Purchase this article with an account.
Shankar Swaminathan, Hong Lu, Robert W. Williams, Lu Lu, Monica M. Jablonski; Genetic Factors Affecting Iris Transillumination Defects In Recombinant Inbred Strains Of Mice: Correlation Analysis, Mapping And Mathematical Modeling. Invest. Ophthalmol. Vis. Sci. 2012;53(14):1538.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
To investigate the relative contributions of Tyrp1 and Gpnmb to the iris transillumination defect (TID) over the course of the disease and to evaluate the plausible role of other known pigmentation genes (PG) along with potential interactions using novel systems genetics methods and mathematical modeling.
3564 mice of different age groups from 91 strains of BXD RI mice and their parents were used to acquire TID phenotype. The severity of the iris disease was categorized into 5 grades that were applied to all images. Affymetrix M430V2 microarrays and RNA-seq were used to measure gene expression levels from whole eyes. Quantitative trait locus (QTL) mapping was used to map the genomic region(s) that contribute to the TID phenotype. Within any suggestive QTLs, we evaluated if any known PG were plausible candidates. TID grades were scored as a function of the age, strain and genotypes for each gene candidate and the data were analyzed ANOVA with genotypes as fixed factors. Different permutations and combinations of the PG (i.e., Gpnmb, Tyrp1, Oca2, Myo5a and Dct) were fed into the model and the contribution of individual genes and potential synergisms/interactions were studied.
Our mapping studies indicate that while Tyrp1 contributes to the phenotype at all ages, Gpnmb only contributes individually in the oldest mice tested. Composite interval mapping that controlled for Tyrp1 exposed secondary eQTL peaks for two known PG viz. Oca2 and Myo5a in younger animals (1-2 mo) and Dct in mice older than 13 mo. Analysis of interactions confirmed the contributions of other predicted candidates. The best mathematical models were achieved using all 5 PG predictors. In the youngest age group, contribution of Tyrp1 was highly significant as was the interaction of Gpnmb and Myo5a (Gpnmb*Myo5a). We also found marginal statistical significance of: Oca2; Tyrp1*Oca2; Tyrp1*Gpnmb*Oca2; and Gpnmb*Oca2*Myo5a. In oldest age group, the contribution of both Tyrp1 and Gpnmb were highly significant, while marginal significance was observed only for the interaction of Gpnmb*Dct.
These results demonstrate that, along with Tyrp1 and Gpnmb, Oca2, Myo5a and Dct modify or contribute to TID. Oca2 and Myo5a have a more significant contribution at younger ages while Dct has a stronger contribution in animals greater than 13 mo.
This PDF is available to Subscribers Only