Phenotypic Consequences Of Imprinting Perturbations At Rasgrf1 In Mouse

Abstract

Rasgrf1 is imprinted and paternally-expressed in neonatal mouse brain. At weaning, expression becomes biallelic. Manipulating the elements regulating imprinted Rasgrf1 expression produces mice with transient perturbations in Rasgrf1 imprinting, which we can use to assay the consequences of aberrant imprinting during a specific developmental time. As neonates, monoallelic-paternal (wild-type) mice paternally express Rasgrf1; biallelic mice express Rasgrf1 from both alleles; null mice do not express Rasgrf1; and monoallelic-maternal mice reciprocally express Rasgrf1, from the maternal allele. This last genotype represents a unique opportunity to study the effects of expression derived from either of the two parental alleles. All genotypes biallelically express Rasgrf1 around weaning. Two phenotypes appear when Rasgrf1 imprinted is perturbed: a difference in size, and a difference in performance on an associative odorlearning task at postnatal day 8 (P8). Size differences persist through adulthood and reflect the level of Rasgrf1 expressed pre-weaning: biallelics are larger, nulls are smaller, and the two monoallelically-expressing genotypes are intermediate and indistinguishable from one another. We determined that neonatal Rasgrf1 expression affects the functioning of the growth hormone/IGF-1 axis (GH/IGF-1 axis), with an input as far upstream as the hypothalamus, and that Rasgrf1 helps to set growth parameters early in development that persist through adulthood. Importantly, these data are the first experimental validation of one of the critical assumptions of the “conflict hypothesis,” which describes the evolution of genomic imprinting in mammals. Imprinting perturbations at Rasgrf1 also produce performance differences on an associative odor-learning task in P8 neonates. The presence of a wild-type paternal allele is critical for proper task performance, as null mice, and those with maternally-derived Rasgrf1 expression, do not perform as well. These differences in neonatal performance are accompanied by differences in Rasgrf1 expression level and alterations in the amount of activated Ras and Rac protein in various brain structures. Differences in learning are not present in adult mice, suggesting that the role Rasgrf1 plays in these phenotypes differs with respect to age, as well as the sensitivity of the system to Rasgrf1 expression level.