Structural basis for substrate specificity

While it is axiomatic that enzymes are specific for their substrates, few general principles have emerged. Studies with aLP provide new insights into the role that coupled motions play in modulating substrate specificity. Previously we had used a combination of mutagenesis, functional and crystallographic analyses to probe the structural basis for substrate specificity. One particular class of mutations (typified by Met190->Ala) had the remarkable property of high activity and extremely broad specificity. Structural analyses demonstrated that the enzyme displays a high degree of plasticity in being able to adapt to substrate sidechains ranging from Ala to Phe. The structural distortions found with different enzyme-ligand complexes were suggestive of concerted motions around the active site. Further, mutations of a residue quite distant from the active site (I172) had a pleiotropic effect on specificity attributable to altered dynamics.

We have used a combination of x-ray crystallographic, solution NMR, and computational methods to explore the role of dynamics in specificity. At room temperature aLP is a rather rigid molecule having a mean B factor of 12-14. At cryogenic temperatures (120K), thermal motion is quenched and a large percentage of the atomic B factors approach 0. However, there are several regions where the B factors remain elevated. Since there is clearly negligible mosaicity, we postulated that these regions corresponded to areas where multiple conformational substates were trapped by cooling. Multi-conformation refinement (16 structures) using fixed B factors allowed us to explore the structural properties of these trapped substates. These regions correspond to the walls of the binding pocket. Solution dynamics using NMR confirmed that these regions were undergoing slow motions (Chemical exchange, msec - µsec) at room temperature. In solution, these motions and not the fast dynamics were quenched by ligand binding. Together, these data suggest that slow binding pocket motions pre-exist at room temperature and can be trapped at 120K.

Our hypothesis is that some of our mutations alter substrate specificity by changing binding pocket dynamics. This was examined via normal modes analysis (NMA) which seeks to calculate low frequency, concerted protein motions. NMA revealed that for the wild-type protein, the walls of the binding pocket were moving in parallel with one another - preserving the size binding pocket even though they were dynamic. By contrast, in the M190A broad specificity mutant, the walls of the binding pocket became uncoupled, allowing the binding pocket to sample smaller and larger configurations. Thus a single mutation could alter function by altering the dynamic properties of the molecule. Further, this suggests that evolution also seeks to optimize dynamical properties as a means to optimize function.

Relevant Publications

Wilson, C., Mace, J.E., and Agard, D.A. (1991). A computational method for the design of enzymes with altered substrate specificity. J. Mol. Biol. 220: 495-506. (pdf).

Wilson, C. and Agard, D.A. (1991). Engineering substrate specificity. Curr. Opinions Cell Biol. 1: 617-623. (pdf).

Bone, R. and Agard, D.A. (1991). Mutational remodeling enzyme specificity. Methods Enzymol. 202: 643-672. (pdf).

Bone, R., Frank, D., Kettner, C., and Agard, D.A. (1989). Structural analysis of specificity: α-lytic protease complexes with analogs of reaction intermediates. Biochem. 28: 7600-7609 (pdf).

Bone, R., Silen, J.L., and Agard, D.A. (1989). Structural plasticity broadens the specificity of an engineered protease. Nature 339: 191-195.(pdf).