Mammalian preimplantation embryos normally develop within the protected environment of the female reproductive tract, which virtually precludes studies on embryogenesis in situ. Information must therefore be derived from experiments on cultured embryos. Consequently, studies on the epigenetic regulation of embryogenesis have long been interwoven with efforts to formulate culture media capable of sustaining normal development. In this review, comparative information on epigenetic regulation of embryo development is discussed, including information on energy substrate and amino acid preferences of embryos. Advantages of simple versus complex culture media, and of substituting serum albumin or synthetic macromolecules for serum, are discussed. Some potential pitfalls of co-culture are described. Culture appears to induce anomalies in embryo metabolism, which may derive from disturbed intracellular pH. Rationales for selecting endpoints to evaluate the outcome of experiments are considered, including incorporation of timing of embryo development into the analysis. Poor experimental design and/or data analysis can detract from or even negate the value of data obtained from embryo culture; examples are examined to help correct this problem. All of these points are discussed with a view to using data on the needs of embryos for making improvements in the design of culture media, so that higher yields and increased viability of embryos are achieved.

A major challenge for fluorescence imaging of living mammalian cells is maintaining viability following prolonged exposure to excitation illumination. We have monitored the dynamics of mitochondrial distribution in hamster embryos at frequent intervals over 24 h using two-photon microscopy (1,047 nm) while maintaining blastocyst, and even fetal, developmental competence. In contrast, confocal imaging for only 8 h inhibits development, even without fluorophore excitation. Photo-induced production of H 2 O 2 may account, in part, for this inhibition. Thus, twophoton microscopy, but not confocal microscopy, has permitted long-term fluorescence observations of the dynamics of three-dimensional cytoarchitecture in highly photosensitive specimens such as mammalian embryos.Keywords two-photon microscopy; laser scanning confocal microscopy; live cell fluorescence imaging; embryo; mitochondrial dynamics; mammal; hamster The detection of specific cellular components by imaging techniques such as wide-field epifluorescence or laser scanning confocal microscopy (LSCM) requires exposure to high intensity light that can cause cellular damage 1 . Consequently, the quantity or quality of images that can be collected is limited or, even worse, the reliability of the images may be compromised. This is a particular problem when imaging events that occur over periods of time ranging from hours to days, such as embryonic development. For this reason, much of our current understanding of subcellular morphological changes during mammalian embryonic development is based on images of fixed or static specimens at different developmental stages [2][3][4][5][6] . Thus, it can be difficult to interpret dynamic processes accurately, because the continuity of events must be inferred. The establishment of long-term fluorescence imaging methods that maintain the viability of live specimens is critical for advancing our understanding of cell biology and embryonic development in areas such as ion dynamics 7 , cytoplasmic reorganization, compaction and blastocoel formation, embryonic development in exotic species (where specimens are heterogeneous and difficult * Corresponding author (jsquirre@students.wisc.edu). to obtain), use of fluorescent tags for the preselection of embryos for subsequent embryo transfer 8 , and studies of protein expression in living cells using green fluorescent protein 9 . HHS Public AccessEmbryos of some mammals, particular hamsters, are very sensitive to culture conditions 10 . Furthermore, studies suggest that mammalian oocytes and embryos are adversely affected by exposure to visible light [11][12][13] . Because of this sensitivity, mammalian embryos are ideal to test live-cell imaging techniques. In addition, there are obvious morphological changes associated with differentiation, namely compaction and blastocoel formation, which can be used to assess viability. The embryo must undergo cell division during and after imaging as well as maintain a level of developmental competence that allows it to initiate dif...

Eight-cell embryos were recovered from mated golden hamsters that had been superovulated with pregnant mare's serum gonadotropin (PMSG) and human chorionic gonadotropin (hCG). Embryos were cultured for 24 or 32 h in a defined medium (modified Tyrode's solution) designed for fertilization of hamster oocytes in vitro. This medium was supplemented in some experiments with amino acids (glutamine, phenylalanine, methionine and isoleucine) and with vitamins (Eagle's Minimum Essential Medium vitamin supplement). At the end of the culture period, the numbers of embryos developing to the blastocyst stage were recorded. In other experiments, the effects of varying the osmotic pressure (225, 250, 275 and 300 m0smol/kg) and the pH (6.8 and 7.4) of the culture medium on blastocyst formation were examined. A difference was found between the ability of early 8-cell embryos (approx. 54 h post-egg activation) and late 8-cell embryos (approx. 62 h post-egg activation) to develop in culture. In the unsupplemented culture medium, only 2% of early 8-cell embryos developed to the blastocyst stage compared with 22% of late 8-cell embryos. A marked effect of the four amino acids on development was found. In the presence of amino acids 36% of early 8-cell embryos developed into blastocysts (18-fold increase). The amino acids also increased the percentage of late 8-cell embryos that developed into blastocysts from 22% to 66%. These data suggest that an important metabolic change may occur in hamster embryos during a critical period at the 8-cell stage of development. No additional effect on development was observed when vitamins were included in the culture medium. No significant effect of either osmotic pressure of pH of the culture medium on development was found. When blastocysts formed from cultured 8-cell embryos were transferred surgically to pseudopregnant hamsters, about 25% developed into normal-looking fetuses and 5 normal-looking young were born, 4 of which have survived. These results represent an approach towards achieving complete preimplantation development of hamster embryos in vitro.

The failure of hamster 2-cell embryos to develop in vitro (2-cell block) was examined with experiments in which concentrations of glucose and phosphate in the culture medium were varied. Embryos were cultured in a protein-free modified Tyrode's solution that normally contains 5.0 mM glucose and 0.35 mM sodium dihydrogen phosphate. In the presence of 0.35 mM phosphate but without glucose, 23% of 2-cell embryos reached the 4-cell stage or further after culture for 1 day and 27% after 2 days. Glucose inhibited embryo development even at 0.1 mM (4% development to greater than or equal to 4-cells after culture for 2 days); there was no dose-related inhibition above this glucose concentration. In a second experiment, phosphate levels were varied in the absence of glucose. Phosphate was highly inhibitory to development, with 97% of 2-cell embryos reaching the 4-cell stage or further after culture for 1 day in the absence of phosphate compared to 9-21% in the presence of 0.1-1.05 mM phosphate. After culture for 2 days, 26% of embryos reached the 8-cell stage or further when phosphate was absent compared to 0% development to 8-cells with 0.1 mM phosphate or higher. In a factorial experiment, phosphate blocked development when glucose was present or absent, whereas glucose did not block embryo development in the absence of phosphate. However, 2-deoxyglucose (a non-metabolizable analogue of glucose) inhibited embryo development in the absence of phosphate. These data show that the in vitro block to development of hamster 2-cell embryos is caused at least in part by glucose and/or phosphate. Deletion of these compounds from the culture medium eliminates the 2-cell block to development in virtually all embryos, and approximately 25-75% of embryos develop to the 8-cell or morula stages in vitro. The observations provide a possible explanation for the 2-cell and 4-cell blocks that occur in conventional culture media: stimulation of glycolysis by glucose and/or phosphate may result in inefficient adenosine triphosphate (ATP) production. The data indicate marked dissimilarities in the regulation of in vitro development of early cleavage stage hamster embryos compared with embryos of inbred mice, since the latter have an inactive glycolytic pathway prior to the 8-cell stage of development and will grow from 1-cell to blastocyst with both phosphate and glucose in the culture medium.