Is there an association between genetic polymorphisms and immune responses to rotavirus vaccination in South African infants?

Abstract

Rotavirus (RV) vaccines have demonstrated low efficacy in sub–Saharan Africa. Recent studies have reported that RVs bind to histoblood group antigens (HBGAs), and thus they may act as receptors for this virus. Variation in HBGA phenotypes have been suggested as the cause of RV vaccines efficacy reduction. Thus, the aim of this study was to investigate the associations between polymorphisms in HBGA genes (ABO/FUT2/FUT3) and selected innate immune response genes, and ABO/Secretor (Se)/Lewis (Le) phenotypes with immune response (seroconversion) to RV vaccine called Rotarix (RV1) in Black South African infants. A total of 215 saliva samples collected before and after exposure to RV vaccination in infants were available from a previous study. Previously published RV–specific IgA titres pre– and post–vaccination were used to classify samples as seroconverters or non–seroconverters. Pre–vaccination IgA titre in babies was significantly associated with seroconversion and was used as a covariate in multivariable analysis of factors influencing baby immune response to vaccination. We reported minor allele frequencies (MAF) of the 50 SNPs. A haplotype in FUT2 linked to the se428 mutation (rs601338) was the most common haplotype in the cohort (42.8%); overall the frequency of non–Secretor status predicted from genetic data was 16%. The most common FUT3 haplotype causing predicted Le absence was linked to the le508 (rs3745635) mutation, found in 29% of haplotypes. Overall, the frequency of Le negative phenotype predicted from genetic data was 20%. ABO genotypic data were used to predict ABO blood type frequencies in the cohort, that were similar to previous frequencies reported in the SA Black population. Se and Le phenotypes, FUT2 genotypes and ABO genotypes were not significantly associated with seroconversion in the current study. Only one FUT3 SNP rs761121538 (le862) was associated significantly with seroconversion in an allelic model (P= 0.002, OR= 0.070; P= 0.042 after Bonferroni correction, P= 0.051 in multivariable analysis). Other missense mutations in FUT3 were not significantly associated with seroconversion in the current study. However, when FUT2 and FUT3 genetic data were combined and used to predict Le antigen phenotypes, there was an association between predicted Le phenotype and seroconversion (P= 0.026). Epistasis analysis also suggested that the interaction between some SNPs in the FUT2se428 haplotype and the FUT3 le667 SNP (creating the Le (a–b–) phenotype) was significantly associated with seroconversion. These findings support the hypothesis that combinations of FUT2–FUT3 variants may contribute to seroconversion after exposure to Rotarix vaccine. With regards to our study of genetic variation in genes influencing innate immunity, the TLR8SNP rs2129377 was significantly associated with seroconversion (P= 0.0146, OR= 5.351; P >0.05 after Bonferroni correction) and one six SNP haplotype found on the X chromosome in the TLR7–TLR8 region was significantly associated with seroconversion (P= 0.042, P > 0.05 after Bonferroni correction). Overall, it can be concluded that interaction between SNPs in FUT2 and FUT3 genes, and variation in innate sensors of viral RNA, might play a significant role towards variation in immune response to Rotarix vaccine in this South African cohort