Is the Case for a Diels-Alderase Dead?

It's natural for chemists to wonder whether the ingenious reactions that have been invented in the laboratory over the past century were actually harnessed millions of years ago by enzymes, nature's complex and powerful catalysts.

Take the case of the 77-year-old Diels-Alder reaction, perhaps the most powerful chemical synthetic tool in use today. A conjugated diene and an alkene or alkyne combine during this reaction to create two carbon-carbon bonds in one concerted step, forming the cyclohexene rings that are ubiquitous in industrial and pharmaceutical chemistry. Is it possible that some enzymes have long been performing this remarkable reaction in the course of their own synthesis of biochemical compounds?

Answering that question, which chemists have been exploring avidly for the past 10 years, has proven to be anything but clear-cut. Although more than 100 proteins have been flagged as potential "Diels-Alderases," only three--lovastatin nonaketide synthase, solanapyrone synthase, and macrophomate synthase, all from fungi--have been isolated.

Things had looked especially promising for macrophomate synthase (MPS), which converts 2-pyrone derivatives to their corresponding benzoate analogs. Two years ago, chemistry professor Hideaki Oikawa and colleagues at Hokkaido University, Sapporo, Japan, determined MPS's crystal structure (C&EN, March 17, 2003, page 32), finally making it possible to move forward with substantive theoretical studies of the system.

Although definitive proof that MPS and the other two enzymes catalyze the Diels-Alder reaction remains elusive, a tenor of certitude permeates some of the literature, where these proteins are referred to simply as Diels-Alderases.

Now, Yale University chemistry professor William L. Jorgensen and colleagues have tossed a bucket of cold water on the Diels-Alderase crusade with a new theoretical study suggesting that macrophomate synthase likely doesn't catalyze a Diels-Alder reaction after all (J. Am. Chem. Soc. 2005, 127, 3577). Rather, their calculations show, a much more energetically probable scenario involves the sequential-step Michael-aldol reaction, whose transition states are 17.7 and 12.1 kcal per mol more stable than a model for the Diels-Alder transition state.

In addition to the MPS crystal structure, advances in computational methods played a key role in the new work, Jorgensen says. "In the last half dozen years, people have really begun to do sophisticated calculations on enzyme mechanics," he says.

Oikawa remains undaunted. "We regard the JACS paper as an opinion and not a conclusion," he says.

Observers say the new paper's results can't be dismissed. "They seem to unambiguously prove the Michael-aldol pathway to be the most likely," says Georg Pohnert, chemistry professor at Max Planck Institute for Chemical Ecology, in Jena, Germany.

Robert M. Williams, chemistry professor at Colorado State University, Fort Collins, believes the papers on Diels-Alderases made overambitious claims. Merely noting that the substrates and products, and the resulting stereochemistry of these systems, look tantalizingly like the hallmarks of a Diels-Alder reaction "is very far from proving a concerted mechanism," Williams says. "I was delighted to finally see somebody published a theoretical paper on this to give a sober perspective."

Even John C. Vederas, chemistry professor at the University of Alberta, Edmonton, whose lab purified lovastatin nonaketide synthase (LNKS) and is hot on the trail of its potential Diels-Alderase activity, says the new work "makes a strong case for the stepwise process" in MPS, while still not detracting from the "beauty and elegance" of Oikawa's crystallographic results.

These experiments are difficult because the proteins are each involved in several processes. For example, the transformation catalyzed by MPS consists of three distinct reactions: the decarboxylation of oxalacetate to form pyruvate enolate, the formation of two C-C bonds between the pyruvate enolate and 2-pyrone, and the decarboxylation and dehydration of the resulting bicyclic intermediate. Isolating the suspected Diels-Alder pathway in the middle has been a challenge.

Several experimental strategies, although complicated and arduous, could help nail down the mechanism in these enzymes, researchers note. The kinetic isotope effect, in which reaction rates are altered by the substitution of an isotope such as ¹³C for ¹²C, could help elucidate the synchronicity of bond formations. "There are lots of ways of distinguishing stepwise and concerted reactions," says Donald Hilvert,, chemistry professor at the Swiss Federal Institute of Technology, Zurich, who is working on mutant and kinetic isotope studies of MPS in collaboration with Oikawa's group. "Some are more ironclad than others."

The other two enzymes--LNKS and solanapyrone synthase--are still prime Diels-Alderase contenders.