Unusual cancerous allies

Researchers have described an unusual mechanism at work in certain types of B-cell lymphomas that requires the joint activities of normal and mutant forms of an enzyme.

DNA wrapped around histone proteins
Image: Wikimedia commons, Zephyris

In a linkurl:paper;http://www.pnas.org/content/early/2010/11/08/1012525107.abstract published this week in the __Proceedings of the National Academy of Sciences__, a team of scientists from the pharmaceutical company Epizyme describes how in cells heterozygous for a mutation in EZH2—an enzyme that induces epigenetic changes in one of the histone proteins—the normal and mutant form of the enzyme work together to silence genes involved in suppressing tumors."It's a fascinating paper," said linkurl:Olivier Elemento,;http://physiology.med.cornell.edu/faculty/elemento/lab/ a cancer biologist at Weill Cornell Medical College in New York City who was not involved in the study. To his knowledge, it's the first time such a mechanism has been described in cancer. "It is definitely a very unusual situation where you have a mutated gene working in cooperation with a normal gene." EZH2 catalyzes the addition of one to three methyl groups to the lysine in position 27 of the H3 histone—a member of the family of proteins responsible for wrapping up DNA into chromatin. Earlier this year, researchers from the University of British Columbia and other institutions identified a recurring genetic mutation in two types of human lymphomas that arise from the antibody-producing B cells. They linkurl:reported;http://www.nature.com/ng/journal/v42/n2/abs/ng.518.html that these mutations affect a single amino acid of EZH2—a tyrosine in position 641—and seemed to impair the enzyme's ability to methylate the histone.But this presented linkurl:Robert Copeland,;http://www.epizyme.com/about-us/management.asp executive vice-president of R&D and chief scientific officer at Epizyme, with a conundrum. "That's contrary to everything we know about EZH2's role in cancer," he said. For one, the opposite scenario—overexpression of EZH2—is implicated in a variety of cancers, since its trimethylation of histone's lysine leads to the silencing of genes involved in tumor suppressionMoreover, the patients identified with these mutations were always heterozygous—meaning they carried both a mutant and a regular copy of EZH2. So it was unclear why a loss of function of EZH2 would result in cancer, particularly when the patients still had 50 percent of normal EZH2 activity from the gene that wasn't mutated."So we decided to look at those enzymes in much more biochemical detail," Copeland said.To do this, Copeland and his team at Epizyme compared how efficiently normal EZH2, as well as the four mutant versions identified in the human lymphomas, methylated short peptides they had synthesized to represent the region of the H3 histone where the lysine 27 is found. They found that the mutant versions performed poorly compared to wildtype EZH2. But when they repeated the experiment using natural histone wrapped in DNA that had been purified from chicken blood (in place of the short synthetic peptides) they found that both mutant and normal EZH2 were perfectly capable of catalyzing the methylation."Something was obviously different," said Copeland, who suspected the mutants were sensitive to methyl groups already attached to the naturally obtained histones. So the team retried the original experiment, this time using peptides with different methylation states.They found that while the EZH2 mutants were bad at attaching a methyl group to an unmethylated peptide, "if there was already one methyl group on the lysine, then the mutants were able to put a second and third one," Copeland explained.In fact, in cases when at least one methyl group was already present, the EZH2 mutants were much better at attaching the subsequent methyl groups than the wildtype EZH2 was. Based on the structures of the mutant enzymes, the researchers came up with an explanation for why the mutations resulted in their variable methylating abilities. These mutants had the tyrosine in position 641 replaced with a phenylalanine, histidine, asparagine, or a serine—all of which are smaller than the original amino acid.By replacing the tyrosine with a smaller amino acid, the mutant enzyme can now fit in a lysine that's already carrying one or two methyl groups with greater ease. At the same time, these mutations impair the mutant's ability to initially latch on to the unmodified lysine.But in cells expressing the wildtype EZH2, the normal enzyme takes care of adding that first methyl group, so cells end up with a greater number of histones with trimethylated lysines, and therefore a greater suppression of anti-proliferative genes. This ultimately translates to B cells dividing uncontrollably."Essentially you end up with the same result as if EZH2 is overexpressed," Elemento said.Because cells that contain these mutations are uniquely dependent on EZH2 activity for survival, "small molecule inhibitors that bind to mutant enzymes should be able to kill those cells but not have an effect on cells that don?t have the mutants," Copeland said.This is crucial when dealing with a universally important gene such as EZH2, Elemento said. "You don't necessarily want to target every single EZH2 in the body."Developing such a targeted therapy for lymphoma patients with EZH2 mutations is the team's primary focus at the moment, Copeland added.C.J. Sneeringer, et al., "Coordinated activities of wild-type plus mutant EZH2 drive tumor-associated hypertrimethylation of lysine 27 on histone H3 (H3K27) in human B-cell lymphomas," PNAS, doi:10.1073/pnas.1012525107, 2010.
**__Related stories:__***linkurl:Silenced genes drive viral cancers?;http://www.the-scientist.com/blog/display/55410/
[9th February 2009]*linkurl:A new epigenetic cancer;http://www.the-scientist.com/blog/display/55684/
[11th May 2009]*linkurl:Methylation mystery;http://www.the-scientist.com/article/home/53645/
[1st October 2007]