After a cross-country move with a new baby in one arm and a newly jealous dog in the other, Tim Mitchison is a busy man. Mitchison has left the University of California at San Francisco (UCSF), where he had worked for many years on how microtubules behave, make spindles, and move chromosomes, to be the codirector of a new institute attached to Harvard Medical School, the Institute for Chemistry and Cell Biology (ICCB). He leaves behind a tradition of raucous parties, great science, and trashy fishnet stockings on Halloween, and moves into an environment where he can fully apply the power of chemical synthesis to the solution of cell biological problems. If Bostonians are lucky, he will have also brought along the fishnets.

"I always knew I wanted to be a scientist, certainly by the age of six or seven," says Mitchison. "Science was definitely a family business." Mitchison's father, Avrion, was a noted immunologist who worked on transplantation immunity (which at the time was "totally opaque . . . absolutely mystifying"). His uncle Murdoch was a yeast physiologist who trained Paul Nurse and Kim Nasmyth, and his great-uncle was the evolutionary biologist J.B.S. Haldane.

Mitchison started at Oxford and then moved to UCSF for his Ph.D. He toyed with the idea of working on tumor virology with Michael Bishop, but a series of lectures by Marc Kirschner on timing and spatial organization in cells turned him on to microtubules. "The ideas were pretty mind-blowing," says Mitchison. "He was taking problems that epitomize biology, that make living systems alive, and saying that now the time was ripe to understand these at a molecular level."

The first step for Mitchison was purification of centrosomes, the eukaryotic structures that direct microtubule growth and form the poles of the mitotic spindle. While studying these isolated complexes, he noticed that reducing the concentration of tubulin, the subunit of microtubules, did not result in a uniform reduction in the rate of microtubule growth. Some microtubules continued to grow rapidly, while many more underwent a "catastrophic" conversion to a rapid shortening phase. The net result - the only parameter that other workers had looked at - was less polymerization. The pattern of alternate stable states of growing and shrinking, which Mitchison dubbed "microtubule jerking" but Kirschner saved with the name "dynamic instability," explains how microtubule arrays can be broken down and reformed. "It was very inspiring to feel that with this very simple system one could create what felt like a living process," says Mitchison.

When Mitchison first presented the idea of dynamic instability at a meeting, "the microtubule mafia were totally surprised and didn't really believe it," he says. Once verified, however, it became a staple in cell biology textbooks. "It allowed people to see how plastic the system was and how the cell could control microtubule arrays," says Conly Rieder of the Wadsworth Center in Albany, New York.

After brief interludes as a postdoc in the Kirschner lab, and as a research fellow at the National Institute for Medical Research in Mill Hill, London, Mitchison started as an assistant professor at UCSF. "The first version of the lab was an exciting, lively place," he says. "People were really sparring with each other to do their best work."

Mitchison had just shown that microtubule attachment to the chromosome (at a site called the kinetochore) is dynamic, with tubulin subunits constantly coming and going. Inspired by David Trenton, an organic chemist at Mill Hill, he made a "caged" fluorescein to further prove this point. When caged fluorescein linked to tubulin was added to cells, it was incorporated into microtubules but did not fluoresce. A bar of light on the spindle, however, knocked off a chemical group and gave a stripe across the microtubules that Mitchison could track as it moved steadily away from the kinetochore and toward the spindle pole.

Repeating this experiment with actin, a student then showed that actin remains fixed relative to the underlying substrate as cells move. Mitchison now uses the intracellular motility of the pathogenic bacterium Listeria as a model for eukaryotic cell motility, and has recently isolated proteins involved in this process.

Studies of chromosome movement began when a student first isolated eukaryotic chromosomes, then showed that they can move in opposite directions along microtubules depending on the amount of phosphorylation of the kinetochore. Others in the lab have used frog extracts to isolate proteins involved in spindle formation, the change in microtubule dynamics in mitosis, and even chromosome condensation. "Everything he touches he leaves a big impact on," says Rieder.

Mitchison is conscious of his shift from describing phenomena to identifying specific proteins that do the work. He is perhaps more keen on the "how" than the "what," but says he does "want to get to the level of molecular mechanism."

This is part of the reason for his shift to the ICCB. Once he has a protein, he wants to define its function by switching it on and off at will. "This is the major stumbling block in modern cell biology," he says. "The high road is biochemical reconstitution [combining all the necessary components in a test tube]. For Listeria and actin I think we can do that soon, but mitosis is too complicated at the moment. In the next few years we will probably have to be satisfied with reconstitution of simple subprocesses in mitosis." Genetics has its limitations, and methods such as antisense are far from perfect; the solution, says Mitchison, is specific chemical inhibitors.

"The overall goal of the [ICCB] is to find a routine way to make a small molecule ligand," says Rebecca Ward, the director of research affairs for the institute. "It's not so much a collaboration as it is team building. The entire institute is one big team working on one big problem."

Chemists will synthesize large libraries of compounds, and biologists will then assay for compounds that can inhibit specific proteins or cell biological processes. The chemistry will use natural products as inspiration: compounds will be large, rigid, and stereochemically complex. These properties, which will allow many functional groups to be presented to a target, increase the chances of finding inhibitors of protein-protein interactions. The libraries will be made by the split/pool method, which starts with many beads, each with a different chemical attached. The beads are split into pools, and each pool is reacted with a different monomer. Several rounds of mixing and reacting the beads can yield millions of different compounds. Each bead, however, has just one compound attached to it.

Of the approximately fifty members of the institute, perhaps ten will be devoted to synthesizing chemical monomers, tags, and linkers, and to decoding the structure of active compounds. Apart from a handful of independent fellows, other chemists and biologists will be affiliated with Stuart Schreiber, the other codirector; Matthew Shair, Eric Jacobsen, Greg Verdine (all from Harvard); Mitchison (whose primary lab adjoins the institute) or Marc Kirschner (Harvard Medical School).

Harvard Medical School has provided 10,000 square feet of space, start-up funds, and a new professorship for Mitchison. Discussions are ongoing with a number of large pharmaceutical companies for additional funding. The ICCB will allow Mitchison to fully merge his interests in chemistry and biology. "What I am bringing to [the ICCB]," he says, "is more sociological. I can talk to an organic chemist in their own language." Any active compounds will quickly be applied to biological studies. For example, an inhibitor of KCM, a kinesin that Mitchison suspects is involved in chromosome movement, could be added to cells just after they begin pulling chromosomes to either end of the spindle.

There have been distractions at the start: the unloading of moving trucks, the changing of his child's diapers, and perhaps the reconstruction of the "love shack" in the back garden, complete with makeshift hot tub (a bath with a Coleman stove underneath). But with chemistry at his disposal, Mitchison will be in a position to pull apart mitosis, movement, and more.