Several roads lead to elevated blood sugar levels in diabetes. The best-known mechanism is insulin deficiency, whether through destruction of pancreatic islet cells in Type I diabetes, or deficiencies in insulin secretion or action in Type II diabetes.

Therapeutic efforts have by and large targeted insulin levels directly - from the 1982 approval of South San Francisco-based Genentech Inc.'s Humulin (human insulin) as the first recombinant DNA product, to the most recent addition in the therapeutic arsenal, Amylin Pharmaceuticals Inc.'s Byetta, which was approved by the FDA last month. (See BioWorld Today, May 2, 2005.)

But in an unfortunate synergistic effect, the liver also tends to produce too much glucose in diabetics in the first place. And research published by scientists from San Diego-based Metabasis Therapeutics Inc. and Harvard University in the May 31, 2005, edition of the Proceedings of the National Academy of Sciences, reported preclinical details of their compound CS-917 (now moved into clinical trials), that shows promise in shutting down that route to elevated blood glucose levels, as well.

"For a long time, the view of diabetes was that it is caused by inadequate insulin levels," said Mark Erion, executive vice president of research and development at Metabasis. Increased liver production was seen as secondary. "It is now being recognized that over time, the liver becomes a very important problem."

The researchers decided to target FBPase, an enzyme that catalyzes the second-to-last step in glucose generation, because several pathways feed into it and success at targeting that late step would shut down glucose production from pretty much any starting point. They focused on the enzyme's binding site for adenosine monophosphate (AMP). AMP is leftover from the cell's energy generation via ATP use. "A lot of pathways have AMP binding sites, and they tend to be synthesis pathways," Erion told BioWorld Today. "It's nature's way of turning down these pathways when there's an energy crisis" and the organism shifts from builder to survival mode.

Targeting the AMP site with a noncompetitive inhibitor rather than a substrate binding site had two advantages. "In gluconeogenesis, all enzymes bind highly charged molecules; for medicinal chemists, these are a nightmare," Erion said. Additionally, blocking the substrate binding site leads to buildup of substrate, which makes it all the more important to design a high-affinity competitive inhibitor - exactly what is difficult if the binding site is very polar.

Using structure-based drug design principles, the researchers first identified a series of compounds that could inhibit glucose production in liver cell cultures. The compound that was chosen for further testing, CS-917, inhibited glucose production in liver cells.

In acute and chronic animal studies, CS-917 was able to lower blood glucose levels "both in early stage and what was very advanced diabetes." Insulin-targeting drugs depend on having a minimum level of insulin available in the first place, and in the later stages of diabetes, as beta cells become less and less able to function, there is less and less insulin available. "We've demonstrated that our drug works even in those late stages," Erion said, "and we are intrigued by that."

The preclinical side effect profile also was encouraging. Administration of CS-917 caused no hypoglycemia and only mild elevation of lactate levels. At least in the case of lactate levels, though, the preclinical data has been overtaken by clinical events; in March, a pair of clinical studies was halted after two serious adverse events involving lactic acidosis were reported. One was testing a combination regimen on CS-917 and metformin, and the other relatively high doses of CS-917 as a solo agent. (See BioWorld Today, March 17, 2005.)

Erion said that Metabasis and its partner, Tokyo-based Sankyo Co. Ltd., are evaluating the compound's fate. One Phase I study is continuing, and Erion is optimistic about the drug's future, partly because of its structure-based drug design genesis.

Metformin also targets glucose synthesis, but its precise mechanism of action is not fully understood, either for inhibiting gluconeogenesis or the production of lactate, a known side effect of the drug. The same lack of molecular detail holds true for some of the drugs that target insulin. In contrast, Erion said, "we know exactly how this drug works."