Schistosome Soluble Egg Antigen Decreases Mycobacterium tuberculosis–Specific CD4⁺ T-Cell Effector Function With Concomitant Arrest of Macrophage Phago-Lysosome Maturation

Abstract

Helminth-infected individuals possess a higher risk of developing tuberculosis, but the precise immunologic mechanism of Mycobacterium tuberculosis control remains unclear. We hypothesized that a perturbation of the M. tuberculosis–specific CD4⁺ T-cell response weakens the ability of macrophages to contain M. tuberculosis. We exposed peripheral blood mononuclear cells from M. tuberculosis–infected humans to schistosome soluble egg antigen (SEA) and then profiled M. tuberculosis–specific CD4⁺ T cells via multiparametric flow cytometry. SEA decreased the frequency of cells producing interferon γ (6.79% vs 3.20%; P = .017) and tumor necrosis factor α (6.98% vs 2.96%; P = .012), with a concomitant increase in the median fluorescence intensity of interleukin 4 (IL-4; P < .05) and interleukin 10 (IL-10; 1440 vs 1273; P < .05). Macrophages polarized with SEA-exposed, autologous CD4⁺ T-cell supernatant had a 2.19-fold decreased colocalization of lysosomes and M. tuberculosis (P < .05). When polarized with IL-4 or IL-10, macrophages had increased expression of CD206 (P < .0001), 1.5-fold and 1.9 fold increased intracellular numbers of M. tuberculosis per macrophage (P < .0005), and 1.4-fold and 1.7-fold decreased colocalization between M. tuberculosis and lysosomes (P < .001). This clarifies a relationship in which helminth-induced CD4⁺ T cells disrupt M. tuberculosis control by macrophages, thereby providing a mechanism for the observation that helminth infection advances the progression of tuberculosis among patients with M. tuberculosis infection.

Although Mycobacterium tuberculosis infects one-third of the world population, current evidence suggests that only 5%–10% of infected individuals progress to active M. tuberculosis infection [1]. The immune system ensures the majority of infected individuals remain asymptomatic, but there exist processes that undermine protective immunity against M. tuberculosis to permit the development of active M. tuberculosis infection. Helminths, which infect one-quarter of the world population, are a prominent risk factor for developing latent M. tuberculosis infection [2], increased incidence of progression from latent to active M. tuberculosis infection [3], and more-severe tuberculosis [4]. Investigating how helminths perturb the immune system to promote tuberculosis progression may be a means to clarify protective immunity to M. tuberculosis infection.

The mechanism by which CD4⁺ T lymphocytes control M. tuberculosis is unknown, but there is substantial observational evidence in humans for this phenomenon. Impaired T-cell function, for example via human immunodeficiency virus infection, tumor necrosis factor α (TNF-α) inhibitors, interferon γ (IFN-γ) signaling defects, and Mendelian susceptibility to mycobacterial disease, is associated with an increased risk of tuberculosis [5]. Helminth infection has profound consequences in the CD4⁺ T-cell compartment, as well. In both human and murine experimental systems, helminths induce interleukin 10 (IL-10) production, T-helper cell type 2 (Th2) polarization (interleukin 4 [IL-4], interleukin 5, and interleukin 13), and increased frequencies of Foxp3⁺CD25⁺ T regulatory cells, collectively promoting wound healing and pathogen compartmentalization, rather than pathogen clearance [6, 7]. This helminth-induced profile results in a so-called negative bystander effect, downregulating effector T-cell responses and Th17 immunity to tetanus toxoid [8], cholera toxin [9], and bacillus Calmette-Guerin vaccine [10]. The negative bystander effect of helminth infection on M. tuberculosis–specific T-cell immunity [11, 12] offers a model of CD4⁺ T-cell function perturbation that may illuminate a mechanism of tuberculosis control. We used schistosome soluble egg antigen (SEA) to recapitulate the helminth immune phenotype [13–15] and evaluated its effect on the M. tuberculosis–specific CD4⁺ T-cell response. We hypothesized that SEA would skew the response away from an effector profile characterized by interleukin 2 (IL-2), TNF-α, and IFN-γ [16] toward a helminth-induced profile highlighted by IL-10 and IL-4 production.

Since macrophages phagocytize M. tuberculosis, their interaction with the bacterium determines its fate and, hence, the health of the host. Macrophages eliminate intracellular M. tuberculosis via the formation of a phago-lysosome; the phagosome containing M. tuberculosis fuses with a lysosome containing acidic hydrolases [17, 18]. We hypothesize that the functional response by M. tuberculosis–specific CD4⁺ T cells influences the M. tuberculosis–macrophage interaction to promote either disease control or development. By extension, we hypothesize that helminth-induced perturbation of CD4⁺ T-cell function should negatively influence the maturation of the macrophage phago-lysosome and, thus, promote intracellular M. tuberculosis survival. While it has been published that alternative macrophage activation is permissive to mycobacterial growth and that IL-4 neutralization improves macrophage control of M. tuberculosis [19, 20], the actual mechanism(s) underlying these observations has not been elucidated. We analyzed the macrophages for their ability to internalize M. tuberculosis and acidify internalized M. tuberculosis via the formation of a phago-lysosome [17, 18]. Whereas most of the literature on tuberculosis investigated either the macrophage–M. tuberculosis interaction or M. tuberculosis–specific T-cell responses individually, our study aims to define the relationship between these 2 immune compartments essential for M. tuberculosis control.

MATERIALS AND METHODS

Ethics Statement

All samples were obtained from subjects in compliance with institutional guidelines for the Protection of Human Subjects and the Declaration of Helsinki. The study was approved by Houston Methodist Hospital and Baylor College of Medicine institutional review boards (IRBs), and written informed consent was obtained from subjects. For M. tuberculosis–specific enzyme-linked immunosorbent assay (T-SPOT TB assay) positive peripheral blood mononuclear cells (PBMCs), as recommended by the Department of Health and Human Services policy and guidance materials [21] and with IRB approval, excess PBMCs marked for discard were used after de-identification of clinical information.

Antigens

For M. tuberculosis–specific cell stimulations, cells were stimulated with overlapping peptide pools of the early-secreted antigenic target (ESAT-6) and culture filtrate protein (CFP-10) of M. tuberculosis. These reagents (peptide Arrays NR-34824 and NR-34825) were obtained through BEI Resources, National Institute of Allergy and Infectious Diseases, National Institutes of Health. SEA was derived from Schistosoma mansoni as previously described by the Schistosomiasis Resource Center at the Biomedical Research Institute (Rockville, MD) [22]. We quantified the endotoxin in this batch of SEA (final concentration, 0.003 ng/mL or 0.03 endotoxin units/mL), which is acceptably low for immunologic study.

SEA Exposure

PBMCs from 26 subjects were isolated from ACD blood by a density gradient technique, using Ficoll; washed in Roswell Park Memorial Institute (RPMI) and Aim5 media; and then kept in R20 media (RPMI, 20% fetal calf serum, and 1% penicillin/ streptomycin) for 24 hours until the T-spot result was known. PBMCs that had positive ESAT-6 or CFP-10 IFN-γ T- SPOT TB responses were divided and cultured in 12-well plates with 10⁶ PBMCs in R10 media (10% FBS, RPMI, and penicillin/streptomycin) with or without SEA (50 ng/mL) and 50 units/mL of interleukin 2 (IL-2). On day 2, M. tuberculosis–specific peptides ESAT-6 and CFP-10 (1 ng/mL) were added to PBMCs that were or were not exposed to SEA. Media were replenished every 3–4 days.

Multiparametric T-Cell Profiling

On day 14 of SEA exposure, cells were stimulated with M. tuberculosis–specific antigens ESAT-6 and CFP-10 (5 µg/mL) for 12 hours. Dimethyl sulfoxide and staphylococcal enterotoxin B were used as negative and positive controls, respectively. Cell stimulations occurred in the presence of brefeldin A (1 µg/mL), Golgi-Stop (0.7 µg/mL), and CD28/CD49 (2.5 µg/mL). Following stimulation, cells were stained for viability (Ghost dye); the surface antigens CD3, CD8, CD4, PD-1, and CTLA-4; and the intracellular proteins IFN-γ, TNF-α, IL-2, IL-4, and IL-10. Cells were then acquired on a BD LSR II Fortessa flow cytometer as previously described [23]. Representative gating is illustrated in Supplementary Figure 1. Reported values were corrected for background from the negative stimulation. Data represent the median responses of 26 patients.

Monocyte-Derived Macrophage (MDM) and Immune Polarization

Macrophages were derived from monocytes using plastic adhesion. Fresh PBMCs were plated onto chamber slides. After 4 hours, the nonadherent cells were washed off, and the adherent cells were cultured for 7 days in RPMI with 10% human AB serum and 10 ng/mL each of macrophage colony-stimulating factor and granulocyte macrophage colony-stimulating factor. After 7 days, macrophages were polarized using supernatant from SEA-exposed autologous CD4⁺ T cells, IL-4 (20 ng/mL), IL-10 (20 ng/mL), or IFN-γ (50 ng/mL). After 48 hours, the cells were incubated with live Td Tomato-labeled M. tuberculosis H37Rv at a multiplicity of infection of 3:1 [24]. After incubation for 4 hours, nonphagocytized bacilli were removed with washing, and the cells were permeabilized and fixed. Cells were stained with lysosomal marker CD107a and either the nuclear stain DRAQ5 or F-actin. All tests were performed in triplicate, with at least 20 macrophages counted per condition.

Image Acquisitions and Analyses

Confocal images were acquired with a Leica SP8 confocal microscope. Detection of CD107a and M. tuberculosis fluorescence was thresholded greater than twice the level of macrophage autofluorescence, and colocalization was measured using Manders 2 coefficient (Volocity). Images were acquired and analyzed in the x-, y-, and z-axes at 0.5-μm-thick slices with a median thickness of 10 μm (range, 4–21 μm) per image. To measure internalization, Volocity automated cell counting identified M. tuberculosis and MDMs with thresholds set at twice the level of macrophage autofluorescence. Samples were also acquired on an imaging flow cytometer (ImageStream X Mark II, Amnis). Approximately 1000 cells were acquired per condition using a 60× lens.

Statistics

Data analyses were performed using FlowJo X (TreeStar, Ashland, OR); GraphPad 6.0 (GraphPad Software); Volocity 3D Image Analysis Software, version 6.3 (Perkin Elmer); IDEAS, version 5.0 (AMNIS); and SPICE (available at: http://exon.niaid.nih.gov/spice). Statistical differences between the frequencies of paired CD4⁺ T cells producing cytokines with or without SEA exposure were calculated using a Wilcoxon-rank test. Comparison of the CD4⁺ T-cell functional distribution was evaluated in SPICE, using the χ² partial permutation test [25]. For paired samples, comparison between categories was evaluated using a Wilcoxon rank test. For colocalization of phagocytized M. tuberculosis with lysosomal marker CD107a, Manders coefficient was calculated.

RESULTS

SEA Diminishes the M. tuberculosis–Specific CD4⁺ T-Cell IFN-γ and TNF-α Response

To test the premise that helminths subvert the M. tuberculosis–specific CD4⁺ T-cell response, we devised an experimental system in which human PBMCs from subjects with M. tuberculosis infection, as defined by T-SPOT positivity, were stimulated with peptides pools, representing the M. tuberculosis antigens CFP-10 and ESAT-6, in the presence and absence of SEA, a classic model for studying the immunomodulatory effects of helminth infection [22, 26–28]. Previous studies correlated defects in TNF-α and IFN-γ production with an increased risk for tuberculosis progression, suggesting a role for these effector cytokines in protection from M. tuberculosis [5]. SEA pretreatment significantly diminished the frequency of CD4⁺ T cells responding with effector cytokines to the M. tuberculosis proteins ESAT-6 and CFP-10 (Figure 1A and 1B): for IFN-γ, the non-SEA median value was 6.79%, compared with the SEA median value of 3.20% (n = 26; P < .0001, by the Wilcoxon matched-pairs signed rank test); for TNF-α, the non-SEA median value was 6.98%, compared with the SEA median value of 2.96% (n = 26; P < .0001, by the Wilcoxon matched-pairs signed rank test). These data indicate that helminth infection alters antigen-specific T-cell responses but does not cause alterations to the general T-cell compartment.

Figure 1.

Schistosome soluble egg antigen (SEA) decreases the frequency of Mycobacterium tuberculosis–specific CD4⁺ T cells producing tumor necrosis factor α (TNF-α) and interferon γ (IFN-γ). Peripheral blood mononuclear cells (PBMCs) from M. tuberculosis–infected subjects were exposed to SEA for 14 days and then restimulated with M. tuberculosis–specific peptide pools (early-secreted antigenic target [ESAT]-6 and culture filtrate protein [CFP]-10) to measure CD4⁺ T-cell functionality. A, Dot plots illustrating the definition of CFP-10–specific IFN-γ and TNF-α responses. Top row, cells not exposed to SEA; bottom row, cells exposed to SEA. The numbers next to the gates indicate proportion of CD4⁺ T cells. Data are representative of all similar analyses. Representative staphylococcal enterotoxin B (SEB) dot plots are shown in the supplement. B, Comparison of the proportion of total M. tuberculosis–specific (ESAT-6⁺CFP-10⁺) CD4⁺ T cells producing a particular cytokine (y-axis) with and without SEA exposure (x-axis; left). Reported values were subtracted from background no stimulation response. Each data point represents a response from an individual patient (n = 26). SEB superantigen stimulation was a positive control (right). C, The bar graph communicates the absolute frequency of M. tuberculosis–specific CD4⁺ T cells that produced TNF-α, IFN-γ, and/ or interleukin 2 (IL-2) after SEA exposure (blue bars) or no SEA exposure (red bars). The overall M. tuberculosis–specific CD4⁺ T-cell response is illustrated as a pie graph. Each pie slice corresponds to a discreet functional subpopulation, outlined in the bar graph below. The black arcs denote the proportion of the response accounted for by IFN-γ (P < .05, by the χ² partial permutation test, for the difference between the pies). Data represent the mean responses from 26 patients.

A polyfunctional (TNF-α⁺IL-2⁺IFN-γ⁺) M. tuberculosis–specific CD4⁺ T-cell response has been associated with asymptomatic control of M. tuberculosis infection, while a monofunctional response dominated by TNF-α was shown to distinguish active tuberculosis [16]. Therefore, we evaluated the effect of SEA on the polyfunctional nature of M. tuberculosis–specific CD4⁺ T cells. M. tuberculosis–specific CD4⁺ T cells exposed to SEA prior to stimulation were less capable of the simultaneous production of TNF-α and IFN-γ (median, 2.1% for no SEA vs 0.8% for SEA; n = 26, P < .05, by the Wilcoxon signed rank test; Figure 1C). This double-positive subpopulation constitutes the largest proportion (55.8%) of the total M. tuberculosis–specific CD4⁺ T-cell response from our non–SEA-exposed cells, confirming the importance of these cytokines in M. tuberculosis control. Exposure to SEA reduced the contribution of the TNF-α⁺IFN-γ⁺ subpopulation to the total M. tuberculosis–specific CD4⁺ T-cell response to less than half of the normal response (Figure 1C).

SEA Exposure Increases Relative IL-4 and PD-1 Expression on M. tuberculosis–Specific CD4⁺ T Cells

Concomitant to the reduced frequency of M. tuberculosis–specific TNF-α⁺IFN-γ⁺CD4⁺ T cells was a rise in the proportion of cells with neither of the 3 effector functions measured. Since only 3 functions were quantified, it could be that the cells still responded, except the functional landscape had shifted toward a different profile. To this end, we expanded our panel of cytokines to include those traditionally associated with helminth infection, IL-4 and IL-10. While SEA exposure did not yield an absolute increase in IL-4 production (Figure 1B), there was a relative increase among the total response of M. tuberculosis–specific CD4⁺ T cells responding with only IL-4 (2.5% vs 0.5%; P < .05, by the χ² test; Figure 2A). Although there was no increase in the absolute frequency of IL-10–producing M. tuberculosis–specific CD4⁺ T cells, the median fluorescence intensity (MFI; a measure of the amount produced per single cell) of IL-10 increased after SEA exposure (MFI, 1440 for SEA and 1273 for non-SEA; P < .05; Figure 2C). These data reveal that M. tuberculosis–specific CD4⁺ T cells experience a modification to their functional capacity upon exposure to SEA.

Figure 2.

Decrease in Mycobacterium tuberculosis–specific effector CD4⁺ T-cell function is associated with an increase in interleukin 4 (IL-4) production. Peripheral blood mononuclear cells (PBMCs) from M. tuberculosis–infected subjects were exposed to schistosome soluble egg antigen (SEA) for 14 days and then restimulated with M. tuberculosis–specific peptide pools (early-secreted antigenic target [ESAT]-6 and culture filtrate protein [CFP]-10) to measure CD4⁺ T-cell functionality. A, Proportion of total M. tuberculosis–specific (ESAT-6 and CFP-10) CD4⁺ T cells that produced interferon γ (IFN-γ), IL-4, interleukin 10 (IL-10), interleukin 2 (IL-2), and/or tumor necrosis factor α (TNF-α) after SEA exposure (blue bars) or no SEA exposure (red bars). Above, the overall M. tuberculosis–specific CD4⁺ T-cell response is illustrated as a pie graph. Each pie slice corresponds to a discreet functional subpopulation, outlined in the bar graph below. The black arcs denote the proportion of the response accounted for by IFN-γ, while the teal and green arcs denote the proportion of IL-4 and IL-10 response, respectively (P < .05, by the χ² partial permutation test, for the difference between the pies). Data represent the mean responses for 26 patients. B, Dot plots illustrating the definition of CFP-10–specific IFN-γ and IL-4 responses. The numbers next to the gates indicate proportion of CD4⁺ T cells. C, Quantity of IL-10 as measured by median fluorescent intensity (MFI; P < .0164 by the Wilcoxon matched-pairs signed rank test). Abbreviation: MFI, median fluorescence intensity.

Many studies of CD4⁺ T cells in the context of helminth and M. tuberculosis infections have reported a correlation between disease progression and an exhausted T-cell profile [12, 29, 30]. To further characterize the helminth-induced phenotype of M. tuberculosis–specific CD4⁺ T cells in our system, we quantified the expression of the putative exhaustion molecules PD-1 and CTLA-4. After SEA exposure, there was an 83.3% increase in the proportion of M. tuberculosis–specific CD4⁺ T cells expressing PD-1 (P < .05, by the χ² test; Figure 3A). The majority of cells expressing PD-1 did not perform any function. In contrast, there was no increase in CTLA-4 expression on the surface of M. tuberculosis–specific CD4⁺ T cells (data not shown). Since PD-1 is associated with an increase in apoptosis [31], we measured the amount of Annexin V on M. tuberculosis–specific CD4⁺ T cells: after 4 days of SEA exposure, we detected an increase in Annexin V expression (Figure 3B). Together, the data suggest that SEA reduces the reactive capacity of M. tuberculosis–specific CD4⁺ T cells.

Figure 3.

Schistosome soluble egg antigen (SEA) exposure induces PD-1 expression on Mycobacterium tuberculosis–specific CD4⁺ T cells. M. tuberculosis–infected subjects were exposed to SEA for 14 days and then restimulated with M. tuberculosis–specific peptide pools (early-secreted antigenic target-6 and culture filtrate protein-10) to measure CD4⁺ T-cell functionality. A, Proportion of M. tuberculosis–specific CD4⁺ T cells that express PD-1 (yellow arcs) or interferon γ (IFN-γ; black arcs), with and without exposure to SEA (P < .05, by the χ² partial permutation test, for the difference between the pies). Below, a bar graph illustrates the distribution of M. tuberculosis–specific CD4⁺ T-cell functional subpopulations with simultaneous PD-1 expression. B, SEA modulation of Annexin V expression (y-axis) on CD4⁺ T cells. Abbreviations: IL-2, interleukin 2; IL-4, interleukin 4; TNF, tumor necrosis factor.

Cytokine Polarization Undermines M. tuberculosis–Specific Macrophage Function

Since CD4⁺ T cells are critical mediators of protection against M. tuberculosis disease progression and M. tuberculosis infects macrophages initially, we speculated that CD4⁺ T cells and macrophages must cooperate to contain M. tuberculosis. Given our disparate functional profiles between M. tuberculosis–specific CD4⁺ T cells that were or were not exposed to SEA, we hypothesized that the cytokines produced in each setting may influence the macrophage–M. tuberculosis interaction differentially. It is well known that helminths induce an alternative activation profile by macrophages, resulting in a loss of M. tuberculosis containment [19, 20], but the cell biology mechanism explaining this observation is not known.

Virulent strains of M. tuberculosis might thrive inside hostile macrophages by blocking the formation of the phago-lysosome, created by the fusion of the M. tuberculosis–ingested phagosome with the acidic lysosome [17, 32]. We polarized human monocyte-derived macrophages (MDMs) with the supernatant from SEA-exposed and M. tuberculosis–stimulated autologous lymphocytes. Then, using high-resolution confocal microscopy, we evaluated phago-lysosome formation by quantifying the degree of colocalization between the lysosomal associated membrane protein, LAMP-1, and Td-tomato H37Rv M. tuberculosis bacilli (Figure 4A). Macrophages polarized with SEA-treated supernatants demonstrated 2.1-fold lower colocalization between M. tuberculosis and LAMP-1 as compared to non-SEA supernatants (P = .031, by the Mann–Whitney test; Figure 4B).

Figure 4.

Schistosome soluble egg antigen (SEA)–polarized lymphocytes decrease macrophage phago-lysosome maturation. Peripheral blood mononuclear cells from Mycobacterium tuberculosis–infected subjects were exposed to SEA for 14 days. After 14 days of SEA exposure, supernatant was used to polarize autologous human monocyte-derived macrophages (MDMs) for 48 hours. MDMs were then incubated with Td Tomato–labeled M. tuberculosis strain H37Rv in a biosafety level 3 facility for 4 hours, and the noninternalized M. tuberculosis were washed off before permeabilization, fixation, and staining with LAMP-1 (CD107a) and F-actin. Experiments performed in triplicate, with 86 cells measured. A, Representative 100× MDM images (LAMP-1 in green, H37Rv M. tuberculosis in red, and F-actin in blue). B, Median and interquartile range of Manders-2 colocalization of M. tuberculosis and LAMP-1 (P = .031, by the Mann–Whitney test).

Although SEA clearly induced less phago-lysosome formation than no SEA treatment, there was significant variability (Figure 4B), a likely product of our heterogeneous human population To investigate the consequences to macrophages by helminth-induced perturbation of CD4⁺ T-cell function in a more robust way, we polarized human MDMs with the cytokines most affected by SEA in our polyfunctional analyses, which also induce an alternative activated macrophage phenotype in mice [7]. Macrophages polarized with either IL-4 or IL-10 resulted in significantly decreased colocalization of M. tuberculosis with LAMP-1 when compared with nonpolarized or IFN-γ/lipopolysaccharide (LPS)–stimulated MDMs (P < .001; Figure 5). The MFI of LAMP-1 was not lower in both IL-4– and IL-10–polarized conditions, compared with that in nonpolarized cells (P = .748 and .348, respectively, by the Mann–Whitney test; data not shown).

Figure 5.

Decreased maturation of the phago-lysosome in interleukin 10 (IL-10)– and interleukin 4 (IL-4)–polarized macrophages. Human monocyte-derived macrophages (MDMs) were polarized with IL-4, IL-10, interferon γ (IFN-γ) and lipopolysaccharide (LPS), or standard medium for 48 hours. MDMs were then incubated with Td Tomato–labeled Mycobacterium tuberculosis strain H37Rv in a biosafety level 3 facility for 4 hours, and noninternalized M. tuberculosis was washed off before permeabilization, fixation, and staining with LAMP-1 (CD107a) and the nuclear stain DRAQ5. Experiments were performed in triplicate, with 40–60 cells measured. A, Representative 100× MDM images (LAMP-1 in green, H37Rv M. tuberculosis in red, and nuclei in blue). B, Median and interquartile range of Manders-2 colocalization of M. tuberculosis and LAMP-1 (P < .001, by the Mann–Whitney test).

The mannose receptor, CD206, is a prominent receptor with which macrophages bind and internalize M. tuberculosis [33, 34]. It has also been reported that CD206 expression increases in the presence of IL-4 [31]. Thus, we evaluated its expression after macrophage polarization, using imaging flow cytometry. We found that both IL-4– and IL-10–polarized MDMs increase their expression of CD206, compared with nonpolarized macrophages (P < .0001, by the Mann–Whitney test; Figure 6A and 6B). To determine whether an increase in CD206 intensity correlates with an increase in intracellular M. tuberculosis, macrophages were incubated with live H37Rv M. tuberculosis for 4 hours and then visualized by confocal microscopy. Macrophages polarized with IL-4 and IL-10 possessed a higher amount of internalized M. tuberculosis bacilli as compared to nonpolarized or IFN-γ– and LPS-polarized MDMs (P < .0005, by the Mann–Whitney test; Figure 6C and 6D).

Figure 6.

Increased Internalization of Mycobacterium tuberculosis in interleukin 4 (IL-4)– and interleukin 10 (IL-10)–polarized Macrophages. Human monocyte-derived macrophages (MDMs) were polarized with IL-4, IL-10, interferon γ (IFN-γ) and lipopolysaccharide (LPS), or standard medium for 48 hours. Experiments were performed in triplicate with at least 18 macrophages counted per experiment. A, Representative image flow cytometry of MDMs. B, Expression of mannose receptor (CD206) based on polarization (P < .0001; 4929 MDMs analyzed). For analysis of M. tuberculosis internalization, MDMs were polarized under the same conditions and incubated with M. tuberculosis strain H37Rv for 4 hours. Then, noninternalized M. tuberculosis were washed off before permeabilization, fixation, and staining with CD107a (LAMP-1) and the nuclear stain DRAQ5. C, Representative 20× automated Volocity counting of MDMs and M. tuberculosis. MDMs were then incubated with Td Tomato–labeled M. tuberculosis strain H37Rv in a biosafety level 3 facility for 4 hours. Then, noninternalized M. tuberculosis were washed off before permeabilization, fixation. and stained with LAMP-1 (CD107a) and the nuclear stain DRAQ5. Supplementary Figure 2 shows representative 20× confocal images. D, M. tuberculosis internalization in MDMs polarized with IL-10 and IL-4 (P < .0005, by the Mann–Whitney test).

Collectively, the data are evidence that the helminth-induced perturbation to the M. tuberculosis–specific CD4⁺ T-cell profile, which diminishes TNF-α and IFN-γ in favor of IL-4 and/or IL-10 production, leads to an impairment in macrophage acidification and, hence, to an inability to eliminate ingested bacilli [17, 32]. Thus, we provide a mechanism at the cell biology level to explain how helminth infection promotes M. tuberculosis to thrive.

DISCUSSION

Consistent with the epidemiologic evidence that helminth infection increases one's risk for tuberculosis progression [35], this study provides supportive mechanistic immune evidence. Helminth infection alters the quality of the M. tuberculosis–specific CD4⁺ T-cell response, which is associated with an impairment in the macrophage function that contains M. tuberculosis. These data are significant because they establish a human model for the cooperation of CD4⁺ T cells and macrophages in controlling M. tuberculosis and elaborate a mechanism at which helminth infection perturbs this control. This model is an instructive foundation upon which to elucidate the complex, coordinated lymphocyte response with downstream effects on macrophage cell biology.

We identify a mechanism by which the helminth-altered CD4⁺ T-cell response perturbs macrophage control of M. tuberculosis: the cytokines associated classically with helminth infection (IL-4 and IL-10) endorse M. tuberculosis internalization by macrophages but retard the development of the phago-lysosome, which is imperative for degradation of the pathogen. We designated the expression of the mannose receptor, CD206, as a correlate of M. tuberculosis uptake, since it is the primary macrophage receptor for the engulfment of M. tuberculosis [33, 34]. The increased level of M. tuberculosis internalization after IL-4 and IL-10 polarization could reflect increased phagocytosis and/or could signify a decreased rate of M. tuberculosis elimination. If an immature phagosome cannot fuse with a lysosome to achieve acidification, the internalized M. tuberculosis will not only survive but thrive [17, 18, 36]. By showing that, after SEA exposure, IL-4 and IL-10 polarization inhibit the development of the phago-lysosome, we create a direct link between helminth-induced perturbation of CD4⁺ T-cell function and M. tuberculosis survival. This lends mechanistic insight to published studies of models demonstrating that helminth infection increases the intracellular survival of M. tuberculosis in alternatively activated macrophages [20].

The simultaneous measurement of multiple T-cell functions (polyfunctionality) has transformed our understanding of protective immunity in general [23, 37]. With regard to tuberculosis immunity, multiparametric T-cell profiling has revealed that M. tuberculosis–specific CD4⁺ T cells capable of IL-2, TNF-α, and IFN-γ production simultaneously are associated with latent, subclinical infection, while active infection is characterized by a dominant TNF-α monofunctional response [16]. Here, we show that, after SEA exposure, there is a lower frequency of CD4⁺ T cells producing TNF-α and IFN-γ, with a concomitant increase in frequency of CD4⁺ T cells that produce IL-4, express PD-1, express apoptotic markers, and/or are nonfunctional. Previous studies have shown that blocking IFN-γ results in a relative increase in IL-4 production, a default pathway presumed to be active when the CD4⁺ T cells cannot access IFN-γ [38]. However, we observed a consistent subpopulation of M. tuberculosis–specific CD4⁺ T cells that produced IL-4 and IFN-γ simultaneously. This observation is consistent with literature reports of the existence of hybrid Th1-Th2 populations [39, 40]. Thus, we propose that M. tuberculosis–specific T-cell responses should be defined by their functional breadth, rather than by an imprecise Th1/Th2 dichotomous classification system.

The prevailing dogma in helminth immunology is that helminth infection stimulates robust IL-10 production and that this cytokine is instrumental in effectuating the profound immunomodulatory consequences of the infection. While SEA exposure did not increase the frequency of IL-10–producing M. tuberculosis–specific CD4⁺ T cells, there was a statistically significant increase in the quantity of IL-10, as measured by median fluorescence intensity. Although modest, this increase could have dramatic consequences for the immune system, as previous studies of helminth infection demonstrated that a picogram-level change in IL-10 levels induces perturbations to the immune system [41, 42]. Still, granulocytes, dendritic cells, monocytes, macrophages, natural killer cells, and innate lymphoid cells have all been shown to release IL-10 [43], which could account for the majority of IL-10 produced during helminth infection. The perturbation in macrophage function we show as a result of IL-10 polarization attests to this cytokines' potency and highlights the need to identify its source cell type(s).

The immune checkpoint inhibitors CTLA-4 and PD-1 are associated with chronic immune activation states, including helminth and M. tuberculosis infections [12, 44]. We show that SEA-exposed CD4⁺ T cells have increased PD-1 expression, which is associated with a diminished effector function profile and increased apoptosis. Interpreting these results requires caution, as murine [44, 45], zebra fish [46], and macaque [47] models all suggest the need for a balanced immune response. For example, while PD-1^−/− mice had increased TNF-α, interleukin 1, interleukin 6, interleukin 17, and IFN-γ, this was associated with higher lung bacterial counts, severe pulmonary necrosis, and increased mortality [44, 45]. These findings emphasize the complexity of tuberculosis immunology and the continued need for an improved understanding of the balance of proinflammatory and antiinflammatory cytokines that promotes disease control or progression [46, 47].

Helminth coinfection of M. tuberculosis–infected subjects represents a natural model to elucidate the immunologic mechanism of tuberculosis progression. We found that SEA exposure leads to an alteration of the functional response by M. tuberculosis–specific CD4⁺ T cells and that these changes to the cytokine milieu could impair the macrophage function that controls M. tuberculosis. Our experimental system recapitulates reasonably well the epidemiologic premise that helminths operate to fundamentally change the immunologic landscape before and during a response to M. tuberculosis. Our data argue for a longitudinal study to demonstrate the clinical significance between the immunologic response and clinical outcomes and specifically to determine whether the helminth-induced changes in T-lymphocyte and macrophage function predict acquisition of M. tuberculosis infection and/or progression to tuberculosis.

Supplementary Data

Supplementary materials are available at http://jid.oxfordjournals.org. Consisting of data provided by the author to benefit the reader, the posted materials are not copyedited and are the sole responsibility of the author, so questions or comments should be addressed to the author.