189 TRANSCRIPTIONAL PROFILING OF HISTONE-MODIFYING GENES DURING BOVINE PRE-IMPLANTATION EMBRYO DEVELOPMENT IN VITRO

Abstract

The proper removal of gametic epigenetic marks and coordinated re-establishment of the epigenome is critical to mammalian embryonic development. This global reprogramming of the embryonic genome includes fluctuations in both DNA methylation and histone modifications that are necessary to control chromatin structure and thus gene expression. In the bovine model, epigenetic changes occur from fertilization through blastocyst stages; in particular, and concurrent with the maternal-embryonic transition, de novo DNA methylation begins at the 8-cell stage. In order to understand which factors might be playing key roles in this epigenetic process, we used quantitative real-time PCR to characterize the temporal expression profiles of several genes involved in DNA and/or histone methylation: G9a, SetB1, Suv39h1, Suv420h1, SmyD3, Suz12, and LSH. Bovine ova and embryos were produced via in vitro maturation, fertilization, and culture from multiple pools of ova. Groups of 12–25 bovine ova or embryos, pooled at the 2-, 4 to 7-, mid 8-, late 8-, 12 to 16-cell, morula, and blastocyst stages, were washed twice through 1X PBS and stored in RNA lysis buffer at –80°C until further use. RNA was isolated from each sample using the RNeasy® Mini kit (Qiagen, Valencia, CA, USA), optimized for isolating RNA from single embryos, and treated to remove any contaminating genomic DNA. cDNA was generated with iScript™ reverse transcriptase (Bio-Rad Laboratories, Hercules, CA, USA) and diluted 1:10 with RNase/DNase-free water for further use in real-time PCR. Relative gene expression from each RNA sample was calculated in triplicate using the SYBR Green comparative Ct method (Applied Biosystems, Foster City, CA, USA) adjusted for individual PCR efficiencies (Bustin 2003) and normalized to the geometric mean Ct of 3 endogenous controls (GAPDH, YWHAZ, and SDHA) in order to account for differences in both cell number and amount of total mRNA present in each sample (Goossens et al. 2005). G9a and SetB1, both lysine-specific methyltransferases, were expressed at their highest levels in the metaphase II (MII) oocyte and 2-cell stage, before expression decreased gradually to basal levels by the morula and blastocyst stages. Suv39h1, Suv420h1, and SmyD3, also lysine-specific methyltransferases, all shared a similar pattern of expression: transcript levels were fairly high in the MII oocyte, increased at the 2-cell stage, then gradually dropped off around the 8–16-cell stage to basal levels by the morula stage. Interestingly, Suz12 and LSH both showed low expression from the MII oocyte until the 4 to 7-cell stage, increased dramatically at the 8-cell stage, then decreased again by the morula stage. Suz12 is a member of several Polycomb group complexes (PRCs); LSH associates with PRC-mediated gene silencing as well as DNMT3a and 3b. These data suggest that Suz12 and LSH may be implicated in bovine embryonic genome activation, while the latter genes are active during earlier cleavage events. Ongoing studies will evaluate the role of each of these epigenetic modifiers in bovine pre-implantation embryos by selective silencing via RNA interference.