From Oxyaldehydes to Carbohydrates

There's got to be an easier way to make simple carbohydrates than what is currently available. David W. C. MacMillan, a chemistry professor at California Institute of Technology, envisioned a much shorter route through repetitive aldol reaction of two-carbon -oxyaldehydes. The first step would give an erythrose. The second reaction--the erythrose plus another -oxyaldehyde--would give a hexose.

Although the strategy is conceptually simple, the reactions to make it happen had yet to be developed. Now, MacMillan and coworkers Alan B. Northrup, Ian K. Mangion, and Frank Hettche have discovered an aldol reaction that accomplishes the first part of the strategy. With L-proline as catalyst, the reaction unites two -oxyaldehydes to form an erythrose in high yield and high enantioselectivity [Angew. Chem. Int. Ed., 43, 2152 (2004)].

The work is groundbreaking, opening up an entirely new way of making chiral polyols with exquisite stereospecificity, says Malcolm MacCoss, vice president for basic chemistry at Merck Research Laboratories, Rahway, N.J. "The chemistry has far-reaching ramifications, and I'm sure it will quickly become a go-to methodology for synthetic chemists interested in chiral chemistry."

Carbohydrates are essential tools in efforts to understand biological processes, notes Scott G. Nelson, an associate professor of chemistry at the University of Pittsburgh. "The ability to custom design and synthesize specific carbohydrate derivatives is essential for these investigations. MacMillan's work constitutes a novel and practical entry to catalytic asymmetric carbohydrate construction."

The use of aldol reactions to prepare complex polyols enantioselectively is well established. But only recently has the reaction been used with aldehydes as substrates. In 2002, MacMillan and Northrup showed that L-proline catalyzes the enantioselective coupling of -alkylated aldehydes [J. Am. Chem. Soc., 124, 6798 (2002)]. Before this, most chemists thought it couldn't be done. In the new work, the Caltech team shows that a similar reaction can be achieved with -oxygenated aldehydes, the required substrates for making carbohydrates. "It does not seem like a big leap, but there is a vast difference in the reactivities of -oxygenated and -alkylated carbonyls," MacMillan explains.

Beyond getting chemists halfway through MacMillan's two-step route to carbohydrates, the new chemistry could dramatically simplify the synthesis of complex macrolides such as the antibiotic erythromycin, MacMillan says. The most efficient route to that compound takes about 30 steps, making its industrial preparation by synthesis extremely difficult. Methods based on the new reaction could shorten the route to fewer than 10 steps, he points out.

As to the rest of the way to carbohydrates, MacMillan says his group has developed methods to execute the second step in his strategy. Details will be disclosed in a separate paper.