IL6 174 G/C Promoter Polymorphism Influences Susceptibility to Mucosal but Not Localized Cutaneous Leishmaniasis in Brazil

Transcription

1 MAJOR ARTICLE IL6 174 G/C Promoter Polymorphism Influences Susceptibility to Mucosal but Not Localized Cutaneous Leishmaniasis in Brazil Léa Castellucci, 1,7 Eliane Menezes, 1 Joyce Oliveira, 1 Andrea Magalhães, 1 Luiz Henrique Guimarães, 1 Marcus Lessa, 1 Silvana Ribeiro, 1 Jeancarlo Reale, 1 Elza Ferreira Noronha, 3 Mary E. Wilson, 4 Priya Duggal, 5 Terri H. Beaty, 6 Selma Jeronimo, 2 Sarra E. Jamieson, 7 Ashlee Bales, 4 Jenefer M. Blackwell, 7 Amélia Ribeiro de Jesus, 1 and Edgar M. Carvalho 1 1 Serviço de Imunologia, Hospital Universitário Professor Edgard Santos, Universidade Federal da Bahia, Salvador, Bahia, 2 Centro de Biociências, Departamento de Bioquímica, Universidade do Rio Grande do Norte, Natal, Rio Grande do Norte, and 3 Universidade de Brasília, Distrito Federal, Brazil; 4 University of Iowa and the Veterans Affairs Medical Center, Iowa City; 5 Inherited Disease Research Branch, National Human Genome Research Institute, National Institutes of Health, and 6 Johns Hopkins School of Public Health, Baltimore, Maryland; 7 Cambridge Institute for Medical Research, University of Cambridge School of Clinical Medicine, Cambridge, United Kingdom Background. Mucosal leishmaniasis (ML) is associated with exaggerated tumor necrosis factor a and interferon-g responses and tissue destruction. ML follows localized cutaneous leishmaniasis (CL) caused by Leishmania braziliensis infection. Interleukin (IL) 6 down-regulates T helper (Th) cell type 1 differentiation and drives Th2 cell differentiation. The IL6 174 G/C polymorphism is associated with proinflammatory diseases and IL-6 regulation. Methods. The 174 G/C polymorphism was genotyped in population samples and families with CL and ML from Brazil. Genotype frequencies were compared among patients with ML, patients with CL, and 2 control groups by logistic regression and family-based association test (FBAT) analysis. IL-6 levels were measured in macrophages. Results. The C allele was more common in patients with ML than in patients with CL (odds ratio [OR], 2.55 [95% confidence interval {CI}, ]; P p.005), than in patients who were leishmanin skin-test positive (OR, 2.23 [95% CI, ]; P p.009), and than in neighborhood control subjects (OR, 2.47 [95% CI, ]; P p.01). FBAT analysis confirmed an association between allele C and ML under both additive ( z p 4.295; P p ) and dominant ( z p 4.325; P p ) models. Significantly lower levels of IL-6 were measured in unstimulated macrophages from CC individuals than from GG individuals ( P p.003) as well as after stimulation with soluble leishmania antigen ( P p.009). Conclusions. IL-6 may regulate type 1 proinflammatory responses, putting individuals with low macrophage IL-6 levels at increased risk for ML. Leishmaniasis is a protozoan parasitic infection transmitted by sand flies. Different species of Leishmania are Received 10 January 2006; accepted 9 March 2006; electronically published 10 July Potential conflicts of interest: none reported. Financial support: National Institutes of Health (NIH; grants AI and AI ); Howard Hughes Medical Institute; Department of Veterans Affairs. J.M.B. and S.E.J. are funded by the Wellcome Trust. A.R.d.J. and E.M.C. are funded by the Brazilian National Research Council and are members of the Insttituto de Investigação em Imunologia, Institutos do Milênio. J.O. and A.M. are funded by the Fogarty International Center, NIH (grant 1 D43 TW ). P.D. is supported by intramural research at the National Human Genome Research Institute, NIH. Reprints or correspondence: Dr. Amélia Ribeiro de Jesus, HUPES, Serviço de Imunologia, 5 andar, Rua João das Botas, s/n, Canela, Salvador, BA, Brazil (imuno@ufba.br or amelia@ufba.br). The Journal of Infectious Diseases 2006; 194: by the Infectious Diseases Society of America. All rights reserved /2006/ $15.00 associated with distinct clinical forms of disease. After L. braziliensis infection, most patients develop localized cutaneous leishmaniasis (CL) that can eventually selfheal, but most patients need antimony treatment. A fraction of patients with CL (!5%) develop mucosal disease, either at the time of ulceration or several months or years later. Mucosal leishmaniasis (ML) is associated with nasal, oral, pharyngeal, and/or laryngeal mucosal lesions that can be destructive and are difficult to treat using standard antimony therapy [1 4]. Immunologically, CL is associated with a predominantly type 1 immune response, in which the proinflammatory cytokines tumor necrosis factor (TNF) a and interferon (IFN) g [3, 5] mediate macrophage activation to control the parasite. In ML, the proinflammatory and IL6 Polymorphism and Mucosal Leishmaniasis JID 2006:194 (15 August) 519

2 Table 1. Comparison of demographic and epidemiological data for the case and control groups. Case patients Control subjects Parameter ML CL NC DTH positive Sex Male Female Age, years Mean % CI Time (years) in current municipality? Mean % CI Ever worked on a farm? No Yes Period of time (years) worked on a farm? Mean % CI Time (AM) start work on farm? Mean % CI Time (PM) finish work on farm? Mean % CI Occupation? Farmworker Domestic Child/student Other Do you have electricity? No Yes Do you hunt at night? No Yes Do you keep animals? No Yes Do you keep dogs? No Yes Do you keep chickens? No Yes Chicken coop near the house? No Yes If yes, distance (m) to coop? Mean % CI Pig sty near the house? No Yes (continued)

3 Table 1. (Continued.) Parameter If yes, distance (m) to pig sty? Case patients Control subjects ML CL NC DTH positive Mean % CI to Corral near the house? No Yes If yes, distance (m) to corral? Mean % CI to Farm near the house? No Yes If yes, distance (m) to farm? Mean % CI Forest near the house? No Yes If yes, distance (m) to forest? Mean 144.1a % CI NOTE. Data are no. of subjects, unless otherwise specified. CI, confidence interval; CL, cutaneous leishmaniasis; DTH positive, positive for the leishmanin delayed-hypersensitivity skin-test response; ML, mucosal leishmaniasis; NC, neighborhood community. a Distance from forest is significantly different ( P p.04, unpaired t test) between the ML and CL groups but not between the ML group and the NC or DTH-positive groups. type 1 cellular immune response is exaggerated [3, 5], leading to adverse pathological manifestations and tissue destruction. Modulation of the proinflammatory TNF-a response by combining pentoxifylline with antimony therapy leads to improved outcome [6], providing further support for the hypothesis that overproduction of TNF-a is detrimental in ML. Recent studies of leishmaniasis have focused on regulation of the cellular immune response and the need for an appropriate balance between proinflammatory and anti-inflammatory cytokines to achieve cure and avoid immune-mediated pathological damage or reactivation of disease [7, 8]. To date, attention has focused on the roles played by transforming growth factor (TGF) b and interleukin (IL) 10 in regulating the proinflammatory response. In particular, in human CL caused by L. braziliensis infection, the proinflammatory TNFa/IFN-g response is modulated in antigen-stimulated peripheral-blood mononuclear cells (PBMCs) by adequate levels of the immunomodulatory cytokine IL-10, whereas, in ML, high antigen-stimulated TNF-a/IFN-g levels are observed in the presence of low levels of IL-10 [5]. In patients with ML, both IL-10 and TGF-b show a decreased ability to modulate TNFa/IFN-g responses in cell culture, compared with cells from patients with CL. Previous studies have also shown that pretreatment with recombinant human IL-6 inhibits IFN-g and TNF-a mediated activation of human macrophages for killing of L. amazonensis [9], and IL-6 has been shown to downregulate the expression of TNF-a membrane receptors [10]. In addition, maxidilan, the vasodilatory peptide of the L. braziliensis sand-fly vector Lutzomyia longipalpis salivary glands, has immunosuppressive properties associated with the induction of both IL-10 and IL-6 production in vitro [11]. Hence, it is possible that IL-6 may also contribute to the immune modulation of proinflammatory TNF-a/IFN-g responses. IL-6 has traditionally been thought of as a proinflammatory cytokine. It is produced by cells from innate and adaptive immune responses and contributes to differentiation and/or activation of macrophages and T cells, growth and terminal differentiation of B cells, and support of multipotential colony formation by hematopoietic cells [12]. However, IL-6 is also produced by professional antigen-presenting cells and drives the differentiation of CD4 + Th2 cells by the early induction of IL-4 in CD4 + T cells [13]. A recent study demonstrated the mechanism by which IL-6 induces Th2 cell differentiation it induces IL-4 gene expression by up-regulating the transcrip- IL6 Polymorphism and Mucosal Leishmaniasis JID 2006:194 (15 August) 521

4 tional activity of nuclear factor of activated T cells (NFAT) and increases levels of NFATC2. This finding was confirmed by the inability of IL-6 to promote Th2 cell differentiation in CD4 + T cells lacking NFATC2 [13]. Moreover, this cytokine acts as a negative regulator of CD4 + Th1 cell differentiation by upregulating suppressor of cytokine signaling 1 expression in activated CD4 + T cells, specifically by interfering with signal transducer and activator of transcription 1 phosphorylation to prevent autoregulation by IFN-g during CD4 + T cell activation and Th1 differentiation [14, 15]. Thus, IL-6 promotes CD4 + Th2 cell differentiation and inhibits Th1 cell differentiation by 2 independent molecular mechanisms. These potential immunoregulatory properties of IL-6 led us to reconsider its role in CL and ML, in relation to both genetic polymorphisms in IL6 that might modulate disease and its influence on the regulation of IL-6 production. The human IL6 gene maps to chromosome band 7p21. A single-nucleotide polymorphism (SNP) described at nucleotide position 174 of the promoter region of the IL6 gene has been associated with functional differences in IL-6 levels [16, 17] and with a variety of inflammatory diseases [18]. The aim of the present study was to evaluate the role played by the IL6 174 SNP in ML and CL, using both case-control and familybased sampling from the area of Corte de Pedra, Bahia, Brazil, where L. braziliensis is endemic. MATERIALS AND METHODS Case patients, control subjects, and study design. The present study was conducted in the area of Corte de Pedra, Bahia, Brazil, where L. braziliensis is endemic. Although all case patients included in the present study are past case patients from whom parasites were not isolated, Corte de Pedra has long been known as an area where L. braziliensis infection predominates [19, 20], and L. braziliensis is the only species that has been identified in all new cases of CL and ML in this endemic area during the past 5 years. The Corte de Pedra area comprises 20 municipalities in a total area of 10,000 km 2 around the Corte de Pedra Public Health Post, the referral center for treatment of leishmaniasis in that region. Corte de Pedra is in a region of rural rain forest, where agriculture underpins the local economy. For the present genetics study, both case-control and family-based cohorts were studied. Index cases of ML were ascertained from medical records of the Corte de Pedra Public Health Post, and the families and neighborhoods of the patients with these cases were revisited, to establish the study population. Informed consent was obtained from all of the participants, and the research was approved by the ethics committee of the Hospital Universitário Professor Edgard Santos. All participants in the study were enrolled between January 2001 and November All were of equivalent socioeconomic status (table 1). Initially, 60 index case patients with ML (47 males and 13 females; mean SD age, years) were selected and matched by age and sex to 60 unrelated case patients with CL (47 males and 13 females; mean SD age, years) and 60 unrelated neighborhood control (NC) subjects (47 males and 13 females; mean SD age, years). These NC subjects had no clinical history of disease or leishmaniasis scars, but their leishmanin skin-test status was unknown. For this reason, a second control group was included in the study, composed of 60 age- and sex-matched unrelated individuals (47 males and 13 females; mean SD age, years) from another site within the Corte de Pedra area of endemicity who had been recently screened and were known to be positive for the leishmanin delayed-hypersensitivity skin-test response (hereafter, DTH positive ). The advantage of this second control group is that they had been infected but had no clinical symptoms of disease. This second site was geographically and demographically equivalent to the first neighborhood studied (table 1). The only significant difference in environmental risk factors was that the case patients with ML lived significantly ( P p.04) closer to forest than did the case patients with CL, although they did not live closer than the NC or DTH-positive control subjects. The 60 index case patients with ML were also used to ascertain a total of 67 multicase-leishmaniasis (mixed for CL and ML) pedigrees (101 nuclear families; table 2), providing a total of 91 ML cases (i.e., 31 additional cases) and 223 CL cases (exclusive of the 60 CL cases used in the case-control study). Unaffected members of the pedigrees were included in the study to contribute genotype information, to increase the statistical power of the family-based association test (FBAT) analysis, especially for families with missing parents. Full demographic and epidemiological information in relation to the multicase families used in this study have been presented elsewhere [21]. Sample collection and DNA extraction. Blood (8 ml) was taken by venipuncture and collected into dodecyl citrate acid containing Vacutainers (Becton Dickinson). Genomic DNA was prepared using the proteinase K and salting-out method [22]. The IL6 174 G/C promoter polymorphism was genotyped by polymerase chain reaction (PCR) with sequence-specific primers, using a commercially available kit (One Lambda). PCR was performed using a Perkin-Elmer 9700 Thermocycler, in accordance with the manufacturer s instructions, and PCR products were visualized by agarose gel electrophoresis. IL-6 ELISAs. After genotyping, a subgroup of individuals ( n p 12/genotype) with different IL6 174 G/C genotypes were selected for measurement of IL-6 production by adherencepurified macrophages, with or without stimulation with soluble leishmania antigen (SLA) from L. braziliensis [23] or lipopolysaccharide (LPS; Sigma). To ensure that the TNF-a 308 G/A polymorphism, for which the rare A allele has previously been shown to be associated with ML [24], did not influence IL JID 2006:194 (15 August) Castellucci et al.

5 Table 2. Family structure for multicase families with leishmaniasis caused by Leishmania braziliensis infection from the study area in Corte de Pedra, Bahia, Brazil. Category Leishmaniasis CL ML Families 67 Nuclear families 101 Nuclear families with 1 affected sibling affected siblings affected siblings affected siblings affected siblings affected siblings Affected offspring Affected parents Total affected individualsa NOTE. Data are no. of families or individuals. The distribution of disease in nuclear families is further stratified by cutaneous leishmaniasis (CL) and mucosal leishmaniasis (ML). a The total no. of affected individuals is not the sum of the no. of affected offspring plus the no. of affected parents because some affected individuals are both offspring and parents. production in our study, the individuals selected for the macrophage assays were all homozygous for the common G allele. PBMCs were obtained from heparinized peripheral blood by ficoll-hypaque gradient (Organon Tecknika). Individuals selected for the IL-6 assays were a mixture of cured case patients and NC subjects for each genotype. Macrophages were derived from PBMCs by adherence to petri dishes, as described elsewhere [5]. Briefly, PBMCs in RPMI 1640 supplemented with 10% fetal calf serum (FCS) were plated onto plastic petri dishes (10-cm diameter; Falcon; Becton Dickinson) and incubated for 40 min at 37 C. Nonadherent cells were removed by 2 washes with warm saline solution, and cold saline was added to remove monocytes. After centrifugation, cells were resuspended in RPMI 1640 with 10% FCS, the cell concentration was adjusted to cells/ml, and the adherent-cell subpopulation was plated onto 24-well plates. These enriched subpopulations were incubated with medium alone or stimulated with 10 mg/ml SLA or 10 mg/ml LPS for 24 h, which were determined to be the optimal conditions after preliminary experiments comparing incubation times (12 and 24 h) and concentrations of SLA (1 and 10 mg) and LPS (0.1, 1, and 10 mg/ml). IL-6, TNF-a, and IL-10 levels were measured using an ELISA sandwich kit (for IL-6, R&D Systems; for TNF-a, Duo Set from R&D Systems; and for IL-10, Duo Set from Pharmingen). Standards included serial double dilutions of recombinant human IL-6, TNF-a, and IL-10. Statistical analyses. A test for deviation from Hardy-Weinberg equilibrium was performed on a set of unrelated individuals selected from the families (parents and individuals marrying into the pedigree). Tests for Hardy-Weinberg equilibrium were done using STATA (version 8.2; available at: with the freely available GenAssoc package (available at: stata/). Logistic regression analysis was performed using STATA to determine allele-wise (1 df test) and genotype-wise (2 df test) associations at the IL6 174 SNP, comparing the ML, CL, NC, and DTH-positive groups. Global test statistics were generated for both the 1 df test and the 2 df test, and odds ratios (ORs) with 95% confidence intervals (CIs) were computed to compare the risk of ML and CL for the variant C allele relative to the common G allele. A likelihood ratio test comparing the 1 and 2 df tests provided a test for dominance effects. PedCheck [25] was used to determine and delete Mendelian inconsistencies within families. Family-based allelic-association tests were done using FBAT [26, 27] under the null hypothesis of no linkage or no association and using the offset value obtained via the o option in the fbat command, to generate an offset value that minimizes the variance of the test statistic. This option maximizes the statistical power of the test by including both affected and unaffected individuals in the analysis. FBAT allelicassociation analyses were done under both additive and dominant models. For the functional IL-6 assays, an unpaired Mann-Whitney U test was performed to test for statistical differences between the 3 genotypic groups. Tests were considered to be statistically significant if the probability of a type I error was!0.05. Bioinformatics. Inspection of the IL6 gene promoter region for putative transcription factor binding sites was undertaken for sequences bearing the alternative 174GorC allele using AliBaba (version 2.1; available at: and MatInspector (version 8; available at: RESULTS Population-based analysis of the IL polymorphism. There was no evidence of deviation from Hardy-Weinberg equi- Table 3. Allele and genotype distributions for study group comparisons. Category Genotype Case patients Control subjects ML CL NC DTH positive GG 18 (31) 34 (57) 29 (49) 38 (64) GC 36 (61) 24 (40) 30 (51) 15 (26) CC 5 (8) 2 (3) 0 (0) 6 (10) Allele G 54 (70) 80 (85) 73 (83) 83 (86) C 23 (30) 14 (15) 15 (17) 13 (14) NOTE. Data are no. (%) of subjects. CL, cutaneous leishmaniasis; DTH positive, positive for the leishmanin delayed-hypersensitivity skin-test response; ML, mucosal leishmaniasis; NC, neighborhood community. IL6 Polymorphism and Mucosal Leishmaniasis JID 2006:194 (15 August) 523

6 Table 4. Results of logistic regression analyses for study group comparisons. Comparison Global test 2 df 1 df IL6 174 allele OR (95% CI) P (CL + ML) vs. (NC + DTH positive) ML vs. non-ml (CL + NC + DTH positive) C vs. G 2.29 ( ).001 ML vs. (NC + DTH positive) C vs. G 2.25 ( ).003 ML vs. CL C vs. G 2.55 ( ).005 ML vs. NC a C vs. G 2.47 ( ).01 ML vs. DTH positive C vs. G 2.23 ( ).009 CL vs. NC a C vs. G 0.86 ( ).67 CL vs. DTH positive C vs. G 1.02 ( ).94 NOTE. The 2 df test represents the genotype-wise test, and the 1 df test represents the allele-wise test. CI, confidence interval; CL, cutaneous leishmaniasis; DTH positive, positive for the leishmanin delayed-hypersensitivity skin-test response; ML, mucosal leishmaniasis; NC, neighborhood community; OR, odds ratio. a The 2 df test was invalid, because of a zero value for CC genotype in the NC control group. librium using unrelated individuals in these families. Table 3 presents the frequency distribution for genotypes and alleles in the different population-based patient and control groups. To determine initially whether the IL6 174 SNP was associated with susceptibility to leishmaniasis per se, we first compared the ML and CL groups against the NC and DTH-positive groups (table 4). This comparison was not significant by the global allele-wise or genotype-wise tests, indicating that the IL6 174 SNP was not a gene for susceptibility to leishmaniasis per se in this population. In contrast, when the ML group was compared with the 3 non-ml groups, either individually or pooled, significant allele-wise and genotype-wise (where valid) associations were observed. No dominance effects were observed in the logistic regression analysis, indicating that a multiplicative model explains these data. In all comparisons, the OR for ML was significantly associated with the C allele at the IL6 174 SNP, with similar ORs, P values, and 95% CIs for each comparison (table 4). Notably, the OR for ML was 2.55 (95% CI, ; P p.005) compared with the CL group, 2.47 (95% CI, ; P p.01) compared with the NC group, and 2.23 (95% CI, ; P p.009) compared with known DTH-positive control subjects. These results indicate that the IL6 174 SNP is specifically associated with susceptibility to ML disease. This conclusion is further supported by the failure to observe a statistically significant association when the CL group was compared with either the NC or the DTHpositive control group (table 4). Confirmation of the association between the IL SNP and ML by FBAT analysis. One concern with genetic analysis using a case-control design in this Brazilian population is that ethnic admixture could lead to false-positive results. Although the low prevalence of ML precluded sampling of a completely independent patient group for replication, the 60 index cases were used to ascertain a total of 67 multicase families, which also included an independent sample of cases of CL (table 2). We undertook further genotyping of all members of these pedigrees, thus increasing the power of our FBAT analysis to evaluate transmission of the C or G allele at the IL6 174 SNP from heterozygous parents to ML- or CL-affected offspring. This analysis confirmed the association between the C allele and ML under both additive ( z p 4.295; P p ) and dominant ( z p 4.325; P p ) models. No association was observed for CL alone, further confirming our case-control findings. Association between genotypic variation at the IL SNP and differences in baseline IL-6 cytokine release by macrophages. To compare, in vitro, IL-6 production between individuals with known IL6 174 SNP genotypes, we measured cytokine levels in unstimulated macrophages and in macrophages stimulated with SLA or LPS. For unstimulated macrophages, we observed significantly ( P p.003) less IL-6 release in macrophages from CC individuals than from GG individuals, with cells from GC individuals having an intermediate phenotype (figure 1A). A similar result was observed for SLAtreated macrophages ( P p.009), and a trend was observed for LPS-treated macrophages ( P p.057). However, the CC and GC genotypic groups clearly had a stronger response to LPS stimulation than did the GG group, as was evidenced by stimulation indices (LPS values divided by unstimulated values) of , , and for the GG, GC, and CC genotypes, respectively. Baseline levels of IL-10 and TNFa were low in individuals of all 3 genotypes and remained low after SLA stimulation (figure 1B and 1C). LPS enhanced both IL-10 and TNF-a levels, but there were no significant differences between the IL6 174 SNP genotypes (figure 1B and 1C). These results suggest that IL-6 was not acting in a direct feedback loop to influence macrophage pro- or anti-inflammatory responses. Bioinformatics. Analysis using AliBaba confirmed the presence of a nuclear factor (NF) 1 binding site introduced by 524 JID 2006:194 (15 August) Castellucci et al.

7 Figure 1. Interleukin (IL) 6 (A), tumor necrosis factor (TNF) a (B), and IL-10 (C) cytokine production in unstimulated (medium), soluble leishmania antigen (SLA) stimulated, and lipopolysaccharide (LPS) stimulated macrophages from individuals carrying the GG, GC, or CC genotype at the IL6 174 promoter region polymorphism. Bars represent the median response. the 174 C variant in the IL6 promoter, as has been alluded to previously [16]. Using Mat-Inspector, we found that the 174 C variant also introduces a perfect Smad4 binding site. DISCUSSION The data presented here provide both population-based and family-based evidence for an association between the C allele of the 174 SNP at IL6 and susceptibility to ML. In our study, the C allele was associated with reduced baseline expression of IL-6 in unstimulated macrophages and in macrophages stimulated with SLA. Because macrophages are the primary site of infection, we hypothesize that low IL-6 production in carriers of the C allele may contribute to a reduced capacity to induce Th2 cell differentiation and regulate the activity of CD4 + Th1 cell generated cytokines (such as IFN-g and TNF-a) that contribute to the destructive pathological manifestations associated with ML. In contrast, the high baseline IL-6 production observed in unstimulated and SLA-stimulated macrophages from GG individuals protects by contributing to the regulation of the proinflammatory response, since IL-6 produced early after Leishmania infection during antigen presentation up-regulates IL-4, which promotes Th2 cell and inhibits Th1 cell differentiation. Previous studies have also identified associations between the IL6 174 SNP and a range of inflammatory conditions and autoimmune diseases [28 32]. However, there are inconsistencies among these studies, both in terms of the disease-associated allele and when attempting to determine whether different genotypes are functionally associated with the production of different IL-6 levels. The basis of these inconsistencies is likely 2-fold: (1) because IL-6 might contribute differently to the pathological manifestations of these diseases due to its pleiotropic effects in regulating both type 1 and type 2 immuneresponse pathways, plus the complexities of the immunopathogenesis of these different diseases [14, 33]; and (2) because the 174 SNP is not the sole polymorphic determinant of differential and cell type specific promoter activity driving IL6 gene transcription [16, 17]. In relation to our own study, it is notable that reporter-gene studies in HeLa cells using constructs that incorporate the 174 G versus C alleles in isolation from IL6 Polymorphism and Mucosal Leishmaniasis JID 2006:194 (15 August) 525

8 other known polymorphic variants in the promoter region have demonstrated lower levels of transcription of the C allele [16]. This is consistent with the observation that the C allele generates potential binding sites for the transcription factors NF- 1 and Smad4, which are not present with the G allele sequence. NF-1 comprises a family of transcription factors that are variously active in many cell types; however, in HeLa cells, NF-1 acts as repressor of gene expression [34]. Smad4 is a central component of a family of proteins that serve as intracellular mediators to regulate TGF-b superfamily signaling [35]. Although not previously associated with IL6 gene transcription, Smad4 does act as a transcriptional inhibitor for target genes of the IL-6 induced pathway [36]. Introduction of a Smad4 binding site in the IL6 promoter itself could contribute to the repressed production of IL-6 we observed in macrophages from individuals carrying the C allele. Associations between the IL6 174 SNP and different diseases and phenotypes might also relate to cell type specific effects observed in studies of reporter-gene expression [17] and, in turn, to cell type specific expression of these transcription factors. Other polymorphic elements, in particular a variable AnTn tract at 392 to 373, also influence transcription of 174 G allele and 174 C allele carrying haplotypes [17], suggesting that analysis of more-complex regulatory haplotypes extending further upstream of the 174 promoter may be required to properly define disease associations [37]. Terry et al. found that reporter-gene constructs carrying variable AnTn tract repeats on the background of the AGG hapotype for 597 G/A, 572 G/C, and 174 G/C SNPs showed variable reporter-gene activity [17] that is, that the AnTn tract itself could influence IL6 gene transcription. Interestingly, however, the AnTn tract (A8T12, usually present on an AG8/12C haplotype) that had the biggest experimental effect on AGG-driven (i.e., AG8/12G-driven) transcription is rare (frequency, 0.01). Conversely, the AnTn tracts (A10T11 and A9/T11) most commonly found in a haplotype with the 174 G allele are either absent (A10T11) or rare (A9/T11; frequency, 0.01) on haplotypes with the 174 C allele [17]. Hence, in our study, the 174 SNP is likely to have captured the important association with ML, even if there is collaboration with upstream elements driving IL6 gene transcription. In this respect, it is interesting that the repressive activity of the 174 C allele observed here in macrophages was completely overcome by the activation of LPS signaling pathways, the major transcription factor binding sites (NF-kB) for which lie 100 bp downstream of the 174 position in the IL6 promoter. Although LPS signaling itself would occur only in patients with ML who have secondary infections, Leishmania parasites are known to trigger the Toll-like receptor 2 induced NF-kB pathway [38], and TNF-a and IL-1b, the proinflammatory cytokines associated with ML pathogenesis, are themselves potent triggers of the NF-kB pathway. Hence, morecomplex interactions during infection in vivo may modulate the role played by the IL6 174 SNP as a risk factor for ML. Nevertheless, our studies have demonstrated that carrying allele C for this polymorphism is an important risk factor for ML. Further functional characterization of this and other polymorphic immunomodulatory proteins in larger numbers of patients particularly during active disease and in relation to age, sex, and disease severity should help our understanding of the molecular pathogenesis of this disease. Acknowledgments We thank Ednaldo Lima do Lago and other people from the Corte de Pedra Public Health Post, for helping to recruit the patients; Drs. Paulo Machado, Roque Almeida, and Albert Schriefer, for helping to diagnose the patients; and Elbe M. Silva and Lúcia Reis, for secretarial assistance. References 1. Carvalho EM, Johnson WD, Barreto E, et al. Cell mediated immunity in American cutaneous and mucosal leishmaniasis. J Immunol 1985; 135: Castes M, Agnelli A, Rondon AJ. Mechanisms associated with immunoregulation in human American cutaneous leishmaniasis. Clin Exp Immunol 1984; 57: Castes M, Trujillo D, Rojas ME, et al. Serum levels of tumor necrosis factor in patients with American cutaneous leishmaniasis. Biol Res 1993; 26: Marsden PD, Lessa HA, Oliveira MR, et al. Clinical observations of unresponsive mucosal leishmaniasis. Am J Trop Med Hyg 1998; 59: Bacellar O, Lessa H, Schriefer A, et al. Up-regulation of Th1-type responses in mucosal leishmaniasis patients. Infect Immun 2002; 70: Lessa HA, Machado P, Lima F, et al. Successful treatment of refractory mucosal leishmaniasis with pentoxifylline plus antimony. Am J Trop Med Hyg 2001; 65: Belkaid Y, Piccirillo CA, Mendez S, Shevach EM, Sacks DL. CD4+CD25+ regulatory T cells control Leishmania major persistence and immunity. Nature 2002; 420: Mendez S, Reckling SK, Piccirillo CA, Sacks D, Belkaid Y. Role for CD4 + CD25 + regulatory T cells in reactivation of persistent leishmaniasis and control of concomitant immunity. J Exp Med 2004; 200: Hatzigeorgiou DE, He S, Sobel J, Grabstein KH, Hafner A, Ho JL. IL- 6 down-modulates the cytokine-enhanced antileishmanial activity in human macrophages. J Immunol 1993; 151: Bermudez LE, Wu M, Petrofsky M, Young LS. Interleukin-6 antagonizes tumor necrosis factor-mediated mycobacteriostatic and mycobactericidal activities in macrophages. Infect Immun 1992; 60: Soares MB, Titus RG, Shoemaker CB, David JR, Bozza M. The vasoactive peptide maxadilan from sand fly saliva inhibits TNF-alpha and induces IL-6 by mouse macrophages through interaction with the pituitary adenylate cyclase-activating polypeptide (PACAP) receptor. J Immunol 1998; 160: Van Snick J. Interleukin-6: an overview. Annu Rev Immunol 1990;8: Diehl S, Chow CW, Weiss L, et al. Induction of NFATc2 expression by interleukin 6 promotes T helper type 2 differentiation. J Exp Med 2002; 196: Diehl S, Anguita J, Hoffmeyer A, et al. Inhibition of Th1 differentiation by IL-6 is mediated by SOCS1. Immunity 2000; 13: JID 2006:194 (15 August) Castellucci et al.

9 15. Diehl S, Rincon M. The two faces of IL-6 on Th1/Th2 differentiation. Mol Immunol 2002; 39: Fishman D, Faulds G, Jeffery R, et al. The effect of novel polymorphisms in the interleukin-6 (IL-6) gene on IL-6 transcription and plasma IL-6 levels, and an association with systemic-onset juvenile chronic arthritis. J Clin Invest 1998; 102: Terry CF, Loukaci V, Green FR. Cooperative influence of genetic polymorphisms on interleukin 6 transcriptional regulation. J Biol Chem 2000; 275: Bidwell J, Keen L, Gallagher G, et al. Cytokine gene polymorphism in human disease: on-line databases. Genes Immun 1999; 1: Cuba-Cuba CA, Marsden PD, Barreto AC, et al. Identification of human stocks of Leishmania spp. isolated from patients with mucocutaneous leishmaniasis in Três Braços, Bahia, Brazil. Trans R Soc Trop Med Hyg 1984; 78: Rosa AC, Cuba CC, Vexenat A, Barreto AC, Marsden PD. Predominance of Leishmania braziliensis braziliensis in the regions of Tres Bracos and Corte de Pedra, Bahia, Brazil. Trans R Soc Trop Med Hyg 1988; 82: Castellucci L, Cheng LH, Araujo C, et al. Familial aggregation of mucosal leishmaniasis in northeast Brazil. Am J Trop Med Hyg 2005; 73: Sambrook J, Fritsch EF, Maniatis T. Molecular cloning: a laboratory manual. Vol. 2. Cold Spring Harbor, New York: Cold Spring Harbor Laboratory Press, Carvalho LP, Passos S, Dutra WO, et al. Effect of LACK and KMP11 on IFN-g production by peripheral blood mononuclear cells from cutaneous and mucocutaneous leishmaniasis patients. Scand J Immunol 2005; 61: Cabrera M, Shaw M-A, Sharples C, et al. Polymorphism in TNF genes associated with mucocutaneous leishmaniasis. J Exp Med 1995; 182: O Connell JR, Weeks DE. PedCheck: a program for identification of genotype incompatibilities in linkage analysis. Am J Hum Genet 1998; 63: Laird NM, Horvath S, Xu X. Implementing a unified approach to family-based tests of association. Genet Epidemiol 2000; 19(Suppl 1): S Horvath S, Xu X, Laird NM. The family based association test method: strategies for studying general genotype-phenotype associations. Eur J Hum Genet 2001; 9: Herbert A, Liu C, Karamohamed S, et al. The -174 IL-6 GG genotype is associated with a reduced risk of type 2 diabetes mellitus in a family sample from the National Heart, Lung and Blood Institute s Framingham Heart Study. Diabetologia 2005; 48: Sawczenko A, Azooz O, Paraszczuk J, et al. Intestinal inflammationinduced growth retardation acts through IL-6 in rats and depends on the -174 IL-6 G/C polymorphism in children. Proc Natl Acad Sci USA 2005; 102: Pawlik A, Wrzesniewska J, Florczak M, Gawronska-Szklarz B, Herczynska M. IL-6 promoter polymorphism in patients with rheumatoid arthritis. Scand J Rheumatol 2005; 34: Bittar MN, Carey JA, Barnard J, et al. Interleukin 6 G-174C polymorphism influences outcome following coronary revascularization surgery. Heart Surg Forum 2005; 8:E140 5; discussion E Testa R, Olivieri F, Bonfigli AR, et al. Interleukin G1C polymorphism affects the association between IL-6 plasma levels and insulin resistance in type 2 diabetic patients. Diabetes Res Clin Pract 2005; 71: Rincon M, Anguita J, Nakamura T, Fikrig E, Flavell RA. Interleukin (IL)-6 directs the differentiation of IL-4-producing CD4+ T cells. J Exp Med 1997; 185: Rein T, Forster R, Krause A, Winnacker EL, Zorbas H. Organization of the alpha-globin promoter and possible role of nuclear factor I in an alpha-globin-inducible and a noninducible cell line. J Biol Chem 1995; 270: Grady WM. Transforming growth factor-beta, Smads, and cancer. Clin Cancer Res 2005; 11: Zauberman A, Lapter S, Zipori D. Smad proteins suppress CCAAT/ enhancer-binding protein (C/EBP) beta- and STAT3-mediated transcriptional activation of the haptoglobin promoter. J Biol Chem 2001; 276: Fife MS, Ogilvie EM, Kelberman D, et al. Novel IL-6 haplotypes and disease association. Genes Immun 2005; 6: de Veer MJ, Curtis JM, Baldwin TM, et al. MyD88 is essential for clearance of Leishmania major: possible role for lipophosphoglycan and Toll-like receptor 2 signaling. Eur J Immunol 2003; 33: IL6 Polymorphism and Mucosal Leishmaniasis JID 2006:194 (15 August) 527