ABSTRACT
Visceral leishmaniasis (VL) is a chronic parasitic disease caused by Leishmania infantum in the Americas. During VL, several proinflammatory cytokines are produced in spleen, liver, and bone marrow. However, the role of interleukin-32 (IL-32) has not been explored in this disease. IL-32 can induce production of proinflammatory cytokines in innate immune cells and polarize the adaptive immune response. Herein, we discovered that L. infantum antigens induced expression of mRNA mainly for the IL-32γ isoform but also induced low levels of the IL-32β transcript in human peripheral blood mononuclear cells. Furthermore, infection of human IL-32γ transgenic mice (IL-32γTg mice) with L. infantum promastigote forms increased IL-32γ expression in the spleen and liver. Interestingly, IL-32γTg mice harbored less parasitism in the spleen and liver than wild-type (WT) mice. In addition, IL-32γTg mice showed increased granuloma formation in the liver compared to WT mice. The protection against VL was associated with increased production of nitric oxide (NO), interferon gamma (IFN-γ), IL-17A, and tumor necrosis factor alpha by splenic cells restimulated ex vivo with L. infantum antigens. In parallel, there was an increase in the number of Th1 and Th17 T cells in the spleens of IL-32γTg mice infected with L. infantum. IL-32γ induction of IFN-γ and IL-17A expression was found to be essential for NO production by splenic cells of infected animals. These data indicate that IL-32γ potentiates the Th1/Th17 immune response during experimental VL, thus contributing to the control of L. infantum infection.
INTRODUCTION
Visceral leishmaniasis (VL) is a chronic, debilitating disease caused by Leishmania spp. transmitted by sand flies. It is estimated that there are 400,000 new cases of VL annually, with about 30,000 deaths (1). VL is caused by Leishmania donovani in Asia, India, and East Africa and by Leishmania infantum in the Americas and in the Middle East, Central Asia, and China (2, 3). Brazil, together with India, Nepal, Bangladesh, Sudan, and Ethiopia, accounts for 90% of VL cases worldwide (2). The transmission of L. infantum is zoonotic, in which humans are accidental hosts and dogs are the main peridomestic reservoirs (1, 4). The clinical manifestations of VL range from an asymptomatic self-resolving phenotype to a progressively severe disorder that is characterized by fever, cachexia, weight loss, anemia, leukopenia, thrombocytopenia, hypergammaglobulinemia, and hepatosplenomegaly and that may be fatal if left untreated (5, 6).
In general, protection against VL is dependent on Th1 lymphocyte responses and interferon gamma (IFN-γ) production (7–9), which activates macrophages to produce leishmanicidal molecules, such as nitric oxide (NO) (10). In addition to Th1 cells, Th17 lymphocytes produce interleukin-17A (IL-17A) and are also likely to be protective in VL. IL-17A acts synergistically with IFN-γ to increase NO production by murine macrophages infected with L. infantum (11–13). On the other hand, Th2 and T regulatory (Treg) cell cytokines, especially IL-10, are associated with susceptibility and progression of VL (14–17).
Interleukin-32 (IL-32) is an intracellular cytokine produced by immune and nonimmune cells, which induces the production of proinflammatory cytokines and chemokines, such as tumor necrosis factor alpha (TNF-α), IL-6, IL-8, and IL-1β, via activation of p38 mitogen-activated protein kinase (MAPK) and NF-κB (18–20). IL-32 can polarize the acquired immune response to the Th1 and Th17 profiles by enhancing antigen presentation and IL-12 and IL-6 induction (21). There are nine known IL-32 isoforms generated by the alternative splicing of IL-32γ, which is the most biologically active isoform (19, 22). IL-32 has been associated with the immunopathology and/or protection of numerous infectious diseases, such as those caused by Mycobacterium tuberculosis, M. avium, human immunodeficiency virus (HIV), influenza A virus, and hepatitis B virus (23–28).
Although rodents do not present any gene homologous to IL-32, murine cells are able to respond to this cytokine (18, 29). On the basis of this, transgenic mice expressing the human IL-32γ gene (IL-32γTg mice) were generated and have been successfully used in studies on the role of IL-32 in in vivo infections (30–32). We recently showed that IL-32 expression increases in IL-32γTg mice or humans infected with parasites that cause cutaneous leishmaniasis (32–34). Importantly, IL-32 was associated with the production of proinflammatory cytokines and leishmanicidal molecules, such as NO and antimicrobial peptides, during the infection of human macrophages with Leishmania amazonensis or Leishmania braziliensis (34). In addition, IL-32 contributed to the control of L. amazonensis dissemination from cutaneous lesions to the spleens in the IL-32γTg mice (32). Due to the aforementioned properties of IL-32 and the fact that IL-32 is highly expressed in the livers and spleens of IL-32γTg mice (35), we investigated the effect of IL-32 overexpression in a murine model of VL induced by infection with L. infantum.
RESULTS
Leishmania infantum induces IL-32γ expression in human PBMCs.We have demonstrated that L. amazonensis and L. braziliensis, which cause cutaneous leishmaniasis, induce expression of IL-32γ in human peripheral blood mononuclear cells (PBMCs) (32–34). To assess whether L. infantum also induces IL-32 expression, human PBMCs were incubated with L. infantum antigens. IL-32 protein levels were significantly increased in cell lysates of PBMCs stimulated with L. infantum for 24 h and 48 h compared to unstimulated cells (P < 0.05) (Fig. 1A). The IL-32γ mRNA appeared 4 h after the stimulus and was maintained at high levels for up to 48 h (Fig. 1B). Interestingly, IL-32β mRNA expression was detected only after 24 h of stimulation of the PBMCs with L. infantum (Fig. 1C). The IL-32α isoform was not detected at any of the aforementioned incubation times. Comparatively, L. infantum was able to induce higher levels of IL-32γ than IL-32β (Fig. 1D). The level of IL-32γ mRNA induced by L. infantum was comparable to the level of this isoform induced by lipopolysaccharide (LPS) and higher than that induced by L. braziliensis (Fig. S1).
L. infantum induces expression of IL-32γ in human PBMCs. PBMCs (5 × 106/ml) from healthy donors were cultured with total antigen (Ag) of L. infantum (50 μg/ml) for 4 h, 24 h, or 48 h. RPMI 1640 complete medium was used for unstimulated cells. (A) IL-32 production was assessed by ELISA of the cell lysates. (B) Expression of IL-32γ mRNA by qPCR. (C) Expression of the IL-32β isoform by qPCR. (D) Comparison of the relative expression of the γ and β isoforms of IL-32. Data represent the median with interquartile range (n = 6 donors). *, P < 0.05 compared to the control at each time by the Mann-Whitney U test (A to C) or P < 0.05 (IL-32γ versus IL-32β) by the Mann-Whitney U test (D).
IL-32γ controls systemic infection caused by L. infantum.After demonstrating that L. infantum can upregulate both mRNA and protein expression of IL-32, we examined the role of this cytokine in a murine model of VL. To investigate whether IL-32γ could modulate VL-associated immunopathogenesis, wild-type (WT) and IL-32γTg mice were infected with stationary-phase L. infantum promastigotes. It was observed that L. infantum infection increased IL-32 production in both the spleen and liver of IL-32γTg mice (Fig. 2A). As was observed in human PBMCs, L. infantum predominantly induced the IL-32γ isoform in the spleen and liver of IL-32γTg mice (Fig. 2B). The peak of IL-32γ mRNA expression and IL-32 production was on day 45 postinfection (p.i.) in the spleen and on day 30 p.i. in the liver (Fig. 2A and B). IL-32β mRNA was expressed only on the 45th day p.i. in both organs (Fig. 2B).
IL-32γ improves parasite control in mice infected with L. infantum. WT and IL-32γTg mice were infected i.p. with 1 × 107 L. infantum promastigotes. After 15, 30, 45, and 60 days p.i., the mice were euthanized. (A) The spleens and livers of uninfected (0 h) and infected WT and IL-32γTg mice were macerated, and total IL-32 production was assessed by ELISA. (B) Expression of IL-32γ and IL-32β mRNA by qPCR. (C) Evaluation of parasite burden by limiting dilution. (D and E) Organ weight relative to animal body weight (in percent). The data show the mean ± SEM for 9 animals per group from 3 independent experiments. *, P < 0.05 compared to uninfected mice (0 day), by one-way ANOVA with Bonferroni's posttest (A and B) or P < 0.05 by Student's t test (C to E); #, P < 0.05 compared to uninfected mice (0 day) in the same group by one-way ANOVA with Bonferroni's posttest.
IL-32γTg mice had fewer parasites in both the spleen and liver than WT mice over the course of the 60-day infection period (Fig. 2C; see also Fig. S2 in the supplemental material). Despite the lower tissue parasitism, IL-32γTg mice exhibited an earlier and greater increase in the weights of the spleen and liver than WT mice (15 and 30 days after infection). However, during the late phase of infection (45 and 60 days p.i.), there were no significant differences in the relative weights of the spleen and liver between IL-32γTg and WT mice (Fig. 2D and E). These data suggest that endogenous IL-32γ is important for the control of tissue parasitism during experimental VL.
IL-32γ increases granuloma formation in the liver during murine experimental visceral leishmaniasis.VL is characterized by granulomatous lesions in the viscera, located mainly in the spleen and liver, caused by a chronic inflammatory process induced by the parasites. Therefore, since we showed that IL-32γ decreases the parasite load in these organs, we investigated the association of IL-32 expression with histopathological alterations triggered by the host immune response to the parasite. Infection with L. infantum led to the loss of the lymphoid architecture of the spleen with a predominance of red pulp cells, including an increase in the number of megakaryocytes. These findings were similar in the spleens of the L. infantum-infected WT or IL-32γTg mice (Fig. S3). However, we observed a considerable increase in the number of liver granulomas in the IL-32γTg mice compared to the WT mice (Fig. 3A and B). In addition, we observed that the hepatic granulomas of L. infantum-infected IL-32γTg mice occupied larger areas in the liver than the hepatic granulomas of WT mice (Fig. 3C). We also observed the presence of necrotic areas in the infected IL-32γTg mouse liver larger than those seen in the infected WT mouse liver (Fig. 3A). No differences in the histopathological aspects of liver and spleen were found between uninfected IL-32γTg and WT mice (Fig. 3A and S3). Thus, histopathological analyses indicated that IL-32γ promoted the formation of larger and more necrotic hepatic granulomas during murine L. infantum infection.
IL-32γ increases hepatic granuloma formation during experimental infection with L. infantum. WT and IL-32γTg mice were infected i.p. with 1 × 107 L. infantum promastigotes for 30 days, and fragments of spleen and liver were prepared for histopathological analysis (hematoxylin-eosin staining) under a light microscope. (A) Representative photomicrographs of the liver of uninfected or infected WT and IL-32γTg mice. Magnifications, ×100 (left and middle) and ×400 (right). Arrows, granulomas; asterisks, necrotic areas. (B) Number of hepatic granulomas from 30-day-infected WT and IL-32γTg mice per 10 fields. (C) Total area of hepatic granulomas in 10 fields. In panels B and C, the data represent the median and individual values for 8 animals from 3 independent experiments. *, P < 0.05 by the Mann-Whitney U test.
IL-32γ induces proinflammatory cytokines during experimental visceral leishmaniasis.After 30 days of in vivo infection with L. infantum, the splenic cells of the IL-32γTg mice produced significantly higher levels of IFN-γ, IL-17A, TNF-α, and IL-10 following ex vivo stimulation with total L. infantum antigen. In WT splenic cell cultures, only IFN-γ was induced by lysates of L. infantum, compared to nonstimulated WT spleen cells (Fig. 4A to D). Exposure of the splenic cell culture to concanavalin A (ConA) modestly induced IFN-γ production in cells from both infected WT and IL-32γTg mice as well as IL-10 in cultures of cells from WT mice (Fig. 4A to D). In addition, IFN-γ, IL-17A, TNF-α, and IL-10 were found in higher levels in both splenic and liver lysates of infected IL-32γTg mice than in those of infected WT mice (Fig. 4E to H). In IL-32γTg mice, a positive correlation between the tissue levels of IL-32 and TNF-α or IL-10 levels was found in the spleen (Fig. S4). All evaluated cytokines were detected at higher levels in the liver than in the spleen (normalized for tissue weight) from IL-32γTg mice. In addition, the levels of IL-32 in the liver significantly correlated with the IFN-γ, IL-17A, or IL-10 levels after infection with L. infantum (Fig. S4B). These data indicate that IL-32γ favors a mixed Th1 and Th17 profile during experimental VL in mice.
IL-32γ induces proinflammatory cytokines during murine infection with L. infantum. WT and IL-32γTg mice were infected i.p. with 1 × 107 L. infantum promastigotes for 30 days. The spleens were removed, and splenocytes (5 × 106 cells/ml) were stimulated with L. infantum antigen (50 μg/ml) or ConA (5 μg/ml) for 24 h for TNF-α or 72 h for the other cytokines. The levels of the IFN-γ (A), IL-17A (B), TNF-α (C), and IL-10 (D) cytokines were measured by ELISA. The spleens and livers from 30-day-infected WT and IL-32γTg mice were macerated, and the lysates were used to assess cytokine production by ELISA: IFN-γ (E), IL-17A (F), TNF-α (G), and IL-10 (H). The data show the mean ± SEM for 9 animals per group from 3 independent experiments. P values were determined by Student's t test. #, P < 0.05 compared with medium in the same group; *, P < 0.05 for the indicated comparisons.
Th1 and Th17 lymphocytes are increased in IL-32γ transgenic mice infected with L. infantum.Since the presence of IL-32γ induced both Th1- and Th17-associated cytokines during experimental VL, we investigated whether IL-32γ increased both Th1 and Th17 T cell numbers after L. infantum infection. IL-32γTg mice infected with L. infantum for 30 days possessed a higher percentage of T helper lymphocytes (CD3+ CD4+) in the spleens than infected WT mice (Fig. 5A), whereas there was no difference in the percentage of cytotoxic T cells (CD3+ CD8+) between the groups (Fig. 5B). Although there was the production of large amounts of proinflammatory cytokines in the livers of IL-32γTg mice, we did not observe differences between the percentage of CD4+ or CD8+ T cells present in the livers of WT and IL-32γTg mice (Fig. 5A and B). The antigen-specific proliferation of Th lymphocytes was assessed by the carboxyfluorescein succinimidyl ester (CFSE) dilution method (Fig. S5). CD4+ T cells from the spleens of infected IL-32γTg mice displayed higher antigen-specific proliferation than those from infected WT mice (Fig. 5C), whereas a similar level of proliferation was obtained after polyclonal activation of T lymphocytes from IL-32γTg and WT mice with ConA (Fig. 5C). In addition, 30-day-infected IL-32γTg mice exhibited a higher percentage of IFN-γ-producing CD4+ T cells (Fig. 5D) and IL-17A-producing CD4+ T cells (Fig. 5E) in the spleens than infected WT mice. Together, these data indicate that IL-32γ upregulates Th1- and Th17-associated immune responses during experimental VL.
IL-32γ increases the generation of Th1 and Th17 cells during mouse infection with L. infantum. WT and IL-32γTg mice were infected i.p. with 1 × 107 L. infantum promastigotes for 30 days. Spleen and liver cells were labeled with anti-CD3-FITC, anti-CD4-PE, and anti-CD8-PerCP-Cy5.5 and analyzed by flow cytometry. (A and B) The percentage of T CD3+ CD4+ (A) and T CD3+ CD8+ (B) lymphocytes in spleen and liver from 30-day-infected WT and IL-32γTg mice. (C) Splenic cells were labeled with 5 μM CFSE and stimulated with total L. infantum antigen (50 μg/ml) or ConA (5 μg/ml) for 4 days. The cells were labeled with anti-CD4-FITC for flow cytometric analysis of proliferation of CD4+ T cells. (D and E) Splenic cells (5 × 106 cells/ml) from infected WT and IL-32γTg mice were stimulated with 50 μg/ml of total L. infantum antigen for 72 h. The cells were treated with brefeldin A (10 μg/ml) for an additional 6 h, and the cells were stained with anti-CD4-PE. Intracellular cytokines were assessed with anti-IFN-γ–Alexa Fluor 488 (D) or anti-IL-17A–APC (E). The data show representative contour plots of IFN-γ-positive (IFN-γ+) CD4+ (D) or IL-17A-positive (IL-17A+) CD4+ (E) T cells in the left and middle and the mean ± SEM for 6 animals per group from 2 independent experiments on the right. *, P < 0.05 by Student's t test.
Increased nitric oxide production in IL-32γTg mice is dependent on IFN-γ and IL-17A.Since NO is one of the major molecules involved in the killing of Leishmania spp. in mice (10), we evaluated NO production—as measured by the level of nitrite, the metabolite of NO—in the culture supernatant of splenic cells from 30-day-infected IL-32γTg mice stimulated with L. infantum antigen ex vivo. After incubation with total L. infantum antigen, the level of NO production by splenic cells from infected IL-32γTg mice compared to those from WT mice was increased. ConA stimulus did not induce significant differences in NO production between the splenic cells from WT and IL-32γTg mice (Fig. 6A). The nitrite levels in the splenic cell cultures activated with L. infantum antigens were positively correlated with the IL-32 levels present in the spleens of infected-IL-32γTg mice (Fig. 6B). Although infection with L. infantum induced IL-32γ mRNA in bone marrow-derived macrophages (BMDMs) from IL-32γTg mice (Fig. 6C), BMDMs from WT and IL-32γTg mice produced similar NO levels after in vitro infection with L. infantum promastigotes and stimulation with IFN-γ alone or with IFN-γ–LPS (Fig. 6D). In fact, there was no significant difference in the percentage of infected macrophages (Fig. 6E) or the number of amastigotes per macrophage (Fig. 6F) between WT and IL-32γTg mice (Fig. 6E). These findings indicate that the protective effect of IL-32γ during experimental VL is not solely due to phenotypic alterations in the macrophages themselves of the IL-32γTg mice.
Splenic cells from IL-32γTg mice infected with L. infantum produced larger amounts of NO after stimulation with L. infantum antigen ex vivo. (A) WT and IL-32γTg mice were infected i.p. with 1 × 107 L. infantum promastigotes. After 30 days, the splenic cells (5 × 106 cells/ml) were ex vivo stimulated with L. infantum antigen (50 μg/ml) or ConA (5 μg/ml) for 72 h. Nitric oxide production in culture supernatants was evaluated by the Griess method. (B) Correlation between the levels of IL-32 in the spleen and the nitrite concentration in the supernatant of the splenic cell culture. (C) BMDMs from IL-32γTg mice were infected with L. infantum (MOI, 5:1) for 3 h or 24 h, and IL-32 mRNA was analyzed by qPCR. (D) BMDMs from WT or IL-32γTg mice were infected with L. infantum (MOI, 5:1) for 3 h. Cells were washed to remove noninternalized parasites and cultured for additional 48 h in the absence or presence of IFN-γ (10 ng/ml) alone or IFN-γ (10 ng/ml) plus LPS (1 μg/ml). The NO levels in the 48-h culture supernatant were evaluated by the Griess method. (E and F) The percentage of infected macrophages (E) and the number of parasites per infected cell (F) were evaluated under a light microscope. In panel A, the data show the mean ± SEM for 9 animals per group from 3 independent experiments. *, P < 0.05 by Student's t test. In panel B, the data show the individual values for 9 animals and the linear regression curve. r and P values were from the Pearson correlation test. In panels C to F, the data represent the mean ± SEM from 3 independent experiments performed in duplicate.
Since WT and IL-32γTg mouse BMDMs produced NO at similar levels, we reasoned that the increased L. infantum antigen-specific NO production observed in IL-32γTg mouse spleen cells was secondary to the increased Th1/Th17 cytokine levels induced by IL-32γ. Splenic cells from 30-day-infected WT and IL-32γTg mice were then incubated with total antigen of L. infantum in the absence or presence of neutralizing antibody to IFN-γ or to TNF-α or a pharmacological inhibitor of Th17 cell differentiation and IL-17A production in vivo and in vitro (digoxin) (36, 37). Neutralization of IFN-γ or Th17/IL-17A inhibition significantly reduced the antigen-specific NO production by IL-32γTg mouse splenic cells. Simultaneous inhibition of IFN-γ and IL-17A further increased this NO inhibition (P < 0.05) (Fig. 7). In WT mice, only neutralization of IFN-γ reduced NO production. TNF-α neutralization had no significant effect on NO production by splenic cells from both groups of mice (Fig. 7).
Increased NO production in splenic cells from IL-32γTg mice infected with L. infantum is dependent on IFN-γ and IL-17. WT and IL-32γTg mice were infected i.p. with 1 × 107 L. infantum promastigotes for 30 days. The spleens were removed, and splenocytes (5 × 106 cells/ml) were stimulated ex vivo with L. infantum antigen (50 μg/ml) in the absence or presence of rat IgG isotype control, anti-IFN-γ or anti-TNF-α (5 μg/ml) antibodies, and/or DMSO or digoxin (10 μM) for 72 h. NO production was evaluated by measuring the nitrite concentrations in the supernatant of the WT and IL-32γTg mouse splenic cells. The data show the mean ± SEM for 6 animals per group from 2 independent experiments. *, P < 0.05 by one-way ANOVA with Bonferroni's posttest.
Taken together, these data demonstrate that IL-32γ can induce antigen-specific NO production during murine VL by increasing the production of IFN-γ and IL-17A. In fact, we observed that IFN-γ and IL-17A levels were positively correlated with NO production in cultures of spleen cells from IL-32γTg mice. On the other hand, there was no significant correlation between IFN-γ or IL-17A and nitrite levels in the cultures of spleen cells from WT mice. There was no correlation between TNF-α and nitrite levels in both animal groups (Fig. S6).
Thus, the protective effect of IL-32γ against murine VL appears to be dependent on the activation of an axis comprised of IL-32γ, Th1/IFN-γ, Th17/IL-17A, and macrophage/NO components.
DISCUSSION
In this study, we showed that L. infantum mainly induces the IL-32γ isoform and, to a lesser extent, the IL-32β isoform in PBMCs from healthy individuals. Our previous studies demonstrated enhanced expression of IL-32γ in human American tegumentary leishmaniasis (ATL) lesions and also in human PBMCs and macrophages after stimulation with L. braziliensis or L. amazonensis (32–34). We report here that L. infantum induces even higher levels of IL-32γ in human PBMCs than other Leishmania spp. or Candida albicans. We found that Leishmania protozoans are potent inducers of IL-32 expression and that L. infantum was the most potent inducer of IL-32. These data prompted us to investigate the role of IL-32γ in a murine model of VL.
The present study demonstrated that transgenic mice that constitutively express human IL-32γ also displayed increased induction of IL-32 in both the spleen and liver with L. infantum infection, with peak IL-32γ levels occurring at 30 and 45 days p.i. This result was unexpected, since the IL-32γ gene was driven by the chicken β-actin promoter. Although the constitutive expression of this gene is expected, different tissues and organs differentially expressed IL-32 (35). The 5′ flanking region of the β-actin gene promoter contains various cis-regulatory elements, including highly conserved CCAAT and TATA binding domains, present in the promoter region of β-actin found in rodents, humans, and chickens (38, 39). Indeed, several evidences have shown that β-actin can be induced by innumerable biological agents, proinflammatory cytokines, and cell proliferation. Furthermore, it is known that age and tissue type influence the expression of β-actin (40–43). It is already known that different cell types differentially express β-actin, which may explain, at least partially, the differences in the expression of IL-32 in the different organs. As we further showed that L. infantum infection increased inflammatory infiltrate in both spleen and liver, the increase in cellularity after infection may also explain the increased production of IL-32.
Of biological significance, the IL-32γTg mice were more resistant to L. infantum infection than WT mice, with reduced parasitism being found in the spleen and liver. It is likely that the relatively high expression of IL-32γ at about 30 days of infection contributed to the hepatomegaly and splenomegaly. Indeed, the liver was involved earlier in IL-32γTg mice than in WT mice, characterized in the former by an increase in the number, size, and necrosis of the granulomas, which likely accounted for the transient hepatomegaly. The hepatomegaly regressed from day 45 p.i. in both group of animals, but the splenomegaly was still significant in WT mice. The results of this study are in agreement with previous reports, which demonstrated that C57BL/6 mice develop VL after L. infantum infection and control the parasite after 60 to 80 days of infection (12, 44, 45). C57BL/6 mice that developed splenomegaly and hepatomegaly after L. infantum infection have improved control of parasitism (12, 44), possibly because of a tissue increase of a host-protective inflammatory response. Our data demonstrated that IL-32γ increased hepato- and splenomegaly and that the transgenic mice were even more resistant to L. infantum infection than WT mice, indicating the salutary effect of IL-32 in controlling VL.
The formation of granulomas is important to limit L. infantum infection, which can further lead to the killing of the parasites when associated with a strong Th1 immune response (46–48). Our data showed that IL-32γ favors the development of both Th1 and Th17 lymphocytic profiles. There was an increase of proinflammatory cytokines, such as IFN-γ, TNF-α, and IL-17A, during experimental VL in IL-32γTg mice. These data are also consistent with results showing that C57BL/6 mice displayed both host-protective Th1 and host-protective Th17 responses against L. donovani and L. infantum (12, 49). Interestingly, in our study, the Th1 and Th17 responses were improved in IL-32γTg mice, as demonstrated by the increase of IFN-γ and IL-17A in tissue lysates as well as after ex vivo stimulation of spleen cell cultures with a specific antigen. These findings are supported by studies showing the induction of Th17 cells and IL-17A by L. infantum and L. donovani, respectively (12, 50). Thus, IL-17 emerges to be an important host-protective cytokine during VL. We showed here that IL-32γ induces strong Th1 and Th17 responses against L. infantum. Moreover, IL-32γ appears to reduce the immunosuppression caused by infection with this Leishmania sp.
One of the mechanisms by which IL-32γ induces both Th1 and Th17 responses might be the improvement of dendritic cell maturation and subsequent enhanced production of IL-12 and IL-6 by dendritic cells in a phospholipase C/Jun N-terminal protein kinase/NF-κB-dependent mechanism (21). A previous study showed that IL-32 knockdown resulted in a drastic reduction of Th1 cytokines by human PBMCs (51). Indeed, proinflammatory cytokines such as TNF-α and IFN-γ induce IL-32, and in turn, IL-32 induces the production of these cytokines in a feedback loop that amplifies the inflammatory response (19, 52). IL-17A is also an inducer of IL-32 in rheumatoid arthritis, a disease in which there is a loop involving TNF-α, IL-32, and IL-17A (53). Thus, in the present VL mouse model, IL-32γ, IFN-γ, and IL-17A are important players in coordinating the immune response against L. infantum parasites.
The limitation of Leishmania growth is mainly dependent on the production of NO in mice, which is dependent on Th1 cytokines (54). We have demonstrated that IL-32γ induces NO production in splenic cells of infected IL-32γTg mice and IL-32 levels strongly correlated with nitrite levels. Indeed, our results indicate that the high levels of production of IL-32γ-induced IFN-γ and IL-17A appear to be essential for NO production and likely contributed to parasite control during VL. In fact, inhibition of IL-17-producing cells indicated that Th17 cells enhanced the ability of Th1 cells and macrophages to produce NO. Moreover, the correlation between NO and IFN-γ or IL-17A production was especially strong in IL-32γTg mice. We have demonstrated that IL-32γ is associated with NO production by human macrophages infected with L. amazonensis and L. braziliensis (34). Although macrophages isolated from WT and IL-32γTg mice produced amounts of NO similar to those induced by infection with L. infantum, our results indicated that the increases in both Th1 and Th17 levels in IL-32γTg mice can induce even more NO production by splenic macrophages during VL.
The role of Th1/IFN-γ production in NO production and the control of parasitism in VL is well established (10, 55, 56); however, the role of IL-17A in leishmaniasis is controversial (57). Terrazas et al. demonstrated that IL-17A promotes susceptibility to experimental VL, since IL-17A-deficient mice showed less parasitism in the spleen and liver than control mice (50). Conversely, it was recently shown that IL-17A synergized with IFN-γ to augment NO induction and is essential for resistance in experimental VL (12, 13, 45, 58) and other infectious diseases (59, 60). Significantly, the Th17 response and IL-17A production are associated with protection in patients with VL (11, 61). Indeed, Th1 and Th17 are important for induction of inducible nitric oxide synthase/NO and the increased resistance of dogs to VL (62). In human cutaneous leishmaniasis, Novoa et al. showed a protective role of IL-17 in patients infected with L. braziliensis (63). However, Th17 is associated with pathology and disease severity in mice and human models of cutaneous leishmaniasis, especially due to an increase in neutrophil recruitment, which is associated with tissue damage without the benefit of controlling parasite replication (64–67). Thus, despite the controversy, most previous studies have demonstrated that IL-17A contributes to the protective response against VL. It is important to note that a Th17 response alone does not protect patients with VL and that the Th1 response appears to be essential in protection against human VL even in the presence of large amounts of IL-17A (68).
IL-32γ splicing is important because different IL-32 isoforms may have different functions during immune responses (22, 69). IL-32β is especially known to induce the production of IL-10 (69, 70). We evaluated three main isoforms of IL-32, and only IL-32γ and IL-32β were detected during L. infantum infection, corroborating our previous hypothesis that Leishmania spp. can inhibit IL-32γ splicing (33, 34). Here, we detected the expression of IL-32β during experimental VL, but the levels of mRNA for this isoform were very low relative to the levels of mRNA for IL-32γ. Therefore, it is difficult to ascribe the higher levels of IL-10 in L. infantum-infected IL-32γTg mice than in WT mice solely to an increase of IL-32β during infection.
Although we have shown that IL-32γ is increased in human PBMCs in the presence of L. infantum, the role of this cytokine in human infection remains unknown. Mice present a weaker inflammatory process than humans, and they are susceptible to L. infantum infection (6); thus, we hypothesized that the expression of IL-32 could contribute to better control of the parasites by increasing the host-protective inflammatory response. In fact, we showed that there is an improved response to the parasite but not healing, as occurs in nontreated VL patients. Together with other proinflammatory cytokines, such as IFN-γ, TNF-α, and IL-17A, IL-32γ contributes to the control of the parasitic infection, but it also likely participates in the immunopathology of the disease (6, 71). In human VL, there is a large mixture of pro- and anti-inflammatory factors (6), but the network of their interactions is poorly understood. Our data highlight the relevance of future studies to investigate how the expression of IL-32 varies with the clinical course of the disease in humans.
In conclusion, the present study shows that IL-32γ is important in inducing an adaptive immune response profile during murine VL and that IL-32γ contributes to the control of L. infantum. These results support the need to evaluate IL-32 in human VL, where Th1 and Th17 cells are important to the control of this infection. Humanized mouse models allow a more relevant characterization of the host immune response to human infections, and herein we have demonstrated that human IL-32 is an important contributor to the host immune response to leishmaniasis. Furthermore, the role of IL-32 in improving both Th1 and Th17 responses during VL places IL-32 as a target to augment in future VL vaccine and adjuvant developments. Augmenting IL-32 expression in future leishmania vaccine development—with IL-32 functioning as an adjuvant itself and/or being induced by more traditional adjuvants—has the potential to improve existing vaccines against leishmaniasis (72).
MATERIALS AND METHODS
Ethics statement.The human sample study was approved by the Commission on Human Research-CMO Arnhem-Nijmegen (NL32357.091.10). Buffy coats from healthy donors were obtained after the donors provided written informed consent (Sanquin Blood Bank, Nijmegen, The Netherlands). Donor samples were kept anonymous, and the use of the samples received Commission on Human Research-CMO Arnhem-Nijmegen review board approval, as described above. All animal procedures followed were in accordance with the guidelines and legislation on ethics of the Brazilian Society of Science in Laboratory Animals (SBCAL) and the National Council of Control of Animal Experimentation (CONCEA) and were approved by the Committee on Ethics in the Use of Animals/CEUA-PRPI-UFG (protocol no. 042/16).
IL-32 expression in human PBMCs upon exposure to L. infantum antigens.Peripheral blood mononuclear cells (PBMCs) were obtained from blood samples after the samples were overlaid on a density gradient (Ficoll-Paque; Pharmacia Biotech, Piscataway, NJ, USA), as previously described (73, 74). PBMCs were cultured in RPMI 1640 medium (Sigma-Aldrich, St. Louis, MO, USA) supplemented with 10% fetal calf serum FCS (Gibco, Life Technologies, São Paulo, SP, Brazil), 2 mM l-glutamine, 11 mM sodium bicarbonate, 100 U/ml penicillin, and 100 μg/ml streptomycin (Sigma-Aldrich), referred to as complete medium. PBMCs (5 × 105) were cultured in the absence or presence of total L. infantum antigen (50 μg/ml) for 4 h, 24 h, and 48 h at 37°C in 5% CO2. After stimulation, the cells were lysed with phosphate-buffered saline (PBS)–0.5% Triton X-100 (Sigma-Aldrich) containing protease inhibitor cocktail (Sigma-Aldrich) for measuring the IL-32 protein level by enzyme-linked immunosorbent assay (ELISA; R&D Systems, Minneapolis, MN, USA), according to the manufacturer's protocol (detection limit, 31 pg/ml). In addition, expression of IL-32 mRNA was also evaluated as described below.
IL-32 mRNA expression by qPCR.RNA isolation was performed as described before (75). Briefly, mRNA was transcribed into cDNA using an iScript cDNA synthesis kit (Bio-Rad, Hercules, CA, USA). cDNA was used for quantitative real-time PCR (qPCR) analysis by using StepOnePlus sequence detection systems (Applied Biosystems, Foster City, CA, USA) with SYBR green Mastermix (Applied Biosystems). The primer sequences for IL-32 (Biolegio, Nijmegen, The Netherlands) used here are presented in Table S1 in the supplemental material and were used as previously described (76). The mRNA analysis was performed with the 2ΔCT × 1,000 method, and mRNA levels were normalized to those of the GAPDH (glyceraldehyde-3-phosphate dehydrogenase) housekeeping gene (human samples and PBMC lysates) or 18S rRNA (mouse samples, BMDMs, and spleen and liver lysates).
Animals and parasites.IL-32γTg mice were generated by Choi et al. (35) using the C57BL/6 mouse strain, donated by Charles Dinarello (University of Colorado Denver, USA), and maintained at the animal facility of the Federal University of Goiás/IPTSP, Brazil. Six- to 8-week-old- C57BL/6 wild-type (WT) and IL-32γTg mice were used in the experiments.
L. infantum (MHOM/BR/74/PP/75) was cultured in Grace's insect medium (Gibco-Life Technologies) supplemented with heat-inactivated 20% fetal calf serum (FCS; Gibco-Life Technologies), 0.2 mM l-glutamine (Sigma-Aldrich), and 100 U/ml of penicillin-streptomycin (Sigma-Aldrich) at 26°C. Promastigotes in stationary phase (5th day) of growth were washed three times with sterile phosphate-buffered saline (PBS), pH 7.4 (1,000 × g, 10 min, 10°C), suspended in PBS, and quantified by use of a hemocytometer after dilution in PBS–0.1% formaldehyde. Total antigen of L. infantum promastigotes was obtained as previously described (32).
In vivo infection and evaluation of parasitism and tissue lesions.WT and IL-32γTg mice were injected intraperitoneally (i.p.) with 107 promastigotes in the stationary phase of growth. After 15, 30, 45, and 60 days of infection, the animals were euthanized for collection and weighing of the organs and evaluation of parasitism in the spleen and liver by the limiting dilution method (77, 78). The percent splenomegaly and hepatomegaly was obtained by the following equation: (organ weight/total animal weight) × 100.
Paraformaldehyde-fixed spleen and liver tissues were embedded in paraffin for histopathological analysis. After staining with hematoxylin-eosin, the histopathological aspects of the spleen and liver and granuloma formation in liver sections were evaluated under a light microscope. The number and area of granulomas in 10 consecutive fields were quantified by using ImageJ software.
Mouse spleen cell cultures.Spleen cells from 30-day-infected mice were cultured at 5 × 106 cells/ml in RPMI 1640 complete medium and stimulated with 50 μg/ml of L. infantum antigen or 5 μg/ml concanavalin A (ConA; Sigma-Aldrich) for 24 h or 72 h at 37°C in 5% CO2. In some experiments, the cells were stimulated for 72 h with L. infantum antigen or ConA in the presence of neutralizing monoclonal antibodies to IFN-γ (clone XMG1.2) or to TNF-α (clone MP6-XT3). Moreover, the Th17/IL-17A pharmacological inhibitor digoxin (10 μM; Sigma-Aldrich) (36) was used. Normal rat IgG was used as the isotype control, and dimethyl sulfoxide (DMSO) was used as the digoxin control.
Preparation of tissue homogenates.To measure cytokine concentrations in the organs, tissue samples were harvested, weighed, and macerated in 0.5 ml of PBS–0.1% Triton X-100 (Sigma-Aldrich) to which protease inhibitor cocktail (Sigma-Aldrich) was added for cytokine measurement or in 0.5 ml of the TRIzol reagent for RNA isolation as described above.
Mouse cytokine and nitrite measurements.Murine IL-17, TNF-α, and IL-10 were evaluated by the use of commercial enzyme-linked immunosorbent assay (ELISA) kits (R&D Systems), according to the manufacturer's protocol (detection limits, 39 pg/ml for IL-17A and 78 pg/ml for TNF-α and IL-10). IFN-γ was evaluated by ELISA using monoclonal antibodies obtained from hybridoma cultures as described previously (79), and its detection limit was 39 pg/ml. The nitrite concentration was evaluated by the Griess method to evaluate NO production in the cultures (80). Human IL-32 from cell lysates was evaluated by ELISA (R&D Systems), according to the manufacturer's protocol, as described above for human cells.
Flow cytometry and lymphocyte proliferation assays.Spleens and livers from 30-day-infected mice were harvested and macerated to obtain single-cell suspensions. For cell surface marker analysis, cells were fixed, washed, and stained with anti-mouse CD3-fluorescein isothiocyanate (FITC) (clone 145-2C11), anti-mouse CD4-phycoerythrin (PE) (clone GK1.5), anti-mouse CD8-peridinin chlorophyll protein (PerCP)-cyanine 5.5 (Cy5.5) (clone 53-6.7), or their respective isotype controls (all from eBioscience, San Diego, CA, USA). For intracellular cytokine analysis, spleen cells were cultured in the presence of 50 μg/ml of L. infantum antigen for 72 h. Brefeldin A (10 μg/ml; eBioscience) was added to the culture, and the culture was incubated for an additional 6 h. The cells were washed with fluorescence-activated cell sorting buffer (eBioscience) and stained with anti-mouse CD4-PE (clone GK1.5; eBioscience) antibody, washed, fixed with 100 μl of intracellular fixation buffer (eBioscience), and permeabilized with permeabilization buffer (eBioscience). The cells were subsequently stained with anti-mouse IFN-γ–Alexa Fluor 488 (clone XMG1.2; eBioscience) and anti-mouse IL-17A–allophycocyanin (APC) (clone eBio17B7; eBioscience). The CD4+ Th cells were gated individually for determining the population of IFN-γ- or IL-17A-producing cells.
Antigen-specific proliferation analysis was performed by the carboxyfluorescein succinimidyl ester (CFSE) method, as described before (81). Briefly, spleen cells (5 × 106 cells/ml) were incubated with 5 μM CFSE (Sigma-Aldrich) at 25°C for 5 min. The cells were washed in PBS, resuspended in RPMI 1640 complete medium, and incubated at 37°C in 5% CO2 for 4 days in the presence of 50 μg/ml of L. infantum antigen or ConA (5 μg/ml). The cells were then collected and analyzed by flow cytometry.
All experiments were performed in a BD Accuri C6 flow cytometer (BD Bioscience, San Jose, CA, USA), and the data were analyzed using FlowJo software (TreeStar, Ashland, OR, USA).
L. infantum infection of mouse BMDMs and measurement of microbicidal activity and NO.Mouse bone marrow-derived macrophages (BMDMs) were obtained as described before (82). BMDMs (1 × 106 cells/ml) were cultured in RPMI 1640 complete medium and infected with stationary-phase L. infantum promastigotes (multiplicity of infection [MOI], 5:1) for 3 h. They were then washed to remove noninternalized parasites and incubated with complete medium in the absence or presence of murine recombinant IFN-γ (10 ng/ml; Sigma-Aldrich) and lipopolysaccharide (LPS; 1 μg/ml; Sigma-Aldrich). After 3 h and 48 h at 37°C in 5% CO2, the coverslips were collected and the cells were fixed in methanol (Labsynth, Diadema, SP, Brazil) and stained with Giemsa (Merck KGaA, Darmstadt, Germany). The percentage of macrophages containing internalized parasites and the number of parasites per infected cell were evaluated under a light microscope. At least 200 cells were counted. Supernatants were collected for nitrite measurement as described above.
Statistical analysis.Data were expressed as means ± standard errors of the mean (SEM), and Student's t test and one-way or two-way analysis of variance (ANOVA) with Bonferroni's posttest were used. The median with the interquartile range and the Mann-Whitney U test were used for data with a nonparametric distribution. Prism (version 6.0) software (GraphPad Software, La Jolla, CA, USA) was used, and a P value of <0.05 was considered significant.
ACKNOWLEDGMENTS
We thank the Pathology Section of the Instituto Goiano de Oncologia e Hematologia (INGOH) for technical support.
We thank CNPq/FAPEG for financial support (grant no. 465771/2014-9 to F.R.-D. and L.A.B.J. from the National Institute of Science and Technology [INCT] of the strategies in host-pathogen interaction, Brazil). F.R.-D., M.A.P.O., and M.M.T. are research fellows of CNPq. R.S.G., M.V.T.S., and J.C.D.S. were financially supported by CAPES.
The funders had no role in study design, data collection and interpretation, or the decision to submit the study for publication.
FOOTNOTES
- Received 2 November 2017.
- Returned for modification 11 December 2017.
- Accepted 3 February 2018.
- Accepted manuscript posted online 26 February 2018.
Supplemental material for this article may be found at https://doi.org/10.1128/IAI.00796-17.
- Copyright © 2018 American Society for Microbiology.
REFERENCES
- 1.↵
- 2.↵
- 3.↵
- 4.↵
- 5.↵
- 6.↵
- 7.↵
- 8.↵
- 9.↵
- 10.↵
- 11.↵
- 12.↵
- 13.↵
- 14.↵
- 15.↵
- 16.↵
- 17.↵
- 18.↵
- 19.↵
- 20.↵
- 21.↵
- 22.↵
- 23.↵
- 24.↵
- 25.↵
- 26.↵
- 27.↵
- 28.↵
- 29.↵
- 30.↵
- 31.↵
- 32.↵
- 33.↵
- 34.↵
- 35.↵
- 36.↵
- 37.↵
- 38.↵
- 39.↵
- 40.↵
- 41.↵
- 42.↵
- 43.↵
- 44.↵
- 45.↵
- 46.↵
- 47.↵
- 48.↵
- 49.↵
- 50.↵
- 51.↵
- 52.↵
- 53.↵
- 54.↵
- 55.↵
- 56.↵
- 57.↵
- 58.↵
- 59.↵
- 60.↵
- 61.↵
- 62.↵
- 63.↵
- 64.↵
- 65.↵
- 66.↵
- 67.↵
- 68.↵
- 69.↵
- 70.↵
- 71.↵
- 72.↵
- 73.↵
- 74.↵
- 75.↵
- 76.↵
- 77.↵
- 78.↵
- 79.↵
- 80.↵
- 81.↵
- 82.↵