A genetic means of identifying "pregnancy competent" oocytes is provided. The means comprises detecting the level of expression of one or more genes that are expressed at characteristic levels (upregulated or downregulated) in cumulus cells derived from pregnancy competent oocytes. This characteristic gene expression level, or pattern referred to herein as the "pregnancy signature", also can be used to identify subjects with underlying conditions that impair or prevent the development of a viable pregnancy, e.g., pre-menopausal condition, other hormonal dysfunction, ovarian dysfunction, ovarian cyst, cancer or other cell proliferation disorder, autoimmune disease and the like. In preferred embodiments the pregnancy signature will comprise one or more of ABCA6, NCAM1, OLFML3, PTPRA, SDF4, GPR137B, DDIT4, DUSP1, GPR137B, IDUA, KCTD5, NDNL2, SLC26A3, and TERF2IP.

ObjectiveDespite its limited accuracy, subjective morphological criteria continue to be the standard for embryo selection for transfer during IVF. An objective and accurate method to select embryos for transfer could increase pregnancy rates while also promoting a move towards universal single embryo transfer. Gene expression profiling of cumulus cells (CC) offers an attractive approach to reach this goal.DesignMultisite retrospective study. Gene expression in CC from single oocytes were analyzed and correlated to pregnancy outcome using a qRT-PCR platform.Materials and methodsPatients < 36 years from three clinical sites undergoing either IVF or ICSI and patient-specific stimulation protocols were recruited for the study (n=55). Gene expression of a selected group of genes in CC corresponding to 86 individual oocytes was analyzed using qRT-PCR. Two independent CC sample sets, A (n=49) and B (n=37), from two different sets of patients, were evaluated. CC samples were divided into two groups: "live birth samples" corresponded to CC of single oocytes which produced embryos that resulted in healthy live birth; "nonpregnant samples" corresponded to CC of single oocytes which produced embryos that were transferred into the patient but failed to implant.ResultsPreviously, a candidate set of <250 genes was identified on microarray as predictive of pregnancy outcome. A selection of these genes was validated on sample set A using qRT-PCR, and a refined panel of <20 genes was identified as predictive of pregnancy outcome. The predictive power of this signature was then tested on sample set B. An established method for determining diagnostic utility of a test, Receiver Operating Characteristic analysis, was employed, and confirmed predictive power of the signature (P<0.0001).ConclusionTranscriptional profiling of CC across multiple sites and stimulation protocols can be used to predict those oocytes and embryos with the capacity to advance to healthy live birth. ObjectiveDespite its limited accuracy, subjective morphological criteria continue to be the standard for embryo selection for transfer during IVF. An objective and accurate method to select embryos for transfer could increase pregnancy rates while also promoting a move towards universal single embryo transfer. Gene expression profiling of cumulus cells (CC) offers an attractive approach to reach this goal. Despite its limited accuracy, subjective morphological criteria continue to be the standard for embryo selection for transfer during IVF. An objective and accurate method to select embryos for transfer could increase pregnancy rates while also promoting a move towards universal single embryo transfer. Gene expression profiling of cumulus cells (CC) offers an attractive approach to reach this goal. DesignMultisite retrospective study. Gene expression in CC from single oocytes were analyzed and correlated to pregnancy outcome using a qRT-PCR platform. Multisite retrospective study. Gene expression in CC from single oocytes were analyzed and correlated to pregnancy outcome using a qRT-PCR platform. Materials and methodsPatients < 36 years from three clinical sites undergoing either IVF or ICSI and patient-specific stimulation protocols were recruited for the study (n=55). Gene expression of a selected group of genes in CC corresponding to 86 individual oocytes was analyzed using qRT-PCR. Two independent CC sample sets, A (n=49) and B (n=37), from two different sets of patients, were evaluated. CC samples were divided into two groups: "live birth samples" corresponded to CC of single oocytes which produced embryos that resulted in healthy live birth; "nonpregnant samples" corresponded to CC of single oocytes which produced embryos that were transferred into the patient but failed to implant. Patients < 36 years from three clinical sites undergoing either IVF or ICSI and patient-specific stimulation protocols were recruited for the study (n=55). Gene expression of a selected group of genes in CC corresponding to 86 individual oocytes was analyzed using qRT-PCR. Two independent CC sample sets, A (n=49) and B (n=37), from two different sets of patients, were evaluated. CC samples were divided into two groups: "live birth samples" corresponded to CC of single oocytes which produced embryos that resulted in healthy live birth; "nonpregnant samples" corresponded to CC of single oocytes which produced embryos that were transferred into the patient but failed to implant. ResultsPreviously, a candidate set of <250 genes was identified on microarray as predictive of pregnancy outcome. A selection of these genes was validated on sample set A using qRT-PCR, and a refined panel of <20 genes was identified as predictive of pregnancy outcome. The predictive power of this signature was then tested on sample set B. An established method for determining diagnostic utility of a test, Receiver Operating Characteristic analysis, was employed, and confirmed predictive power of the signature (P<0.0001). Previously, a candidate set of <250 genes was identified on microarray as predictive of pregnancy outcome. A selection of these genes was validated on sample set A using qRT-PCR, and a refined panel of <20 genes was identified as predictive of pregnancy outcome. The predictive power of this signature was then tested on sample set B. An established method for determining diagnostic utility of a test, Receiver Operating Characteristic analysis, was employed, and confirmed predictive power of the signature (P<0.0001). ConclusionTranscriptional profiling of CC across multiple sites and stimulation protocols can be used to predict those oocytes and embryos with the capacity to advance to healthy live birth. Transcriptional profiling of CC across multiple sites and stimulation protocols can be used to predict those oocytes and embryos with the capacity to advance to healthy live birth.

The low developmental competence of somatic cell nuclear transfer (SCNT)-derived embryos has been attributed to aberrant or incomplete epigenetic reprogramming. Tri-methylation of Histone H3 at lysine 27 (H3K27me3) is established by polycomb group genes (Ezh2, Eed, and Suz12) and is associated with stable and heritable gene silencing. The aim of this study was to characterize the expression of polycomb genes during bovine preimplantation development and to compare the dynamics of H3K27me3 after IVF, parthenogenetic activation (PG), and SCNT. Embryos were produced using our standard protocols (Ross et al. 2006 Biotechniques 41, 741–750). MII oocytes and pronuclear-, 2-cell-, 4-cell-, 8-cell-, morula-, and blastocyst-stage embryos were collected for quantitative RT-PCR or immunofluorescence (IF). IF analysis of embryos was performed using primary antibodies against H3K27me3, Eed, Ezh2 (Abcam, Cambridge, MA, USA), and Suz12 (Santa Cruz Biotechnology, Santa Cruz, CA, USA) and imaged by confocal microscopy. The average background-corrected nuclear fluorescence intensity for each embryo was analyzed by ANOVA using the MIXED procedure of SAS (SAS Institute, Inc., Cary, NC, USA). Ezh2, Eed, Suz12, and control transcript levels from oocytes and IVF-derived embryos were determined by absolute quantification as described by Bettegowda et al. (2006 Mol. Reprod. Dev. 73, 267–278). RNA levels were normalized to the external control and analyzed by ANOVA (mixed model). A steady decrease in total transcripts was observed from the MII to the 8-cell stage (P < 0.05). During this period, the levels of polyadenylated Suz12 and Eed transcripts remained steady, while Ezh2 transcripts increased at the 2-cell stage (P < 0.05) and then decreased from the 2- to 4-cell stage (P = 0.10). After embryonic genome activation, all three genes displayed an increase in transcript abundance by the blastocyst stage (P < 0.05). Concurrent IF analysis revealed that Ezh2 localized to the nucleus throughout preimplantation development, Eed localized to the cytoplasm at the 2-, 4- and 8-cell stages (expression confirmed by westernblot) and to the nucleus at morula and blastocyst stages, and Suz12 was exclusively nuclear and detected only at morula and blastocyst stages. IF analysis of H3K27me3 revealed that, for all embryo types, nuclear fluorescent intensity was highest at the MII stage, steadily decreased to reach a nadir at the 8-cell stage (P < 0.05), and then significantly increased at the blastocyst stage (P < 0.05). H3K27me3 fluorescent intensity was significantly higher in SCNT blastocysts than in IVF and PG blastocysts (P < 0.05). We conclude that H3K27me3 is efficiently removed from the embryonic and somatic cell chromatin, but is abnormally reestablished in SCNT embryos.

Faulty epigenetic reprogramming is a likely major cause of the low success rate observed in all mammals produced through somatic cell nuclear transfer (SCNT). It has been reported that treatment of reconstructed mouse embryos with the potent histone deacetylase inhibitor, trichostatin A (TSA), results in significantly increased developmental capacity of SCNT preimplantation embryos and live offspring (Kishigami et al. 2006 Biochem. Biophys. Res. Commun. 240, 183–189; Rybouchkin et al. 2006 Biol. Reprod. 74, 1083–1089; Kishigami et al. 2006 J. Reprod. Dev. 53, 165–170). Studies investigating similar reprogramming capabilities of TSA in bovine SCNT embryos report conflicting results (Akagi et al. 2007 Reprod. Fertil. Dev. 19, 24 abst; Iwamoto et al. 2007 Reprod. Fertil. Dev. 19, 48 abst). In this study, the effects of TSA treatment on in vitro development of bovine SCNT embryos were examined. Bovine fetal fibroblasts were cultured under contact inhibition for 2 to 5 days and used as donor cells for SCNT. Oocytes were aspirated from abattoir-derived ovaries, and matured in vitro for 18 h prior to enucleation. Reconstructed SCNT couplets were electrofused, and then activated 24 h post-maturation using 5 µm ionomycin followed by 2 mm dimethylaminopurine (DMAP) for 4 h. SCNT embryos were subjected to 0 (control; C-NT) or 50 nm TSA for 13 h post-ionomycin (hpi) TSAa-NT) or 13 hpi + 6 h starting from 40 hpi (TSAb-NT). IVF embryos were produced as an additional control. All embryos were cultured in KSOM supplemented with 3 mg mL–1 BSA for 7.5 days, with 5% FBS added on Day 3. Experiments were repeated 3 or 7 times, and data were analyzed a -way ANOVA procedure. Developmental rates to the blastocyst stage and total cell number of blastocysts were determined. Total cell numbers were determined by fixing blastocysts in 4% paraformaldehyde, and staining with bisbenzimide 33342, followed by microslide mounting and visualization using an epifluorescence microscope. No difference was observed in cleavage rates among the four treatment groups, C-NT, TSAa-NT, TSAb-NT, and IVF, with the rates being 66%, 75%, 73.1%, and 82.3%, respectively (P = 0.33); nor was any improvement seen in the rate of blastocyst development of TSAa-NT or TSAb-NT over C-NT embryos: 36%, 40.2%, and 30.2%, respectively (P = 0.22). Furthermore, there was no significant difference in mean total cell number of blastocysts among treatment groups: C-NT, 120.2; TSAa-NT, 124.2; TSAb-NT, 129.3; and IVF, 141.1 (P = 0.29). These results suggest that 50 nm TSA treatment immediately following activation does not affect the development of bovine SCNT preimplantation embryos.