ANN ARBOR—Building on recent discovery of a potent potential antibiotic, University of Michigan College of Pharmacy researchers have found a previously unknown family of metal-requiring enzymes in bacteria. They have also demonstrated that the antibiotic compound they are studying effectively inhibits enzymes in this family.

The latest work by Prof. Ronald Woodard and colleagues follows from research they reported in the Sept. 27, 2000, Journal of the American Chemical Society. In that project, the researchers screened some 150,000 compounds and found one, PD 404182, that was 10,000 times more effective than other known inhibitors of a key enzyme in Gram negative bacteria. The compound targets the enzyme KDO 8-P synthase, which plays a vital role in the formation of antenna-like lipopolysaccharides on the surfaces of bacterial cells. Lipopolysaccharides have numerous functions—helping bacteria defend themselves against antibiotics and host immune responses, for example.

The new research focuses on the target enzyme, KDO 8-P synthase. Collaborating with Prof. Domenico L. Gatti at Wayne State University School of Medicine, Woodard’s group had previously determined the crystal structure of the enzyme in the bacterium E. coli. However, it was impossible to use those crystals to observe how PD404182 binds to the active site of KDO 8-P synthase, because the holding solution in which KDO 8-P synthase crystals are stable contains high concentrations of salt, which interfere with binding.

To get around that problem, Woodard’s group isolated and crystallized KDO 8-P synthase from another bacterium, Aquifex aeolicus, which lives at high temperatures in hydrothermal vents. A. aeolicus does not cause human disease, but can serve as a useful model for pathogenic organisms. Because the A. aeolicus enzyme is active only at high temperatures, the researchers could introduce substrates to the crystallized enzyme at lower temperatures without causing a complete reaction. The enzyme—which is very similar in structure to the one from E. coli—bound the substrates and changed shape, but the enzyme-substrate complex did not react to form the end product KDO 8-P. As a result, the researchers were able to determine eight intermediate structures of KDO 8-P synthase, representing critical steps in the catalytic cycle. This part of the work will be published in the March 16 issue of the Journal of Biological Chemistry. In addition to Woodard and Gatti, authors are U-M research fellow Henry Duewel and Sergei Radaev and Jian Wang of Wayne State University.

“This is useful information for drug design, as we see almost see the actual structure of the intermediate immediately before catalysis,” says Woodard. “As far as we know, this is the first time that a crystal structure has been determined using active enzyme and all of the necessary substrates.”

In the process of working with KDO 8-P synthase from A. aeolicus, Woodard’s group discovered that, unlike its counterpart from E. coli, this enzyme requires a metal for activity. Furthermore, it is not affected by the potential antibiotic PD 404182.

“Before the current study, the presence of any metal-requiring KDO 8-P synthase was unknown and unsuspected,” says Woodard. “If such a family did exist, then, based on our initial studies, a different approach might be needed to design a drug—or to modify PD 404182—against the metallo-KDO 8-P synthases.”

To explore that possibility, doctoral student Matthew Birck searched existing DNA/protein databases and found 31 KDO 8-P synthases. He selected 25 for which the most complete information was available and then used statistical techniques to predict which of these might require metal. The analysis, which was published in the Journal of Molecular Evolution last month, showed two distinct classes of KDO 8-P synthase. In experiments on eight KDO 8-P synthases, another student, David Howe, confirmed that they do indeed fall into two classes—one that requires a metal and one that does not.

To determine whether these enzymes, like the A. aeolicus enzyme, are resistant to the prospective antibiotic PD 404182, Birck tested the compound on metal-requiring KDO 8-P synthase from Helicobacter pylori J99, the bacterium implicated in gastric ulcers and cancers. PD 404182 inhibited the enzyme, making Woodard and colleagues hopeful that the compound may someday be useful against a wide range of disease-causing bacteria.

Because PD 404182 weakens, but does not kill, living bacteria, Woodard’s group hopes to modify the compound in ways that will enhance its antibacterial activity.

College of PharmacyRonald WoodardDomenico L. Gattipart of the work