Journal list menu

Volume 284, Issue 2 p. 211-217
Discovery-in-Context Review
Free Access

Nuclear transplantation, the conservation of the genome, and prospects for cell replacement

J. B. Gurdon

Corresponding Author

J. B. Gurdon

Wellcome Trust/Cancer Research UK, Gurdon Institute, Cambridge, UK


J. B. Gurdon, Wellcome Trust/Cancer Research UK, Gurdon Institute, Tennis Court Road, Cambridge CB2 1QN, UK

Tel: +44 (0)1223 334090

E-mail: [email protected]

Search for more papers by this author
First published: 14 December 2016
Citations: 6


Initial nuclear transplantation experiments in Xenopus eggs provided the first evidence for the conservation of the genome after cellular differentiation. This Discovery-in-Context Review recounts the early experiments that led to successful nuclear transfer in amphibians and the establishment of totipotency of a differentiated cell and shows how these discoveries paved the way for similar cloning experiments in other organisms.

When starting my graduate work in the 1950s, it was not known whether all different kinds of cells had the same set of genes in the same organism. It had indeed been suggested by Weismann [1] that genes no longer required in development might be lost or permanently inactivated. For example, skin genes might no longer be needed for the lineage which gives rise to the brain, and the permanent inactivation or loss of genes no longer required in development could help direct cell differentiation in desired directions. Rauber [2] published a paper describing the implantation of a frog nucleus into a toad egg and vice versa. He was curious to know what would happen in such an experiment, but he merely reported that development did not take place. It is unclear whether he actually did such an experiment or not. It was clear that the ideal experiment would be one in which the nucleus of a specialized cell is injected into an unfertilized egg whose own chromosomes had been removed. The clear question was whether the nucleus of a cell which had embarked on one pathway of differentiation could nevertheless support development of other, unrelated cell types. The first real success in transplanting living nuclei from one cell to another was achieved by Briggs and King in 1952 [3] (see also Table 1). They showed that the blastula nucleus of a Rana pipiens egg could be transplanted to the enucleated egg of the same species and, in a significant number of cases, they were able to obtain swimming tadpoles. They also reported, in the same paper, that if they took the nucleus from a more advanced embryo, for example, from a neurula embryo, the same experiment was not successful. They concluded, quite reasonably, that some change had occurred during early development such that the neurula nucleus was no longer able to substitute for the egg and sperm nuclei of a zygote.

Table 1. An abbreviated time-line for the original and early somatic cell nuclear transfer experiments. Details can be found in Ref. [16] and in many chapters in Ref. [17]
Author Date Experimental results Conclusion References
Spemann H 1938 Embryonic development and induction One cell can influence the differentiation of another [18]
Briggs and King 1952 Normal swimming tadpoles from transplanted blastula nuclei in Rana pipiens The nucleus of a blastula embryo is multipotent [3]
Gurdon JB 1962 Normal swimming tadpoles from intestinal epithelium cells of a feeding tadpole in Xenopus laevis The nucleus of a differentiated cell type is multipotent [19]
Gurdon and Uehlinger 1966 Fertile adult frogs from intestinal epithelium cells of a feeding tadpole in X. laevis The nucleus of a differentiated cell is totipotent [6]
Di Berardino and Hoffner 1970 Extensive chromosome damage can arise in somatic cell nuclei transplanted to amphibian eggs [7]
McGrath and Solter 1984 Mammal cloning is claimed to be ‘biologically impossible’ [10]
Willasden SM 1986 First account of successful nuclear transfer in sheep, using cell fusion and egg values The first successful nuclear transfer experiments in mammals [20]
Tsunoda et al. 1987 Full-term development of mouse embryo resulting from nuclear transfer using 4- and 8-cell donor embryos [21]
Prather et al. 1987 Full-term development of bovine nuclear transfer embryos [22]
Campbell and Wilmut 1996 Sheep cloned by nuclear transfer Successful cloning from nuclei of mammalian cells [8]
Wilmut et al. 1997 Dolly the sheep Successful cloning from nuclei of adult mammalian cells [9]
Wakayama et al. 1998 Full-term development of mice resulting from transplanted nuclei of terminally differentiated adult somatic cells First account of successful cloning from adult mouse cells [23]
Wakamatsu et al. 2000 Fertile adult fish from transplanted embryo nuclei [24]
Polezaeva et al. 2000 Nuclear transfer in pigs using cultured adult somatic cells [25]

When starting my graduate work, my supervisor, M. Fischberg, advised me to repeat the experiments of Briggs and King but to use the frog Xenopus laevis instead of R. pipiens. He took this view for two very good reasons. First, Xenopus embryos can be grown to sexual maturity in less than a year, whereas the same process takes up to four years in R. pipiens. Second, Xenopus frogs can be induced to lay eggs throughout the year, following hormone injection, whereas R. pipiens, like most European frogs, lay eggs only in the spring of each year. Experiments with Rana, therefore could not be done for 10 months out of each year. My view, at that time, was that if I could make nuclear transfer work in Xenopus, I might expect that either my results would reproduce those of Briggs and King, or that I might get more successful nuclear transfer embryo development. In either case there would be an opportunity to investigate the mechanisms that follow nuclear transfer and it was even possible that I might get more normal development than Briggs and King using the nuclei of differentiated cells.

There were two major obstacles in achieving nuclear transfer in Xenopus. The first was that Xenopus eggs are surrounded by an extremely elastic jelly present around the membrane that encloses the egg, and this viscous jelly turned out to be impenetrable by even the sharpest of microneedles (Fig. 1). The second problem was that the method of enucleation of the egg used by Briggs and King was not successful in Xenopus whose eggs suffered so much damage by cautery or physical attempts to remove the nucleus that they could not be used for such experiments. The solution to these problems emerged, more by good luck than judgment, within a year. First, an ultraviolet light used for microscopy turned out to be very effective at killing the egg mitotic chromosomes which, fortunately, were located on the surface of the egg. Ultraviolet light penetrated an egg only to a depth of about 30 μm, and did not therefore significantly harm a Xenopus egg with a diameter of 1200 μm. Had the chromosomes not been in this position they could not have been killed by this means. The second piece of exceptionally good fortune was that the wavelength of the ultraviolet light used not only killed the egg chromosomes, as might be expected, but also denatured the elastic properties of the jelly surrounding the egg so that, at the right dose, a needle could penetrate the egg without damaging the egg itself. Even then, I had to make a special microforge so as to put a sharp point on a micropipette that was also small enough to break a donor cell without damaging its nucleus (Fig. 2). In retrospect, this last piece of good fortune was not entirely surprising. After all, sperm enter the egg without causing damage.

Details are in the caption following the image
Even/not even a sharpened micropipette can puncture the highly elastic jelly that surrounds Xenopus eggs.
Details are in the caption following the image
A microforged injection pipette had to be sharp enough not to damage a recipient egg, but with a sufficiently smooth opening not to break a nucleus. It also had to be small enough to break the donor cell plasma membrane.

Another piece of exceptional good fortune, or wisdom on the part of my supervisor, resulted in a novel genetic marker being discovered for Xenopus. It happened, at that time, that he had a student, Sheila Smith, studying haploid development for which it was necessary to fertilize enucleated eggs and so to obtain haploid cells which were known to be characterized by a single nucleolus (which contains the ribosomal DNA-encoding genes) in each cell, whereas normal diploid embryos usually have two nucleoli representing the two sets of ribosomal genes, in each nucleus (Fig. 3). It was known that haploid embryos do not survive long enough to commence feeding and die as crippled embryos. The student concerned found that, unexpectedly, she could obtain normal embryos which had only one nucleolus and which were diploid. Fortunately, my supervisor did not follow the normal route, when the results are unexpected, which would have been to ask the student to remake all their solutions and start the experiment all over again with a new set of eggs. He asked the student to go through the whole Xenopus colony and see if she could find a female which had this ability to lay eggs whose derived embryos were diploid with only one nucleolus. Such an animal was, surprisingly, discovered and was the stock from which the well-known one-nucleolated diploid embryos (using the anucleolate mutant) were obtained [4]. Thus, it was possible to carry out a nuclear transfer experiment such that an unfertilized Xenopus egg could be harmlessly penetrated and, most importantly, a genetic marker could be used to demonstrate unequivocally that the egg chromosomes had been killed and that any resulting embryos did indeed result from the genetic material of the transplanted nucleus.

Details are in the caption following the image
Development of the nucleolate mutant in Xenopus. Heterozygotes develop entirely normally into fertile adult frogs. Homozygote mutants with ribosomal RNA genes die as swimming tadpoles before feeding; they develop this far by use of maternal ribosomes in the egg. Figure reproduced from Ref. [4].

It soon became apparent that, when I transplanted nuclei from the neurula stages of embryos as used by Briggs and King, I found no significant decrease in the normality of development [5]. This gave strong encouragement to continue these experiments using cells that were progressively more differentiated. I focused on the endoderm lineage of embryos because the cells of this cell-type are very large and easy to handle because of their large content of yolk. This made it possible for me to eventually transplant nuclei from larval intestine cells, derived from the endoderm, and to achieve the transplantation of a nucleus from a cell which had become committed to a particular type of cell differentiation. I had been able to carry out a large number of such experiments and the resulting embryos, from the intestinal epithelium, had begun to metamorphose into young frogs. At that point I had committed myself to a postdoctoral position in California, where I was asked to concentrate on bacteriophage genetics. I therefore left my young froglets in the hands of my supervisor Michael Fischberg who had just moved to Geneva as a Professor of Zoology. He asked his technician to look after the froglets and grow them up to adulthood. By the time I returned to Oxford from my postdoctoral period in California I went back immediately to the embryological work which had served me well. By this time the intestinal epithelium-derived frogs had reached sexual maturity and a considerable number of these frogs could be induced to lay eggs as females or fertilize eggs as males. In Oxford, I found that the intestine-derived frogs were indeed fertile and that their offspring showed no evidence of any defects beyond those which affect all laboratory-reared animals. This led to the end of that phase of my early career and I was able to publish a paper entitled, ‘Fertile-intestine nuclei’ [6]. This gave the first decisive evidence that the nucleus of a specialized cell is totipotent.

I should add that, as cells become increasingly differentiated, their transplanted nuclei become progressively less able to support normal development of enucleated eggs, as clearly shown by the Briggs and King experiments [3]. I found dramatic differences between nuclei from the same donor embryo in the serial nuclear transfer clones made from them (Fig. 4). This was probably due to damage (possibly random) sustained in the first nuclear transfers as shown by Di Berardino and Hoffner [7].

Details are in the caption following the image
Serial cloning of a transplanted endoderm nucleus yields dramatically different survival of embryos. Figure reproduced from Ref. [5].

At that time I was subject to considerable criticism. It was reasonably thought that a young graduate student was most unlikely to be able to overturn the conclusions from very well-established and highly respected workers in the field, namely Briggs and King. In all, it took about 10 years for my results to be generally accepted. Critical to the acceptance of these results was the benefit of using a nuclear marker, that is, the single nucleolus mutant discovered by my supervisor. This made it almost impossible to reject the results of my experiments. Therefore, the conclusion was eventually reached that as cells undergo progressive differentiation, their nuclei nevertheless retain the ability to bring about, in combination with unfertilized egg cytoplasm, almost all cell types which characterize an individual. This conclusion is now generally accepted for all animals and plants. However, it took nearly 40 years for the same conclusions to be reached with mammalian nuclear transfer [8, 9] (Table 1). The reason it took so long for the Xenopus experiments to be reproduced in mammals was, at least in part, due to the decision by those attempting nuclear transfer in mammals to use fertilized recipient eggs that were subsequently enucleated [10]. In Xenopus, and indeed in R. pipiens, unfertilized eggs had been used to receive transplanted nuclei. Once fertilization has taken place it is much harder to replace the zygote nucleus with a somatic cell nucleus. In the end, and as it now is, the original design of Amphibian experiments needed to be repeated exactly in mammals and, under these conditions, and as done in the sheep work referred to above, nearly all mammals can now be used for successful somatic cell nuclear transfer.

It is conceivable that the enucleation of Xenopus eggs by ultraviolet light gave them some advantage compared to enucleation of Rana eggs by microsurgery. Ultraviolet light destroys DNA but will leave most of the proteins which are associated with chromosomes, or present but unassociated with chromosomes in the nucleus, still intact. It has now been shown that there is a dispersal of some nuclear proteins into the cytoplasm at mitosis [11]. If such components are important in permitting successful somatic cell nuclear transfer, they might as well have been removed or destroyed by the physical removal of an egg nucleus or its chromosomes, whereas they might remain intact and available for reprogramming the nucleus after ultraviolet destruction of DNA.

Looking back, I was extremely fortunate to be supervised by Michael Fischberg (Fig. 5) and directed onto a really important problem very early in my career. He was extremely generous in giving me every facility to use in his laboratory. He was also generous enough to let me publish most of my work as a single author without attaching his name to the work that I did in his laboratory. Nowadays, he, in his capacity as Head of the research group, would certainly have expected to be a subsidiary author on all of my early Xenopus nuclear transfer work. I am, however, glad that he was first author on the initial paper describing success with nuclear transfer in Xenopus [12].

Details are in the caption following the image
Michael Fischberg.

One might ask, ‘What next?’ I have devoted the rest of my scientific career in attempts to understand the mechanism of nuclear reprogramming which takes place after nuclear transfer in Xenopus (e.g., Jullien et al. [13]). We would very much like to know what are the molecular events which permit this remarkable reprogramming of the somatic nucleus to take place efficiently and quickly. It was some half-century after these early Xenopus nuclear transfer experiments that Takahashi and Yamanaka [14] announced the amazing result that they could reprogram a small percentage of somatic cells back to an embryonic state by overexpression of selected transcription factors. This celebrated work has rightly captured the interest and enthusiasm of almost all laboratories and has led to the founding of numerous ‘regenerative medicine’ laboratories in many if not most countries. It would, however, be unrealistic to use nuclear transfer to eggs as a route toward regeneration of replacement cells in humans, because human eggs are hard to obtain in any number.

The success of the Takahashi/Yamanaka procedure is already beginning to reach the point where clinical benefit can be made available to humans, in particular with respect to the replacement of the pigmented epithelium of the eye for patients with age-related macular degeneration [15].

If we could understand the details of how the egg cytoplasm can so efficiently reprogram a somatic cell nucleus, this could greatly improve the success of reprogramming accessible somatic cells such as those of skin or blood to give large numbers of new cells of many different kinds. In this way, the reprogramming of somatic cells is likely to be of enormous benefit to humans for the alleviation of disease or aging.