Why it's easy to twig

The significance of the title and first sentence of Marie Antoinette Di Berardino's new book Genomic Potential of Differentiated Cells, may be lost on some readers. But the genomic potential of our cells is, among a few other things, what makes us different from trees and is fundamental to the ethical debates raging about the practice of human cloning.

Cloning a tree is easy. Most of us by the age of five have planted a cutting under the watchful eyes of our parents or school teachers. Cloning a human or even a frog or goldfish is not so easy. The best cell and developmental biology took over a century to clone the celebrated Dolly the sheep - the first animal to be cloned from another adult's somatic cells. The reason that just about anyone can clone a tree is that tree cells (and those of most plants) have the property of being "totipotent". This simply means that they retain their potential to act like a fertilised egg might in an animal: any cell in a tree body can become the starting lineage of a new tree.

Animal cells, intriguingly, are different. Shortly after the zygote or single-celled stage of a developing animal, the cells begin to differentiate into cell types. Some will become backbones, some will become neurons, some will become teeth or eyes, or fingernails. This differentiation is associated with what has long been suspected to be irreversible changes to the cell's chromosomes. Differentiated cells lose the ability to behave like an embryonic cell and thus to be the zygote or starting cell to a new body. Once a cell starts down a particular developmental pathway its fate is more or less determined. Once a tooth always a tooth, or at least approximately so. Every now and then something goes wrong and tooth or eye cells will appear in the odd arm or kidney (and the owner will end up on one of those vulgar television programmes), but, happily, to a first approximation the "once-always" rule is correct.

Researchers like Di Berardino try to understand precisely what it is about the genetic changes that happen to a cell and cause it to lose its totipotency. It is not a simple matter of losing or altering genes; all of our highly differentiated cells carry the same genetic information. Instead, quite unbelievably precise and complicated instructions are acquired within cells at a supra-genetic level. These act to turn on or off particular sets of genes and at particular times. Knowing how cells accomplish this is the key to understanding how cells acquire a fate, and knowing this is the key to understanding development; how, that is, one gets eyes and ears and fingernails and teeth.

Ironically, developmental scientists have little interest in cloning per se. The opening lines of the Dolly cloning experiments article that got so many bishops, broadcasters, scientists, and self-appointed ethicists-to-humankind talking read "transfer of a single nucleus at a specific stage of development, to an enucleated unfertilised egg, provided an opportunity to investigate whether cellular differentiation to that stage involved irreversible genetic modification". Not really the stuff of Shelley. All those researchers wished to show was that they could take an "adult" fully differentiated somatic cell and make it behave like one from a zygote, because doing so demonstrates some level of mastery of the complex processes of differentiation. Dolly was the inevitable outcome of a successful experiment. Sadly for Di Berardino, the results of this experiment appeared just as her book was coming out, although this hardly detracts from its effectiveness. Large sections are given over to reporting the results of experiments on insects, amphibians, fish and mammals, that attempt to coax reluctant somatic cells to become totipotent again.

A century after August Weissman's pioneering work on the separation of germ and soma in the bodies of higher animals, the accumulated knowledge about tiny objects of mind-boggling complexity and potential leaves us poised on the brink of discovering the fountain of youth, for our somatic cells anyway. Regeneration of severed limbs and nerves, promotion of growth, and many other useful technologies may be in sight. Part biography, part scientific history, and part pure science, Di Berardino's book authoritatively brings to a close the first long chapter in the search to understand the processes of cellular differentiation and development.

John Gerhart and Marc Kirschner's Cells, Embryos and Evolution traces the details of how evolution by natural selection has produced the reliable and functioning package we call a body. Functioning bodies require cells that can differentiate into characteristic phenotypes such as eyes and teeth, but also cells that reliably interact with other cells to produce a stable, adapted, and adaptable organism. Their task is tantamount to writing a recipe for a successful body. The genetic and biochemical instructions alone are almost unimaginably complicated. No one has come close to producing such a recipe but Gerhart and Kirschner make it look possible.

Several general themes of development emerge. Many of the genetic programmes that specify the gross details of body plans in contemporary animals (including humans) were established 500 to 600 million years ago in our earliest multi-cellular ancestors. The received wisdom is that these genetic programmes have persisted because they got lucky: the bodies they inhabit just managed to avoid the mass extinctions that have regularly swept across the earth owing to various climatic and cosmological events. Gerhart and Kirschner sweep away this dogma, suggesting that these ancient programmes have been retained essentially unchanged for 600 million years because in addition to being efficient at specifying useful body plans, they are also evolvable; they have the capacity to generate new forms. The ability to introduce novelty in response to opportunities is a good long-term strategy. Simply being good at what you do in your current environment does not suffice if you want to be part of the great chain of being.

Contingency to a developmental geneticist refers to a cell's capacity to change its actions or functions in response to environmental demands. Skin cells produce additional melanin when exposed to the sun, and your two sets of leg-bone cells must divide the same number of times lest you end up lopsided. Armies execute contingency-based flexibility routinely by sending commands down along a well rehearsed chain of command. But the army analogy utterly fails when applied to bodies because bodies are paragons of decentralised control; almost as if the creator were a liberal democrat. This enables bodies to be flexible without long command chains, but places great responsibility on local controllers to read the environment properly and take appropriate action. Gerhart and Kirschner suggest a simple and effective way to make this happen: contingent cell functions should depend upon conditions as near the site of cell action as possible. This ensures that the acting cells are also the best informed about local environmental conditions. It is also why, remarkably, you can write your name on your arm using sunblock.

The most enduring metaphor of natural selection pits individuals against one another in the struggle for existence. Gerhart and Kirschner emphasise that it is also in the interest of individuals regularly to stage competitions within themselves. Early in development visual neurones migrate from centres near the eye into the brain. Many start the journey but only a few are eventually wired up. To sculpt organ systems such as the kidney, the body produces many more cells than are required for the completed organ. Cells then must compete to the death for a limited amount of "trophic" or growth factors. In both instances competition possibly ensures that only the fittest or best cells survive. These are examples of cellular selection.

The essential and vexing feature of the selection competitions is that what constitutes "best" performance can vary with circumstance: for example, an undernourished body may have different requirements from a well-nourished one. To be successful a body must produce a range of cells hoping that one or more will match the specific contingency: too little variability and the eventual winner may not be well suited to the task; too much and many of the contestants will be hopelessly outclassed and wasted. The authors speculate that evolution by natural selection may over many eons finely tune how much variability a given body produces for its internal competitions. Like the examples of evolvability in genetic programmes, this is tantalisingly close to "foresight" in an evolutionary context.

Gerhart and Kirschner have produced a serious and demanding work of remarkable breadth and detail, drawing on research in paleontology, embryology and cell and developmental biology. Like Di Berardino's book, Gerhart and Kirschner's signals the end of the first generation of developmental studies, now being replaced by a maturing discipline that is beginning to understand the rules by which bodies are constructed. Thankfully their book has some lighter moments. and here are some stunning colour plates. But the overriding impression to be gleaned from these two books is that even if Dolly the sheep had not been cloned she just might one day have been built from the ground up by researchers such as Di Berardino, Gerhart and Kirschner.

Mark Pagel is senior research fellow in the department of zoology, University of Oxford.