A chemist investigates how proteins assume their shape

When proteins are first made in our cells, they often exist as floppy chains until specialized cellular machinery helps them fold into the right shapes. Only after achieving this correct structure can most proteins perform their biological functions.

Many diseases, including genetic disorders like cystic fibrosis and brittle bone disease, and neurodegenerative diseases like Alzheimer’s, are linked to defects in this protein folding process. Matt Shoulders, a recently tenured associate professor in the Department of Chemistry, is trying to understand how protein folding happens in human cells and how it goes wrong, in hopes of finding ways to prevent diseases linked to protein misfolding.

“In the human cell, there are tens of thousands of proteins. The vast majority of proteins must eventually attain some well-defined three-dimensional structure to carry out their functions,” Shoulders says. “Protein misfolding and protein aggregation happen a lot, even in healthy cells. My research group’s interest is in how cells get proteins folded into a functional conformation, in the right place and at the right time, so they can stay healthy.”

In his lab at MIT, Shoulders uses a variety of techniques to study the “proteostasis network,” which comprises about a thousand components that cooperate to enable cells to maintain proteins in the right conformations.

“Proteostasis is exceedingly important. If it breaks down, you get disease,” he says. “There’s this whole system in cells that helps client proteins get to the shapes they need to get to, and if folding fails the system responds to try and address the problem. If it can’t be solved, the network actively works to dispose of misfolded or aggregated client proteins.”

Building new structures

Growing up in the Appalachian Mountains, Shoulders was homeschooled by his mother, along with his five siblings. The family lived on a small farm near Blacksburg, Virginia, where his father was an accounting professor at Virginia Tech. Shoulders credits his grandfather, a chemistry professor at Ohio Northern University and Alice Lloyd College, with kindling his interest in chemistry.

“My family had a policy that the kids helped clean up the kitchen after dinner. I hated doing it,” he recalls. “Fortunately for me, there was one exception: If we had company, and if you were in an adult conversation with the company, you could get out of cleaning the kitchen. So I spent many hours, starting at the age of 5 or 6, talking about chemistry with my grandfather after dinner.”

Before starting college at nearby Virginia Tech, Shoulders spent a couple of years working as a carpenter.

“That’s when I discovered that I really liked building things,” he says. “When I went to college I was thinking about fields to get into, and I realized chemistry was an opportunity to merge those two things that I had begun to find very exciting — building things but also thinking at the molecular level. A big part of what chemists do is make things that have never been made before, by connecting atoms in different ways.”

As an undergraduate, Shoulders worked in the lab of chemistry professor Felicia Etzkorn, devising ways to synthesize complex new molecules, including stable peptides that mimic protein functions. In graduate school at the University of Wisconsin, he worked with Professor Ronald Raines, who is now on the faculty at MIT. At Wisconsin, Shoulders began to study protein biophysics, with a focus on the physical and chemical factors that control which structure a given protein adopts and how stable the structure is.

For his graduate studies, Shoulders analyzed how proteins fold while in a solution in a test tube. Once he finished his PhD, he decided to delve into how proteins fold in their natural environment: living cells.

“Experiments in test tubes are a great way to get some insight but, ultimately, we want to know how the biological system works,” Shoulders says. To that end, he went to the Scripps Research Institute to do a postdoc with professors Jeffery Kelly and Luke Wiseman, who study diseases caused by protein misfolding.

Neurodegenerative diseases like Alzheimer’s and Parkinson’s diseases are perhaps the best known protein misfolding disorders, but there are thousands of others, most of which affect smaller numbers of people. Kelly, Wiseman, and many others, including the late MIT biology professor Susan Lindquist, have shown that protein misfolding is linked to cellular signaling pathways involved in stress responses.

“When protein folding goes awry, these signaling pathways recognize it and try to fix the problem. If they succeed, then all is well, but if they fail, that almost always leads to disease,” Shoulders says.

Disrupted protein folding

Since joining the MIT faculty in 2012, Shoulders and his students have developed a number of chemical and genetic techniques for first perturbing different aspects of the proteostasis network and then observing how protein folding is affected.

In one major effort, Shoulders’ lab is exploring how cells fold collagen. Collagen, an important component of connective tissue, is the most abundant protein in the human body and, at more than 4,000 amino acids, is also quite large. There are as many as 50 different diseases linked to collagen misfolding, and most have no effective treatments, Shoulders says.

Another major area of interest is the evolution of proteins, especially viral proteins. Shoulders and his group have shown that flu viruses’ rapid evolution depends in part on their ability to hijack some components of the proteostasis network of the host cells they infect. Without this help, flu viruses can’t adapt nearly as rapidly.

In the long term, Shoulders hopes that his research will help to identify possible new ways to treat diseases that arise from aberrant protein folding. In theory, restoring the function of a single protein involved in folding could help with a variety of diseases linked to misfolding.

“You might not need one drug for each disease — you might be able to develop one drug that treats many different diseases,” he says. “It’s a little speculative right now. We still need to learn much more about the basics of proteostasis network function, but there is a lot of promise.”