New map of proteins may help in understanding mutations linked to cancer and ALS
Researchers at the University of Toronto and the Hospital for Sick Children (SickKids) have released a "map" that can make predictions about the biological functions of thousands of mysterious regions in human proteins known as intrinsically disordered regions (IDRs).
The breakthrough could be an important step toward understanding the proteins that are mutated in diseases like cancer, neurodevelopmental and neurodegenerative disorders, such as amyotrophic lateral sclerosis (ALS).
The research is described in the paper, “A functional map of the human intrinsically disordered proteome,” published in the Proceedings of the National Academy of Sciences.
The lead author is Iva Pritišanac, a former postdoctoral researcher at U of T and SickKids, who is currently a research group leader at the Institutes of Computational Biology and Structural Biology at Helmholtz Munich. Her lead co-authors include Alan Moses, a computational biologist and professor in the Department of Cell & Systems Biology in the Faculty of Arts & Science; and Julie Forman-Kay, senior scientist, Molecular Medicine, SickKids; and professor (status only), Department of Biochemistry, Temerty Faculty of Medicine.
Proteins do many of the major jobs in our cells; they enable the digestion of lactose, copy our DNA, carry oxygen in your blood, and more.
Classically, proteins are known to do their jobs because of their 3-dimensional shape structures. Thanks to the deep-learning/AI revolution, major breakthroughs in the last few years — including Nobel prize-winning work — have led to computer programs that can predict the 3D structures for most proteins. Knowing these 3D structures can tell scientists a lot about the jobs performed by the thousands of proteins still poorly understood.
However, the same structure prediction programs strongly predict that many protein regions don’t have stable 3D structures and instead are intrinsically disordered. These IDRs are found among proteins that cause diseases like cancer and neurological disorders when mutated, such as the proteins encoded by the BRCA genes that are mutated in breast cancers. Because most of their sequences don’t have stable 3D structures, it’s been challenging for scientists to determine the normal jobs of these kinds of proteins.
The team, led by Moses and Forman-Kay, invented a new computer program to predict the functions of IDRs and applied it to nearly 20,000 proteins encoded in the human genome.
As a result, the biological functions for thousands of human IDRs are now freely available for the first time. According to Moses, the key breakthrough was to transform the proteins into data points in a more than 100-dimensional space of chemical and amino acid sequence properties, like electric charge, ability to dissolve in water, and presence of amino acids motifs that enable binding to other proteins.
“Projecting into a high-dimensional space actually made the problem of predicting IDR function easier,” says Moses. “When we first discovered this, it was one of the most surprising moments I can remember.”
“Alan and I developed this approach a number of years ago and first applied it to yeast,” says Forman-Kay, a world-renowned expert in IDRs. “Our first published paper about it was in 2019.
“This new work is our application of that theory to the human proteome, and the potential for understanding disease and health issues in humans is really exciting. We've been working on the theory for years, but this is the systematic application to the human proteome.”
The researchers’ new approach could provide insight into the BRCA1 and BRCA2 proteins and how mutations in them cause breast cancer. The proteins are mostly disordered and have been studied for decades, but their functions are still not understood.
"Exploration of the map of the human IDRs led me to predict that BRCA1's normal job is to help position the recombination enzymes on the meiotic chromosomes,” says Moses.
“Recombination is the exchange of genes between chromosomes, and usually happens during meiosis, when sperm and egg cells are made. At first this seems to have nothing to do with cancer, but in fact, cancer causing mutations often lead to genome instability, exactly what we might expect to happen if recombination isn't working correctly.
“If we can understand that function of these proteins using this new approach, then when the mutations happen, we could potentially fix the mutations and prevent the cancer,” says Moses. “That's the most exciting implication of this kind of work.”
There is still a lot to learn about IDRs, and Forman-Kay, Moses and several other groups around the world are now working to understand how exactly they do their jobs without stable 3D shapes.
“We have the hypothesis that highly dynamic interactions of IDRs in the cellular environment are key to the mechanisms of IDR function” says Forman-Kay. “Although I’ve been working on IDRs for many years, this new computational approach makes it an exciting time for our field.”
This story by Christopher Sasaki was originally published on the Arts & Science website.