Nonspecies-specific effects of mouse oviducts on the development of bovine IVM/IVF embryos by a serum free co-culture

In Experiment 1, development of bovine embryos derived from in vitro-matured (IVM) and in vitro-fertilized (IVF) oocytes was examined under 4 culture conditions: 1) co-culture with mouse ampullae continuously for 8 d, 2) co-culture with mouse ampullae that were replaced with fresh ampullae at 48-h intervals, 3) co-culture with bovine granulosa cell monolayers, and 4) culture in medium alone. Culture medium consisted of tissue culture medium 199 (TCM-199) supplemented with 1% fetal calf serum (FCS). Inseminated oocytes were transferred to each of the culture treatment 24 h after insemination and were cultured for 8 d. The number of blastocysts per number of cleaved ova obtained after co-culture with mouse ampullae (42.9%) was significantly (P<0.05) higher than that obtained after co-culture with granulosa cell monolayers (28.3%) or culture without cells (4.2%). In Experiment 2, the developmental ability of bovine IVM/IVF embryos co-cultured with mouse ampullae supplemented with or without serum was examined. When serum was excluded from the culture medium, 26.4% (33 125 ) of the total number of embryos cultured were able to develop to the blastocysts stage using this co-culture system. This value was comparable to that obtained in a serum-supplemented co-culture system (30.7%; 39 125 ). In addition, the developmental ability of embryos that reached to the 4-cell stage or beyond at 46 to 48 h after insemination was not significantly different when the embryos were co-cultured with mouse ampullae with (38.5 vs 44.6%) or without (37.0 vs 33.8%) serum.

180 EFFECT OF Dnmt1p mRNA KNOCK DOWN ON Dnmt1 PROTEIN TRANSLATION IN MOUSE TESTIS

We have reported that there was a significant reduction (P < 0.05) in the relative amount of Dnmt1p mRNA in spermatozoa from aged male mice (Kato et al. 2007 Reprod. Fertil. Dev. 19, 277 abst.). The reduction of mRNA levels of Dnmt1p in spermatozoa would lead to altered epigenetic modification of the genome. Dnmt1p is one of 5′ exon alternative isoforms of Dnmt1 and its mRNA is specifically expressed in the pachytene spermatocyte. However, the function of Dnmt1p still has not been elucidated. In this study, we tried to elucidate the function of Dnmt1p in the male mouse reproductive system. This was accomplished by suppressing the expression level of Dnmt1p in the whole testis by short hairpin RNA (shRNA) expression, which was specifically designed from mouse Dnmt1p mRNA. Dnmt1p cDNA was cloned from total RNA extracted from a piece of testis from a C57BL/6J male mouse. Four shRNA expression vectors were constructed and the knock-down efficiency of each shRNA expression vector was evaluated by flow cytometry. From these results, the 589 shRNA expression vector was picked out for further experimentation. The 589 shRNA expression vector was linearized and injected into the pronuclei of C57BL/6J mouse embryos. After injection, the embryos were cultured for 24 h and cleavage was evaluated. Cleaved embryos were transferred into oviducts of recipient ICR mice. After 18 to 19 days, fetuses were delivered by C-section. Two weeks after the birth, the existence of the 589 shRNA expression gene construct in its genome was evaluated by PCR. From founders with 589 shRNA expression gene construct in its genome, finally 1 TG strain was established and used for further experimentation. Two F2 -generation male mice with the 589 shRNA expression gene construct, 2 F2-generation male mice without the 589 shRNA expression gene construct and 1 C57BL/6J wild-type male mouse were used for evaluating the expression level of Dnmt1p mRNA in the whole testis by quantitative PCR. Then, thin sections of testis derived from the F2-generation mouse, which showed a suppressed expression level of Dnmt1p, was evaluated by immunostaining for Dnmt1 protein. The survival rate of mouse embryos after gene injection was 76.8% (202/263) and the cleavage rate of gene-injected embryos was 69.8% (141/202). The developmental rate of transferred embryos to the birth was 19% (27/139). The rate of newborn mice with the 589 shRNA expression gene construct was 37% (10/27). The fertility of established TG strain mouse was normal and there was no abnormality in the thin section figure of testis stained with hematoxylin-eosin double staining method. The relative expression level of Dnmt1p mRNA in the whole testis of the F2 TG mouse was ∼25% of C57BL/6J wild-type male mouse (P < 0.05, Student's t-test). There was Dnmt1 protein in seminiferous tubules, especially in spermatids of C57BL/6J wild-type male mouse. However, there was no Dnmt1 protein in seminiferous tubules of the F2 TG mouse. From these results, it was concluded that the expression of Dnmt1p mRNA was associated in some way with the translation of the Dnmt1 protein in the mouse testis.

62 RECOVERY OF CELL NUCLEI FROM 15 000-YEAR-OLD MAMMOTH TISSUES AND INJECTION INTO MOUSE ENUCLEATED MATURED OOCYTES

Since the report of “Dolly” (Wilmut et al. 1997 Nature 385, 810-813), cloned animals have been successfully produced by somatic cell nuclear transfer (SCNT). In those reports, cultured or fresh cells were mainly used as nuclear donor cells in SCNT. Meanwhile, there was a report of the birth of offspring by SCNT with the nuclear donor cell derived from frozen whole body kept at -20°C for 16 years without cryoprotectant (Wakayama et al. 2008 Proc. Natl. Acad. Sci. USA 105, 17 318-17 322). This report suggests that if tissues or cells are cryopreserved postmortem without cryoprotectant, resurrection of the postmortem animal individual might be possible by SCNT. Furthermore, in cases of extinct or endangered species, it is difficult to use oocytes derived from same species as the recipient cytoplasm. However, there was a report of the birth of offspring by SCNT with oocytes derived from closely related species (Lanza et al. 2000 Cloning 2, 79-90). This result suggests that if oocytes from closely related species are available as the recipient cytoplasm, it would be possible to resurrect extinct or endangered species by SCNT. Woolly mammoth (Mammuthus primigenius) is a famous animal that existed during the ice ages and became extinct at the end of the last ice age. Its relics have been excavated from permafrost. Recently, the prospect of the resurrection of woolly mammoth by SCNT has heightened. However, even if the soft tissue of the woolly mammoth is excavated, it is not clear whether its cell nuclei retain their biological characteristics for more than several thousand years. In this study, we recovered cell nuclei from 15 000-year-old mammoth tissues and injected them into mouse enucleated matured oocytes as the nuclear donor in SCNT. Frozen mammoth legs were excavated from Northeast Siberian permafrost at the point of 71°34′56.9″ N and 141°37′37.6″ E. The conventional radiocarbon age of the sample after 13C correction was 13 100 to 12800 BC or 12400 to 11 800 BC. Cell nucleus-like structures were successfully recovered from skin and muscle tissues. Those nuclei were then injected into enucleated BDF1 mouse oocytes, and more than half of the injected oocytes were able to survive (67%; 12/18 for oocytes injected with skin-derived nucleus, 55%; 72/131 for oocytes injected with muscle-derived nucleus and 33%; 26/80 for oocytes injected with frozen-thawed mouse bone marrow-derived nucleus as the control group). Injected nuclei were not taken apart and retained their nuclear structure. Those oocytes did not show disappearance of nuclear membrane or premature chromosome condensation at 1 h after injection and did not form pronuclear-like structures at 7 h after injection. Because one-half of the oocytes injected with nuclei derived from frozen-thawed mouse bone marrow cells were able to form pronuclear-like structures (46%; 12/26), it might be possible to promote the cell cycle of nuclei from ancient animal tissues by suitable pretreatment in SCNT. This is the first report of SCNT with nuclei derived from mammoth tissues. This study was funded in part by Gifu Prefecture, Japan.

70 FIBROBLAST GROWTH FACTOR 4 PROMOTES THE DEVELOPMENT OF SOMATIC CELL NUCLEAR TRANSFER EMBRYOS IN MICE

Somatic cell nuclear transfer (SCNT) embryos can develop during the preimplantation period; however, most of these die after implantation period. A transcription factor, Cdx2, promotes differentiation of extraembryonic tissues and appears to be involved in the segregation of inner cell mass (ICM) and trophectoderm (TE) in preimplantation embryos. So far, we have demonstrated that the expression of Cdx2 in mouse SCNT embryos is delayed and its expression level is significantly lower than that in intracytoplasmic sperm injection (ICSI) embryos. Moreover, the ectopic expression of Oct-3/4 was observed in the TE tissues of SCNT blastocysts, but not in ICSI blastocysts. Fibroblast growth factor (FGF) receptor 2 (FGFR2) is specifically expressed in 8-cell to morula-stage embryos and trophectoderm (TE) and is essential for implantation; however, FGFR2 expression in SCNT embryos significantly decreases compared with IVF embryos. Therefore, it is likely that abnormality of differentiation that is controlled in development of pre-implantation in SCNT embryos leads to a rapid decrease of subsequent developmental ability. Then, we investigated the effects of FGF4 on development of SCNT embryos. Mouse SCNT embryos were produced according to the method reported previously (Wakayama et al. 1998). B6D2F1 and B6C3F1 female mice were used for the collection of recipient oocytes and donor cells, respectively. Data were analyzed by Student’s t-test. First, the timing to start adding FGF4 was decided by FGFR2 expression time about 54 h after cell injection and treated for 3, 6, 12, 24, and 42 h thereafter. In the case of FGF4 concentration at 25 ng mL-1 with treating time of 6 h from the 4- to 8-cell stages, SCNT embryos significantly promoted the development to morula and blastocyst stages (91 and 45%, respectively) compared with IVF embryos (80 and 30%, respectively; P < 0.05). However, longer treatment of 42 h with FGF4 made their morphology considerably worse. Then, concentrations of FGF4 at 5, 25, 50, 250, and 500 ng mL-1 with treating time of 6 h was examined. In case of FGF4 concentration at 25 and 50 ng mL-1, SCNT embryos significantly promoted the development to morula and blastocyst stages (P < 0.05). Immunohistochemical analysis showed segregation of the expression of Oct-3/4 and Cdx2 in ICM and TE, respectively, in FGF4-treated SCNT embryos, unlike in the case of nontreated SCNT embryos, which showed an ectopic expression of Oct-3/4 in TE tissues. Furthermore, after the transplantation of SCNT embryos treated with FGF4 at 50 ng mL-1 and the treating time of 6 h to recipient mice, most of the transferred embryos implanted and cloned mice were successfully produced as well as nontreated SCNT embryos. Therefore, FGF4 facilitates the development of SCNT embryos especially to the morula and blastocyst stages. This work was supported by a Grant-in-Aid for Scientific Research from the Japan Society for the Promotion of Science.

Early morphological events of in vitro fertilized bovine oocytes with frozen-thawed spermatozoa

Bovine follicular oocytes were matured and inseminated in vitro with spermatozoa capacitated in vitro. The first evidence of sperm penetration was observed at 3 h after insemination. The penetration rate increased until 5 h, and reached a maximum rate (92%) at 5 h. Decondensation of the sperm head and pronuclear formation were observed 4 h and 7 h after insemination, respectively. Female chromatins of all penetrated oocytes were activated at 3 h, and female pronuclei were formed at 7 h after insemination. Percentages of oocytes with male and female pronuclei at 9 h were 88 and 94%. Polyspermy (4, 7, 19 and 29% at 4, 5, 7 and 9 h after insemination, respectively) and abnormal development of male pronuclei (6 and 7% at 7 and 9 h after insemination, respectively) were also seen.

Effect of visual light on in vitro embryonic development in the hamster

The effect of light exposure during collection and culture of hamster embryos on their subsequent development in vitro was examined. When embryos were collected under dark conditions (70 lux) within 10 minutes and then cultured in a HECM-1 medium in 5% CO2 in air, the developmental rates of 1-cell embryos to the 4- and 8-cell stages were 88.6% (93/105) and 66.7% (70/105), respectively. These rates were significantly higher than those under light conditions (1600 lux): 51.9% (56/108) and 34.3% (37/108). Light irradiation during the culture of 1-cell embryos suppressed subsequent development. The degree of suppression correlated inversely with duration of light irradiation, and light irradiation of 30 minutes or more completely blocked development to the 2-cell stage. When 1-cell embryos were irradiated through a yellow filter, cutting the light wavelengths to less than 500 nm, embryonic development was still suppressed. However, the degree of the suppression varied and 45.7% (53/116), 6.0% (7/116), and 0.9% (1/116) of the embryos developed to the 2-, 4-, and 8-cell stages, respectively, under 30 minute light irradiation. Inhibitory effects of light irradiation on the development of 2- and 8-cell embryos were also observed, showing an inverse correlation with duration; the developmental rates of 2-cell embryos to the 8-cell stage under 0, 10, and 30 minutes of irradiation were 85.6% (107/125), 1.6% (2/122), and 0% (0/129), respectively, and those of 8-cell embryos to the blastocyst stage were 79.8% (91/114), 74.8% (86/115), and 0% (0/110), respectively. These findings indicate that early-stage embryos are sensitive to light exposure; however, severe light exposure adversely affects the development of embryos at any stage. Thus, the protection of embryos from light exposure at all stages of embryo manipulation, from collection to culture, is essential.

179 MODULATION OF RHOPHILIN-2 MAY REGULATE THE PROGRESSION OF CELL DIVISION IN FERTILIZED MOUSE EGGS

It has been reported that activity of Rho, one of the GTPases, is essential for division of nuclei and cytoplasm of fertilized mouse eggs. Since it has been reported that alteration of activities of GTPases modifies their ability to attach to each of their effector proteins in somatic cells, effector proteins seem to be able to control not only progression but also repression of cell division by changing their cellular localizations through activities of GTPases. For this reason, Rhophilin-2, one of the effector proteins of Rho, seems to be involved in the decision of progression of division of fertilized mouse eggs. To examine whether this involvement works in fertilized mouse eggs, cellular localization of Rho and Rhophilin-2 in fertilized mouse eggs that were treated with Rho inhibitor were analyzed. Moreover, cellular localization of GABA A receptor association protein (GABARAP), which was identified in our previous study (Matsuoka et al. 2006 Reprod. Fertil. Dev. 18, 176–177) as a protein that interacts with Rhophilin-2, was also analyzed. Fertilized mouse eggs were obtained from in vitro fertilization technique. One group of fertilized eggs was obtained at 24 h after insemination as experimental control. To obtain the mouse eggs in which Rho activities were inhibited, Clostridium botulinum C3 exoenzyme (C3-CB), an inhibitor of Rho activity, was injected into the other group of fertilized mouse eggs at 12 h after insemination, and were collected after 12 h of subsequent culture. Cellular localization of Rho (n = 100), Rhophilin-2 (n = 10). and GABARAP (n = 10) in the collected oocytes was analyzed by using immunofluorescence. Our results showed that Rho and Rhophilin-2 were co-localized at the midbody microtubule, which is an important device for cytoplasmic division in control eggs. However, the inhibition of Rho activity did not modify the co-localization of Rho and Rhophilin-2. On the other hand, localization of GABARAP was modified by the inhibition of Rho activity, and GABARAP was detected around the nuclei of fertilized eggs in which Rho activity was inhibited. In the next experiment, we examined whether interaction of Rhophilin-2 and GABARAP was modified by the inhibition of Rho activity by using a co-immunoprecipitation assay (co-IP) (n = 100). The interaction of Rhophilin-2 and GABARAP was found to disappear after inhibition of Rho activity. These results suggest that activity of Rho seems to regulate cytoplasmic division through Rhophilin-2 modification. Moreover, Rho seem to modulate the nuclear division of fertilized mouse eggs by regulating the interaction between Rhophilin-2 and GABARAP. This study was supported by a Grant-in-Aid for the 21st Century COE Program of the Japan Mext and by a grant for the Wakayama Prefecture Collaboration of Regional Entities for the Advancement of Technology Excellence of the JST.

40 EARLY DEVELOPMENT OF RECONSTRUCTED MOUSE EMBRYOS USING BONE MARROW CELLS FROZEN WITHOUT CRYOPROTECTANT

Animal cells frozen with suitable cryoprotectants have been successfully cryopreserved for long periods of time, maintaining viability upon thawing. Animal cells frozen without cryoprotectant, however, may suffer serious damage and not be useful as donors in somatic cell nuclear transfer (SCNT). However, in some cases, old animal samples were frozen only as a whole body or a piece of tissue without cryoprotectant. If the cells from such old samples could be useful for SCNT, then there are potentially many candidates where individual animals could be reproduced. In this study, we examined the possibility of using mouse bone marrow cells frozen without cryoprotectant as nuclear donors in SCNT. Thigh bones were collected from B6C3F1 mice and frozen in either a –25�C or a –80�C freezer for more than one month. Thawing of frozen bones was performed by placing them in an incubator at 37�C. Bone marrow cells were collected by washing the bone cavity with saline. Recipient oocytes for SCNT were collected from B6D2F1 female mice. The enucleation of recipient oocytes and the injection of nuclei were performed as previously reported (Wakayama et al. 1998 Nature 394, 369–374) with a piezo-actuated micromanipulator system. In this study, 4 groups of mouse cells (fresh bone marrow cells, bone marrow cells frozen at –25�C, bone marrow cells frozen at –80�C, and fresh cumulus cells) were used as the nuclear donors in SCNT. After nuclear injection, embryos were kept in mCZB medium for 1 h at 37�C. Subsequently, embryos were cultured for 3 h with 5 µg mL–1 cytochalasin B and 10 mm SrCl2 for activation and cultured for an additional 20 h in mKSOM medium. The nuclear dynamics of SCNT embryos in each donor cell group was observed using 42,6-diamidino-2-phenylindole (DAPI) staining and a fluorescent microscope at 0, 1, 7, and 24 h after nuclear injection. Data were analyzed by Student's t-test. The cell viability after thawing by trypan blue vital staining was about 20% regardless of freezing temperature. At 7 h after nuclear injection, the SCNT embryos injected with frozen bone marrow cells, regardless of freezing temperature, had more single pronuclei (67%, 54/81; P < 0.05) than SCNT embryos injected with either fresh bone marrow cells (36%, 26/73) or cumulus cells (28%, 67/236). At 24 h after nuclear injection, fewer SCNT embryos injected with bone marrow cells, either fresh or frozen, developed to the 2-cell stage (fresh: 11%, 6/56; frozen at –25�C: 21%, 5/24; frozen at –80�C: 20%, 10/49) than SCNT embryos injected with cumulus cells (58%, 185/319; P < 0.05). There was no difference in the embryonic development to the 2-cell stage among SCNT embryos injected with either fresh or frozen bone marrow cells. Further studies are required to determine whether cells frozen without cryoprotectant are capable of resulting in viable clones.

58 PRODUCTION OF CLONED BOVINE EMBRYOS DERIVED FROM AMNIOTIC CELLS OF PREGNANT COWS

Progeny tests are widely used for selection of sires for beef and dairy cattle. A less costly method might be to clone the sire candidates at their earliest developmental stage possible. To produce cloned bulls, we obtained amniotic cells as donors for nuclear transfer by transvaginal aspiration of pregnant cows. However, the collected cells may include some maternal cells. In this study, we examined collection methods to obtain only fetal cells from the collected fluid. We also examined the developmental capacity of the embryos cloned from these cells. Amniotic fluids were aspirated from pregnant cows by ultrasound-guided aspiration. We collected amniotic fluids from 27 pregnant Japanese black beef cattle (between 58 and 132 days of gestation). In Method 1, cells were recovered from the whole amniotic fluid (approximately 15 mL). In Method 2, the initial 5 mL of aspirated fluid was discarded and then the next 10 mL sample was collected. Cells were recovered from the collected fluids. The cells in the fluids were washed twice by centrifugation and then cultured in AmnioMAX™-II medium (GIBCO, Grand Island, NY, USA). After 3–4 passages, the sex of the cell lines was determined by the loop-mediated isothermal amplification (LAMP) method (Eiken Chemical Co., Ltd., Tokyo, Japan). For the cell lines that were determined as 'male' by the LAMP method we further analyzed the sex of individual cells (137–620 cells of each cell line) by fluorescent in situ hybridization (FISH) using a bovineY chromosome-specific probe (Kobayashi et al. 1998 Mol. Reprod. Dev. 51, 390–394). The percentage of male cells obtained from Methods 1 and 2 were 0–0.4% (from 4 animals) and 93.7–99.5% (from 6 animals), respectively. Then, we used confluent amniotic cells from 3 cell lines obtained by Method 2 as donor cells for nuclear transfer and examined the developmental capacity of the cloned embryos. Bovine fibroblasts cultured under serum starvation were used as a control. The cells were electrically fused (2.7 kV cm–1, 11 µs, 2 times) with enucleated bovine oocytes, and activated with a calcium ionophore and cycloheximide. They were subsequently cultured in mSOF until 168 h post-activation. The data were analyzed with Fisher's protected least-squares difference (PLSD) test following ANOVA. The rates of fusion, cleavage, and development to the blastocyst stage of the cloned embryos were the same as those of the control embryos (78% v. 81%, 75% v. 75%, and 22% v. 27%, respectively; P > 0.05). Furthermore, the rate of male blastocysts derived from the cloned embryos with the three cell lines was 95% (19/20). These results indicate that the amniotic fluids collected from pregnant cows by Method 2 contained fewer maternal cells, and that the embryos cloned from the cells developed in a manner similar to that of embryos cloned from the fibroblasts. This work was supported byWakayama Prefecture CREATE, JST.

163 SEARCH FOR GENES OF WHICH THE AMOUNTS OF TRANSCRIPTS OSCILLATE EVERY 24 h IN THE MOUSE OVARY

Perturbation of circadian rhythm is believed to be detrimental to the physiology of organs, including the mammalian ovary. However, the molecular mechanisms that are regulated by circadian rhythm in the ovary have not been identified. To identify the molecular mechanisms that are regulated by circadian rhythm and to speculate on the physiologies that are likely to be damaged by perturbation of circadian rhythm in the ovary, we searched for genes in which the amount of transcripts oscillates every 24 h in the mouse ovary. To achieve this, expression profiles of circadian genes (per1, per2, and bmal1) that code transcription-regulation factors for which transcription activities are known to oscillate every 24 h in almost all organs, and wee1, the transcription activity of which circadian genes regulate and which is known to elongate the G2 phase in the cell cycle, were analyzed in this study. Six-week-old female ICR mice were kept individually under a lighting schedule with lights on for 14 h followed by lights off for 10 h. A vaginal smear of each mouse was collected every day to determine its estrous cycle. Ovaries of 3 mice were collected continuously every 4 h over a 4-day period from the start of the light period on the day of proestrus. Total RNA was extracted from each ovary, and 500 ng each was used for cDNA synthesis. Transcripts of each gene and of tbp were quantified by real-time PCR, and the amount of the transcripts of each gene in each sample was divided by the amount of tbp transcripts. The obtained relative values in each sample were used as the representative data of the amount of transcripts of each gene. The amounts of per1, per2, and bmal1 clearly oscillated every 24 h. The maximum and minimum values of per1 and per2 were observed at 16 and 4 h, respectively, after onset of the light period each day. The maximum and minimum values of bmal1 were observed at the time of onset of the light period and at 12 h after onset of the light period each day. Averages of the maximum values of per1, per2, and bmal1 each day were significantly greater than averages of the minimum values (per1, 3.60 � 0.10 and 1.38 � 0.09; per2, 0.82 � 0.08 and 0.27 � 0.06; bmal1, 0.61 � 0.05 and 0.17 � 0.01; P < 0.05). The cyclicity in the oscillation of the amount of wee1 transcripts was weaker than that observed in circadian genes, but the average of values that were obtained from 12 to 20 h after onset of the light period each day was significantly greater than that obtained from 0 to 8 h (0.29 � 0.02 and 0.22 � 0.01; P < 0.05). Our results suggested that the cell cycle of ovarian cells is regulated in a circadian manner through wee1 transcription, which is regulated by circadian genes of which the amounts of transcripts oscillate every 24 h. Because an abnormal cell cycle seems to trigger the development of tumors or follicular cysts, perturbation of circadian rhythm may cause those ovarian diseases.

38 EFFECTS OF TRICHOSTATIN A ON DNA METHYLATION IN CLONED BOVINE EMBRYOS

Recently, the efficiency of full-term development of somatic cloned mouse embryos was significantly increased by treatment with trichostatin A (TSA), an inhibitor of histone deacetylase (Kishigami et al. 2006 Biochem. Biophys. Res. Commun. 340, 183–189; Rybouchkin et al. 2006 Biol. Reprod. 74, 1083–1089). We have shown that TSA treatment improved the rate of development of the cloned bovine embryos to the blastocyst stage (Iwamoto et al. 2007 Reprod. Fertil. Dev. 19, 142 abst). Higher levels of DNA methylation have been shown in early cloned bovine embryos than in in vitro-fertilized (IVF) embryos (Dean et al. 2001 Proc. Nat. Acad. Sci. USA 98, 13734–13738; Santos et al. Curr. Biol. 13, 1116–1121). In this study, we examined the effects of TSA on DNA methylation levels in cloned bovine embryos by immunostaining with an antibody to 5-methyl cytosine (5-MeC). Bovine fibroblasts were cultured under serum starvation (0.4% FCS) for 7 days before they were used as donor cells. The cells were electrofused with bovine enucleated matured oocytes, and activated with a calcium ionophore and cycloheximide. Atotal of 131 cloned embryos were produced. The NT embryos were exposed to 0 (control) and 50 nmTSA from the start of activation to 48 h post-activation (hpa). They were then cultured in an mSOF medium. At 60 hpa, only embryos developed to the 8-cell stage were used for assessment of DNA methylation levels. Sixteen TSA-treated, 22 non-treated, and 19 IVF embryos were immunostained with 5-MeC antibody. For quantitative analysis of the DNA methylation levels, 5-MeC signals in the fluorescent images were determined using an image analyzer system (Aqua Cosmos; Hamamatsu Photonics, Shizuoka, Japan). The data were analyzed with Tukey-Kramer post hoc test for multiple comparisons following ANOVA. Relative levels of DNA methylation of TSA-treated cloned and IVF embryos did not differ (P > 0.05), but were lower than those of non-treated cloned embryos (P < 0.05). The results indicate that TSA treatment of cloned bovine embryos leads to a reduction of DNA methylation levels of their genome. The data suggest that the TSA treatment decreased the DNA methylation levels of cloned bovine embryos to the levels of IVF embryos, resulting in improved blastocyst development of the cloned embryos.

46 EXPRESSION OF TRANSCRIPTION FACTORS SPECIFIC TO THE TROPHOBLAST LINEAGE IN MOUSE SOMATIC NUCLEAR TRANSFER EMBRYOS

Somatic cell nuclear transfer (SCNT) embryos can develop at relatively high rates during the preimplantation period; however, most of these fail after implantation. Development of extraembryonic tissue is indispensable for normal embryonic development. Hence, an abnormality of trophoblast development might be a significant factor in post-implantation lethality of SCNT embryos. A transcription factor, caudal-related homeobox 2 (Cdx2), appears to be involved in the segregation of ICM and trophectoderm (TE) in preimplantation embryos (Niwa et al. 2005 Cell 123, 917–929). Both Cdx2 and Oct3/4 are expressed in all cells at the morula stage, and then Cdx2 expression becomes restricted to the TE and Oct3/4 to the ICM as the blastocyst develops. Mouse embryos deficient in Cdx2 are able to develop to normal blastocysts but die soon after implantation, probably because of defects in the TE lineage. Moreover, dysplasia of the spongiotrophoblast layer might attribute to an abnormality of Tpbpa expression in mouse SCNT embryos (Wakisaka-Saito et al. 2006 Biochem. Biophys. Res. Commun. 349, 106–114). In this study, we examined the expression profiles of transcription factors implicated in trophoblast development in mouse SCNT embryos and intracytoplasmic sperm injection (ICSI) embryos by immunohistochemistry and real-time PCR analysis. SCNT embryos were produced according to the method reported previously (Wakayama et al. 1998 Nature 394, 369–374). In brief, B6D2F1 and B6C3F1 female mice were used for the collection of recipient oocytes and donor cells, respectively. After nuclear transfer, the oocytes were activated and cultured in KSOM to the morula and blastocyst stages. Immunohistochemical analysis demonstrated that in ICSI embryos Cdx2 was only partially expressed at the 8-cell stage but completely in early morulae. In contrast, in SCNT embryos, it was absent at the 8-cell stage and appeared partially at the early morula stage. Thereafter, Cdx2 expression became restricted to the TE cells in both the ICSI and the SCNT blastocysts. However, ectopic expression of Oct3/4 was observed in the TE cells of SCNT, but not in ICSI blastocysts. Real-time PCR analysis showed that at the 8-cell stage, Cdx2 was expressed in ICSI but not in SCNT embryos. In addition, the expression level of Cdx2 in SCNT embryos at the blastocyst stage was only half that in ICSI embryos (P < 0.05). However, there was no significant difference in expression level of Oct3/4 between ICSI and SCNT embryos. Eomesodermin (Eomes) is also implicated in trophoblast development and its expression depends on Cdx2, BMP4, and FGF4. In SCNT embryos, the expression level of Eomes was also only half that in ICSI embryos. These results indicate that the delayed expression of Cdx2 in SCNT embryos may lead to the ectopic expression of Oct3/4 in blastocysts and, along with the limited expression of Cdx2 and Eomes, may contribute to disorders in the function of the trophoblast lineage for normal placental development. This work was supported by a Grant-in-Aid for the 21st Century Center of Excellence Program of the MEXT, Japan, and by a Grant-in-Aid for Scientific Research from the Japan Society for the Promotion of Science.

Effects of volume, culture media and type of culture dish on in vitro development of hamster 1-cell embryos

We examined effects of medium volume and two different culture media (HECM-3 and HECM-4) on in vitro development of hamster embryos. Groups of 5 to 8 1-cell embryos were cultured for 72 h in either < or =100 or > or =100 microl volumes. In the first experiment, embryos were cultured in Petri dishes with 2, 5, 20, 50 or 100 microl of medium using the two media (2 x 5 factorial experiment). Optimal volumes for morula and blastocyst development were 100 microl of HECM-3 and > or =50 microl of HECM-4; in HECM-4, > or =20 microl volumes were suitable whereas in HECM-3 < or = 50 microl volumes were unsuitable. In the second experiment, embryos were cultured in 100, 200, 500 and 1000 microl of HECM-3 and HECM-4 using organ culture dishes. Controls were 100 microl drops in Petri dishes. In organ culture dishes, blastocyst development was < or =6% in HECM-3 and 33-41% in HECM-4, and suitable volumes for development to at least morulae were > or =200 microl of HECM-3, and > or =100 microl of HECM-4. In both experiments development to morula and blastocyst stages with 100 microl volume in Petri dishes was significantly higher with HECM-4 (96 and 85% in Experiment 1 and 2, respectively) than that with HECM-3 (52 and 40% in Experiment 1 and 2, respectively; P < 0.05). These results indicate that attention should be paid to both type and volume of medium and interaction with type of culture dish for optimizing development of embryos in vitro.

284 IDENTIFICATION AND CHARACTERIZATION OF THE 5′-FLANKING REGION OF THREE MOUSE MATERNAL GENES (HISTONE H1OO, NUCLEOPLASMIN 2, AND ZYGOTE ARREST 1): TRANSCRIPTIONAL ACTIVITY IN MOUSE OOCYTES

Maternal transcripts that accumulate during oocyte growth are involved in the meiotic maturation, the initiation of the first mitosis, and the later pre-implantation development. Although the conserved E-box sequences in promoter region of some maternal genes (for example, Zp3 and Gdf-9), which are important in regulating gene transcripts as binding sites of transcriptional factors, may play a role in the oocyte-specific expression in ovary, the molecular mechanism that regulates the expression of the maternal genes is still not known. In this study, we have focused on the transcriptional activity of promoter regions to clarify the molecular mechanism of transcriptional regulation of these maternal genes [Histone H1oo (H1oo), Nucleoplasmin 2 (Npm2), and Zygote arrest 1 (Zar1)]. First, we observed the expression of firefly luciferase expression vectors under promoter regions of 3 maternal genes in oocytes isolated from 10- to 12-day-old mice, which is mainly NSN-type, transcriptionally active form (Bouniol-Baly et al. 1999 Biol. Reprod. 60, 580–587). Transcriptional activities of H1oo (-3975), Npm2 (-2610), and Zar1 (-5187) promoters were detected in oocytes, the relative luciferase activities being an average of 70, 130, and 12, respectively. On the other hand, these promoter activities were not detected in embryos at the 2-cell stage. Furthermore, deletion analysis of promoter elements (E-boxes) of H1oo and Npm2 was done by microinjecting deletion constructs into oocytes. In the H1oo promoter, deletion of sequences between -3975 and -72 bp from the transcription start site resulted in one-third of the level obtained in the full H1oo (-3975) promoter. In addition, deletion of sequence -68 bp resulted in no detection of luciferase activity. These findings indicate that the putative distal promoter sequences exist at the 52-flanking region (-3975 to -759) of the H1oo gene and that the region (-314 to -68) including the E-box region (-72) may be required for high-level transcriptional activity of the H1oo promoter. In the Npm2 promoter, deletion of sequences between -2610 and -180 bp from the transcription start site resulted in one-third of the level of wild-type activity of the Npm2 (-2610) promoter. In addition, deletion of sequence -101 bp resulted in no detection of luciferase activity. These findings also indicate that 3 putative distal promoter sequences exist at the 52-flanking region (-2610 to -210) of the Npm2 gene and that the region (-210 to -101) that includes the E-box region (-180) is crucial for high-level transcriptional activity of the Npm2 promoter. In conclusion, the E-box may be a key regulatory region for the expression of two of the maternal genes (H1oo and Npm2) examined. Currently, we are attempting to identify the transcriptional factor binding sites by DNase I footprint analysis and gel mobility shift assay.

48 EFFECTS OF TRICHOSTATIN A ON DEVELOPMENT OF BOVINE SOMATIC CELL NUCLEAR TRANSFER EMBRYOS

Although cloning by somatic cell nuclear transfer (SCNT) has been achieved in various mammalian species, its efficiency has been very low (Han et al. 2003 Theriogenology 59, 33–44). Successful cloning requires conversion from differentiated donor nuclei to embryonic nuclei after transfer of the somatic nuclei into enucleated oocytes. Reprogramming of the transferred somatic nuclei must be completed by the time when normal activation of the embryonic genome occurs (Solter 2000 Nat. Rev. Genet. 1, 199–207). Recently, both full-term development and pre-implantation development of mouse SCNT embryos were significantly enhanced by treatment with trichostatin A (TSA), an inhibitor of histone deacetylase (Kishigami et al. 2006 Biochem. Biophys. Res. Commun. 340, 183–189; Rybouchkin et al. 2006 Biol. Reprod. 74, 1083–1089). The objective of this study was to investigate the effects of TSA on the development of bovine SCNT embryos. Bovine fibroblasts were cultured under serum starvation (0.4% FCS) for 7 days and then used as donor cells. The cells were electro-fused with bovine enucleated matured oocytes, and activated with a calcium ionophore and cycloheximide. They were subsequently cultured in mSOF medium until 168 h post-activation (hpa). The NT embryos were exposed to 0 (control), 5, 50, and 500 nM TSA from the start of activation to 48 hpa. Experiments were repeated 3 times, and the data were analyzed with Fisher's PLSD test following ANOVA. The cleavage rates were the same among the groups (60 to 80&percnt;; P &gt;0.05). However, the blastocyst rate of NT embryos treated with 50 nM TSA was higher than that of control embryos (40&percnt; vs. 19&percnt;, respectively; P &lt; 0.05). On the other hand, the blastocyst rate was lower with 500 nM TSA than with 5 or 50 nM TSA (7&percnt; vs. 33&percnt; or 40&percnt;; P &lt; 0.05). These data suggest that proper TSA treatment after somatic cloning improves the rate of development of bovine cloned embryos to the blastocyst stage. Further research is needed to examine whether NT embryos derived from different cell lines or types have similar susceptibility to TSA.

324 EFFECT OF AGING ON AMOUNTS OF DNA METHYLTRANSFERASE mRNA IN MOUSE SPERMATOZOA

The spermatozoon is a specially differentiated cell designed to carry a haploid male genome into an oocyte at fertilization. It recently was reported that a matured spermatozoon contains several kinds of mRNAs and these are delivered into the oocyte at fertilization (Ostermeier et al. 2004 Nature 429, 154). The physiological role of paternally derived mRNAs is not clear; however, there is a report that the DNA methyltransferase (Dnmt) mRNA level in spermatozoa from male rats exposed to ethanol was significantly reduced (Bielawski et al. 2002 Alcohl. Clin. Res. 26, 347–351). The reduction of mRNA levels of Dnmt in spermatozoa would lead to altered epigenetic modification of the genome. Because factors such as age may affect spermatozoa mRNA levels, this study evaluated the effect of individual aging on the expression levels of Dnmts during spermatogenesis. This was accomplished by determining expression levels of Dnmts in the whole testis and in spermatozoa from young and aged mice by quantitative reverse-transcription-PCR. Seven- (young) and 68- (aged) week-old C57BL/6N male mice (n = 3/group) were sacrificed by cervical dislocation and whole testes and matured spermatozoa were collected. Total RNA was extracted and purified from each sample. In this study, 5 Dnmts (Dnmt1s, Dnmt1p, Dnmt3a, Dnmt3b, and Dnmt3l) and glyceraldehyde-3-phosphate dehydrogenase (GAPDH), a reference gene, were examined for expression levels in whole testis and spermatozoa using SYBR Premix Ex Taq (Takara Bio, Inc., Otsu, Shiga, Japan) and the 7300 Real-Time PCR System (Applied Biosystems, Foster City, CA, USA). Real-Time PCR runs for each Dnmt and GAPDH were repeated 3 times using different RNA batches from different individuals. The GAPDH expression level was used to normalize the expression levels of each Dnmt. Data were analyzed by Student&apos;s t-test. Relative expression levels of each Dnmt in testis from aged males compared to that of young males were 0.94, 1.15, 0.91, 1.15, and 1.14 (Dnmt1s, Dnmt1p, Dnmt3a, Dnmt3b, and Dnmt3l, respectively). There was no difference in the expression levels of the 5 Dnmts examined between testes from aged and young males. On the other hand, the relative amounts of each Dnmt mRNA in spermatozoa from aged males compared to that of young males were 0.87, 0.01, 0.54, 1.07, and 1.75 (Dnmt1s, Dnmt1p, Dnmt3a, Dnmt3b, and Dnmt3l, respectively). There was a significant reduction (P &lt; 0.05) in the amount of Dnmt1p mRNA. The reason why the amount of Dnmt1p mRNA in spermatozoa from aged male mice showed such reduction is not clear. There was no difference in the relative expression levels of Dnmt1p in testis irrespective of male age. Dnmt1p is only translated in the spermatocyte during the pachytene stage in meiosis and its physiological role is not clear. To elucidate this male, age-related reduction of the amount of Dnmt1p mRNA in spermatozoa would clarify part of physiological role of Dnmt1p. This work was supported by Wakayama Prefecture Collaboration of Regional Entities for the Advanced of Technological Excellence, Japan, and by a Grant-in-Aid for the 21st Century Center of Excellence Program of the MEXT, Japan.

229 DERIVATION OF PRESUMPTIVE GONOCYTES IN VITRO FROM PRIMATE EMBRYONIC STEM CELLS

Embryonic stem cells (ESCs) of nonhuman primates are important for research into human gametogenesis, because of similarities between the embryos and fetuses of nonhuman primates and those of humans. Recently, the formation of germ cells from mouse ESCs in vitro has been reported. In this study, we established cynomolgus monkey ES (cyES) cell lines and attempted to induce their differentiation into germ cells in order to obtain further information on the development of primate germ cells by observing the transcripts of some markers reported as specific for germ cells. CyES cell lines were established using blastocysts produced by intracytoplasmic sperm injection (ICSI). For inducing superovulation, females were treated with 25 IU kg-1 pregnant mare serum gonadotropin once a day for 9 days, followed by 400 IU kg-1 hCG. Oocytes were collected at 40 h after injection of hCG. After sperm injection, embryos were cultured in mCMRL medium to the blastocyst stage. For cyES cell establishment, inner cell masses (ICMs) were isolated by immunosurgery. The ESC colonies developed at about 10 days after ICM plating, and 3 cell lines were successfully established (3/11; 27.3%). All cell lines expressed Oct3/4, SSEA-4, and ALP activity. These ESCs formed teratomas containing 3 different embryonic layers when injected into SCID mice. And the cells could be passaged over 50 times without losing their original properties. To observe in vitro gametogenesis, we attempted to induce differentiation by non-adherent conditions. When cyES cells differentiated spontaneously, the aggregated structures (i.e. embryoid bodies; EBs) accumulated vasa, the expression of which is restricted to germ cells, and some meiotic markers such as dmc1 and sycp1 that exist only in synaptonemal complexes in meiosis. The existence of these markers was also confirmed by immunocytochemistry on cryosections. Interestingly, these products expressed oct4 and nanog again at Day 16, though the expression of both genes diminished at once with onset of differentiation. In vivo, it is reported that vasa, oct4, and nanog are expressed in migrating PGCs, posibly throughout the development of germ cells into spermatocytes/oocytes. Given the results obtained with the meiotic markers, it is possible that developing germ cells such as PGCs or gonocytes could be formed in cynomolgus EBs as in previous cases with mouse or human EBs. These results demonstrate that cyES cells might contribute to putative germ cells in vitro by differentiating into EBs and could be used as a model for studying mechanisms of germ cell development. This study was supported by a Grant-in-Aid for the 21st Century COE Program of the Japan Mext and by a grant for the Wakayama Prefecture Collaboration of Regional Entities for the Advancement of Technology Excellence of the JST.

44 NUCLEAR TRANSFER IN CATTLE USING SOMATIC CELLS FROM FROZEN TESTICLES WITHOUT CRYOPROTECTANTS

Obtaining somatic cells from preserved organs or tissues is useful for the conservation and regeneration of genetic resources by nuclear transfer (NT). Bovine cells for NT have been obtained from cooled carcasses stored at 0�C for several days (Arat et al. 2005 Reprod. Fert. Dev. 17, 164 abst) and from fetal skin tissue cryopreserved with DMSO (Fahrudin et al. 2001 J. Vet. Med. Sci. 63, 1151–1154). However, frozen storage of organs or tissues without cryoprotectants was considered to be quite inappropriate for obtaining viable cells. We report here that viable donor cells for NT were obtained from bovine testicles after frozen storage without cryoprotectants. In the first experiment, we investigated whether viable cells can be recovered from frozen testicles castrated from Japanese Black bulls. The testicles were frozen at -80�C in a freezer for several days; then some were stored in liquid nitrogen for 10 months without cryoprotectants. Before thawing, the testicles were divided into 3 pieces, caput epididymis, cauda epididymis, and testis. Each piece was then put in saline at 42�C for quick thawing. Thawed tissues were minced into 5-mm pieces and incubated at 39�C for 2 h in DMEM containing 0.1% collagenase and 0.2% dispase. After filtration through a 250-�m nylon mesh filter, the filtrates were centrifuged at 250 � 4g for 5 min. Then precipitates were resuspended with MF-start� primary culture medium (TM Cell Research Inc., Fukui, Japan) and incubated at 38.5�C under the atmosphere of 5% CO2 in air with high humidity. After 5 days of incubation, the medium was replaced and nonadherent debris was discarded. Viable cells were obtained from the caput epididymis. These cells actively proliferated and expanded. In the next experiment, to determine whether these cells can be used for NT, the cells were electrically fused with enucleated bovine oocytes. Bovine fibroblasts taken from unfrozen ear tissue were used as controls. The NT embryos were activated by Ca-ionophore treatment, followed by treatment with cycloheximide for 6 h, and then cultured in mSOF for 168 h. NT embryos reconstructed from testicle cells did not significantly differ from NT embryos made with control cells with regard to blastocyst rates (22.1% and 20.2%), cell number of blastocysts [130 � 43 and 121 � 43 (mean � SD)], and ICM ratio (21.1% and 22.6%), respectively (ANOVA). These results suggest that somatic cells derived from bovine frozen testicles can be used for nuclear transfer. Further studies are needed to examine whether viable cells can be obtained from other frozen organs or tissues. This study was partially supported by a grant from the Wakayama Prefecture Collaboration of Regional Entities for the Advancement of Technological Excellence, JST.

50 DNA METHYLATION PROFILES OF UPSTREAM ELEMENTS OF Oct-3/4 GENE IN IN VITRO FERTILIZATION (IVF) AND SOMATIC CELL NUCLEAR-TRANSFERRED (SCNT) EMBRYOS

Cloning by adult somatic cell nuclear transfer (SCNT) has proven to be successful for the production of clones from many species (Keith 2004 Cytogenet. Genome Res. 105, 285). However, somatic cloning is currently highly inefficient. One of the reasons for this is that SCNT is believed to be associated with epigenetic errors including abnormal DNA methylation of the reconstructed embryo. The Oct-3/4 gene, a member of the POU transcription factor family, is expressed throughout the pre-implantation embryo. Abnormal expression of the Oct-3/4 gene in the nuclear-transferred embryo is either directly or indirectly caused by nuclear transfer and is suggested to be indicative of a general failure to reset the genetic program (Boiani et al. 2002 Genes Dev. 16, 1209). In this study, we investigated the DNA methylation profiles of the Oct-3/4 gene in the genome of SCNT embryos, using bisulfite sequencing analysis. Then, we observed the detailed subcellular localization of Oct-3/4 proteins in SCNT embryos using immunocytochemical (ICC) analysis. Nuclear transfer of cumulus cell nuclei was carried out as previously described (Wakayama et al. 1998 Nature 394, 369). After nuclear transfer, embryos were subsequently cultured in KSOM media to the morula and blastocyst stages. We compared the methylation profiles of 3 transcriptional control elements (distal enhancer, DE; proximal enhancer, PE; and promoter) of the upstream region of the Oct-3/4 gene with the genome of in vitro fertilization (IVF) and SCNT embryos. The methylation rate of CpG sites in the DE and promoter regions of both IVF and SCNT embryos was low at both the morula and the blastocyst stages. What&apos;s interesting is that there was a significant difference in the methylation level on CpG sites in the PE element between IVF and SCNT embryos. At the morula stage, the methylation level on CpG sites in the PE element was very low in the IVF embryo and moderately high in the SCNT embryo (0.9&percnt; and 26.3&percnt;). Conversely, at the blastocyst stage, CpG sites in the PE element showed high methylation in the IVF embryo and low methylation in the SCNT embryo (55.2&percnt; and 10.5&percnt;). CpG sites in the PE element were lightly methylated (3.0&percnt;) in the inner cell mass (ICM) of the IVF embryo. This means that the main portion of methylation in the IVF blastocyst embryo occurred at the trophectoderm (TE). On the other hand, in ICM of the SCNT embryo, the methylation level of each embryonic cell was almost the same in the whole blastocyst embryo (9.8&percnt; and 10.5&percnt;). As a result, it is highly possible that the CpG sites in the PE element of ICM were methylated as in the TE. ICC analysis revealed that some SCNT embryos showed aberrant Oct-3/4 expression in the TE. These results indicate that the methylation of CpG sites in the Oct-3/4 PE element may be related to expression of Oct-3/4 in the mouse IVF and SCNT embryos. These differences in methylation level between IVF and SCNT embryos were reflected as abnormal expressions of Oct-3/4 on SCNT embryos. This study was supported by the 21st COE Program of MEST. M.K. is a JSPS Research Fellow and supported by Grant-in Aid for Scientific Research (No. 1751132) of JSPS.

298 EFFECTS OF TRICHOSTATIN A DURING IN VITRO FERTILIZATION OF BOVINE OOCYTES ON SUBSEQUENT DEVELOPMENT, CELL NUMBER, AND ALLOCATION OF RESULTING EMBRYOS

Histone acetylation is one of the major mechanisms of epigenetic reprogramming of gamete genomes after fertilization or of transferred cell genomes after nuclear transfer to establish a totipotent state for normal development. In the fertilization of bovine oocytes, asynchronous histone acetylation occurs during pronuclear formation in the manner that modification of the paternal genome precedes that of the maternal genome (Wee et al. 2006 J. Biol. Chem. 281, 6048–6057). In the present study, the effects of trichostatin A (TSA), an inhibitor of histone deacetylase, during in vitro fertilization (IVF) of bovine oocytes on subsequent embryonic development were investigated. Cumulus-enclosed oocytes obtained from slaughterhouse ovaries were in vitro-matured (IVM) for 21 h in TCM-199 supplemented with 5% v/v FCS, 0.5 mM sodium pyruvate, 0.02 AU mL-1 FSH, and 1 �g mL-1 estradiol-17β at 39�C under 5% CO2 in air. After IVM, the oocytes were subjected to IVF with 3 � 106 mL-1 of Percoll gradient-selected sperm in a defined medium (Brackett and Oliphant 1975 Biol. Reprod. 12, 260–274) supplemented with 0 (control), 5, 50, and 500 nM TSA for 18 h. After IVF, presumptive zygotes were freed from cumulus cells and cultured in mSOF medium until 168 h post-insemination (hpi) at 39�C under 5% CO2, 5% O2, and 90% N2 with high humidity. Cleavage and blastocyst development were assessed at 48 and 168 hpi, respectively. Inner cell mass (ICM) and trophectoderm (TE) of blastocysts were differentially stained by the method of Thouas et al. (2001 Reprod. Biomed. Online 3, 25–29) to assess cell number and ICM/TE ratio. Experiments were replicated 3 times. Data are presented as means � SEM and statistically analyzed by multiple comparison with the Holm method. Rates of cleavage (0 nM: 71.0 � 7%, n = 102; 5 nM: 75.5 � 5%, n = 106; 50 nM: 68.8 � 6%, n = 105; and 500 nM: 71.7 � 4%, n = 98) and blastocyst formation (21.4 � 5%, 22.3 � 6%, 17.8 � 2%, and 18.2 � 2%, respectively) were similar among the groups. However, 500 nM TSA significantly (P &lt; 0.05) increased ICM and total cell numbers (59.8 � 4 and 143.5 � 7, respectively, n = 31) compared with the control (43.1 � 3 and 120.9 � 7, n = 31). In addition, ICM/TE ratios were higher in the 50 nM (0.81 � 0.08, n = 29) and 500 nM (0.92 � 0.2, n = 31) groups than in the control (0.59 � 0.04, P &lt; 0.05). These results suggest that TSA treatment during IVF of bovine oocytes does not affect the blastocyst rate but alters the cell numbers and their allocation to ICM and TE. Overriding epigenetic modification of the genome during fertilization may have a carryover effect on cell proliferation and differentiation in pre-implantation embryos. This study was supported by a grant from Wakayama Prefecture Collaboration of Regional Entities for the Advancement of Technological Excellence, JST.

281 EXPRESSION PROFILE AND KNOCKDOWN ANALYSIS OF A FUNCTIONALLY UNKNOWN DD2-2 GENE IN MOUSE PRE-IMPLANTATION EMBRYOS

Zygotic gene activation (ZGA) starts at the G2 phase at the 1-cell stage in the mouse. However, the molecular mechanism of ZGA has not been completely elucidated. We have investigated the molecular functions of many gene clusters, DD clones obtained by differential display assays for ovulated eggs at the M II stage and 1-cell stage embryos at the G2 phase. As a result, we have identified a functionally unknown gene, whose sequence did not match a known transcript in the gene bank DD2-2 gene. Here, we report the expression profile and knockdown analysis of the DD2-2 gene in mouse pre-implantation embryos. Nucleotide sequence analysis of the DD2-2 cDNA revealed that the open reading frame of 1056 bp encodes a protein of 351 amino acids with a predicted molecular mass of 41.5 kDa. The deduced amino acid sequence indicated that DD2-2 protein might be a soluble protein without a signal peptide. We first investigated the expression profiles of DD2-2 in pre-implantation embryos by quantitative real-time PCR using an ABI PRISM 7300 Sequence Detection System (Applied Biosystems, Foster City, CA, USA). To investigate the effect of knockdown of the DD2-2 gene on the development of pre-implantation embryos, we injected pβ-act/antisenseDD2-2/IRES/EGFP into male pronuclei of embryos at 7 to 9 h after insemination (hpi) and observed the development of embryos that showed EGFP expression at 24 hpi. Real-time PCR analysis of pre-implantation embryos showed that maternal DD2-2 mRNA at a low level significantly increased up to the early 2-cell stage, and significantly decreased by the 4-cell stage and later, suggesting that DD2-2 gene specifically expresses at major ZGA. In the knockdown analysis, EGFP-positive embryos with pβ-act/antisenseDD2-2/IRES/EGFP showed a lower rate of development to the 4-cell stage and later, compared with that of EGFP-positive embryos with pβ-act/luc+/IRES/EGFP [72% (94/130) vs. 54% (71/131); P &lt; 0.05], indicating that the knockdown of DD2-2 by antisense RNA resulted in a inhibition of pre-implantation development. In conclusion, the DD2-2 gene, a functionally unknown gene, may play an important role in pre-implantation development. This study was supported by a Grant-in-Aid for the 21st Century COE Program of the Japan Mext and by a grant for the Wakayama Prefecture Collaboration of Regional Entities for the Advancement of Technology Excellence of the JST.

227 DIFFERENTIATION OF HEPATOCYTE-LIKE CELLS FROM MOUSE EMBRYONIC STEM CELLS IN A MONOLAYER CULTURE SYSTEM

Embryonic stem (ES) cells can be differentiated in vitro into a variety of cell lineages. In vitro differentiation of ES cells, therefore, provides a model system for organogenesis as well as an innovative approach for regenerative medicine. However, current in vitro hepatic differentiation systems from ES cells require embryoid body formation, and such systems achieve quite low differentiation efficacy. In this study, in order to examine a system for preparation of significant numbers of hepatocytes from ES cells, mouse ES cells were directly differentiated into hepatocyte-like cells using monolayer culture conditions, and their hepatic characteristics were examined. ES cells were cultured on gelatin-coated plates for 3 days in ES cell medium containing leukemia inhibitory factor (100 units mL-1) and retinoic acid (10-8 M). After 3 days, the cells were harvested by trypsinization and cultured on gelatin-coated plates for 5 days in ES cell medium containing acidic fibroblast growth factor (aFGF; 100 ng mL-1), FGF4 (20 ng mL-1), and human growth factor (50 ng mL-1). After 5 days, the cells were passaged on type I collagen-coated plates for 2 days in ES cell medium containing Oncostatin M (10 ng mL-1), followed by additional culture in modified William&apos;s E medium containing 5% FBS and dexamethasone (10-7 M) for 7 days. For characterizing the differentiated ES cells, the cells were examined for albumin production by immunocytochemistry, and the expression profiles of hepatic genes by RT-PCR. For analyzing their hepatic function, the cells were examined for glycogen production by PAS staining, and their metabolic activities of ammonia elimination and urea synthesis were measured. After 10 days of culture, almost all cells showed albumin production by immunofluoresence. RT-PCR analysis showed expression of various genes specific for hepatocytes, such as ALB, AFP, TTR, and HNF3�, but not HNF4α, and genes specific for matured hepatocytes, such as PEPCK, TAT, G6P, and Cyp7a1. After 11 days of culture, glycogen production was partially detected in the hepatocyte-like cells by PAS staining. Although ammonia elimination activity of hepatocyte-like cells was shown to be significantly lower than that of the primary culture of fetal hepatocytes after 24 and 48 h of treatment (P &lt; 0.05), its metabolic activity was acquired at the level of about 40% of that of fetal hepatocytes. Hepatocyte-like cells also showed lower activity of urea synthesis than fetal hepatocytes at 6, 12, and 24 h of treatment (P &lt; 0.05), but this activity increased thereafter and reached about 70% of that of fetal hepatocytes after 48 h of treatment. These results demonstrated that monolayer culture conditions for ES cell differentiation could provide an in vitro system to produce significant numbers of hepatocytes. This work was supported by a Grant-in-Aid for the 21st Century Center of Excellence Program of the MEXT, Japan, and by a Grant-in-Aid for Scientific Research from the Japan Society for the Promotion of Science.

69 RELATION OF SPATIAL GENE EXPRESSION PATTERNS IN BOVINE EMBRYOS RECONSTRUCTED WITH SOMATIC CELLS TO BLASTOCYST DEVELOPMENT

Recently, enhanced development to full term was obtained with embryos reconstructed with bovine early G1 cells rather than with G0 cells (Kasinathan et al. 2001 Nat. Biotechnol. 19, 1176-1178; Urakawa et al. 2004 Theriogenology 62, 714-728). However, the reason why donor somatic cells at the early G1 phase are better for embryo reconstruction is unclear. In this study, we investigated the relation of spatial gene expression patterns at the 4- to 8-cell stage to blastocyst development of embryos reconstructed with early G1 cells. Bovine fibroblasts stably transfected with �-act/luc+/IRES/EGFP were used for embryo reconstruction. M phase cells were prepared as described by Urakawa et al. (2004). Early G1 cells were obtained from cultured M phase cells soon after the M phase cells divided. Quiescent cells (cultured in 0.4% serum for 7 days) were used as G0 cells for a control. The cells were electrofused with enucleated bovine oocytes matured in vitro, and activated with a calcium ionophore and cycloheximide. The reconstructed embryos were cultured until 60 hours post fusion (hpf), and zonae pellucidae of 4- to 8-cell embryos were removed by pronase. To determine gene expression, the LUC+ activity (luminescence) in the embryo blastomeres was detected with an imaging photon counter (Hamamatsu Photonics, Hamamatsu City, Shikuoka Prefecture, Japan) for 10 min. The embryos were categorized as being positive, mosaic, or negative depending on whether all, some or no blastomeres were luminescent, respectively. The embryos were cultured in mSOF medium individually until 168 hpf to assess development to the blastocyst stage. Blastocyst development of reconstructed embryos without detection of luminescence was also examined. Experiments were repeated three times, and the data were analyzed with Fisher's PLSD test following ANOVA. At 60 hpf, 75% (74/99) of embryos reconstructed with early G1 cells and 55% (46/83) of embryos with G0 cells developed to 4- to 8-cell stage embryos. The difference is significant (P < 0.05). The percentages of positive, mosaic, and negative embryos with G1 cells were 49, 35 and 16%, and blastocyst rates were 30, 11, and 0%, respectively. With G0 cells, the percentages were 32, 56, and 12%, and the blastocyst rates were 15, 4, and 0%, respectively. More positive embryos were obtained with early G1 cells than with G0 cells (P < 0.05). Blastocyst rates of the positive embryos with early G1 cells were the same as with G0 cells (P > 0.05). Blastocyst development of positive embryos was higher than that of mosaic and negative embryos in early G1 and G0 groups (P < 0.05). Without detection of luminescence, the blastocyst rates from the reconstructed embryos were 43% (35/81) and 16% (20/125) with early G1 and G0 cells, respectively (P < 0.05). These results suggest that the higher developmental capacity of embryos reconstructed with early G1 cells might be related to the appropriate spatial gene expression at the 4- to 8-cell stage. A part of this study was supported by a grant from the Wakayama Prefecture Collaboration of Regional Entities for the Advancement of Technological Excellence of the JST.

136 FUNCTIONAL ANALYSIS OF RHOPHILIN-2 GENE IN PRE-IMPLANTATION MOUSE EMBRYO

Zygotic gene activation (ZGA), which starts at the G2 phase at the 1-cell stage (Latham 1999), promotes the reprogramming of gene expression and is critical for the subsequent development of pre-implantation embryos. We have investigated the molecule function of many gene clusters, DD clones obtained by Differential-Display assays for ovulated eggs at the M II stage, and 1-cell embryos at the G2 phase. The differential expression of rhophilin-2 shown in DD assays was also confirmed by 3 independent real-time PCR analyses (P < 0.05). For these reasons, in this study, we focused on the rhophilin-2 gene, which regulates cytoskeletal organization (Peck et al. 2002). At first, we identified a protein that interacts with the Rhophilin-2 protein by a yeast 2-hybrid system. To confirm the interaction between Rhophilin-2 and the putative protein obtained by a yeast two-hybrid system, we used a co-immunoprecipitaion assay. We also investigated the expression profiles of rhophilin-2 and the transcripts of the identified protein in ovary and pre-implantation embryos using real-time PCR and immunofluorescence (IF) analysis. The ICR mice at 48 h after PMSG priming were primed with hCG, and ovaries were collected at 7 h after hCG priming. Pre-implantation embryos were collected at 1-cell, 2-cell, and 4-cell stages, and cDNA was produced by mRNA isolated from 10 oocytes or embryos in each group and was subjected to real-time PCR using a TaqMan Probe system (ABI). Sectioned ovaries and pre-implantation embryos were analyzed by IF analysis using antibody of Rhophilin-2 and the identified protein. This is the first report that GABA receptor-association protein (GABARAP) was identified as a protein that interacts with Rhophilin-2, as a result of using the yeast 2-hybrid system and subsequent co-immunoprecipitation assay. After fertilization, transcript levels of rhophilin-2 significantly decreased from the 1-cell stage to the 2-cell stage (P < 0.05), but transcript levels of GABARAP significantly increased from the 1-cell stage to the 2-cell stage (P < 0.05). The IF analysis revealed localization of Rhophilin-2 and GABARAP at the nucleolus of all follicle stage in the ovary. Moreover, Rhophiln-2 and GABARAP were found to be localized on the microtubules of 1-cell and 2-cell embryos, but no signal of Rhophilin-2 was detected in 4-cell embryos. These results suggest that Rhophilin-2 protein regulates the cytoskeletal organization in 1-cell to 2-cell embryos and is involved in the molecular mechanism of cell division by coupling with GABARAP. This study was supported by a Grant-in-Aid for the 21st Century COE Program of the Japan Mext and by a grant for the Wakayama Prefecture Collaboration of Regional Entities for the Advancement of Technology Excellence of the JST.

138 IN VITRO CULTURE OF CD9-EXPRESSING CELLS ENRICHED BY MAGNETIC CELL SORTING FROM TESTES OF CRYPTORCHID ADULT AND PUP IN MICE

Recently, studies on cell surface markers of spermatogonia in combination with germ cell transplantation techniques have made possible the functional analysis of germline stem cells (GS cells). The GS cells are downstream of the stem cells such as ES cells and embryonic germ cells (EG cells), which are derived from primordial germ cells (PGCs). Therefore, GS cells are expected to be useful in the production of genetically modified animals. In this study, we examined the enrichment and cultivation of mouse GS cells by magnetic cell sorting (MACS). Testicular cell suspensions were collected from C57BL/6J cryptorchid adult testes at 2 to 3 months after surgery and ICR pup (6 to 8 dpp) testes. They were digested by 0.1% collagenase followed by 0.25% trypsin with gentle shaking. Dissociated cell suspensions were filtrated through a glass-wool column followed by a Falcon cell strainer (40-�m mesh). They were then treated with biotin-conjugated anti-mouse CD9 antibody, whose antigen, CD9, is localized on the GS cell surface, followed by the streptavidin-microbeads treatment. The cell suspension was passed through a MACS-separation column. In Experiment I, MACS-treated fractions were analyzed by flow cytometry (FCM) on the rates of recovery and enrichment and their cellular characteristics. In Experiment II, CD9-positive (CD9+) cells were cultured on gelatin-coated MultiDish (176740, Nunc) with 4-5 � 105 cells/well in StemPro34-SFM supplemented with 1% fetal bovine serum, leukemia inhibitory factor, GDNF, bFGF, EGF, insulin, transferrin, putrescine, MEM vitamin solution, MEM-NEAA and some other reagents at 32�C or 37�C under 5% CO2 in air. They were examined for their proliferation and cytological changes such as CD9, �6-integrin and Oct-1 expression by immunohistochemistry. In Experiment I, MACS selection effectively enriched CD9+ cells from mouse testes. However, FCM analysis revealed that the CD9-negative (CD9-) cells partially remained in MACS-selected fraction from cryptorchid adult testes. In contrast, the CD9+ subpopulation could be successfully separated from CD9- subpopulation from pup testes. Therefore CD9+ subpopulation from pup testes was used for the following cultivation. In Experiment II, the cells proliferated in the first few days in suspension. Then they attached to the dish and formed colonies after 5 days or 3 days of culture at 32�C or 37�C, respectively. Immunohistochemical analysis showed that the cells maintained the expression of CD9 for at least 14 days, but their expression of �6-integrin gradually diminished. It was demonstrated by immunohistochemistry and FCM analysis that the cells in colonies expressed Oct-1, and its expression level was stronger in culture at 37�C than at 32�C. These findings indicate that the CD9+ cells collected from mouse pup testes have stem cell properties. This work was supported by the Wakayama Prefecture Collaboration of Regional Entities for the Advanced Technological Excellence, JST; by a Grant-in-Aid for the 21st Century Center of Excellence Program of the MEXT, Japan; and by a Grant-in Aid for Scientific Research from the Japan Society for the Promotion of Science.

306 ANALYSIS OF GLOBAL DNA METHYLATION IN BOVINE SPERMATOZOA

Methylation of cytosine base in the CpG dinucleotide sequence (DNA methylation) is a major epigenetic modification of the genome and plays an important role in gene expression. Recently, global DNA methylation in genome was studied by using a restriction landmark genomic scanning (RLGS) method and/or a representational difference analysis (RDA) method. However, these methods are complicated and need to use restriction enzymes. Therefore, the information derived from those methods is restricted to the region of the DNA sequence which is able to be cleaved by restriction enzymes. In this study, to establish a simple method to estimate global DNA methylation level in bovine spermatozoa, we tried to develop the DNA methylation analyzing method by using immunostaining of 5-methylcytosine. The immunostaining method for 5-methylcytosine in this study was based on the method developed by Benchaib et al. (2003 Fertil. Steril. 80, 947-952) for human spermatozoa. Because of the species difference, we modified some treatments to apply to bovine spermatozoa. Frozen-thawed bovine spermatozoa were washed by using 30 and 45% Percoll gradient solutions. After washing, spermatozoa were treated with 0.25 dithicthreitol M (DTT) and 1% sodium dodecyl sulfate (SDS) at room temperature (RT). Then, treated spermatozoa were spread on a slide glass with Cytospin4 (30g, 5 � 104 cells/mL) and air-dried at RT. Air-dried bovine spermatozoa specimens were fixed in methanol: glacial acetic acid (3:1) solution at RT and treated with 1% Triton X and 1% SDS at RT; DNA was denatured with 6 N HCl at RT. After the denaturation, 5-methylcytosine in sperm DNA was analyzed by immunofluorescence technique with mouse anti 5-methylcytosine antibody and fluorescein isothiocyanate (FITC)-conjugated goat anti mouse IgG antibody. The total sperm DNA was counterstained with propidium iodide (PI). Stained samples were observed with a confocal laser scanning microscope and obtained images were analyzed with fluorescence image analysis software. The area that was clearly stained with PI in each sperm head was designated and measured as the area of total sperm DNA, and the number of the dots that showed FITC fluorescence within the total sperm DNA area was designated and measured as the area of 5-methylcytosine in total sperm DNA. The area measurement was performed with fixed light strength. Three bovine spermatozoa samples derived from different bulls, used daily for calf production by AI, were examined. The ratio of the mean total area of the 5-methylcytosine in sperm DNA to the mean total area of the sperm DNA was 34.1% in bull A (9.13 � 5.66 �m2, 26.75 � 5.29 �m2, n = 57), 45.2% in bull B (16.60 � 3.79 �m2, 36.74 � 5.95 �m2, n = 41) and 43.9% in bull C (14.66 � 4.27 �m2, 33.45 � 7.13 �m2, n = 22). There was significant difference in the ratio between bull A and bulls B and C (P < 0.01). More research is required to evaluate the meaning of this individual difference of DNA methylation between bulls. This work was supported by Wakayama Prefecture Collaboration of Regional Entities for the Advanced of Technological Excellence, Japan, and by a Grant-in-Aid for the 21st Century Center of Excellence Program of the MEXT, Japan.

39 DEVELOPMENTAL COMPETENCE OF RECONSTRUCTED EMBRYOS PRODUCED BY NUCLEAR TRANSFER BETWEEN CYNOMOLGUS MONKEY FIBROBLAST CELLS AND RABBIT OOCYTES

Interspecies nuclear transfer has been used as an invaluable tool for studying nucleus-cytoplasm interactions and it may also be used for rescuing endangered species whose oocytes are difficult to obtain. In this study, we investigated interaction of the cynomolgus monkey cell as a nuclear donor with the rabbit oocyte as a host cytoplasm. Whole cynomolgus fibroblast cells were injected into the rabbit enucleated oocytes (cynomolgus-rabbit cloned embryos) and cultured in TCM-199 and RPMI 1640 culture media. Rabbit-rabbit cloned embryos we used as control in this study. Karyotype analyses confirmed that genetic material of blastocysts was derived from the cynomolgus donor cells at blastocyst stage. Mitochondrial constitution analysis of the cynomolgus-rabbit cloned embryos indicated that mitochondria from both donor cells and enucleated oocytes coexisted. After culture for 168 h post-nuclear transfer, all cynomolgus-rabbit cloned embryos in TCM-199 were arrested at the 8-cell stage, but some of them developed to the blastocyst stage in RPMI 1640 (11/59, 18.6%). In this experiment, the nutrition requirement in vitro and the cleavage rate at each 24 h were examined. When TCM-199 was supplemented with lactate, some of these embryos developed to the blastocyst stage (15.3%, 2/13). This means that cynomolgus-rabbit cloned embryos might be controlled by the donor nucleus even in these early developmental stages. However, the timing of cleavage of cynomolgus-rabbit cloned embryos is very similar to that of the rabbit-rabbit cloned embryos. Time of cleavage may depend on the protein accumulated in the cytoplasm. In the prolonged culture of reconstructed embryos on feeder cells, adhesion cells were observed. These cells are also very similar to the cells derived from cynomolgus embryos by the same method. Our results suggest that: (1) a cynomolgus nucleus can co-ordinate with rabbit oocyte cytoplasm in early embryo development, (2) the 8- to 16-cell stage block in the cynomolgus-rabbit cloned embryos may due to the same reason as that in the cynomolgus embryos, and (3) ooplasmic factors that control time of cleavage are highly conservative between the species.

118 EXPRESSION PROFILES OF CIRCADIAN CLOCK GENES IN MOUSE OOCYTES AND PRE-IMPLANTATION EMBRYOS

Matsuo et al. reported that circadian clock genes regulate the timing of cell division in mouse regenerating liver cells (2003). Their results suggested the importance of circadian clock genes for organs or tissues for which functions are characterized by cell division, such as pre-implantation embryos. To obtain basic information on the molecular functions of circadian clock genes in pre-implantation embryos, we investigated the expression profiles of transcripts and proteins of some circadian clock genes, clock, bmal1, cry1, and per2, in mouse germinal vesicle oocytes (GV), MII oocytes (MII), and pre-implantation embryos using real-time PCR and immunocytochemistry (ICC). Germinal vesicle oocytes were collected from ICR females at 48 h after PMSG priming. The mouse at 48 h after PMSG priming was primed with hCG, and MII were collected at 15 h after hCG priming. The pre-implantation embryos were collected at 6, 12, 24, 36, 48, 60, 72, 84, and 96 h after insemination, and they corresponded to early 1-cell, late 1-cell, early 2-cell, late 2-cell, 4-cell, 8-cell, early morula, late morula, and blastocyst stages, respectively. cDNA was produced by mRNA isolated from 20 oocytes or embryos using oligo dT and was subjected to real-time PCR using a TaqMan Probe system (ABI). Three sets of 20 oocytes or embryos at each developmental stage were applied to mRNA extraction and real-time PCR analysis to ensure equal mRNA extraction efficiency between samples. The level of mRNA of each clock gene contained in 3 samples from each developmental stage was almost the same. Statistical analysis of the transcripts of each gene were done by ANOVA. Germinal vesicles, MII and embryos collected at each time point were subjected to ICC using antibodies of CLOCK, BMAL1, CRY1, and PER2. The oocytes or embryos treated with only secondary antibody did not produce any signal. All of the examined genes except per2 were expressed in oocytes and pre-implantation embryos. The transcript level of clock, bmal1, and cry1 in MII were significantly lower than those in GV (P < 0.05). After fertilization, transcript levels of clock, bmal1, and cry1 significantly decreased from early 1-cell stage to late 2-cell stage (P < 0.05). These decreased transcript levels were maintained until the blastocyst stage after the late 2-cell stage. Immunocytochemistry analysis showed the nuclear localization of CLOCK and BMAL1 in early and late 2-cell embryos and of CRY1 in early 2-cell embryos but no signals of PER2 in oocytes or pre-implantation embryos. Because mouse oocytes and 1- to 2-cell embryos are transcriptionally inert, the abundant transcripts of clock, bmal1, and cry1 in these stages seemed to indicate that they were synthesized and stored during the oocyte growth phase. Moreover, the nuclear localization of CLOCK, BMAL1, and CRY1 in the oocytes and 1- to 2-cell stage embryos suggested that some clock genes were translated and worked for oocyte maturation and early embryogenesis. This study was supported by a Grant-in-Aid for the 21st Century COE Program of the Japan Mext and by a grant for the Wakayama Prefecture Collaboration of Regional Entities for the Advancement of Technology Excellence of the JST.

207 ESTABILSHMENT OF CYNOMOLGUS MONKEY EMBRYONIC STEM CELL LINES AND CONFIRMATION OF THE POSSIBILITY FOR GERMINAL COMPETENCY

Embryonic stem cells (ESCs) of nonhuman primate are important tools for human gametogenesis research. Generally, ESCs, embryos, and fetuses of nonhuman primates are similar to these of human. Recently, germ cell formation of mouse ESCs in vitro has been reported. In this study, we established new cynomolgus monkey ES (cyES) lines and determined germinal competency by assessing expression of mRNA markers. CyES lines were established using blastocysts produced by intracytoplasmic sperm injection (ICSI). For inducing super-ovulation, females were treated with 25 IU/kg pregnant mare serum gonadotropin (PMSG) once a day for 9 days, followed by 400 IU/kg hCG. Oocytes were collected 40 h after injection of hCG. After sperm injection, embryos were cultured in mCMRL medium to the blastocyst stage. For ES line establishment, inner cell masses (ICMs) were isolated by immunosurgery. ESC colonies emerged at about 10 days after ICM plating; three cyES cell lines were successfully obtained (3/11; 27.3%). We characterized these lines by immunocytochemistry for Oct-3/4, SSEA-3, and SSEA-4, which are diagnostic markers for primate ESCs, and by assay for alkaline phosphatase (ALP) activity. All cell lines expressed Oct-3/4, SSEA-4 and ALP activity. The previously reported SSEA-3 weak expression in cyES cells was not observed. These lines differentiated spontaneously when they were replaced in non-adherent culture (embryoid body: EB) or injected into SCID mice subcutaneously. To assess germ cell competency in vitro, we analyzed for the presence of vasa mRNA which shows a restricted expression pattern to germ cell formation, and DMC1 and SYCP1 which show specific existence on synaptonema complex in meiosis. Detection of these germ cell markers was performed by RT-PCR with total cDNA from ESCs and EBs. Nanog mRNA was detected only in ESCs. Oct-4 was detected in gonadal tissue of both sexes, ESCs, and EBs. Vasa was expressed in testis, but not in ESCs or somatic cells. Interestingly, we recognized weak expression of Vasa in Day 12-16 EBs. DMC1 and SYCP1 as meiosis markers were not detected. Because Oct-4 and Vasa mRNA are transcribed simultaneously, similar to that in the early part of gametogenesis such as the latter period of primordial germ cell (PGC) migration, PGC formation in cynomolgus EBs could occurr as in some cases of mouse or human EBs previously reported. Although detailed properties such as the functions of these Vasa-positive cells have not been confirmed, these results demonstrate that cyES cells obtained in the current study might contribute to putative germ cells in vitro by differentiating to EBs. This study was supported by a Grant-in-Aid for the 21st Century COE Program of the Japan Mext and by a grant for the Wakayama Prefecture Collaboration of Regional Entities for the Advancement of Technology Excellence of the JST.

Role of cumulus cells during maturation of porcine oocytes in the rise in intracellular Ca2+ induced by inositol 1,4,5-trisphosphate

At the time of fertilization, release of inositol 1,4,5-trisphosphate (IP3) into the cytoplasm of oocytes is said to be induced by hydrolysis of phosphatidylinositol bis phosphate (PI2) via activation of phospholipase C and is responsible for the Ca2+ oscillation in oocytes immediately after sperm penetration. On the other hand, cumulus cells have been reported to play an important role in cytoplasmic maturation of mammalian oocytes and to affect embryonic development after fertilization. To obtain more information on the role of cumulus cells in cytoplasmic maturation of oocytes, the effects of cumulus cells on the rise in [Ca2+]i and the rates of activation and development of porcine mature oocytes induced by IP3 injection were investigated. Mature porcine oocytes that had been denuded of their cumulus cells in the early stage of the maturation period had a depressed rise in [Ca2+]i (4.0-6.0) and reduced rates of activation (31.4-36.8%) and development (10.0-24.4%) induced by IP3 injection compared with those of their cumulus-enclosed counterparts (7.3, 69.1% and 43.8%; P < 0.05). The [Ca2+]i rise and the rates of activation and development depressed by the removal of cumulus cells were restored by adding pyruvate to the maturation medium. Furthermore, the IP3 injection-induced depression of [Ca2+]i rise in mature oocytes derived from cumulus-denuded oocytes (DOs) was restored when they were cultured in a medium with pyruvate (3.9-6.3, P < 0.05). Also, mature oocytes from cumulus-oocyte complexes (COCs) cultured in a medium without glucose had a lower rise in [Ca2+]i than that in mature oocytes from COCs cultured with glucose (7.4-6.0, P < 0.05). Cumulus cells supported porcine oocytes during maturation in the rise in [Ca2+]i induced by IP3 and the following activation and development of porcine oocytes after injection of IP3. Moreover, we inferred that a function of cumulus cells is to produce pyruvate by metabolizing glucose and to provide oocytes with pyruvate during maturation, thereby promoting oocyte sensitivity to IP3.

40 CHARACTERIZATION OF EARLY G1 CELLS AS NUCLEAR DONORS FOR SOMATIC CELL CLONING IN CATTLE

In somatic cell cloning, the cell cycle phase of the donor cells has critical impact on nuclear reprogramming and chromosomal normality of the reconstructed embryos. Recently, enhanced development to full term was obtained with embryos reconstructed with bovine fibroblasts soon after cell division (early G1 cells, Kasinathan P et al. 2001 Nat. Biotech. 19, 1176–1178; Urakawa M et al. 2004 Theriogenology 62, 714–728). In this study, to investigate the detailed cell cycle characteristics and gene expression of the early G1 cells as nuclear donors, we examined the cell proliferating and nuclear activity by detecting PCNA and Ki-67 in the cells, and the gene expression in the cells transfected with the luciferase gene. Bovine fibroblasts were transfected with chicken β-actin/firefly luciferase fusion gene (β-act/luc+), and stably transfected; cloned cells were used for cell analysis. We compared cell cycle characteristics for quiescent cells (0.4% serum for 7 days), cell doublets (early G1 cells) prepared by the “shake-off” method, and proliferating (30 to 40% confluency) cells. The presence of PCNA and Ki-67 and the incorporation of BrdU in the cells were determined by immunohistochemical analysis. The LUC+ signal (luminescence) in the cells was detected with an imaging photon counter for 10 consecutive min. Embryos reconstructed with these cells were cultured for 168 h for examination of blastocyst development. Experiments were repeated three times, and the data were analyzed with Fisher's PLSD test following ANOVA. Incorporation of BrdU was observed only in proliferating cells (24% of the cells). Neither PCNA nor Ki-67 signals were detected in the quiescent cells. PCNA was detected but Ki-67 was not detected in early G1 cells. Both PCNA and Ki-67 were detected in the proliferating cells. A strong LUC+ signal (6354 ± 673 pixels/cell) was detected in the proliferating cells, and weak signals were detected in the early G1 (2044 ± 303 pixels/cell, P < 0.05) and quiescent cells (617 ± 59 pixels/cell, P < 0.05). The rate of blastocyst development with early G1 cells was higher (45/133, 32%) than that with starved and proliferating cells (47/233, 21%, and 41/258, 14%, respectively, P < 0.05). These results indicate that early G1 cells were actively proliferating cells because of the positive PCNA signals, but their nuclei were silent because of the absence of Ki-67 signals and the weak LUC+ signals. These characteristics of the early G1 cells might enhance the development of the reconstructed embryos. This study was supported by a Grant-in-Aid for the 21st Century COE Program of the Japan MEXT, and by a grant from the Wakayama Prefecture Collaboration of Regional Entities for the Advancement of Technological Excellence of the JST.

276 THE EFFECT OF CUMULUS CELLS DURING MATURATION ON THE RISE IN THE CONCENTRATION OF INTRACELLULAR Ca2+ ([Ca2+]i) OF PORCINE OOCYTES INDUCEDBY INOSITOL 1,4,5-TRISPHOSPHATE

Increase of inositol 1,4,5-triphosphate (IP3) in the cytoplasm of mammalian oocytes is said to be responsible for [Ca2+]i oscillation observed in the oocytes immediately after sperm penetration, and the [Ca2+]i oscillation is known to be essential for the development of embryos. On the other hand, cumulus cells have been reported to play an important role in cytoplasmic maturation of oocytes and affecting the embryonic development. To obtain more information about the role of cumulus cells in cytoplasmic maturation, the effects of cumulus cells during maturation on the rise in [Ca2+]i and on the rate of activation of porcine mature oocytes induced by IP3 injection were investigated. The immature porcine oocytes were divided into three groups: COCs (intact cumulus-oocyte complexes), DOs (oocytes denuded of their cumulus cells), Co-culture (DOs attached to separated cumulus cells). These groups of immature oocytes were cultured in NCSU23 46 h for maturation. To examine the function of cumulus cells, two groups of immature oocytes were also prepared: DOs + pyruvate (DOs put into NCSU23 with pyruvate) and COCs-glucose free (COCs put into NCSU23 without glucose). The mature oocytes from each group were loaded with Ca2+ indicator fluorescent dye Fura2-AM, and then were irradiated by 340 nm and 360 nm of ultraviolet immediately after the injection of IP3. The intensities of emission light caused by the irradiation of 340 nm and 360 nm ultraviolet were recorded as E340 and E360. Since coupling of Ca2+ and the dye intensifies E340, but does not change E360, the level of [Ca2+]i was shown as R (ratio = E340/E360) in this study. Activation rate was calculated by counting the number of the oocytes that formed pronuclei by injection of IP3. ANOVA and Student's t-test were used in this study. Transient rise in [Ca2+]i was observed in the mature oocytes from every group. The peak R of the rise in [Ca2+]i of the mature oocytes derived from COCs, Dos, and Co-culture and induced by IP3 were 7.2, 4.0, and 6.9, respectively. The R of DOs was significantly lower than those of the others (P < 0.05). Also, the activation rate of the mature oocytes from DOs was significantly lower than those from COCs and Co-culture (31, 66, and 66%). The mature oocytes from DOs + pyruvate showed the same level of peak R compared with those from COCs (7.4 and 6.3), but COCs-glucose free showed a slight but significantly lower peak R compared with the mature oocytes from COCs (6.0 and 7.4, P < 0.05). In conclusion, cumulus cells appeared to support the rise in [Ca2+]i of porcine oocytes induced by IP3 during maturation and the following activation. Moreover, a function of cumulus cells supposedly produces pyruvate by metabolizing glucose and provides it to oocytes during maturation for promoting the cytoplasmic maturation. A part of this study was supported by a Grant-in-Aid for the 21st Century COE Program of the Japan MEXT, and by a grant from the Wakayama Prefecture Collaboration of Regional Entities for the Advancement of Technological Excellence of the JST.

228 EXPRESSION OF zag1 IN MOUSE PRE-IMPLANTATION EMBRYOS

Embryonic gene activation (EGA) first occurs during the second half of the mouse 1-cell embryo (Latham KE 1999 Int. Rev. Cytol. 193, 71–124). Moreover, precise regulation of EGA is considered to be essential for normal embryo development. To understand the molecular basis for the regulation of EGA, we have focused on the identification and functional characterization of genes activated at the late 1-cell stage of the mouse embryo. Recently, we have identified and isolated a novel gene, termed zag1 (zygotic activating gene 1), transcribed specifically at the EGA, using a fluoro-differential display method with oocytes and embryos at 15 h post-insemination. Messenger RNA of zag1 expressed at lower level in the oocyte than that in the embryo at 15 h post-imsemination. In this study, we investigated the potential function of zag1 by analysis of mRNA expression and protein distribution in mouse tissues and pre-implantation embryos. Nucleotide sequence analysis of zag1 cDNA revealed that the open reading frame of 1726 bps encodes a protein of 575 amino acids with a predicted molecular mass of 66 kDa. The deduced amino acid sequence indicated that zag1 protein might be a soluble protein with a bipartite nuclear targeting sequence, a NACHT NTP domain, and an APT/GTP binding site motif as a predicted functional domain. Two μg of Poly(A)+ RNA from various tissues of adult mice were subjected to Northern blot analysis using the mouse zag1 cDNA probe. We detected this gene abundantly expressed in mouse testis and ovary by approximately 2- to 3-fold compared with one in other mouse tissues (heart, liver, kidney, lung, brain, skeletal muscle, and spleen). zag1 transcript and protein, as assessed by RT-PCR and immunoblotting, respectively, were slightly present in ovulated oocytes, gradually decreased in the early 1-cell embryos, but re-expressed in the late 1-cell and early 2-cell stage embryos which coincided with the mouse EGA. Subsequent to microinjection of an expression vector encoding zag1-enhanced green fluorescent protein (EGFP), fused protein into male pronucleus of 1-cell embryos was detected in the nuclei of 2-cell embryos. These findings suggest that zag1 may be functionally associated with early embryonic development. This study was supported by a Grant-in-Aid for the 21st Century COE Program of the Japan MEXT, and by a grant from the Wakayama Prefecture Collaboration of Regional Entities for the Advancement of Technological Excellence of the JST.

222 METHYLATION OF THE 52-UPSTREAM REGION OF THE H19 GENE IN MOUSE SOMATIC CELL, GAMETES, WILD TYPE AND ANDROGENETIC ES CELLS

In mammals, several genes influenced by the phenomenon of genomic imprinting are critical during development. Recently, Kono et al. (2004 Nature 428, 860–864) reported the production of intact female mouse individuals that had only two haploid sets of maternal genome. They obtained these mice by combining a normal haploid maternal genome and a mutant haploid maternal genome with a 13 k base deletion in the H19 gene and its 5′-upstream region. This genomic combination resulted in the appropriate expression of the Igf2, H19, and other imprinted genes. In the mouse genome, there are four CTCF-binding sites in the 5′-upstream region of the H19 gene. The binding of CTCF to these binding sites regulates the expression of the Igf2 and H19 genes. The binding of CTCF to its binding sites is regulated by methylation of CpG sites in binding sites. In this study, as the first step to elucidate the role of the paternal genomic imprinting during development, we investigated the methylation of CpG sites in the 5′-upstream region of the H19 gene in mouse somatic cells, gametes, and two types of ES cells. Genomic DNA was isolated from BDF1 (C57BL/6N × DBA/2N) mouse's tail (male and female somatic tissue, mST and fST, respectively), spermatozoa (S), oocytes (O), and wild type and androgenetic embryonic stem cells (wtES and agES, respectively). The methylation of CpG sites was evaluated by using the bisulfite sequencing assay. There were 13 CpG sites and a CTCF-binding site in the region from −4413 to −3976 in the H19 gene relative to the transcription start site. The percentages of CpG sites in this region that were methylated were 88% (160/182), 79% (27/130), 93% (230/247), 8% (10/130), 77% (10/13) and 89% (314/351) for mST, fST, S, O, wtES, and agES, respectively. In the CTCF-binding site core motif (CCGCGTGGTGGCAG), the percentages of methylated CpG sites were 93% (26/28), 80% (16/20), 95% (36/38), 0% (0/20), 50% (1/2) and 96% (52/54) for mST, fST, S, O, wtES, and agES, respectively. The CpG sites in the sequence of agES were highly methylated similar to the finding in spermatozoa. However, an aberrant methylation pattern was observed in some clones of agES. From these results, it was concluded that the methylation of CpG sites in the genomic sequence of agES was well conserved and, therefore, agES is useful to elucidate the role of the paternal genomic imprinting during development. This work was supported by Wakayama Prefecture Collaboration of Regional Entities for the Advanced of Technological Excellence, Japan, and by a Grant-in-Aid for the 21st Century COE Program of the Japan MEXT.

313 EFFECTS OF ETHANOL TREATMENT AFTER INTRACYTOPLASMIC SPERM INJECTION (ICSI) ON SPERM AFTER FORMATION AND THE MICROTUBULE ORGANIZATION OF BOVINE OOCYTES

The cleavage rate of bovine embryos is very low without activation of oocytes after intracytoplasmic sperm injection (ICSI), although both male and female pronuclei are formed. We previously reported that the stimulus due to the injected sperm alone was sufficient to lower the MPF activity of bovine oocytes after ICSI, and the activation treatment of oocytes with ethanol at 4 h after ICSI served to maintain the low levels of MPF activity until the next cell cycle started (Fujinami et al. 2004 J. Reprod. Dev. 50, 171–178). These results suggested that activation treatment is necessary to improve the embryonic development after bovine ICSI. In bovine fertilization, the sperm introduces the centrosome into the oocyte. The centrosome acts as the microtubule-organizing center and microtubules are organized within the oocyte. It is reported that the sperm aster is important for the normal fertilization process. Therefore, failure of sperm aster formation possibly causes the failure of cleavage following fertilization. To investigate the reason of the low cleavage rate after bovine ICSI without artificial activation treatment, we examined sperm aster formation and the microtubule organization in bovine oocytes with or without activation treatment after ICSI. Bull spermatozoa immobilized by piezopulse was injected into bovine oocytes matured in vitro. At 4 h after ICSI, oocytes were treated with 7% ethanol in TCM199 for 5 min for activation. Oocytes were fixed at 6 and 12 h after ICSI, and the microtubule organization was examined by using specific antibodies and immunofluorescence microscopy. The cleavage rate (51% vs. 15%) and the developmental rate to the blastocyst stage (13% vs. 3%) were increased by ethanol treatment after ICSI (with or without ethanol treatment, respectively, P < 0.05). In oocytes activated with ethanol after ICSI, both the sperm aster formation rate at 6 h and the microtubule organization rate at 12 h after ICSI were significantly higher than in oocytes without activation treatment (58%, 80% vs. 12%, 26%, P < 0.05). It was reported that the sperm aster has an important role for the pronuclear movement to make the male and female pronuclei come into close apposition. From these results, it was concluded that oocyte activation after bovine ICSI promoted sperm aster formation and microtubule organization, and was effective to improve embryonic development. This study was supported by a Grant-in-Aid for the 21st Century COE Program of the Japan MEXT, and by a grant from the Wakayama Prefecture Collaboration of Regional Entities for the Advancement of Technological Excellence of the JST.

62 RELATION OF INTENSITY OF GENE EXPRESSION IN BOVINE RECONSTRUCTED EMBRYOS TO SUBSEQUENT DEVELOPMENT

During embryo development, embryonic gene activation (EGA) is the first critical event. We previously showed that EGA is also critical for further development in somatic cell-cloned embryos (Saeki K et al. 2004 Reprod. Fertil. Dev. 16, 157–158 abst). To show this, we reconstructed bovine embryos with bovine somatic cells transfected with chicken β-actin/firefly luciferase fusion gene (β−act/luc+) and showed that only luminescent embryos at 60 hours post-fusion (hpf) developed to the blastocyst stage. In this study, we examined the relation between the intensity of expression of the same reporter gene in embryos reconstructed with bovine β−act/luc+ fibroblasts and their subsequent development to the blastocyst stage. Bovine fibroblasts were transfected with β−act/luc+ as described earlier (Saeki K et al. 2004 Reprod. Fertil. Dev. 16, 157–158 abst). The stably transfected and cloned cells were cultured for several passages. The cells were cultured under serum starvation (0.4% FCS) for 7 days and then used as donor cells. In vitro-matured bovine oocytes derived from slaughterhouse ovaries were enucleated at 20 h post maturation. Enucleated oocytes were electrofused with the cells, and activated with a calcium ionophore and cycloheximide. The LUC+ signal (luminescence) in the embryos was detected in medium containing 500 μg mL−1 luciferin with an imaging photon counter (ARGUS 50, Hamamatsu, Japan) for 30 consecutive min at 60 hpf. The intensity of luminescence in embryos (4- to 8-cell stage) was graded as being strong (>10 × 104 pixels/embryo), intermediate (5 to 10 × 104 pixels/embryo), weak (<5 × 104 pixels/embryo), or absent. The embryos were cultured separately until 168 hpf, and examined for blastocyst development. Experiments were repeated four times, and the data were analyzed with Fisher's PLSD test following ANOVA by Stat View software (Ver. 5.0; abacus Concepts, Berkeley, CA, USA). Of 125 embryos that were reconstructed, 74 (59%) developed to the 4- to 8-cell stage at 60 hpf. The luminescence was strong in 29 (39%) of the embryos, intermediate in 12 (16%), weak in 19 (26%), and absent in 14 (19%). Blastocysts were obtained from a group of embryos that exhibited strong luminescence (10/29, 34%), but none of the embryos from the other groups developed to blastocysts. These results suggest that active gene expression in embryos reconstructed with somatic cells is important for their subsequent development. This study was supported by a Grant-in-Aid for the 21st Century COE Program of the Japan Ministry of Education, Culture, Sports, Science, and Technology, and by a grant from the Wakayama Prefecture Collaboration of Regional Entities for the Advancement of Technological Excellence of the JST.

115 CYTOLOGICAL ANALYSIS OF HEPATIC GENE EXPRESSION AND IMMUNOLOGICAL RESPONSE OF MHC ANTIGENS IN MOUSE AMNIOTIC EPITHELIAL CELLS

Human amniotic epithelial cells (hAECs) have been reported to have unique properties. They express almost no class I and class II MHC antigens and lack response to interferon-γ (IFN-γ) which mediates the expression of those MHC molecules. Moreover, hAECs express some genes characteristic of hepatic cells. Therefore, hAECs seem to have multipotency and are expected to substitute for hepatic tissues in part. We aimed to develop the experimental model for investigating AECs in mice (mAECs). In this study, we examined the induction of MHC molecules by IFN-γ and the hepatic gene expression in mAECs. Murine amniotic membranes were collected from C57BL/6J females at 17.5 days of gestation. They were digested by 0.03% hyaluronidase followed by 0.2% collagenase treatment. Dissociated mAECs were cultured on dishes in DMEM supplemented with 10% FBS at 37°C under 5%CO2 in air. Embryonic fibroblasts (EFs) collected from C57BL/6J fetuses at 13.5 dpc were cultured in the same condition as mAECs. In Experiment I, the effect of IFN-γ on induction of MHC molecules in mAECs was examined. mAECs and EFs cultured in the presence or absence of IFN-γ at 1 × 103 U mL−1 for 72 h were recovered and incubated with FITC-conjugated antibodies against mouse H-2 MHC class I or I-A/I-E MHC class II antigens. The cells were analyzed by flow cytometry. In Experiment II, the expression of the genes in mAECs was examined by RT-PCR. mRNA was purified from adult liver, EFs, fresh mAECs, and mAECs cultured for 5 days. As the genes characteristic for hepatic cells, HNF-3α, HNF-3β, HNF-3γ, HNF-4, transthyretin (TTR), albumin, α-fetoprotein (AFP), glucose-6-phosphatase (G6P), and asialoglycoprotein receptor-1 (Asgr1) were examined. In Experiment I, cell-surface expression of class I and class II MHC antigens in response to IFN-γ was observed weakly in mAECs as compared to EFs, suggesting different property in hAECs which lack the expression of those antigens. In Experiment II, RT-PCR analysis showed that all of the genes except G6P were expressed in fresh mAECs. However, the expression of transcription factors such as HNF-3α, HNF-3β, HNF-4, and TTR, serum proteins such as albumin and AFP, and Asgr1 decreased after in vitro culture, contrary to the case of hAECs in which, for example, albumin appeared after cultivation. In conclusion, it was evident that mAECs have quite different properties, both in the inductivity of MHC molecules and the expression of hepatic genes, from hAECs. This work was supported by Wakayama Prefecture Collaboration of Regional Entities for the Advancement of Technological Excellence of the JST, and by a Grant-in-Aid for the 21st Century COE Program of the Japan MEXT.

339CORRELATION BETWEEN FOLLICLE SIZE AND QUALITY OF OOCYTES FROM THE SUPEROVULATED CYNOMOLGUS MONKEY

Variability in the superovulatory response continues to be one of the most frustrating problems with the application of assisted reproductive technologies in non-human primates. Superstimulation of donor animals with equine chorionic gonadotrophin (eCG) and human chorionic gonadotrophin (hCG) is widely used, but individual responses to these hormones vary widely. In human in vitro fertilization, follicle size is commonly used as a marker to determine the timing of hCG treatment in order to acquire oocytes matured in vivo. Incorporation of techniques designed to control follicular size in humans may improve superstimulatory response in non-human primates. In this study, we measured follicle size and examined oocyte quality in Cynomologus Monkeys superstimulated as described below. On the first day of spontaneous menses, monkeys were treated with long-acting GnRH (Luprin: Takeda Pharm, Co. Ltd., Osaka, Japan;; Day 0). A dose of 25IU/kg/day eCG (Serotoropin;; Teikokuzoki Pharm, Co. Ltd., Tokyo, Japan) was administered i.m. from Day 1 to Day 9, followed by 200IUkg−1 hCG (Puberogen;; Sankyo Pharm, Co. Ltd., Tokyo, Japan) i.m. on Day 10. Dominant follicle sizes were measured on Days 7 and 9 by ultrasonography. Oocytes were collected by laparoscopy in anesthetized monkeys 40h after the hCG injection. Oocytes were classified by nuclear status. Immature oocytes at the stages of germinal vesicles (GV) and metaphase (MI) were cultured until reaching the stage of Metaphase II (MII). Matured oocytes (MII) were fertilized by ICSI and cultured for 7 days. At the end of culture, the developmental stage of oocytes was examined. The ovaries with different-sized follicles on Day 7 were divided into two groups;; ovaries with large follicles (&gt;4.5mm) were in the first group and ovaries with small follicles (&lt;4.0mm) were in the second group. On Day 9, follicles in first group grew to more than 5.0mm and follicles in second group remained less than 5.0mm. Sixty-two percent of oocytes from follicles in first group were at MI or MII stage, while only 15% of oocytes in second group reached the MI or MII stage. After ICSI, 42% of MII oocytes from first group developed to the blastocyst stage, while no blastocyts were observed in second group. These results suggest that the size of dominant follicle was a limiting factor for the developmental ability of oocytes in vitro. For production of Cynomolgus monkey blastocysts derived from ICSI, the diameter of dominant follicle was required to be at least 5mm before hCG in order to collect MI and MII oocytes. Incorporation of hormonal treatments designed to optimize follicular size probably reduced the variability in quality of oocytes. Therefore, we expect that an adjustment of dose and duration of eCG and hCG treatment may improve developmental ability of oocytes from follicles that had not reached 5mm.

72EFFECTS OF GENE EXPRESSION IN BOVINE EMBRYOS RECONSTRUCTED WITH FIBROBLASTS TRANSFECTED WITH LUCIFERASE GENE ON THE SUBSEQUENT DEVELOPMENT

During embryo development, embryonic gene activation (EGA) is one of the first critical events. Inappropriate EGA results in failure of further development. We have reported that gene expression in bovine embryos reconstructed with fibroblasts begins at 48 hours postfusion (hpf) and reaches a maximum level at 60hpf as detected by their bioluminescence following injection of chicken β-actin/firefly luciferase fusion gene (β-act/luc+) into their nuclei (Saeki et al., 2001 Theriogenology 55, 289). In the present study, effects of gene expression in embryos reconstructed with bovine fibroblasts transfected with luciferase gene on their subsequent development to the blastocyst stage were examined. Cultured bovine fibroblasts taken from an ear of a female calf were transfected with plasmid containing β-act/luc+/IRES/EGFP and neor using GeneJammer (StrataGene, La Jolla, CA, USA). Neomycin-resistant cells were selected by culturing with G418. Then, EGFP-positive colonies were further selected under fluorescence microscopy to obtain stably transfected cells. The transfected cells were cultured for several passages. Growing (50 to 60% confluence, GCs) and serum-starved cells (SCs) were used as donor cells. In vitro-matured bovine oocytes derived from slaughterhouse ovaries were enucleated at 20h post maturation. Enucleated oocytes were electrofused with the cells, and activated with calcium ionophore and cycloheximide. Luminescence in the embryos was detected with an imaging photon counter at 0 and 60hpf. Luminescence-positive (P) and -negative (N) embryos were cultured separately at each detection time. Embryos were cultured until 168hpf, and examined for cleavage and blastocyst development. Experiments were repeated 3 times, and totals of 91 and 123 embryos were reconstructed with GCs and SCs, respectively. Data were analyzed with Fisher’s PLSD test following ANOVA by Stat View software (Ver. 5.0). At 0hpf, luminescence was detected in 55 and 4% of embryos reconstructed with GCs and SCs, respectively. At 60hpf, luminescence was detected in 47 and 28% of P and N embryos with GCs, and 17 and 40% of P and N embryos with SCs at 0hpf, respectively. Cleavage rates were not different among groups (P&gt;0.05). Blastocysts were obtained only from the groups of embryos that were N at 0hpf and P at 60hpf (8% with GCs and 17% with SCs). No embryos in the other groups developed to the blastocyst stage. These results suggest that appropriate gene expression in embryos reconstructed with somatic cells is important for their subsequent development and that detecting the reporter gene expression can be used for selection of viable cloned embryos.

191IN VIVO AND IN VITRO DIFFERENTIATION OF EMBRYONIC STEM CELLS DERIVED FROM PARTHENOGENETIC EMBRYOS IN MICE

Availability of embryonic stem (ES) cells opens the prospect for regenerative medicine. However, ES cells genetically mismatched to diseased individuals cause immunological rejection. In this study, we established ES cells from parthenogenetic embryos in mice and examined their pluripotency. Oocytes were collected from (C57BL/6xDBA)F1 mice (BDF1) by superovulation. Parthenogenetic diploid embryos were produced by activation treatment in 5mM SrCl2 in Ca2+-free KSOM medium for 2h, followed by cultivation in 5μgmL−1 cytochalasin B for 6h. The zonae pellucidae of embryos developed to the blastocyst stage in vitro were removed by a 5-min incubation in 0.5% pronase. Inner cell masses (ICMs) isolated immunosurgically were seeded on the feeder layers (mitomycin C-treated mouse embryonic fibroblasts) in DMEM supplemented with 15% Knock-Out Serum Replacement (Invitrogen), 2mM L-glutamine, non-essential amino acids, β-mercaptoethanol and 103UmL−1 of Leukemia inhibitory factor (LIF) at 37°C in a humidified atmosphere with 5% CO2 in air. The attached ICM cells were mechanically disaggregated and seeded on the fresh feeder layers. After several passages, parthenogenetic ES (PnES) cell lines were established. The efficacy of establishing PnES cell lines was 66% (37/56). To examine the characteristics of PnES cell lines, seven lines were subjected to histochemical and immunohistochemical analysis. All showed alkaline phosphatase activity and immunoreactivity to anti-SSEA-1 and anti-Oct4 antibodies. They maintained euploid sets of choromosomes at 29; 59%. PnES cells from two of the seven lines were injected into 59 host blastocysts obtained from ICR mice, resulting in 16 chimeric offspring (27%). In another experiment, injection of ICM cells and ES cells obtained from fertilized BDF1 blastocysts and ICM cells obtained from BDF1 parthenogenetic blastocysts also produced chimeric offspring (35%, 7/20; 46%, 6/13; and 53%, 10/19, respectively). However, no chimeric mouse with germline transmission was obtained from PnES cells. Injection of 1×107 of PnES cells into SCID mice formed teratocarcinomas. Immunohistochemical analysis showed cells positive for nestin (specific to neuroepitherial stem cells), Tu-J (class III β-tublin), NF-M (neurofilament), desmin (muscle), and albumin (hepatocytes), which indicated their differentiation potency to the cells derived from all three germ layers. Simple embryoid bodies produced from these cell lines were plated on tissue culture dishes under conditions for induction of differentiation. Immunohistochemistry and RT-PCR analysis showed their differentiation into neurons (NF-M, nestin), cardiomyocytes and hepato-like cells (albumin, α-fetoprotein). Our results indicate that PnES cells are pluripotent similar to the ES cells from fertilized embryos except for germline transmission and should be tested in cell replacement animal models.

Establishment of embryonic stem cell lines from cynomolgus monkey blastocysts produced by IVF or ICSI

Human embryonic stem (ES) cells are predicted to be a valuable source for producing ES-derived therapeutic spare tissues to treat diseases by controlling their growth and differentiation. To understand the regulative mechanisms of their differentiation in vivo and in vitro, ES cells derived from nonhuman primates could be a powerful tool. We established four ES cell lines from cynomolgus monkey (Macaca fascicularis) blastocysts produced by in vitro fertilization (IVF) or intracytoplasmic sperm injection (ICSI). The ES cells were characterized by the expression of specific markers such as alkaline phosphatase and stage-specific embryonic antigen-4. They were successfully maintained in an undifferentiated state and with a normal karyotype even after more than 6 months of culture. Pluripotential competence was confirmed by the formation of teratomas containing ectoderm-, mesoderm-, and endoderm- derivatives after subcutaneous injection into SCID mice. Differentiation to a variety of tissues was identified by immunohistochemical analyses using tissue-specific antibodies. Therefore, we established pluripotent ES cell lines derived from monkeys that are widely used as experimental animals. These lines could be a useful resource for preclinical stem cell research, including allogenic transplantation into monkey models of disease.

Defining a common region of DNA amplification at 22q11.2-12 in head and neck squamous cell carcinomas by quantitative FISH analysis

By comparative genomic hybridization (CGH) with 17 head and neck squamous cell carcinoma (HNSCC) cell lines, we previously detected an amplified region as a distinct peak at 22q11.2-12 in 3 cell lines. Because the possible presence of an oncogene was strongly suggested, the region was mapped in more detail by defining the minimum region that was commonly amplified by using fluorescence in situ hybridization (FISH) with a series of cosmids. Eighteen cosmid clones on 22q11.2-12 were assigned to their locations as a fractional length map and hybridized to cells from three HNSCC cell lines. The three cosmid clones, cHKA-118, cHKAD-26, and D22S938, showed the highest levels of amplification, and the size of the amplicon was calculated to be approximately 1.7 Mb in the OM1 and HSC6 cell lines. Several genes related to oncogenesis, including PRKM1, map to this locus. Thus, the definition of the common region with the highest level of copy number increases by FISH provides a starting point for identifying the gene that may play an important role in the development of HNSCC.

Timing of completion of the first meiotic division in bovine oocytes after maintenance of meiotic arrest with cycloheximide and their subsequent development

This study examined the timing of completion of meiosis I of bovine oocytes in which meiotic resumption had been inhibited by cycloheximide (CHX), and also determined the optimum interval of maturation in culture for subsequent fertilization and development. Most oocytes treated with CHX reached metaphase II at 16 hr in the maturation culture, while control oocytes did at 20 hr. CHX-treated oocytes cultured for 16 hr were normally fertilized but failed to develop into blastocysts. Maturation in culture for 20 hr resulted in comparable development for control oocytes. The results indicate that nuclear maturation of CHX-treated oocytes was completed 4 hr faster than for control oocytes, however the same interval of maturation as that of control oocytes is necessary for subsequent development to blastocysts.