PROMOTER POLYMORPHISMS OF THE CANINE SCL11A1 GENE ARE CORRELATED WITH SUSCEPTIBILITY TO CANINE LEISHMANIOSIS

1Laboratory of Parasitology and Parasitic Diseases, Veterinary Faculty, School of Health Sciences, Aristotle University of Thessaloniki, Thessaloniki, Greece; 22nd Department of Paediatrics, Medicine Faculty, School of Health Sciences, Aristotle University of Thessaloniki, Thessaloniki, Greece; 3Diagnostic Laboratory, Department of Clinical Sciences, Veterinary Faculty, School of Health Sciences, Thessaloniki, Greece; 4Department of Genetics, Development and Molecular Biology, School of Biology, Aristotle University of Thessaloniki, Thessaloniki, Greece


INTRODUCTION
Canine leishmaniosis (CanL) is a complex, life threatening disease with a variety of clinical manifestations [1]. Epizootical studies using molecular techniques in enzootic areas have confirmed that CanL has a prevalence involving as much as 67%-80% of the dog population [2,3]. However, this high prevalence estimate is accompanied by a low rate of apparent clinical disease. One determining factor for the polarity of clinical manifestations is the balance between humoral and cellular immune responses [1]. Protective immunity and thus resistance against Leishmania parasites is mediated by CD4+ Th1 cellular responses, where the cytokines interferon gamma (IFN-γ), tumor necrosis factor alpha (TNF-α) and interleukin-12 (IL-12) are released inducing macrophage activation and nitric oxide (NO) killing of parasites. On the contrary, the failure to control infection is related to the Th2 immune response, increased activity of humoral immunity and the production of IL-4, IL-5, IL-6, IL-10 and IL-13 [4]. Other genetic factors, such as major histocompatibility complex (MHC) class II and TNF-α have been analyzed for CanL and significant correlations have been reported [5,6]. SLC11A1 (Solute Carrier Family 11 member 1), also known as Nramp1 (Natural resistance associated macrophage protein 1), is a metal ion transporter protein, which is expressed in the lysosomal compartment of professional phagocytes [7]. The SLC11A1 protein is crucial for the initial activation of macrophages and is critical for resistance to intracellular pathogens [8]. Moreover, SLC11A1 is implicated in the regulation of the expression of TNF-α and IL-1b [9] and it enhances the induction of nitric oxide synthase (iNOS) and NO, which are toxic to Leishmania parasites [10]. SLC11A1 is encoded by the positionally cloned Ity/Lsh/Bcg gene, which was firstly described in genetic studies of mouse models for its role in regulating susceptibility and resistance to Leishmania, Salmonella and Mycobacterium [8,11]. The canine SLC11A1 gene has been located on dog chromosome CFA37 and spans for 9kb, including a promoter region of 700 base pairs (bp) and 15 exons (GenBank accession number AF091049) [12]. Up to date, there is increasing evidence that genetic variation within the SLC11A1 can determine susceptibility or resistance to CanL [13,14]. Three single nucleotide polymorphisms (SNPs) within a region of 491 bp, mapped in the promoter of the canine SLC11A1 gene have been identified by sequence variant analysis conducted by direct sequencing of the cDNA and characterized in susceptible dogs from various breeds, i.e. SNP1 T151C, SNP2 Α180G και SNP3 G318A [13,14].
The aim of the study was to corroborate the association of these three sequence variants in the promoter region within the canine SLC11A1 gene with susceptibility or resistance to CanL in an extended case-control cohort.

Animals and DNA extraction
Infected dogs (n=148) from 14 different breeds were included in the study. All dogs were living in Northern Greece and had owners. From each dog we used the residual of peripheral whole blood, which was taken anyway at routine check ups by practitioners. For all dogs Leishmania infection was verified, either by positive cytology following lymph node aspiration and/or by positive anti-Leishmania antibody titre. In detail, for the detection of antibodies against Leishmania spp. an in-house enzyme-linked immunosorbent assay (ELISA) was used as described previously [15]. Furthermore, it was confirmed that all dogs were free of symptoms of any other disease with possible immunosuppressive effect and were seronegative (spot ELISA) for Ehrlichia canis and Dirofilaria immitis. Dogs were divided in two groups: resistant and susceptible. Breeds were equally distributed to both groups so as to exclude breed effect ( Table 1). The group of resistant dogs comprised of 73 apparently healthy dogs (39 males and 34 females). All of these dogs were older than 7 years and never exhibited any clinical manifestations of CanL, although examined regularly by a practitioner. The group of susceptible dogs consisted of 75 symptomatic dogs (34 males and 41 females) that exhibited clinically severe disease and/or numerous relapses. Genomic DNA was extracted from 0.3 ml of peripheral whole blood of the examined animals using the Gentra Puregene Blood Kit according to the manufacturer's manual (Qiagen, USA). Genotyping analyses in regard to the three SNPs were accomplished by enzymatic digestion of the DNA samples in question with specific restriction endonucleases.
In particular, the enzymes Hpy188I, Bsp1286Ι and AciI (New England Biolabs) were selected using NEBcutter V2.0 (http:/tools.neb.com/nebcutter2) to discriminate between SNP1 T151C, SNP2 A180G and SNP3 G318A, respectively ( Table 2). All restriction reactions were done in a final volume of 15 μl containing 6 μl PCR product, 1.5 μl NE buffer (10x), 0.3 μl of the enzyme and 7.2 μl distilled water with incubation at 37 °C for 16h. The restricted fragments were electrophorized on 2.5% ethidium bromide stained agarose gels and visualized under UV illumination. Informed consent: Informed consent for the use of the residual 0.3 ml blood and every dog's medical record in the form of written permission has been obtained from the owners for all dogs included in the study.

RESULTS
For SNP1 T151C only the TT genotype was detected in all dogs ( Figure 1) and thus a chi-square test was not conducted. For SNP2 A180G both genotypes (AA and AG) were detected in both groups ( Figure 2). The AA genotype was detected in 70 resistant and 51 susceptible dogs. The AG genotype was detected in 3 resistant and 24 susceptible dogs. SNP2 Α180G was statistically correlated with susceptibility to CanL (p < 0.05) ( Table 3). For SNP3 G318A both genotypes (GA and AA) were detected in both groups ( Figure 3). The GA genotype was detected in 71 resistant and 57 susceptible dogs. The AA genotype was detected in 2 resistant and 18 susceptible dogs. Subsequent statistical analysis demonstrated that SNP3 G318A might be a susceptibility factor since the presence of G318A polymorphism in homozygous state (AA) was remarkably higher (p < 0.083) in susceptible dogs (Table 4). Finally, according to the statistical analyses all loci were in Hardy-Weinberg equilibrium for both populations (p > 0.05).

DISCUSSION
In CanL some dogs are resistant and thus remain asymptomatic, while others are susceptible and develop the clinical form of the disease [1,2]. A number of genetic loci have been involved in susceptibility or resistance, although the mechanisms are not yet thoroughly clarified [3]. In a cohort of Brazilian mongrel dogs naturally exposed to L. infantum the presence of MHC class II allele DLA-DRB1*01502 increased the risk of infection [5]. A correlation of two SNPs, i.e. TNF-α -40 C/Α and TNF-α -1243 C/G, located upstream the canine TNF-α open reading frame with susceptibility or resistance to CanL, respectively, was determined from a study of naturally infected dogs in Greece [6].
SLC11A1 is highly conserved among humans, mice and dogs [12] and it is implicated in susceptibility to L. infantum infection in mice [8,16]. In humans one repeated GT promoter sequence and 823C/T in exon 8 have been associated with susceptibility to visceral leishmaniosis [17][18][19], while SNPs in the 3'UTR region, in exon 8 and the D543N variant have been correlated to cutaneous leishmaniosis [20,21].
In the present study three previously described SNPs (SNP1 T151C, SNP2 Α180G και SNP3 G318A) in the promoter of the canine SLC11A1 gene were further analyzed in a case-control study of 148 dogs. Statistical analyses revealed a significant association either for the SNP2 Α180G in heterozygosity (AG) or for the SNP3 G318A in homozygosity (AA) with susceptibility to CanL while in other studies all three of these SNPs were implicated in the control of infection [13,14,22]. SLC11A1 is implicated in Th0 cell differentiation of infected animals. Studies have shown that congenic mice carrying the wild-type SLC11A1 allele mounted a predominantly Th1 immune response to vaccination and resolved lesions upon challenge with Leishmania major whereas mice carrying the mutant allele mounted a Th2 immune response and lesions expanded following challenge [23]. The mechanism of this action is not clarified, although it has been proven that the ability of SLC11A1 to induce the pathways of IL-10 and IL-12 in macrophages and dendritic cells (DCs) is involved [24]. The balance between these cytokines reflects the development of the Th1 cellular immune response [25]. It has been proven that the ratio of IL-10 production to IL-12 is higher in DCs from mutant mice than in those from wild-type mice [26].
SLC11A1 influences innate immunity against Leishmania parasites by interfering directly with antigen processing [20,27], probably by regulating the activity of proteases in the endosomal compartment [23], which is dependent on metal ions [28]. It has been proven that macrophages of congenic mice carrying the wild type of SLC11A1 have impaired iron homeostasis and deficient antigen processing compared to the ones bearing the polymorphism [27]. SLC11A1 also influences antigen processing by regulation of MHC class II molecules [29]. DCs of mice carrying the wild-type allele show increased expression of MHC class II than DCs from mutants [26].
In dogs silico analysis has revealed that within the vicinity of these SNPs are putative transcription factor binding sites [14]. This fact affects the binding ability of transcription factors and results in changes of transcriptional machinery [30]. One SNP of the human SLC11A1 gene, which was associated with infectious diseases, encodes four alleles, which differ in their ability to drive gene expression [17]. This may explain the diverse expression of SLC11A1 gene in dogs bearing the polymorphic sites [12].
In conclusion, in the present study the presence of SNP2 Α180G in heterozygosity (AG) and the presence of SNP3 G318A in homozygosity (AA) were correlated with susceptibility to CanL. Future studies of more SNPs in candidate genes are required so as to clarify the roles of different polymorphic variants in determining CanL severity.