Epigenetically immature oocytes lead to loss of imprinting during embryogenesis.

Loss of imprinting (LOI) is occasionally observed in human imprinting disorders. However, the process behind the LOI is not fully understood. To gain a better understanding, we produced embryos and pups from mouse oocytes that lacked a complete methylation imprint using a method that involved transferring the nuclei of growing oocytes into the cytoplasm of enucleated fully grown oocytes following in vitro fertilization (IVF). We then analyzed the imprinting statuses. Our findings show that the incomplete methylation imprint derived from growing oocytes results in epigenetic mosaicism or a loss of methylation imprint (LOM) at maternal alleles in embryos. In some embryos, both hypo- and hypermethylated maternal Kcnq1ot1 alleles were detected, whereas either hypo- or hypermethylated maternal Kcnq1ot1 alleles were detected in others. Such tendencies were also observed at the Igf2r and Mest loci. Gene expression levels of imprinted genes were linked with their methylation statuses in some but not all embryos. Possible explanations of the inconsistency between the data from DNA methylation and gene expression include epigenetic mosaicism in embryos. Pups were successfully produced from growing oocytes at a quite low frequency. They exhibited an obese phenotype and LOI with respect to Igf2r, Snrpn and Mest. Our finding suggests the possibility that LOI/LOM at maternal alleles in human concepti could be derived from epigenetically immature/mutated oocytes.

Expression and localization of cFLIP, an anti-apoptotic factor, in the bovine corpus luteum.

The objective of the present study was to investigate the potential mechanisms regulating cellular FLICE-like inhibitory protein (cFLIP), an anti-apoptotic factor, in the bovine corpus luteum (CL). Expression of cFLIP mRNA was highest at the developing stage and then decreased significantly during the mid, late and regressed stages (P<0.05). Western blot analysis revealed that expression of the long isoform of cFLIP (cFLIP(L)) protein was high during the early and developing luteal stages, remained steady during the mid and late luteal stages and then decreased significantly (P<0.05) by the regressed stage. However, the expression levels of the short isoform of cFLIP (cFLIP(S)) remained low during the early, developing and mid luteal stages. Immunostaining of cFLIP was strongest in the cytoplasm of luteal and non-luteal cells, including endothelial and immune cells, remained high during the early, developing and mid luteal stages and then decreased significantly (P<0.05) in the late and regressed luteal stages. Immunostaining of cFLIP was observed only in macrophage-like cells in the regressing CL. However, cultured mid luteal cells had a higher percentage of cFLIP-positive cells and a lower percentage of TUNEL-positive cells than luteal cells treated with tumor necrosis factor alpha (TNF)/interferon gamma (IFNG; P<0.01). These results indicate downregulation of cFLIP during structural luteal regression, suggesting that cFLIP plays a survival role in the bovine CL.

The expression of the nuclear receptors NR5A1 and NR5A2 and transcription factor GATA6 correlates with steroidogenic gene expression in the bovine corpus luteum.

The corpus luteum (CL) is the major site of progesterone (P4) production during the luteal phase of the estrous cycle in cattle. To better understand the molecular mechanisms underlying P4 production, we compared the mRNA and protein expression profiles of key components of the steroidogenic pathway (StAR, CYP11A, and 3beta-HSD) during the bovine CL luteal phase with that of several transcription factors (NR5A1, NR5A2, GATA4, GATA6) known for their roles in the control of steroidogenic gene expression. In the bovine CL, StAR, CYP11A, and 3beta-HSD mRNA and protein levels remained constant at the mid and late luteal phases but markedly declined at the regressed luteal stage. NR5A1 and NR5A2 exhibited a similar pattern with a significant decrease in expression at the regressed luteal stage. Both GATA4 and GATA6 mRNA and proteins could be detected in bovine CL; GATA6 levels, however, were generally higher. Although GATA4 expression did not change during the luteal phase, GATA6 showed a marked decrease at the regressed luteal stage, like NR5A1, NR5A2, and the other steroidogenic markers. Thus, we suggest that NR5A1, NR5A2, and GATA6, but not GATA4, contribute to the transcriptional regulation of steroidogenic gene expression, and hence P4 production, in the bovine CL. Furthermore, we have demonstrated the association of NR5A1 and NR5A2 with the bovine StAR promoter in the mid-luteal CL using chromatin immunoprecipitation, suggesting that these factors have definitive roles in the regulation of StAR gene transcription in vivo.

Lysophosphatic acid modulates prostaglandin secretion in the bovine uterus.

Lysophosphatidic acid (LPA) modulates prostaglandin (PG) synthesis via LPA receptor 3 (LPAR3) in the murine endometrium. The lack of functional LPAR3 in mice may lead to embryo mortality. In the present study, we examined the role of LPA in the bovine uterus. We confirmed that LPA is locally produced and released from the bovine endometrium. Moreover, there are enzymes involved in LPA synthesis (phospholipase (PL) D(2) and PLA2G1B) in the bovine endometrium during estrous cycle and early pregnancy. Expression of the receptor for LPA (LPAR1) was positively correlated with the expression of PGE(2) synthase (PGES) and negatively correlated with the expression of PGF(2alpha) synthase (aldose reductase with 20 alpha-hydroxysteroid dehydrogenase activity - PGFS) during early pregnancy. In vivo LPA induced P4 and PGE(2) secretion was inhibited by LPAR1 antagonist (Ki16425). The overall results indicate that LPA is locally produced and released from the bovine endometrium. Moreover, LPAR1 gene expression in the endometrium during the estrous cycle and early pregnancy indicates that LPA may play autocrine and/or paracrine roles in the bovine uterus. LPAR1 gene expression is positively correlated with the expression of the enzyme responsible for luteotropic PGE(2) production (PGES) in endometrium. In cow, LPA stimulates P4 and PGE(2) secretion. Thus, LPA in the bovine reproductive tract may indirectly (via endometrium) or directly support corpus luteum action via the increase of P4 synthesis and the increase of PGE(2)/PGF(2)(alpha) ratio. It suggests that LPA may serve as an important factor in the maintenance of early pregnancy in cow.

Cortisol is a suppressor of apoptosis in bovine corpus luteum.

Glucocorticoid (GC) acts as a modulator of physiological functions in several organs. In the present study, we examined whether GC suppresses luteolysis in bovine corpus luteum (CL). Cortisol (an active GC) reduced the mRNA expression of caspase 8 (CASP8) and caspase 3 (CASP3) and reduced the enzymatic activity of CASP3 and cell death induced by tumor necrosis factor (TNF) and interferon gamma (IFNG) in cultured bovine luteal cells. mRNAs and proteins of GC receptor (NR3C1), 11beta-hydroxysteroid dehydrogenase type 1 (HSD11B1), and HSD11B2 were expressed in CL throughout the estrous cycle. Moreover, the protein expression and the enzymatic activity of HSD11B1 were high at the early and the midluteal stages compared to the regressed luteal stage. These results suggest that cortisol suppresses TNF-IFNG-induced apoptosis in vitro by reducing apoptosis signals via CASP8 and CASP3 in bovine CL and that the local increase in cortisol production resulting from increased HSD11B1 at the early and midluteal stages helps to maintain CL function by suppressing apoptosis of luteal cells.

Hypoxia promotes luteal cell death in bovine corpus luteum.

Low oxygen caused by a decreasing blood supply is known to induce various responses of cells, including apoptosis. The present study was conducted to examine whether low-oxygen conditions (hypoxia) induce luteal cell apoptosis in cattle. Bovine midluteal cells incubated under hypoxia (3% O(2)) showed significantly more cell death than did those incubated under normoxia (20% O(2)) at 24 and 48 h of culture, and had significantly lower progesterone (P4) levels starting at 8 h. Characteristic features of apoptosis, such as shrunken nuclei and DNA fragmentation, were observed in cells cultured under hypoxia for 48 h. Hypoxia increased the mRNA expressions of BNIP3 and caspase 3 at 24 and 48 h of culture. Hypoxia had no significant effect on the expressions of BCL2 and BAX mRNA. Hypoxia also increased BNIP3 protein, and activated caspase-3. Treatment of P4 attenuated cell death, caspase-3 mRNA expression, and caspase-3 activity under hypoxia. Overall results of the present study indicate that hypoxia induces luteal cell apoptosis by enhancing the expression of proapoptotic protein, BNIP3, and by activating caspase-3, and that the induction of apoptosis by hypoxia is partially caused by a decrease in P4 production. Because hypoxia suppresses P4 synthesis in bovine luteal cells, we suggest that oxygen deficiency caused by a decreasing blood supply in bovine corpus luteum is one of the major factors contributing to both functional and structural luteolysis.

The role of glucocorticoid in the regulation of prostaglandin biosynthesis in non-pregnant bovine endometrium.

To determine whether glucocorticoids (GCs) play a role in regulating uterine function in cow, the present study examined the expression of mRNA encoding GC receptor (GC-R) alpha, 11beta-hydroxysteroid dehydrogenase (11-HSD) type 1 and type 2, and the activity of 11-HSD1 in bovine endometrial tissue throughout the estrous cycle. We also studied the effects of cortisol on basal, oxytocin (OT)- and tumor necrosis factor-alpha (TNFalpha)-stimulated prostaglandin (PG) production. A quantitative real-time PCR analysis revealed that GC-Ralpha mRNA was expressed more strongly in the mid-luteal stage (days 8-12) than in the other stages. In contrast to GC-Ralpha mRNA expression, 11-HSD1 mRNA expression was greater in the follicular stage than in the other stages, whereas 11-HSD2 mRNA expression was lowest in the follicular stage. The activity of 11-HSD1 was greater in the follicular stage and estrus than in the other stages and was lowest in the mid-luteal stage. Cortisone was dose-dependently converted to cortisol in the cultured endometrial tissue. Although cortisol did not affect either the basal or OT-stimulated production of PGs in the cultured epithelial cells, the production of PGs stimulated by TNFalpha in the stromal cells was suppressed by cortisol (P < 0.05). Cortisol suppressed basal prostaglandin (PG)F2alpha without affecting basal PGE2 production in the stromal cells. The overall results suggest that the level of cortisol is locally regulated in bovine endometrium throughout the estrous cycle by 11-HSD1, and that cortisol could act as a luteoprotective factor by selectively suppressing luteolytic PGF2alpha production in bovine endometrium.

DNA methylation imprints on the IG-DMR of the Dlk1-Gtl2 domain in mouse male germline.

Mouse genomes show a large cluster of imprinted genes at the Dlk1-Gtl2 domain in the distal region of chromosome 12. An intergenic-differentially methylated region (IG-DMR) located between Dlk1 and Gtl2 is specifically methylated in the male germline; IG-DMR regulates the parental allele-specific expression of imprinted genes. Here, we show the resetting of IG-DMR methylation marks during male germ-cell differentiation. For parental allele-specific methylation analysis, polymorphisms were detected in a 2.6-kb IG-DMR in three mouse strains. Bisulfite methylation analysis showed erasure of the marks by E14 and re-establishment before birth. The IG-DMR methylation status was maintained in spermatogonia and spermatocytes of mature testes. The IG-DMR methylation status established before birth is thus maintained throughout the lifetime in the male germline.

Effects of storage and passage of bovine luteal endothelial cells on endothelin-1 and prostaglandin F2alpha production.

To establish a storage system for isolated bovine luteal endothelial cells (LECs), we investigated the basal and tumor necrosis factor (TNF) alpha-stimulated production of endothelin-1 (ET-1) and prostaglandin (PG) F2alpha in unfrozen and frozen-thawed LECs until passage 10. LECs were obtained from developing corpora lutea (CL; days 5-7 of the estrous cycle) using enzymatic digestion and magnetic beads coated with lectin BS-1. The LECs were frozen at -80 C or further cultured and/or passaged until passage 10 in DMEM/Ham's F-12 supplemented with 10% calf serum. The hormonal productions of unfrozen and frozen/thawed LECs were compared through passages 2-10. When both the unfrozen and frozen/thawed cells reached confluence, the culture medium was replaced with fresh medium containing 0.1% bovine serum albumin (BSA), and the cells were incubated with TNFalpha (50 ng/ml) for 12 h. The basal productions of ET-1 and PGF2alpha by the unfrozen and frozen/thawed LECs were similar at passage 2. The basal production of PGF2alpha by LECs was not altered by passage and storage at -80 C, whereas the basal production of ET-1 decreased from passage 2 and 3 to passage 4 in the unfrozen LECs and from passage 2 to passage 3 in the frozen/thawed LECs. However, production of ET-1 by the unfrozen and frozen/thawed LECs was similar between passages 4-10 and passages 3-10, respectively. Exposure of LECs to TNFalpha increased (P<0.05) ET-1 and PGF2alpha production by the unfrozen and frozen-thawed LECs in all passages examined. Thus, LECs obtained from developing CLs and stored until passage 10 can be used for study of the physiology of LECs in vitro.

Disruption of parental-specific expression of imprinted genes in uniparental fetuses.

In mammals, imprinted genes show parental origin-dependent expression based on epigenetic modifications called genomic imprinting (GI), which are established independently during spermatogenesis or oogenesis. Due to GI, uniparental fetuses never develop to term. To determine whether such expression of imprinted genes is maintained in uniparental mouse fetuses, we analyzed the expression of 20 paternally and 11 maternally expressed genes in androgenetic and parthenogenetic fetuses. Four genes of each type were expressed in both groups of fetuses. Furthermore, quantitative analysis showed that expression levels deviated from the presumed levels for some imprinted genes. These results suggest that mechanisms acting in trans between paternal and maternal alleles are involved in the appropriate expression of some imprinted genes.

Possible roles of intracellular cyclic AMP, protein kinase C and calcium ion in the apoptotic signaling pathway in bovine luteal cells.

Structural luteolysis occurs by apoptosis of luteal cells. The present study examined the effects of activators of well-characterized second messengers on Fas and caspase-3 mRNA expression and on P4 production in luteal cells in order to trace the pro- and anti-apoptotic factors in the bovine corpus luteum (CL). Cultured bovine mid luteal cells were treated for 24 h with a cyclic AMP analogue (8-bromo cyclic AMP; 8br-cAMP; 2.5 mM), a protein kinase C (PKC) activator (phorbol 12-myristate 13-acetate; PMA; 10 microM), or calcium ionophore (A23187; 10 microM). Fas and caspase-3 mRNA expression was inhibited by 8br-cAMP and PMA but was increased by A23187 (P<0.05). In addition, P4 production by bovine luteal cells was stimulated by 8br-cAMP and PMA, whereas it was inhibited by A23187, compared with untreated controls (P<0.05). The overall results suggest that cAMP and PKC suppress apoptosis in bovine luteal cells through inhibition of Fas and caspase-3 mRNA expression and through stimulation of P4 production. Therefore, substances that activate cAMP or PKC may act as survival factors in the bovine CL. Furthermore, substances that mobilize Ca2+ may act as apoptotic factors by stimulating Fas and caspase-3 expression in the bovine luteal cells.

Oocyte growth-dependent progression of maternal imprinting in mice.

In mammals, some genes categorized as imprinted genes are exclusively expressed either from maternal or paternal allele. This parental-origin-specific gene expression is regulated by epigenetic modification of DNA methylation in differentially methylated region (DMR), which is independently imposed during oogenesis and spermatogenesis. It is known that methylation of DMR in the female germ line is established during oocyte growth phase. However, the cause of the progression of methylation on DMR, due to either aging of mice or growth-size of oocyte was unclear up to now. Here, we analyzed the methylation of DMR for each eight imprinted genes (Igf2r, Lit1, Zac1, Snrpn, Peg1/Mest, Impact, Meg1/Grb10, and H19) by bisulfite sequencing methylation assay, using oocytes from 10 dpp (days post partum), 15 dpp, 20 dpp, and adult mice. To find whether the size of oocytes is the cause of methylation, above oocytes were classified into seven groups (each oocyte diameter ranging from 40 to 75 microm with intervals of 5 microm). The results from juvenile mice oocytes showed that DMR methylation progressed according to oocyte growth each imprinted gene. More than 85% of DMR methylation was achieved for both Igf2r, Zac1 & Lit1 with oocyte size of reaching 55 microm and Snrpn, Peg1/Mest, Impact, and Meg1/Grb10 with oocyte size of reaching 60 microm. Preferential methylation of maternal allele was observed in Zac1 and Peg1/Mest of juvenile oocytes and in Snrpn of juvenile and adult oocytes. The oocyte size-dependent-methylation progressed equally for all three different-age juvenile mice. The size-dependent-methylation was also recognized in the growing oocytes collected from adult mice, although the progress is slightly slower than that of juvenile mice. From these results, we concluded that DNA methylation is established with oocyte size dependent manner, not with aging of mice.

Nuclei of oocytes derived from mouse parthenogenetic embryos are competent to support development to term.

Mouse parthenotes result in embryonic death before 10 days of gestation, but parthenogenetic embryos (ng/fg PE) that contain haploid sets of genomes from nongrowing (ng) oocytes derived from newborn fetuses and fully grown (fg) oocytes derived from adults can develop into 13.5-day-old fetuses. This prolonged development is due to a lack of genomic imprinting in ng oocytes. Here, we show maternal genomes of oocytes derived from ng/fg PE are competent to support normal development. After 28 days of culture, the ovaries from ng/fg PE grew as well as the controls, forming vesicular follicles with follicular antrums. The oocytes collected from the developed follicles were the same size as those of the controls. To determine whether maternal primary imprinting had been established in the oocytes derived from ng/fg PE, we examined the DNA methylation status in differentially methylated regions of three imprinted genes, Igf2r, Lit1, and H19. The results showed that maternal-specific modifications were imposed in the oocytes derived from ng/fg PE. Further, to assess nuclear competence to support development, we constructed matured oocytes containing a haploid genome derived from ng/fg PE oocytes by serial nuclear transfer. After in vitro fertilization and culture and embryo transplantation into recipients, two live pups were obtained. One developed normally to a fertile adult. These results revealed that oocytes derived from ng/fg PE can be normally imprinted during oogenesis and acquire competence to participate in development as female genomes.