Selection for genetic variation inducing pro-inflammatory responses under adverse environmental conditions in a Ghanaian population

Abstract

Background: Chronic inflammation is involved in the pathogenesis of chronic age-associated, degenerative diseases. Pro-inflammatory host responses that are deleterious later in life may originate from evolutionary selection for genetic variation mediating resistance to infectious diseases under adverse environmental conditions.

Methodology/principal findings: In the Upper-East region of Ghana where infection has remained the leading cause of death, we studied the effect on survival of genetic variations at the IL10 gene locus that have been associated with chronic diseases. Here we show that an IL10 haplotype that associated with a pro-inflammatory innate immune response, characterised by low IL-10 (p = 0.028) and high TNF-alpha levels (p = 1.39 x 10(-3)), was enriched among Ghanaian elders (p = 2.46 x 10(-6)). Furthermore, in an environment where the source of drinking water (wells/rivers vs. boreholes) influences mortality risks (HR 1.28, 95% CI [1.09-1.50]), we observed that carriers of the pro-inflammatory haplotype have a survival advantage when drinking from wells/rivers but a disadvantage when drinking from boreholes (p(interaction) = 0.013). Resequencing the IL10 gene region did not uncover any additional common variants in the pro-inflammatory haplotype to those SNPs that were initially genotyped.

Conclusions/significance: Altogether, these data lend strong arguments for the selection of pro-inflammatory host responses to overcome fatal infection and promote survival in adverse environments.

Figure 1. IL10 gene structure, cytokine production and changes in allele frequencies with age.

(A) IL10 gene structure with flanking gene regions (∼23.5 kb) and location of the genotyped SNPs. Pair-wise linkage disequilibrium (LD, D') as observed in the Ghana study population (n = 4336) is also depicted. Frequencies of the different haplotypes (if >1%) are presented with the minor alleles of each SNP indicated in red; (B) The association between the minor allele of each IL10 SNPs and ex vivo cytokine production in response to co-stimulation with LPS and zymosan (n = 615). The production of IL-10 and TNFα is expressed as z-scores with s.e.m, which indicate the deviance from population mean (zero-value) for carriers of at least one copy of the minor allele. Data were obtained using linear regression adjusted for age, sex, socioeconomic status and tribe, * p<0.05; ** p<0.001; (C) Minor allele frequencies of the IL10 SNPs in three age groups: ≤5 years (n = 1014), 20-45 years (n = 1462) and in ≥60 years (n = 727). Differences between groups were studied using linear regression adjusted for sex, socioeconomic status and tribe; * p<0.05; ** p<0.001.

Figure 2. Environmental interaction with gene expression in Ghana.

(A) Minor allele frequencies of the IL10 SNPs for people who make use of wells/rivers (n = 802) or boreholes (n = 3284) as a drinking source (* p<0.05); (B) Age-specific mortality risks during a five-year follow-up period for people drinking from wells/rivers and in interaction with the rs3024490 SNP. People drinking from wells/rivers had higher age-specific mortality risk compared to those drinking from boreholes (hazard ratio (HR) 0.79, 95% CI [0.67–0.92], p = 3.0×10⁻³). In addition, carriers of the pro-inflammatory rs3024490 SNP experienced a survival benefit when using water from wells/rivers but suffered a slight mortality risk when boreholes were used as a drinking source (p_interaction = 0.040); (C) Mortality risks for carriers of the IL10 SNPs separately for those who make use of wells/rivers, or boreholes, as a drinking source and interaction terms (indicated with a big star, p<0.05).

Figure 3. Location of genetic variants identified by resequencing the IL10 gene, and haplotype structure as inferred by PHASE.

Haplotypes were obtained from unphased resequencing data for 74 chromosomes (37 individuals) and were grouped into haplotype 1 and haplotypes 2 or 3 based on the initial criteria for selecting individuals for resequencing. Characteristic of the originally selected haplotype 1 was the presence of the rs1800871, rs1800872, rs3024490 and rs1554286 SNPs (indicated in boxes). The major and minor alleles are indicated in blue and yellow, respectively.