Evolution of Genomic Imprinting: Re-Visiting the Kinship Theory

Abstract

Genomic imprinting describes the monoallelic expression of specific loci based on parent of origin of the allele. The prevailing theory on the evolution of imprinting is known as the “Kinship Theory,” which postulates that imprinting is the result of a conflict between the paternal and maternal genomes over the growth and development of the embryo. However, this theory, like many others, was based upon the very small sample size of known imprinted genes in mammals (~126 in mice). In 2005, Leudi et al. claimed to have identified ~600 new imprinted mouse genes using machine learning classifiers based on sequence data. The goal of this thesis was to rigorously examine which of these mouse genes identified by Leudi et. al are truly imprinted, and attempt to reconcile their molecular functions and parental expression bias with the kinship theory. Upon filtering the Leudi et al. genes based on more stringent criteria for imprinting, 6 new imprinted clusters were identified and mapped. Subsequent GO analysis surprisingly showed a strong enrichment of GO terms associated with neurological development. I conclude that these results are, in fact, consistent with the Kinship Theory because behaviors during infancy are critical to an offspring’s growth. Therefore, the genes underlying them are mediated by imprinting in the same way genes underlying fetal growth are mediated. In this way, the Kinship Theory’s intragenomic conflict extends itself beyond nutrient allocation in utero; it mediates imprinting after birth, when the infant is still dependent on its mother.