Replication of the association of chromosomal region 9p21.3 with generalized aggressive periodontitis (gAgP) using an independent case-control cohort

Abstract

Background: The human chromosomal region 9p21.3 has been shown to be strongly associated with Coronary Heart Disease (CHD) in several Genome-wide Association Studies (GWAS). Recently, this region has also been shown to be associated with Aggressive Periodontitis (AgP), strengthening the hypothesis that the established epidemiological association between periodontitis and CHD is caused by a shared genetic background, in addition to common environmental and behavioural risk factors. However, the size of the analyzed cohorts in this primary analysis was small compared to other association studies on complex diseases. Using our own AgP cohort, we attempted to confirm the described associations for the chromosomal region 9p21.3.

Methods: We analyzed our cohort consisting of patients suffering from the most severe form of AgP, generalized AgP (gAgP) (n = 130) and appropriate periodontally healthy control individuals (n = 339) by genotyping four tagging SNPs (rs2891168, rs1333042, rs1333048 and rs496892), located in the chromosomal region 9p21.3, that have been associated with AgP.

Results: The results confirmed significant associations between three of the four SNPs and gAgP. The combination of our results with those from the study which described this association for the first time in a meta-analysis of the four tagging SNPs produced clearly lower p-values compared with the results of each individual study. According to these results, the most plausible genetic model for the association of all four tested SNPs with gAgP seems to be the multiplicative one.

Conclusion: We positively replicated the finding of an association between the chromosomal region 9p21.3 and gAgP. This result strengthens support for the hypothesis that shared susceptibility genes within this chromosomal locus might be involved in the pathogenesis of both CHD and gAgP.

Figure 1

Linkage disequilibrium (LD) map for the SNPs genotyped in this study. The LD region 1 as described by Schaefer et al. (2009) [17] is represented by the three SNPs rs2891168, rs1333042, and rs1333048. The LD region 2 is represented by SNP rs496892. The derived LD block structure is shown for the Greifswald case-control cohort (n = 469) described in this study. Numbers represent pair wise percent R²-values.

Figure 2

Transcriptional map of the human 9p21.3 locus associated with CHD and gAgP. The transcripts specifying CDKN2A, CDKN2B, ARF, and ANRIL are shown as well as the four SNPs tested in this study (rs496892, rs2891168, rs1333042, rs1333048) and the two SNPs rs10757278 and 1333045 analyzed extensively in the studies of [26] and [25], respectively (illustration adopted from [26] and modified, by courtesy of Norman E. Sharpless).

Figure 3

Linkage disequilibrium (LD) map for the three SNPs within LD region 1 genotyped in this study (rs2891168, rs1333042, and rs1333048) as well as for rs10757278 and the putative causative SNP rs1333045. The derived LD block structure is derived from the HAPMAP data. Numbers represent pair wise percent R²-values.