50 years ago in Biochemistry

Demystifying a Key Biochemical Reaction

Ronald Breslow
May 1, 2008

Editor's note: Citation Classics Commentaries were written by the authors of some of studies that were the most highly cited papers between 1961 and 1975. The essays were originally published between 1977 and 1993 in Current Contents, a publication of the Institute for Scientific Information (ISI), now Thomson Scientific. (ISI was founded by Eugene Garfield, also the founder of The Scientist.) This essay has been edited for space.

View other Citation Classics

As a young chemist at Columbia University in 1956, one of Ronald Breslow's first research goals was to solve the mystery of how the enzyme thiamine pyrophosphate catalyzed biochemical reactions such as the Krebs cycle and photosynthesis. He used nuclear magnetic resonance to capture the subatomic properties of the reaction. Breslow can't recall exactly what he exclaimed when he first saw a hydrogen atom from the compound exchange with deuterium oxide. "Maybe I shouted 'Hot Dog!' I...

<figcaption>Reference ? R. Breslow, " />
Reference ? R. Breslow, "On the mechanism of thiamine action. IV. Evidence from studies on model systems," J Am Chem Soc, 80:3719-26, 1958. Credit: image ? 1958 American Chemical Society Reprinted with permission from J Am Chem Soc 80(14) 1958.

Mechanistic chemists can generally propose sensible possible pathways for chemical or biochemical reactions. However, as an undergraduate learning biochemistry in the 1950s, I saw one baffling exception. Thiamine pyrophosphate (ThPP) is the coenzyme for several important biochemical processes. The intermediates in enzymatic reactions involving ThPP are formally acyl anions, which are unstable species. The function of ThPP must be to directly bond, so as to convert the anions instead to new stable species, but nothing in the structure of ThPP suggested how it might do this.

Japanese chemists had accidently discovered that treatment of benzaldehyde with a thiazolium salt led to the production of benzoin, but here too there was no idea or evidence about how this catalysis by a thiazolium salt occurs. The benzoin reaction also involves a formal acyl anion intermediate, while the most unusual feature of ThPP is the presence of a thiazolium ring. Thus, I thought it likely that solving the mechanism of the benzoin thiazolium catalysis would give an insight into the biochemical function of ThPP.

"People at Columbia still remember my shout when I dissolved a thiazolium salt in D2O and saw a new C-D band appear in the infrared."

When I came to Columbia, I worked on this benzoin reaction. After other possibilities were systematically excluded, it became clear that the catalysis must involve the carbon atom between sulfur and nitrogen of the thiazolium ring. If the C-H bond ionized (for which there was no precedent), the resulting carbon anion should be able to act as catalyst. The test was whether this hydrogen atom would exchange with D2O. People at Columbia still remember my shout when I dissolved a thiazolium salt in D2O and saw a new C-D band appear in the infrared. Definitive proof came from NMR studies.

Using a home-built 30 MHz instrument in the laboratory of Ben Dailey, I watched a single proton signal in the NMR of thiazolium salts disappear as that proton exchanged with D2O. From its NMR position it was clearly the proton on the carbon between sulfur and nitrogen; from the rate of exchange I could estimate the acidity of that hydrogen atom. This was the first application of NMR to such a mechanistic problem. I could then propose a pathway for the formation of benzoin catalyzed by thiazolium salts and related pathways for the biochemical processes in which ThPP plays a role.

Chemical models can suggest how biological processes occur, and most of the interest in this paper came from that aspect. However, information can also flow in the other direction. Our studies stimulated chemists to invent new synthetic methods, using thiazolium salts as catalysts. Also, we and others have prepared enzyme mimics that incorporate thiazolium salts into synthetic or natural binding sites and perform useful synthetic steps with characteristic enzyme-like selectivity.

There are still other poorly understood biochemical processes waiting to teach us some new chemistry while we solve a biochemical mystery. In the years since our paper, I have had the pleasure of introducing many PhD students and postdoctorals to this field. They are now some of the most active and successful investigators of enzyme models and their mechanisms. While I have received a number of awards and honors for this earliest work and later research, the success of so many of my former coworkers is at least as gratifying.

Ronald Breslow, Department of Chemistry, Columbia University, June 28, 1993.

Correction (posted May 21) A previous version of the article incorrectly stated Ronald Breslow's first name. The Scientist regrets the error.