A major part of the drugs manufactured today contains enantiopure molecules [8]; hence, highly enantioselective reactions for the production of enantiopure building blocks are of great industrial importance. Enzymes are an attractive class of catalysts often used in the synthesis of enantiopure compounds. They usually exhibit high enantioselectivity, operate under mild reaction conditions and have a large substrate scope [9]. However, with substrates that are very different from the natural substrate, enzymes can display low enantioselectivity and/or poor reactivity. A solution to this problem is to genetically modify the enzyme and thereby increase the substrate acceptance or enantioselectivity.
Designing enantioselectivity of enzymes is one of the most attractive but challenging trials in the field of protein engineering for synthesis of enantiometically pure compounds, there is, however, no practical theory for introducing mutations into any enzyme to change its enantioselectivity. One way would be to modelsubstratelipasesandesterasesbysubstrate-imprinteddocking, which takes into account the substrate transition states, productive and non-productive hydrogen bonds as well as com-