of CE-specific T cells. Consistent with the results obtained with other challenging antigens, addition of CE to these DCs also resulted in a decrease in the capacity of CE-specific T cells to secrete IL-2 and IFN-g, which indicates that these T cells were nonproliferative. Furthermore, the decrease in the levels of IFN-g was also accompanied by modest changes in the levels of IL-10 in some DC subsets, such as those observed in the case of CD8a⫺ DCs, which are known to induce Th2 responses [18]. The fact that the IL-10 levels were, by and large, not affected and the decrease in IFN-g levels point to a change in the overall T helper responses at sites of infection, whereby a decrease in the proinflammatory responses could contribute to down-regulation of immune responses to mycobacterial antigens. These results may thus indicate a possible putative role for secretory antigens in modulating immune responses during a mycobacterial infection. This is further supported by the results obtained with CFPs, where a similar trend was obtained with respect to the attenuation of IL-2 and IFNg responses from CE-specific T cells when cocultured with CEpulsed, CFP-matured DCs.
Maturation of DC precursors by secretory antigens may have important bearings in the spread of infection, as noted in a recent report by Bodnar et al. [37], which examined the use of DCs as transport vehicles to migrate to secondary lymphoid organs. Our results on the retention of phagocytic ability of MTSA-matured DCs suggest that this might constitute one of the mechanisms that the bacterium may exploit to achieve this objective. Although maturation also results in changes in the chemokine receptor profiling that would now also change their receptiveness to various chemokines and would allow the DCs to migrate, it is nevertheless possible and probable that these DCs may still be able to phagocytose mycobacteria that are released from macrophages, because development of a chemokine gradient is essential for proper migration of DCs [66]. However, more-detailed experiments are required to support the current hypothesis. In light of the above, it is tempting to speculate that maturation of DCs by MTSA and other secretory antigens might provide additional frequency of such transport vehicles for the bacterium to migrate to secondary organs.
Acknowledgements
We thank J. T. Belisle (Colorado State University, Fort Collins), for the kind gift of whole-cell lysate and culture filtrate proteins (provided through National Institutes of Health, National Institutes of Allergy and Infectious Diseases grant AI75320), and Pawan Malhotra (International Center For Genetic Engineering and Biotechnology, Aruna Asaf Ali Marg, New Delhi, India), for the kind gift of Plasmodium falciparum merozoite surface protein–119 kDa.
References 1. World Health Organization. Global tuberculosis control. World Health Organization. Geneva: 1998. 2. Dolin PJ, Raviglione MC, Kochi A. Global tuberculosis incidence and mortality during 1990–2000. Bull World Health Organ 1994; 72:213–20. 3. Bloom BR, Murray CJL. Tuberculosis: commentary on a re-emergent killer. Science 1992; 257:1055–64. 4. Colditz GA, Brewer FT, Berkey CS, et al. Efficacy of BCG vaccine in the prevention of tuberculosis: meta-analyses of the published literature. JAMA 1994; 271:698–702. 5. Kaufmann SHE. Is the development of a new tuberculosis vaccine possible? Nat Med 2000; 6:955–60. 6. Fine PE. Variation in protection by BCG: implications of and for heterologous immunity. Lancet 1995; 346:1339–45. 7. Hess JH, Schaible UE, Raupach B, Kaufmann SHE. Exploiting the immune system: toward new vaccines against intracellular bacteria. Adv Immunol 2000; 75:1–8. 8. Manabe YC, Bishai WR. Latent Mycobacterium tuberculosis: persistence, patience, and winning by waiting. Nat Med 2000; 6:1327–9. 9. Foote S. Mediating immunity to mycobacteria. Nat Genet 1999; 21:345–6. 10. Flynn JA, Chan J. Immnuology of tuberculosis. Annu Rev Immunol 2001; 19:93–129. 11. Beatty WL, Russel DG. Identification of mycobacterial surface proteins released into subcellular compartments of infected macrophages. Infect Immun 2000; 68:6997–7002.
12. Anderson P. Effective vaccination of mice against Mycobacterium tuberculosis infection with a soluble mixture of secreted mycobacterial proteins. Infect Immun 1994; 62:2536–44. 13. Roberts AD, Sonnenberg MG, Ordway DJ, et al. Characteristics of protective immunity engendered by vaccination of mice with purified culture filtrate protein antigens of Mycobacterium tuberculosis. Immunology 1995; 85:502–8. 14. Roche PW, Triccas JA, Avery DT, Fifts T, Billman-Jacobe H, Britton WJ. Differential T cell responses to mycobacteria-secreted proteins distinguish vaccination with bacille Calmette Gue´rin from infection with Mycobacterium tuberculosis. J Infect Dis 1994; 170:1326–30. 15. Anderson PA, Anderson P, Sorenson L, Nagai S. Recall of long-lived immunity to Mycobacterium tuberculosis infection in mice. J Immunol 1995; 154:3359–72. 16. Banchereau J, Steinman RM. Dendritic cells and the control of immunity. Nature 1998; 392:245–52. 17. Steinman RM. Dendritic cells. In: Paul WE, ed. Fundamental immunology. Philadelphia: Lippincott-Raven, 1999:547–73. 18. Reid SD, Penna G, Adorini, L. The control of T cell responses by dendritic cells subsets. Curr Opin Immunol 2000; 12:114–21. 19. Sousa CR. Dendritic cells as sensors of infection. Immunity 2001; 14: 495–8. 20. Colangeli R, Spencer JS, Bifani P, et al. MTSA-10, the product of the Rv3874 gene of Mycobacterium tuberculosis, elicits tuberculosis-specific, delayed-type hypersensitivity in guinea pigs. Infect Immun 2000; 68: 990–3. 21. Dillon DC, Alderson MR, Day CH, et al. Molecular and immunological characterization of Mycobacterium tuberculosis CFP-10, an immunodiagnostic antigen missing in Mycobacterium bovis BCG. J Clin Microbiol 2000; 38:3285–90. 22. Lewinsohn DM, Zhu L, Madison VJ, et al. Classically restricted human CD8⫹ T lymphocytes derived from Mycobacterium tuberculosis–infected cells: definition of antigenic specificity. J Immunol 2001; 166:439–46. 23. Arend SM, Ottenhoff TH, Andersen P, van Dissel JT. Uncommon presentations of tuberculosis: the potential value of a novel diagnostic assay based on the Mycobacterium tuberculosis–specific antigens ESAT6 and CFP-10. Int J Tuberc Lung Dis 2001; 5:680–6. 24. Brock I, Munk ME, Kok-Jensen A, Andersen P. Performance of whole blood IFN-gamma test for tuberculosis diagnosis based on PPD or the specific antigens ESAT-6 and CFP-10. Int J Tuberc Lung Dis 2001; 5: 462–7. 25. Trajkovic V, Singh G, Singh B, Singh S, Sharma P. Effect of Mycobacterium tuberculosis–specific 10-kilodalton antigen on macrophage release of tumor necrosis factor alpha and nitric oxide. Infect Immun 2002; 70:6558–66. 26. Latchumanan VK, Singh B, Sharma P, Natarajan K. Mycobacterium tuberculosis antigens induce the differentiation of dendritic cells from bone marrow. J Immunol 2002; 169:6856–64. 27. Natarajan K, Sahoo NC, Rao KVS. Signal thresholds and modular synergy during expression of costimulatory molecules in B lymphocytes. J Immunol 2001; 167:114–22. 28. Vijayakrishnan L, Natarajan K, Manivel V, Raisuddin S, Rao KV. B cell responses to a peptide epitope. IX. The kinetics of antigen binding differentially regulates the costimulatory capacity of activated B cells. J Immunol 2000; 164:5605–14. 29. Inaba K, Inaba M, Romani M, et al. Generation of large numbers of dendritic cells from mouse bone marrow cultures supplemented with granulocyte/macrophage colony-stimulating factor. J Exp Med 1992; 176:1693–702. 30. Scheicher C, Mehlig M, Zecher R, Reske K. Dendritic cells from mouse bone marrow: in vitro differentiation using low doses of recombinant granulocyte/macrophage colony-stimulating factor. J Immunol Methods 1992; 154:253–64. 31. Gonzalez-Juarrero M, Orme IR. Characterization of murine lung dendritic cells infected with Mycobacterium tuberculosis. Infect Immun 2001; 69:1127–31. 32. Jeannin P, Renno T, Goetsch L, et al. OmpA targets dendritic cells,
Maturation of Dendritic Cells by MTSA • JID 2003:187 (15 March) • 927
33. 34.
35.
36.
37. 38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
induces their maturation and delivers antigen into the MHC class I presentation pathway. Nat Immunol 2000; 1:502–9. Takeuchi O, Kawai T, Mulradt PF, et al. Discrimination of bacterial lipoproteins by Toll-like receptor 6. Int Immunol 2001; 13:933–40. Maldonado-Lopez R, Smedt TD, Chel PM, et al. CD8a⫹ and CD8a⫺ subclasses of dendritic cells direct the development of distinct T helper cells in vivo. J Exp Med 1999; 189:587–92. Souji R, Matsumoto M, Tekeuchi O, et al. Maturation of human dendritic cells by cell wall skeleton of Mycobacterium bovis bacillus Calmette-Guerin: involvement of toll-like receptors. Infect Immun 2000; 68:6883–90. Henderson RA, Watkins SC, Flynn JA. Activation of human dendritic cells following infection with Mycobacterium tuberculosis. J Immunol 1997; 159:635–43. Bodnar KA, Serbina NV, Flynn JL. Fate of Mycobacterium tuberculosis within murine dendritic cells. Infect Immun 2001; 69:800–9. Demangel C, Palendira U, Feng CG, Heath AW, Bean AG, Britton WJ. Stimulation of dendritic cells via CD40 enhances immune responses to Mycobacterium tuberculosis infection. Infect Immun 2001; 69:2456–61. Demangel C, Bean AG, Martin E, Feng CG, Kamath AT, Britton WJ. Protection against aerosol Mycobacterium tuberculosis infection using Mycobacterium bovis bacillus Calmette Gue´rin–infected dendritic cells. Eur J Immunol 1999; 29:1972–9. Hertz CJ, Kiertscher SM, Godowski PJ, et al. Microbial lipopeptides stimulate dendritic cell maturation via Toll-like receptor 2. J Immunol 2001; 166:2444–50. Brandt L, Oettinger T, Holm A, Andersen P. Key epitopes on the ESAT6 antigen recognized by mice during the recall of protective immunity to Mycobacterium tuberculosis. J Immunol 1996; 157:3527–33. Elhay MJ, Oettinger T, Andersen P. Delayed type hypersensitivity response to ESAT-6 and MPT64 from Mycobacterium tuberculosis in the guinea pig. Infect Immun 1998; 66:3454–6. Oettinger T, Andersen AB. Cloning and B-cell-epitope mapping of MPT64 from Mycobacterium tuberculosis H37Rv. Infect Immun 1994; 62:2058–64. Harboe M, Oettinger T, Wiker HG, Rosenkrands I, Andersen P. Evidence for occurrence of the ESAT-6 protein in Mycobacterium tuberculosis and virulent Mycobactrium bovis and its absence in Mycobacterium bovis BCG. Infect Immun 1996; 64:16–22. Sorensen AL, Nagai S, Houen G, Andersen P, Andersen AB. Purification and characterization of a low-molecular-mass T-cell antigen secreted by Mycobacterium tuberculosis. Infect Immun 1995; 63:1710–7. Haga S, Yamaguchi R, Nagai S, Matsuo K, Yamazaki A, Nakamura RM. Delayed-type hypersensitivity to a recombinant mycobacterial antigen MPT64 in guinea pigs sensitized by Mycobacterium tuberculosis or Mycobacterium bovis BCG. J Leukoc Biol 1995; 57:221–5. Roche PW, Winter N, Triccas JA, Feng CG, Britton WJ. Expression of Mycobacterium tuberculosis MPT64 in recombinant Mycobacterium smegmatis: purification, immunogeneticity and application to skin tests for tuberculosis. Clin Exp Immunol 1996; 103:226–32. Brandt L, Elhay M, Rosenkrands I, Lindblad EB, Andersen P. ESAT-6 subunit vaccination against Mycobacterium tuberculosis. Infect Immun 2000; 68:791–5. Ratliff TL, McGarr JA, Abou-Zeid C, et al. Attachment of mycobacteria to fibronectin-coated surfaces. J Gen Microbiol 1988; 134:1307–13.
928 • JID 2003:187 (15 March) • Natarajan et al.
50. Tanghe A, D’Souza S, Rossesls V, et al. Improved immunogenicity and protective efficacy of a tuberculosis DNA vaccine encoding Ag85 by protein boosting. Infect Immun 2001; 69:3041–7. 51. Kamath AT, Briscoe H, Britton WJ. Co-immunization with DNA vaccines expressing granulocyte-macrophage colony-stimulating factor and mycobacterial secreted proteins enhances T-cell immunity, but not protective efficacy against Mycobacterium tuberculosis. Immunology 1999; 96:511–6. 52. Pieters J. Entry and survival of pathogenic mycobacteria in macrophages. Microbes Infect 2001; 3:249–55. 53. Binder RJ, Anderson K, Basu S, Srivastava PK. Cutting edge: heat shock protein gp96 induces the maturation and migration of CD11c⫹ cells in vivo. J Immunol 2000; 165:6029–35. 54. Gursel M, Verthelyi D, Klinman DM. CpG oligodeoxynucleotides induce human monocytes to mature into functional dendritic cells. Eur J Immunol 2002; 32:2617–22. 55. Giacomini E, Iona E, Ferroni L, et al. Infection of human macrophages and dendritic cells with Mycobacterium tuberculosis induces a differential cytokine gene expression that modulates T cell response. J Immunol 2001; 166:7033–41. 56. Remoli ME, Giacomini E, Lutfalla G, et al. Selective expression of type I IFN genes in human dendritic cells infected with Mycobacterium tuberculosis. J Immunol 2002; 169:366–74. 57. Hickman SP, Chan J, Salgame P. Mycobacterium tuberculosis induces differential cytokine production from dendritic cells and macrophages with divergent effects on naive T cell polarization. J Immunol 2002; 168: 4636–42. 58. Fortsch D, Rollinghoff 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. 59. Kournilsky P, Truffa-bachi P. Cytokine fields and the polarization of the immune response. Trends Immunol 2001; 22:502–9. 60. Kalinski P, Hilkens CM, Weirnenga EA, Kapsenberg ML. T-cell priming by type-1 and type-2 polarized dendritic cells: the concept of the third signal. Immunol Today 1999; 20:561–7. 61. Giacomini E, Iona E, Ferroni L, et al. Infection of human macrophages and dendritic cells with Mycobacterium tuberculosis induces a differential cytokine gene expression that modulates T cell response. J Immunol 2001; 166:7033–41. 62. Feng CG, Bean AG, Hooi H, Briscoe H, Britton WJ. Increase in gamma interferon–secreting CD8⫹, as well as CD4⫹ T cells in lungs following aerosol infection with Mycobacterium tuberculosis. Infect Immun 1999; 67:3242–7. 63. Pulendran B, Smith JL, Caspary G, et al. Distinct dendritic cell subsets differentially regulate the class of immune response in vivo. Proc Natl Acad Sci USA 1999; 96:1036–41. 64. Feng CG, Demangel C, Kamath AT, Macdonald M, Britton WJ. Dendritic cells infected with Mycobacterium bovis bacillus Calmette Gue´rin activate CD8⫹ T cells with specificity for a novel mycobacterial epitope. Int Immunol 2001; 13:451–8. 65. Sousa AO, Mazzaccaro RJ, Russell RG, et al. Relative contributions of distinct MHC class I–dependent cell populations in protection to tuberculosis infection in mice. Proc Natl Acad Sci USA 2000; 97:4204–8. 66. Proudfoot AE, Power CA, Wells TN. The strategy of blocking the chemokine system to combat disease. Immunol Rev 2000; 177:246–56.
