To reduce the burden of Tuberculosis (TB) worldwide, two key components will be necessary: 1) prevention of new TB infection (TBI) and 2) prevention of the progression from TBI to TB disease (1). The global prevalence of TBI is 24.8% (95% CI: 19.7-30.0%) and 21.2% (95% CI: 17.9-24.4%) based on the results of the interferon-gamma release assays (IGRAs) and tuberculin skin test (TST), respectively (2). Prevalence among people who are at higher risk for TBI, including people deprived of liberty (PDL), is generally higher than the community at large in the same geographic location and can be as high as 88.8% (3).

Control of the progression to TB disease relies on accurate TBI diagnosis. However, this is one of the significant challenges in the path toward TB elimination. The main limitations of current tests (TST and IGRAs) are their inability to distinguish between TBI and TB disease or to predict TB progression to active disease (4–8). Despite numerous articles reporting immune markers associated with TBI and TB disease, there is high heterogeneity in the design, participant selection, sample processing (including the stimulation protocols), measured markers and analysis that limits comparing study results (9).

TB disease occurs most frequently among people newly infected with TB, mainly within the first year of acquiring mycobacterial infection (10). Identifying and prioritizing those with new TBI within the first year of mycobacterial infection to offer TBI treatment could be an effective measure to decrease TB transmission.

This study focused on incarcerated population because of the high risk of exposure to TB and progress to TB disease (11). We analyzed data and samples collected from two prisons where we have previously shown high rates of TB infection and disease (12, 13). This study aimed to determine the plasma concentration of 18 cytokines/chemokines associated with the presence of new Mycobacterium tuberculosis (MTB) infection and compare it with the concentration among people with pulmonary TB diagnosis and exposed but uninfected people.

The main results of our study were: 1) the concentration of immune mediators in persons with a new TBI varies according to the time of incarceration (short or long incarceration); and 2) Among people with short incarceration, high concentrations of MIP-3α were associated with a higher risk of a new TBI, and higher concentrations of Eotaxin was associated with a lower risk of a new TBI. Higher concentrations of sCD14 and TNF-α were associated with a higher risk of TB disease, and higher concentrations of IL-18 and MCP-1 were associated with a lower risk of TB disease.

In this study, immune mediators’ concentration varied among people newly infected by the time of incarceration. Levels of sCD14 and IL-18 were increased in individuals with short incarceration and IP-10 in those with long incarceration. These variations between people could occur for several reasons: 1) early exposure to M. tuberculosis can induce increased macrophage activation, leading to elevated sCD14 concentration. 2) Continuously exposed individuals (≥1 year in prison) could eliminate the bacteria repeatedly, owing to an effective immune response. 3) Other factors associated with prison entry include stress, physical activity, or dietary patterns. Several studies have documented changes in the production of pro-inflammatory cytokines when people are under a chronic stressful situation (25), variable levels of physical activity/exercise (26), or when a stress-generating situation is compounded by an infectious process, such as influenza vaccination (27). People deprived of liberty are exposed to acute and chronic stressors, given the social and safety conditions that prevail in prisons in which they live or the changes associated with freedom deprivation (28). In other words- the inflammatory milieu induced by environmental factors related to imprisonment may alter the immune response, thereby increasing the risk of MTB infection (29).

We found that increased circulating levels of MIP-3α were associated with a higher risk of having new TBI after short incarceration. The MIP-3α/CCL20 is a dendritic cell, T cell, B cell and monocyte chemoattractant involved in lymphocyte homeostasis and trafficking, cell proliferation and activation (21, 30, 31). Rivero-Lezcano et al. reported that the expression of MIP-3α increased up to 39-fold when monocytes from healthy donors were infected with M. tuberculosis ( 32), and Lee et al. showed that MIP-3α was up-regulated in PBMC and bronchoalveolar lavage fluids from people diagnosed with TB compared to healthy controls after in vitro stimulation with the 30 kDa antigen (Ag) of M. tuberculosis ( 33). MIP-3α concentration in serum/plasma samples is an attractive target for further evaluation in newly acquired MTB infection studies among people with documented recent exposure. In our research, PDL with a new TBI and with short incarceration had lower circulating levels of Eotaxin. Eotaxin has been reported as inhibited after in vitro infection by M. tuberculosis ( 34), decreased in people diagnosed with TB disease, and with increasing levels after anti-TB treatment (35).

After adjusting for the other co-variables, individuals with elevated concentrations of MCP-1, a potent chemotactic factor for monocytes (36), had a lower risk of TB disease. In a model of macaques previously vaccinated with BCG, the production of β-chemokine MCP-1 increased in lung lesions of animals after five weeks of infection with MTB (37). These immune molecules are mainly associated with the recruitment of monocytes/macrophages, in keeping with the observation that animals had a higher frequency of macrophages (CD14+ CD68+) in peripheral blood three weeks post-infection with MTB (37). In this case, the increase in cells and β-chemokines may be a sign of the recruitment of these cells from peripheral blood to the site of infection in the lung. Our results show that the concentrations of MCP-1 were lower among the TB disease group compared to non-infected individuals; however, most reports about this chemokine have shown increased concentration among people diagnosed with TB (38, 39). Results may be different because of stimulation with mycobacterial antigens or the absence of stimulation in whole blood samples. Increased concentrations of MCP-1 have been reported in stimulated samples from individuals diagnosed with active TB, but higher concentrations in un-stimulated samples among non-infected individuals compared to individuals diagnosed with TB (40). Some additional explanations for those differences between the studies are the ancestry of the population, the specimen (serum, whole blood, or supernatants of cultured PBMC), and the presence of genetic polymorphisms regulating IL-18 or MCP-1 production (41) in the populations, which could alter the final structure, concentration, or function of the protein.

We also found that higher concentrations of sCD14 and TNF-α were associated with a higher risk of TB disease. These immune mediators are highly involved in monocyte/macrophage activation and trafficking pathways (42, 43). Macrophages are the main niche for the growth and survival of M. tuberculosis ( 44); likewise, macrophages are the most important cells for infection control in animal and human models and critical cells in the host response during active disease (44–46) trying to limit the systemic spread of mycobacteria (47). Soluble CD14, whose membrane-bound portion is highly expressed in these cells, is a mediator of macrophage activation and serves as a receptor for mycobacterial lipoarabinomannan (43). It has been reported among people diagnosed with TB disease (48, 49), with or without concurrent HIV infection (47, 50), and individuals with diabetes mellitus (51). A published article by Lawn SD et al. suggests that sCD14 might increase the concentration of TNF-α due to the high load of mycobacterial antigens among people living with HIV and co-infected with TB disease (47). In our study, people diagnosed with TB had higher plasma concentrations of TNF-α, presumably contributing to cell recruitment, the production of other pro-inflammatory cytokines, and apoptosis of MTB-infected cells (52). Similar to our study, TNF-α has been reported to be increased in adults (49, 53–56) and children (57) diagnosed with TB disease.

Our findings show lower concentrations of IL-18 and MCP-1 among the TB group compared to non-infected individuals. Other researchers have shown that M. tuberculosis-stimulated culture supernatants from people diagnosed with TB have lower concentrations of IL-18 compared to those from healthy TST converters (58). IL-18 has an important function in TB as a pro-inflammatory cytokine (59); it plays an important role in the T-cell-helper type 1 (Th1) response, primarily by its ability to induce IFN-γ production in T cells and natural killer (NK) cells (60) and in combination with IL-12 triggers the antimicrobial protein cathelicidin and autophagy, resulting in inhibition of intracellular mycobacteria in macrophages and lung epithelial cells (61). Still, studies in humans are not conclusive. For example, Yamada G et al. (62) showed that increased serum IL-18 concentrations were associated with TB disease compared to healthy individuals (including TST converters).

Future studies are needed to validate and complement our cytokine and chemokine findings.

We did not find associations between IP-10, IL-17, IL-10, and a new TBI (the first one was not different between groups, and the latter two had concentrations below the lower detection limit), contrary to other publications (63, 64). This discrepancy may be attributed to the measurement of immune mediators after mycobacterial antigen and mitogen stimulation in prior studies (53, 56, 65–75), in contrast to the unstimulated measurements in our study. In the case of IL-17, there are six members in the IL-17 family, including IL-17A. IL-17B, IL-17C, IL-17D, IL-17E, and IL-17F (76, 77). Our study only measured Il-17A. Therefore, it is possible that other not measured subunits, such as IL-17F, were increased in plasma samples. In the same way, IL-17 is described as essential in the lung during recent infection (60). Still, perhaps blood sampling in the context of recent infection may have contributed to the reduced systemic level of IL-17.

Another reason for discrepant results may be that individuals included in other longitudinal studies were evaluated with only one TST administration, and those with a negative TST may represent a false negative result (61). In our previous studies, a second administration of TST identified an additional 11.6% positive TST individuals (78).

Our study’s most crucial distinguishing feature is the study design (cohort). In cross-sectional studies, the duration an individual has had TBI cannot be quantified. The main limitation of a cross-sectional approach is that human and primate studies have demonstrated there is a spectrum of TB stages (5, 6, 79, 80), on which there is little published work (4, 6), and that potentially could alter the concentration of the immune parameters according to the stage of the infection. The cohort design allowed us to quantify the concentration of cytokines at a temporal point close to the time of infection (new infection) and identify individuals in whom infection occurred recently. Borgstrom WE et al., using mathematical models and CD4+ T-cell flow-cytometry data, showed that the most specific prediction of recent TBI was a high proliferative CD4+ response to CFP-10 and PPD, and a low response to ESAT-6 at ≤1 month after exposure (81).

Our study has other strengths: rigorous selection criteria among people diagnosed with TB, including only those with microbiological confirmation and with less than 15 days of treatment, rule out of booster effect, and rigorous monitoring and follow-up every three months to all uninfected participants.

The main limitation of this study is that some bacterial factors may modify the response to cytokines, such as the virulence of some strains; however, in people with TBI, it is not feasible to isolate the mycobacterium, and therefore the role of MTB strain cannot be assessed yet.

Our study found that immune markers vary according to the time of incarceration. Among people with short incarceration, high concentrations of MIP-3α were associated with a higher risk of a new TBI, and higher concentrations of Eotaxin were associated with a lower risk of a new TBI. Higher concentrations of sCD14 and TNF-α were associated with a higher risk of TB disease, and higher concentrations of IL-18 and MCP-1 were associated with a lower risk of TB disease. It is necessary to have more cohort studies that evaluate the changes over time of these and other cytokines/chemokines to understand the immune mechanisms across the spectrum of TB.

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 Ethics Committees of the Universidad Pontificia Bolivariana and the University of Manitoba. The patients/participants provided their written informed consent to participate in this study.

Substantial contributions to the conception or design of the work: MH, YK, ZR. Acquisition of data: MH. Analysis and interpretation of data: MH, YK, LL, DM, ZR. Drafting the article and revising it critically for important intellectual content: MH, YK, LV, PM, ZR. Final approval of the version to be published: MH, YK, LV, LL, DM, PM, ZR. Agreement to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved: MH, YK, LL, DM, LV, PM, ZR. Principal investigator and funding acquisition: ZVR. All authors contributed to the article and approved the submitted version.