Blocking of CD1d Decreases Trypanosoma cruzi–Induced Activation of CD4⁻CD8⁻ T Cells and Modulates the Inflammatory Response in Patients With Chagas Heart Disease

Abstract

The control of inflammatory responses to prevent the deadly cardiac pathology in human Chagas disease is a desirable and currently unattained goal. Double-negative (DN) T cells are important sources of inflammatory and antiinflammatory cytokines in patients with Chagas heart disease and those with the indeterminate clinical form of Chagas disease, respectively. Given the importance of DN T cells in immunoregulatory processes and their potential as targets for controlling inflammation-induced pathology, we studied the involvement of CD1 molecules in the activation and functional profile of Trypanosoma cruzi–specific DN T cells. We observed that parasite stimulation significantly increased the expression of CD1a, CD1b, CD1c, and CD1d by CD14⁺ cells from patients with Chagas disease. Importantly, among the analyzed molecules, only CD1d expression showed an association with the activation of DN T cells, as well as with worse ventricular function in patients with Chagas disease. Blocking of CD1d-mediated antigen presentation led to a clear reduction of DN T-cell activation and a decrease in the expression of interferon γ (IFN-γ) by DN T cells. Thus, our results showed that antigen presentation via CD1d is associated with activation of DN T cells in Chagas disease and that CD1d blocking leads to downregulation of IFN-γ by DN T cells from patients with Chagas heart disease, which may be a potential target for preventing progression of inflammation-mediated dilated cardiomyopathy.

CD1d, double-negative T cells, cardiomyopathy, Chagas disease, pathology, Trypanosoma cruzi, immunoregulation, cytokines

T-cell activation is a critical process for establishing immune responses that can lead to the control of chronic infections caused by different pathogens and to the maintenance of homeostasis. However, if this efficient immune response is not further controlled, it may lead to tissue damage, mediating the onset of pathology, as often occurs in parasitic diseases. Human infection by the protozoan parasite Trypanosoma cruzi causes Chagas disease, which affects approximately 15 million people, leading to 12 000 deaths annually [1, 2]. Most individuals with chronic T. cruzi infection have the indeterminate form, characterized by a lack of clinical signs and symptoms of the disease, while approximately 30% present with severe cardiac or digestive disorders [3]. Dilated cardiomyopathy-induced heart failure is the main cause of death associated with Chagas disease [4]. The rate of progression from the indeterminate form to cardiomyopathy has been reported to be 1.85% per year [5]. While the factors that influence the clinical progression of Chagas disease have not been completely elucidated, there is consensus that activated T cells play an important role in this process [6, 7].

Most immunological studies concerning antigen presentation and T-cell activation in human Chagas disease have focused on the contribution of major histocompatibility (MHC) complex class I and II molecules in the presentation of peptide antigens [8, 9]. It has been shown that increased expression of MHC on human monocytes is accompanied by increased expression of costimulatory molecules and immunoregulatory cytokines, as well as by efficient T-cell activation [10]. In addition to the importance of MHC class I and II for T-cell activation during experimental infections with T. cruzi [11], it has also been shown that antigen presentation via the CD1 nonclassical MHC molecules is able to activate cells that are specific to parasite-derived glycoconjugates [12, 13], which predominantly decorate the surface of T. cruzi [14, 15], triggering an inflammatory response [16].

CD1 molecules belong to a nonpolymorphic family of glycoproteins resembling MHC class I, with hydrophobic binding sites that recognize and present lipid- and glycolipid-derived antigens to T cells [17, 18]. Human antigen-presenting cells express 5 isoforms of CD1: CD1a, CD1b, CD1c, CD1d, and CD1e, with the latter being an intracellular chaperone molecule [19]. Most glycolipid-specific CD1-restricted T cells display a CD4⁻CD8⁻ (double negative [DN]) phenotype [20–22]. In physiological conditions, DN T cells correspond to 1%–3% of total circulating lymphocytes and can be further divided in 2 subpopulations, according to the expression of αβ or γδ T-cell receptors (TCRs) [23].

Previous studies by our group have shown that αβ and γδ subpopulations of DN T cells are expanded during the chronic phase of human Chagas disease, suggesting a role for CD1-mediated antigen presentation. Although both αβ and γδ DN T-cell subpopulations were able to produce inflammatory and antiinflammatory cytokines, the γδ subpopulation was markedly biased toward the production of the antiinflammatory cytokine interleukin 10 (IL-10) in patients with the indeterminate form, as opposed to the production of inflammatory cytokines such as tumor necrosis factor α (TNF-α) and interferon γ (IFN-γ) in patients with Chagas heart disease [24]. Moreover, patients with the indeterminate clinical form maintained a regulatory ratio of IL-10–producing DN T cells, which was associated with better cardiac function, as compared to patients with Chagas heart disease.

To better understand the mechanisms underlying DN T-cell activation in human Chagas disease and its possible influence in the establishment of pathology, we evaluated (1) the expression of CD1-presenting molecules on CD14⁺ monocytes, (2) the association of such expression with the functional activation status of T. cruzi–specific DN T cells and with the heart function of patients with Chagas disease, and (3) the effect of CD1 blocking on the activation profile of T. cruzi–specific DN T-cell subpopulations from patients with Chagas heart disease.

Our data show that the blocking of CD1d shifts the cytokine expression toward a less inflammatory profile in patients with Chagas heart disease. Thus, the effects of blocking CD1d on the modulation of the immune response may open new possibilities for immune-based therapies and prophylactic strategies for prevention of Chagas cardiomyopathy.

PATIENTS, MATERIALS, AND METHODS

Patients

This study had a cross-sectional design involving patients with Chagas disease from areas of endemicity in Minas Gerais, Brazil, who are followed at the outpatient clinic of the Universidade Federal de Minas Gerais.

Twenty-two volunteers were enrolled. Fifteen volunteer patients with chronic Chagas disease who had positive T. cruzi–positive results of serologic tests and 7 negative for Chagas disease serological tests were included as a control group. Clinical status of patients was classified as previously defined [25]. Patients with Chagas disease were classified with either the indeterminate clinical form (characterized by positive results of serologic testing and a lack of clinical manifestations or alterations upon clinical, radiological and echocardiographic examination) or the cardiac clinical form (characterized by positive results of serologic testing and ≥1 of the following: right and/or left ventricular dilation, global left ventricular dysfunction, and/or alterations in electrocardiogram impulse generation/conduction, chest radiography findings, or echocardiography findings). Left ventricular diastolic diameter (LVDD) was used as a clinical parameter of ventricular function for patients with Chagas disease [25]. Table 1 summarizes the characteristics of the study group. The study was approved by the Ethical Committee of the Universidade Federal de Minas Gerais (COEP-UFMG–ETIC006/05) and accords with the Helsinki Declaration.

Table 1.

Healthy Individuals and Patients With Chagas Disease Analyzed in the Study

Patient ID	Clinical Form of Chagas Disease	Age, y	Sex	LVDD, mm
NI1	Noninfected	43	Female	ND
NI2	Noninfected	27	Female	ND
NI3	Noninfected	23	Female	ND
NI4	Noninfected	31	Female	ND
NI5	Noninfected	21	Male	ND
NI6	Noninfected	19	Male	ND
NI7	Noninfected	20	Male	ND
I1	Indeterminate	55	Female	55
I2	Indeterminate	38	Female	52
I3	Indeterminate	50	Female	50
I4	Indeterminate	34	Female	46
I5	Indeterminate	50	Female	ND
I6	Indeterminate	39	Female	46
I7	Indeterminate	38	Female	49
D1	Cardiac	50	Male	64
D2	Cardiac	63	Male	65
D3	Cardiac	44	Female	ND
D4	Cardiac	49	Male	ND
D5	Cardiac	55	Male	63
D6	Cardiac	60	Female	66
D7	Cardiac	38	Female	63
D8	Cardiac	59	Female	65

Abbreviations: ID, identifier; LVDD, left ventricular diastolic diameter; ND, not determined.

Parasites

Trypomastigotes (TRPs) of the Y strain of T. cruzi for infection studies were grown in Vero cells, as previously performed by us [7]. Briefly, cells were infected with 10 TRPs/cell, washed with culture medium, and maintained in Roswell Park Memorial Institute (RPMI) medium enriched with 5% fetal calf serum and antibiotic (penicillin 500 U/mL and streptomycin 0.5 mg/mL) for 5 days. After this period, TRPs ruptured the cells and were collected from the supernatant, centrifuged, and resuspended in RPMI medium. Contamination with amastigote forms was always <10%.

TRP Infection of Blood Cells From Patients With and Those Without Chagas Disease

Infection of peripheral blood cells was performed using a ratio of 10 TRPs/cell, as previously described by us [7]. In short, cells and parasites were incubated at 37°C in 5% CO₂ for 3 hours, followed by washing and centrifugation with phosphate-buffered saline (PBS) for removal of free TRPs. After centrifugation, supernatant was removed, and RPMI medium supplemented with antibiotic/antimicotic (amphotericin B 0.25 μg/mL, penicillin 200 U/mL, and streptomycin 0.1 mg/mL) and L-glutamine (1 mM) at an equal volume to the amount of blood initially incubated was added to the tubes. Infected cells were incubated at 37°C in 5% CO₂ for 14 hours, followed by brefeldin A (which retains intracellular proteins; 1 μg/mL) for an additional 4 hours, after which further analyses were performed (described below). Reagents were from Sigma (St. Louis, Missouri). For all individuals, blood specimens were submitted to the same procedures described above but, in the absence of parasites, were used as control (ie, medium-only) cultures.

Immunophenotyping of DN T Cells and Monocytes

Infected or noninfected blood cells were harvested after the final 18 hours of culture and submitted to specific staining. Combinations of fluorochrome-labeled monoclonal antibodies (mAbs) specific for human leukocyte cell-surface markers were used, including anti-CD14-PE-Cy5.5, to detect monocytes; anti-CD4-BV510 or PE-Cy5.5, anti-CD8-PECy5.5 or APCCy7, anti-TCRαβ-FITC or BV421, and anti-TCRγδ-FITC or APC, to identify the specific subpopulations of DN T cells; anti-CD1a-FITC, anti-CD1b-FITC, anti-CD1c-FITC, or anti-CD1d-FITC; and anti-HLA-PE, to assess the frequency of antigen-presenting and activation molecules in CD14⁺ cells. Cells were harvested and plated at 200 000 cells/well with a 40-µL mixture of surface antibodies for 15 minutes at 4°C. Samples were then washed 3 times in PBS containing 1% bovine serum albumin and fixed for 20 minutes with 2% formaldehyde/PBS. After centrifugation and washing with PBS, the cells were permeabilized by incubation for 15 minutes with a 0.5% saponin solution, washed, and subjected to intracellular staining. Samples were incubated with anti-IL-10-PE, anti-IFN-γ-PE, or anti-CD69-PECy7 monoclonal antibodies for 20 minutes at room temperature, washed twice with 0.5% saponin solution, resuspended in PBS, and read in a flow cytometer. A minimum of 40 000 gated events from each sample were collected and analyzed using FlowJo software (TreeStar, Ashland, Oregon). All antibodies were from BioLegend (San Diego, California).

CD1d Blocking Assay

Blood cells from patients with Chagas heart disease were infected with TRPs. Cells were washed by centrifugation with PBS for removal of free TRPs, and anti-CD1d specific blocking antibody (BioLegend) was added at a concentration of 25 µg/mL. Noninfected controls, as well as infected cultures in the absence of blocking antibody or in the presence of an isotype control, were performed. Cells, under the different conditions, were incubated at 37°C in 5% CO₂ for 14 hours in RPMI-supplemented medium (as described above). At the end of this period, immunophenotyping was performed using the same markers and protocol described above.

Statistical Analysis

The paired t test and the Wilcoxon matched pairs test were used to ascertain differences among data with and without a Gaussian distribution, respectively. Correlation analyses were done using the Pearson correlation coefficient. Differences that returned P values of ≤.05 were considered statistically significant from one another.

RESULTS

In Vitro Infection With TRPs Promotes Increased Expression of CD1 Antigen-Presenting Molecules by Monocytes From Patients With Chagas Disease

To determine the effects of T. cruzi infection on the expression of antigen-presenting molecules, we evaluated CD1 levels, as well as the expression of the MHC class II molecule, HLA-DR, by CD14⁺ monocytes from healthy donors and patients with Chagas disease following culture in medium alone or after infection with TRPs. Patients with Chagas disease showed a significant increase in the mean fluorescence intensity for all CD1 isoforms (CD1a, CD1b, CD1c, and CD1d) but not in HLA-DR after in vitro infection with live TRP, as compared to nonstimulated cultures (Table 2). This increase in CD1 molecules was also seen when patients were stratified on the basis of indeterminate or cardiac clinical forms (Table 2). No significant difference in the expression of these markers was seen when cells from individuals without Chagas disease were infected in vitro with TRPs, as compared to control cultures, except for CD1a expression (Table 2).

Table 2.

Expression of CD1 Isoforms and HLA-DR Molecules on CD14⁺ Cells From Patients With Chagas Disease and Noninfected Controls

Group, Culture Condition	Intensity of Expression in CD14+ Cells, MFI, Median (IQR)
	CD1a	CD1b	CD1c	CD1d	HLA-DR
Noninfected controls
Medium only	25.5a (21–40)	44.2 (38–147)	24.7 (19–33)	27.6 (25–31)	77.3 (51–99)
TRP	61.3 (39–127)	159.5 (108–280)	47.4 (45–71)	85.3 (57–105)	135.9 (120–166)
Patients with Chagas disease, by clinical form
Any
Medium only	21.8a (19–31)	54.2a (30–67)	20.3a (18–30)	37.9a (27–43)	126.6 (92–155)
TRP	56.5 (45–78)	113 (92–232)	54.3 (45–74)	76.4 (46–93)	108.1 (89–138)
Cardiac
Medium only	20.8a (20–22)	43.2a (31–56)	28.7a (19–39)	28.7a (22–35)	151.0 (94–207)
TRP	64.8 (56–73)	198.4 (113–284)	122.6 (69–177)	90.4 (79–102)	125.6 (108–143)
Indeterminate
Medium only	23.3a (18–35)	54.2a (29–77)	20.3a (17–25)	38.4a (28–48)	126.6 (90–154)
TRP	53.4 (45–79)	107.6 (80–179)	48.4 (44–59)	62.1 (41–90)	106.5 (82–133)

Abbreviations: IQR, interquartile range; MFI, mean fluorescence intensity; TRP, Trypanosoma cruzi trypomastigote.

^a Significantly different from cultures with TRP.

CD1d Expression Is Associated With a Proinflammatory Profile in DN T Cells From Patients With Chagas Disease, as Well as With Worse Cardiac Function

To determine whether there is an association between CD1 expression and cytokine production by DN T cells, we performed statistical correlation analysis between the expression of CD1 by CD14⁺ monocytes and expression of IFN-γ and IL-10 by DN αβ and γδ T cells from patients with Chagas disease in stimulated and nonstimulated cultures. A significant positive correlation between the intensity of expression of CD1d by CD14⁺ monocytes and the expression of IFN-γ by DN αβ and γδ T cells was observed in TRP-stimulated cultures from patients with Chagas disease (Figure 1A). Interestingly, a negative correlation between the intensity of expression of CD1d by CD14⁺ cells and the expression of IL-10 by DN TCR γδ cells after stimulation was also observed (Figure 1B). No significant correlations were seen in unstimulated cultures (Figures 1A and 1B). No significant correlation between the expression of CD1a, CD1b, or CD1c by CD14⁺ monocytes and expression of cytokines by DN T cells was found.

Figure 1.

Correlative analysis of the intensity of expression of CD1d by CD14⁺ cells and the frequency of double-negative (DN) T cells expressing interferon γ (IFN-γ) or interleukin 10 (IL-10). Cells were incubated in medium alone (MED) or in the presence of the trypomastigote form of Trypanosoma cruzi (TRP) and stained for CD14 and CD1d (x-axis) or for αβ and γδ DN T cells (as indicated) expressing IFN-γ (A) or IL-10 (B; y-axis), as described in Patients, Materials, and Methods. r² and P values are indicated in each graph. Abbreviation: MFI, mean fluorescence intensity.

To determine whether the expression of CD1d was correlated with better or worse cardiac function in patients with Chagas disease, we performed correlation analysis between CD1d and LVDD, in which higher values reflect worse cardiac function. We observed that the higher the intensity of expression of TRP-induced CD1d expression, the larger the LVDD of the patients with Chagas disease (Figure 2D), indicating worse cardiac function. No correlation was observed between the expression of CD1a, CD1b, or CD1c and LVDD (Figure 2A–2C, respectively).

Figure 2.

Correlative analysis of the intensity of expression of CD1a, CD1b, CD1c, and CD1d by CD14⁺ cells and left ventricular diastolic diameter (LVDD). Cells were incubated in the presence of trypomastigote forms of Trypanosoma cruzi and stained for CD14 and CD1a, CD1b, CD1c, and CD1d (x-axis), as described in Materials and Methods. Correlative analyses were performed with LVDD (y-axis), as determined by clinical examinations described in the Patients subsection of Patients, Materials, and Methods. r² and P values are indicated in each graph. Abbreviation: MFI, mean fluorescence intensity.

CD1d Blocking Reduces Activation of DN γδ Cells

To verify whether the antigen presentation of T. cruzi via CD1d was associated with activation of DN T cells, we performed a blocking assay by using anti-CD1d monoclonal antibody and evaluated its effects on the activation status of DN T cells via the expression of CD69. Figure 3A shows the analysis strategy, using as an example the γδ DN T-cell subpopulation (a similar strategy was used for αβ analysis).

Figure 3.

Analysis of the effects of CD1d blocking on the activation of double-negative (DN) T cells. Cells from 6 patients with Chagas heart disease were incubated with medium in the presence (BLK) or absence (MED) of anti-CD1d blocking antibodies or with the trypomastigote form of Trypanosoma cruzi in the presence (TRP + BLK) or absence (TRP) of anti-CD1d blocking antibodies. The frequency of expression of the activation marker CD69 was evaluated in the different T-cell subpopulations, as described in Patients, Materials, and Methods. A, Representative fluorescence-activated cell-sorting plots for the gating strategy of the γδ DN T cells (top) and for the analysis of expression of CD69 within the gated cells under each condition (bottom). A similar strategy was used for αβ DN T-cell analysis. B and C, Frequency of αβ and γδ DN T cells expressing the activation marker CD69 under the different conditions. D and E, Frequency of CD4⁺ and CD8⁺ T cells expressing the activation marker CD69 under the different conditions. Bars indicate averages, and lines in each bar indicate standard deviation. Statistically significant result is indicated in the graph. Abbreviations: FSC, forward scatter; SSC, side scatter; TCR, T-cell receptor.

Our analysis showed that, although CD1d blocking reduced CD69 expression in both TRP-activated αβ and γδ DN T-cell subpopulations (Figure 3B and 3C), this decrease was only statistically significant within the DN γδ T cells (Figure 3C). The effect of CD1d blocking on T-cell activation was specific for the DN subpopulation, as expression of CD69 by CD4⁺ and CD8⁺ T cells did not decrease upon blocking of CD1d (Figure 3D and 3E). These data reinforce that the presentation of T. cruzi antigens via CD1d are directly associated with the activation of DN T cells, especially γδ DN T cells in Chagas disease.

CD1d Blocking Reduces Expression of IFN-γ by DN γδ Cells

To evaluate whether the decrease in activation observed in DN T cells after CD1d blocking also affected cytokine production, we evaluated the expression of IFN-γ and IL-10 by αβ and γδ DN T cells. We observed that, although CD1d blocking did not significantly change the expression of IFN-γ by αβ DN T cells (Figure 4A; P = .06), it led to a significant decrease in the expression of IFN-γ by γδ DN T cells (Figure 4B). Expression of IL-10 was not altered by CD1d blocking in either subpopulations of DN T cells (Table 3). We also observed a slight decrease in the IFN-γ to IL-10 ratio (P = .09), particularly in γδ DN T cells, after CD1 blocking (Table 3). Together, these results suggest that CD1d blocking alters the cytokine profile of DN T cells toward a less inflammatory profile.

Table 3.

Effect of the Blocking of CD1d on Monocytes From Patients With Chagas Heart Disease in the Expression of the Immunoregulatory Cytokines Interferon γ (IFN-γ) and Interleukin 10 (IL-10), as Well as the Ratio of Inflammatory (IFN-γ) to Noninflammatory (IL-10) Cytokines

Culture Condition	DN TCR αβ Cells, Median (IQR)			DN TCR γδ Cells, Median (IQR)
	IFN-γ, %	IL-10, %	Ratio	IFN-γ, %	IL-10, %	Ratio
Medium only	1.2 (0.6–1.9)	2.8 (1.4–3.3)	0.7 (0.2–1.2)	1.3a (1.3–1.4)	2.1 (0.5–5.7)	0.6 (0.6–2.0)
Medium + BLK	0.9 (0.0–3.0)	3.5 (0.7–5.4)	0.4 (0.2–0.8)	0.2 (0.0–0.9)	5.3 (2.9–7.9)	0.08 (0.0–0.2)
TRP	6.1 (3.2–7.2)	2.6 (1.4–3.1)	1.5 (0.2–2.0)	3.4b (2.9–6.4)	2.1 (1.7–2.6)	1.1b (0.6–1.4)
TRP + BLK	2.7 (2.6–3.6)	1.8 (1.3–4.2)	1.0 (0.7–2.5)	2.2 (1.6–3.0)	2.1 (1.4–3.9)	0.5 (0.2–0.6)

Abbreviations: BLK, anti-CD1d blocking antibody; DN, double negative; IQR, interquartile range; TCR, T-cell receptor; TRP, trypomastigote.

^a Significantly different from cultures with medium plus BLK.

^b Significantly different from cultures with TRP plus BLK.

Figure 4.

Analysis of the effects of CD1d blocking on the expression of interferon γ (IFN-γ) by double-negative (DN) T cells. Cells from 6 patients with Chagas heart disease were incubated with medium in the presence (BLK) or absence (MED) of anti-CD1d blocking antibodies or with the trypomastigote form of Trypanosoma cruzi in the presence (TRP + BLK) or absence (TRP) of anti-CD1d blocking antibodies. The frequency of expression of IFN-γ was evaluated in the different T-cell subpopulations, as described in Patients, Materials, and Methods. A and B, Frequency of αβ and γδ DN T cells expressing IFN-γ under the different conditions. Bars indicate averages, and lines in each bar indicate standard deviations. Statistically significant results are indicated in the graphs. Abbreviation: TCR, T-cell receptor.

DISCUSSION

Previous studies by our group have shown that different T-cell populations, especially DN T cells, were expanded and displayed an exuberant cytokine production in patients with Chagas disease, significantly contributing to the inflammatory and antiinflammatory cytokine microenvironment observed in patients with the cardiac and indeterminate clinical forms, respectively [24, 26]. This is of particular interest, given that DN T cells recognize lipid and/or glycolipid antigens [17, 18], which predominate on the surface of T. cruzi [14, 15]. Moreover, previous studies have shown that DN T cells are expanded and activated in the peripheral blood of patients with autoimmune diseases such as lupus [27], as well as in another trypanosomatid infection, leishmaniasis [28]. In this study, we assessed the expression of antigen-presenting molecules by CD14⁺ monocytes from patients with Chagas disease to better understand the mechanisms underlying DN T-cell activation and function. Here we demonstrated that T. cruzi stimulation induces an increase in the expression of CD1 isoforms by CD14⁺ monocytes from patients with Chagas disease and that the increase of CD1d is associated with the expression of an inflammatory cytokine by DN T cells and with worse cardiac function. Importantly, blocking CD1d decreased DN T-cell activation and IFN-γ production in patients with Chagas heart disease. These results show that the inflammatory profile associated with Chagas heart disease can be modulated by CD1d blocking, opening possibilities for immune-based therapies to control Chagas heart pathology.

We observed that the expression of CD1a, CD1b, CD1c, and CD1d molecules was significantly increased in CD14⁺ monocytes from patients with Chagas disease after in vitro infection with T. cruzi. We have previously shown that in vitro infection with T. cruzi induces TNF-α expression by monocytes from patients with Chagas disease [7, 10]. Given that TNF-α has been shown to induce expression of CD1 isoforms [29], it is possible that this cytokine plays a role in increasing the expression of CD1 in CD14⁺ monocytes from patients. CD1 molecules are particularly important for the presentation of lipid and glycolipid antigens [19, 20], which are abundant on the surface of TRPs [14, 15] and induce macrophage activation and inflammatory responses in mice [16]. Thus, it is possible that antigenic components derived from these glycosylated molecules lead to the upregulation of CD1 molecules via induction of proinflammatory cytokines. In turn, the upregulation of CD1 would improve the presentation of lipid/glycolipid antigens, potentiating the immune response.

Interestingly, among the 4 isoforms of CD1 evaluated, only the CD1d isoform was associated with increased DN T-cell function, as evidenced by positive correlation with IFN-γ expression. Different populations of DN T cells can be subdivided according to the expression of the αβ or γδ TCR, antigen recognition, and repertoire. The αβ DN T-cell subpopulation includes cells that display a polyclonal repertoire with MHC class restriction [21, 30, 31], as well as natural killer T (NKT)–cell type I and type II, which have an oligoclonal [32] or polyclonal [33] repertoire, respectively. NKT cells recognize glycolipid antigens [34, 35] restricted to CD1d presentation [36]. Several studies have shown that glycolipid antigens derived from pathogens, including T. cruzi, as well as self-antigens, presented via CD1d, are able to mediate the activation of NKT cells [12, 37]. Thus, it is possible that the association between CD1d expression and IFN-γ production by αβ DN T cells reflects the activation of NKT cells, which are within the total αβ DN T-cell population studied here.

γδ DN T cells can be activated by self-ligands [38], as well as by common microbial products [39]. Their repertoire in vivo is dominated by specific subsets that recognize a limited number of antigens, suggesting a somewhat oligoclonal repertoire [40]. The frequency of CD1d expression also correlated positively with the expression of IFN-γ by the γδ DN subpopulation from patients with Chagas disease but negatively with the expression of IL-10 by the same cells, suggesting the induction of a strong proinflammatory profile. γδ DN T cells can recognize antigens presented via CD1 molecules, particularly CD1c [41] and CD1d [42], which is consistent with our findings.

Given that we observed a proinflammatory response associated with CD1d expression, we evaluated the association of CD1d expression and heart function, as measured by LVDD [43]. Our analysis showed a positive correlation between CD1d expression and higher LVDD (an indicator of worse cardiac function), which points to the possible clinical relevance of CD1d in Chagas heart disease pathology.

Once we established associations between increased CD1d and the inflammatory profile of DN T cells, as well as with worse heart function, we performed CD1d blocking studies, using cells from patients with the severe cardiac form of Chagas disease to (1) verify the relationship between antigen presentation via CD1d and activation of DN T cells and (2) check whether CD1 blocking could control the inflammatory response of DN T cells, with the aim of evaluating them as possible targets for prevention or control of inflammation-induced Chagas heart disease.

The CD1d molecule may present endogenous or exogenous glycolipids, within which the long-chain fatty acid molecule α-galactosylceramide is the most studied [44]. Other antigens presented by CD1d have been reported, such as lipophosphoglycan from Leishmania donovani [45] and glycophosphoinositol from T. cruzi [13]. Our results showed that blocking of CD1d decreased the activation of DN T cells, especially in the γδ subpopulation, as measured by the expression of CD69. This shows that T. cruzi–derived antigens presented via CD1d are important in the activation of DN T cells, particularly for the subpopulation mostly related to the inflammatory profile. It is noteworthy to mention that, while CD1d presents antigens to NKT cells contained within the αβ DN T cells, it can also present antigens to γδ DN T cells [34–36]. At least 2 characteristics of γδ DN T cells can explain the maintenance of activation even after blocking CD1d. First, γδ DN T cells can directly recognize soluble proteins and nonpeptide antigens independent of presentation, and second, they can also recognize antigens via MHC class I and II [40]. As expected, we did not observe any effects of CD1d blocking on the activation of CD8⁺ or CD4⁺ T cells, given that these cells recognize MHC class I– and class II–presented antigens, respectively.

When calculating the ratio of IFN-γ to IL-10, as a measure of the inflammatory versus antiinflammatory cytokine environment, we observed that CD1d blocking induced a slightly less inflammatory profile, likely a consequence of the decrease in IFN-γ production after CD1d blocking. These data suggest that blocking of CD1d could be a strategy to control inflammation and prevent or decrease the severity of Chagas heart disease. Treatment strategies blocking CD1d have been proven efficient in the control of cytokine release in a primate asthma model [46], as well as in reducing septic mortality [47] and eradicating established tumors [48] in mice. Although other cells that are not activated via CD1d also contribute to the inflammatory profile in Chagas disease, it is important to mention that DN T cells are critical sources of these cytokines [24]. Moreover, blocking of a particular molecule and not obliterating the expression of inflammatory cytokines but, rather, decreasing its expression would be a better strategy of disease control, especially considering the role of inflammatory cytokines in keeping the parasite levels in check. These data may have implications in guiding new clinically relevant immune-based strategies to control cardiac damage in patients with Chagas disease. Given the activation of DN T cells by lipids and carbohydrates and the importance of these molecules in the immune response and parasite functions [49, 50], identifying the molecules responsible for their activation could also present new disease protection strategies to target.

Notes

Acknowledgments. W. O. D., K. J. G., L. R. d. V. A., and M. C. P. N. are Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq) fellows. L. M. D. M. and L. S. A. P. are Coordenação de Aperfeiçoamento de Pessoal de Nível Superior fellows.

Financial support. This work was supported by Instituto Nacional de Ciência e Tecnologia Doenças Tropicais, Fundação de Amparo `a Pesquisa do Estado de MInas Gerais, and CNPq.

Potential conflicts of interest. K. J. G. is the founder of BRISA. All other authors report no potential conflicts. All authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors consider relevant to the content of the manuscript have been disclosed.