Probes of hydrogen tunneling with horse liver alcohol dehydrogenase at subzero temperatures

The temperature dependence of steady-state kinetics has been studied with horse liver alcohol dehydrogenase (HLADH) using protonated and deuterated benzyl alcohol as substrates in methanol/water mixtures between +3 and -50 degrees C. Additionally, the competitive isotope effects, k(H)/k(T) and k(D)/k(T), were measured. The studies indicate increasing kinetic complexity for wild-type HLADH at subzero temperatures. Consistent with earlier findings at 25 degrees C [Bahnson et al. (1993) Biochemistry 31, 5503], the F93W mutant shows much less kinetic complexity than the wild-type enzyme between 3 and -35 degrees C. An analysis of noncompetitive deuterium isotope effects and competitive tritium isotope effects leads to the conclusion that the reaction of F93W involves substantial hydrogen tunneling down to -35 degrees C. The effect of methanol on kinetic properties for the F93W mutant was analyzed, showing a dependence of competitive KIEs on the NAD(+) concentration. This indicates a more random bi--bi kinetic mechanism, in comparison to an ordered bi-bi kinetic mechanism in water. Although MeOH also affects the magnitude of the reaction rates and, to some extent, the observed KIEs, the ratio of ln k(H)/k(T) to ln k(D)/k(T) for primary isotope effects has not changed in methanol, and we conclude little or no change in kinetic complexity. Importantly, the degree of tunneling, as shown from the relationship between the secondary k(H)/k(T) and k(D)/k(T) values, is the same in water and MeOH/water mixtures, implicating similar trajectories for H transfer in both solvents. In a recent study of a thermophilic alcohol dehydrogenase [Kohen et al. (1999) Nature 399, 496], it was shown that decreases in temperatures below a transition temperature lead to decreased tunneling. This arises because of a change in protein dynamics below a break point in enzyme activity [Kohen et al. (2000) J. Am. Chem. Soc. 122, 10738-10739]. For the mesophilic HLADH described herein, an opposite trend is observed in which tunneling increases at subzero temperatures. These differences are attributed to inherent differences in tunneling probabilities between 0 and 100 degrees C vs subzero temperatures, as opposed to fundamental differences in protein structure for enzymes from mesophilic vs thermophilic sources. We propose that future investigations of the relationship between protein flexibility and hydrogen tunneling are best approached using enzymes from thermophilic sources.