A team at the University of California San Diego has developed a large-scale screening approach that identifies proteins controlling alternative polyadenylation (APA), a fundamental step in gene expression that occurs in more than 70% of human genes. The study was published in Molecular Cell.
APA determines where an RNA molecule is cut and finished before it is translated into protein, influencing the stability, localization, and function of thousands of genes. Abnormalities in APA have been implicated in cancer, neurological disorders, and other diseases, yet many of the proteins that regulate the process have remained unknown.
To address this gap, the research team screened 879 human RNA-binding proteins using a custom reporter system designed to measure how individual proteins influence APA. The screen identified 63 high-confidence activators of poly(A) site usage—the most important step of the APA process. Excluding known positive controls, only seven of these proteins had previously been associated with APA, meaning the vast majority represent newly identified regulators. Follow-up experiments confirmed that many of these proteins alter RNA processing in cells and affect distinct groups of genes involved in specific biological functions.
Among the most unexpected findings were new roles for two proteins, GRB2 and RNPS1, neither of which had previously been associated with APA. The study showed that both proteins can directly interact with components of the cellular machinery responsible for APA.
The researchers also trained a protein language model to predict APA regulators directly from protein sequences. The model successfully identified activators in an independent validation set and highlighted regions of proteins that appear critical for their function—an approach that could help accelerate the discovery of RNA regulatory proteins and provide insight into how they work.
Beyond cataloguing new regulators, the team developed a programmable RNA-targeting platform capable of recruiting specific proteins to particular poly(A) sites, creating a potential framework for manipulating RNA processing in a targeted manner. This could offer new possibilities for treating diseases linked to APA dysregulation.
APA determines where an RNA molecule is cut and finished before it is translated into protein, influencing the stability, localization, and function of thousands of genes. Abnormalities in APA have been implicated in cancer, neurological disorders, and other diseases, yet many of the proteins that regulate the process have remained unknown.
To address this gap, the research team screened 879 human RNA-binding proteins using a custom reporter system designed to measure how individual proteins influence APA. The screen identified 63 high-confidence activators of poly(A) site usage—the most important step of the APA process. Excluding known positive controls, only seven of these proteins had previously been associated with APA, meaning the vast majority represent newly identified regulators. Follow-up experiments confirmed that many of these proteins alter RNA processing in cells and affect distinct groups of genes involved in specific biological functions.
Among the most unexpected findings were new roles for two proteins, GRB2 and RNPS1, neither of which had previously been associated with APA. The study showed that both proteins can directly interact with components of the cellular machinery responsible for APA.
The researchers also trained a protein language model to predict APA regulators directly from protein sequences. The model successfully identified activators in an independent validation set and highlighted regions of proteins that appear critical for their function—an approach that could help accelerate the discovery of RNA regulatory proteins and provide insight into how they work.
Beyond cataloguing new regulators, the team developed a programmable RNA-targeting platform capable of recruiting specific proteins to particular poly(A) sites, creating a potential framework for manipulating RNA processing in a targeted manner. This could offer new possibilities for treating diseases linked to APA dysregulation.