Scientists are starting to solve one of the great mysteries of biology - how we get from an egg to an embryo, says John Maynard Smith
THERE is a revolution going on in our understanding of development, the process whereby an apparently formless egg turns into a structured adult.
Development has long been a puzzle. An adult animal is formed of many different kinds of cell - blood, bone, muscle, nerve and so on - arranged in a complex three-dimensional pattern, yet every cell contains the same set of genes. It has been clear for some time that, somehow, different genes must be switched on in different cells, but how?
Progress in answering this question is at last being made by using the techniques of molecular biology. It is now possible to identify genes that are active early in development and, for each gene, to find out when and where it is active, to "knock it out" and see what happens to the embryo, to cause it to be active in unusual locations, and even to transfer it to another kind of animal.
One remarkable experiment, by Walter Gehring and his colleagues in Switzerland, illustrates these possibilities. There is, in the mouse, a gene called "small eye". If this gene mutates, or is experimentally knocked out, the mouse develops without eyes. What this means is that the unmutated gene plays a role in normal eye development.
Gehring and his colleagues transferred this gene into the fruitfly Drosophila. The result was the growth of an eye on the fly's leg.
Of course, the eye that grew was not a mouse eye, but a compound fly eye. A single mouse gene could not tell the cells of a fly how to make a mouse eye that would require very many genes. But it does tell the developing embryo "make an eye here". Presumably, it does so by switching on other genes needed for eye development.
But why should a mouse gene work in a fly? The answer has to be that the common ancestor of mouse and fly had a gene that acted to locate some simple sense organ on its head, perhaps a group of light-sensitive cells, and that the gene has remained almost unchanged for more than 500 million years. It is turning out that many of the signalling genes that act early in development are astonishingly conserved throughout the animal kingdom.
One recently discovered gene of this kind provides a curious historical twist. In 1830 there was a famous debate at the French Academy between two distinguished anatomists, Cuvier and Geoffroy St Hilaire. The philosophically minded Geoffroy argued that all animals are built on a common plan; the more empirical Cuvier argued for the existence of four major groups, between which parallels cannot be drawn. It is today accepted that parallels exist between, for example, the fin of a fish, the arm of a man and the wing of a bird, and, less obviously, between a bone bracing the jaw articulation in fish, and a bone conducting sound in our ears.
But Geoffroy had difficulty comparing vertebrates, whose nerve cord runs along their backs, and the invertebrates, whose nerve cord is ventral, that is, towards the belly. He solved the problem by suggesting that vertebrates are simply invertebrates upside down. When I was a student, this story was told as an example of how mistaken scientists of the past could be, particularly if French. It now looks as if Geoffroy was right.
A gene that acts dorsally in vertebrates, determining which is back and which is belly, is similar to a gene that acts ventrally in insects. In evolution, it seems, our ancestors took to swimming upside down. The idea is not absurd; there are animals today, such as the back-swimming water boatman, that do the same.
I have spoken of genes being "switched on". This is a process that has been well understood for 30 years, since the work of two French biologists, Jacob and Monod, showed how genes could be switched on or off in bacteria.
But in the development of many celled organisms there is an additional problem: how are different genes switched on in different places? This question is now being answered. One example must suffice.
When the egg of a fruitfly is made in its mother's ovary, maternal cells introduce into one end of the egg the product of a particular gene, "bicoid". As a result, bicoid protein, produced at one end of the egg, diffuses backwards, setting up a concentration gradient. Other genes in the egg are switched on (or kept switched off) by the concentration of bicoid protein.
Many cases are now known in which the product of one gene acts to switch other genes on or off. The example of bicoid is unusual in that the source of the gradient lies outside the egg itself, in the ovary; usually such signals are generated within the embryo.
Thus a picture is emerging of development as a process in which cells become different because genes can regulate the activity of other genes. It is very much the kind of picture one would expect to emerge at a time when information technology is changing our lives; it is a picture of development being programmed by genes. Can this be the whole story?
There is an alternative view. Physicists and mathematicians, in particular, are fascinated by the complex patterns that can be generated by physical processes - whirlpools, snowflakes, waves - without the intervention of genes (see above). Surely, they argue, such dynamic processes must be relevant to development. Perhaps.
The difference between the head and tail end of a fruitfly may be determined by a simple concentration gradient, but no one imagines that the stripes on a zebra occur because a chemical gradient is set up from nose to tail, and that the cells respond locally to 50 different concentrations by making black pigment, and to intermediate values by not doing so.
Surely, we are looking at some kind of wave. There are plausible theories about what kind of wave it could be, but as yet no direct evidence.
At present, the revolution in our understanding of development is being driven by the notion of genetic control.
As a geneticist, it is something I have long expected, and that I welcome. But I think that before we fully understand development, we will also need ideas from the theory of complex dynamic systems.
