The School of Molecular and Cellular Biology at the University of Illinois at Urbana-Champaign

The Catalytic Power of Enzymes

Enzymes are catalysts of unparalleled power and specificity, which essentially allow chemical reactions to take place on a timescale conducive to life. Enzyme catalyzed reactions have been shown to have rate accelerations of up to 10²⁰, relative to uncatalyzed reference reactions. This acceleration can be ascribed to a number of factors, one of which is transition state stabilization (i.e., the enzyme binds tightly to the transition state). This is usually achieved by highly specific interactions between the enzyme and transition state, in the form of hydrogen bonding and electrostatic stabilization. Other factors that affect reaction rates include dynamical recrossing, quantum mechanical tunneling and non-equilibrium distributions of the total degrees of freedom (non-Boltzmann distribution). While there are a myriad of terms that have been employed to describe the origins of enzyme catalyzed rate enhancements, stabilization of the transition state is thought to be the primary contributor.

The goal of my research is to understand the determinants of enzyme catalytic power from a chemical and physical perspective. It is hoped that a detailed understanding of the chemical and physical principles that govern enzyme catalyzed reactions will lead to significant scientific and medical advances. One of the ways this will be accomplished is by determining the precise transition state structures and energetics for enzymes of medical relevance. Enzymes that catalyze very large rate accelerations via tight binding to a transition state should, in principle, exhibit exceedingly tight binding to an appropriately designed transition state analogue. This can be accomplished by obtaining the precise transition state structure for the enzyme catalyzed reaction, and finding a transition state analogue that closely approximates the native binding determinants.

Obtaining Transition State Structures

Enzymes represent a unique target class for drug design. Unlike cell surface receptors, DNA, ion channels etc., enzymes catalyze covalent bond making and breaking. Thus, in addition to exploiting a binding event, one may also capitalize on the highly specific chemical events that take place in the enzyme's active site, vis-à-vis drug design efforts. There are 317 FDA approved drugs that target enzymes (human, bacterial, viral, fungal and protazoal). The majority of these drugs (~65%) rely on some type of substrate mimicking (direct interaction with cofactor, structural resemblance to substrate, T.S. analogue, etc.). In order to exploit these highly specific chemical interactions, one must fully understand the enzyme mechanism, and perhaps even obtain a detailed transition state structure. This level of detail can only be accomplished by a combination of kinetic isotope effect methodologies and computational chemistry, which will be discussed below.

Use of Kinetic Isotope Effects

Substitution of light isotopes for their heavy counterparts frequently leads to changes in the observed reaction rates for chemical reactions in which the isotope is involved. The measurement of such kinetic isotope effects is a very powerful tool for the diagnosis of reaction mechanisms.

For example, when a proton (protium) is replaced by deuterium or tritium, changes in the rate of the reaction as large as several fold may take place. However, such substitutions can also yield small changes (10 - 20 %) in the rate, requiring specialized techniques to define their magnitude. Larger effects usually result when the isotope is transferred in the transition state that limits the overall rate of the reaction. Such reactions give what are referred to as primary kinetic isotope effects.

The reason that isotopic substitution is such a sensitive probe is that we can completely separate electronic and nuclear motion, due to the large mass difference between electrons and nucleons (the Born-Oppenheimer Approximation). This means that there is no change in the vibrational force constants (which govern the motion of atomic nuclei) upon isotopic substitution. Thus, in principle, isotopic substitution does not alter the transition state structure, making it an exquisitely sensitive probe. Isotope effects are primarily due to differences in the zero point energies (i.e., the zeroth vibrational state) between the light and heavy atoms. The difference in the isotopically sensitive vibrational frequencies between the light and heavy atoms in the reactants is the major contributor to the isotope effect. This energy difference results in the light reacting faster than the heavy isotope.

Kinetic isotope effects are measured in order to obtain information on either 1) the mechanistic sequence of a multi-step chemical reaction, or 2) the transition state structure of a simple reaction. Enzymes are highly efficient catalysts due to their ability to bind very tightly to and thus stabilize transition state structures. They generally catalyze chemical reactions via a multi-step sequence. Therefore, KIE values are extremely useful tools for the analysis of enzyme mechanisms since they can provide information on both the nature of the rate-limiting step(s) and the transition state structure of the rate-limiting step, if it is sensitive to isotopic substitution

Questions to be Addressed

• What are the determinants for transition state binding for a number of pharmacologically important cofactor-independent amino acid racemases and epimerases (e.g., diaminopimelate epimerase and glutamate racemase)? We will answer this by determining the free energy profiles and intrinsic hydrogen kinetic isotope effects, which will guide our computational efforts in determining the various transition state structures.
• What are the physicochemical contributions of the cofactor pyridoxal-5'-phosphate (PLP) to the catalytic power and efficiency of a number of PLP-containing enzymes of pharmacological import?
• Is it possible, by means of directed evolution methodologies, to improve the catalytic efficiency of wild-type enzymes that are far from chemically perfect (i.e., not diffusion-controlled)? Is it possible to evolve a chemically perfect enzyme? Would there be significant biological disadvantages to these chemically perfect enzymes?