How Dolly Was Designed
(Something slightly overlooked in the frenzy of speculation about Dolly, the sheep cloned by Scottish scientists last week, are the details of the experiment that resulted in her creation. We asked Andrew Berry, a geneticist, to explain how Dolly came into being, and why, from a biologist's point of view, she matters.)
First, we have to look at what we mean by "cloning." We've been cloning genes for some 25 years now. In other words, we can propagate small strips of DNA outside their parent organisms. (We put a piece of, say, sheep DNA into a bacterial cell, which reproduces itself asexually--doubling its genetic material and splitting--and each time, it replicates our sheep gene.) We've been cloning cells for longer. Certain cells can be "cultured," reproducing themselves like bacteria in a petri dish. Both forms of cloning constitute major technological accomplishments, but neither comes close to reconstituting an organism. That is a much bigger problem.
Not an insurmountable one. As every gardener knows, you can regenerate entire plants from the smallest cutting. But animals are not so straightforward. The only kind of whole-organism cloning we've been able to perform up till now has involved DNA taken from early embryonic tissue. In the 1950s, biologists were able to generate entire frogs from DNA that came from embryos less than 80 hours old. Any older, and no new frog developed. That is because cells in very early tissue are what is known as undifferentiated--not yet committed to a particular developmental path. What's revolutionary about Dolly is that the DNA that created her was taken from a grown-up, differentiated cell.
u ntil now, the conventional wisdom in biology has had it that while early embryonic cells are "totipotent," capable of becoming any kind of tissue, once cells have differentiated into a particular tissue type they must remain that type. Pre-Dolly, differentiation in animal tissue was regarded as an irreversible process. If we cloned muscle cells, we got more muscle cells. Muscle cells were not going to generate liver cells.
How does differentiation work? We don't really know. We do know that every cell contains an identical and complete set of genetic instructions, the genome. As cells mature, however, they switch on some parts of those instructions, and switch off others. The cells in your muscle tissue switch on muscle genes. The cells in your liver switch on liver genes. But we can't explain this process--why some cells express some genes, and not others. The process of development (or differentiation) is one of the least understood in biology. At one end of the black box you put in a simple cell, the fertilized egg, and out of the other you get a sheep, an earthworm, a human--a mature animal of staggering structural complexity.
D olly is living proof that the process can go backward. Dolly is derived from a single mammary cell. Conventional wisdom would have dictated that mammary cells, which are highly differentiated and specialized, could produce only more mammary cells. But the mammary DNA that gave rise to Dolly became dedifferentiated, capable of generating the full range of different cell types that make up a sheep.
Given the high-tech world of modern biology, the method by which the Scottish team managed to provoke this extraordinary dedifferentiation is almost old-fashioned in its simplicity. In Dolly's case, they took a mammary cell and grew it on a petri dish in cell culture so that it produced more copies of itself. They then essentially starved the cells so that they shut down normal metabolic function, entering a "quiescent" state, in which dedifferentiation apparently occurs. The DNA from one of these cells was then transferred into an unfertilized sheep egg cell from which they had carefully removed the DNA. (Unfertilized eggs contain only the mother's half of genes; by adding the genes from an adult mammary cell--which has both its mother's and father's genes--the egg cell then had the full amount of DNA found in adult sheep cells.) The egg cell was biochemically pre-programmed to commence development, and that, apparently, was enough to jerk the mammary-cell DNA out of its quiescent state. The egg was put into a surrogate sheep mother and, 21 weeks later, we got Dolly.
One question that inevitably comes up is whether there is something peculiar about the way sheep mammary tissues differentiate. A second result in the Scottish study that has been largely ignored by the media--but which is, in biological terms, equally important--suggests that the answer is no. Biologist Ian Wilmut's team also cloned lambs out of DNA derived from sheep-fetus fibroblasts, cells found in connective tissue. Even in a fetus, a fibroblast is as highly specialized and fully differentiated as a mammary cell. Dolly is more of a media magnet than her unnamed fibroblast-derived cousins because she came from adult tissue. Biologically, though, the two experiments are comparable: In both cases differentiated DNA became sufficiently dedifferentiated to generate a whole new sheep. The result is, therefore, not mammary-specific, but more general.
Dolly and her fibroblast-derived cousins have changed forever the way we think about animal development. Cells in mature animal tissues are not, as we had thought, irreversibly differentiated. Understanding differentiation is the key to understanding development, and Dolly embodies the extraordinary possibility of manipulating the process--of doing experiments to identify the basic construction rules used in putting animals together.