New Route To Treat Diabetes | October 20, 2008 Issue - Vol. 86 Issue 42 | Chemical & Engineering News
Volume 86 Issue 42 | pp. 34-36
Issue Date: October 20, 2008

New Route To Treat Diabetes

Glucokinase activators in development show promise as an alternative for diabetics
Department: Business
Activating Moment
Roche chemists study the crystal structure of a drug bound to glucokinase.
Credit: Roche
Activating Moment
Roche chemists study the crystal structure of a drug bound to glucokinase.
Credit: Roche

AMERICANS SEEM to have awoken to the dual epidemics of obesity and diabetes. Consumers are starting to read food labels more closely, schools are taking sugary drinks and snacks out of the lunchroom, and entire U.S. cities have gone on diets. Still, the incidence of type 2 diabetes, a by-product of obesity and the most common form of the disease, is growing by leaps and bounds. According to the Centers for Disease Control & Prevention, some 24 million Americans—about 8% of the country—have some form of diabetes, up by 3 million people in just two years. Another 57 million are labeled as "prediabetic."

Type 2 diabetes occurs when the body starts losing its ability to respond to the insulin it produces. Because insulin is responsible for shuttling sugar to cells, glucose begins to accumulate in the bloodstream, depriving cells of energy and straining multiple organs. The disease can sometimes be controlled by diet and exercise, but many patients end up needing drugs to help lower the level of glucose in their blood.

Newer and better drugs for type 2 diabetes have made their way to patients in recent years. Byetta, a peptide-based injectable drug marketed by Amylin Pharmaceuticals and Eli Lilly & Co., lowers blood glucose and offers modest weight loss, and Merck & Co.'s pill Januvia avoids some of the side effects that have kept some patients on older drugs from sticking to their regimens.

But drug companies still see room for improvement in controlling blood sugar safely, consistently, and without side effects. Glucokinase, an enzyme involved with sensing glucose in tissues, has emerged as an attractive new target that researchers hope will provide another option for diabetics. Big pharma and biotech firms alike are eyeing the opportunity, as evidenced by the sudden explosion of patent applications for glucokinase activators last year. To date, companies including Roche, Lilly, and AstraZeneca have managed to get their molecules into the early stages of clinical trials.

The opportunity for effective diabetes treatments is vast: Sales of diabetes drugs totaled $12 billion in the U.S. last year, according to the health care research firm Datamonitor. Merck's Januvia franchise, which includes Januvia and a combination of Januvia and the older diabetes drug metformin, brought in $754 million last year.

Glucokinase belongs to a family of enzymes called hexokinases, which catalyze the phosphorylation of glucose to glucose-6-phosphate, a critical first step in metabolizing sugar. Hexokinases are generally marked by their ubiquity—several serve housekeeping functions and are thus found in nearly every tissue in the body—and their tight bond to glucose. "At first glance you'd think it would be difficult, if not impossible, for a compound to interact with and actually activate the enzyme," says David E. Moller, vice president of endocrine and cardiovascular research at Lilly.

But glucokinase is something of a black sheep among hexokinase kin. It is found in relatively fewer tissues, and its affinity for glucose is delicate. In the pancreas it is believed to "sense" just the right concentration of glucose in β cells to signal the release of insulin, explains Joseph Grimsby, senior research leader at Roche. And in the liver glucokinase initiates the first step of glucose metabolism, kicking into action after a meal and later sensing when the body is in a fasting state and needs to store glucose.

THOSE DIFFERENCES made it an intriguing target for drug companies looking for new approaches to treating diabetes. In diabetes, those critical glucose-sensing processes have gone awry, says Array BioPharma's director of medicinal chemistry, Steven A. Boyd. "The amount of insulin secreted is insufficient to bring down the blood glucose, and the liver is putting out too much glucose and not taking up enough," he says.

But glucokinase became even more compelling as a target in the 1990s, after two forms of diabetes were directly linked to a genetic mutation in the enzyme. Scientists found that a rare form of diabetes called maturity-onset diabetes of the young type 2 (MODY-2) is caused by a mutation that renders glucokinase less sensitive to glucose; without the right signal from glucokinase, not enough insulin is secreted, and blood sugar levels increase. Subsequently, researchers discovered an even less common mutation that causes glucokinase to be overactive, causing patients to have higher insulin levels and not enough glucose circulating in their bodies.

"This is very distinct and unlike any other drug target in the diabetes field in that there is extremely strong human genetic validation in both directions—loss and gain of function," Lilly's Moller explains. "Putting it all together, it is about as compelling as you can get in terms of a concept on paper."

Yet the enzyme also poses a unique challenge for researchers. Small molecules have traditionally been used to block an enzyme—akin to simply flipping a switch off. In the case of glucokinase, medicinal chemists had to think for the first time about the more nuanced process of activating an enzyme. "With an inhibitor, you don't think about it all that much, you just inhibit the enzyme," Grimsby points out. "But here, you need to fine-tune how your activator activates the enzyme."

Initially, companies took the brute force approach to finding molecules that would interact with glucokinase: They screened gigantic compound libraries in hopes of a hit.

Armed with those first hits, scientists were able to develop a better mechanistic understanding of the enzyme. X-ray crystallography has shown that glucokinase has a pocket that serves as an allosteric binding site, one whose conformation changes when the right small molecule attaches. The interaction between molecule and enzyme was then confirmed by what is observed in patients. Researchers were able to show that the genetic mutations that either ramped up or dampened the activity of glucokinase were all clustered around the same binding site, Array's Boyd explains. Scientists were then able to design or optimize compounds based on that site.

Tapping into the body's metabolic pathways is a tricky and potentially unsafe business.

THE CURRENT CROP of compounds is in the early stages of clinic trials. AstraZeneca has molecules in Phase I and II trials, and Roche's lead glucokinase activator, R1511, is in Phase I trials. In January 2007, Lilly licensed OSI Pharmaceutical's lead glucokinase activator, a compound now in Phase I trials. Array's lead compound, ARRY-403, is in advanced preclinical studies, and the company hopes to initiate Phase I trials in the first half of 2009.

Although they are early in clinical development, glucokinase activators could have some advantages over existing diabetes drugs, researchers believe. Because glucokinase mediates insulin release in the pancreas and regulates glucose metabolism in the liver, molecules that activate it are expected to have an effect on both organs. Januvia and Byetta, in contrast, affect only the pancreas.

Overall, companies developing glucokinase activators think their approach could be better at a "net degree of glucose lowering," Lilly's Moller says. He notes that the first group of potential patients would likely be those who are no longer responding to Januvia.

The companies developing glucokinase activators are keenly aware that tapping into the body's metabolic pathways is a tricky and potentially unsafe business. Indeed, several diabetes drugs have stumbled over safety concerns in recent years. Sales of GlaxoSmithKline's Avandia, which acts on the peroxisome-proliferator-activated receptor (PPAR), plummeted after a report last year linked the drug to increased risk of heart attacks.

Other PPAR agonists had also fallen to safety issues. In 2000, Warner-Lambert pulled Rezulin from the market due to liver toxicity, and in 2005, Merck and Bristol-Myers Squibb pulled their New Drug Application for the PPAR agonist Pargluva, citing cardiovascular safety concerns. Byetta, meanwhile, has been linked to several cases of pancreatitis, resulting in two deaths.

Glucokinase researchers say they are trying to incorporate tests into their discovery and clinical strategies to ensure their products are safe. The Food & Drug Administration has indicated that its biggest concern is whether a prospective diabetes drug increases the risk of cardiovascular events, because type 2 diabetics already have a high chance of heart attacks. Because glucokinase is primarily expressed in the pancreas and liver—and not in the heart—companies appear confident their drugs won't have the cardiovascular risks that the PPAR agonists do.

"Our strategy was to look at cardiovascular safety and make sure we weren't having any negative impact," Array's Boyd says. In animal studies during the optimization of ARRY-403, the company monitored changes in blood pressure and the electrophysiology of the heart. It also tested whether cholesterol was raised. "Both we and our competitors have shown that glucokinase activators are at worst neutral but in some cases have a beneficial effect," Boyd says.

According to Lilly's Moller, the main concern with turning on glucokinase is hypoglycemia, because instances of low blood sugar have been observed in animals given a high dose of a glucokinase activator. But he points out that physicians already have to work with patients to determine the optimal dosage of other diabetes drugs, particularly insulin. "We expect that it will be manageable," he says.

And although it is too early to predict how often patients will have to take any glucokinase activator that makes its way into the clinic, Moller says a pill that would last between six and 24 hours would be ideal because it could be quickly flushed from the system if a patient did develop hypoglycemia.

Chemical & Engineering News
ISSN 0009-2347
Copyright © American Chemical Society

Leave A Comment

*Required to comment