Strategy for Computational Enzymology

Prediction of the three-dimensional structure of proteins remains a major goal in computational biology and chemistry. Typical proteins are simply way too large to be treated with any ab initio or DFT method. Even with a force field, the conformational space is often too large to be fully sampled. Furthermore, the inherent 

Sinee optimization of the structure of the protein from seratch is so difficult, most computational enzymology studies begin with a known protein structure, typically from an X-ray crystallography experiment. In an ideal situation, one might hope to obtain the structure of the protein bound with its substrate, even more ideally, obtain the structure of the TS within the enzyme active site. However, since enzymes are efficient catalysts, it is impossible to isolate either of these states. Rather, the next best thing is to obtain the X-ray structure of the enzyme bound with an inhibitor. This structure can then be used as the starting point for a computational examination. 

With modem computational techniques and computers, the QM region is typically evaluated with some DFT method. As with any decision to study an organic reaction, care must be taken while choosing a functional appropriate for the reaction at hand. Similarly, one needs to carefully select a basis set that provides sufficient flexibility while keeping an eye on the basis set size in order to keep the computations from becoming too long. 

There are a variety of force fields that have been designed for protein modeling. The most popular are the Amber and Charmnp- variants. Work is ever ongoing in the development of more accurate force fields. A discussion of force fields is outside the scope of this work, and the interested reader is directed toward other sources for further information.  

Computation of the reaction potential energy surface requires identification of reactant, TS, and product. Techniques that were discussed in Chapter 1 (and throughout the book) can be applied to enzyme/substrate systems too. Given the large number of atoms in proteins and, therefore, the large number of coordinates and gradients that must be minimized, other approaches are often more suitable. 

---