| Popis: |
In this chapter, the application of quantum chemical methodsfor screening of enzyme activity is discussed. Using a linearinterpolation approach, where the enzyme structure is divided intodiscrete structural frames along the reaction coordinate of the rate-determining step and the wave function derived semi-empiricalenergies of each frame are evaluated to provide an approximatereaction profile, the amide hydrolysis activity in Candida antarctica lipase B (CalB) and glycosidic bond hydrolysis activity in Bacillus circulans xylanase (Bcx) and derived mutants is screened. The application of this approach to a programmatically generatedlibrary of mutant structures allows ranking and identification ofmutants with likely increased activity toward an artificial substrate.Experimental verification of the CalB results yields a prediction ratewhere 15 of 22 mutants are correctly characterized in terms ofboth increased and decreased activity. The screening of Bcx mutantsallows identification of structural patterns among the most activemutants providing a new hypothesis for rationalizing the catalytic-enhancing determinants of the Bcx active site. At the time of writing,the presented applications are the largest quantum chemical screensof enzyme catalysts and present a valuable tool for computationalenzyme engineering. |
