Ocular chlamydial genomic variants and disease severity in trachoma: a cross-sectional population-based genome-wide association study
Chlamydia trachomatis is the commonest cause of bacterial sexually transmitted infection and the commonest infectious cause of blindness (trachoma) worldwide. Underlying host–pathogen factors responsible for diverse clinical outcomes are unclear. We investigated C trachomatis genomic associations with disease severity using an in-vivo conjunctival phenotype and C trachomatis whole-genome sequence (WGS) analysis.
A cross-sectional population-based trachoma survey was undertaken (on the Bijagós Archipelago, Guinea-Bissau). Conjunctival samples and detailed clinical phenotype data were collected. Droplet digital PCR was used to detect and estimate C trachomatis load. C trachomatis WGS data were obtained with next-generation Illumina sequencing (Illumina, San Diego, CA, USA). Single nucleotide polymorphisms (SNPs) were called against the reference genome C trachomatis A/HAR-13. A permutation-based C trachomatis genome-wide association study (GWAS) was performed to investigate SNP associations with conjunctival phenotype.
1507 conjunctival samples and corresponding detailed clinical phenotype data were collected from 38 villages across four islands. C trachomatis was detected in 220 samples, and quantified in 184. We obtained WGS data from 126 samples. After stringent quality filtering of WGS data we retained 71 C trachomatis WGS in the final analysis. We identified 129 SNPs in coding and intergenic regions on the C trachomatis genome. Ordinal GWAS analysis found six SNPs associated with disease severity at genome-wide significance within the genes pmpE (odds ratio 0·08, p=0·009), glgA (9·71, p=0·022), alaS (0·10, p=0·032), and trmD (8·67, p=0·037), the coding locus CTA0273 (8·36, p=0·042), and the intergenic region CTA0744-CTA0745 (0·13, p=0·043).
These data provide insight into host–pathogen interactions that are likely to be important in C trachomatis pathogenesis. Further studies will determine whether these antigens are important vaccine candidates. There are limitations with statistical power in this GWAS analysis, and we conducted a permutation analysis to address this. We intend to test these associations with further in-silico, in-vitro, and replication cohorts to validate our findings.