| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Reproductive Technology |
Division of Gene Expression and Development,4 Roslin Institute, Roslin EH25 9PS, United Kingdom
Department of Biomedical and Clinical Laboratory Sciences,5 University of Edinburgh, Edinburgh EH8 9XD, United Kingdom
Human Genetics Unit,6 MRC, Western General Hospital, Edinburgh EH4 2XU, United Kingdom
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
developmental biology, early development, embryo, gene regulation
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Prior to fertilization, the genomes of both sperm and metaphase II oocytes are transcriptionally inactive. Oocyte chromatin exists in a nucleosomal form with incorporated histones, while sperm DNA is highly compacted around protamines and is rapidly remodeled into a nucleosomal form postfertilization. In the mouse, rat, pig, and human, but not in sheep or rabbit, dramatic loss of DNA methylation has been reported in the sperm pronucleus shortly after fertilization [6–9]. However, the fate of the methylation status of the somatic nucleus in the first postfusion cell cycle, which has the nucleosomal chromatin structure of the oocyte but is transcriptionally active, is unclear. In early mouse cleavage stages, the genome-wide DNA methylation level further declines after fertilization to 30% of the typical somatic level and the embryo is largely demethylated by the time it reaches the blastocyst stage [8–13]. Genomic remethylation in the mouse embryo is apparent in the inner cell mass of the blastocyst, while the trophectoderm remains hypomethylated [14]. While the sheep embryo undergoes a much more limited global demethylation up to the eight-cell stage, the remethylation of the genome between eight cell and blastocyst observed upon purely visual inspection of sheep embryos (and probably bovine embryos; [9]) is in fact a visual artifact associated with decreasing nuclear intensity between progressive cell division [6]. Thus, there is no evidence of global remethylation of the normally fertilized sheep embryo either before or at the blastocyst stage.
Although mammalian cloning by somatic cell nuclear transfer (SCNT) has now been achieved in at least seven species [7], the overall efficiency is still low (typically 2– 10% development to term; reviewed in [15]). A genome-wide demethylation/remethylation event during early preimplantation development clearly provides an attractive mechanism for inducing reprogramming of somatic cells and is partly supported by recent studies in SCNT bovine embryos [9, 16]. Many reconstructed bovine embryos are more highly methylated compared with in vitro fertilized embryos [9, 16, 17], indicating inefficient reprogramming. However, bisulphite sequencing in SCNT pig embryos revealed efficient demethylation of all sequences analyzed [18], whereas the development to term of pigs produced by SCNT is no higher than that of the cow (reviewed in [15]). The current study was designed to investigate the dynamics of methylation reprogramming of fetal fibroblast donor cells after nuclear transfer in the sheep embryo and to provide further insight into the role of DNA methylation in determining the success of mammalian somatic cell nuclear transfer.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
All animal procedures were under strict accordance with UK Home Office regulations and within a project license issued under the Animal (Scientific Procedures) Act of 1986. For ovine in vitro fertilization, oocytes derived from abattoir-collected ovaries were matured and incubated with Texel ram sperm as previously described [19]. Fertilized oocytes recovered 24 h later were subsequently cultured in Synthetic Oviduct Fluid (SOF) supplemented with amino acids and bovine serum albumin (BSA) in 5% oxygen, 5% CO2, 39°C until the required developmental stages for fixation. For surgical recovery of in vivo-fertilized ovine oocyte and embryo stages, Scottish Blackface ewes were synchronized using progestagen-releasing intravaginal sponge inserted for 12–16 days. Five days before surgery, superovulation was induced by subcutaneous injections of 9 mg ovine FSH (oFSH) twice daily (0800 and 1700 h) over 4 days. A single intramuscular injection of 150 IU eCG was given with the second oFSH injection. Sponges were removed on the seventh oFSH injection and 2 ml Receptal (GnRH analogue; Intervet, Milton Keynes, UK) was administered intramuscularly 24 h later. Oocytes were recovered 26–29 h after Receptal injection by flushing the oviducts with PBS containing 1% fetal calf serum (FCS). Embryos were recovered from ewes naturally mated after Receptal injection at Day 2–7 postmating.
Cell Culture and Nuclear Transfer
Sheep fetal fibroblasts prepared from a Day 35 Black Welsh sheep fetus (Roslin line BLW1) were cultured for 5 days in serum-reduced medium (0.5% fetal calf serum) before use as donor karyoplast. Recipient oocytes were surgically retrieved from superovulated ewes 26–29 h after GnRH injection. Metaphase II-arrested oocytes were then washed through calcium-free HEPES-buffered SOF (HSOF) medium containing 10% FCS and, if required, cumulus cells were removed by gentle pipetting in calcium-free HSOF containing 300 IU hyaluronidase. Oocytes were enucleated in the presence of 7.5 µg/ml cytochalasin b and 5 µg/ml Hoechst 33342 in calcium-free HSOF to prevent spontaneous activation. Enucleated oocytes were held in bicarbonate-buffered SOFaaBSA at 39°C, 5% CO2 until reconstruction. Donor cell fusion and oocyte activation were achieved simultaneously with three 80-sec pulses of 1.25 kv/cm2 in fusion medium (0.3 m mannitol, 0.1 mm MgCl2 and 0.05 mm CaCl2). Pulsed embryos were held in SAFaaBSA containing 7.5 µg/ml cytochalasin b for 1 h at 39°C, 5% CO2 before assessment of fusion. Fused embryos were subsequently washed and cultured in SOFaaBSA as described above.
5-Methylcytosine Immunodetection in Embryos
Embryos were washed in phosphate-buffered saline (PBS) and fixed for 30 min at room temperature in 4% paraformaldehyde. To allow immunostaining of several developmental stages on the same day, some embryos were kept in 4% paraformaldehyde at 4°C for 1–7 days (control experiment showed no difference in staining within embryos from the same stage when processed within a 7-day interval). In all cases, embryos were washed extensively in PBS before further processing, then permeabilized with 0.5% Triton X-100 for 30 min at room temperature and treated with 4 N HCl for 1 h at 37°C. After several washes with 0.05% Tween-20, embryos were blocked for 1 h in PBS containing 2% BSA (PBS-BSA 2%). Methylated DNA was visualized with a mouse monoclonal antibody against 5-methylcytosine (5-mC; gifted from Alain Niveleau, Universite de Grenoble, La Tronche, France). Incubation with this antibody was performed at 37°C for 1 h (1:100 dilution in PBS-BSA 2%), followed by washes with 0.05% Tween-20 (30 min) and 1 h incubation at room temperature with a fluorscein isothiocyanate (FITC)-conjugated anti-mouse secondary antibody (1:200 dilution; Jackson ImmunoResearch, West Grove, PA). After several washes in 0.05% Tween-20, embryos were postfixed overnight at 4°C in 4% paraformaldehyde. For observation, chromatin was stained 30 min with propidium iodide (PI; 25 µg/ml) followed by a ribonuclease A treatment (1 mg/ml, 1 h at 37°C) and washes with 0.05% Tween-20. Embryos were mounted with Citifluor anti-fading agent (Citifluor Limited, Leicester, UK) on multiwell glass microscope slides.
5-Methylcytosine Immunodetection in Fibroblasts
The immunolabeling procedure was the same for sheep fibroblasts except that cells were grown on chamber slides to subconfluence before being fixed. In depletion experiments, 1–30 µg of 5-methylcytosine or cytosine (Sigma, St. Louis, MO) were mixed with 5 µl of the monoclonal antibody, diluted to 100 µl in PBS-BSA 2%, and incubated for 1 h at room temperature before being centrifuged for 15 min at 10 000 x g. The resulting supernatant was then used as above for immunofluorescence staining.
Confocal Microscopy
Observations were performed on an upright Optiphot-2 Nikon microscope (Nikon Corporation, Tokyo, Japan) equipped with the Bio-Rad MRC600 laser scanning confocal imaging software (Bio-Rad Laboratories, Hercules, CA) using a Nikon Plan Apo x60 oil immersion objective (NA = 1.4) and excitation wavelengths of 488 and 514 nm. For each wavelength, serial optical sections (Z-series) were collected at 1-µm intervals through the specimens, with Kalman averaging and slow scan speed. Collection of each color channel was done sequentially. For each experiment, the same gain, black-level, and aperture parameters were used. These Z-series were later merged with the ImageJ Software, v.1.30a (National Institues of Health, Bethesda, MD), to produce a two-dimensional image depicting the staining patterns and total intensities of all the nuclei present. All processed images were assembled and pseudocolored with Adobe Photoshop v 6.0 (Adobe Systems, San Jose, CA).
Quantitative Analysis
For quantitative measurements of the integrated fluorescence emitted by each nucleus, we used the merged images corrected for background by subtracting the mean intensity of the cytoplasmic area to the whole image. Nuclear intensities were measured by manually outlining all nuclei, except in blastocysts. The total fluorescence intensity emitted by each individual nucleus was measured using the SimplePCI imaging software (Compix Inc. Imaging Systems, Cranberry Township, PA) and averaged per embryo. For blastocysts, we analyzed 25 nuclei in each, randomly selected within the inner cell mass or trophectoderm cell populations based on the cellular morphology (small and compacted inner mass cells versus elongated trophectoderm cells). Student t-test was used to compare the values of different embryo stages and treatments.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Prior to using the 5-mC antibody on interphasic nuclei in sheep preimplantation embryos, we confirmed its specificity on cultured sheep fibroblasts. Visual observation of the pattern of immunolocalization of 5-methylcytosine compared with DNA (stained with PI) reveal that the signal could be subdivided into two components, larger foci that coincide with condensed chromatin of high PI fluorescence and a continuous labeling of the whole nuclear area (Fig. 1A). We verified the absence of light leak-through from one channel to another and that the methylation pattern did not depend on the fixation/denaturation protocol used by testing ethanol and paraformaldehyde fixations, as well as different HCl concentrations and incubation times for denaturation (data not shown). Antibody specificity was demonstrated by preincubation of the antibody with increasing concentrations of 5-methylcytosine, which led to complete disappearance of nuclear labeling, whereas high concentrations of unmethylated cytosine did not (Fig. 1B).
|
Heterogeneity in First-Cycle Methylation Status Is Recapitulated at the Two-Cell Stage
Because the sheep oocyte does not appear to demethylate the sperm-derived pronucleus after fertilization, we examined the effect of transferring ovine fetal fibroblast nuclei into ovine metaphase II-arrested oocytes. We first analyzed serum-starved fetal fibroblasts used in our standard SCNT procedure. These cells presented 27% lower levels of methylation (77.51 ± 23.90, n = 350; Fig. 2A) than in serum-fed cells (106.45 ± 12.06, n = 260; P < 0.001), although the global distribution pattern was similar and homogeneous between cells. When reconstructed embryos were fixed 4 h after fusion of the somatic cells with the recipient oocyte (n = 30), we observed three different types of methylation pattern (Fig. 2B). The first group (type 1) was very similar to the somatic cell pattern with intense methylated foci colocalized to heterochromatin (Fig. 2, A and B). The second group (type 2) presented a more homogeneous staining, covering all the nucleoplasm and more resembling the early in vivo embryonic pattern (Figs. 2B and 3). Furthermore, type 2 embryos exhibited 43% less methylation than type 1 and were 48% hypomethylated relative to the fibroblast donor cells. Type 1 and type 2 patterns were observed in 33% and 30% of the embryos, respectively. Finally, 37% of the embryos (type 3) showed no interphasic nucleus but compacted DNA remains. Type 3 embryos are unlikely to undergo any cleavage but instead arrest at the one-cell stage.
|
|
Indeed, by 2 days after SCNT, 31% of the reconstructed embryos had the same characteristics as type 3 embryos and did not cleave (n = 23 out of 75 Day 2 SCNT embryos analyzed from three independent replicates). The type 3 population probably corresponds with oocytes that did not activate after the somatic cell transfer and remained blocked in a metaphase II-like cell cycle state. We also observed another category of uncleaved embryos, accounting for 21% of the SCNT embryos analyzed on Day 2 after SCNT. These embryos had a nucleus that was still decondensed but appeared to be close to the oocyte membrane (type 4, Fig. 2B). Usually, these nuclei presented a methylation staining similar to type 1 and type 2 embryos, suggesting that these embryos did not develop beyond the activation process.
Out of the 48% of SCNT embryos that actually cleaved to the two-cell stage (Table 1), half presented a homogeneous methylation staining throughout both blastomere nuclei (similar to normal embryos; Figs. 3 and 4A), whereas the other half presented higher methylation levels with fibroblast-like intense methylated foci colocalized to heterochromatin (Fig. 3). It is therefore tempting to suggest that type 2 one-cell reconstructed embryos with a remodeled and partially demethylated somatic nucleus correspond with two-cell embryos with a methylation pattern similar to normal embryos and that type 1 reconstructed one-cell embryos (in which the somatic nucleus is not remodeled) correspond with the population of two-cell embryos with abnormal staining. Finally, it is of note that, in two-cell SCNT embryos with intense methylated foci (but not in in vivo two-cell embryos), we observed subnuclear compartmentalization of the foci (Fig. 3). Altogether, it seems that the somatic nucleus requires both successful activation and nuclear remodeling before the first mitosis to correctly reprogram development.
|
|
Abnormal Methylation and Nuclear Reorganization in SCNT Sheep Embryos
Because a role for methylation in early embryonic development might be reflected by changes in the distribution as well as in the overall levels of methylation, we examined and quantified all in vivo-derived and SCNT preimplantation stages from two-cell to blastocyst with the antibody described above and SimplePCI Imaging software. As shown in Figure 4A, the signal detected in the in vivo-derived embryo until the eight-cell stage is dispersed homogeneously throughout the entire blastomere nucleus, with no evidence of the asymmetrically demethylated paternal component described previously in the mouse [8, 20]. This observation is consistent with our observed lack of major paternal pronuclear demethylation during the first cell cycle in sheep [6]. Additional foci were detected at the eight-cell stage, clearly localized in regions of high chromatin density, where the methylation pattern reflects the distribution of DNA (Figs. 3 and 4A). In many two- to eight-cell SCNT embryos, the staining is characterized by the presence of larger and even more intensely methylated foci, also colocalized with condensed DNA (Figs. 3 and 4A). These foci are similar to the ones observed in somatic cells, suggesting that the somatic profile has not been reprogrammed into an embryonic profile where methylation pattern changes are linked to reorganization of DNA [6]. In comparison with in vivo-derived embryos, the majority of SCNT embryos present abnormally high levels of methylation (P < 0.01; Fig. 4, A and B), although the SCNT embryos do still follow the limited initial demethylation trend of in vivo-derived embryos (exhibiting 44% demethylation from the two- to the eight-cell stage; Fig. 4B). Thus, it seems that the minor demethylation event seen in cleavage-stage embryos derived both in vivo and SCNT occurs independently of chromatin reorganization and is likely a maternally programmed event.
Developmental Checkpoints at 8- to 16-Cell and Blastocyst Stages
Interestingly, the visual and quantitative differences between normal and SCNT embryos then temporarily disappear at the 16-cell and morula stages. During these stages of development, both SCNT and in vivo-derived embryos present a punctate pattern of methylation and have similar methylation intensities (Fig. 4, A and B). This suggests that only those reconstructed embryos that can sufficiently remodel their chromatin and epigenetic status by the time of embryonic genome activation (8- to 16-cell stage in sheep) can develop further. Increased nuclear compaction from the eight-cell to morula stages gives the false visual impression that methylation intensity increases over this period of normal development [6] and the same is true of our SCNT embryos (results not shown). Intriguingly, however, 48% of the SCNT embryos that develop to the blastocyst stage once again show quantitative methylation abnormalities (Table 1). The trophectoderm cells of in vivo-derived sheep embryos present 50% less methylation than the cells of the inner cell mass (ICM) (which maintain the same level of methylation as morula cells). While SCNT ICM cells present similar methylation intensities to those of morula and also to the ICM cells of normal embryos, around half of SCNT blastocysts are characterized by an absence of normal demethylation in the trophectoderm cells (Fig. 4, A and B). This may result in a further developmental checkpoint around the blastocyst stage.
Methylation Abnormalities in SCNT Embryos Concur with Low Cleavage Rate
We then further investigated any link between our observations of methylation defects with embryo developmental potential in culture. Cleavage rates were compared with embryos produced and cultured in vitro (IVF) at the same time as controls, based on our previous observation that IVF embryos exhibit similar methylation dynamics as embryos collected in vivo [6]. In terms of DNA methylation, we observed most heterogeneity between individual SCNT embryos in the earliest cleavage stages. Interestingly, most developmental arrests also occurred at these stages. At the second cleavage, for example, 40% of four-cell SCNT embryos present abnormal staining relative to fertilized embryos (data not shown) and only 68% of them survive this cleavage point in contrast with 92% of IVF embryos (Table 1). As development progressed, the heterogeneity in methylation defects in the population of SCNT embryos disappeared and the losses between the 16-cell and blastocyst stages were similar to the one observed in IVF embryos (Table 1). Thus, DNA methylation anomalies are strongly associated with developmental arrest in the early cleavage stages of ovine SCNT embryos. Because the anomalies reflect both abnormally high and misdistributed methylation, despite evidence of some demethylation, the possibility that abnormal chromatin distribution is the prime cause of developmental failure rather than lack of methylation reprogramming cannot be ruled out.
Finally, we examined SCNT embryos that arrested development between the two-cell and morula stages (n = 38). Very few had blastomeres of equal size or the expected one nucleus per cell. In fact, 66% of these embryos had very small cells without nuclei and 34% had cells that contained more than one nucleus per cell (Fig. 4C). In all cases, the embryos presented high methylation levels with punctate patterns. However, whether these embryos more likely stopped developing due to severe mitotic defects independent of DNA methylation abnormalities is unclear.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The mouse monoclonal antibody against 5-methylcytosine used in this study is sensitive to subtle changes in methylation status of both euchromatin and heterochromatin and allows quantification on individual nuclei [24–26]. After fusion with the recipient sheep oocyte, the condensed methylated foci of the fibroblast donor nucleus appear to either stay methylated or become partially demethylated. Demethylation may occur by the putative active demethylation mechanism observed in several mammalian zygotes soon after fertilization. This possibility is supported by our observations that the sheep oocyte is capable of partially demethylating murine and bovine sperm after interspecies intracytoplasmic sperm injection (unpublished results). At the two-cell stage, both nuclear types can still be distinguished; however, highly methylated embryos do not seem to survive the next cleavage divisions beyond the eight-cell stage. This suggests that the initial nuclear reorganization after fusion is crucial for further development and that an important step in reprogramming is remodeling of the somatic chromatin by factors present in the oocyte cytoplasm. In support of this, aberrant nuclear remodeling in mouse SCNT embryos has recently been described [27]. Intriguingly, the two-cell SCNT sheep embryos with highly methylated foci showed an asymmetric subnuclear localization of 5-mC. Asymmetric localization of normal mouse paternal chromosomes at the two-cell stage is detected by 5-mC immunostaining as a result of active male pronuclear demethylation [8, 20]. However, because the sheep male pronucleus does not demethylate and the sheep in vivo two-cell embryo does not show asymmetric 5-mC immunostaining [6], whether the SCNT subnuclear localization reflects parent-specific or sequence/heterochromatin-specific methylation is not known.
The question also remains whether the type 1 (methylated foci as in donor cells) or type 2 (diffuse pronuclear pattern) embryos we commonly observe result from heterogeneous reprogramming capacities in the oocytes (e.g., arising from the use of superovulated oocytes with heterogeneous methylating capacities; see [21]) or from differences between the somatic cells used for SCNT that are undetectable by 5-methylcytosine immunodetection. It is possible that our serum-starved donor cells were heterogeneous—i.e., even if most cells were in G0, some would probably be in G1-S or G2 phase [28]. However, we detected little heterogeneity in the methylation level per cell in our serum-starved population, with more than one quarter less methylation than in serum-fed cells. A recent study showed that, in quiescent cells, the steady level of DNA methytransferase (DNMT)1 falls rapidly [29], consistent with our observed demethylation. Because serum starvation is considered an important contributor to the success of somatic cell nuclear transfer in sheep [30, 31] and cattle [32], correlation with an epigenetic change in methylation is of interest. This raises the possibility that further experimental manipulation of methylation levels in donor cells can increase nuclear transfer success. In the cow, Kang et al. [17] reported no noticeable differences in the methylation-sensitive PCR fingerprints of growing or serum-starved bovine fibroblasts although only the Satellite I region was analyzed, whereas opposite observations resulted from total methylation assays [33]. Interestingly, Dean et al. [9] reported that nonstarved fetal fibroblasts were globally demethylated in cloned bovine zygotes, whereas Bourc'his et al. [16] did not observe any demethylation of quiescent somatic fibroblasts after introduction into bovine oocytes. Because our quiescent sheep fibroblasts were at least partially demethylated in 33% of sheep SCNT embryos 4 h after fusion, it is not yet clear whether procedural or species differences underlie the differing observations. Altogether, this underlines the importance of consistency in the choice of donor cells and the protocols used in methylation analysis of reconstructed embryos both within and between species.
Beyond the first cell cycle, the progressive loss of methylation during the first cleavage stages after fertilization described in the mouse and (to a lesser extent) bovine embryos was observed in both in vivo-derived and SCNT sheep embryos, with both types losing around 42% between the two- and eight-cell stages. Thus, although SCNT embryos tended to be hypermethylated relative to their in vivo-derived counterparts during this period (probably due to the hypermethylation of fetal fibroblasts relative to both sheep gametes [34]), the passive demethylation process occurs to the same extent as in normal development and thus seems unaffected by the putative introduction of different forms of DNMT with the somatic cell [23]. While insufficient demethylation of sequences silenced in the fibroblast but required in the early embryo may of course lead to disrupted gene expression and arrested/abnormal development, the failure of somatic chromatin to reorganize its heterochromatic component within the nucleus may actually be the cause of inadequate demethylation rather than lack of appropriate/sufficient epigenetic regulatory apparatus. This concurs with our observation that a large proportion of SCNT embryos have highly methylated foci at the two- and four-cell stages, akin to those typically observed in somatic cells. In normal embryos, they are only detected in embryonic blastomeres after the eight-cell stage, when nuclear reorganization accompanying zygotic genome activation occurs. Abnormal maintenance of heterochromatic foci after SCNT may prevent the activation of genes essential to regulate preimplantation development, consistent with the observation that 25% of the transcripts detected in mouse preimplantation stages are only ever expressed during this period of development [35].
The global genomic de novo methylation of the mouse genome beyond the 16-cell stage [10–14] is not apparent in either the in vivo-derived or the SCNT sheep embryo, at least until the blastocyst stage. However, sheep embryos do not normally implant until more than 14 days after they reach the blastocyst stage because the trophectoderm first elongates extensively [36]. Because we observed a demethylation event in the trophectoderm of the normal sheep blastocyst, it is still possible that de novo methylation occurs in the embryonic disc before implantation. Interestingly, although all SCNT embryos at the 16-cell and morula stages have similar levels and patterns of methylation as in vivo-derived, normal embryos, almost half of the SCNT embryos that survive to the blastocyst stage present abnormally methylated trophectoderm cells. Any abnormality in these cells may therefore lead to perturbation in placental development and implantation and provide a second key developmental checkpoint. That trophectoderm demethylation was usually deficient in our sheep SCNT blastocysts may account for our previous observation of 67% embryo loss of transferred blastocysts created with the same fibroblast cell line as in the present study by Day 35 of gestation [19]. Many fetuses surviving to Day 35 demonstrated lack of placental vasculature and reduced or absent cotyledons [19]. Because the trophectoderm contributes to the trophoblast and chorion components of the placenta and presumably signals to the inner cell mass-derived allantois, abnormal gene expression associated with perturbed methylation in this lineage may well contribute to the observed embryo loss and placental defects. In particular, release of trophoblast-specific factors such as the interferon tau protein family are critically required for maternal recognition of pregnancy and implantation (reviewed in [37]).
However, it is not clear if the hypermethylation of trophectoderm cells results from incomplete reprogramming at earlier stages or from perturbation of other developmental factors. Abnormal methylation reprogramming may indeed result in precocious transcriptional activation as already reported in SCNT bovine and mouse embryos [22, 38–40]. Differential methylation in imprinted genes could also be altered as described in mouse SCNT-derived placenta [22, 41]. However, immunodetection with antibodies against 5-methylcytosine cannot be directly related with the methylation status and expression of specific sequences.
We demonstrate that the high percentage of abnormally methylated embryos observed in early development of sheep SCNT embryos is consistent with the very high losses observed during the cleavage divisions, in comparison with IVF embryos. This suggests that the small proportion of SCNT embryos with normal levels and patterns of methylation observed at the two-cell stage correspond to the surviving 16-cell stage SCNT embryos, which all look similar to normal embryos. This may explain the surprising observation of consistently successful DNA methylation reprogramming in four-cell pig embryos [18]. The four-cell embryos examined may only reflect those that reprogrammed and survived to the stage of major genome activation in the pig (earlier stage embryos were not examined). Our hypothesis is supported by a recent study in mouse showing that the proportion of aberrant methylation in two-cell IVF embryos correlates with the number of embryos that do not develop in vitro to the blastocyst stage [20].
This study supports the hypothesis that methylation reprogramming is pivotal for successful and normal development after SCNT in mammals; however, the possibility that it may be secondary to and perhaps dependant on DNA reorganization within the somatic nucleus requires further investigation. Identification of epigenetic changes induced both in normal fertilized embryos and in the reprogrammed somatic cell nucleus after nuclear transfer is likely to define the blueprint epigenetic state of totipotent cells and an indication of how to achieve this state at the mechanistic level. Ultimately, this information may prove critical to developing novel means to transdifferentiate a somatic cell into a totipotent embryonic cell without the need for reprogramming in an oocyte cytoplasm. This study emphasizes the need to obtain epigenetic blueprints in the species of interest, and species differences in epigenetic remodeling at the blastocyst stage even during normal development may underlie the phenotypic differences also observed between embryonic stem cells from human and mouse. While DNA methylation appears to be a marker of reprogramming in all mammalian species examined to date, it is not yet clear to what extent it is a determinant.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
2 Correspondence and current address: Lorraine Young, Institute of Genetics and Division of Obstetrics and Gynaecology, University of Nottingham, Queens Medical Centre, Nottingham NG7 2UH, United Kingdom. FAX: 44 0115 970 9234; lorraine.young{at}notthingham.ac.uk
3 Current address: Biologie du Développement et Reproduction, INRA/ ENVA UMR 1198, Domaine de Vilvert, 78352 Jouy-en-Josas cedex, France
Received: 29 January 2004.
First decision: 9 February 2004.
Accepted: 1 March 2004.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  D J Kwon, C K Park, B K Yang, and H T Cheong Control of nuclear remodelling and subsequent in vitro development and methylation status of porcine nuclear transfer embryos Reproduction, May 1, 2008; 135(5): 649 - 656. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. M. Giraldo, D. A. Hylan, C. B. Ballard, M. N. Purpera, T. D. Vaught, J. W. Lynn, R. A. Godke, and K. R. Bondioli Effect of Epigenetic Modifications of Donor Somatic Cells on the Subsequent Chromatin Remodeling of Cloned Bovine Embryos Biol Reprod, May 1, 2008; 78(5): 832 - 840. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. H. Shi, J. S. Ai, Y. C. OuYang, J. C. Huang, Z. L. Lei, Q. Wang, S. Yin, Z. M. Han, Q. Y. Sun, and D. Y. Chen Trichostatin A and nuclear reprogramming of cloned rabbit embryos J Anim Sci, May 1, 2008; 86(5): 1106 - 1113. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Fulka, I. Barnetova, T. Mosko, and J. Fulka Epigenetic analysis of human spermatozoa after their injection into ovulated mouse oocytes Hum. Reprod., March 1, 2008; 23(3): 627 - 634. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Yamagata Capturing Epigenetic Dynamics During Pre-implantation Development Using Live Cell Imaging J. Biochem., March 1, 2008; 143(3): 279 - 286. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Niemann, X C. Tian, W A. King, and R. S F Lee Epigenetic reprogramming in embryonic and foetal development upon somatic cell nuclear transfer cloning Reproduction, February 1, 2008; 135(2): 151 - 163. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. Yang, R. Hao, B. Kessler, G. Brem, E. Wolf, and V. Zakhartchenko Rabbit somatic cell cloning: effects of donor cell type, histone acetylation status and chimeric embryo complementation Reproduction, January 1, 2007; 133(1): 219 - 230. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Yang, S. Yang, N. Beaujean, Y. Niu, X. He, Y. Xie, X. Tang, L. Wang, Q. Zhou, and W. Ji Epigenetic Marks in Cloned Rhesus Monkey Embryos: Comparison with Counterparts Produced In Vitro Biol Reprod, January 1, 2007; 76(1): 36 - 42. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Suteevun, R. Parnpai, S. L. Smith, C-C. Chang, S. Muenthaisong, and X. C. Tian Epigenetic characteristics of cloned and in vitro-fertilized swamp buffalo (Bubalus bubalis) embryos J Anim Sci, August 1, 2006; 84(8): 2065 - 2071. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N Cobbe Why the apparent haste to clone humans? J. Med. Ethics, May 1, 2006; 32(5): 298 - 302. [Full Text] [PDF]
|
|
|
|
|
|
R. C. Ribas, J. E. Taylor, C. McCorquodale, A. C. Mauricio, M. Sousa, and I. Wilmut Effect of Zona Pellucida Removal on DNA Methylation in Early Mouse Embryos Biol Reprod, February 1, 2006; 74(2): 307 - 313. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Gioia, B. Barboni, M. Turriani, G. Capacchietti, M. G. Pistilli, P. Berardinelli, and M. Mattioli The capability of reprogramming the male chromatin after fertilization is dependent on the quality of oocyte maturation Reproduction, July 1, 2005; 130(1): 29 - 39. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. D. Morgan, F. Santos, K. Green, W. Dean, and W. Reik Epigenetic reprogramming in mammals Hum. Mol. Genet., April 15, 2005; 14(suppl_1): R47 - R58. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Fulka, M. Mrazek, O. Tepla, and J. Fulka Jr DNA methylation pattern in human zygotes and developing embryos Reproduction, December 1, 2004; 128(6): 703 - 708. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||
