BLOCKADE The proapoptotic protein Bak (green ribbon) binds to a hydrophobic cleft on Bcl-xL (space-filling model). The Abbott/Idun team used NMR to identify two small molecules (orange) that bind in this same cleft. These were then stitched together and optimized to give ABT-737.
	© NATURE 2005

Organic chemist Patrick Harran of the University of Texas Southwestern Medical Center comments that the development of ABT-737 "is one more step toward balance, where ultimately the clinical oncologist will be able to target both aberrant growth stimuli and the suppression of apoptosis," the two main strategies cancer cells use to proliferate.

Apoptosis is regulated by two opposing types of proteins: those that initiate cell suicide and those that inhibit it. In healthy cells, antiapoptotic proteins such as Bcl-x_L and Bcl-2 block apoptosis by binding to proapoptotic proteins, counteracting these proteins' death signals. This balance is disrupted in damaged cells, thereby triggering apoptosis. Cellular damage that characterizes cancer cells would typically trigger apoptosis, notes Stephen Fesik of Abbott. Many cancer cells avoid this fate by overproducing antiapoptotic proteins such as Bcl-2 and Bcl-x_L, he adds.

Developing small molecules that prevent these proteins from binding to (and thereby inactivating) their proapoptotic partners "is a sound strategy for cancer drug discovery," says cancer biologist Yi Sun of the University of Michigan, Ann Arbor. But it's proven difficult to find small molecules capable of disrupting those protein-protein interactions.

ABT-737 binds tightly to a hydrophobic cleft on Bcl-2 and Bcl-x_L, the same spot these proteins use to bind proapoptotic proteins. The research team--which included Idun's Tilman Oltersdorf and Abbott's Fesik, Steven W. Elmore, Alexander R. Shoemaker, and Saul H. Rosenberg-- first used nuclear magnetic resonance spectroscopy to screen a library of diverse small molecules to identify ones that would bind Bcl-x_L's hydrophobic cleft. They found two library members bound to two adjacent sites within the cleft, and using a structure-based approach and parallel synthesis, the team stitched them together. ABT-737 emerged after further optimization to reduce nonspecific protein binding and to enhance binding to Bcl-2.

ABT-737 binds to Bcl-2 and Bcl-x_L 100- to 1,000-fold more tightly than do previously reported compounds, Fesik notes. Administered alone, it kills lymphoma and small-cell lung cancer cell lines. When used with chemotherapy and radiation, it enhances the effects of treatment on many other cancer cell lines. In mice carrying human small-cell lung tumors, the compound caused tumor regressions and was relatively well tolerated, Fesik adds. If these promising results can be duplicated in human trials, "ABT-737 or its derivatives should stand out in the crowds of many other apoptosis-inducing agents for future development of a novel class of targeted cancer drug," Sun says. Abbott is actively pursuing both intravenous and oral formulations of ABT-737 and related compounds, Fesik notes.