Duke Chemists Synthesize Fungus Compound That Could Lead To Oral Diabetes Drugs

The compound, called demethylasterriquinone B1 or "DAQ" for short, "has this fascinating property of being able to activate insulin receptors in cells in basically the same way as insulin, and yet it's not a protein," said Michael Pirrung, a Duke chemistry professor who prepared his team's work for presentation Wednesday at the American Chemical Society's national meeting in San Diego.

"Instead, it's just a regular organic molecule like you take for your allergies. And because it's a regular organic molecule you can take it orally, which is such a big deal," Pirrung added in an interview at Duke. Insulin to treat diabetes requires injections. "The other big thing is that it's a naturally occurring molecule," he noted. "And one area that combinatorial chemistry has previously not really made any inroads into is naturally occurring compounds."

Combinatorial chemistry seeks to efficiently synthesize and identify chemicals with desired traits by mixing and matching large "libraries" of different molecular building blocks to eventually create the desired structure. Pirrung helped pioneer "solid phase synthesis" techniques for performing combinatorial chemistry reactions on the surfaces of small glass chips.

However, "the DAQ molecule is not particularly amenable to synthesis while its pieces are attached to a solid," he said. Fortunately, Pirrung noted, "the principles of combinatorial chemistry have now been expanded to molecules that are made in solution."

Instead of fixing molecular libraries to a surface, tiny samples of each library entry can be placed in a different small well, with different reagents then pipetted in to interact there, he said.

Merck Research Laboratories scientists originally discovered DAQ by systematically screening more than 50,000 combinations of synthetic and natural compounds for molecules that activate human insulin receptors, according to articles in the May 7, 1999 issue of the journal Science.

In a report in the Jan. 12, 2001 issue of the American Chemical Society journal Organic Letters, Pirrung, Duke post-doctoral fellow Kaapjoo Park and graduate student Zhitao Li described the start of now-completed efforts by Duke combinatorial chemists to synthesize DAQ more efficiently.

"The molecule itself is highly modular," Pirrung said. "It has an indole ring, a quinone ring, and a different indole ring. So that makes it perfect for combinatorial chemistry. If you have three variants of the first module, three of the second and three of the third, just by combining all those forms there are 27 possible compounds to make. So that enables you to make very large collections of related molecules."

Indole, a protein decomposition product occurring in some flower oils, is a ring-shaped organic molecule made of carbon, hydrogen and nitrogen atoms. Quinones are ring-shaped compounds containing double-bonded carbon and oxygen groups.

"One of the things that I think is really interesting about this project is that it's enabling us to address basic questions about the insulin receptors and how they work," said Pirrung, who also is a key member of Duke's Program in Biological Chemistry. To study insulin receptors, his lab will begin raising the fruit fly Drosophila, an insect that uses insulin receptors to regulate growth rather than blood sugar levels.

The authors of the Science report also noted that DAQ does not activate the receptor for insulin-like growth factor-1 (IGF-1), which is similar to the insulin receptor and has been tied to both prostate and breast cancer.

Pirrung noted that Nicholas Webster, a researcher at the University of California San Diego, has suggested evaluating DAQ-like compounds for use against prostate cancer.

In addition to its support from the American Diabetes Association, Pirrung's lab has now begun studying the chemistry of the IGF-1 receptor with funding from CapCURE, a foundation begun by Michael Milken to address prostate cancer. "There are many other so-called growth factors in the cell that have similar though not the same kind of receptor," he said. "And we're thinking that these molecules we've been working on might have what medicinal chemists sometimes call a ‘privileged structure.'

"There are certain kinds of structures that keep showing up over and over again in biologically active molecules, steroids being an obvious example. What we're hoping is that this initial hit has gotten us into chemicals that have privileged structures for growth factor receptors.

"So having a big selection of molecules to look at, with a large number of different growth factor receptors, could be very powerful in finding new molecules that could selectively turn the receptors off or on."