Frank Jordan

of action, and folding of two groups of enzymes. Our research on thiamin
diphosphate (the vitamin B1 coenzyme)-dependent enzymes has followed a dual
approach. Part of our group is studying how a-keto acid decarboxylascs work,
including both nonoxidative and oxidative functions. Questions being asked
include determining the 3-D struc-ture of the protein, the conformation and
activation of the coenzymes in catalysis, the nature of proton transfer events
in catalysis, and the mechanism of electron and group transfer in the oxidative
processes. Tools used in these studies include high resolution X-ray
crystallography (pyruvate decarboxylase, the simplest of these enzymes was
solved to a resolution of 2.3A), nuclear magnetic resonance spectroscopy,
molecular genetics to change amino acids in the structure to establish
structure-function relationships, steady-state and pre-steady-state kinetics,
and immunochemical methods. Our enzyme studies are being complemented with model organic chemical studies, in which the elementary steps of the multi-step enzymatic reactions are being modeled. From such chemical models, one can learn which step and by how much the enzyme must accelerate.

Our research on serine proteases has two principal aims. First, we are
attempting to define the active center electronic structure in the enzymes
in the absence and in the presence of both synthetic and protein protease
inhibitors. From these studies, a picture is emerging that can differentiate
ground-state vs. transition-state vs. acyl-enzyme type inhibitors, providing
a convenient tool for drug design. The second part of our research focuses
on the varied roles of the pro-sequence in enzymes expressed as
pre-pro-proteins. Subtilisin, an alkaline bacterial protease, is the model
being used (27.5 kDa), since its pro-sequence (8.5 kDa, 77 amino acids) has
been shown to be important both for folding of the mature sequence, as well
as for inhibiting it.