January 3, 2007
Dana-Farber scientists home in on key cancer signal
Jean Zhao and Tom Roberts led research on blocking a protein in a cell-signaling pathway.
For more than 20 years, Dana-Farber scientists have been looking for the best way to attack a complex cell-signaling pathway known as PI3K that often goes awry in breast, colon, and other cancers.
Now, Tom Roberts, PhD, Jean Zhao, PhD, and their colleagues in the Department of Cancer Biology appear to have found such a target. When they knocked out one form of a protein in the PI3K pathway in mouse cells, the cells became resistant to factors that normally would make them cancerous, they report.
"When you remove this subunit, p110-alpha, from the PI3K pathway, oncogenes quit working and growth factor receptors shut down, and almost nothing can transform these cells into cancer cells," explains Roberts. The p110-alpha molecule is an "isoform" one of several slightly different forms of the PI3K enzyme.
The findings have thrust p110-alpha into the spotlight, because the Dana-Farber experiments strongly suggest cancer therapy could be improved by blocking the p110-alpha molecule.While a p110-alpha inhibitor is likely to cause some significant side effects, they should be manageable, say Roberts and Zhao, an assistant professor at Dana-Farber and Harvard Medical School, the lead author on the paper published Oct. 31 in the Proceedings of the National Academy of Sciences.
"When Jean goes to a meeting, she attracts a crowd of people from the pharmaceutical companies," quips Roberts, the paper's senior author.
PI3K (phosphoinositide 3-kinase) is an enzyme which, when mistakenly activated by overactive growth signals, turns on a multitude of other genes and proteins that can trigger cancerous cell changes, including uncontrolled growth and escape from the mechanism that is supposed to destroy damaged cells.
Roberts, chair of Cancer Biology, and Lewis Cantley, PhD, of Beth Israel Deaconess Medical Center and the Dana-Farber/Harvard Cancer Center, discovered the PI3K enzyme and its role in cancer in the 1980s. Mutations that improperly turn on PI3K activity are present in about 32 percent of colon cancers, 30 percent of breast cancers, 27 percent of brain cancers, and 25 percent of stomach cancers.
Because cancer cells so often contain mutations affecting the PI3K signaling pathway, scientists at Dana-Farber and elsewhere have had high expectations of PI3K inhibitors in cancer therapy.
Drug companies have begun testing such drugs, but Roberts and Zhao believe that more selective inhibitors that would block only the p110-alpha subunit of PI3K, and not other proteins in the PI3K pathway, might have fewer serious side effects. Even with the selective knockout of p110-alpha in the mice, the two investigators found as a side effect that the cells' insulin response was inactivated. However, the possibility that patients taking p110-alpha inhibitors might develop diabetes "doesn't stop you dead in the clinic," Roberts says. "You can ameliorate that problem."
Previously, two inhibitors of PI3K were tested, but because they blocked several different enzymes related to PI3K, they were found too toxic to be drug candidates, Roberts says. He believes the new Dana-Farber findings should prompt companies to develop improved p110-alpha inhibitors.
– Richard Saltus
Richard_Saltus@dfci.harvard.edu
