The Story of DNMT3A
With DNMT3A mutations, faulty programming makes leukemia more likely.
No one knew about the relationship DNMT3A has to cancer – and various forms of leukemia in particular – until 2010. That’s when researchers led by Timothy Ley from Washington University in St. Louis reported that they had sequenced the complete genome of a woman who died of acute myelogenous
leukemia (AML) in her 50s, along with the genome of her cancer cells. Those genomic comparisons between healthy and cancerous cells can lead researchers right to the genetic roots of the problem. In this case, her leukemia cells carried a mutation not found in her other cells in the gene known as DNMT3A (for “DNA methyltransferase”).
That by itself wasn’t necessarily such a big deal. But when they looked for
this mutant gene in the leukemia cells of more than 280 other patients with the same diagnosis, they turned up DNMT3A mutations a whopping 21 percent of the time, in 62 individuals. What’s more, those mutations showed up in more than a third of patients who were deemed, based on other features, as being at “intermediate risk” for leukemia. The findings suggested a potential role for this gene in determining who might be at greater risk of an unfavorable outcome and how aggressively the leukemia should be treated as a result.
It’s now clear that changes in DNMT3A are exceedingly common in AML and other blood cancers. The genetic mutation may be a culprit in a variety of other solid tumors as well. Mutations or other abnormalities in the activity of this gene turn up in lymphomas, as well as gastric and lung cancers.
These findings have led to recognition that DNMT3A is a critically important new tumor suppressor representing what some consider a new class of oncogene. While most oncogenes cause cancer by encouraging cells in one way or another to grow selfishly out of control, DNMT3A represents another path to the
disease. The gene may be mutated, but ultimately the problem is not one of genetics directly but of epigenetics.
The prefix “epi-” means “over,” “outside of,” or “around.” These changes around genes are made by enzymes that modify DNA or its packaging with the addition or subtraction of chemical marks. The presence or absence of epigenetic marks can change the way genes work despite the fact that the underlying sequence
stays the same. Some people compare epigenetic marks to punctuation marks, which tell the cell’s machinery how or whether our genes should be read. Others explain epigenetics by likening the genome to a computer’s hardware. In this analogy, epigenetics is the software that tells our “computer” what to actually do.
Generally speaking, if lots of chemical marks are added, genes are switched off. Loss of those chemical marks, on the other hand, turns genes on.
Now back to DNMT3A, which is the protein produced by the gene DNMT3A. Just as its name suggests, the protein’s job is to transfer chemical methyl marks to DNA, writing the initial epigenetic programming within cells. If DNMT3A doesn’t do its job properly, cells aren’t automatically destined for cancer, but they are surely prone to future problems.
While knowledge of DNMT3A’s role may be new, the basic idea that cancer cells may have faulty epigenetic programming isn’t. Scientists first noticed epigenetic aberrations to cancer cells’ DNA a couple of decades ago. The thinking was that changes to those chemical marks were in some cases switching important tumor suppressor genes off. Or, in other cases, perhaps too few chemical marks made our genomes less stable.
There’s still plenty left for researchers to learn about the underlying molecular biology of DNMT3A in cancer, but Ley’s discovery made perfect sense from the start to Margaret Goodell, an expert in blood-forming stem cells at Baylor College of Medicine in Houston. When Ley published his initial results, Goodell had been studying the gene in stem cells for about three years and recognized its cancer-causing potential. Not long after Ley’s discovery, Goodell says there were dozens more studies on DNMT3A and cancer.
“It was a remarkable explosion,” she says. “[Scientists had] been studying the gene for at least 20 years. It was right in front of our noses that this gene might be implicated in hematologic malignancies and no one had noticed.”
The loss of DNMT3A in blood stem cells seems to push them into a state of persistent self-renewal. As those DNMT3A-deficient stem cells go on dividing, they outcompete other normal cells and gradually start to accumulate. A mutation in DNMT3A isn’t an instant recipe for leukemia. Blood stem cells can persist in the body for many years, decades even. Researchers suspect that DNMT3A mutations produce a kind of “pre-leukemic state,” upping the odds for the future development of the disease.
Mutations in DNMT3A often may arise early in the path to leukemia, only progressing to full-blown disease when additional hits come into play. For those with AML who go into remission, the lingering presence of these mutations may up the odds for a relapse.
Existing drugs namely decitabine and 5-azacytidine, are DNA methytransferase inhibitors, which appear to influence the DNMT3A protein and related enzymes. But those drugs were developed before the discovery of DNMT3A and no one knows exactly how they work. Clinical trials of these epigenetic drugs are currently underway for the treatment of AML and other cancers. Goodell says her group is also working on studies in laboratory mice with leukemia, with the goal to explore the drugs’ effects without all the complexity found in human cancers.
The hope is that new and improved DNA methyltransferase inhibitors will be found. DNMT3A can be mutated in different ways, but most of the time it seems to happen in one particular spot. That consistency might be useful in the search for drugs designed to specifically target the aberrant enzymes and restore the epigenetic program.
“We want to understand better who will respond and to think about which drugs can be modified or how to make new drugs to treat patients with [these] specific mutations,” Goodell said.