Antileishmanial Drugs Modulate IL-12 Expression and Inflammasome Activation in Primary Human Cells

The leishmaniases are neglected tropical diseases caused by protozoan parasites. There are 20 Leishmania species pathogenic for humans, responsible for three main types of disease classified by clinical symptoms: cutaneous leishmaniasis (CL), mucocutaneous leishmaniasis, and visceral leishmaniasis (VL) (1). Globally, this disease group mainly affects the poorest countries, and according to the World Health Organization, >1 billion people are at risk for infection (1). Among the leishmaniases, VL is the most dangerous if untreated. In the order of 300,000 cases of VL are reported each year with >20,000 deaths (1). Leishmania exhibit a pronounced tropism for macrophages and neutrophils (2). Treatment of leishmaniasis relies on specific antileishmanial drugs, such as amphotericin B (AmB), miltefosine (hexadecylphosphocholine [HePC]), and pentamidine, among others (3). These drugs were introduced as second-line treatment for VL to replace antimonials, which are very harmful for patients. However, they also have demonstrated serious side effects in patients (4–6). Undesirable effects of HePC include gastrointestinal side effects, hepatotoxicity, and nephrotoxicity (7), whereas pentamidine may cause hypotension, insulin-dependent diabetes mellitus, and cardiotoxicity (4, 8). Although pentamidine is no longer used in VL because of its toxicity, it is still used for CL as a systemic treatment (9) and as secondary prophylaxis in HIV-infected individuals (9–11). AmB therapy is known to induce acute toxicity with infusion-related reactions and nephrotoxicity (6, 12). It is interesting to note that treatments are not completely effective in immunosuppressed patients, supporting the importance of having a fully competent immune system during therapy (13–15). It has previously been reported that HePC as well pentamidine reduce cell proliferation in mice (16–18). Nevertheless, potential side effects remain, so far, poorly studied in primary human cells in relation to cell death and immune cell function.

AmB, which is an antifungal drug, binds to ergosterol in the cell membrane of the parasite, inducing pore formation and consequently cell death (19, 20). HePC, an anticancer drug, is the only oral treatment available for VL. It interferes with lipid metabolism in the protozoan (21, 22) and affects the mitochondria, leading to apoptosis-like death (23–26). Pentamidine, which is a cationic aromatic diamine, induces disintegration of the protozoan kinetoplast and also acts on mitochondria (27–29). Therefore, given the action of these antileishmanial drugs in inducing mitochondria damage and cell death, it cannot be excluded that such therapies would also impact the host’s immune cells, interfering with cell survival and immune activation. Thus, the impact of such antileishmanial drugs on primary human cells deserves further detailed analysis.

Whereas monocytes constitute a considerable systemic reservoir of myeloid precursors, they are more prone to undergoing apoptosis than differentiated macrophages (30–34) while being more efficient than macrophages in IL-1β secretion (35). Caspases, which are the main effectors of programmed cell death, have been demonstrated to be essential for the processing of inflammatory cytokines leading to the release of IL-1β and IL-18 (36, 37). It is generally accepted that the inflammasome is dependent of caspase-1, the adaptor protein PYCARD often mentioned as apoptosis-associated speck-like protein containing a C-terminal caspase recruitment domain (ASC) (38–40), and members of NLRP family (41, 42). But, a noncanonical inflammasome pathway has also been described in which caspase-11 and its human orthologs, caspase-4 and caspase-5 in the cytoplasm, induced inflammasome activation. Darisipudi et al. (43) reported that in the presence of AmB, LPS-primed murine macrophages induce the release of IL-1β through an NLRP3–ASC–caspase-1–dependent pathway. However, there is a fundamental difference between human and rodent cells in the release of IL-1. Unlike murine cells, human blood monocytes do not require a second signal for LPS-mediated IL-1β release (44–47). Whereas in mice, this process may be followed by rapid lytic cell death known as pyroptosis, which contributes to the release of inflammatory cytokines (39, 48), and the release of IL-1β can proceed independently of pyroptosis and K⁺ efflux in primary human monocytes (49). Therefore, whether antileishmanial drugs interfere with host cell survival and induce an inflammatory process of human cells is so far unknown.

In the current study, we assessed the impact of HePC, pentamidine, and AmB on human PBMC. We observed that only the highest concentrations (>20 μM) of antileishmanial drugs induce cell death (phosphatidylserine [PS] exposure) and mitochondrial membrane depolarization. Nevertheless, at lower concentrations, we found that antileishmanial drugs inhibited CD4 and CD8 T cell proliferation and the expression of Th1-associated cytokines (IL-12 and TNF-α) after stimulation of primary human monocytes with LPS. In contrast to HePC and pentamidine, only AmB induced the release of IL-1β and IL-18 as well as the expression of mRNA encoding for the inflammasome machinery. Our results provide important observations that may be of interest in the treatment of VL, where it would seem that a competent immune system is essential for these drugs to exert full control over the microbial infection.

PBMC were isolated from peripheral blood of healthy donors (R75069, Etablissement Français du Sang). All donors provided informed consent to Etablissement Français du Sang.

Cells were isolated by density gradient centrifugation using Ficoll-Hypaque (GE Healthcare) and then incubated with antileishmanial drugs, including AmB, HePC, and pentamidine purchased from Sigma-Aldrich. Staurosporine (Sigma-Aldrich) was used as a positive control for inducing cell death. Monocyte (CD14⁺) cells were isolated using CD14 microbeads (Miltenyi Biotec) according to manufacturer’s instructions. Cells were cultured in RPMI 1640 supplemented with 10% FCS (PAA Laboratories), penicillin/streptomycin (50 U/ml; Life Technologies), glutamine (2 mM; Life Technologies), sodium pyruvate (1 mM; Life Technologies), and HEPES buffer (10 mM; Life Technologies).

PBMC and CD14⁺ cells (6 × 10⁵ cells per well) were incubated in the absence or presence of antileishmanial drugs at 37°C and 5% CO₂. Cell death was assessed by measuring PS exposure using labeled annexin V (Beckman Coulter Coultronics) as previously described (50). Briefly, cells were stained with anti-CD3-PE-CF594 (clone SP34-2; BD Biosciences), anti-HLA-DR-allophycocyanin (clone G46-6; BD Biosciences), and anti-CD20-PE-Cy7 (clone 2H7; BD Pharmingen) for 30 min at 4°C. Cells were then washed and labeled with annexin V for 15 min in the dark at 37°C. Cellular membrane integrity was assessed using propidium iodide (PI) (Molecular Probes) and mitochondrial membrane potential (Δψ_m) using 3,3′-dihexyloxacarbocyanine iodide (DiOC₆; Molecular Probes). Samples were evaluated by flow cytometry (Cytomics FC500; Beckman Coulter) and analyzed using FlowJo software (Tree Star).

Cell proliferation was determined using CFSE probe (Thermo Fisher Scientific) as described by Quah et al. (51). Briefly, PBMC were suspended in PBS with 10 μM CFSE. After incubation (15 min), cells were washed and stimulated with anti-CD3 mAb (200 pg/ml, clone OKT3; BioLegend) and incubated for 4 d in the absence or presence of antileishmanial drugs. As indicated, cells were also treated with recombinant human IL-12 (20 ng/ml; R&D Systems) or with 10 ng/ml anti–IL-10 (Clone JES3-19F1; BioLegend). At day 4, cells were stained with anti-CD4-allophycocyanin (clone L200; BD Biosciences) and anti-CD8-PE-Cy7 (clone RPA-T8; BD Biosciences). Analyses were performed on Cytomics FC 500 (Beckman Coulter) and further analyzed using FlowJo software (Tree Star). CFSE profiles were determined as described by Roederer et al. (52), namely the fraction diluted (percentage of CFSE^low), which represents the fraction of cells in the final culture that divided at least once.

Cells were stimulated with 1 μg/ml LPS (InvivoGen). After 4 h of stimulation, brefeldin A (BD GolgiPlug; BD Biosciences) was added overnight to the culture to assess the expression of TNF-α and IL-12 by flow cytometry using anti-TNF-PE (clone Mab11; BD Biosciences), anti–IL-12–allophycocyanin (clone C8.6; Miltenyi Biotec), anti-CD14-FITC (clone TUK4; Miltenyi Biotec), and anti-HLA-DR-ECD (clone immu 357; Beckman Coulter). Cells were analyzed on a Cytomics FC 500 (Beckman Coulter).

From the supernatants of activated human CD14⁺ cells (2 × 10⁶), we used ELISA to quantify the amount of IL-1β (human IL-1 β/IL-1F2; R&D Systems), IL-18 (human total IL-18; R&D Systems), IL-10 (human IL-10 ELISA Kit; Invitrogen), and IL-6 (LEGENDplex; BioLegend). Assays were performed according to manufacturers’ instructions. Cells were also treated with Q-VD-OPH (20 μM; SM Biochemicals), a pan-caspase inhibitor (50, 53), or with MCC950 (1 μM; Merck), an NLRP3 inhibitor.

Human CD14⁺ cells were seeded at 2 × 10⁶ cells per well and incubated with 10 ng/ml GM-CSF (R&D Systems). Cells were treated in the absence or presence of either AmB (Sigma-Aldrich), HePC, or pentamidine for 24 and 48 h at 37°C. Total RNAs were extracted using TRIzol reagent (Thermo Fischer Scientific). RT-PCR was performed with 150 ng of RNA by SensiFAST cDNA synthesis kit (Bioline), and gene expression was assessed by quantitative PCR (qPCR) using SensiFAST SYBR Hi-ROX kit (Bioline). Assays were performed in 10-μl reactions with 10 ng of cDNA. Thermocycling settings consisted of one hold of 15 min at 95°C followed by a two-step temperature (95°C for 15 s and 60°C for 30 s) over 40 cycles in CFX384 Touch Real-Time PCR Detection System (Bio-Rad). Human-specific primers are described in Table I. The delta threshold cycle (Δct) values for each tested gene were obtained by calculating the difference between the ct value for the gene of interest and the geometric mean of the ct values of two housekeeping genes (GAPDH and RPS18).

Human CD14⁺ cells were stimulated with LPS (1 μg/ml) or AmB (5 μM) for 24 h and lysed using 1% NP40 buffer (150 mM NaCl, 50 mM Tris-HCl [pH 7.4], 10 mM EDTA) supplemented with a mixture of antiproteases and antiphosphatases. Cell extracts were resolved by SDS-NuPAGE 4–12% Bis-Tris gel (Novex, Les Ulis, France) and transferred to nitrocellulose membrane (Amersham Biosciences, Les Ulis, France). Nonspecific sites were blocked by incubation with 5% of nonfat milk or BSA for 1 h at room temperature. The Abs used were rabbit polyclonal anti-caspase-1 (p20 form; Sigma-Aldrich), mouse monoclonal anti-caspase-1 (Clone 661228; R&D Systems), rabbit monoclonal anti–IL-1β (clone D3U3E; Cell Signaling), rabbit monoclonal anti-caspase-5 (clone D3G4W; Cell Signaling), mouse monoclonal anti-NLRP3 (clone Cryo-2; Adipogen Life Sciences), and monoclonal anti-caspase-4 (clone 4B9; MBL Life Science). Equal protein loading was assessed using mouse monoclonal anti-actin (clone C4; Merck Millipore). Membranes were treated with HRP-linked goat anti-rabbit or anti-mouse secondary Abs (Amersham Biosciences). Immunoreactive proteins were detected by chemiluminescence, and densitometric analysis was performed using ImageJ software.

Statistics were performed with the GraphPad Prism 5 software. Data are presented as mean ± SEM. A one-way ANOVA followed by a Dunnett test was used for comparison between untreated and treated cells, which incorporates the dependencies between these groups. Statistical differences were also assessed using the Mann–Whitney U test (*p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001).

PS exposure on the cell surface is an early event associated with programmed cell death, which can be detected using fluorescent-labeled annexin V. After treatment of PBMC with HePC and pentamidine for 24 h, PS exposure was determined by flow cytometry. The percentages of PBMC stained with annexin V in the presence of 50 μM either HePC or pentamidine compared with medium alone are shown in Fig. 1A. Our results demonstrated that PS exposure on the cell surface is drastically increased at a nonphysiological dose of antileishmanial drugs (200 μM, Fig. 1B). In the presence of HePC, 55% ± 12.6 of the PBMCs displayed annexin V staining, but only 20% ± 8.9 displayed annexin V staining in the presence of pentamidine (Fig. 1B). The percentage of annexin V–positive cells in the presence of staurosporine (at the dose of 10 μM), used as a positive control, reached nearly 90%. To analyze in more detail the nature of the cells more prone to die, cells were stained with anti-CD3 and anti-CD20 Abs (Fig. 1C, 1D). Interestingly, B cells (CD20⁺ cells) were more sensitive to death compared with T cells. Thus, HePC (Fig. 1C) and pentamidine (Fig. 1D) induced PS exposure on B cells in a dose-dependent manner, reaching a dose of 50 μM, 44.1% ± 12.5 and 41.1% ± 11.0 of annexin-positive cells, respectively (Fig. 1C, 1D). Our results demonstrated that only higher doses (>20 μM) of antileishmanial drugs induce the death of human lymphoid T cells.

We then assessed whether antileishmanial drugs interfere with the capacity of T cells to proliferate. T cell proliferation was measured using the CFSE probe, which is a fluorescein derivative molecule diluted according to cell division (51, 54). In Fig. 2A and 2B, we show the typical profile of CFSE dilution (CFSE^low) in CD4 and CD8 T cell populations (gating on live cells) at day 4 in the absence or presence of anti-CD3 mAb. In the presence of 20 μM HePC, CD4 and CD8 T cell divisions decreased (CD4, 25.7% CFSE^low and CD8, 18.7% CFSE^low) and were completely blunted in the presence of pentamidine (Fig. 2B). By using a lower dose of pentamidine (5 μM), a significant reduction of both CD4 and CD8 T cell proliferation is observed (Fig. 2C). Thus, cell proliferation (CFSE^low) reached 53.1% ± 13.4 and 59.7% ± 12.3 for CD4 and CD8 T cells, respectively, compared with CD3-stimulated untreated cells (78.0% ± 11.0 and 77.7% ± 11.9, respectively) (Fig. 2C). Furthermore, at the same dose of 5 μM, HePC had only a slight impact on T cell proliferation, mostly on CD8 T cells (Fig. 2C). Although pentamidine has an impact on T proliferation, this effect is cell death independent as shown in Fig. 2D and consistent with Fig. 1. Therefore, we also decided to analyze T cell proliferation in the presence of AmB. For CD4 and CD8 T cell proliferation, in the presence of 5 μM, AmB reached 41.4% ± 5.6 and 45.8% ± 17.4 CFSE^low, respectively, compared with CD3-stimulated T cells (69.0% ± 7.5 and 61.9% ± 20.4, respectively, Fig. 2E). At the highest dose of AmB tested (20 μM), CD4 and CD8 T cell proliferation was almost blunted (12.7% ± 5.2 CSFE^low for CD4 and 10.8% ± 10.7 CSFE^low for CD8, Fig. 2E). This reduction of cell proliferation was not related to increase in cell death because even if the difference is statistically significant, we observed that cell death did not exceed 20% for CD4 or 10% for CD8 T cells (Fig. 2F). Taken together, our results showed that antileishmanial drugs interfere with CD4 and CD8 T cell proliferation.

Innate cells, in particular monocytes, are essential for correct T cell activation. Therefore, we analyzed the impact of HePC and pentamidine on CD14⁺ cells. First, we assessed the effect of PS exposure on CD14⁺ cells after overnight culture in the absence or presence of antileishmanial drugs. At a dose of 20 μM, only pentamidine induced death (PS⁺) of monocytes (49.3% ± 23.1, Fig. 3A, 3B), whereas the effect of HePC at these concentrations was negligible. We then extended the analyses to study ΔΨm and cellular membrane integrity using the fluorescent probes DiOC₆ (3) and PI, respectively. After 48 h of incubation, pentamidine at the dose of 20 μM induced ΔΨm loss (DiOC₆^low) in 47.5% ± 12.5 of CD14⁺ cells (Fig. 3C, 3D). HePC, at the dose of 20 μM, slightly affected mitochondria potential (29.3% ± 10.5) compared with untreated cells (19.2% ± 6.0; Fig. 3C, 3D). In contrast to PS exposure, few necrotic and/or late apoptotic cells are observed by flow cytometry using PI staining (Fig. 3C, 3E). Thus, despite the observation that 47.5% ± 12.5 of CD14⁺ cells display ΔΨm loss in the presence of 20 μM pentamidine, only 15.5% ± 12.2 showed a loss in cellular membrane integrity (PI⁺), and this difference was not statistically significant compared with untreated cells (Fig. 3E). HePC did not induce cellular membrane permeabilization (Fig. 3E). In contrast, AmB treatment at the dose of 20 μM increased monocyte cell death as indicated by annexin V staining concomitantly with mitochondrial membrane depolarization (DiOC₆^low) and cell membrane permeabilization (PI⁺), although to a lower extent compared with ΔΨm loss (Fig. 3F). At the dose of 5 μM, we did not observe any impact on monocyte survival (Fig. 3F). Thus, our results indicated that pentamidine and AmB are reducing monocyte survival and ΔΨm only at the highest concentration used (20 μM).

The production of proinflammatory cytokines expressed by monocytes is a key event in regulating T cell proliferation such as IL-12 (55) and is previously shown to be essential in the context of Leishmania infection (56, 57). We assessed the capacity of monocytes to express IL-12 and TNF-α by flow cytometry after LPS stimulation, a TLR-4 agonist, in the absence or presence of antileishmanial drugs. At a dose of 20 μM, pentamidine completely blunted the expression of both cytokines induced by LPS stimulation (TNF-α: 2.9% ± 2.3 and IL-12: 4.6% ± 2.1) by comparison with LPS-stimulated untreated cells (TNF-α: 32.9% ± 23.0 and IL-12: 21.5% ± 9.4; Fig. 4A–D). At the dose of 5 μM, pentamidine reduced LPS-mediated TNF-α and IL-12 expression (16.3% ± 10.8 and 11.4% ± 5.0, respectively; Fig. 4C, 4D). Interestingly, no difference in the secretion of IL-6 production was observed at this dose of 5 μM, suggesting that the alteration is specific to IL-12 and TNF-α (Fig. 4E). This change is also not related to an increase in cell death as indicated by PS exposure compared with LPS stimulation alone (Fig. 4F, 4G). Furthermore, our results indicated that the incubation of monocytes in the presence of HePC, at the dose of 20 μM, also decreased the expression of both TNF-α and IL-12 (TNF-α: 21.7% ± 14.4 and IL-12: 16.7% ± 9.9; Fig. 4A–D), which could be related to cell death (PS⁺, Fig. 4F, 4G). This is of interest because HePC alone has no effect on monocyte survival but seems to act in synergy with LPS to induce cell death. At the lower dose (5 μM), the expression of IL-12 is not decreased (Fig. 4B, 4D), and HePC has no impact on the levels of IL-6 secretion (Fig. 4E). Our results indicated that AmB also decreases IL-12 production but only at the highest dose of 20 μM (Fig. 4H); a dose that also induces monocyte cell death (Fig. 3F). These results demonstrated that antileishmanial drugs impair the capacity of monocytes to express IL-12 and TNF-α when treated with LPS.

We then assessed whether in vitro addition of IL-12 restores T cell proliferation. In the presence of IL-12 and AmB, CD8 T cell proliferation was mostly recovered (67.4 ± 24.3) compared with CD4 T cell proliferation (34.9 ± 29.2) (Fig. 5A, 5B). In contrast, IL-12 has a minor impact on the restoration of T cell proliferation in the presence of HePC (CD4, 12.1 ± 1.2 versus CD8, 21.8 ± 2.1) (Fig. 5A, 5B), and IL-12 is unable to restore pentamidine-mediated immunosuppression (data not shown). It has previously been shown that IL-12 antagonizes T cell immunosuppression mediated by IL-10 as well in regulating monocyte response (31, 58–60). We found that unlike HePC and pentamidine, only AmB induced IL-10 secretion from monocytes, although the level of IL-10 secretion was donor dependent (Fig. 5C). The addition of Ab against IL-10 restored T cell proliferation suppressed by AmB in a range similar to IL-12 (Fig. 5D). Altogether, our results indicate that the balance of IL-10/IL-12 regulates AmB-mediated T cell immunosuppression, whereas it has a minor impact on HePC and pentamidine-mediated immunosuppression.

In addition to IL-12/IL-10, IL-1β that acts as a soluble mediator of inflammation provides costimulatory signals to T cells (35, 61, 62). Circulating human monocytes produce IL-1β (35) after the cleavage of its proform by the IL-1β–converting enzyme (ICE), better known as caspase-1. A multiprotein complex, termed inflammasome, is formed after activation (38, 63) involving caspase-1, members of the NLR family, and the protein PYCARD (64, 65), leading to the release of IL-1β. Human monocytes were incubated in the presence of antileishmanial drugs, and mRNA expressions were quantified by qPCR (Table I). As expected, LPS induced inflammatory gene expression (IL-1β, IL-1α, and IL-6 mRNAs), and of the drugs tested, only AmB induced a similar profile in primary human monocytes (Fig. 6A). In human cells, as well as caspase-1, two additional caspases may contribute in the release of IL-1β, caspase-4 and caspase-5, which are the equivalent of caspase-11 in mice (41, 66–68). Our results indicate that LPS activation induced the expression of caspase-4 and caspase-5, with a transient increase in caspase-1 and caspase-4 observed in the presence of AmB (Fig. 6B). In contrast, we observed lower levels of IL-1β, IL-1α, and IL-6 mRNAs in the presence of pentamidine, whereas neither HePC nor pentamidine induced caspase mRNA expression, with pentamidine tending to decrease the expression of caspase-1 and caspase-4 mRNA at 48 h (Fig. 6A, 6B). Extending the analysis to NLRP3 and PYCARD, our data indicated that none of the antileishmanial drugs induced NLRP3 mRNA expression compared with LPS stimulation (Fig. 6C). Finally, we observed that PYCARD is transiently downregulated in human monocytes treated with either LPS or AmB, whereas pentamidine downregulated PYCARD at 48 h (Fig. 6C).

To monitor inflammasome activation, we quantified the release of IL-1β and IL-18, also known as IFN-γ–inducing factor (IGIF), a substrate for caspase-1 (42, 69). Whereas it was expected that LPS would induce the release of IL-1β from primary human monocytes, we found that AmB but not the other antileishmanial drugs also induced IL-1β secretion (Fig. 7A). Although the levels of IL-18 secretion were donor dependent, it is clear that among the drugs tested, only AmB induced IL-18 secretion (Fig. 7B). The amount of IL-1β and caspases was assessed by Western blot. In addition to the increase in IL-1β detected (Fig. 7C, 7D), we also observed increased levels of caspase-1 (p45) and caspase-5 (p50) proforms in both LPS- and AmB-treated monocytes compared with untreated cells (Fig. 7C, 7D). There was also a slight increase in caspase-4 compared with untreated cells but to a lower extent because of its constitutive expression (Fig. 7C, 7D). In nonstimulated human monocytes, the majority of caspase-1 was in the 20-kDa form that may explain why after LPS stimulation, IL-1β can be rapidly released. The amount of caspase-1 in its p20 form was lower in LPS- and AmB-treated monocytes, consistent with IL-1β release (Fig. 7C, 7D). Furthermore, we observed that the active form of caspase-5 (p35) is detected only in the presence of LPS with only a slight band detected in the presence of AmB (Fig. 7C, 7D). Our data indicated increased amounts of NLRP3 in LPS- as well as in AmB-treated monocytes. We used a pan-caspase inhibitor, Q-VD-OPh, to block caspase activation. We demonstrated that in the presence of Q-VD, IL-1β and IL-18 secretions from monocytes treated with LPS and AmB are decreased, whereas IL-6 secretion, as expected, is independent (Fig. 7E, 7F). Furthermore, we used a synthetic compound, MCC950, that inhibits NLRP3 and consequently inflammasome activation. We observed that MCC950 completely abolishes the secretion of IL-1β and IL-18 (Fig. 7E). Interestingly, AmB at the highest dose (20 μM), which was demonstrated to induce monocyte cell death, induced less production of IL-1β but higher levels of IL-18, which is generally associated with tissue injury (70) (Fig. 7E). Altogether, our results indicate that AmB is the compound among the antileishmanial drugs tested that leads to inflammasome activation in primary human monocytes.

This study shows that antileishmanial drugs have the capacity to interfere with CD4 and CD8 T cell proliferation in peripheral blood human cells, although this is not directly related to the death of T cells. We found that the expression of IL-12 induced by LPS stimulation of human monocytes was inhibited and that only the defected T cell proliferation mediated by AmB could be restored by the addition of IL-12 or the neutralization of IL-10, suggesting different mechanisms of action among drugs. Furthermore, our results highlight that AmB induces inflammasome activation, leading to the secretion of IL-1β and IL-18. These results demonstrate that a clear impact of antileishmanial drugs on primary human immune cells may be of importance in the context of leishmania therapy.

Consistent with previous reports in mice (16–18), we showed that the antileishmanial drugs, HePC, pentamidine, and AmB, interfere with CD4 and CD8 T cell proliferation in human blood cells. However, cell death was observed only with HePC and pentamidine at the highest concentrations (>20 μM) used, and B cells (CD20⁺) were particularly sensitive. The drug concentrations used in this study are lower than the plasma doses achieved in vivo. Indeed, it has previously been shown that the plasma concentration of HePC varies between 75 and 150 μM at a dose of 2.5 mg/kg/d (71, 72). Several studies performed in patients treated with oral HePC for CL have reported that this drug accumulates in PBMC, with an intracellular concentration 2- to 3-fold higher than that present in the plasma (72, 73). In rodents, HePC and pentamidine accumulate principally in the kidneys, lung, and liver but also in the spleen (74–76), which is the main organ for T and B cell regulation. Interestingly, an earlier report showed that pentamidine at a concentration of 10 μM reduces splenic mouse B cell proliferation (18). Thus, one side effect of these antileishmanial drugs in humans might be premature death of B cells compared with T cells, not only in the blood but also in the spleen.

Our results also highlight that antileishmanial drugs have an effect on blood monocyte survival. Thus, at a dose of 20 μM, we observed that AmB induces both PS exposure associated with ΔΨm loss and cellular membrane permeabilization, and pentamidine induces PS exposure and ΔΨm loss only, whereas HePC has only a minor impact on ΔΨm and PS exposure. However, we found that all three drugs tested inhibit the expression of IL-12 induced by LPS stimulation of blood monocytes. Mitochondrial metabolism is a critical component of macrophage polarization as well of survival (77). Thus, although mitochondria depolarization in monocytes (CD14⁺) may contribute in part to the loss of IL-12 expression and cell death (77), the inhibition of IL-12 expression occurred at a lower dose in the absence of ΔΨm loss, supporting the hypothesis that additional mechanisms may be involved.

IL-12 is a key cytokine in the genesis of Th1 cell–mediated immunity and for Leishmania infection via the production of IFN-γ for killing intracellular parasites (57, 78). It is well known that IL-12 is also essential in sustaining T cell proliferation (79–82). This may have a more global impact because IL-12 is also involved in enhancing cytotoxicity of NK cells (82). Thus, a major impact of these different antileishmanial drugs could be related to their capacity to inhibit IL-12 expression. IL-10 is also a potent immunosuppressive cytokine, impairing myeloid functions and IL-12 secretion (83). However, we observed that only AmB induces IL-10 secretion from blood monocytes. We showed that in vitro addition of IL-12, as well as anti–IL-10 Abs, restored almost completely the proliferation capacity of CD8 T cells treated with AmB. This effect of IL-12 on CD8 T cell proliferation is consistent with its original description as a cytotoxic lymphocyte maturation factor (CLMF) (84–86). Severe VL is characterized by a reduction of CD8 and Th1 cells (87–89), and treatment with AmB, although active against the parasites, might reduce the capacity of the host to develop an immune response efficient enough to clear the parasite completely during therapy, particularly in immunosuppressed patients. On the contrary, the addition of IL-12 had no impact on pentamidine-mediated immunosuppression, supporting the hypothesis of alternative mechanisms of action.

Pentamidine has been reported to reduce TNF-α induced by LPS or by bacteria (90, 91) as well as IL-1β, IL-1α, and IL-6 expression (91–93), which may provide additive signals for T cell proliferation (35, 61, 62). HePC, at a dose of 50–100 mg/kg, was reported to attenuate allergic sensitization in a model of OVA-induced delayed-type hypersensitivity in mice (17) and to reduce intestinal inflammation mediated by T cell transfer in mice (16). However, other studies performed in rodents or in cell lines such as THP-1 have reported that HePC increases pro-Th1 cytokines (IL-12 and IFN-γ) (94, 95) and transcripts of TLR-4 and TLR-9 (95). However, growing evidence indicates that mouse macrophages or cell lines such as THP-1 differ in their susceptibility to express proinflammatory cytokines compared with primary human monocytes (44, 45, 96). Thus, unlike human blood monocytes, murine and THP-1 cells require not only LPS but also a second signal for the release of IL-1β (44–47). Furthermore, human monocytes are more prone to undergo apoptosis than differentiated macrophages (30–34). Thus, a fundamental difference occurs in the subsequent IL-1/IL-18 release between human and mouse cells. Our results highlight that only AmB, not pentamidine or HePC, triggers the inflammasome machinery and the secretion of both IL-1β and IL-18 in the absence of a second signal and the level of IL-18 secretion is increased with a dose of AmB that induces monocyte death.

Production of mature IL-1β and IL-18 requires their proteolytic cleavage by inflammatory caspases (caspase-1, -4, or -5). Caspase-4/5 have been implicated in a noncanonical inflammasome pathway promoting IL-1β secretion by human monocytes (47, 97). By using a caspase inhibitor, Q-VD-OPh, we reduced the secretion of IL-1β and IL-18. Although AmB did not enhance the expression of the mRNA encoding for the sensor NLRP3, the use of MCC950, an inhibitor of NLRP3, abolished IL-1β and IL-18 production completely, indicating the inflammasome may be implicated in the presence of AmB. Darisipudi et al. (43) have reported that in LPS-primed murine macrophages, the release of IL-1β in the presence of AmB is dependent on NLRP3–ASC–caspase-1 proteins. In this study, we demonstrate that AmB is able to induce IL-1β and also IL-18 secretion by human blood monocytes. It has been assumed that AmB mediates an indirect immunomodulatory effect against microbes through IL-1/NO production (98). The observation that AmB induces IL-18 secretion could be of interest because IL-18 is a potent inducer of IFN-γ (69, 70) and may contribute to the control of Leishmania infection (99). Alternatively, another report indicated that in susceptible BALB/c mice, NLRP3 inflammasome has a more detrimental effect, favoring Leishmania major infection (100). Our results also indicate that despite the idea that AmB may activate myeloid cells through TLR-2/4 signaling pathways (101), the observation that IL-18 is induced only by AmB, and not by LPS, supports an alternative pathway of TLR-2/4 signaling in primary human cells consistent with a role of the noncanonical inflammasome activation. Taken together, these data reveal that antileishmanial drugs may have distinct impact on inflammation in primary human cells.

Immunosuppression is a hallmark of protozoan infection (2, 102) and, in particular, a pronounced effect on Th1 cell–mediated immunity. Particularly, individuals coinfected by HIV in whom cell-mediated immunity is more defective have a higher risk of parasitemia relapse. Our findings indicate a profound impact of antileishmanial drugs on the capacity of myeloid cells to secrete IL-12, support T cell proliferation, and induce inflammasome activation. Therefore, development of novel drug therapies for protozoan infections require prior testing on primary human cells to establish any potential side effects on immune cell function.

The authors have no financial conflicts of interest.