D. Bourc'his *, D. Le Bourhis †,‡ , D. Patin *, A. Niveleau §, P. Comizzoli †, J.-P. Renard † and E. Viegas-Péquignot *^@

[*] Institut National de la Santé et de la Recherche Médicale U383, Hôpital Necker-Enfants Malades, 149 rue de Sèvres, 75743 Paris cedex 15, France
[†] Unité de Biologie du Développement, Institut Nationale de la Recherche Agronomique,78352 Jouy-en-Josas, France
[‡] Union Nationale des Coopératives Agricoles d'Elevage et d'Insémination Artificielle, Services Techniques, 93703 Maisons Alfort, France
[§] Université J. Fourier de Grenoble, 38706 La Tronche, France

^@ Correspondence: E. Viegas-Péquignot   Fax: + 33 1 47 83 32 06
E-mail: viegas@necker.fr

Full-term development has now been achieved in several mammalian species by transfer of somatic nuclei into enucleated oocytes [1, 2] . Although a high proportion of such reconstructed embryos can evolve until the blastocyst stage, only a few percent develop into live offspring, which often exhibit developmental abnormalities [3, 4] . Regulatory epigenetic markers such as DNA methylation are imposed on embryonic cells as normal development proceeds, creating differentiated cell states. Cloned embryos require the erasure of their somatic epigenetic markers so as to regain a totipotent state [5]. Here we report on differences in the dynamics of chromosome methylation between cloned and normal bovine embryos before implantation. We show that cloned embryos fail to reproduce distinguishable parental-chromosome methylation patterns after fusion and maintain their somatic pattern during subsequent stages, mainly by a highly reduced efficiency of the passive demethylation process. Surprisingly, chromosomes appear constantly undermethylated on euchromatin in morulae and blastocysts, while centromeric heterochromatin remains more methylated than that of normal
embryos. We propose that the abnormal time-dependent methylation events spanning the preimplantation development of clones may significantly interfere with the epigenetic reprogramming, contributing to the high incidence of physiological anomalies occurring later during pregnancy or after clone birth.

The success of cloning depends on the reprogramming of the donor nuclei that must assume a gene expression program typical of the zygote genome. In normal embryos, reprogramming is time limited, and the restructuring of both parental genomes is completed by the time the embryonic genome is activated [6]. The time lag between somatic nuclear reprogramming and normal gametic reprogramming in oocyte cytoplasm may be involved in cloning inefficiency. Because in normal development both gametes are transcriptionally silent at the time of fertilization [7], activation of the embryonic genome requires extensive chromatin remodeling. The resetting of adequate epigenetic markers involved in chromatin organization is essential for cloning efficiency. Methylation changes are one aspect of nuclear reprogramming that may contribute with other events to restore totipotency.

DNA methylation is an epigenetic mark associated with compact and inactive chromatin structure [8]. The biological consequences of DNA methylation are mediated by methyl-CpG binding proteins, such as MeCP2, that recruit histone deacetylases. Silencing conferred by DNA methylation and MeCP2 may be released by inhibitors of histone deacetylases, thus creating a remodeled, transcriptionally active chromatin [9]. Mammalian somatic cells show elevated methylation levels. In comparison, gametes are less methylated, with sperm being more methylated than oocyte DNA [11]. In normally fertilized embryos, active [12] and passive [13] mechanisms erase most of the gametic methylation patterns before implantation, so that cells from blastocysts are globally undermethylated. New patterns of methylation are established after implantation, and defective patterns have been related to pathological situations [14].

Global genomic methylation changes can be followed on chromosomes by immunofluorescence with antibodies against 5-methylcytosine (m⁵C) [15]. Earlier work on normal preimplantation mouse embryos has shown that paternal chromosomes appear less methylated than maternal ones at the first mitosis after fertilization [13]. The asymmetrical staining of sister chromatids from the two-cell stage showed that passive demethylation of both parental genomes started at the first S phase of the zygote and proceeded on the whole embryo genome during cleavage divisions. In late preimplantation stages such as morulae and blastocysts, embryos are thus undermethylated [10]. We used this approach to evaluate the ability of bovine cloned and fertilization-derived embryos of the same genetic background to undergo these programmed methylation changes when developing under the same environmental conditions. Cloned embryos were obtained from the fusion of adult skin fibroblasts to enucleated metaphase II oocytes via a procedure proven to be compatible with the birth of normal offspring [3].

In Figure 1 are shown the methylation patterns observed in chromosomes and nuclei of donor fibroblasts. In fertilized bovine embryos, m⁵C antibodies generated the same patterns previously observed in mouse embryos (Figures 2a,b and 3) , suggesting that these patterns are common features of mammalian species. Cloned embryos presented several flagrant discrepancies in comparison with fertilized ones. The donor nucleus appeared as methylated as a somatic nucleus (Figure 2c,d) during the interval preceding the first S phase, which under our experimental conditions started after 7 hr postfusion (unpublished). In normally fertilized embryos, pronuclei appeared 7–8 hr after the exposure of oocyte to spermatozoa, and the first S phase started around 13–15 hr [16]. In clones, the moment somatic cells are fused to enucleated oocytes corresponds, in normal embryos, to the time pronuclei are formed. Thus, no evidence for an active demethylation process, as reported in normal mouse zygotes for the paternal genome before the first replication [12], was observed in bovine cloned embryos (Figure 2c,d). These results may indicate that the signal promoting the active demethylation may depend on a specific sperm nucleus rather than on an oocyte cytoplasm factor or, alternatively, that the somatic composition of the transplanted nucleus is resistant to this process.

Figure 1:

Fibroblast methylation pattern after immunofluorescence with m⁵C antibody. (a) Chromosomes from bovine fibroblast cells used to generate cloned embryos. Euchromatin is methylated and displays a banding-like pattern (large arrows). Centromeric heterochromatin is fully methylated (small arrows). (b) Nucleus from the donor somatic fibroblast cells. Arrow indicates the large spots of methylated heterochromatin.

Figure 2:

Methylation pattern after immunofluorescence with m⁵C antibody. (a,b) Normal cattle nuclei fixed at 8 and 14 hr, respectively, after in vitro fertilization. The formation of pronuclei occurs at 7–8 hr, and the first S phase starts around 13–15 hr postfertilization in bovine species [16]. Note the decreased staining of the decondensed male pronucleus at 14 hr postfertilization. (c,d) Nuclei of reconstructed bovine embryos, 5 and 7 hr after fusion, respectively. Pronuclei are not formed in bovine cloned embryos. The somatic methylation pattern of the somatic donor nucleus is globally maintained (see Figure 1a). Arrows indicate methylated heterochromatin. 

Figure 3:

Chromosome methylation patterns after immunofluorescence with m⁵C antibody in normal bovine preimplantation embryos. (a) Chromosomes from normal embryos at the one-cell stage (zygote). Paternally derived chromosomes are undermethylated (P), and maternally derived ones are methylated (M). There is a spatial separation of the two parental chromosome sets. (b) Chromosomes from normal embryos at the four-cell stage. Arrows indicate asymmetrically methylated chromosomes. Asymmetry indicates a passive demethylation mechanism. (c) Chromosomes from normal embryos at the blastocyst stage. Chromosomes are undermethylated on both chromatids (arrows). Note that centromeric heterochromatin is partially methylated.

In accordance with the absence of active demethylation, chromosomes of cloned embryos exhibited a somatic-like methylation pattern at the first metaphase (Figure 4a). There was no evidence of two differentially methylated parental chromosome sets, as seen in biparental embryos (Figure 3a). No topological separation of the homologous parental chromosomes, as described for normal embryos until the four-cell stage [17]; Figure 3a,b), was observed. The somatic-like profile was maintained after two cell divisions (two- and four-cell stages; Figure 4b). However, few chromosomes appeared asymmetrically methylated in less than 10% of the metaphases as early as the two-cell stage and without increasing frequency during embryo cleavage ( Figure 4b, inset). In normal embryos, 100% of asymmetrical chromosomes are observed at the two-cell stage. As cleavage progresses, the proportion of asymmetrical chromosomes halves and they are replaced by palely stained (undermethylated) symmetrical chromosomes. Chromosome asymmetry is a typical indication of a passive demethylation, i.e., a cell division-dependent demethylation resulting from a failure of  maintenance methylation [13].

Figure 4:

Chromosome methylation patterns in bovine cloned preimplantation embryos. (a) Chromosomes of reconstructed bovine embryos were obtained after the first post-fusion replication phase (equivalent to the one-cell stage). There is no evidence of two differentially methylated parental sets as observed in normal embryos (see Figure 3a). Chromosomes display a somatic-like methylation pattern in euchromatin and heterochromatin. (b) Chromosomes of reconstructed embryos were obtained after the third postfusion replication phase (equivalent to the four-cell stage). There is persistence of the somatic-like methylation patterns. In a few cells (inset) after the second and the third S phases, some asymmetrical chromosomes are observed (arrows). (c,d) Chromosomes from cloned embryos at the fourth and fifth postfusion stages (equivalent to the eight- and sixteen-cell stages), respectively. The first signs of an undermethylation of euchromatin appear at the 8-cell stage. Undermethylation persists until the blastocyst stage. (e) Chromosomes at the blastocyst stage (day 7 postfusion). Euchromatin is undermethylated, but centromeric heterochromatin tends to be more methylated than that of normal embryos (arrows).

Thus, the methylation patterns typical of biparental chromosomes are not reproduced, and passive demethylation proceeds with greatly reduced efficiency in cloned embryos. The somatic chromatin composition could play a significant role in these anomalies. Moreover, the fusion procedure potentially introduces the somatic form of Dnmt1, the maintenance mammalian methyltransferase, which is not normally present in preimplantation embryos. This form, contrary to the oocyte-specific variant (Dnmt1o) that is mainly retained in cytoplasm [18], could be operating to perpetuate the somatic-like methylation patterns in early cloned embryos. The sporadic occurrence of individual asymmetrical chromosomes in cloned embryos indicates only partial and rare passive demethylation.

A decrease in chromosome arm (euchromatin) staining of cloned embryos appeared as cleavage proceeded (Figure 4c–e). Under our experimental conditions, a first sign of a loss of euchromatin methylation was evidenced in some metaphases at the eight-cell stage (Figure 4c), this cellular heterogeneity being probably related to the asynchronous division of blastomeres from the third division cycle [19]. In contrast, heterochromatin remained stained. We note that these changes in euchromatin methylation coincide with a critical survival period for cloned embryos (our unpublished data); this survival period occurs between the four- and eight-cell stages, so that less than half of them (about 40% for bovine clones) will reach the blastocyst stage.

At morula and blastocyst stages, reconstructed embryos were systematically less methylated than the earliest stages, indicating that they adopted mainly in euchromatin an undermethylated pattern (Figure 4e) similar to that of normal embryos before implantation (Figure 3c). Centromeric heterochromatin, which is incompletely methylated in normal blastocysts, remained more methylated in cloned embryos, however (Figure 4e). The origin of this difference and its biological consequences need to be determined.

Because the passive demethylation process is poorly efficient in bovine cloned embryos, an active demethylating mechanism could act late in preimplantation development. Such a possibility was suggested earlier for normal mouse embryos between the 8- and 16-cell stages [20, 21] . Additional studies are required to determine the precise moment this mechanism could take place in clones and whether it occurs also in normal embryos.

The euchromatin demethylation we observed has some similarities to the previously reported delayed and incomplete nuclear reprogramming of reconstructed embryos [22, 23] . For instance, induction of telomerase activity, which normally occurs at the time of zygotic activation in fertilized bovine embryos (eight-/sixteen-cell stage), becomes apparent only by the blastocyst stage during postcloning development [24]. The apparent recovering of an undermethylated pattern in euchromatin is consistent with a removal of at least some somatic epigenetic markers in clones, as previously noted in mouse cloned embryos for the X inactivation process [25]. However, this late removal is apparently not sufficient to ensure successful pre- and postnatal development, as attested by the low cloning efficiency at birth (1%–3%), the perinatal death, and the high incidence of postnatal abnormal symptoms [4, 6] . Moreover, the persistence of a somatic pattern after fusion at a time when parental genomes are normally differentially methylated [12, 13] may have deleterious effects on the developmental potential of cloned embryos. There is significant evidence that inappropriate gene regulation during the period of nuclear reprogramming can have long-term detrimental effects [7]. The delayed or incomplete reprogramming of the somatic nuclei in oocyte cytoplasm may influence the normal course of events because a precise control over the time of embryonic genome activation is essential for normal embryogenesis

Although the euchromatic methylation pattern before implantation is similar in normal and cloned embryos, we cannot conclude that about the same degree of methylation occurs at individual sequences. As previously underlined, the chromosome methylation approach is efficient for topologically detecting clustered methylated CpGs, located either in highly repeated sequences (heterochromatin) or in interspersed repeated sequences enriched in methylated CpG sites [13, 15] . Methylation of individual CpG sites cannot be ascertained by this method. Two studies on cloned mice obtained from either cumulus or embryonic-stem (ES) cells indicated that cloned animals can exhibit methylation and hypomethylation at CpG islands of tissue-specific genes [26], as well as anomalies of imprinted gene methylation and expression [27], with important variations among individual clones. Compared to that of normal embryos, the genome of cloned embryos seems therefore to be a mixture of
normal and aberrantly methylated sequences. New information will be needed to clarify the causes and the extension of the observed variability.

The heavy methylation of centromeric heterochromatin in cloned embryos contrasts with the low methylation observed in normally fertilized embryo heterochromatin before implantation. This observation is in agreement with the relatively homogeneously methylated pattern of centromeric CpG-rich satellite I DNA recently described in cloned blastocysts derived from fetal bovine fibroblast cells [28] and analyzed by restriction digestion after bisulphite mutagenesis. For other repeated sequences, significant variations in the degree of methylation among individual cloned blastocysts were observed. Our approach, which allows the direct visualization of euchromatin and heterochromatin, the two major genome compartments clearly points out their different methylation levels. The persistence of a high methylation level in centromeric heterochromatin in cloned embryos may also be a source of disturbance of early embryonic activity because heterochromatin has been involved in gene silencing in mammals and other organisms [29, 30] .

In conclusion, the disturbance in methylation dynamics during the earliest stages of clones' embryogenesis, in which the embryonic genome is gradually prepared for activation, may thus be one of the factors contributing to persistent abnormalities that compromise survival or normal development of most cloned animals.

Material and Methods:

Production of Fertilized and Reconstructed Embryos:

Oocytes derived from abattoir-obtained ovaries were matured in vitro for 24 hr in TCM199 (Gibco) supplemented with 10% FCS, 10 µg/ml FSH, and 1 µg/ml LH. Fertilized embryos were obtained from metaphase II oocytes coincubated with frozen-thawed semen during 18 hr in Talp modified medium, and the presumptive zygotes were cultured in B2 medium seeded with VERO cells as described in Revel et al. [31]. As for cloned embryos, a single cell obtained from quiescent cultured fibroblasts derived from skin biopsies performed at the ears of adult donor cows was fused to enucleated metaphase II oocytes with a double electric pulse of 2.0 Kv/cm for 30 µs. Reconstructed embryos were activated with 10 µg/ml cycloheximide and 5 µg/ml cytochalasin B for 5 hr and then cultured in B2 medium.

Nuclei and chromosome preparation and immunofluorescence with antibodies against 5-methylcytosine (m⁵C):

Nuclei were analyzed at 8, 10, 12, and 14 hr after in vitro fertilization for normal embryos and after 1, 3, 5, and 7 hr postfusion for cloned embryos. Metaphases were obtained by the treatment of embryos with colchicine at a final concentration of 0.05 µg/ml. Spreading was performed according to modifications of the methods of King et al. [32] and Rougier et al. [13]. In brief, hypotonic treatment was achieved by incubating embryos in 1:6 diluted fetal calf serum and fixation was achieved by dropping 1:1 methanol/acetic acid fixative. Slides were then fixed in 3:1 methanol/acetic acid fixative for 1 hr and kept at -20°C after air drying.

The indirect immunofluorescence method was previously described [13, 15] . Slides were irradiated with UV light for 8–12 hr with a germicidal lamp. The m5C monoclonal antibody was used at 1:10 dilution in PBT (PBS, 0.1% Tween 20; 0.4% BSA) during 45 min at room temperature. After PBS washing, a second antibody (anti-mouse fluorescein-conjugated IgG) was added for 45 min. Slides were rinsed in PBS and examined under a Leica fluorescence/CCD microscope (Alcatel digital imaging system and Adobe Photoshop software). For each time point about 10–20 metaphases were analyzed.

We thank Dr. Timothy Bestor for critical comments on the manuscript. This work was supported by grants from the Ministère de l'Education Nationale, de la Recherche, et des Techniques, from the Association pour la Recherche contre le Cancer, the Ligue contre le Cancer, Comité de Paris, UE BIO4-CT98-0032 and FAIR CT98-4339.

[1] Campbell K.H.S., McWhir J., Ritchie W.A. and Wilmut I. (1996) Sheep cloned by nuclear transfer
from a cultured cell line. Nature, 380:64-66.

[2] Polejaeva I.A., Chen S.-H., Vaught T.D., Page R.L., Mullins J., Ball S., Dai Y., Boone J., Walker S. and Ayares D.L. et al. (2000) Cloned pigs produced by nuclear transfer from adult somatic cells. Nature, 407:86-90.

[3] Vignon X., Chesne P., Le Bourhis D., Flechon J.E., Heyman Y. and Renard J.P. (1998) Developmental potential of bovine embryos reconstructed from enucleated matured oocytes fused with cultured somatic cells.
C R Acad Sci III, 321:735-745.

[4] Renard J.P., Chastant S., Chesne P., Richard C., Marchal J. and Cordonnier N. et al. (1999) Lymphoid neoplasia and somatic cloning. Lancet, 353:1489-1491.

[5] Kikyo N. and Wolffe A.P. (2000) Reprogramming nuclei: insights from cloning, nuclear transfer and heterokaryons. J Cell Sci, 113:11-20.

[6] Solter D. (2000) Mammalian cloning: advances and limitations. Nature Reviews, 1:199-207.

[7] Latham K.E. (1999) Mechanisms and control of embryonic genome activation in mammalian embryos. Int Rev Cytol, 193:71-124.

[8] Bird A.P. and Wolffe A.P. (1999) Methylation-induced repression–belts, braces, and chromatin. Cell, 99:451-454.

[9] Jones P.L., Veenstra G.J., Wade P.A., Vermaak D., Kaas S.U., Lansberger N., Strouboulis J. and Wolffe A.P. (1998) Methylated DNA and MeCP2 recruit histone deacetylase to repress transcription. Nat Genet, 19:187-191.

[10] Monk M., Boubelik M. and Lehnert S. (1987) Temporal and regional changes in DNA methylation in the embryonic, extraembryonic and germ cell lineages during mouse embryo development. Development, 99:371-382.

[11] Howlett S.K. and Reik W. (1991) Methylation levels of maternal and paternal genomes during preimplantation development. Development, 113:119-127.

[12] Mayer W., Niveleau A., Walter J., Fundele R. and Haaf T. (2000) Demethylation of the zygotic paternal genome. Nature, 403:501-502.

[13] Rougier N., Bourc'his D., Molina Gomes D., Niveleau A., Plachot M., Pàldi A. and Viegas-Péquignot E. (1998) Chromosome methylation patterns during mammalian preimplantation development. Genes Dev, 12:2108-2113.

[14] Xu G.-L., Bestor T.H., Bourc'his D., Hsieh C.-L., Tommerup N. and Bugge M. et al. (1999) Chromosome instability and immunodeficiency syndrome caused by mutations in a DNA methyltransferase gene. Nature, 402:187-191.

[15] Miniou P., Jeanpierre M., Blanquet V., Sibella V., Bonneau D. and Herbelin C. et al. (1994) Abnormal methylation pattern in constitutive and facultative (X inactive chromosome) heterochromatin of ICF patients.
Hum Mol Genet, 3:2093-2102.

[16] Comizzoli P., Marquant-Le Guienne B., Heyman H. and Renard J.-P. (2000) Onset of the first S-phase is determined by a paternal effect during the G1-phase in bovine zygotes. Biol Reprod, 62:1677-1684.

[17] Mayer W., Smith A., Fundele R. and Haaf T. (2000) Spatial separation of parental genomes in preimplantation mouse embryos. J Cell Biol, 148:629-634.

[18] Carlson L.L., Page A.W. and Bestor T.H. (1992) Properties and localization of DNA methyltransferase in preimplantation mouse embryos: implications for genomic imprinting. Genes Dev, 6:2536-2541.

[19] Holm P., Shukri N.N., Vajta G., Booth P., Bendixen C. and Callesen H. (1998) Developmental kinetics of the first cell cycles of bovine in vitro produced embryos in relation to their in vitro viability and sex.
Theriogenology, 50:1285-1299.

[20] Shemer R., Kafri T., O'Connell A., Eisenberg S., Breslow J.L. and Razin A. (1991) Methylation changes in the apolipoprotein AI gene during embryonic development of the mouse. Proc Natl Acad Sci USA, 88:11300-11304.

[21] Kafri T., Gao X. and Razin A. (1993) Mechanistic aspects of genome-wide demethylation in the
preimplantation mouse embryo. Proc Natl Acad Sci USA, 90:10558-10662.

[22] Lavoir M.C., Kelk D., Rumph N., Barnes F.L., Betteridge K.J. and King W.A. (1997) Transcription and translation in bovine nuclear transfer embryos. Biol Reprod, 57:204-211.

[23] King W.A., Shepherd D.L., Plante L., Lavoir M.C., Looney C.R. and Barnes F.L. (1996) Nucleolar and mitochondrial morphology in bovine embryos reconstructed after nuclear transfer. Mol Reprod Dev, 44:499-506.

[24] Betts D.H., Bordignon V., Hill J.R., Winger Q., Westhusin M.E. and Smith L.C. et al. (2001)
Reprogramming of telomerase activity and rebuilding of telomere length in cloned cattle. Proc Natl Acad Sci USA, 98:1077-1082.

[25] Eggan K., Akutsu H., Hochedlinger K., Rideout W.I., Yanagimachi R. and Jaenisch R. (2000) X-chromosome inactivation in cloned mouse embryos. Science, 290:1578-1581.

[26] Ohgane J., Wakayama T., Kogo Y., Senda S., Hattori N. and Tanaka S. et al. (2001) DNA methylation variation in cloned mice. Genesis, 30:45-50.

[27] Humpherys D., Eggan K., Akutsu H., Hochedlinger K., Rideout W.M. III and Biniszkiewicz D. et al. (2001) Epigenetic instability in ES cells and cloned mice. Science, 293:95-97.

[28] Kang Y.-K., Koo D.-B., Park J.-S., Choi Y.-H., Lee K.-K. and Han Y.-M. (2001) Aberrant methylation of donor genome in cloned bovine embryos. Nature Genet, 28:173-177.

[29] Brown K.E., Guest S.S., Smale S.T., Hahm K., Merkenschlager M. and Fisher A.G. (1997) Association of transcriptionally silent genes with Ikaros complexes at centromeric heterochromatin. Cell, 91:845-854.

[30] Henikoff S. (2000) Heterochromatin function in complex genomes. Biochim Biophys Acta, 1470:01-08.

[31] Revel F., Mermillod P., Peynot N., Renard J.P. and Heyman Y. (1995) Low developmental capacities of in vitro matured and fertilized oocytes from calves compared with that of cows. J Reprod Fertil, 103:115-120.

[32] King W.A., Linares T., Gustavsson I. and Bane A. (1979) A method for preparation of chromosomes from bovine zygotes and blastocysts. Vet Sci Comm, 3:51-56.

1. Frenster JH,  "Activation of DNA Transcription Within Repressed Chromatin", 14th John Innes Symposium, Sept. 5-8, 2001.