LONDON The magic of the tooth fairy, who visits sleeping children during the night to transform the milk teeth under their pillows into coins, has entered the arena of developmental embryology. Researchers in London have shown that, by manipulating expression of the genes responsible for tooth development, it is possible to get a molar where once an incisor would have grown.

The observed transformation leads scientists into the realm of being able to bioengineer new teeth. The prospect looms of perhaps even being able to induce a replacement adult tooth to grow, by delivering a cocktail of cells and chemicals to the socket in the jaw from which a tooth has just been extracted.

Paul Sharpe, head of the department of craniofacial development at Guy¿s, King¿s and St Thomas¿ School of Dentistry, in London, told BioWorld International: ¿The dream of many people is to be in a position where we had enough information to start work on regenerating teeth in adults. Whenever I give a talk to dentists, I am asked, Will you get to the point where if we extract a molar, you can come along and put a magic potion into the cavity, and a new tooth will grow?¿ I used to be very skeptical, but I think you can no longer say that this will never happen although I¿d hate to say how long it might take to materialize. Certainly, this provides us with the first stepping stone for gaining a very detailed understanding of how teeth develop.¿

Abigail Tucker, Karen Matthews and Sharpe, all from the same institution, report their findings in the Nov. 6 issue of Science, in a paper titled ¿Transformation of tooth type induced by inhibition of BMP signalling.¿

The dentition of each species is as characteristic as its DNA, and the genetic control of tooth development is crucial to a species¿ survival. If animals are to evolve so that they can explore new feeding niches, they must develop new patterns of dentition. Much research has therefore been directed at understanding the key genes that may be involved in producing teeth of different shapes and in different positions in the jaw.

Some clues to the kinds of genes that would be involved came from studies on flies. These insects have homoeotic genes, which, when mutated, can cause a leg to develop in place of an antenna, for example. Vertebrates have similar genes, which have been called homoeobox genes. Their role is to determine position along the embryo, such as the head-tail axis, or the anterior of a limb from its posterior.

¿Each of this large family of homoeobox genes encodes transcription factors which regulate the expression of other genes,¿ Sharpe said. ¿We noticed a few years ago that some of the genes we were working on were expressed in very discrete domains in the developing jaw of the early embryo, long before you get any cell differentiation in the jaw. I therefore put forward an hypothesis that these different domains might control tooth identity.¿

He and his colleagues began to work on proving that this was the case. Using knockout mice, they first showed that when a homoeobox gene that they thought was involved in the development of molars was inactive, no maxillary molars formed. But the incisors of these animals were completely normal.

The researchers wanted to know, however, whether any of the products of these genes were sufficient, on their own, to make a tooth of a particular shape.

They focused first on the products of two homoeobox genes, called Msx-1 and Barx-1. Cells that express Msx-1 are destined to become incisors, while those that express Barx-1 become molars. Barx-1 is never expressed in cells that become incisors.

Msx-1 is activated by a molecule called BMP-4, which is produced by cells of the early oral epithelium in the developing jaw in the mouse embryo. Expression of Barx-1 is switched on by a protein called Fgf-8, whereas BMP-4 inhibits expression of Barx-1.

¿We decided to find out what would happen if we mis-expressed Barx-1 in incisor-forming cells,¿ Sharpe explained.

For their experiments, they worked on the precursor of the jaw from early mouse embryos. This can be removed from the embryo, cultured for three days in vitro as an ¿explant¿ and then, if necessary, inserted into the kidney capsule of an adult mouse, where after about 10 days teeth will form. ¿This is a powerful technique,¿ Sharpe said, ¿because the teeth are essentially the same as they would be if they developed in the embryo and it is very unusual to be able to see the outcome of your manipulations in developmental biology in this way.¿

The group took beads coated in a protein called Noggin, which inhibits BMP-4, and implanted them in the jaw explants at the point where BMP-4 is normally found. Sharpe said: ¿We were able to show that we had inhibited BMP-4 signaling, and that this allowed expression of Barx-1 to extend into regions where it is never normally expressed. The outcome of this was tooth transformation. We got molars instead of incisors. This was very nice because it supports everything we have been saying about these genes and what they do.¿

Harold Slavkin, director of the National Institute of Dental and Craniofacial Research at the National Institutes of Health, said the finding ¿is important because it clearly demonstrates the feasibility of understanding morphogenesis at the molecular levels of biology. The implications are that one day it may be possible to design and fabricate complex biological forms such as tooth organs.¿

Next, Sharpe¿s group plans to find out whether mis-expression of Barx-1 is sufficient to produce the transformation from incisor to molar. ¿We know that, as well as mis-expression of this gene, we are also losing expression of Msx-1,¿ he said. ¿It may be that you need to do both in order to get the transformation.¿