19 GENOME-WIDE ANALYSIS OF DNA METHYLATION IN CLONES AND NONCLONES OF TWO DIFFERENT BREEDS: HOLSTEIN AND JAPANESE BLACK

Abstract

Epigenetic marks, and especially DNA methylation, are at the interplay of both environmental and genetic factors. By facilitating the metabolic adaptation of highly selected rent animals to their environment, DNA methylation could contribute to the phenotypic differences observed between breeds. The aim of this study was to assess to which extent the methylome of 2 specialised cattle breeds – the dairy breed Holstein and the beef breed Japanese Black – could show some variability. We focused on the liver, which has a central role in metabolism and is therefore most susceptible to be affected by genetic and environmental variations. For each breed, both cloned and noncloned animals were included in the study. We used 9 adult Holstein cows aged from 5 to 15 years (5 healthy clones generated from ear skin fibroblasts of 4 genotypes, 2 cell donors obtained by AI and 2 other AI controls of unrelated genotypes, and 11 Japanese Black cows aged from 4 to 10 years (5 healthy clones generated from cumulus cells of one genotype and 6 AI controls of unrelated genotypes). The Holstein breed and Japanese Black breed were therefore represented by 6 and 7 genotypes, respectively. Liver samples were snap-frozen after slaughtering, and genomic DNA was extracted. To identify methylated regions, we used immunoprecipitation of methylated DNA followed by hybridization on a bovine promoter microarray (MeDIP-chip). The microarray targets the upstream region (–2000 to +1360 bp) of 21 416 genes (UMD3.1 assembly). After normalization of the data, enriched probes were identified using ChIPmix (Martin-Magniette et al. 2008). Results of exploratory analysis, including correlation clustering and principal component analysis, show a clear separation between the two breeds. A statistical test based on differences in the proportion of the enriched probes was used to identify differentially methylated regions (DMR) related to cloning and breed (Spatstat R package; http://www.spatstat.org/spatstat/). Only a restricted number of cloning-related DMR could be found (240). Interestingly, most of these DMR showed no overlap between Holstein and Japanese Black animals, maybe reflecting the different origin of the somatic cells used for cloning (fibroblasts v. cumulus cells). In contrast, we identified an important number of breed-related DMR (3642). These DMR were significantly enriched in genes involved in placental development and lactation, suggesting an adaptation of the two breeds to the different metabolic demand during gestation. Whether these epigenetic differences rely on environmental variations or genetic polymorphism remains to be elucidated.Research was supported by grants ANR-09-GENM-012-01 and Egide-JSPS Sakura 2012.