Are there any normal cloned mammals?

Nature Medicine v.8, n.3, Mar02

IAN WILMUT

The finding that cloned mice, produced by transfer of nuclei from cumulus cells, develop obesity but do not

transmit the phenotype to their offspring provides further evidence that cloned embryos are vulnerable to epigenetic change. (pages 262-267)

An extraordinary amount is asked of a single cell when the nucleus is transferred from a differentiated adult cell into an enucleated oocyte, and the aim is to produce viable offspring. The nucleus of the donor cell is organized for tissue-specific function. This function is organized at several different, interacting levels, including the nuclear membrane, localization of specific chromosomal domains within active or inactive regions of the nucleus, assembly of protein complexes around the DNA, post-translational modification of those proteins and modification of the DNA itself [1]. It is assumed that a large part of this organization must be erased after nuclear transfer before remodeling of chromatin and nucleus to ensure normal development (Fig. 1). Exceptions include imprints created during gametogenesis that must be retained [2].

It is not clear whether the changes in nuclear organization must all take place during the first cell cycle or if they may continue in later cycles provided that inappropriate gene expression does not have harmful effects. The changes must be brought about by an oocyte that evolved to receive sperm chromosomes that are packed primarily in protamines, whereas chromosomes in the transferred nucleus are packed in somatic histones. Furthermore, there is no transcription in the sperm head whereas a transferred nucleus is active.

The published results of nuclear transfer are consistent with the judgment that the present methods for nuclear transfer are error-prone (http://www.roslin.ac.uk/pubic/-webtablesGR.pdf). The low overall efficiency is the cumulative result of unusual patterns of failure at all stages of development and after birth. Although the precise pattern of loss may vary between species and cell types, in general the outcome is similar. The greatest part of the loss occurs during very early cleavage, but death of clones occurs at all stages of development as a result of many different abnormalities [3]. Although the abnormalities could arise because of errors in just one of the mechanisms that regulate gene expression, it seems likely that several different mechanisms are involved, at least in some clones. In short, cloning by the present methods is a lottery, a stochastic process. Several coins are thrown and must all come -up as heads if normal life is to result.

With this understanding, it is questionable whether there are any clones that are entirely normal. If several mechanisms of regulating gene expression must be reset, might it be that, in clones produced by the present methods, this never happens perfectly for all genes? And that as a result there are some genes whose expression is not 'normal'? (That is, whose expression is outside the range observed in animals produced by sexual reproduction [4].) This is the context in which to consider the studies of obesity in cloned mice [5,6].

---

Fig. 1 (NOT SHOWN) Molecular mechanisms regulating development of cloned embryos. a, Normal ploidy is maintained by transfer of nuclei from donor cells awaiting DNA replication (GO/G1 phase). The recipient oocyte has a high level of M-phase promoting factor (MPF) activity that causes nuclear membrane breakdown (NEBD) and premature chromosome condensation (PCC). Activation of the oocyte induces decay of MPF activity and allows reformation of the nuclear membrane. After DNA replication, the cell divides to produce a 2-cell embryo. b, It is assumed that during the first cell cycles remodeling of nuclear and chromatin structure occurs, although there are very few direct observations of this process. Whereas genomic imprints established during gametogenesis are maintained, other aspects of organization of nucleus and chromatin are remodeled. These include assembly of a stage-specific histone structure, modification of the histones (for example, by methylation or acetylation) and changes in DNA methylation. In turn, these changes regulate access by specific factors that induce the onset of transcription from the embryonic genome during the 2-cell stage in mice.

---

Two reports by Tamashiro et al., one published a few years ago [5] and the other in this issue [6], indicate that cloned mice develop obesity. In the latter study, cloned mice, produced by transfer of nuclei from cumulus cells, were contrasted with two groups of control mice: stock mice and mice produced after subjecting zygotes and recipient females to many of the processes involved in cloning. Similar results were observed in two genetic backgrounds. Although birth-weight did not differ between groups, after 10 weeks of age the clones became heavier than both groups of controls. Clones ate the same amount as stock mice, but significantly less when considered in relation to body weight. The clones had higher levels of insulin and leptin and their increased body weight was directly attributable to increased accumulation of fat. These mice shared characteristics of human obesity absent from some other murine models. In particular, they were if anything more sensitive to the central melanocortin signaling system, rather than less. Their obesity must be due to errors in another mechanism.

As the obese phenotype was not transmitted to offspring of the cloned mice produced by natural mating, it probably resulted from epigenetic errors in-donor cells or errors that arose as a result of inadequate nuclear remodeling. Some of the changes described in cloned mice were also present to a lesser extent in the second control group of mice. In this group, zygotes were recovered and cultured in the same system as the cloned embryos, before transfer of a few embryos and ultimately delivery by Cesarean section. Such perturbations in development that result from manipulating embryos are well documented in sheep and cattle [7], but Tamashiro et al. provide the clearest set of observations in a different order of eutherian mammals. They provide further evidence that mammalian embryos are vulnerable to genetic change and suggest that clinicians should be cautious in the introduction of new procedures-such as transfer of oocyte cytoplasm-in human reproductive medicine.

Mice offer important opportunities to study the effects of cloning. They have a short generation interval are available in large numbers, and there is a great deal of background information about them. In this study, the authors have defined a distinct phenotype arising as a result of cloning. It seems that all of the clones became unusually large, although there was more variation among clones than stock mice. The comparative uniformity in phenotype is different from the outcome in most experiments. For example, in cattle-cloning experiments only some clones are unusually heavy [8]. The findings reported by Tamashiro et al. do not indicate that all clones are likely to be large. Rather, the study defines the outcome with one specific experiment and confirms that there may be epigenetic changes after cloning that are compatible with life. Many factors may be expected to influence the result of other cloning experiments, including species used, choice of donor cell, methods of nuclear transfer and enabling procedures, such as embryo culture. There is likely to be far less uniformity in experiments with out-bred livestock populations because of variation among animals in their oocytes and donor cells.

The question "Are there any normal cloned mammals?" as posed here cannot yet be answered meaningfully because it is not possible to measure expression of all genes in every tissue. However, it will serve a purpose if more groups are stimulated to document the health and physiology of their cloned animals, as was done by Tamashiro et al. and others [9]. It is important that studies continue through the expected life span of clones. Detailed observations of clones will assist in understanding and overcoming the limitations of the present procedures. They are also important before large-scale use of the technology in medicine or in farm animal production.

1. Thompson, E.M. Chromatin structure and gene expression in the preimplantation mammalian embryo. Reprod. Nutr. Dev. 36, 619-635 (1996).

2. Jaenisch, R. DNA methylation and imprinting: Why bother? Trend Genet. 13, 323-329 (1997). 3. McCreath, K.J. et al. Production of gene-targeted sheep by nuclear transfer from cultured somatic cells. Nature, 405, 1066-1069 (2000).

4. Jaenisch, R. & Wilmut, I. don't clone humans. Science 291, 2552 (2001).

5. Tamashiro, K.L.K. et al. Postnatal growth and behavioural development of mice cloned from adult cumulus cells. Biol. Reprod. 63, 328-334 (1999).

6. Tamashiro, K.L.K. et al. Cloned mice have an obese phenotype that is not transmitted to their offspring. Nature Med. 8, 262-267 (2002)

7. Young L.E. & Fairburn H.R. Improving the safety of embryo technologies: possible role of genomic imprinting. Theriogenology 53, 627-648 (2000).

8. Kato, Y., Tani, T. & Tsunoda, Y. Cloning of calves from various somatic cell types of male and female adult, newborn and fetal cows. 1. Reprod. Fert. 120, 231-237 (2000).

9. Lanza, R.P. et al. Cloned cattle can be healthy and normal. Science 294, 1893-1894 (2001).

Department of Gene Expression and Development
Roslin Institute Roslin, UK
Email: ian.wilmut@bbsrc.ac.uk