CONSERVATION OF IMPRINTING IN SWINE AND COMPARATIVE ASPECTS OF IMPRINTING

Abstract

Imprinted genes have been identified in placental mammals, and marsupials but not in egg-laying mammals (monotremes), birds, fishes, reptiles, or amphibians, suggesting that the phenomena of genomic imprinting and placentation have co-evolved. While marsupials are not placental mammals per se, they develop a non-invasive yolk-type placenta that, in the case of the opossum, is functional for a period of 3 days. The parental-conflict hypothesis proposes that paternal imprinting evolved as a way to enhance fetal growth, and maternal imprinting evolved as a way to control fetal growth. The hypothesis is supported by uniparental pregnancies, direct experimental evidence from null knockout mice, as well as human conditions associated with aberrant imprinting, all of which indicate that dysregulation of imprinted genes have a drastic effect on placental morphology and function. Thus, the evolutionary and experimental evidence supporting the co-evolution of placental function and imprinting reinforces the importance of concentrating on this family of genes when searching for candidate genes capable of affecting placental efficiency and fetal growth. This is important not only for human disorders such as intrauterine growth restriction and pre-eclampsia but also to elucidate the placental abnormalities commonly seen in somatic cell nuclear transfer clones of certain species, and to understand their impact upon litter size in species such as swine. As the placenta is one of the most divergent organs among placental mammals (ranging from highly invasive haemochorial to non-invasive epitheliochorial), we have utilized comparative genomics to better understand the role of imprinted genes in placental function. We have elected to focus our efforts on the non-invasive porcine placenta and have undertaken a systematic approach to identify and compare the family of imprinted genes in this species. Pregnancies generated from embryos derived from only maternal DNA (gynogenotes or parthenotes) produce small fetuses and placentas; in contrast, pregnancies generated with only paternal DNA (androge-notes) have large, overgrown placentas but only vestigial fetuses. As such, we conducted transcriptional profiling experiments of day 30 control and parthenogenetic porcine embryos using Affymetrix Porcine GeneChip microarrays. Four tissues were profiled: brain, fibroblast, liver, and placenta. At q< 0.05, we identified 13 known imprinted genes differentially expressed in at least one of the four tissue types: DCN, DIRAS3, IGF2, MEST, NDN, NNAT, PEG3, PEG10, PHLDA2, PLAGL1, SGCE, SLC38A4, and SNRPN. DIRAS3 was detected as differentially expressed in all tissues except placenta, MEST and SGCE in all tissues except liver, and IGF2 (known to be imprinted in swine) in all tissues except brain. NNAT was detected as differentially expressed only in brain and fibroblast. NDN, PEG10, PLAGL1, and SNRPN were detected as differentially expressed in all four tissues. To confirm the expression analysis results we followed up our screen for epigenetic asymmetry by interrogating the imprinting status of a subset of differentially expressed genes by pyrosequencing using an interbreed model. Our data also supports the lack of imprinting of GABRA5 and GNAS. These findings point to the usefulness of comparative models to clarify the tissue-specific epigenetic asymmetry of mammalian imprinted loci. This research was supported by USDA-CSREES grant 524383, and NIH grant HD048510 to JP and by a NSF Graduate Research Fellowship to ST. This work was performed as part of an initiative from the Center for Comparative Medicine and Translational Research (CCMTR) at the North Carolina State University College of Veterinary Medicine.