Tuberculosis caused by Mycobacterium tuberculosis still remains a global threat with an estimated 1.5 million deaths annually. Cellular immunity plays a critical role in mediating the protection against tuberculosis. Indeed, IFN-γ-producing CD4⁺ T helper type 1 cells are critical for the control of M. tuberculosis in humans and murine models (1–4). Thus, general paradigm for tuberculosis vaccines has largely focused on enhancing the Th1/IFN-γ response (5, 6). However, despite enhancing IFN-γ response, the recombinant MVA85A vaccine failed to protect infants from tuberculosis (7). Therefore, it is pertinent to decipher the role played by other CD4⁺ T cell subsets and their cytokines in mediating immunity against M. tuberculosis.

Th17 cells that express transcription factor RORC and secrete archetype cytokine IL-17A represent a distinct subset of CD4⁺ T cells. Th17 cells also produce IL-17F, IL-21, IL-22, GM-CSF, and IL-26 and mediate pro-inflammatory responses (8–11). Th17 cells are associated with the pathogenesis of several autoimmune and inflammatory diseases (12–14). Evolutionarily, Th17 response is conserved to mediate protection at mucosal surfaces and against extracellular pathogens (10, 15). Recent reports have also indicated that Th17 response may play a crucial role in mediating protection against intracellular pathogens, such as Francisella tularensis and Chlamydia muridarum (16–18). These data thus indicate the diverse role of Th17 cells in various physiopathologies.

Mycobacterium tuberculosis employs a plethora of mechanisms to suppress both innate and adaptive immune responses. The role of Th17 response to M. tuberculosis is largely pursued in mice, and it remains highly controversial (19–25). Recent reports in tuberculosis patients indicate that active disease and its severity are associated with low Th17 response (26, 27). Of note, anti-tuberculosis therapy is associated with enhanced Th17 response, suggesting that M. tuberculosis suppresses Th17 response as one of the immune evasion mechanisms (28).

Programed death-1 (PD-1)–programed death ligand-1 (PD-L1)/PD-L2 pathway occupies a unique place in the immune evasion strategies employed by M. tuberculosis. Recent data highlight the role of PD-1–PD-L1/PD-L2 axis in modulating regulatory T cell (Treg) and Th1 response to M. tuberculosis (29–33). Whether this pathway also regulates Th17 response to M. tuberculosis is not known. Therefore, in the present study, we have evaluated the role of PD pathway members (PD-L1, PD-L2, and PD-1) in mediating human monocyte- and dendritic cell (DC)-mediated Th17 response to M. tuberculosis.

Several reports have shown that DCs promote Th17 responses to either M. tuberculosis or its antigens (34–37). We found that monocytes and DCs have differential capacity to promote Th17 response to M. tuberculosis and M. tuberculosis-derived antigens. Notably, a prominent IL-17 response was mediated by DCs in comparison to monocytes. Although both monocytes and DCs did not express PD-L2, PD-L1 was significantly enhanced upon stimulation with M. tuberculosis. Similarly, M. tuberculosis stimulation of monocyte/DC–CD4⁺ cocultures also lead to significant increase in the frequency of PD-1⁺CD4⁺ T cells. Importantly, blocking PD-L1 or PD-1 neither significantly altered the frequencies of Th17 cells nor augmented IL-17 secretion from CD4⁺ T cells. Analysis of key Th17-polarizing cytokines indicated that the production of IL-1β was crucial in the establishment of Th17 response to M. tuberculosis. These results thus reveal that the outcome of Th17 response to M. tuberculosis is dictated by the capacity of human innate cells to secrete key Th17-polarizing cytokine (IL-1β) and not expression of members of the PD pathway.

M. tuberculosis-loaded aerosols that enter the lungs interact with resident phagocytes, which include alveolar macrophages and DCs. Different subpopulations of DCs are known to coexist in the human lungs (44). Circulating monocytes form a major reservoir for tissue macrophages and different subsets of DCs (45–47). Moreover, monocytes are known to give rise to DCs in vivo at mucosal surfaces, such as skin (48) and lungs (49, 50), thus suggesting an in vivo relevance of monocyte-derived DCs in mediating immune response to mucosal pathogens, such as M. tuberculosis. In comparison to other infections, there is a delay in the onset of adaptive immune response upon M. tuberculosis infection allowing the bacteria to form a niche in the lungs (51). While it is clear that Th1/IFN-γ response is indispensable for the protection against tuberculosis, the role of Th17/IL-17 in mediating protection is unclear.

Initial studies in mice suggested that the IL-23/Th17 axis is dispensable for the overall protection against M. tuberculosis challenge but is required for enhancing the formation of granuloma in the absence of an active IL-12/Th1 axis (52). Surprisingly, Th17 response mediated protection to a highly virulent M. tuberculosis strain HN878 by activating macrophages and curtailing bacterial burden in the lungs. However, it was dispensable to the less virulent strains of M. tuberculosis thus ascribing a protective role for Th17 to emerging virulent strains (53). Earlier reports in tuberculosis patients indicated that virulent strains, such as multi-drug resistant M. tuberculosis, strongly induce Th17 response. However, enhanced IL-17 was correlated to the severity of the disease and high bacterial burden in the lungs, suggesting a detrimental role for Th17 response in humans (54). More recent reports have demonstrated that tuberculosis patients display lower antigen-specific Th17 response. Incidentally, anti-tuberculosis therapy not only enhanced Th1 response but also augmented Th17 response (28). Furthermore, individuals with biallelic RORC mutations with a defective Th17 response are associated with a compromised Th1 response and are susceptible to fungal and mycobacterial infections (55). Thus, it appears that the protective roles of Th17 response may vary depending on the stage of infection. Th17 response contributes to vaccine-mediated protection and during the earlier stages of infection by recruiting lymphocytes and promoting Th1 response. However, chronic exposure or during the later stages of infection it may be detrimental due to neutrophil recruitment that can mediate tissue damage leading to immunopathology.

Previous studies have shown that DCs can mediate Th17 response to M. tuberculosis by signaling through TLR-2, dectin-1, DC-SIGN, and mannose receptors (36, 53, 56). In the present study, we show that monocytes and DCs have differential capacity to modulate Th17 responses to M. tuberculosis and M. tuberculosis-derived antigen fractions. Notably, DCs evoked a much stronger IL-17A production from CD4⁺ T cells than monocytes. This might be due to the capacity of DCs to secrete large quantities of IL-23 than monocytes. IL-23 plays an important role in stabilizing and sustaining the ensuing Th17 response. Interestingly, unlike monocytes that promoted Th17 response to cell wall, cell membrane, and cytoplasmic fractions of M. tuberculosis, DCs displayed differential response to these antigen fractions. Thus, cell wall fraction triggered strong DC-mediated IL-17A production from CD4⁺ T cells; cell membrane fraction promoted intermediate IL-17A response, and the cytoplasmic fraction did not significantly modulate Th17 response. As immuno suppressor T cells have the ability to inhibit effector T cell responses, we surmised whether increased frequency of IL-10⁺CD4⁺ T cells or FoxP3⁺CD4⁺ Treg cells was responsible for low-level induction of DC-mediated Th17 responses by cytoplasmic fraction of M. tuberculosis (57–60). However, IL-10⁺CD4⁺ T cells were not induced under any stimulatory conditions. Although cytoplasmic fraction significantly enhanced the frequency of FoxP3⁺CD4⁺ Treg cells, this increase was similar to that observed with M. tuberculosis and other antigen fractions.

Suppression of cellular immunity reckons a major evasion strategy employed by M. tuberculosis. Of the numerous evasion strategies employed by M. tuberculosis, the exploitation of PD-L1/PD-L2–PD-1 axis occupies a central role due to its implication in the expansion of Treg response and suppression of effector Th1 response (61, 62). Considering that induced Tregs and Th17 responses are reciprocally regulated, the possible role of PD-L1/PD-L2 in regulating Th17 response has not been questioned. It is well established that M. tuberculosis infection induces PD-L1 on APCs and PD-1 on T cells. Interaction of PD-L1 with PD-1 promotes Treg differentiation and expansion by activating SHP1/2. Induced SHP1/2 suppresses STAT1 functions, abrogates IFN-γ production, and abolishes its inhibitory effect on FOXP3 leading to Treg expansion (63). Since PD-1 axis plays a critical role in mediating immune tolerance and loss of which can predispose to inflammatory conditions and autoimmune diseases (64, 65), it is deleterious to completely abrogate PD-L1–PD-1 signaling during M. tuberculosis infection. This is evidenced by the fact that PD-L1-deficient and PD-1-deficient mice are susceptible to M. tuberculosis infection and display exacerbated inflammation (66, 67).

In the present study, we found that monocytes stimulated with M. tuberculosis and its antigen fractions significantly induce PD-L1 expression on human monocytes. On the contrary, only cell wall and cell membrane fractions induced PD-L1 expression on DCs, whereas the cytoplasmic fraction failed to enhance PD-L1 on DCs. PD-L2 was not expressed on both monocytes and DCs under any stimulatory conditions. The inability of cytoplasmic fraction to induce PD-L1 on DCs was not due to its lack of stimulatory capacity as cytoplasmic fraction inducted DC maturation similar to that of cell wall and cell membrane fractions.

To decipher the role PD-L1 in regulating Th17 response to M. tuberculosis, we employed blocking antibodies to PD-L1. Our results demonstrate that PD-L1 blockade did not significantly alter the frequencies of Th17 cells. On the other hand, IFN-γ production was significantly enhanced when PD-L1 was blocked, thus confirming the previous data on role of PD axis in regulating Th1 responses (31, 32). However, we did observe a marginal (~50 pg/ml) but insignificant increase in IL-17 production upon PD-L1 blockade. As Th17 cells and Tregs are reciprocally regulated and moreover Tregs are known to suppress Th17 response (66, 68, 69), it is likely that blockade of PD-L1 can indirectly favor Th17 response to a certain extent. This could explain minimal augmentation of IL-17A production that we observed upon PD-L1 blockade.

In addition to PD-L1, M. tuberculosis significantly enhanced PD-1 on CD4⁺ T cells. However, PD-1 blockade had no repercussion on either monocyte- or DC-mediated Th17 response. Our data contradict a recent report that indicated that Mycobacterium-induced PD-1 orchestrates suppression of Th17 response in tuberculosis patients (28). This discrepancy might be due to the fact that previous report focused on active disease where prolonged infection can enhance additional co-stimulatory molecules that might not be present during the earlier phases of stimulation. For example, APCs from active disease express both the co-stimulatory molecules PD-L1 and PD-L2 (31), whereas our current data and previous reports show that DCs from healthy individuals stimulated with M. tuberculosis both at the transcript level as well as the protein level induce only PD-L1 and not PD-L2 (30, 33). It is essential to note that we have used γ-irradiated M. tuberculosis or M. tuberculosis-derived cell wall, cell membrane, or cytoplasmic fractions in our experiments, and these conditions may not completely mimic in vivo situation of infection with live bacteria. Therefore, further work is necessary to decipher the role of PD-1–PD-L1/PD-L2 pathway in tuberculosis. Nevertheless, our report indicates that PD pathway does not contribute to negative regulation of human Th17 response to M. tuberculosis. Also, cytoplasmic fraction of M. tuberculosis failed to induce Th17 response despite lack of induction of PD-L1 on DCs.

To decipher the possible mechanism that explains inability of cytoplasmic fraction to promote DC-mediated Th17 response, we analyzed key Th17-polarizing cytokines secreted by monocytes and DCs. Our results indicate that cytoplasmic fraction failed to induce IL-1β, the key cytokine that promotes Th17 response. The amounts of secretion of IL-1β and Th17 responses were significantly correlated. Importantly, exogenous supplementation of IL-1β was sufficient to significantly augment Th17 response by the cytoplasmic fraction. These results indicate that inability of cytoplasmic fraction to induce the production of IL-1β from DCs resulted in its failure to prime a robust Th17 response. Further work on mechanisms underlying the differential production of IL-1β from monocytes and DCs upon stimulation with cytoplasmic fraction is warranted (70). Distinct expression of pattern-recognition receptors that sense cytoplasmic fraction on monocytes and DCs might have resulted in the differential IL-1β secretion upon stimulation with the cytoplasmic fraction. Therefore, antigens that come in direct contact with innate immune cells, such as the cell wall antigens, are attractive candidates for future tuberculosis vaccines. It is important to note that IL-1β is a critical mediator of immunity to M. tuberculosis (71, 72). Our results suggest that this function of IL-1β is in part via enhancement of Th17 responses. As tuberculosis patients were reported to exhibit lower antigen-specific Th17 response (28), supplementing IL-1β might enhance Th17 response.

Taken together, our report demonstrates that DCs have differential capacity to mediate Th17 response to various antigen fractions of M. tuberculosis. Therefore, cell wall antigens of M. tuberculosis constitute potential subunit vaccine candidates. Importantly, we demonstrate that PD pathway is critical for regulating Th1, but not Th17, response to M. tuberculosis. IL-1β however is necessary for promoting a prominent Th17 response to M. tuberculosis.

ES-V and JB designed the research. ES-V, VS, MD, AK, CS, ML, and CG performed the research. ES-V, VS, MD, AK, CS, ML, CG, SK, and JB contributed to data analyses and data interpretation. ES-V and JB wrote the manuscript. ES-V, VS, MD, AK, CS, ML, CG, SK, and JB revised the manuscript critically for important intellectual content and approved the final version.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.