As a postdoctoral associate at Harvard Medical School, Andrew Camilli needed a research topic. Inspired by ongoing research in the lab, he was struck by an idea. Could he build a molecular tool to systematically identify the deadly virulence proteins that cholera bacteria express as the infection proceeds?

Nine years later, that brief brainstorming session yielded a breakthrough method for analyzing diverse bacterial pathogens. Today, Camilli is doing research to understand how two such pathogens—Vibrio cholerae, which causes cholera, and Streptococcus pneumoniae, which causes pneumonia and middle-ear infection—become so virulent. His primary goal is to identify and characterize the function and regulation of specific genes and proteins that V. cholerae and S. pneumoniae express during different stages of infection. Ultimately, he hopes this work will lead to more effective vaccines and treatments.

Putting his postdoctoral ideas into action, Camilli has developed several tools to identify bacterial virulence factors and expressed genes in vivo—inside an infected host. His best-known invention is a genetic technique called recombination-based in vivo expression technology (RIVET), which marks bacterial genes that only activate during infection. Using RIVET, Camilli's lab has shown that V. cholerae expresses a critical virulence factor—intensifying cholera symptoms—only after an adhesive organelle on the surface of the bacteria has mediated attachment to tissues in the small intestine.

Camilli also became the first researcher to use DNA microarray technology to study the gene expression of a pathogen—again, V. cholerae—derived from naturally infected humans. He and colleagues compared gene expression levels and degree of infectivity of cholera bacteria grown in the lab with bacteria taken from stool specimens of cholera patients. They found that cholera bacteria shed from patients have a distinct gene expression pattern and have become orders of magnitude more infectious. Recent findings indicate that this enhanced infectivity may be mediated by regulation of bacterial chemotaxis.

In other work, Camilli and colleagues learned how a transcription factor dubbed VieA modulates expression of cholera toxin. They found that VieA has an independently functioning domain that hydrolyzes and thus lowers the concentration of a signaling molecule, cyclic diguanylate, which in turn activates V. cholerae virulence genes and simultaneously represses the bacteria's ability to form a biofilm for environmental survival. By engineering the bacteria to cause misregulation of cyclic diguanylate levels in the cytoplasm, they can reduce V. cholerae's virulence 20-fold.

Separately, his lab is analyzing the S. pneumoniae genes that allow this bacterium to stably colonize the nasopharynx and to infect the lungs, causing pneumonia. These and other findings are being applied toward the development of novel vaccines.