Beta Cell Discovery Holds Promise for Diabetes Treatment
Shivatra Talchai wasn’t the only researcher in Domenico Accili’s lab at Columbia University Medical Center who was working toward a cure for diabetes, but at 25, she was certainly the youngest. Nicknamed “Noi,” which translates roughly—and appropriately—to “junior” in her native Thai, she had yet to complete her doctorate in metabolic biology, but that didn’t stop her from approaching Accili as a graduate student in 2004.
Something of a legend in the world of diabetes research, Accili had a reputation for being a nurturing and supportive mentor to the researchers who worked with him. Talchai was determined to land one of the 10 or so coveted spots in his lab, even though she was two years shy of earning her PhD, a prerequisite for working with Accili. “I was a graduate student when I started, so I was quite the exception,” she says. “I wouldn’t take no for an answer.” In the incremental world of scientific discovery, where years pass with no guarantee of success, that sort of persistence would serve her well. She had no way of knowing it at the time, but the discovery she would make about what causes beta cells to lose their ability to make adequate amounts of insulin—the hallmark of diabetes—would represent what Accili would call “a sea change” in the way the scientific community views the disease.
An Auspicious Beginning
Arguably the most promising development for people with diabetes since the discovery of insulin began with the roundworm, of all things. In the late 1990s, scientists observed some connections that made them wonder if a gene originally identified for its effects on the lifespan of roundworms might also play a role in insulin action. The idea that a gene with the unwieldy name of Forkhead box O—FoxO, for short—could have implications for people with diabetes seemed farfetched at the time. How could something that happens in a spaghetti-like parasite known mostly for wreaking havoc on the intestines of puppies be of any consequence to people, never mind people with diabetes? “We worry about curing diabetes in people, so any time we try to reproduce the disease in an animal—be it a dog, a squirrel, a monkey, or a mouse—we’re straying further away from the actual person who [has] diabetes,” says Accili, the Russell Berrie Foundation professor of diabetes and director of the Diabetes Research Center at Columbia University College of Physicians and Surgeons. “When it was proposed that the FoxO gene played a role in insulin action in the roundworm, there was really no way to predict that it would be relevant to a human being.”
But there was plenty to suggest it might be. For starters, the discovery of FoxO in worms answered a question researchers had been unable to address: How does insulin control which genes are turned on and off? Scientists had been working on targeting various enzymes that break down glucose. But if they could understand how insulin affects these enzymes, they might be able to identify a single target—what they call a “master regulator”—to treat diabetes more effectively.
Imagine “an old house with leaky faucets in the bathrooms and kitchen,” says Accili. “The first thing you want to do is stop the leaks, so you look for the water main. Looking for the way in which insulin regulates genes was like looking for the water main. Can we control that so we can then go back and fix the individual faucets? FoxO [seemed to] meet the requirements of a gene that could be modulated by insulin to affect many genes at once.” In other words, insulin could alter FoxO (the water main) in a way that affected a lot of genes (the leaky faucets) at the same time. Or at least that’s the way it looked on paper.
The Path to Discovery
Laying the foundation for the kind of discovery that leads to new therapies and treatments is often the work of many scientists. Other scientists then add building blocks—one painstaking lab finding at a time—until, finally, what emerges is a game-changer.
Case in point: In the late 1990s, Accili was one of many researchers to show that insulin regulates the FoxO gene by changing where it’s situated in the liver cell. FoxO sits in the nucleus of the cell, where its job is to turn on some genes and turn off others. When cells are exposed to insulin, FoxO responds by leaving the nucleus very quickly. For someone who might be prone to type 2 diabetes, that quick exit from the nucleus is a good thing because it stops activating the genes that are linked to diabetes.
That finding helped lay the foundation for what came next. In 2003, Accili and his team were able to show in mice that when FoxO’s location inside the cell changes, glucose levels do, too. “This was interesting, but not nearly as interesting as what we found going further into this,” says Accili. “We had been focusing on the liver because that’s the glucose factory for the body and it’s the key site of the disease in type 2 diabetes. But we know that diabetes affects virtually all the organs in the body—from the kidneys to the lungs, the eyes, and the brain.” His next challenge: determine how this mechanism explains what happens in these areas, as well as in the insulin-producing beta cell.
“For people with type 2 diabetes, the beta cell was on the receiving end of years and years of metabolic stress,” says Accili. “The beta cell would first try to adapt to the body’s changing needs for insulin by making more insulin but would eventually give up once the need became overwhelming. We knew this process—so-called beta cell failure—was not really understood. The assumption was that the beta cell just dies of exhaustion, but I never found this convincing.”
Accili thought of the people with type 2 he treated as a young doctor in the ’80s. Once they began following a healthier eating and exercise regimen, they “would rebound with robust insulin properties,” he says. He knew this wasn’t consistent with the long-held notion that their beta cells had died. He just couldn’t explain it.
Maybe Talchai, his young protégé with a particular knack for creative thinking, could.
The Aha Moment
By 2006, the FoxO gene had all but taken over the Accili lab. Virtually every researcher was focused on FoxO in various organs and tissues (the brain, the liver, the heart, the gut, the fat cells) and not just in type 2 diabetes, but in type 1, as well. Talchai—at the time 27 and now armed with a PhD—was assigned the beta cell, and this mystery: What was the physiological significance of Accili’s research showing the relationship between FoxO’s location in the cell and the change in glucose levels?
She began her experiment by removing FoxO from the beta cells of a group of mice. Then she waited (and waited) for them to age. Initially, she noticed, the mice were normal. But over time, as they experienced regular stressors of life—specifically, multiple pregnancies in the female mice and aging in the males—they developed high blood glucose, decreased insulin secretion, and other signs of type 2 diabetes, the same way people with type 2 do. That combination means the insulin that should be made in order to lower blood glucose isn’t being produced, suggesting something is wrong with the beta cell.
What happened to the beta cells of these mice that now had diabetes? By 2009, Talchai still had no definitive answer. After almost four years on the case, she was ready to give up. At the outset, Accili told her what he tells every researcher taking on a project in his lab: “Don’t worry about how long it takes or how much it costs. My concern is making sure you’ll be cared for, and your concern is devoting 100 percent of your waking life energies to the project.” And that’s what Talchai had done. But her 60- and 70-hour work weeks yielded nothing conclusive.
On the day in 2009 she showed up at the lab ready to shut down the experiment, she spotted something she’d never seen before. In looking at the pancreas of a pregnant mouse with diabetes, she saw that the beta cells hadn’t died. She had been using a genetic trick to visualize beta cells and insulin with colored labels. When the beta cells were making insulin, as they should, they would appear yellow under the microscope; if beta cells were dying, she would have seen fewer yellow cells. But what she saw instead were green cells in place of yellow ones, indicating the cells that used to make insulin were still alive. In other words: The beta cells weren’t dead; they were essentially sleeping.
“People had always assumed the main reason beta cells failed in people with type 2 diabetes was because the beta cells died,” says Talchai. “But on that day, I saw that that was not true. The beta cells were actually alive. They did not produce insulin anymore, but they did not die.”
As aha moments go, this discovery—a process known as dedifferentiation—was “a real stunner,” says Accili, who was awarded the American Diabetes Association’s 2017 Banting Medal for Scientific Achievement for long-term contributions to the understanding, treatment, or prevention of diabetes. “We’ve been telling you, ‘Too bad—your beta cells are gone. Let’s give you insulin to replace them,’ ” he says. “But now we’re telling you, ‘The beta cells are still there and we know that if we act early enough we can bring them back to life so that you may never need to go on insulin.’ Nowhere has the scope of scientific inquiry been more advanced by this line of investigation on FoxO than in the pancreatic beta cell.”
Now that scientists know the beta cells aren’t dead but are merely sleeping, there’s the potential to wake them up through lifestyle modifications such as exercise, diet, and certain drugs. Scientists are still trying to figure out how to do that.
“It’s small advances that turn into big changes that benefit patients,” says Accili. “You get to the right answer through a process of unending mistakes. Once we’ve stripped off the mistakes, what’s left behind is the new discovery.”
3 Questions for Domenico Accili, MD
1. People who live with diabetes hear about all the research that’s being done and wonder why it’s taking so long to find a cure. Why should they be hopeful about the current research?
“The tools to understand the biology of disease have never been as powerful as they are today. Our understanding of [the way diabetes starts and develops], insulin resistance, and beta cell failure has reached a critical stage. We are now positioned to design new classes of drugs that will tackle root causes of diabetes and consign the disease to the realm of medical curiosities, much the same way as polio or smallpox.”
2. How close are we?
“The road ahead is fraught with uncertainty, and I don’t want to minimize the obvious difficulties of translating this promising biology. In many ways, this is only a start. But I do want to emphasize how a formerly intractable problem has been de-convoluted into its simplest parts by way of methodical, if boring, scientific advance and is now ripe for transformative drug discovery.”
3. How ripe is the field for drug discovery?
“Patients are frustrated by our putting dates on things we can’t yet foresee. It could literally be tomorrow and it could literally be 20 years from today. We have all the ingredients—it’s up to us to make them work.”
What About Type 1?
The FoxO gene also has implications for type 1 diabetes. In a study published in a 2014 issue of the journal Nature Communications, Columbia University’s Domenico Accili, MD, and his team showed that by turning off FoxO in the gut, they were able to convert human gastrointestinal cells into insulin-producing cells. The hope is that someday a drug could trigger the same effect in humans.
Additional sources: Utpal Pajvani, MD, PhD, assistant professor of medicine in the Division of Endocrinology at Columbia University Medical Center; and Rebecca Haeusler, PhD, assistant professor of pathology and cell biology at the Naomi Berrie Diabetes Center at Columbia University Medical Center.