Visceral leishmaniasis (VL) is the most severe form of leishmaniasis caused by Leishmania donovani and Leishmania infantum (chagasi). Annual incidence of VL is approximately 500 000 cases and the mortality rate in most endemic countries is almost 10%, even when treatment is available [1,2]. In Morocco, VL is exclusively caused by L. infantum [3,4]. Each year, 150 cases are recorded, 93% mostly in children under 10 years old [4]. On the other hand, the incidence of LV in Morocco is low and the number of clinical cases does not reflect the reality of the parasite infection, it seems that asymptomatic carriage is the rule in Leishmania infection, this suggests the existence of other factors including genetics that could influence the outcome of Leishmania infection.

Tumor necrosis factor-α (TNFα) (cachetin) is an inflammatory cytokine primarily produced by activated macrophages and lymphocytes T and B. It is involved in the innate phase of the immune response with a central role in the defense against intracellular pathogens [5]. The tumor necrosis factor-β (TNFβ) is a Th1 cytokine, primarily produced by activated lymphocytes T and B. TNFβ (lymphotoxin α) is also a key mediator in the initiation of a local vascular inflammatory response. Genetic polymorphisms in TNFα and TNFβ locus affect expression level of their genes [6]. The transition G to A at base pair 308 in the promoter region of TNFα has been identified (termed the A allele) [7]. A polymorphism in first intron of the TNFβ at position +252 (A>G) leads to two allelic forms; the common allele is TNFβ A and TNFβ G is variant allele [8]. Variant alleles of TNF2 and TNFβ G seem to have a strange transcriptional activation, which leads to their higher serum levels [9]. In this study we aimed to examine whether polymorphic alleles at these two loci are involved in the susceptibility versus resistance to VL in children living in Moroccan leishmaniasis endemic area.

We involved 102 children with active VL, diagnosed clinically by serological and parasitological examinations. Two groups of controls comprised 132 unrelated children from the same endemic region: 92 asymptomatic healthy volunteers, with no history of leishmaniasis and positive leishmanin skin test (DTH+); 40 healthy volunteers, with no history of leishmaniasis and negative leishmanin skin test [negative skin test delayed-type hypersensitivity (DTH-)]. Leishmanin skin test was performed by intradermal injection of 0.1 ml. Approval for the study was provided by the Ethical Committee of the Institute Pasteur of Morocco and with children’s parents consent. There were no significant differences in the distribution of the mean age and sex between cases and controls (P>0.05), suggesting that the matching based on these two variables was adequate. Molecular identification of the causative agent of VL was done by ITS1 PCR-RFLP for 55 patients; all were found to be infected by L. infantum. Genotyping of the –308 TNFα and the +252 TNFβ was performed by PCR-RFLP (18).

Briefly, 0.5 μg of DNA was added to 20 μL of reaction mixture containing 20 pmol of each primer, (TNF-: αα: 3’AGGCAATAGGTTTTGAGGGCCAT5’; 3’TCCTCCC TGCTCCGATTCCG5’; TFN- 3’CCGTGCTTCGTGCTTTGGACTA5’; 3’AGAGCT GGTGGGACATGTCTG5’) with an annealing temperature of 60°C for TNFα and 65°C for TNFβ . In a final volume of 15 μL, 10 μL of the PCR products was digested by 5 IU of NcoI (BioLabs, New England), for 3h at 37°C. Restriction fragments were separated in 3% agarose gel electrophoresis. For TNFα , an amplified product of 107 bp, containing the G to A transition at position -308, was obtained and restriction digests generated products of 87 and 20 bp for TNF1 allele and 107 bp for TNF2 allele. For TNFβ amplification followed by NcoI digestion generated a fragment of 740 bp for allele 1 (TNFβ 2 ) and 555 plus 185 bp for allele 2 (TNFβ1 ). Statistical analysis was performed using Statistical Package for the Social Sciences (SPSS Inc., Chicago, IL, USA).

All SNPs were in Hardy-Weinberg equilibrium. Associations were detected between group with active VL and DTH- control group for the heterozygote genotype (P=0.021), the recessive model: 1/2+2/2 (P=0.006) and the minor allele 2 (P=0.019) (Table 1). The minor allele 2 frequencies in TNFα promoter region are 21.6% in VL, 15.8% in DTH+ and 10.3% in DTH- groups. Regarding the polymorphism of TNFβ, the frequencies of the minor allele 2 are 29.4% in VL, 20.1% in DTH+ and 22.5% in DTH- groups. According to our statistical results, the allelic frequencies distribution in this SNP did not significantly differ between all groups. By using 1/1 genotype as the reference group, we showed a significant association under a recessive model, when comparing VL patients with DTH+ group (OR 0.558 [0.316-0.987]; P=0.044). The recessive genotype 2/2 was associated with VL patients compared to DTH- group (OR 0.245 [0.065- 0.925]P=0.023).

Table 1: Genotypes, allelic and recessive model frequencies distribution of -308 TNFα and +252 TNFβ polymorphisms in the VL, DTH+ and DTH− groups (CI: confidence interval; OR: odds ratio).

The pro-inflammatory cytokine TNF plays a role in innate and adaptive immune responses, and is also implicated in a wide variety of infectious and autoimmune human diseases. In the present work, no association was found in genotypic and allelic frequencies of TNF-α and TNF-β between patient group and asymptomatic infected group (DTH+). Associations were detected between VL patients and DTHcontrol group for the heterozygote genotype, the minor allele2 and the recessive model for TNF-α , and for homozygote genotype 2/2 of TNF-β , but due to smaller number of DTH- cases, this result must be regarded as preliminary.

A few studies with controversial results have been performed to evaluate the TNF polymorphisms in different clinical types of leishmaniasis. In Brazil, an association was found between the outcome of Leishmania chagasi infection and alleles at TNF locus. The strongest association was found between asymptomatic infection DTH + and a polymorphism in the TNF locus and haplotypes containing TNF2 were associated with symptomatic VL [10]. In agreement with this finding, a case control study of 46 patients with mucocutaneous leishmaniasis caused by Leishmania braziliensis suggested that the frequency of allele 2 at the -308 TNF-α gene polymorphism is significantly higher in patient than asymptomatic group [11]. Other studies showed that TNF polymorphisms are not responsible for the resistance versus susceptibility to cutaneous leishmaniasis [11,12]. In Tunisia, Meddeb-Garnaoui failed to find associations between either the -308 of TNF-α , gene polymorphism or the first intron of TNF-β gene polymorphisms and susceptibility to Mediterranean VL [13]. Although controversial, the majority of the data support a direct role for the -308 TNF2 allele in the elevated TNF levels observed in TNF2 homozygotes [9], many arguments support the central role of TNF in the anti-Leishmania immune response has been strengthened by observations in Leishmania major (L. major ) (strain BNI) infected B6. TNF –/– mice, which were not able to develop an efficient immune response against parasites and died 6–8 weeks after infection from visceral leishmaniasis [14]. Conversely, Ritter et al. showed that infection of B6. TNF –/– mice with L. major (strain FRIEDLIN) resulted in an attenuated form of disease, even though the animals were not able to resolve the local lesions but developed a chronic form of cutaneous leishmaniasis. Thus, in addition to the extensively studied host factors, the biological properties of Leishmania strains play an important role in the outcome of leishmaniasis [15]. Considering numerous genetic variations, that influence the apparently divergent findings, our data show that -308 TNF-α and +252 TNF-β genotypes do not influence susceptibility versus resistance to visceral leishmaniasis. However, some associations were detected between VL patients and DTH- control group, but due to smaller number of DTHcases, these results should be interpreted with caution. These associations however, point out the need for further studies with a larger sample size of control groups, to understand the specific role of TNF that confer protection against leishmaniasis.

This work was funded by Institute Pasteur of Morocco and EMROCOMSTECH Grant RP 04/47. The authors are grateful to the team of Department of Parasitology, Ministry of Health, the Health Delegations of the Provinces of Zouagha My Yacoub, and Taounate. We also thank the local authorities of the provinces.