Our Scientists

Ever since he was a child, David Agard has wanted to understand how things work. Today, as a biomedical scientist, he uses that curiosity to figure out at an atomic scale how "machines" of different sizes inside a cell function.

When Agard was a child, an uncle who was a physicist introduced him to the wonders of science and electronics. As he grew up, he initially thought that he too wanted to be a physicist. But ultimately, Agard was drawn to biology because of life's enormous complexity. "You could work a lifetime [in biology] and still be amazed at every new thing you saw," Agard said.

Nevertheless, his early exposure to inorganic systems influenced him. One of his goals as an investigator is to bring more physics and chemistry to bear on biology, with the idea of making biology, to some degree, an engineering science. Agard believes that by breaking components of a cell apart, examining the function of each of the pieces, and seeing how the parts come together at the molecular level, scientists will be able to predict how the cell's equipment works.

To that end, Agard for the past two decades has been deconstructing the mechanics of single macromolecules, small molecular complexes, and large supramolecular structures. To delve as deep as possible into his model systems, Agard has developed new types of biochemical analyses and visualization methods.

His philosophy as a scientist is to have a "big picture" approach. He searches for projects that might lead to a better understanding of general biochemical principles.

An appreciation for macromolecular structure began when he was an undergraduate at Yale and continued throughout his graduate studies at CalTech and postdoctoral studies at the MRC Laboratory of Molecular Biology in Cambridge, England. In the early 1980s, Agard felt that the time was right to combine the then new technology of molecular biology with structural biology to better understand the complex relationship between gene sequence, protein structure, and biological function at a molecular level. A gene codes for the linear sequence of amino acids that make up a protein. Proteins in a cell act as enzymes performing chemical reactions or as structural elements. When forming, a string of amino acids folds into a three-dimensional shape, which determines the molecule's function. Change part of the gene's code, and the resulting protein's form and activity may be impaired or altered in an interesting way.

In the early 1980s, when he came to the University of California, San Francisco, Agard particularly began to focus on alpha-lytic protease, an enzyme that breaks down nutrients for a soil bacterium. Agard chose the enzyme to study because its ability to survive outside of the microbe in the harsh and unpredictable environment of the soil might provide insights about other proteins. Ultimately, Agard found something unexpected about the enzyme: when it folded into its functional form it was unstable and not at its "lowest energy condition."

Up until that point, scientists thought proteins naturally fold into their lowest energy state. But Agard showed that during the protease's formation, it uses a molecular scaffold, called the pro region, to create a higher-energy, protein structure. Although thermodynamically unstable, the mature form of this protein is functionally stabilized by a large kinetic barrier that effectively blocks its unfolding. Simultaneously, this allows the mature form to be quite rigid, providing powerful protection from other bacterial enzymes and changes in soil chemistry. Since then, scientists have found that nature makes other thermodynamically unstable proteins for a variety of purposes.

Agard continues to study alpha-lytic protease and has expanded the complexity of his research to multiple-molecule interactions. He is particularly interested in the Hsp90 molecular chaperone protein because it binds to and facilitates the folding and function of hundreds, or possibly thousands, of proteins inside the cell. "It has to be doing something very interesting to affect all these proteins," Agard said.

Larger macromolecular complexes, such as centrosomes, organelles that help the cell divide, also fall under his scrutiny. To help see these and other structures, Agard and his long-term collaborator John Sedat have invented new microscopy techniques. One, deconvolution light microscopy, which images a specimen at different focal positions and uses computer modeling to remove the out-of-focus distortions, allows better visualization of small, three-dimensional structures than conventional light microscopy. Together with another colleague, Mats Gustafsson, they have developed a series of new light microcopy methods that obtain resolutions well beyond the classical diffraction limit of light. Additionally, Agard and Sedat have worked to automate and improve tomographic electron microscopy, which involves taking a series of two-dimensional pictures of a structure from all angles and then using software to reconstruct a three-dimensional image.

With these novel microscopy methods, Agard has been able to see organized structures in the centrosome in what was believed to be an amorphous area. The finding allowed him to do additional experiments that revealed how the centrosome organizes fibers in the cell called microtubules. Microtubules form and collapse as the cell divides. Agard's results have provided new insights into the process of microtubule assembly. His data also suggest that microtubule dynamics rely on instability and metastability akin to that of alpha-lytic protease. Agard hypothesizes that other complicated cellular assemblies may employ such thermodynamic tactics to form.

Agard likens his role as a scientist to that of a reverse engineer of nature. Much remains to be learned, he said, about the moving parts of life. Exposing the molecular mechanics of life's gears, shafts, and engines continues to motivate him to come to work every day.