Tuberculosis (TB) is the leading cause of death due to a single infectious agent (1). Although TB is largely a curable disease with a global treatment success rate of ~85%, almost 1.5 million people succumbed to the disease in 2019 (1). This is likely to worsen due the rapid emergence and global spread of drug resistant forms of the disease which threaten to derail advances in TB control (2). Eradication of TB is only likely to be achieved with an effective vaccine. Currently, the only licensed TB vaccine is an attenuated strain of Mycobacterium bovis (M.bv), BCG (3). BCG confers protection against severe forms of TB disease in children but offers limited protection (~30%) against adult forms of pulmonary disease (4). Data from recently trialled vaccine candidates have not offered much hope. For instance, MVA85A did not show any improvement compared to BCG despite compelling pre-clinical animal data (5), and the more recently trailed M72/AS01 performed much better but was still only associated with an efficacy of only ~50% (6). These studies and their outcomes highlight our incomplete understanding of the immune mechanisms underpinning protective immunity and bacterial persistence.

One of the central hallmarks of active TB disease is a failed T-cell effector immune response (7). However, there is still much that is not known about the specific immune mechanisms underpinning failed T-cell immunity (8). In active TB disease immunity associated with recovery is traditionally thought to be primarily driven by pro-inflammatory effector T-cells, and cytotoxic CD8 cells (8). However, the host immunity also involves several regulatory mechanisms for suppressing T-cell responses against M. tb-specific antigens, which may be leveraged by pathogens to their advantage. Such strategies may also underpin the extensive immunopathology associated with TB.

Myeloid-derived suppressor cells (MDSCs) is one type of regulatory cell that potently regulates cancer immunity at the site of pathology (9). MDSCs have a morphology similar to granulocytes and/or monocytes (10). In healthy individuals, immature myeloid cells are generated in the bone marrow and rapidly differentiate into mature macrophages, dendritic cells, or granulocytes (10). MDSCs can broadly be classified into three groups: 1) early-stage-MDSC (LIN1^-HLA-DR^-/low CD11B⁺CD33⁺), 2) polymorphonuclear-MDSC (HLA-DR^-/low CD14^-CD15⁺CD33^+/dim), and 3) monocytic-MDSC (HLA-DR^-/lowCD14⁺CD15^-CD33⁺) (10). This report specifically focuses on polymorphonuclear-MDSCs (PMN-MDSCs), also known as granulocytic-MDSCs.

Others have previously shown that regulatory pathways (11, 12), including MDSCs are upregulated in patients with active TB (13–17), their frequency in peripheral blood decreases with successful treatment (14), and MDSCs are able to inhibit T-cell proliferation (13). However, data are limited to a handful of studies (less than 5), and there are hardly any data from humans or TB-endemic settings (13). Furthermore, there are no published data about the biological significance of these cells and the ability of MDSCs to directly restrict mycobacterial growth. Thus, their role during M. tb infection remains unclear. To address these knowledge gaps, we explored the levels of M-MDSCs in patients with active TB versus those who are latently infected, and whether they abrogated the ability of effector T-cells to contain mycobacterial growth.

We found that patients with active TB disease had significantly higher frequencies of circulating MDSCs. These cells suppressed the proliferative capabilities of PPD-specific effector T cells and attenuated mycobacterial containment in vitro.

The role of MDSCs in TB has been contentious. We have shown, for the first time in humans, that MDSCs can modulate T-cell activity, thus enhancing mycobacterial growth in vitro. Our results indicate that MDSCs can modulate an immune response that favours M.tb survival. The mechanism by which this occurs deserves further study. Previous murine studies that examined the direct effect of MDSCs on mycobacterial survival, showed that infected MDSCs were unable to directly kill mycobacteria (19), despite their ability to produce nitric oxide (20). Knaul et al. found that M.tb-infected MDSCs, similar to macrophages, can release pro-inflammatory (IL-1 and IL-6), but also anti-inflammatory cytokines (IL-10), thus providing a possible mechanism for our observations (21). However, further work in now required to study an array of possible mechanisms by which might potentially including Th2 cytokines, cell-to-cell contact, and other humoral mechanisms etc.

We found the levels of MDSC were significantly higher in patients with active TB compared to those with presumed-LTBI, and that MDSCs suppressed T-cell proliferation. Indeed, MDSCs were first described as suppressor cells recruited to tumor sites (10), where the increased levels of MDSCs significantly correlated with aggressive disease, poor patient prognosis and immune escape (18). More recently, MDSCs were shown to be upregulated during active TB disease in adults (13, 14) and in HIV-TB co-infected children (17), and in the same patients are able to suppress T cell proliferation (13, 14) and alter cytokine expression (13). However, although we confirm these observations, we have extended these findings to show that the anti-proliferative and immune-modulatory effects translate into sub-optimal mycobacterial containment.

There are several limitations to our findings. The immuno-phenotype of study participants was analysed in the peripheral blood compartment, and whether these effects occur at the site of disease remain unknown. However, harvesting cells from the lung is constrained by cost, ethical and infection control considerations. Second, our study sample size was limited, and we only recruited patients from one geographical setting. Nevertheless, we were able to demonstrate statistically significant differences between the groups, which together with the impact on proliferation, suggest that these are likely biologically meaningful observations. Third, we did not explore the immune mechanism by which MDSCs subverted the ability of T-cells to kill M.tb. However, this was limited by available funding but is likely to begin shortly.

Fourth, containment was used to define how MDSCs can modulate the function of effector T cells to control M. tb. As described in our previous publication (22), it is difficult to quantify anti-mycobacterial activity as our assay cannot distinguish between organisms that have been killed or those that have entered a non-replicating state. However, containment is a much more biologically meaningful outcome measure as compared to biomarker proxies, including cytokines or receptor signaling pathways, that may indicate protection. Although the latter are commonly used in human studies, they may merely represent a bystander effect rather than having a causal link.

In conclusion, persons with active TB disease, compared to those with presumed-LTBI, demonstrate an altered immuno-phenotype characterized by a high frequency of MDSCs. MDSCs isolated from patients with active TB disease were highly suppressive and could attenuate mycobacterial containment in vitro. Collectively, these data suggest a contributing role for MDSCs in the immuno-pathogenesis of TB and MDSCs may be potential targets that could be exploited to design vaccines or host-directed therapies.

LS is currently based at Stellenbosch University, but all work outlined in this study was performed when she was based at the Centre for Lung Infection and Immunity at the University of Cape Town.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The studies involving human participants were reviewed and approved by University of Cape Town Research committee. Patients/participants provided their written informed consent to participate in this study.

Study concept and design: MD, AP, and KD. Laboratory work: MD, AP, LS, and MT. Analysis and interpretation: MD, AP, LS, MT, and KD. Drafting of manuscript: MD, AP, LS, MT, and KD. All authors contributed to the article and approved the submitted version.

KD and the work presented here was supported by the South African MRC (RFA-EMU-02-2017) and the EDCTP (TMA-2015SF-1043 & TMA- 1051-TESAII).

The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fimmu.2021.676679/full#supplementary-material

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.