The presence of MBD4, a member of the methyl binding domain family, was investigated in both murine and bovine oocytes and pre-implantation embryos. MBD4 is the only MBD family member that is involved in DNA repair but not active in transcriptional repression or in the formation of complexes with histone deacetylase complexes (HDACs). It contains a mismatch-specific glycosylase domain that acts to repair G:T mismatches within a CpG context. Bovine cumulus oocyte complexes were collected from abattoir-derived ovaries, matured in vitro and used for IVF as described previously (Ruddock et al. 2004 Biol. Reprod. 70, 1131–1135). Samples were analyzed at all steps in this process. Murine oocytes were collected from superovulated mice (C57BL6 × CBA) and subjected to conventional IVF. A polyclonal antibody derived in the rabbit against human peptides from specific regions of MBD4 (Imgenex, San Diego, CA, USA) was used to localize MBD4 protein. This antibody was tested at a variety of concentrations against both human HL60 leukemia cells and bovine embryos. Staining of HL60 cells was optimum at 32–64 μg/mL and embryos at 64 μg/mL. Briefly, the staining protocol consisted of fixing cells and zona-free oocytes or embryos in 4% paraformaldehyde for 15 min, followed by 15 min in 0.1% Triton X-100. Primary antibody incubation was performed overnight at 4°C. Embryos were then washed in blocking buffer for 1 hr prior to incubation at 4°C in mouse anti-rabbit IgG conjugated to FITC in blocking buffer for 30 min in the dark. Lastly, embryos were incubated in 10 μg/L Hoescht 33342 for 15 min, and then washed and mounted with Vectashield (Vector Labs, Burlingame, CA, USA). Negative controls contained no primary antibody. Mounted cells/embryos were viewed by epifluorescence microscopy. MBD4 was found to be expressed in both murine and bovine oocytes and pre-implantation embryos. In the cow, faint nuclear expression was detected at the 2-cell stage, followed by exclusion of the protein from the nucleus until the blastocyst stage of development. At this stage, staining was primarily nuclear and quite intense. In the mouse, staining was cytoplasmic at the 2 pronuclear stage, but was then concentrated in the nucleus from the 2-cell stage onward. It will be interesting to determine if this is due to the different timing of embryonic genome activation between the two species, hence implying a role for MBD4 in this important biological process. Further investigations are underway to compare the subcellular localization of the other MBD proteins in both species during preimplantation development and to identify a role for MBD4 in embryonic genome activation.