The question of whether protein motions play a role in the chemical step of enzymatic catalysis has generated much controversy in recent years. Debate has recently reignited over possible dynamic contributions to catalysis in dihydrofolate reductase, following conflicting conclusions from studies of the N23PP/S148A variant of the Escherichia coli enzyme. By investigating the temperature dependence of kinetic isotope effects, we present evidence that the reduction in the hydride transfer rate constants in this variant is not a direct result of impairment of conformational fluctuations. Instead, the conformational state of the enzyme immediately before hydride transfer, which determines the electrostatic environment of the active site, affects the rate constant for the reaction. Although protein motions are clearly important for binding and release of substrates and products, there appears to be no detectable dynamic coupling of protein motions to the hydride transfer step itself.

Dihydrofolate reductase has long been used as a model system to study the coupling of protein motions to enzymatic hydride transfer. By studying environmental effects on hydride transfer in dihydrofolate reductase (DHFR) from the cold-adapted bacterium Moritella profunda (MpDHFR) and comparing the flexibility of this enzyme to that of DHFR from Escherichia coli (EcDHFR), we demonstrate that factors that affect large-scale (i.e., long-range, but not necessarily large amplitude) protein motions have no effect on the kinetic isotope effect on hydride transfer or its temperature dependence, although the rates of the catalyzed reaction are affected. Hydrogen/deuterium exchange studies by NMR-spectroscopy show that MpDHFR is a more flexible enzyme than EcDHFR. NMR experiments with EcDHFR in the presence of cosolvents suggest differences in the conformational ensemble of the enzyme. The fact that enzymes from different environmental niches and with different flexibilities display the same behavior of the kinetic isotope effect on hydride transfer strongly suggests that, while protein motions are important to generate the reaction ready conformation, an optimal conformation with the correct electrostatics and geometry for the reaction to occur, they do not influence the nature of the chemical step itself; large-scale motions do not couple directly to hydride transfer proper in DHFR.

The influence of temperature and pH on the stability and catalytic activity of dihydrofolate reductase (MpDHFR) from the cold-adapted deep-sea bacterium Moritella profunda was studied. The thermal melting temperature was found to be ∼38 °C and was not affected by pH, while activity measurements demonstrated that its stability was maximal at pH 7 and was reduced dramatically below pH 6 or above pH 8. The steady-state rate constant (k(cat)) was maximal at neutral pH and higher temperatures, while the Michaelis constants (K(M)) for both substrate and cofactor were optimal at lower temperatures and at elevated or reduced pH. For both temperature and pH, any change in k(cat) was therefore offset by a similar change in K(M). Both the activation enthalpy and entropy of the MpDHFR-catalysed reaction were lower than those of DHFR from E. coli leading overall to a very small difference in activation free energy and therefore similar steady-state rate constants at the same temperature. The chemical step of the reaction is not rate limiting at pH 7, but becomes progressively more rate limiting as the pH increases. These results demonstrate adaptation of MpDHFR to its environment and show compromises between enthalpic and entropic contributions to the reaction, and between k(cat) and K(M).

No abstract

Protein isotope labeling is a powerful technique to probe functionally important motions in enzyme catalysis and can be applied to investigate the conformational dynamics of proteins.