Biophysics, Computational Biology
University of California, San Diego
Dr. McCammon is also Joseph E. Mayer Chair of Theoretical Chemistry at the University of California, San Diego, and Distinguished Professor of Pharmacology, at the UCSD School of Medicine.
Theory of Biomolecular and Cellular Activity
In 1970, J. Andrew McCammon was drafted for alternative military service. He had just earned a master's degree in physics from Harvard University and thought he'd spend the next two years "sweeping floors and things like that" at Massachusetts General Hospital. But when his supervisors realized they had a budding academic on their hands, they put him to work in a thyroid biochemistry lab.
There he performed routine hormone assays, but he also poked around and became involved in basic research with antibodies—the giant immune molecules that fight off bacteria and viruses. McCammon helped pen an article on the structure and function of antibodies, and his career in biology was launched. "I went back to graduate school with a heightened interest in trying to do something at the interface of theoretical physics on the one hand and structural biology on the other," says McCammon.
In the decades since, McCammon has done just that—pioneering the field of computational biophysics. Harnessing ever more powerful computers, he has developed myriad new ways to model how proteins—the workhorses of biology—wiggle, connect, and otherwise go about their life-making business. Besides fostering a deep understanding of the machinery of biology, McCammon's work has led to two important anti-HIV medicines and prompted the entire pharmaceutical industry to rethink how it invents new drugs.
Not bad for a would-be janitor.
Early in his career, McCammon stuck to pencil-and-paper theoretical chemistry. But in 1976, as a postdoctoral fellow, he teamed with Martin Karplus at Harvard to orchestrate the first computer simulation of protein motion. The pair chose a small target, the bovine pancreatic trypsin inhibitor. The protein itself held little intrinsic interest, but its modest size made it an ideal candidate for computer simulation. Still, it took the IBM supercomputers of the time several days to compute how the protein should move, bend, and twist. "It's probably fair to say that something that fits in your pocket today is much more powerful than those machines were," McCammon says.
That first simulation, like the many thousands that have followed, relied on sheer number crunching to render the motion of atoms as they're buffeted by basic electrical and Newtonian forces. "It's just like tracing out the orbits of planets around the sun, except instead you're tracing the paths of atoms within a large molecule," explains McCammon.
McCammon moved on to become a professor at the University of Houston and spent his time adapting his protein dynamics simulations to ever larger and larger molecules. Then the appearance of HIV and AIDS pushed him to think about applying this basic work to medicine.
"I wanted to study how an enzyme might change its shape a little bit when an inhibitor, maybe a drug molecule, binds to it, and whether that would be helpful in drug discovery," he says.
McCammon signed on as a consultant to Agouron Pharmaceuticals, a start-up launched by researchers from the University of California, San Diego. The company wanted to engage in rational drug design—that is, invent drugs that fit into a known target in a disease-related protein and hence block its action. "But they didn't know how to go about the computational part of that, so they recruited me to help set up that side of the company."
The company designed a molecule to block the action of HIV protease, an enzyme key to the survival of the virus. In 1997, the molecule became nelfinavir (Viracept), still an important drug in the fight against HIV. "When the protease inhibitors, including Viracept, came on stream, HIV for many people became a chronic disease, something they could expect to go on and live with for many decades," says McCammon.
Agouron was soon swallowed up by drug giant Pfizer, which expanded Agouron's facilities to house 1,000 researchers who use structure-based approaches to design drugs for cancer and other diseases. "Agouron was really the first company that rolled up its sleeves and said, 'We're going to try to create new drugs based on the structures of enzymes we want to inhibit,'" McCammon explains. "Now, the entire pharmaceutical industry designs drugs this way."
More recently, McCammon helped find a weakness in the third of HIV's three enzymes, integrase. Drugs against HIV's other two enzymes, protease and reverse transcriptase, revolutionized the care of HIV, but no one knew how to block the action of integrase, which slips HIV's genetic code into the genome of infected white blood cells. By using advanced computational techniques, McCammon discovered a hidden nook on the large, complicated integrase enzyme. "We discovered a binding site that no one expected to be there," he says.
That discovery led to the drug raltegravir (Isentress), the first HIV integrase inhibitor. "It's the first in a new class of drugs, and it's of huge importance for people resistant to existing HIV drugs," McCammon says.
With those two successes fueling his interest in drug design, McCammon is now working on African sleeping sickness and other tropical diseases. "We're pushing hard on the neglected diseases now," he says.
And, of course, McCammon continues to refine and expand his computational models of biological molecules. Computers have grown so powerful that modeling the behavior of even larger cellular structures—such as the synapses between brain cells—is within grasp.
McCammon even envisions the day he'll be able to simulate everything that goes on inside a cell. "It's far-fetched, but it is a goal," he says. "As Howard Hughes investigators, we're encouraged to look over the horizon. This may be way over the horizon, but we're headed in that direction."