부산대학교 도서관

Amelogenesis Imperfecta (AI) is a Mendelian inherited rare disease, affecting the formation of enamel during tooth development. Mutations in various genes have been implicated in AI, however, even after whole exome sequencing (WES) the genetic basis of the disease remains unexplained in between a third and half of AI patients. This project aimed to identify new genes mutated in non-syndromic AI by combining WES in unsolved AI families with the study of the evolutionary history of genes associated with AI. Specifically, whether selective pressure during the genes' evolution created a pattern of selection shared among them, that could provide further evidence to support new candidate genes and variants implicated in AI. The WES of individuals from 33 families, presenting with various AI phenotypes resulted in the identification of 25 variants causing AI. Unrelated families sharing variants were further examined for phenotype overlap and where unrelated families shared variants, these were tested to determine whether there was a common ancestral founder. The unexpectedly high level of diagnostic success obtained in the cohort of families was probably due to the selection of families with clear patterns of dominant or consanguineous recessive inheritance. It is proposed that genes that co-evolved and cooperate to form tooth enamel in mammals should exhibit similar substitution patterns. Loss of enamel or teeth occurred independently multiple times in placental mammals, so representatives of both character states across the phylogeny were included in the analysis. Using molecular evolutionary tools (i.e.: codeml and SLAC), the variation in selective pressure was calculated by examining the substitution patterns in protein-coding sequences of these genes. Positive selection was found to act on genes active during amelogenesis, while genes with broader functions did not exhibit detectable levels of positive selection. Toothless and enamel-less species showed signatures of species-specific positive selection in AI associated genes, indicating a potential shift of function in these species. Surprisingly these genes had been considered pseudogenised, yet a high level of sequence conservation is evident, suggestive of a selective constraint. Future work on these regions is required to determine whether they are producing functional protein sequences.