Method: We conducted a GWAS using HapMap-II imputed genotype data from 4910 European American participants of the Dental ARIC study. Complex phenotypes were derived from PCA of “intermediate” traits including microbial composition and gingival crevicular fluid inflammatory biomarkers. We considered SNPs with minor allele frequency (MAF) of 5% or greater, and used a genome-wide significance threshold of p<5x10-8. Analyses relied on logistic regression models assuming log-additive allelic effects.
Result: The first 4 principle components, which were significantly associated with the presence of periodontal disease, as well as a unique pattern of microbial composition and inflammatory mediator response, demonstrated several statistically significant loci. For example, one locus was identified (chr2) as a mis-sense mutation for a cytokine that is critically involved with the down-regulation of the innate immune response at mucosal surfaces [OR=1.70, (95% CI: 1.53-1.89), p=6.8x10-21, MAF=0.3]. A second significant mis-sense mutation locus (chr1) was identified for an inflammasome protein critical for processing TLR-mediated responses [OR=2.54, (95% CI: 1.83-3.54), p=2.8 x10-8, MAF=0.12].
Conclusion: Using PCA that incorporates microbial composition and GCF inflammatory biomarker data to define complex trait phenotypes is far more informative for identifying subsets of individuals with specific genetic-periodontitis syndromes than GWA using clinical phenotypes. The identification of previously un-reported candidate genes, which modulate the innate immune response, provide promise for establishing relatively common, new genetic syndromes for periodontal disease.