Genetic Association and Gene-Gene Interaction Analysis of Apol1, Myh9 and G6pd Variants in Patients with Chronic Kidney Disease

Abstract

Chronic Kidney Disease (CKD) is a major health crisis globally because of the high prevalence recorded in many developing and developed countries and is estimated to affect about 200 million persons worldwide. Variants in the human Apolipoprotein L1 (APOL1) and the Human non-Muscle Myosin (MYH9) genes are associated with the risk of CKD in people of West African descent, but not all individuals with these mutations develop CKD in their lifetime. There is a need to identify other genetic mutations that predisposes individuals to disease association. This work was aimed at determining the prevalence of G6PD variants A- (202) and A (376) in CKD patients of unknown aetiology and also to determine whether G6PD variants could be a risk factor by interacting with APOL1 and MYH9 gene variants in increasing or reducing the risk of developing CKD in the West African region. This was a case-control study, which utilised archived blood samples from CKD patients of unknown aetiology from Ghana and Nigeria, collected on behalf of the H3Africa Kidney disease Research Network (KDRN) in Ghana. Samples were genotyped using Restriction Fragment Length Polymorphism (RFLP) following PCR amplification of the region of the G6PD gene that carries these mutations. Whiles genotyping of APOL1 and MYH9 variants was done using Ligation Detection Reaction (LDR) assay. A total of 411 samples made up of 202 cases (CKD patients of unknown aetiology) and 209 controls (from the general population) were genotyped for 2 G6PD variants (A- and A), 4 APOL1 variants (rs73885319 (G1), rs60910145 (G1), rs71785313 (G2) and rs9622363) and 4 MYH9 variants (rs20324871, rs4821481 (E1 haplotype) and rs5750248 and rs5750250 (S1 haplotype)). The general prevalence of G6PD A- and A observed for cases was 49% and 54.5% respectively, whiles that of the control population was 44% and 45.9% respectively. In terms of gender, the prevalence observed for the male and female cases were 43.2% and 56% for G6PD A- and 47.7% and 62.6% for G6PD A variants respectively. There was a significant difference between the prevalence observed for the females and the control group (p = 0.001) but not for the male subjects (p = 0.4205) compared to the male control group. Significantly strong associations was observed between females who had at least one risk allele (heterozygous carriers) of G6PD A- and CKD of unknown aetiology (OR 1.96, 95%CI 1.0044-3.3030, p = 0.048) which further increased for those with 2 risk alleles (homozygous carriers) (OR 6.8, 95%CI 2.2225-19.2411, p = 0.001), whilst a similar association was observed between females with 1 risk allele of G6PD A variant and CKD of unknown aetiology (OR 2.0, 95% CI 1.1206-3.6372, p = 0.019). Significantly strong associations was also observed with previously reported APOL1 variants rs73885319 (OR 1.3, 95%CI 1.6435-4.6118, p = 0.02) and rs60910145 (OR 1.3, 95%CI 1.7042-4.9177, p = 0.004) after controlling for age, gender and clinical site location. No significant evidence of association was observed between CKD of unknown aetiology and any of the MYH9 variants after accounting for multiple testing in the sample. Interaction analysis however revealed significant epistatic interactions between deficient G6PD A- and A genotype and heterozygous rs4821481 (p = 0.016), a trend towards association for rs5750250 (p = 0.062), and the APOL1 variants rs60910145 (0.014) and rs73885319 (0.002) which indicates a synergistic epistatic interaction between carriers of these genotypes which could predispose them to CKD of unknown aetiology.