A new way to reset gene expression in cancer cells shows promise for leukemia treatment

At the center of this research are proteins called histones, which provide a physical support structure for the genome, and can also help regulate gene expression. Chemical modifications to histones can turn nearby genes on or off, and the cell interprets these chemical marks with the help of a variety of "reader" proteins. Once they recognize and bind to the chemically modified histones, the reader proteins recruit other factors that coordinate gene activation or inhibition.

This process can become derailed in cancer, and drugs that selectively inhibit a class of readers known as BET proteins have already shown early promise in treating certain tumors.

Now a multi-institutional research team -- led in part by David Allis, Joy and Jack Fishman Professor and head of the Laboratory of Chromatin Biology and Epigenetics -- has uncovered similar therapeutic potential for another, recently-identified class of reader proteins. These proteins share a structural feature called a YEATS domain, which specifically recognizes histones modified with a type of chemical mark called an acetyl group.

"The functional importance of this reading activity by the YEATS domain was unknown," says Liling Wan, a postdoctoral fellow in the Allis lab and lead author on the study, but she also notes strong evidence linking these proteins to cancer.

Dangerous fusion

The genomes of cancer cells can become extremely jumbled, with DNA strands breaking and reattaching to each other in unnatural ways that disrupt gene function. In some people with leukemia, cells feature genomic rearrangements that fuse the gene encoding a protein called MLL with that of various YEATS domain-containing proteins, including one known as ENL. Such rearrangements are linked with a particularly poor prognosis, and are especially common in extremely young patients.

"In infants, it occurs in more than 70 percent of acute lymphoid leukemia and in more than 35 percent of acute myeloid leukemia cases," says Wan.

When she and colleagues set out to explore whether ENL can spur cancer, they selectively deleted its gene from the genome of various leukemia cell lines. They found that mice transplanted with ENL-depleted leukemia cells fared much better than those receiving unmodified leukemia cells: their cancer cells did not divide as quickly, and they survived longer. These results made the researchers suspect ENL acts as an engine for tumor growth -- one that could potentially be stalled with a well-designed inhibitor.

The researchers also demonstrated that ENL activates cancer-related genes by binding to acetyl-modified histone sites throughout the genome. By engineering mutations in cells that render the YEATS domain dysfunctional, they were able to show this part of ENL is critical for the protein's leukemia-promoting effects.

New ideas for leukemia therapy

BET proteins, the class of reader proteins for which drugs are already being developed, don't have a YEATS domain, but they recognize histone acetyl marks via a structural element known as the bromodomain. When the researchers tested the effectiveness of a bromodomain inhibitor drug, JQ1, in mice with leukemia cells carrying YEATS-disrupting mutations in ENL, this two-pronged attack proved much more effective than the drug alone -- suggesting it might be possible to create effective combination therapy regimens.

"We look forward to the development of new drugs that can block YEATS domain function, perhaps working in concert with existing bromodomain inhibitors that are already in clinical trials," says Allis.

Wan says the multi-institutional team of American and Chinese scientists who conducted this study will be extending their collaboration in order to delve deeper into the role of ENL in cancer. Leukemia may not be the only class of malignancy powered by this protein, and Wan says there is evidence for YEATS mutations in certain pediatric kidney cancers.

"It seems to suggest a more broad function for the reader role of ENL in other cancer types that we have no knowledge about so far," she says, "and that's something we are very interested in investigating in the near future."