IL-10 and TNFα Genotypes in SLE

The production of two regulators of the inflammatory response, interleukin 10 (IL-10) and tumor necrosis factor α (TNFα), has been found to be deeply deregulated in SLE patients, suggesting that these cytokines may be involved in the pathogenesis of the disease. Genetic polymorphisms at the promoter regions of IL-10 and TNFα genes have been associated with different constitutive and induced cytokine production. Given that individual steady-state levels of these molecules may deviate an initial immune response towards different forms of lymphocyte activation, functional genetic variants in their promoters could influence the development of SLE. The present review summarizes the information previously reported about the involvement of IL-10 and TNFα genetic variants on SLE appearance, clinical phenotype, and outcome. We show that, in spite of the heterogeneity of the populations studied, the existing knowledge points towards a relevant role of IL-10 and TNFα genotypes in SLE.


Introduction
In spite of its unknown etiology, it is accepted that genetics and environmental factors contribute to systemic lupus erythematosus (SLE) susceptibility and outcome. The levels of various cytokines have been found elevated in SLE patients; so they have been considered essential elements in the etiopathology of the disease. Given that the production of these molecules is controlled at genetic level, functional polymorphisms in their promoters could influence the development and severity of the disease. In particular, the production of interleukin 10 (IL-10) and tumor necrosis factor α (TNFα), two mutually regulated cytokines that play complex and predominantly opposite roles in systemic inflammatory responses, has been found to be deregulated in SLE patients (Figure 1). Besides its stimulated production, various cell types are constitutively capable of producing detectable amounts of these cytokines, mainly cells of myeloid origin and less abundantly T and B lymphocytes. It has been reported that individual steady-state levels of these molecules may deviate an initial immune response towards different forms of T cell activation, influencing the likelihood to transform a limited autoimmune response into an autoimmune disease.
Several evidences suggest that IL-10 could be a strong candidate gene influencing SLE susceptibility. IL-10 is an important immunoregulatory cytokine that inhibits T cell function by suppressing the expression of proinflammatory cytokines such as TNFα, IL-1, IL-6, IL-8, and IL-12 [1, 2]. It also inhibits antigen presenting cells by downregulating major histocompatibility complex class II (MHC-II) and B7 expression [3]. In addition to these inhibitory actions, IL-10 promotes B-cell-mediated functions, enhancing survival, proliferation, differentiation, and antibody production [4]. Hence, increased production of IL-10 could thus explain B cell hyperactivity and autoantibody production, two main features of the immune dysregulation in SLE. In fact, elevated levels of this molecule have been currently reported in SLE patients, frequently associated with indicators of disease activity [5,6]. Moreover, it has been demonstrated that IL-10 plays an important role in murine lupus. Ishida et al. [7] reported that continuous administration of anti IL-10 antibodies in the and TNFα in lupus disease. Both cytokines are produced by multiple cells types of the innate and adaptative immune system, in particular dendritic cells (DCs), monocytes/macrophages, and specific effector T cells. Th1 cells produce the proinflammatory cytokine TNFα which activates DCs and other antigen presenting cells (APCs), and induces the production of IL-10. In addition, TNFα promotes inflammation and apoptosis, generating neoantigens that could result in autoantibody production. On the other hand, IL-10, a Th2 cytokine, antagonize Th1 differentiation and inhibits APCs and T cells. Conversely, IL-10 is a potent stimulator of B cell proliferation, differentiation and antibody production. Thus, B cell activation in presence of neo-antigens may led to autoantibody secretion and immune complexes formation, thus resulting in tissue damage affecting diverse organs. STAT; signal transducer and activator of transcription. murine lupus model New Zealand black/white (NZB/W) F1 delayed the onset of autoimmunity and improved the survival rate from 10 to 80%. Interestingly, Llorente et al.
[8] demonstrated that constitutive IL-10 production by monocytes and B cells in healthy members of multicase families with SLE was significantly higher than that of healthy unrelated controls, but was similar to that of SLE patients, thus suggesting that a genetically controlled high innate IL-10 production may predispose to SLE development.
In the same way, TNFα is a well-known cytokine for its role in the regulation of inflammation and apoptosis, two processes involved in the pathogenesis of SLE. This molecule stimulates the production of inflammatory cytokines, enhances neutrophil activation and expression of adhesion molecules and acts as a costimulator for Tcell activation and antibody production. Accordingly, in vivo and in vitro studies demonstrated that high levels of TNFα lead to exacerbation of the inflammatory response.
These effects, together with its potent immunomodulator activities [9-11], could support the involvement of TNFα in the pathogenesis of SLE [12]. However, initial studies in murine models of SLE showed contradictory results, probably because different strains of lupus prone mice have different phenotypic features. Thus, whereas the (NZB/W) F1 strain produces relatively low level of TNFα and treatment with recombinant TNFα caused a significant delay in the onset of nephritis and an improved survival rate [13], MRLlpr/lpr and BXSB strains constitutively produce relatively high amounts of TNFα, being its effect deleterious in the outcome of the disease [14].
Nevertheless, a number of studies showed higher serum levels of TNFα in SLE patients compared with controls, which were frequently linked to SLE activity [15] or to specific immunological or clinical features, such as elevated autoantibody production [10, 11, 16]. All these data lead to suspect that TNFα has an important role in SLE susceptibility or outcome [17, 18].

Functional IL-10 and TNFα Genetic Polymorphisms
Human IL-10, encoded by a gene located at chromosome 1, is secreted by a variety of cell types in response to several activation stimuli. This cytokine could be also constitutively produced at low levels by immune cells, mainly monocytes, macrophages and dendritic cells. In fact, in contrast to many other cytokines, the synthesis of IL-10 is regulated by the transcription factors Sp1 and Sp3, which are constitutively expressed by different cell types [19]. The striking interindividual differences detected in IL-10 levels at both in vivo constitutive and in vitro following cellular stimulation [20], suggest that its production is genetically controlled. A number of genetic polymorphisms at the promoter region of the IL-10 gene have been reported, some of them associated with different constitutive and induced cytokine production. Among them, the most widely studied include two areas of multiple (CA)n repeats, the microsatellites IL10.R (−4 Kb) and IL10.G (−1.1 Kb) [21,22] with probable association with IL-10 production, and three single nucleotide polymorphisms (SNPs) located at −1082 (G/A), −819 (C/T) and −592 (C/A) positions upstream of the transcription start site [23]. A complete linkage disequilibrium exists between the alleles present at −819 and −592 positions; so these polymorphisms occurred in tandem and only three haplotypes have been found in Caucasian populations (GCC, ACC and ATA). These SNPs have been associated with variability in IL-10 production [24][25][26] and carriers of the GCC/GCC genotype are considered as genetically high producers, being −1082G the most relevant allele [24,27,28] (Table 1) [29][30][31][32][33][34][35][36][37][38][39][40][41][42][43][44][45][46]. The gene encoding TNFα is located at the MHC class III region, placed on chromosome 6p21. Similarly to IL-10, an important genetic diversity at the TNFα promoter has been detected. In vitro studies indicated that TNFα production varied among different alleles of the five microsatellite markers described (a, b, c, d and e). In addition, several SNPs have been identified [47][48][49], being −308 G/A and −238 G/A the most extensively examined. The polymorphic variant −238A is associated with DR3 and DR7 in extended haplotypes [50,51], but no consistent data about their functionality were reported. Polymorphism present at position −308, identified by Wilson et al. [49], has been associated with different levels of cytokine production. The less common TNF2 allele (−308A) has been related to higher TNFα transcription rate than the TNF1 allele (−308G) after in vitro activation of lymphocytes with different stimuli [52,53]. In vivo studies on mRNA constitutive levels confirmed this association [54] ( Table 1). TNF2 is part of the extended haplotype HLA-A1-B8-DR3-DQ2 [55], associated with high TNFα production [56,57] and with predisposition to several autoimmune diseases. Nevertheless, carriage of TNF2 allele-in the presence or absence of other loci-leads to an increase in TNFα production that could modify cytokine homeostasis in favor of the development of pathogenic situations.

IL-10 Genetic Polymorphisms and SLE Susceptibility
The IL-10 gene is situated in a major SLE susceptibility locus (1q31-32) [58]. However, in spite of the considerable number of genetic studies performed, no definitive result about its involvement in SLE susceptibility was achieved. Some works showed significant associations between IL-10 microsatellites or SNPs with SLE susceptibility or with the development of certain clinical or immunological features, while other studies indicated that these polymorphisms did not appear to have any relevance in the disease ( (with a CA repeat number greater than 21) were significantly increased in Mexican-American [70], Italian [30,66] and British [71] patients respectively. On the contrary, an increase in IL10.G4 (short allele) was reported in Chinese patients [29] whereas no significant differences in IL10.G alleles were detected in other cohorts [65,68,72,73]. In addition, a meta-analysis study showed only association of the IL10.G11 allele with SLE susceptibility in the populations analyzed (OR = 1.279; 95% CI: 1.027-1.593; P = .028) [62]. It has been reported that LPS-stimulated cells from individuals carriers of the IL10.G allele with 26 CA repeats presented higher IL-10 production than those from carriers of short alleles [25], suggesting that long alleles might be responsible for a high IL-10 production. Thus, accordingly to these data, high IL-10 producer genotypes (with more than 21 CA repeats) could be associated with SLE susceptibility, while presence of short alleles could confer a protective effect [60,66].
Conflicting results were also obtained after examining the possible association between SLE susceptibility and SNPs at −1082, −819 and −592 positions of IL-10 gene in the different populations in which they were investigated. The frequency of high IL-10 producers (carriers of −1082G allele or GCC haplotype) was found to be increased in several works with Asian [62,74] or European [59,75] patients,

TNFα Genetic Polymorphisms and SLE Susceptibility
Less controversial data exist with regard to TNFα SNPs, since genotypes associated with high cytokine production have been linked to SLE susceptibility in different populations (Table 3) [31, 38-46, 74, 77-90]. Thus, an increased risk of developing SLE, independent of the HLA-DR genotype, has been reported for carriers of TNF2 allele in Caucasian [31, 44-46, 78, 81, 90], African American [91], Chinese [82,85], Colombian [31,38,92] and Mexican [42] populations, while no relation was found in a few works analyzing mestizo Mexican [83], Caucasian [39,84], African Americans [81] or Asian [41,60,74,85] cohorts. In fact, the allele-based comparisons of 21 studies [77], after stratification by ethnicity,  Journal of Biomedicine and Biotechnology detected a significant association of the −308A allele in the European-derived groups, but not in Asian-derived or African-derived populations. Conversely, no association between −238 TNF SNP and SLE was observed in the great majority of the populations analyzed [42,44,74,81,89,92]. On the other hand, the influence of TNFα microsatellite variants in SLE incidence has been poorly investigated. Alleles a2, b3 and d2 have been found to be increased in SLE patients from various European populations [86,87,93], showing linkage disequilibrium with HLA-DR3 haplotypes associated with SLE risk. On the contrary, no association was found in a study with Japanese patients [79].

Influence of IL-10 and TNFα Genotypes on Autoantibody Production
The presence of autoantibodies, mainly directed against nuclear antigens (ANAs), is one of the most characteristic features of SLE. It has been observed that the incidence of ANAs is more frequent among nonaffected family members of SLE patients than in the healthy population, suggesting that presence of autoantibodies may be, at least in part, genetically controlled [94][95][96]. The effect of IL-10 genotypes did not seem to be especially relevant, although it has been reported an increased prevalence of antibodies against several extractable nuclear antigens (anti-ENA) in patients with the allele IL-10.G9 [71], and the presence of anti-Sm antibodies was found significantly overrepresented among patient carriers of G14 and G15 alleles and R2-G15 and R2-G14 haplotypes [65]. On the other hand, an association of the high producer TNFα genotypes (−308 AA or AG) with the presence of autoantibodies has been consistently reported. It has been described an association between carriage of the TNF2 allele and presence of anti-SSa or anti-SSb antibodies [45,82,85]. This finding is in accordance with the increased frequency of TNF2 allele reported in patients with cutaneous lupus erythematosus [76,97], congenital heart block [98] and cutaneous neonatal lupus [99], all of them being pathologies linked to the presence of anti-SSa antibodies. However, it is important to consider that the actions of cytokines may be profoundly conditioned by the presence of other cytokines, and this is particularly true in the case of IL-10 and TNFα, two mutually regulated molecules which have opposite roles in the inflammatory reactions. In fact, the investigations about the effect of combined IL-10 and TNFα genotypes in SLE supported this interaction. Specifically, the highest percentage of antibodies against SSa and SSb was found among carriers of the combined genotype "low IL10 (−1082AA-AG)/high TNFα (−308AA-AG)" [45], the genotype linked to the highest TNFα production [100]. Association of high TNFα genotypes, alone or in combination, and autoantibody appearance has also been described in patients with other autoimmune pathologies such as inflammatory bowel disease or Sjögren's syndrome [101][102][103]. This effect of TNFα could be mediated by its highly proapoptotic activity since, as it has been reported, sera from SLE patients react with proteins phosphorylated during apoptosis [104], probably by recognising new epitopes generated by phosphorylation or proteolysis [105]. Thus, we can hypothesize that the proapoptotic properties of elevated TNFα levels could not be counterbalanced by the low amounts of IL-10 in patients with the high TNFα/low IL-10 genotype, thus triggering an autoimmune response to antigenically modified autoproteins generated during the apoptotic process.

Genetic Polymorphisms and Clinical Outcome
Increased circulating levels of IL-10 and TNFα have been consistently reported in the sera of patients with SLE. However, there were no definitive data on the association of IL-10 or TNFα polymorphisms and specific clinical manifestations, probably due to the heterogeneity of the disease. For instance, renal involvement has been associated with both high (GCC) [27,106] and low (ATA) [107] IL-10 producer genotypes. High prevalence of neuropsychiatric [28] and cardiovascular disorders [108] has been reported in patients with low genetic production whereas high IL-10 production has been linked to an increased incidence of serositis, hematological disorder [29], SLICC/ACR Damage Index [61] and presence of discoid or mucocutaneous lesions [45,68]. This last association was supported by the increased frequency of the high producer −1082G allele observed in patients with discoid lupus erythematosus [45,67] and by the fact that cutaneous manifestations improved in SLE patients under anti IL-10 monoclonal antibody treatment [109]. With respect to TNFα genotype, most works did not find relevant relationships with clinical parameters, although it has been reported an increased frequency of the TNF2 allele in patients with nephritis [85], central nervous system involvement [82] and presence of malar rash, discoid lesions, photosensitivity, oral ulcers or serositis [43]. However, it is worth noting the interesting association detected between TNFα genotype and clinical outcome after antimalarial treatment. Antimalarial drugs (hydroxychloroquine, chloroquine, and quinacrine) have been widely used as diseasemodifying antirheumatic agents mainly in the treatment of SLE and rheumatoid arthritis [110]. Nevertheless, their beneficial mechanisms have not been fully defined. Several in vitro experiments have demonstrated that antimalarials decreased the production of proinflammatory cytokines induced by LPS or CpG oligonucleotides in monocytes and macrophages [111][112][113][114] by a nonlysosomotropic mechanisms [115] and/or by blocking the interaction between TLR9 and CpG in monocyte endosomes [116]. More recently, this valuable antiinflammatory effect has been documented in patients with SLE, in whom antimalarialtreatment has been shown to downregulate serum levels of TNFα [100,117]. But the most interesting finding was that antimalarial effect seems to be influenced by polymorphisms of the genes encoding TNFα and IL-10. Specifically, the greatest beneficial effect of antimalarial treatment appeared in patients carriers of the combined genotype low IL-10/high TNFα, since they presented better clinical response, lower amount of circulating TNFα and increased number and function of CD4 + CD25 high Treg cells [118] as compared with other genotypes. Of note, antimalarial treated SLE patients which were carriers of the opposite high IL-10/low TNFα genotype, presented higher circulating IFNα levels, thus suggesting an interesting relationship between TNFαand IFNα in lupus disease [119].

Conclusions and Perspectives
It seems to be clear that carriage of the high producer TNF2 allele is a risk factor for SLE appearance in Caucasian populations. However, in spite of the wide number of studies performed, conclusive data on the involvement of IL-10 genetic variants have not been obtained. Nevertheless, the inverse relationship existing between both cytokines, could determine a role for IL-10 in the phenotype and/or outcome of the disease. SLE patients carriers of the high TNFα genotype probably developed the disease due to the effect of environmental or genetic factors added to their high TNFα production. The elevated levels of this cytokine may be involved in diverse pathological mechanisms and, therefore, a clinical benefit could be expected under a treatment that diminishes TNFα production. In fact, a strong association was found between carriage of this genotype, in combination with low IL-10 producer alleles, and good response to antimalarial therapy, a treatment that downregulated TNFα levels. Thus, we would expect that carriage of the proinflammatory genotype low IL-10/ high TNFα may predispose to develop anti-SSa/SSb antibodies and mild disease presenting a good course under antimalarial therapy. In addition, the conflicting data about the association of IL-10 and TNFα genotypes with clinical features which were observed by the various studies performed supports the heterogeneity of the disease and the involvement of diverse etiopathogenic factors. Thus, treatments and management of the disease might be individualized depending on IL-10 and TNFα genotypes.