Electron density for two Alrestatin molecules bound to the active site of human aldose reductase
Research in my laboratory is primarily concerned with molecular recognition, or answering the question of "how does molecule A recognize and bind molecule B?" Our research focuses on proteins that are involved in diabetes and other debilitating diseases. We use the methods of mechanistic enzymology, small molecules organic synthesis, site-directed mutagenesis, and X-ray diffraction to establish on an atomic level how each protein functions.
Electron density for two
Alrestatin molecules bound to
the active site of human aldose
Our most developed area of research is on the diabetes related enzyme, human aldose reductase. As a postdoctoral fellow I determined the structure and deduced the mechanism of this enzyme. We are now engaged in mapping a region known as the "specificity" determinant for use in the design of highly specific small molecule inhibitors. The specificity determinant was found when we determined the structure of aldose reductase with the drug Alrestatin bound to the active site. This structure is remarkable in that there are two drug molecules bound in the same active site cavity. The carboxylate of one molecule is positioned in the manner that was expected from the previously determined mechanism of inhibition. The carboxylate of the other drug molecule is binding to a region of the enzyme that is unique to aldose reductase, and not found in other members of the aldo-keto reductase super-family.
The stacking arrangement of the Alrestatin molecules is reminiscent of the structure of ferrocene acetic acid derivatives. Using a combinatorial chemistry approach, we will attach a ferrocene acetic acid to a random peptides and select the peptides which bind aldose reductase tightly, and that do not bind the related enzyme aldehyde reductase. The structure of aldose reductase bound to the appropriate ferrocene acetic acid peptide derivative will be determined, and the geometries between the functional groups will be used in designing highly specific inhibitors.
The specificity determinant and other sites will also be probed using the new technique of "solvent mapping." The diffraction pattern of an organic solvent soaked cross-linked protein crystal is measured. The location of the bound solvent molecules are mapped to the surface of the protein. After a number of solvents are mapped to the surface of the protein, a unique three-dimensional pattern of functional group binding sites emerge. This information is used to design highly specific inhibitors. This technique is a valuable teaching aid, as it allows a student to experience many of the facets of protein crystallography (crystal growth and manipulation, data collection, and electron density fitting) during a three month period of time.
A second project in my laboratory focuses on methylglyoxal synthase an enzyme found on an alternate branch of the glycolytic pathway. Methylglyoxal syntase appears to be fundamental to bacterial life, yet not present in eukaryotes. The enzyme catalyzes the irreversible conversion of dihydroxy acetone phosphate (DHAP) to methylglyoxal and inorganic phosphate. This project is typical of future studies in my laboratory as we have purified the endogenous protein, used its N-terminal sequence to clone and sequence the gene, used site-directed mutagenesis to identify residues critical for catalysis, and we have obtained crystals and are in the process of determining the three dimensional structure. Surprisingly, the enzyme appears to be unrelated to triose phosphate isomerase, an enzyme that converts DHAP to glyceraldehyde phosphate. It will be exciting to learn how these two enzymes which bind the same substrate control their reaction coordinate to avoid making the other enzyme's product. We have also crystallized the next enzyme on this alternate pathway, the ubiquitous enzyme, glyoxalase I, which is important since it detoxifies a major class of cancer chemotherapeutic agents. Our goal is to help in the development of novel glyoxalase I inhibitors that will allow these chemotherapeutic agents to function.
In collaboration with Dr. Henry Miziorko, at MCW, I am also working on the structure of the bacterial form of the photosynthetic protein phosphoribulokinase (PRK). While this enzyme is essential for life on the planet and controls the rate determining step in CO2 uptake, this project also addresses some of the broader issues of phosphoryl transfer common to all kinases. Having successfully determined the structure of the apo-enzyme, we are currently trying to obtain crystals with substrates, products, allosteric effectors, or inhibitors bound, and in this way create some of the "still frames" in a three dimensional movie of the enzyme at work. In collaboration with Dr. Richard Sabina, also at MCW, we have successfully crystallized the AMP-deaminase. This protein, important in muscular disorders, is functionally similar to adenosine daminase, yet the enzymes share little sequence similarity and does not catalyze the deamination of the other's substrate.