We have characterised the changes in preimplantation embryos that occur in the mRNA population during the transition from maternal to zygotic control of embryogenesis. We connected the mRNA differential display method and RT-PCR based method that allows amplification of the whole population of messengers. In the early stages of development we have further characterised the level of individual mRNAs with the help of semiquantitative RT-PCR used with specific primers. This report concerns four of 12 cDNA fragments that appeared to be differentially expressed between the 4- and late 8-cell stages. A transcript corresponding to fragment no. 1/12 appears to be analogous to the maternal mRNA since it is abundant in 1-, 2-, 4- and 8-cell embryos and rapidly decreases in the later stages. A similar pattern of expression was revealed in the transcript corresponding to fragment no. 8/9. A transcript corresponding to fragment no. 20/8 is newly synthesised from the embryonic genome at the late 8-cell stage and its amount rapidly increases during the following stages. This messenger shows a 91.7% identity with mRNA for human S3A ribosomal protein and 92.2% identity with mRNA for Felis domesticus S3A ribosomal protein. A transcript corresponding to fragment no. 8/19 is stage-specific, being newly synthesised from an embryonic genome at the late 8-cell stage and decreasing in the later stages. This messenger shows 86.6% identity with a mouse mRNA for proline-rich protein and 91.6% identity with human mRNA for KIAA-0058 gene. A complex of these molecular markers represents a suitable tool for answering questions concerning the molecular control of major gene activation during bovine embryogenesis.

Chromatin remodelling in the fertilised mouse egg is intimately linked to protein synthesis and degradation, to protamine by histone replacement and to specific histone modifications. The involvement of histone deacetylases (HDACs) in the beginning of development is poorly understood. HDACs are essential for cell proliferation and in the control of gene expression in a wide variety of mammalian systems. Here we focus on mHDAC6, a recently identified class II histone deacetylase, and we analyse its expression and localisation in oocytes and pronuclear zygotes. By indirect immunofluorescence we show that mHDAC6 is detected in the cytoplasm of germinal vesicle (GV) stage oocytes and 1-cell embryos. Ectopic expression of this enzyme after injection into germinal vesicles and pronuclei alters the nuclear structure and causes premature compaction of the chromatin. Our data suggest that the effect of condensation is linked to the ubiquitin-binding activity of mHDAC6, rather than to its function as a deacetylase.

That a mammalian embryo might influence the maternal endocrinological and/or immunological systems whilst still in the Fallopian tube has long seemed possible (Fig. 1). Under normal circumstances, the embryo contains a full diploid complement of genetic information and thus the point at issue is just how soon instructions in the genome are expressed for trophic purposes. It could be argued from several points of view that it would be biologically prudent for an embryo to influence the mother whilst still at an early stage of development. Consideration of the extremely intimate apposition of the zona pellucida and the tubal epithelium - perhaps best appreciated in scanning electron micrographs (Fig. 2) - also suggests scope for a mutual exchange of signals. Even so, one of the earliest recorded maternal responses to a developing embryo is the change in permeability of the capillary bed of the uterus that precedes implantation in rodents: this can be demonstrated by the Pontamine blue reaction, a series of bands of dye in the uterine wall corresponding to the location of individual embryos, and is detectable in rats on the fourth day after mating (Psychoyos, 1960, 1973).

Cortical tissues containing only primordial and primary follicles, or secondary follicles 140-190 μm in diameter, were collected from bovine ovaries and xenografted under the kidney capsules of female severe combined immunodeficient (SCID) mice. Histological examination revealed that all grafts were well vascularised and contained surviving follicles at 4 or 6 weeks after grafting. Primordial and primary follicles survived but did not develop beyond the one-layer stage. Secondary follicles, on the other hand, had formed antra at 4 weeks after grafting. The mean diameter of secondary follicles, which was 165.2 ± 17.0 μm (n = 42) before grafting, had developed to 442.9 ± 77.9 μm (n = 37) and 592.9 ± 116.0 μm (n = 45) in diameter at 4 and 6 weeks after grafting, respectively. The mean diameter of oocytes, which was 55.1 ± 4.9 μm (n = 42) before grafting, also increased significantly (4 weeks: 105.6 ± 6.3 μm; 6 weeks: 122.2 ± 2.6 μm; p < 0.05). Oocytes were recovered from follicles that had developed to more than 400 μm in diameter after 6 weeks, and were subjected to subsequent mature culture. Of these oocytes, 34% (11/32) resumed meiosis and 6% (2/32) matured to the second metaphase. Follicular fluid in bovine antral follicles developed in SCID mice had the 69 kDa protein, which was detected by anti-mouse albumin antibody but not by anti-bovine albumin antibody in immunoblotting analysis. These results demonstrated that bovine secondary follicles develop to the antral stage in SCID mice, and that the oocytes in the follicles acquire the meiotic competence.

Protein kinase C (PKC) is a family of Ser/Thr protein kinase widely distributed in eukaryotes. There is evidence that PKC plays key roles in the meiotic maturation and activation of mammalian oocytes. However, the mechanism of PKC's actions and the PKC isoforms responsible for these actions are poorly understood. In this study, we reveal in mouse eggs and early embryos: (1) the effects of PKC on the meiotic and mitotic cell cycle progression during oocyte maturation, egg activation and embryonic cleavages; (2) the functional importance of classical PKC subclasses in these processes; and (3) the subcellular localization of the PKCα isoform during development from GV stage oocytes to the blastocyst stage embryos. The results indicate that the PKC activator phorbol 12-myristate 13-acetate (PMA) inhibits the meiotic resumption of cumulus-free mouse oocytes by a mechanism dependent not only on classical PKC activity but also on other PKC isoforms. PKC activation after germinal vesicle breakdown leads to the inhibition of mitogen-activated protein kinase phosphorylation and the arrest of cell cycle at MI stage. The second polar body emission and the cleavages of early embryos are blocked after prolonged PKC activation. The subcellular localization of PKCα isoform in mouse oocytes and embryos is developmental-stage associated. All these results suggest that PKC has multiple functional roles in the cell cycle progression of mouse oocytes and embryos.

Incomplete reprogramming of the donor cell nucleus after nuclear transfer (NT) probably leads to the abnormal expression of developmentally important genes. This may be responsible for the low efficiency of cloned animal production. Insulin-like growth factor 2 (IGF2) and IGF2 receptor (IGF2R) are imprinted genes that play important roles in preimplantation development. To obtain an insight into abnormal gene expression after nuclear transfer, we assessed the transcription patterns of IGF2-IGF2R in single in vitro fertilised and cloned embryos by reverse-transcription polymerase chain reaction (RT-PCR). IGF2R expression did not differ significantly but IGF2 was more highly expressed in cloned embryos than in IVF embryos (p < 0.05). This was confirmed by a quantitative RT-PCR method. Thus, incomplete reprogramming may induce abnormal transcription of IGF2 in cloned embryos.

The present study in rabbits compared, in the first experiment, the effect of two commonly used oocyte ages, 13 h and 17 h after ovulation induction treatment, on the technical efficiency of somatic nuclear transfer steps, using fresh cumulus cells as nuclear donors. Recently ovulated metaphase II oocytes (13 h) showed higher fusion (13 h: 83% vs 17 h: 67%, p < 0.05) and in vitro development rates than in vivo slightly aged metaphase II oocytes (morula, 13 h: 74% vs 17 h: 25%, p < 0.05; blastocyst, 13 h: 16% vs 17 h: 8%; p < 0.05). In contrast, activation rate was higher in the 17 h group (13 h: 45% vs 17 h: 67%; p < 0.05). In a second experiment, using recently ovulated oocytes (13 h) as recipients, two donor cell types (from primary cultures of either cumulus cells or fetal fibroblasts) were tested to evaluate their effects on the efficiencies of the different technical steps of somatic nuclear transfer procedure. A better fusion rate was obtained when fetal fibroblasts were used as nuclear donors (cumulus cells: 45% vs fetal fibroblasts: 67%, p < 0.05). No statistically significant differences were detected in cleavage rate regardless of the cell type used (cumulus cells: 44% vs fetal fibroblasts: 60%, p > 0.05). However, in vitro development to morula (cumulus cells: 41% vs fetal fibroblasts: 14%, p < 0.05) and to blastocyst stage (cumulus cells: 27% vs fetal fibroblasts: 3%, p < 0.05) were different between cell types.

Abnormal oocyte spindle is frequently associated with the infertility of aged women. Directly manipulating the metaphase I (MI) spindle may be a feasible method to overcome this kind of problem. Here, we report that the MI meiotic spindle can be removed from MI mouse oocytes and will autonomously divide into two daughter cells with the same size, morphology and an equal number of chromosomes after culture for 5 h in maturation medium. The division rate of the MI spindle reached 56% after 10-15 h of culture. After transferring the MI meiotic spindle into synchronous ooplasm by electrofusion, about 61% of the reconstructed oocytes continued to complete the first meiosis and extruded a normal first polar body. The matured reconstructed oocytes can also be fertilised. Approximately 50% of the 2-cell embryos developed to the morula stage after in vitro culture.

The culture of pig oocytes in the presence of the calcium channel blocker verapamil (0.02 mM) resulted in the blocking of meiosis at the metaphase I stage, and only a small fraction (about 28%) of the oocytes were able to continue their maturation to the stage of metaphase II. Hence, meiotic maturation in pig oocytes is a calcium-dependent process. After isolation of the pig oocytes from their follicles, the intracellular calcium deposits in the oocyte and granulosa cells, detectable using the combined oxalate-pyroantimonate method, are depleted. The amount of calcium deposits in the oocyte and granulosa cells increased during oocyte meiotic maturation in vitro, especially in the nucleus, mitochondria, vacuoles and cytoplasm. The replenishment of calcium deposits is significantly changed under the effect of verapamil. The increase in calcium deposits in the oocyte nucleus was delayed, a much larger amount of deposits was formed in the mitochondria, and the amount of deposits in the vacuoles was demonstrably smaller. A significant peak in the accumulation of calcium deposits was observed in the cytoplasm of verapamil-treated oocytes after 16 h of in vitro culture. We propose that an altered pattern in the replenishment of calcium deposits can disturb intracellular signalling and prevent the exit of oocytes from the metaphase I stage.

Conventional in vitro fertilization (IVF) techniques have been unable to produce normal embryos in any Australian marsupial, largely owing to problems with the early stages of sperm–oocyte binding. This study has used intracytoplasmic sperm injection (ICSI) of in vivo and in vitro matured tammar wallaby oocytes to bypass these processes and achieve fertilization in vitro. The fertilization rate (i.e. development to the two-pronuclei stage) of in vivo and in vitro matured oocytes following ICSI and sham injection was assessed at 17–19 h after injection. Fertilization occurred in 48% (45/93) of in vivo matured oocytes that were injected with spermatozoa. Significantly fewer sham-injected oocytes (6/82, P<0.005) and uninjected control oocytes (5/84, P<0.005) formed two pronuclei. In a direct comparison, the numbers of in vivo and in vitro matured oocytes that formed two pronuclei after ICSI were 22/28 (78.6%) and 23/40 (57.6%), respectively, which are not significantly different. There was also no significant difference in the nuclear response of in vivo and in vitro matured oocytes to sham injection. The numbers of oocytes forming a single pronucleus after sham injection were 10/24 (41.7%) and 24/37 (64.9) for in vivo and in vitro matured oocytes, respectively. Immature germinal-vesicle-stage oocytes were unable to decondense sperm injected during ICSI or to form pronuclei. These results demonstrate that both in vitro and in vivo matured tammar wallaby oocytes can be fertilized by ICSI. The success of ICSI not only offers the opportunity for fundamental analysis of marsupial fertilization but could, in conjunction with development of appropriate culture conditions and embryo transfer technologies, contribute to increased production of offspring from rare or valuable marsupials.

The effect of the p42/44 mitogen-activated kinase (MAPK) inhibitor, PD98059, on MAPK activation and meiosis resumption in mouse oocytes was studied. When germinal vesicle (GV)-stage denuded oocytes (DOs) were cultured continuously in 50 μM PD98059, germinal vesicle breakdown (GVBD) was postponed for 2-3 h. MAPK phosphorylation and activation was delayed as well. However, PD98059 did not impair histone H1 kinase activation. After 14 h of culture there was no significant difference in the rate of DOs reaching metaphase II (MII) arrest in either control or experimental conditions. The effect of PD98059 on MAPK inhibition was further tested in epidermal growth factor (EGF)-treated oocyte–cumulus complexes (OCCs). Exposure of GV-stage OCCs for 5 min to EGF (10 ng/ml) induced a considerable increase in MAPK phosphorylation. After OCCs were further cultured in 50 μM PD98059 a rapid dephosphorylation of MAPK was induced. Already after 1 min of treatment the non-phosphorylated form of MAPK dominated, indicating the high effectivity of PD98059. This result indicates that short EGF/PD98059 treatment of OCCs induced MAPK phosphorylation/dephosphorylation in cumulus cells only. As only a transient delay in MAPK phosphorylation and activation was observed in PD98059-treated DOs we conclude that there is also another PD98059-nonsensitive pathway(s) leading to MAPK activation in mouse oocytes. The data obtained suggest that meiosis resumption in mouse oocytes is somehow influenced by the MEK/MAPK activation pathway.

Ovulated mouse oocytes are activated by exposure to culture medium containing Sr2+ or Ba2+ or by intracytoplasmic injection of the divalent cations. It is known that in vitro matured pig oocytes are activated by the intracytoplasmic injection of Ca2+. In this study, we examined the effect of exposure and of intracytoplasmic injection of Sr2+ or Ba2+ on in vitro matured pig oocytes (MII-oocytes). When MII-oocytes were exposed to the medium containing divalent cations, no oocytes were activated. However, in the case of oocytes that were injected with Sr2+, Ba2+ and Ca2+, at 6 h after injection, 64%, 71% and 86% of the oocytes had been released from MII-arrest, and 51%, 67% and 84% formed female pronuclei, respectively. The initial transient in intracellular Ca2+ concentration ([Ca2+]i) was measured by the Ca2+ indicator dye fluo-4 dextran. Microinjection of Sr2+, Ba2+ or Ca2+ induced a rapid elevation of [Ca2+]i. The exocytosis of cortical granules was examined by staining with fluorescein isothiocyanate (FITC)-labelled peanut agglutinin. After an injection of divalent cations, a release of cortical granules was observed in the oocytes. Maturation promoting factor (MPF) activity declined to a low level after 6 h in all the oocytes injected with divalent cations. To test their developmental ability, injected oocytes were treated with cytochalasin B and then cultured for 168 h in NCSU23 medium. After 168 h, 29% (Sr2+), 29% (Ba2+) and 51% (Ca2+) of the oocytes had developed to the blastocyst stage. These results indicate that intracytoplasmic injection of Sr2+ and Ba2+, like that of Ca2+, induces in vitro matured pig oocytes to be released from MII-arrest and leads them into a series of events related to oocyte activation.

Our previous studies have shown that the addition of 100 μM cysteamine to the in vitro maturation (IVM) medium increased the embryo development of prepubertal goat oocytes. The aim of the present study was to evaluate the effect of adding different concentrations of cysteamine to the IVM medium and to the in vitro embryo culture medium (IVC) on the embryo development of prepubertal goat oocytes selected by the brilliant cresyl blue (BCB) test. Oocytes were exposed to BCB and classified as: oocytes with a blue cytoplasm or grown oocytes (BCB+) or oocytes without blue cytoplasm or growing oocytes (BCB−). In Experiment 1, oocytes were matured in a conventional IVM medium supplemented with 100 μM, 200 μM or 400 μM cysteamine. In Experiment 2, oocytes were matured with 400 μM cysteamine and following in vitro fertilization (IVF) were cultured in SOF medium supplemented with 50 μM and 100 μM cysteamine. In Experiment 1, BCB+ oocytes matured with 100 μM and 200 μM cysteamine showed higher normal fertilization and embryo development rates than BCB− oocytes. Oocytes matured with 400 μM cysteamine did not present these differences between BCB+ and BCB− oocytes. In Experiment 2, the addition of 50 μM and 100 μM cysteamine to culture medium did not affect the proportion of total embryos obtained from BCB+ oocytes (35.89% and 38.29%, respectively) but was significantly different in BCB− oocytes (34.23% and 29.04%, respectively, P<0.05). In conclusion, the addition of 400 μM cysteamine to the IVM improved normal fertilization and embryo development of BCB− oocytes at the same rates as those obtained from BCB+ oocytes. The proportions of morulae plus blastocyst development were not affected by the treatments.

It is still unclear whether nuclear envelope breakdown and premature chromosome condensation are essential for the reprogramming of the donor nucleus following somatic nuclear transfer. To address this, we determined the ability of delayed-activated or simultaneously activated porcine oocytes to undergo nuclear remodelling and development following somatic cell nuclear transfer. A small microtubule aster was observed in association with decondensed chromatin following nuclear transfer, suggesting the introduction of a somatic cell centrosome. In the delayed-activated condition, most fibroblast nuclei divided into two chromosome masses and two pronuclear-like structures following transfer into oocytes. In contrast, fibroblast nuclei in the simultaneously activated condition formed a large, swollen, pronuclear-like structure. Microtubule asters were organised in the vicinity of the nucleus regardless of the number of nuclei. More reconstructed oocytes developed to the blastocyst stage in the delayed-activated condition than in the simultaneously activated condition (p < 0.05). Nine piglets were born from two recipient sows following transfer of delayed-activated reconstructed oocytes, while none developed to full term in the simultaneously activated condition. Fingerprint analysis showed that the PCR-RFLP patterns of the nine offspring were identical to that of the donor pig. These results suggest that the activation of recipient oocytes during nuclear transfer probably relates to the nuclear remodelling process, which can affect the ability of embryos created by somatic cell nuclear transfer to develop.

In this study, the developmental capacity and cytogenetic composition of different oocyte activation protocols was evaluated following intracytoplasmic sperm injection (ICSI) of in vitro matured bovine oocytes. Motile spermatozoa selected by Percoll density gradient were treated with 5 mM dithiothreitol (DTT) and analysed for ultrastructural changes of the head using transmission electron microscopy (TEM). The alterations in sperm morphology after DTT treatment for different times (15, 30 and 60 min) were 10%, 45-55% and 70-85%, respectively. Further, a partial decondensation of sperm heads was observed after DTT treatment for 30 min. Oocytes were injected with sperm treated with DTT for 30 min. In group 1, sperm injection was performed without any activation stimulus to the oocytes. In group 2, sham injection without sperm was performed without activating the oocytes. Oocytes injected with sperm exposed to 5 μM ionomycin for 5 min (group 3), 5 μM ionomycin + 1.9 mM dimethylaminopurine (DMAP) for 3 h (group 4) and 5 µM ionomycin + 3 h culture in M199 + 1.9 mM DMAP (group 5) were also evaluated for cleavage, development and chromosomal abnormality. Cleavage and development rates in groups 1, 2 and 3 were significantly (p < 0.05) lower than those in groups 4 and 5. The incidence of chromosomal abnormality in the embryos treated directly with DMAP after ionomycin (group 4) was higher than in group 5. We conclude that immediate DMAP treatment after ionomycin exposure of oocytes results in arrest of release of the second polar body, and thus leads to changes in chromosomal pattern. Therefore, the time interval between ionomycin and DMAP plays a crucial role in bovine ICSI.

The onset of pronucleus formation and DNA synthesis in porcine oocytes following the injection of porcine or murine sperm was determined in order to obtain insights into species-specific paternal factors that contribute to fertilisation. Similar frequencies of oocytes with female pronuclei were observed after injection with porcine sperm or with murine sperm. In contrast, male pronuclei formed 8-9 h following the injection of porcine sperm, and 6-8 h following the injection of murine sperm. After pronucleus formation maternally derived microtubules were assembled and appeared to move both male and female pronuclei to the oocyte centre. A few porcine oocytes entered metaphase 22 h after the injection of murine sperm, but normal cell division was not observed. The mean time of onset of S-phase in male pronuclei was 9.7 h following porcine sperm injection and 7.4 h following mouse sperm injection. Ultrastructural observation revealed that male pronuclei derived from murine sperm in porcine oocytes are morphologically similar to normal male pronuclei in porcine zygotes. These results suggest that species-specific paternal factors influence the onset of pronucleus formation and DNA synthesis. However, normal nuclear cytoplasmic interactions were observed in porcine S-phase oocytes following murine sperm injection.

This study was conducted to evaluate the effects of insulin-like growth factor I (IGF-I) and other media factors during oocyte maturation, and the presence of different compositions of amino acids in embryo culture medium, on the development of equine embryos. Oocytes recovered from slaughterhouse-derived ovaries were matured in vitro for 24 h and those with a polar body were subjected to intracytoplasmic sperm injection (ICSI) or nuclear transfer with adult fibroblasts (NT). For ICSI embryos, there were no significant differences in rates of morphological cleavage, cleavage with normal nuclei or average nucleus number at 96 h post-ICSI between the absence and presence of IGF-I in maturation medium, or between embryos cultured in G1.2 or a modified CZB medium (CZB-C). Embryos produced by interspecies NT (equine donor cells into bovine cytoplasts) also showed no difference in cleavage rate or average nucleus number whether cultured in G1.2 or in CZB-C. The rates of cleavage, cleavage with normal nuclei and average nucleus number of equine NT embryos were not significantly different among oocytes matured in M199 with FSH in the presence or absence of IGF-I, or in EMMI medium, which contains IGF-I, epidermal growth factor, steroid hormones, FSH and LH. There were no differences in development of equine NT embryos cultured in any of three amino acid treatments (with or without non-essential amino acids, or containing taurine, hypotaurine and cysteine only). The cleavage rate and average nucleus number of parthenogenetically activated oocytes (treated similarly to NT oocytes but not enucleated or subjected to donor cell injection) were significantly (p < 0.05) higher than those for NT embryos. These results indicate that the presence of IGF-I or of EMMI medium during in vitro maturation of equine oocytes does not have a beneficial effect on their developmental competence as assessed at 96 h. Presence or absence of non-essential amino acids in embryo culture medium does not affect development of NT embryos within the first 96 h of culture. Factors associated with enucleation or nuclear transfer decrease the developmental competence of equine NT embryos. CZB-C medium may be used for culture of equine embryos with results similar to those obtained with G1.2 medium, thus providing a base medium that may be modified for further study of culture requirements of equine embryos.

Cloning efficiency following somatic cell nuclear transfer is very low. In order to obtain insights into this problem, mRNA expression patterns of early development-related genes in nuclear transferred embryos were compared with those obtained from in vivo and in vitro fertilization. Semiquantitative reverse-transcription polymerase chain reaction assay was used to compare the gene expression of, the cell adhesion protein E-cadherin, interleukin -6, heat-shock protein 70.1 and bos taurus apoptosis regulator box-a (Bax). The relative abundances of glucose transporter-1, E-cadherin and interleukin-6 were significantly (P<0.05) higher in in vitro fertilized morulae than in vivo derived morulae. Transcription of the gene encoding octamer-binding transcription factor 4 was higher in blastocysts obtained from in vivo fertilization than in those from in vivo blastocysts. The transcript for Bax was markedly upregulated in blastocysts derived from in vitro production and nuclear transfer procedures compared with in vivo fertilization. These results suggest that alterations in mRNA expression of early development genes are more associated with in vitro culture condition than the nuclear transfer procedure itself.

Through the injection of f-aequorin and the use of a photon imaging microscope, we have previously reported that a rhythmic series of intercellular Ca2+ waves circumnavigate zebrafish embryos over a 10 h period during gastrulation and axial segmentation. These waves first appear at about 65% epiboly and continue to arise every 5-10 min up to at least the 16-somite stage. In response to our publication, it was suggested that the waves may be an artefact caused by dechorionation of the embryos and would not be observed during the development of intact embryos (i.e. those with chorions). Here we demonstrate (again initially by aequorin imaging) that the rhythmic intercellular Ca2+ waves that traverse the blastoderm margin can also be observed in embryos that have an intact chorion. In addition, the appearance time, propagation pathway, velocity, duration and Ca2+ rise of the waves, as well as the interwave interval and the timing of wave onset, are approximately the same in both dechorionated embryos and those with an intact chorion. Furthermore, by loading intact embryos with Ca2+-green dextran at the single-cell stage and then using scanning confocal microscopy to obtain high-resolution images, we confirm the presence of circumferential Ca2+ waves and show that they pass through a population of deep cells located at the blastoderm margin. The confirmation of these pan-embryonic Ca2+ waves in zebrafish further corroborates our earlier suggestion that such waves might play a fundamental role in normal embryonic patterning during the gastrula period.

Apoptosis may occur in early embryos in which the execution of essential developmental events has failed. Thus the initiation of the apoptotic mechanism may be related to activation of the embryonic genome. In this way, developmentally incompetent cells or whole embryos are eliminated. It is likely that some link exists between failed resumption of rRNA synthesis and the incidence of apoptosis in cleaving embryos. In this context, decreased developmental potential in cleaving nucleotransferred embryos is consistent with cell loss, and very likely due to programmed cell death. The effects of apoptosis inducers on cleaving embryos have not been characterised in comparable detail to that in the case of somatic cells. Early embryos provide a very good model for study of these processes because of the specificity of rRNA transcription resumption after fertilization. In our experiments three apoptosis inducers (staurosporin 10 mM, actinomycin D 0.05 mg/ml and camptothecin 0.1 mg/ml) were used in a culture medium for 15 h at the 4-cell stage (day 2) of mouse embryos, followed by further development in a pure culture medium until fixation on days 3, 4 and 5. In staurosporin-induced embryos, light microscopy immunostaining of nucleolar proteins (fibrillarin, Nopp140, protein B23) did not reveal changes in nucleolar morphology on day 3. On days 4 and 5, more compact (roundish) nucleoli (in comparison with controls) were observed. The embryos treated with camptothecin displayed a similar staining pattern to those with staurosporin at each day. In actinomycin-D-treated embryos, marked changes in nucleolar appearance were visible as early as day 3. These changes in nucleolar morphology consisted of loss of the reticulation appearance and fragmentation of nucleoli. In addition to nucleolar changes, significantly decreased cell proliferation was observed. The induced embryos did not reach the blastocyst stage. The number of blastomeres was decreased, and staining with Hoechst 33342 revealed a significant percentage of apoptotic nuclei (condensed/fragmented nuclei) from day 4.