UCSD/SDSC Chemist Receives High Honor in Computational Science

Published 03/29/1995

J. Andrew McCammon of the University of California, San Diego (UCSD), and the San Diego Supercomputer Center (SDSC), a leading authority in the rational design of pharmaceuticals, has been named the winner of this year's Computerworld /Smithsonian Breakthrough Computational Science Award, sponsored by Cray Research, Inc..

McCammon holds the Joseph E. Mayer Chair in Theoretical Chemistry at UCSD together with an appointment in the Department of Pharmacology of the UCSD Medical School. He is also a Senior Fellow of the San Diego Supercomputer Center.

Considered one of the highest honors in computational science, the award is part of a series of Computerworld /Smithsonian prizes given annually to corporations and individuals for technical innovation and scientific achievement. The work of the computational science winner is exhibited in the Information Age display at the National Museum of American History of the Smithsonian Institution in Washington, D.C. The award will be presented to McCammon on June 5 at the seventh annual awards dinner in Washington.

"The winning work has to be innovative and carry the promise of major long-term benefits for humanity, and the nominee must demonstrate great persistence in the face of a very challenging problem," according to Robert H. Ewald, President of Cray Research, which sponsors the award. The award, whose full title is the Cray Research Information Technology Leadership Award for Breakthrough Computational Science, is given annually to the individual or team that has best used high-performance computational science to help improve the human condition, to solve or make progress on a previously intractable problem, and to create new tools to effect change.

The work for which McCammon is cited is a computational solution to a long-standing mystery: how do nerves and muscles work so fast? The "off" switch for the nerve-muscle pathways is an enzyme called acetylcholinesterase, abbreviated AChE, McCammon explained.

"It does its job by eliminating the neurotransmitter acetylcholine," he said. "The mystery was, how could AChE process so much acetylcholine in so short a time? The entire reaction takes less than 2 milliseconds, an astounding speed for a biological enzyme, but one obviously necessary and optimized by aeons of evolution to allow rapid fight or flight by organisms."

The crystal structure of the AChE enzyme was solved a few years ago in the laboratory of Joel Sussman and Israel Silman of the Weizmann Institute in Rehovot, Israel.

"But the structure did not by itself explain the speed of AChE," McCammon said. "The active site of the molecule appeared to be in a channel much too narrow to process acetylcholine and expel the products rapidly."

McCammon began work on the mystery with a group that included Michael Gilson (Center for Advanced Research in Biotechnology of the National Institute of Standards and Technology and the University of Maryland), Dan Ripoll and Carlos Faerman (Cornell Theory Center, Cornell University), Paul Axelsen (University of Pennsylvania), T.P. Straatsma (University of Houston), and Sussman and Silman.

The study, published in the March 4, 1994, issue of Science, exploited a molecular dynamics computer program developed in McCammon's lab to simulate the movements of the thousands of atoms of AChE.

What they discovered was that the atomic motion of AChE can widen the main channel to the active site, and even open extra channels to permit acetylcholine to enter and be broken down by the enzyme.

Since then, working with Gilson and with James Briggs, Jan Antosiewicz, and Stan Wlodek at UCSD, McCammon has studied exactly how the electric charges on the surface of AChE act to "steer" acetylcholine into the enzyme. This steering speeds the reaction by a factor of ten.

"Computation is essential in our research. We used the resources of all four NSF computational laboratories, at San Diego, Pittsburgh, the University of Illinois, and Cornell, in our own work. We are now using the massively parallel Intel Paragon machine at San Diego, which has 400 processors, to continue our work, as well as the Cray C90 machine, also at San Diego, and a group of Silicon Graphics machines at Illinois." McCammon's work is supported by grants from the National Science Foundation and the National Institutes of Health.

McCammon joined UCSD earlier this year from the University of Houston, where he was the M.D. Anderson Professor of Chemistry and director of the Institute for Molecular Design. "The opportunity to work closely with colleagues in pharmacology, notably Palmer Taylor, played a role in the lab's decision to move," McCammon said, "and we are particularly pleased to be near the San Diego Supercomputer Center, where we collaborate with colleagues in the Computational Center for Macromolecular Structure."

While at the University of Houston, McCammon invented what has become popularly known as "computational alchemy," a method to predict how large molecules recognize and bind to one another. With this method, researchers use a computer model of one drug-receptor complex and, by changing the atoms in key parts of the drug to produce a new drug-receptor complex, they calculate the relative binding strengths of the two complexes. The tighter the binding, the more effective the drug is likely to be.

The method is being used to predict how changes or mutations in some viruses and bacteria can affect a drug's ability to combine with and neutralize the virus or bacterium. It is mutations like these that have led to drug-resistant strains of pathogens, from influenza to the newly deadly strains of tuberculosis.

"With computational methods, we hope to keep up with the changes and redesign drugs to restore their effectiveness against the mutated strains," McCammon said.

One project in his own laboratory is directed, for example, at designing an inhibitor for an important enzyme belonging to the AIDS virus. Such inhibitors might eventually be an important part of anti-AIDS drugs.

The computational methods take advantage of the largest and fastest computers, like those at the San Diego Supercomputer Center, which enable the simulation of the motions of tens of thousands of atoms over time, picosecond by picosecond (a picosecond is one trillionth of a second). In essence, the computer is used to simulate molecules as they obey the fundamental laws of physics.

"We're trying to mimic the behavior of atoms and molecules in the computer," McCammon said. "We make what is essentially a mathematical model of a molecule, and then we translate the molecular dynamics into visual images that illustrate the behavior over time."

Born in Lafayette, Indiana, in 1947, McCammon received his B.A. in chemistry from Pomona College (where he was a "Times Scholar in Science," sponsored by the Los Angeles Times) and his M.A. and Ph.D. in physics and chemical physics from Harvard University.

While at Harvard, he worked with MIT professor John Deutch on biological applications of statistical mechanics and hydrodynamics. (Deutch, currently U.S. Deputy Secretary of Defense, is President Clinton's nominee to head the CIA.) Also at Harvard (1976-78), McCammon collaborated with noted chemist Martin Karplus in developing computer simulation approaches to protein dynamics. He joined the University of Houston in 1978 and was named to the Anderson Chair in 1981. McCammon has received numerous awards and honors and he is the author, with Stephen C. Harvey, of Dynamics of Proteins and Nucleic Acids (Cambridge University Press, 1987). In 1987, he was given the first annual George H. Hitchings Award for Innovative Methods in Drug Design, sponsored by the Burroughs Wellcome Fund. The award cited McCammon's achievements in developing new theoretical methods for computer-aided design of pharmaceuticals. He has also received the Camille and Henry Dreyfus Teacher-Scholar Award (1982-87), the NIH Research Career Development Award (1980-85), an Alfred P. Sloan Research Fellowship (1980-84), and a Woodrow Wilson Fellowship (1969).

The San Diego Supercomputer Center, a national laboratory for computational science and engineering, is sponsored by the National Science Foundation, administered by General Atomics, and affiliated with the University of California, San Diego. For additional information, call Ann Redelfs, 619-534-5032 (e-mail: redelfs@sdsc.edu).