Major depression (MD or major depressive disorder, MDD) is a complex disease that not only affects the lives of the patients but also their family members and is one of the major causes of disability worldwide, ranking 4th according to a recent survey of global burden of disease and estimated to become 2nd by 2030 1. To date, the most popular explanation of the cause has been the monoamine hypothesis which only encompasses a part of the disturbances producing the disease. Hence, many new theories and prospects for pharmacotherapy are currently being developed 2. Among these, considerable effort has been made to elucidate the role of the innate immune inflammatory response in depression with elevated pro-inflammatory cytokines influencing neurotransmitter metabolism, neuroendocrine function, synaptic plasticity and information processing 3.

Interleukin (IL)-10 and it's recently discovered paralogs IL-19, IL-20, and IL-24 are all implicated in the inflammatory processes and act on the Th1/Th2 cytokine balance 4. All of their genes are located in the IL10 gene cluster in a 200 kb region of chromosome 1 within the locus q31-32. In accordance with the proposed role of these cytokines in various inflammatory diseases, the polymorphisms in respective genes have also been associated with many immunopathological conditions, especially psoriasis 567. In addition, while elevated pro-inflammatory cytokines such as IL-1, IL-6 and TNF-α (tumor necrosis factor α) have been associated with MDD, IL-10 as a counteractive anti-inflammatory cytokine has been studied directly in reference to MDD. Results describing the relationship between MDD and IL-10 production have so far been inconsistent, showing increased 89, unchanged 10 or decreased 11 levels of IL-10 in depressed patients, and also differentiating between depressive subtypes 12. It has been suggested that in case of the increased IL-10 it may indicate a compensatory anti-inflammatory response against generalized inflammatory state in MDD 13. Furthermore, antidepressants have been shown to stimulate the production of IL-10 along with the reduction of the general pro-inflammatory/anti-inflammatory cytokine ratio 14. In respect to neurotransmitter metabolism, IL-10 inhibits the nearly ubiquitously expressed indoleamine 2,3-dioxygenase (IDO), an enzyme responsible for directing tryptophan degradation along the kynurenine pathway 15, thus possibly increasing the availability of tryptophan for serotonin synthesis and decreasing the levels of neurotoxic N-methyl-D-aspartate (NMDA) receptor agonist quinolinic acid and NMDA antagonist kynurenic acid 16. Finally, astrocytes which release IL-10 in the central nervous system and are especially sensitive to the apoptotic effects of quinolinic acid, have been observed to be reduced in MDD 1617.

Considering the active role of IL10 cluster cytokines in inflammatory processes and the impaired immune function in MDD, and also the specific evidences linking IL-10 to MDD, the present study was aimed to investigate the possible association between genetic variations in these genes and MDD. We selected seven SNPs from the IL10 gene cluster to determine their individual and haplotype associations with MDD. It should also be noted that the genome-wide linkage analysis has identified the chromosome 1q31-32 region that contains the IL10 gene cluster as one of the susceptibility loci to bipolar disorder, known for its high comorbidity with MDD 18.

Seven SNPs of the genomic region of IL10, IL19, IL20, and IL24 genes (IL10 cluster) were analyzed in 153 Estonian patients with MDD and 277 healthy control subjects. Genotype frequencies of these polymorphisms did not deviate significantly from the Hardy-Weinberg equilibrium in neither of the groups. Comparing genotype and allele frequencies between MDD patients and controls, none of the analyzed seven SNPs showed statistically significant association with susceptibility to MDD (Table 1).

Haplotype analysis of the IL10, IL19, IL20, and IL24 genes was performed according to the pairwise linkage disequilibrium pattern observed within each of these genes (cases + controls, n = 429). The LD analysis indicated the existence of two recombination sites in the IL10 gene cluster, the first between rs1800872 of IL10 and rs2243188 of IL19 (|D'| = 0.25) and the second between rs2982572 and rs1518108 of IL20 (|D'| = 0.03; Figure 1). This also created two haplotype blocks, one containing rs2243188, rs2243193, and rs2981572 (the first two SNPs from IL19 and the third SNP from IL20 gene; |D'| 0.94–0.96) and the other containing rs1518108, rs1150253, and rs1150258 (the first SNP from IL20 and the rest from IL24 gene; |D'| 0.91–0.99; Figure 1). Four common haplotypes with an estimated frequency ≥ 1% were identified in both blocks, together comprising 98.6% and 99.1% of all haplotypes in block 1 and block 2, respectively. Additionally, the haplotype analysis provided one haplotype significantly associated with increased disease susceptibility. Namely, the block 2 haplotype TGC frequency was significantly higher in patients with MDD compared to the control group (P = 0.0097; OR 3.45; 95% CI 1.28–9.32; Table 2) and the result remained statistically significant after permutations (P_adj = 0.042; Table 2). However, the overall frequency of this haplotype was only 2.1% in the pooled group of patients and controls.

Numerous studies have established statistical associations of the IL10 gene cluster polymorphisms with various inflammatory diseases 567 and the innate immune inflammatory response is suggested to have a role in the etiology of MDD 3. Deriving from that, the present study was intended to determine the associations between seven selected SNPs from the aforementioned genomic region and MDD.

The results indicated that none of the SNPs were individually associated with MDD. On the part of the IL10 gene, this was in line with a previous report also finding no association between the IL10 gene and MDD, analyzing a promoter polymorphism at position -819 in that case 27. Other studies investigating the association between cytokine gene polymorphisms and MDD have lead to diverse results. No significant associations were found between the polymorphisms from IL1B 28, IL6 29 and TNFB 30 genes and MDD, although there was a trend for patients who were homozygous for the -511T allele of the IL1B gene to have less severity depressive symptoms and more favourable fluoxetine therapeutic response than -511C carriers 28. However, polymorphisms from TNFA 31 and CCL2 32 genes were associated with MDD. Also, single polymorphisms and seven haplotypes from inflammation-related genes involved in T-cell functioning PSMB4 and TBX21 were associated with MDD susceptibility 33.

The LD analysis indicated the existence of two recombination sites in the IL10 gene cluster, the first between rs1800872 and rs2243188 (|D'| = 0.25) and the second between rs2982572 and rs1518108 (|D'| = 0.03), thus confirming the formerly established LD pattern of this genomic region 634. The resulting two haplotype blocks consisted of rs2243188, rs2243193, and rs2982572 (|D'| 0.94–0.96); and rs1518108, rs1150253, and rs1150258 (|D'| 0.91–0.99). Additionally, the haplotype analysis revealed that the frequency of the block 2 haplotype TGC was significantly higher among patients compared to controls and was therefore associated with susceptibility to MDD (P = 0.0097; OR 3.45; 95% CI 1.28–9.32). However, this result should be approached with caution, as the overall frequency of this haplotype was only 2.1% in the pooled group of patients and controls. The block 2 contained one SNP from the IL20 gene and two from the IL24 gene and since these SNPs weren't individually associated with MDD, it is possible that the polymorphisms contributing to the increased disease susceptibility are linked to them in the same haplotype block. Considering the elevated levels of inflammatory mediators in MDD and the mainly pro-inflammatory nature of IL-20 and IL-24 cytokines, these functional polymorphisms could affect the immunological states in the context of MDD. In addition, the block 2 extends further from the IL24 gene in 3' direction and includes the genes FAIM3, PIGR, and FCAMR (HapMap, genome build 35), that could also be affected by alternative haplotypes. Unfortunately, specific causal explanations connecting SNPs and MDD are beyond the scope of this research and could be established in functional studies.

The present study established increased risk for MDD related to the IL20 and IL24 haplotype and suggests that cytokines may contribute to the pathogenesis of MDD. Since none of the block 2 SNPs were individually associated with MDD, it is possible that other polymorphisms linked to them contribute to the disease susceptibility. The main limitation in this case was the small sample size that hinders the detection of small effects. Confirmative study with increased number of the SNPs and a larger sample is needed.

The authors declare that they have no competing interests.

TT participated in the molecular genetic studies, performed the statistical analyses and drafted the manuscript. KK participated in the design of the study and the molecular genetic studies, performed the statistical analyses and helped to revise the draft. TE and EM coordinated the collection of the blood samples of the study participants and performed their psychiatric testing. KK participated in the design of the study and helped to revise the draft. VV, EV, and SK conceived the study, participated in its coordination and draft revision. All authors have read and approved the final manuscript.