Genetic variation in Interleukin-32 influence the immune response against New World Leishmania species and susceptibility to American Tegumentary Leishmaniasis

Interleukin-32 is a novel inflammatory mediator that has been described to be important in the immunopathogenesis and control of infections caused by Leishmania parasites. By performing experiments with primary human cells in vitro, we demonstrate that the expression of IL-32 isoforms is dependent on the time exposed to L. amazonensis and L. braziliensis antigens. Moreover, for the first time we show the functional consequences of three different genetic variations in the IL32 (rs4786370, rs4349147, rs1555001) modulating IL-32γ expression, influencing innate and adaptive cytokine production after Leishmania exposure. Using a Brazilian cohort of 107 American Tegumentary Leishmaniasis patients and a control cohort of 245 healthy individuals, the IL32 rs4786370 genetic variant was associated with protection against ATL, whereas the IL32 rs4349147 was associated with susceptibility to the development of localized cutaneous and mucosal leishmaniasis. These novel insights may help improve therapeutic strategies and lead to benefits for patients suffering from Leishmania infections.

by immune cells, but it is additionally expressed in non-immune cells including epithelial and endothelial cells [24][25][26][27][28]. IL-32 mRNA can be processed by alternative splicing, generating several different isoforms. The isoforms are generated from the pre-mRNA for IL-32γ, which in respect to inflammation is the most active isoform. In the context of human diseases, the most studied isoforms are IL-32α, IL-32β and IL-32γ [29,30]. On the functional level, IL-32γ amplifies inflammatory signaling induced by TNFα and synergizes with PRRs for the induction of IL-1β and IL-6 [31,32]. Through these mechanisms, IL-32 drives innate immune cells to generate Th1 and Th17 cellular responses [33]. From a molecular perspective, IL-32 can inhibit proinflammatory cytokine mRNA decay influencing the activation of transcription factors [26]. Furthermore, IL-32γ inhibits Hepatitis B virus activities at the transcriptional level through a non-cytokine function in hepatocytes [34].
We previously described the expression of IL-32 in epithelial, endothelial and mononuclear cells in lesions from patients with LCL and ML [35]. In fact, IL-32 has also been implicated as a key player in controlling parasite load in infections caused by L. braziliensis or L. amazonensis in both the IL-32γ-transgenic mouse model and human cells [36,37]. To date, it is not precisely known how IL-32 isoforms contribute to host defense against Leishmania infections. In the current study, we confirmed that IL32 mRNA and protein are expressed in human PBMCs stimulated with L. amazonensis or L. braziliensis. We further evaluated the functional consequences of three different genetic variations in IL32 (rs4786370, rs4349147, rs1555001) for innate and adaptive cytokine production. Furthermore, by analyzing a cohort of 107 ATL patients and 245 controls, we evaluated whether these genetic variations in IL32 influence the susceptibility to disease and the clinical outcome. Finally, the expression of important inflammatory mediators in lesion fragments from ATL patients was correlated with mRNA levels of IL-32 isoforms.

Patients, controls and tissues samples
The Brazilian cohort of ATL patients: to access genetic variations, ATL patients (n = 107) were asked to provide blood. All patients were diagnosed with cutaneous (n = 68), mucosal (n = 38) or diffuse (n = 1) ATL in Annuar Auad Tropical Disease Hospital, in Goiânia, Goiás, West Central region of Brazil. To be included in the study, besides a leishmaniasis lesion, patients had at least one exam indicating the presence of parasites were included. Among the technical procedures used for the diagnosis, polymerase chain reaction (PCR) for Leishmania genus, histopathological exam with immunohistochemistry for amastigotes forms (IHC), indirect immunofluorescence (IFI), and Montenegro's skin test (MST) were used (Table 1) [38]. For cytokine expression evaluation, fragments of cutaneous and mucosal lesions of 56 ATL patients were obtained. All the fragments were tested for parasite DNA using a polymerase chain reaction (PCR-RFLP) described by [39], and presented positive result for Leishmania Viannia subgenus. For gene expression evaluation, one lesion fragment was stored in TRIzol reagent (Invitrogen, USA). For the controls (n = 245), blood from healthy individuals was obtained at the blood bank of the Goiano Institute of Oncology and Hematology (INGOH), in Goiânia, Goiás. All procedures were approved by Ethics Committee of Hospital das Clínicas and Hospital Anuar Auad (CAE 47895015.0.0000.5078) and of Federal University of Goiás (n. 164/08), Goiânia, Goiás, Brazil and informed consent was signed by all patients and controls. An additional study was performed using venous blood obtained from 3 Crohn's disease patients bearing the NOD2 3020insC mutation, blood of 6 healthy individuals was used as control. All human subjects of the study were adults.

Leishmania cultures and lysates
L. amazonensis (IFLA/BR/67/PH8) reference strain and MHOM/BR/2003/IMG L. braziliensis, a strain from Leishbank IPTSP/UFG, were used. Promastigote forms were cultured in Grace's insect medium (Gibco, Life Technologies, USA) supplemented with 20% of heat-inactivated fetal bovine serum (FBS, Gibco), 100 U/mL of penicillin/streptomycin (Sigma-Aldrich), at 26˚C. Stationary-phase parasites were obtained on the 6th day of growth and washed three times with phosphate-buffered saline (PBS; 1000 x g, 10 min 10˚C). Parasite lysates were obtained by 5 freeze-thaw cycles of promastigotes in the presence of protease inhibitors (Protease inhibitor cocktail, Sigma-Aldrich), in liquid nitrogen followed by thawing in a water bath at 37˚C. Protein quantification was performed using the Pierce BCA protein assay kit (Ther-moFisher Scientific, USA).

Isolation of peripheral blood mononuclear cells (PBMC) from healthy controls and treatments
Peripheral blood mononuclear cells (PBMCs) were isolated from buffy coats obtained after informed consent of healthy volunteers (Sanquin BloodBank, Nijmegen) and from EDTA tubes of homozygous individuals for the NOD2 3020insC mutation, presenting Crohn's disease. PBMCs were obtained by Ficoll-Paque density gradient cell isolation (GE healthcare, UK), as previously described [12,40]. Cells were resuspended in RPMI 1640 culture medium (Roswell Park Memorial Institute medium; Invitrogen, USA) supplemented with 50 μg/mL gentamicin, 2 mM glutamax (Gibco), and 1 mM pyruvate (Gibco) and quantified. In brief, PBMC stimulations with lysates of both L. amazonensis or L. braziliensis (50 μg/mL) were performed with 5 x 10 5 cells/well in round-bottom 96-wells plates for either 24 h or 7 days in the presence of culture medium with 10% human pool serum at 37˚C and 5% CO 2. In addition, PBMCs were stimulated during 24 h with LPG (10 μg/mL) from L. amazonensis or L. braziliensis. In some experiments, PBMCs were pre-incubated for 1 h, at 37˚C, and 5% CO 2 with either Bartonella quintana LPS (5 μg/mL) [40] for TLR4 blocking or with laminarin (100 μg/mL; Sigma-Aldrich) as antagonist of Dectin-1. After the stimulation period, supernatants and cells were stored in either 200 μL of TRIzol (Invitrogen) at -80˚C until RNA extraction or in 100 μL of 0.5% TritonX-100 (Sigma-Aldrich) for intracellular IL-32 measurement.

Genetic analysis in the 200FG cohort
Genetic variation was assessed in the IL-32 gene in healthy individuals of Western European descent from the 200FG cohort [41], (http://www.humanfunctionalgenomics.org/). The volunteers were between 23 and 73 years old and consisted of 77% males and 23% females. PBMCs were isolated as described above. After 24 h and 7 days, supernatants were collected and cytokines were measured by enzyme-linked immunosorbent assay (ELISA) as described below.

Transcriptional analysis
Previously published microarray data of normal skin samples (n = 10), early (n = 8) and late (n = 17) LCL lesion samples were obtained form the publicly available National Center for Biotechnology Information (NCBI) Gene Expression Omnibus (GEO) database (accession number GSE55664) [42]. The expression of selected genes in normal skin versus L. braziliensis lesions were represented in a heatmap presented as dendograms shown as a tree, representing the distance between variables. Unsupervised hierarchical clustering of selected genes was performed using the R package hclust with Spearman correlation and presented using the R package 'corrplot'.

Quantitative PCR
RNA was isolated from PBMCs and ATL patient's lesions stored in TRIzol as previously described [43]. cDNA was made using the iScript kit (Bio-Rad). Primer pairs for the IL32 isoforms are described elsewhere [36]. IL1B, IL1A, IL22, IL-17 and IFNG primers sequences were designed specifically for each transcript using Primer Express 3.0 software (DeNovo Software, Glendale, CA, USA). Primers were produced by Sigma-Aldrich. Diluted cDNA was used for qPCR that was done by using the StepOnePlus qPCR system (Applied Biosystems, Foster City, CA, USA) with SYBR Green Mastermix (Applied Biosystems). Fold change and relative expression were calculated with the 2^-ddCT and 2^-dCT methods, respectively and normalized against the housekeeping gene B2M.

Enzyme-linked immunosorbent assay (ELISA)
Cytokine production was determined in 24 h supernatants using commercial ELISA kits (R&D Systems) for human TNFα, IL-6 and IL-1β. Supernatants of the 7 days stimulation were used to measure IFNγ, IL-17A and IL-22. IL-32 was determined intracellularly in Triton X-100-cell lysates after 24 h and 7 days using commercial ELISA kit (R&D Systems).

Isolation of genomic DNA and genetic assessment of IL32 variations in ATL patients and controls
DNA was isolated from venous blood of ATL patients and healthy controls using the illustra blood genomicPrep Mini Spin Kit (GE Healthcare, Little Chalfont, UK), according to the manufacturer's protocol. SNPs in IL32 were selected based on previously published functional effects on protein function and/or gene expression and characteristics are described in detail in Supplementary Table 1 [44,45]. Genotyping of ATL patients and controls was performed by TaqMan SNP assays (S1 Table) (Applied Biosystems), according to the manufacturer's protocol on the StepOnePlus qPCR system (Applied Biosystems). Quality control was performed by the incorporation of positive and negative controls and duplication of random samples across different plates.

Statistical analysis
All statistical analyses and graphs of cytokine concentrations and mRNA expression in the in vitro PBMC stimulation assays were performed and created using GraphPad Prism 6.00 software (GraphPad Software, San Diego, CA, USA). Data are shown as mean ± SEM. Results were stratified by genotype, and differences in cytokine production between genotypes were analyzed by Mann-Whitney U test, otherwise in each figure, the applied statistical test is indicated in the figure legend. The difference in genotype frequencies between the patients and the control group was analyzed in a dominant and a recessive model using logistic regression. No correction for multiple testing was performed. The effect of the genotypes on ATL susceptibility was estimated by calculating odds ratios (ORs) and their 95% confidence intervals using the same statistical methods. We also performed Fisher test analysis to determine whether clinical manifestations of ATL were associated with one of the IL32 genotypes. Analyses were conducted using the statistical computing evrionment, R (v3.5.0). Overall, statistical tests P-values of < 0.05 were considered to be statistically significant.

Differential IL-32 isoform expression is dependent on duration of exposure to Leishmania antigens
To investigate whether Leishmania species can induce different isoforms of IL32 in PBMCs from healthy individuals in a time-dependent manner, we evaluated IL32 mRNA expression. Both L. amazonensis and L. braziliensis induced IL-32γ and IL-32β expression in PBMCs after incubation for 24 h (Fig 1A and 1B). On the other hand, after 7 days of Leishmania-antigen stimulation, the IL-32β and IL-32α were the main isoforms expressed (Fig 1C and 1D). The production of intracellular IL-32 protein by PBMCs upon exposure to antigens from both Leishmania species was enhanced compared to unstimulated cells at both time points (Fig 1E  and 1F). However, we observed a 5-fold increase in intracellular IL-32 between 7 days ( Fig 1F) and 24 h of incubation ( Fig 1E).

Early induction of IL-32γ is mediated by NOD2, TLR4 and Dectin-1
In order to evaluate which are the receptors involved in early IL-32 induction by Leishmania spp. in PBMCs, we further assessed whether genetic variations in PRRs could affect the expression of IL-32γ expression. Supplemental Fig 1A (S1A Fig) shows a reduction of IL-32γ mRNA expression in PBMCs isolated from individuals carrying a frame shift mutation in the NOD2 receptor (rs2066847-leu1007insC), when compared to individuals carrying no mutation (Wt) after 24 h of exposure to L. amazonensis antigens. Exploring TLR2 and TLR4, we observed that a missense SNP in TLR4 (rs4986791), but not in TLR2 (rs5743708), modulates IL-32γ mRNA expression upon L. amazonensis exposure. In addition, we investigated the role of C-type lectin receptors, more specifically Dectin-1 (CLEC7A rs16910526 SNP) and Mannose receptor (CD206 rs1926736 SNP) role in IL-32γ induction. S1B Fig (S1B Fig) shows that a nonsense SNP in CLEC7A might influence IL-32γ mRNA expression, since one individual carrying the CC genotype presented lower IL-32γ mRNA expression compared to individuals bearing AA and AC genotypes. This effect was seen for both species of Leishmania. No effect on IL-32γ mRNA expression was observed for individuals carrying different genotypes for a missense SNP in CD206 (rs1926736). The influence of NOD2, TLR4 and Dectin-1 signaling on IL-32γ induction upon Leishmania spp. exposure was further validated using biological assays. PBMCs isolated from patients carrying a frameshift mutation in NOD2 expressed lower levels of IL-32γ mRNA than healthy individuals after L. amazonensis stimulation. There were no differences in L. braziliensis-stimulated cells (Fig 2A). In addition, when TLR4 was blocked with Bartonella quintana LPS (TLR4 antagonist [40] there was a significant reduction in Leishmania spp.-induced IL-32γ mRNA expression ( Fig 2B). Next, using laminarin as a Dectin-1 antagonist, we observed a significant reduction in IL-32γ mRNA expression after exposure to both species of Leishmania. (Fig 2C).

Genetic variations in IL-32 gene influence cytokine production upon Leishmania exposure
To further investigate the relevance of IL-32 for Leishmania-induced cytokine production, we used functional genomics searching for SNPs within IL32 that alter gene expression. Therefore, we assessed the response of PBMCs isolated from 200 healthy volunteers (200FG cohort) stimulated with L. amazonensis or L. braziliensis. Fig 3A showed that a CC genotype of the promoter SNP (rs4786370) trends to have higher mRNA IL-32γ expression after L. amazonensis exposure compared to the TT/CT genotypes (p = 0.057). No major differences were observed after L. braziliensis stimulation. In addition, the effect of a SNP (rs4349147) located in a distal regulatory (enhancer) region of IL32 was examined. Fig 3B showed that individuals bearing the AA genotypes trends to lower levels of L. amazonensis-induced IL-32γ mRNA than those bearing GG/GA genotypes (p = 0.08). Again, no differences were observed for L. braziliensis. In addition, the effect of an intronic SNP (rs1555001) on IL32 gene expression was investigated. Once again, although no significant differences were observed (p > 0.05), the AA genotypes trends to lower IL-32γ mRNA expression compared to the TT/AT genotypes after exposure to both Leishmania species (Fig 3C).
Since genetic variation in the IL32 gene modulates IL32γ expression, we investigated whether these genetic variations influence both innate and adaptive cytokine production in human PBMCs cultured in the presence of L. amazonensis or L. braziliensis antigens. Therefore, we analyzed PBMCs isolated from healthy volunteers of the 200FG cohort with or without those SNPs in the IL32 gene. The classical innate cytokines TNFα, IL-6 and IL-1β were evaluated after 24 h of PBMC stimulation (early IL-32γ production). No significant differences were observed in any of the examined cytokines within the genotypes of the IL-32 promoter SNP after both Leishmania spp. exposure ( Fig 4A). In contrast, a lower TNFα, IL-6 and IL-1β production was observed in the group of individuals with the AA genotypes of the enhancer SNP in IL-32 than in PBMCs from individuals bearing the GG/GA genotypes after L. amazonensis stimulation. Moreover, for L. braziliensis antigen stimulation, a reduction in IL-1β production was observed in cells from individuals carrying the AA genotype of the same SNP compared to the GG/GA genotypes. No significant differences were observed in L. braziliensis-induced TNFα and IL-6 production ( Fig 4B).
For the intronic SNP of IL32, significant lower TNFα and IL-1β production was observed in the PBMC cultures of individuals carrying the AA genotype than in those of individuals bearing TT/AT genotypes after stimulation with either species of Leishmania. No differences in IL-6 production were observed between the genotypes (Fig 4C). Subsequently, production of the lymphocyte-derived cytokines IFNγ, IL-17 and IL-22 were determined after 7 days of PBMCs cultured with Leishmania antigens. Fig 5A shows a higher IL-22 production in the cultures of PBMCs from individuals with CC genotype than those with TT/CT genotypes of the IL-32 promoter SNP (rs4786370) after L. amazonensis stimulation. No differences were observed neither in IFNγ nor in IL-17 production when comparing these genotypes. In addition, for the enhancer SNP (rs4349147), IL-22 production was lower in the AA genotype than the GG/GA genotypes after exposure to both Leishmania species (Fi 5B). In line with previous SNP data, no differences were observed in IFNγ neither in IL-17 production. Interestingly, for the intronic SNP (rs1555001), the production of IFNγ, IL-17 and IL-22 was lower in the cells with AA genotype, which is associated with low levels of IL-32γ expression after L. amazonensis stimulation (Fig 5C). Similar results were noted for IFNγ and IL-22 production when PBMCs were stimulated with L. braziliensis. Strikingly, L. braziliensis was not capable to induce IL-17 production in the PBMC cultures, in contrast to L. amazonensis (Fig 5).

Association of IL-32 genetic variations with susceptibility to Leishmania infection
The observation that IL-32 is induced in PBMCs upon stimulation with Leishmania spp. and that genetic variations in the IL32 gene affect IL-32γ expression, and thereby the production of proinflammatory cytokines, led us to the hypothesis that IL32 SNPs could affect susceptibility to ATL. A Brazilian cohort of 107 ATL patients and a control cohort of 245 healthy individuals were genotyped for IL32 rs4786370 and rs1555001 genetic variations (Table 2). Different models of genetic variations were tested, including comparison of frequencies of all three genotypes, a dominant model (comparison of frequencies between homozygous wild-type versus

Association between IL32 genetic variations and ATL clinical manifestations
To assess whether the IL32 genetic variation influences clinical manifestations of ATL, patients were stratified according to the clinical forms, including LCL, ML and DL and compared among the different IL32 genotypes. No statistically significant differences were observed within the frequencies of the genotypes for either IL-32 rs4786370 and rs1555001variations and the clinical outcome of the patients (Table 3). However, a significant association was observed for the IL-32 rs4349147 genotypes, with the G allele overrepresented in the group of patients with LCL, whereas the A allele is overrepresented in the group of ML patients (P = 0.002). These results suggest that the A allele is associated with susceptibility to develop the more severe form of the disease.

IL-32γ expression is strongly associated with proinflammatory cytokines in ATL skin lesions
To unravel whether IL32 expression in skin lesions of LCL patients is associated to genes linked to skin inflammation and host defense, we analyzed the gene expression profile of  (Fig 6A). Next, we evaluated whether all these genes were correlated to each other in LCL skin lesions (Fig 6B, blue: positive; red: negative correlations). A strong positive correlation was observed between functionally related genes, such as IL32 and IL22, which is adjacent to a another cluster of CLEC4E, IL6, IL1RN, IL1A, MMP10, MMP9 genes which are also positively correlated.
Since we demonstrate that genetic variations in IL32 influence proinflammatory cytokine production in PBMCs stimulated with Leishmania spp. (Figs 4 and 5), we were prompted to validate whether IL-32γ expression is associated with innate (IL1B, IL1A) and adaptive cytokines (IFNG, IL17 and IL22) in lesion fragments of our cohort of ATL patients. A significantly positive correlation between IL-32γ expression and IL1A (Fig 6D), IFNG, and IL17 (Fig 6E and  6F) was apparent. Of great relevance for host defense, a strong correlation was found between IL-32γ and IL22 (r 2 = 0.8930, p < 0.0001) (Fig 6G) expressions in ATL lesions. Remarkably, no significant correlation was observed for IL-32γ and IL1B expression ( Fig 6C). Moreover, a significant correlation of IL-32β mRNA expression with IL17 and IL22 mRNA expression was

Discussion
In the present study, we demonstrated that both L. amazonensis and L. braziliensis induce IL-32 expression in primary human PBMCs. Of interest, L. amazonensis is a stronger inducer of both IL-32γ mRNA and protein than L. braziliensis at least during the first 24 h of PBMC stimulation with Leishmania spp. antigens. Intriguingly, when PBMCs were exposed to both Leishmania species for 7 days, enhanced splicing of IL-32γ mRNA occurred with an increase in IL-32α and IL-32β isoforms expression, resulting in a high amount of IL-32 in the cells. The induction of both IL-32 mRNA and protein reveals to be dependent on several PRRs, namely NOD2, TLR4 and Dectin-1. We showed that genetic variants of the IL32 gene influenced IL-32γ mRNA expression. Moreover, we could demonstrate that these investigated genetic variations influenced the production of innate and acquired Leishmania-induced cytokine production by primary human PBMCs. Of high interest, our data suggest that the IL32 SNP (rs4786370) is associated with susceptibility to ATL. In addition, the IL32 SNP (rs4349147) was associated with the clinical outcome of ATL patients, being more frequent among ML than LCL patients. Lastly, here we demonstrated that the IL-32γ mRNA expression in lesions of ATL patients was positively correlated with the expression of several proinflammatory cytokines, such as IL1A, IFNG, IL17, and IL22. In addition, IL-32α mRNA expression was positively correlated with IL1A and IL17 mRNA expression whereas IL-32β mRNA expression was associated with IL17 and IL22 mRNA expression.
In the present study, it was shown that promastigote antigens are able to induce early expression of IL-32γ in human PBMCs (Figs A and B). In contrast, after 7 days of exposure to both Leishmania species, human PBMCs predominantly expressed IL-32α and IL-32β mRNA (Fig 1C and 1D). In agreement with our findings using promastigotes antigens, previously studies have shown that the infection of THP-1 macrophages with live promastigote forms of L. amazonensis or L. braziliensis as well as the adittion of amastigotes forms in PBMCs lead to the early induction of IL-32γ mRNA expression [35,36]. In addition, promastigote antigens from L. amazonensis were able to induce higher levels of IL-32γ than L. braziliensis in PBMCs,  [42]. Columns represent individual patients and rows represent individual confirming the previous results obtained with THP-1 macrophages [36]. In the present study, we have shown that the expression of IL-32 isoforms is dependent on duration of exposure to Leishmania antigens, which might reflect the ability of the parasite to modulate IL-32γ splicing as the infection progress. Moreover, the subset of cells studied, myeloid (24 h) or lymphoid (7 days), may influence the expression of the different IL-32 isoforms. Interestingly, the highest concentrations of intracellular IL-32 protein were seen after 7 days of exposure to Leishmania spp. (Fig 1F). Of note, since IL-32 isoforms have different biological properties in regard to the induction of inflammation [46], to determine which IL-32 isoform is being induced for Leishmania spp. taking in account the subset of cells and the different stages of disease (early or chronic) might be relevant to understand its relation with the immunopathogenesis of the disease.
Although it was reported previously that Leishmania spp. can induced IL-32 in human cells and it is positively associated with cytokine production in vitro and in vivo [35,36], which receptor is involved in this process is unknown. Using a functional genomics approach, we were able to identify the receptors that are important for Leishmania spp. mediated induction of IL-32γ, the most inflammatory isoform of IL-32. Functional genomics revealed to be a powerful approach to identify novel pathways in infectious diseases, such as candidemia, tuberculosis and Lyme disease [47,48]. Of high interest, using a genomics approach and functional validation, we could show that NOD2, TLR4 and Dectin-1 receptors are involved in Leishmania-induced IL-32γ expression. Several C-type lectins, such as MR and Mincle have been described as recognition receptors for Leishmania [49][50][51] [12,52]. It is unlikely that MR is involved in the induction of IL-32γ since there were no SNPs identified that modulated the IL-32γ expression (S1B Fig). Previous data have shown the relevance of IL-32γ for cytokine and microbicidal molecules production during Leishmania infections using THP-1 macrophages [36]. In order to demonstrate the role of IL-32 in Leishmania-induced cytokine production by primary cells, we described genetic variations within IL32 that alter IL-32γ mRNA expression and proinflammatory cytokines produced by cells from innate and acquired immunity; thus, we have shown that depending on the Leishmania spp., genetic variations in IL32 lead to functional consequences modulating IL-32γ expression and the production of proinflammatory cytokines in human primary cells (Figs 3,4 and 5). As all cytokines evaluated in this study have been shown to be involved in immunopathogenesis of leishmaniasis [15,16,53] and in the parasite control [54,55], the present data indicate a pivotal role of IL-32 in the induction of cytokine production after Leishmania spp. exposure. The results presented here confirm that Il-32γ is a crucial cytokine to control cytokine production induced by Leishmania spp. in macrophage cell lines as well as in primary cells.
To investigate whether these genetic variations in IL32 could be associated with susceptibility or resistance to ATL, we evaluated the differences in genotypes frequencies of ATL patients compared to controls. Genetic variations have been described associated with suscepitibility to complicated skin and skin structure infections (cSSSIs) and HIV infections [45,56]. Likewise in our study, we observed an increased frequency of IL32 rs4786370-CC genotype in controls. This genotype was linked with high IL-32γ and IL-22 induction in PBMCs after exposure to L. IL-32 influences host defense against Leishmania species amazonensis antigens, which suggest a possible role of IL-32 in decreasing the risk of developing ATL. Despite of in vitro cytokine alterations, the other IL32 genetic variants (rs4349147; rs1555001) presented similar frequencies between the groups of ATL patients and controls what could be ascribed due the low number of individuals included in the study. In addition to the low number, other relevant aspect for genetic studies is the presence of a heterogeneity population present in Brazil. Using ancestry-informative DNA polymorphisms in individuals from the four most populous regions of the country, a previous study has unraveled great ancestral diversity between and within different regions of the country [57]. Thus, although the sex-related differences in ATL susceptibility was not addressed in the present study, data from across Brazil show that males are more susceptible to develop cutaneous manifestations of leishmaniasis than females [58]. Since this is first study investigating the role of genetic variations in IL32 associated with susceptibility to ATL, the use of a collective dataset including for genetic associated studies with large number of samples as well as matched asymptomatic and healthy controls, epidemiological variables and host biological factors respresent a promising new endeavor within the scientific community in the field of Leishmania research in Brazilian subjects.
After stratifying patients for their IL32 genotype and correlate it with clinical manifestation of the disease, it was observed the allele G of the IL-32 rs4349147 was more frequent in patients with LCL than ML. The G allele was associated with increased levels of IL-32γ, IL-6, IL-1β and IL-22 in PBMCs of healthy individuals exposed to L. amazonensis antigens, and with IL-1β and IL-22 to L. braziliensis antigens. Thus, these findings suggest that allele G might contribute to the development of the less severe form of the disease conferring protection against ML. Future studies exploring whether the IL-32 rs4349147 is associated with the production of immune mediators are worthwhile in a cohort of ATL patients.
Based on the expression of individual genes coding for immune host defense and being described related with IL-32 expression [29], early and late lesions from L. braziliensis skin that are transcriptionally indistinguishable [42] could be clearly separated from normal skin based on mRNA levels. Expression of IL32 and other genes that promote severe inflammation including cytokines, chemokines and also metalloproteinases, which can degrade the extracellular matrix leading to more tissue damage, were higher in L. braziliensis lesions. Positive correlations between IL-32 and IL-22 were observed in L. braziliensis lesions. IL-22, as well as IL-32, have been described to be important for host defense against microorganisms and antimicrobial peptide secretion [28,29,59].
To investigate the associations between Il-32 isoforms and other proinflammatory cytokines in vivo, we evaluated the expression of IL1A, IL1B, IFNγ, IL17, and IL22 in lesions of ATL patients and studied their correlations with Il-32 alpha, beta and gamma. The mRNA of the three IL-32 isoforms, being IL-32γ > IL-32β > IL-32α were detected in the lesions of ATL patients. In the lesions of of ATL patients, IL-32γ mRNA expression was correlated with IL1A, IFNγ, IL17 and IL22. Although IL-32γ is the most prevalent isoform in the lesions of ATL patients (present data; [35]), we also evaluated whether IL-32β and IL-32α isoforms were associated with proinflammatory cytokines in ATL lesions. IL-32β mRNA was correlated with and IL17 and IL22; IL-32α with IL1A and IL17 (S1, S2 and S3 Figs). These results suggests that the IL-32γ isoform can be associated with a mixed Th1/Th17 profile whereas IL-32β and IL-32α can be more associated with a Th17 profile. It is known that IL-32γ can induce a mixed Th1/ Th17 acquired immune response [33,60]. Previous data has shown that in other inflammatory skin diseases such as Psoriasis and Hydradinitis Superativa, IL-32β was the main isotype that was expressed [61]. Although IL-32β is known to present proinflammatory properties [43], it has been shown that IL-32β induces IL-10 production [62]. IL-32α is regarded as anti-inflammatory member of the IL-32 family which is able to inhibit NK cell activation and JAK/STAT pathway [63]. Thus, in our study all three isoforms are associated with proinflammatory cytokines in ATL, it is possible that IL-32β and IL-32α act as anti-inflammatory mediators in the lesions, being important to decrease the inflammatory process, however, whether IL-32β or IL-32α acts as an anti-inflammatory mediators in lesions of ATL patients is not known at this moment.
In the present study we demonstrated that IL-32γ mRNA expression in the lesions of Leishmania-infected patients strongly correlated with cytokines related to skin inflammation, such as IL-17, IL-22 and IL-1α. Of high interest, IL-22 was found to be highly correlated with IL-32γ expression. IL-22 is known as crucial mediator in cutaneous host defense against extracellular microorganisms, however, it seems not relevant to protect against intracellular micoorganisms [64]. In addition, IL-22 presents proinflammatory and anti-inflammatory properties, promoting wound healing/repair by activationg epithelial cells and fibroblasts [64,65]. In L. major-or L. braziliensis-infected mice the production of IL-22 decreases the lesion size without any effect on parasite load, indicating that IL-22 is more associated with the control of inflammation and wound healing than with parasite control in cutaneous leishmaniasis [23,66]. The role of IL-17 in ATL has been studied. Several studies suggest that IL-17 can be pathogenic in ML patients [16,20] whereas other suggests a protective role for IL-17 in LCL [67]. In regard to IL-1 family, it has been shown that IL-1α and IL-1β are expressed in ATL lesions caused by L. braziliensis. The expression of IL-1β in the lesions of L. braziliensis-infected LCL patients has been associated with immunopathology [53]. In opposite to humans, in mouse models, IL-1β has been shown important for protection in C57BL/6 mice infected with L. braziliensis or L. amazonensis [55]. Thus, the exact mechanism in how IL-32 isoforms modulates the production of proinflammatory cytokines and whether these connections contributes to immunopathology or wound healing remains to be investigated in leishmaniasis.
In conclusion, this study raises the concept of an important role for endogenous IL-32 in modulation of the inflammatory response in humans that influences susceptibility to and progression of localized cutaneous and mucosal leishmaniasis. In ATL, IL-32 is a novel target and may have promising therapeutic potential in the future. However, several aspects of the IL-32 biological function of IL-32 in human immune cells remain elusive. The precise molecular mechanisms mediated by IL-32 in lesions of Leishmania patients, including inflammatory pathways and control of parasite load, needs to be elucidated. Moreover, future studies should reveal whether the functional effects of genetic variations in IL32 observed in human PBMCs are also exerted in the ATL lesions environment.
Supporting information S1