Candidate gene and locus analysis of myopia

PURPOSE A previous study has reported evidence of a strong linkage, but no association, between paired box gene 6 (PAX6) and myopia. We attempted to replicate these findings and to conduct a candidate gene and locus evaluation of genetic involvement in common forms of myopia. METHODS Samples were collected from 517 individuals in 123 families with a myopic child participating in the Orinda Longitudinal Study of Myopia or the Contact Lens and Myopia Progression Study. Myopia in the proband children was defined as -0.75 D or more and as being present in both meridians on cycloplegic autorefraction (1% tropicamide). Affected status in parents and siblings was determined by survey. After DNA was extracted from buccal mucosal cells and genotyped using assays for microsatellite markers and single nucleotide polymorphisms (SNPs), DNA was analyzed for linkage disequilibrium. Markers on chromosomes 12 and 18 were selected as regions previously associated with pathological myopia. SNPs were also analyzed in genes where their expression pattern or their association with syndromes conveys myopia as part of the phenotype (FGF2, BDNF, COL2A1, COL18A1, and PAX6). RESULTS The SNP rs1635529 for COL2A1 on 12q13.11 showed highly significant over-transmission to affected individuals (p=0.00007). No SNP for FGF2, BDNF, COL18A1, or PAX6 showed significant over-transmission to affected individuals after correction for multiple comparisons. Markers on chromosome 12 and 18 previously associated with pathological myopia also showed no significant associations with the more common form of myopia in this study. CONCLUSIONS As reported previously by others, PAX6 showed no association with myopia. Associations in the current analysis are suggestive of involvement of COL2A1. Future studies should focus on replication in other samples and in genome-wide approaches.