Hidden inside every human cell, a molecular editing machine quietly shapes which proteins our genes produce. Now, researchers at the University of California San Diego have thrown open the doors on a part of that machinery that has long stayed in the shadows. Their new study, published in Molecular Cell, identifies 63 proteins that control a process called alternative polyadenylation (APA) — a fundamental step in gene expression that influences how more than 70 percent of human genes function. Perhaps most remarkably, 56 of these proteins had never been connected to APA before, representing what the team calls a "vast majority" of newly identified regulators.
The discovery matters because APA determines where an RNA molecule gets cut and finished before it becomes a protein, affecting the stability, localization, and function of thousands of genes. When APA goes awry, it has been linked to cancer, neurological disorders, and other diseases. Understanding the full cast of proteins that govern this process could open new doors for treatment. "Understanding this hidden machinery could offer new possibilities for treating diseases related to APA dysregulation," the researchers noted.
Led by Gene Yeo, a professor of cellular and molecular medicine at UC San Diego who directs both the Center for RNA Therapeutics and Technologies and the Sanford Stem Cell Innovation Center, the team screened 879 human RNA-binding proteins using a custom reporter system. In addition to the sheer scale of new regulators uncovered, the study revealed unexpected findings for two proteins in particular: GRB2 and RNPS1. Neither had ever been associated with APA, yet both were shown to directly interact with components of the cellular machinery responsible for the process.
The researchers also trained a protein language model to predict APA regulators directly from protein sequences, successfully identifying activators in an independent validation set and highlighting regions of proteins that appear critical for their function. This computational approach could dramatically accelerate future discoveries. Beyond identification, the team developed a programmable RNA-targeting platform that can recruit proteins to specific poly(A) sites — creating what they describe as a potential framework for scientists to manipulate RNA processing with precision.
The work was conducted in collaboration with Youngsheng Shi, a professor of microbiology and molecular genetics at UC Irvine. Together, the findings suggest that the molecular toolkit governing gene expression is far richer and more interconnected than scientists realized — and that some of the most important players have been hiding in plain sight.