Promiscuity

V. Moliner

Departament de Química Física i Analítica, Universitat Jaume I, 12071 Castelló (Spain)

Traditional view on enzymatic activity usually remarks on the fact of their high efficiency and specificity. However, it has been recently suggested that this paradigm, that has dominated thinking in this field, could be too simplistic. Many enzymes have been found to present secondary catalytic activity, or catalytic promiscuity, thus being capable of catalyzing secondary reactions at an active site that was specialized to favour a primary reaction. This promiscuity provides a raw starting point for the evolution of enzymes, as a new duplicated gene presenting low activity would provide a start for adaptative evolution. According to previous studies, promiscuous activity exhibit high plasticity as they can be readily increased by means of one or few mutations, allowing reaching the threshold for being improved under selective pressure. Instead, primary activity presents a large robustness against mutations. As a consequence, the active sites of these existing enzymes provide obvious starting points to engineer novel enzymes with new catalytic functions.

The redesign of biological catalyst has not to be restricted to enzymes. In particular, almost three decades ago, immune-globulin proteins have been used to produce Catalytic Antibodies (CA). The immune response to a stable molecule that mimics as closely as possible the presumed structural and electronic features of the Transition State (TS) of a particular reaction, a Transition State Analogue (TSA), is the first step to obtain a CA. This new macromolecule will be able to catalyze the chemical reaction by its presumably affinity for the TS. Nevertheless, their catalytic efficiency still lags far behind those of natural enzymes. The molecular engineering employed to improve the catalytic efficiency of CA are thus similar to the ones applied to the redesign of enzymes.

The methods and techniques of Computational Chemistry have become a promising complementary tool to design biological catalysts. An alternative computational protocol is presented in this communication and, although containing some limitations, it sorts out the difficulties present in previous strategies. This is based on Molecular Dynamic simulations that, using hybrid quantum mechanics and molecular mechanics (QM/MM) methods, allows obtaining the TS of the chemical reaction in the presence of the protein environment. All the specific substrate-protein interactions that are established in the active site of the enzyme can be analyzed, both at the Michaelis complex and at the real TS, which is not necessary equal to the TSA. The knowledge of this pattern of interactions will provide clues to decide which residues of the active site of a CA or an enzyme should be replaced to better stabilize the TS relative to the Michaelis complex, which would presumably enhance the rate constant of the chemical step of a full catalytic process. Nevertheless, as all these peaces are part of a complex puzzle, it is not possible to predict that a single mutation was going to produce a rate enhancement of the chemical reaction without side part effects. In this regard the free energy profile can be traced from reactants to products, via the corresponding TS, rending theoretical predicted barriers comparable with experimental data and allowing testing directed mutations. This theoretical strategy can be considered as a numerical experiment helping experimentalists in active-site redesign by means of a few mutations. Some efforts in this direction will be presented.