Structural study reveals MLL4's dual cancer role as both driver and suppressor

Robert Roeder's laboratory at Rockefeller University has spent years puzzling over a protein that seems to play two entirely different roles depending on where it shows up in the body. MLL4, one of six members of a family of histone-modifying enzymes, acts as a disease accelerant in certain leukemias while simultaneously suppressing solid tumors. The contradiction was maddening—the same protein, the same cellular machinery, opposite outcomes. Now, using a combination of structural imaging, genetic manipulation, and transcription studies, Roeder's team has begun to explain how this paradox actually works, publishing their findings in Molecular Cell.

MLL4 belongs to the mixed-lineage leukemia family of proteins, each responsible for tagging histone 3 at a specific location—lysine 4—which acts as a molecular switch to turn genes on or off. The protein is enormous, the largest in the mammalian nucleus, and it appears in virtually every cell in the body. In leukemia cells, MLL4 performs a protective function: it shields cancer cells from oxidative and genetic stress and keeps leukemia stem cells in a self-renewing state. But in solid tumors, it works alongside p53, the famous tumor-suppressing protein sometimes called the genome's guardian, to activate genes involved in DNA damage response and cell death. The mechanism behind this cooperation remained mysterious until Jianfeng Sun, a structural biologist in Roeder's lab, decided to map the entire nine-subunit complex at high resolution.

Using cryo-electron microscopy, Sun produced the first complete structural model of MLL4 in multiple configurations. The images revealed something unexpected: the protein anchors itself to the nucleosome—the core unit of DNA packaging—with rigid structures, but extends a flexible arm to search out and tag histones with a methylation marker. More striking still, the protein's N-terminal region folds back onto its C-terminal region, creating an architectural arrangement that turns out to be essential not just for its known histone-methylating function but for something entirely new: direct cooperation with p53 in activating tumor-suppressing genes.

When the researchers genetically knocked out MLL4 in cells, p53 lost much of its ability to activate its target genes—those involved in cell cycle arrest, DNA repair, and programmed cell death. Without MLL4, the genome guardian was essentially disarmed. "This was a very surprising finding," Roeder said. "MLL4's primary function in gene transcription is through histone 3 methylation, and yet here we're seeing that it's also essential for p53 target gene transcription as a direct p53 co-activator. That's a second-and entirely new-function." The discovery suggests that MLL4 operates through at least two distinct mechanisms: one through the histone-marking pathway that was already known, and another through direct partnership with p53 that was entirely hidden until now.

The next phase of research will focus on how MLL4 interacts with leukemia-associated transcription factors—proteins that function similarly to p53 but drive cancer progression instead of suppressing it. Understanding these interactions could reveal why the same protein supports disease in one cellular context and suppression in another. Roeder frames the long-term goal plainly: to understand how MLL4 can simultaneously fuel leukemia transcription programs in one setting while enforcing tumor suppression in another. That knowledge could eventually point toward therapies that exploit MLL4's role in leukemia without disrupting its protective function elsewhere.