Over 100 Years Later, Centriole Debate Settled with Potential Cancer Therapeutic

NIH/3T3 cells expressing a red (RFP) histone and a green (GFP) centriole marker. Image courtesy of Karen Oegema

A 108-year-old debate about centrioles may finally be settled. Normal cells have a pair of tiny barrel-shaped centrioles that are used for cell division and growing cilia. Cancer cells often have too many centrioles. Cell biologists have debated whether these centrioles are required for cell division since Theodor Boveri observed their role in cell division, which he published in 1907. Yet flies generated without centrioles in 2006 seem to settle the debate as they developed remarkably well, no centrioles required. New research suggests otherwise in human cells. It also suggests a new approach for protecting normal cells during cancer chemotherapy.

Andrew Shiau and Karen Oegema at the Ludwig Institute for Cancer Research and the University of California, San Diego are corresponding authors on a new study on centrioles published April 30 in Science Express. Their labs generated a new small molecule to remove centrioles from cells. “There had never been a non-genetic means to remove centrioles from cells for multiple cell cycles,” Oegema said. Their compound “centrinone” (a Plk4 inhibitor) successfully depleted centrioles from human cells, both normal and cancer-derived.

In contrast to the data in flies, normal human cells arrested in G1 when their centrosomes were removed and were unable to proliferate. The researchers looked to see if the same phenotype would be true for cancer cells, which they thought might be “addicted” to having too many centrioles.

The cancer cells turned out not to be “addicted” to centrioles. They were able to divide without them. And surprisingly, when centrinone was removed, the cancer cells went back to having the exact same number of centrioles as they did before the drug was added. “I thought that having too many centrioles was a hereditary thing,” Oegema explained, “that was propagated from cells to their progeny. This result told us that isn’t the reason. The centriole number distribution of a cell line is dynamically maintained and results from an equilibrium between ongoing centriole overduplication and removal from the population of cells with too many centrosomes.”

The pause in cell division in healthy cells, the researchers found, is due to increased p53, which is often mutated (therefore decreased) in cancer cells. When the team inactivated p53 in normal cells, they responded to the centrinone like cancer cells, dividing despite loss of their centrosomes. This explains why the cells respond differently to centrinone.

The differing effects of the drug on healthy versus cancer cells could be harnessed to protect healthy cells during cancer treatment, Shiau says. “The idea is that if you treat with centrinone, normal cells will stop dividing, and cancer cells will continue to divide, which could allow you to use another drug to kill the proliferating cancer cells,” Shiau said. This could minimize the toxicity of chemotherapy.

The next step in the project is to see what happens when you block Plk4 in disease models. “Centrinone and a related compound, centrinone B, have relatively poor pharmacokinetic profiles,” Shiau said. “We are currently working to make more drug-like versions of the compounds.”