Skip to main content
Premium Trial:

Request an Annual Quote

Chemical Proteomic Probes Reveal Epigenetic Enzymes at Work in Human Cancer Cells


NEW YORK (GenomeWeb) – Methyltransferases are an important class of epigenetic enzymes and are implicated in cancer biology in a number of ways. Attempts to identify the entire subset of proteins has been resistant to proteome-wide searches using chemical probes, but now a new probe developed by scientists at the Scripps Research Institute could help complete the search.

Scientists led by Benjamin Cravatt have developed a chemical proteomic probe based on the molecule S-adenosyl homocysteine (SAH) to enrich methyltransfersases in a sample and used quantitative proteomics to identify them in human cancer cell lysates. They also showed the method was able to help identify compounds to inhibit the enzymes. The researchers published their results last week in the Journal of the American Chemical Society.

"This method has been around for a long, long time, but the proteome-wide capacity of these probes to enrich methyltransferases had not been investigated," Cravatt told GenomeWeb. Suspected to play a role in tumorigensis, researchers and pharmaceuticals are interested in developing methyltransferase inhibitors as a potential cancer treatment. This importance made it an "obvious" choice for study, he said, adding that the field lacked sufficient tools to enrich them, characterize them, and assess the selectivity of inhibitors.

"While the study didn't identify all of the more than 200 known methyltransferases, the study found many, and more importantly was incredibly specific to those enzymes and other closely associated proteins," he said.

"It's a pretty impressive piece of work," said Jordan Meier, a scientist at the National Cancer Institute who also works with chemical proteomic probes, but who was not associated with the study. "They had to put together a challenging synthetic probe along with really cutting-edge mass spec and that hadn't been done before."

While chemical probes to search proteome-wide for epigenetic enzymes like acetyltransferases and deubiquitinases have been around for a while, probes for methyltransferases have always been elusive. Meier, who has published work on acetyltransferase probes, said simply finding the right place for a photocrosslinking probe was something that had to be done empirically, making it especially challenging.

Prior research into the enzyme class was focused on specific methyltransferases, but "this study is trying to look at the whole family," he said. "It's more ambitious and more useful for discovery biology than understanding the mechanism of a single drug."

The idea behind the chemical probe is to segregate the enzyme family of interest based on its interaction with small molecules, Cravatt said. "That gives you a chance to inventory a portion of the proteome." His lab has previously developed probes for profiling cholesterol-binding proteins.

"Fortunately, the product of the methyltransferase-catalyzed reaction, SAH, is a good affinity ligand for most of the enzymes of the class, so we could use that as a starting point to create a photoreactive probe to label and enrich native proteomes," Cravatt said, adding that study showed that the probes are "remarkably specific." Proteins enriched by the probes were usually already known to be associated with methyltransferases.

But they do have one disadvantage: low yield. That made the quantitative mass spec analysis crucial. "Having the mass spec back end is valuable to get you as deep into the coverage as possible," Cravatt said. "We still have major improvements that can be made on that front to get deeper coverage. If you look at data, there are a handful of proteins that were picked up sparsely. I think it's not because they're not being enriched, it just means the sensitivity was not good enough to see them. As we develop a more advanced protocol, I think that aspect will improve."

Meier added that while it's an important step, a further goal goal is to be able to identify proteins in a living animal. "Right now they're working in complex proteomes or cell extracts.
What you'd really like is to do the survey in intact cells or living organisms, and the current approach isn't compatible with that," he said, although he acknowledged that the authors addressed some ideas to get to those kinds of studies.

In the paper, the Scripps team showed off one immediate application, screening for and assessing selectivity of methyltransferase inhibitors. They even found a novel covalent inhibitor of nicotinaminde N-methyltransferase, an enzyme potentially involved in cancer and other metabolic disorders.

Based on the ability of the probes to identify methyltransferase-associated proteins that didn't themselves interact with SAH, Cravatt suggested it could be used to find other epigenetic enzymes. "Epigenetic proteins often participate in macromolecular complexes, so you could potentially inventory these complexes these proteins are associated with as they bind to the methyltransferase," he said.

Meier thought the method could be even more important than the researchers let on in their paper. "They really focus on drug design in this paper, but another application is in terms of biodiscovery and how endogenous metabolites interact with methyltransfersases" he said. There is a subset of cancers that have mutations in the gene MTAP, leading to high levels of a metabolite that interacts with methyltransferases. "This is believed to drive cancer development, but we really don't understand which enzymes interact more strongly. [The probes] could tell us how metabolic inputs such as nutrition or cancer signaling drive tumorigenesis," Meier said.

"The technology could be used to discover new enzymes associated with aggressive cancer phenotypes, or reveal the mechanism of action of drugs. It has the potential to enable a lot of different applications that affect a really important class of enzymes," he said.