First-in-Class Allosteric Inhibitors of DNMT3A Disrupt Protein–Protein Interactions


DNA methyltransferase 3A (DNMT3A) belongs to a family of highly conserved DNA methyltransferases that catalyse 5-methylcytosine methylation. Regulatory domains of DNMT3A allow interactions with histone methyltransferases and histones to influence gene expression. DNMT3A mutations occur in diverse haematological malignancies with unique mutational profiles. The R882 hotspot mutation occurs most frequently in AML and has been shown to act as a dominant-negative inhibitor of wild-type DNMT3A enzymatic activity.

A team led by researchers at UC Santa Barbara led by Distinguished Professor Norbert Reich, and including collaborators from UC San Francisco and Baylor College of Medicine, has identified two compounds that are more potent and less toxic than current leukemia therapies. The molecules work in a different way than standard cancer treatments and could form the basis of an entirely new class of drugs. What’s more, the compounds are already used for treating other diseases, which drastically cuts the amount of red tape involved in tailoring them toward leukemia or even prescribing them off-label. The findings appear in the Journal of Medicinal Chemistry. The research identified an enzyme that is mutated in leukemia patients and then found a new way of regulating this enzyme, as well as new molecules that are more effective and less toxic to human cells

All cells in your body contain the same DNA, or genome, but each one uses a different part of this blueprint based on what type of cell it is. This enables different cells to carry out their specialized functions while still using the same instruction manual; essentially, they just use different parts of the manual. The epigenome tells cells how to use these instructions. For instance, chemical markers determine which parts get read, dictating a cell’s actual fate. A cell’s epigenome is copied and preserved by an enzyme (a type of protein) called DNMT1. This enzyme ensures, for example, that a dividing liver cell turns into two liver cells and not a brain cell. However, even in adults, some cells do need to differentiate into different kinds of cells than they were before. For example, bone marrow stem cells are capable of forming all the different blood cell types, which don’t reproduce on their own. This is controlled by another enzyme, DNMT3A. This is all well and good until something goes wrong with DNMT3A, causing bone marrow to turn into abnormal blood cells. This is a primary event leading to various forms of leukemia, as well as other cancers.

Most cancer drugs are designed to selectively kill cancer cells while leaving healthy cells alone. But this is extremely challenging, which is why so many of them are extremely toxic. Current leukemia treatments, like Decitabine, bind to DNMT3A in a way that disables it, thereby slowing the progression of the disease. They do this by clogging up the enzyme’s active site (essentially, its business end) to prevent it from carrying out its function. Unfortunately, DNMT3A’s active site is virtually identical to that of DNMT1, so the drug shuts down epigenetic regulation in all of the patient’s 30 to 40 trillion cells. This leads to one of the drug industry’s biggest bottle necks: off-target toxicity. Clogging a protein’s active site is a straightforward way to take it offline. That’s why the active site is often the first place drug designers look when designing new drugs, Reich explained. However, about eight years ago he decided to investigate compounds that could bind to other sites in an effort to avoid off-target effects.

While most of these epigenetic-related enzymes work on their own, DNMT3A always formed complexes, either with itself or with partner proteins. These complexes can involve more than 60 different partners, and interestingly, they act as homing devices to direct DNMT3A to control particular genes. Early work in the Reich lab, led by former graduate student Celeste Holz-Schietinger, showed that disrupting the complex through mutations did not interfere with its ability to add chemical markers to the DNA. However, the DNMT3A behaved differently when it was on its own or in a simple pair; it wasn’t to stay on the DNA and mark one site after another, which is essential for its normal cellular function. Previous work showed that study the most frequent mutations in acute myeloid leukemia patients are in the DNMT3A gene. Surprisingly, the authors had studied the exact same mutations. The team now had a direct link between DNMT3A and the epigenetic changes leading to acute myeloid leukemia.

The research team worked in identifying drugs that could interfere with the formation of DNMT3A complexes that occur in cancer cells. They obtained a chemical library containing 1,500 previously studied drugs and identified two that disrupt DNMT3A interactions with partner proteins (protein-protein inhibitors, (PPIs).  These two drugs do not bind to the protein’s active site, so they don’t affect the DNMT1 at work in all of the body’s other cells. These drugs are more than merely a potential breakthrough in leukemia treatment. They are a completely new class of drugs: protein-protein inhibitors that target a part of the enzyme away from its active site.

 One major advantage is the two compounds the team identified have already been used clinically for other diseases. This eliminates a lot of cost, testing and bureaucracy involved in developing them into leukemia therapies. There’s still more to understand about this new approach, though. The team wants to learn more about how protein-protein inhibitors affect DNMT3A complexes in healthy bone marrow cells.

First-in-Class Allosteric Inhibitors of DNMT3A Disrupt Protein–Protein Interactions - Medicine Innovates

About the author

Professor Norbert O. Reich
Department of Chemistry & Biochemistry
University of California Santa Barbara

Dr. Reich lab is interested in the study of enzyme function, primarily at the levels of catalytic and specificity mechanisms, with a long term goal of developing enzyme inhibitors with potential therapeutic application. The current focus is on both bacterial and mammalian enzymes involved in the epigenetic process of DNA methylation and on telomerase. We also work on the development of new methods and technologies to investigate cell-biological, biochemical and biomedical questions.


Sandoval JE, Ramabadran R, Stillson N, Sarah L, Fujimori DG, Goodell MA, Reich N. First-in-Class Allosteric Inhibitors of DNMT3A Disrupt Protein-Protein Interactions and Induce Acute Myeloid Leukemia Cell Differentiation. J Med Chem. 2022;65(15):10554-10566. doi: 10.1021/acs.jmedchem.2c00725.

Go To J Med Chem.