Substrate selectivity and chemical specificity are the hallmarks of enzyme activity. However, we still know relatively little about how the structure of an enzyme controls which substrate is bound, how substrate orientation is enforced in the active site, and what factors are responsible for chemical specificity in the reactions that are catalyzed by a given enzyme. While crystallography can provide a three-dimensional structure that is invaluable in finding active sites and identifying catalytically important residues, these structures are primarily static in nature. A photograph of a car engine, no matter how detailed, cannot provide much insight into what moves when the engine is running. We need to be able to watch enzymes at work in four dimensions in order to really understand them, and time is the fourth dimension. Nuclear magnetic resonance (NMR) can provide that fourth dimension, allowing us to watch changes in enzyme structure and dynamics in response to changes in substrate, cofactor and mutation.
Imagine you are a plumber fixing pipes in an old wooden house. When it comes time to solder the pipes, you need to be very careful not to burn down the house with your torch. Nature’s problem is similar. We are surrounded by oxygen, a very reactive and corrosive substance. The only thing standing between us and spontaneous combustion is the fact that O2 exists normally in a triplet state (that is, with unpaired electrons) while most of the chemical bonds making up our bodies are singlet (all electrons paired), and don’t react easily with triplet species.
When metabolic processes require a reaction with O2, it must be activated in order to accomplish the reaction. Cytochromes P450 do this job efficiently, and catalyze many important reactions in our bodies, including steroid hormone and prostaglandin biosyntheses, as well as processing and metabolizing drugs. For that matter, most antibiotics are biosynthesized at least in part by P450 enzymes, and many of the microorganisms that we use to clean up toxic wastes and oil spills rely on P450 enzymes to do their work. There are over 20,000 sequenced P450 genes in GenBank, covering all kingdoms and phyla of life. Yet these diverse sequences, with thousands of different substrate/product combinations, all fold to a common, highly conserved three-dimensional structure.
We are interested in how the P450 structure and dynamics are modulated in order to achieve the remarkable substrate selectivity and reaction specificity combined with high catalytic efficiency that many of these enzymes exhibit. Our laboratory uses structural methods, especially NMR, combined with computational methods, bioinformatics, classical chemistry, molecular biology and enzymology to answer basic questions about P450 function. We recently discovered a hidden “spring” mechanism that adjusts the size of the active site in response to the size of bound substrate (Figure 1). Many P450s require specific protein co-factors in order to turn over substrate. NMR methods have revealed that a series of complex changes take place in the P450 structure upon binding of the co-factor, and this has led us to propose a model for how these changes occur and why they are critical to enzymatic activity (Figures 2 and 3). Currently, we are focusing on the specific residues that impart substrate specificity and product selectivity on several P450 enzymes.
Figure 1
Figure 2
Figure 3
We are extending the methodology and expertise that we have developed to the important class of P450s that are involved in drug metabolism. CYP3A4 is a human liver P450 that metabolizes over 70% of currently marketed pharmaceuticals, and defects in function of CYP3A4 are important determinants of whether a person can tolerate a particular drug. With the advent of personal genomes, understanding how particular mutations affect CYP3A4 function are becoming critical. Because most mammalian P450s are membrane associated, they are difficult to work with using standard NMR methods. We are experimenting with a novel membrane "sushi" called nanodiscs, developed by Steve Sligar at the University of Illinois, that solubilize membrane proteins in a native-like environment (Figure 4), so that we can apply high-resolution solution NMR methods to these important enzymes.
Figure 4