The strength of dehalogenase-substrate hydrogen bonding correlates with the rate of Meisenheimer intermediate formation.

4-Chlorobenzoyl-coenzyme A (4-CBA-CoA) dehalogenase catalyzes the hydrolytic dehalogenation of 4-CBA-CoA to 4-hydroxybenzoyl-CoA by using an active site aspartate as the nucleophile. Formation of the corresponding Meisenheimer complex (EMc) is followed by chloride ion expulsion which forms the arylated intermediate (EAr). This is then hydrolyzed to the product. In this paper, we explore the relationship between active site polarizing forces acting on the benzoyl carbonyl and the rate of formation of the Meisenheimer complex. The polarizing forces at the C[double bond]O group were modulated by introducing site-selected mutations (A112V, Y65D, G113A, G113S, G113N, and F64P), near the C[double bond]O binding site. Using either the substrate, 4-CBA-CoA, or the substrate analogue, 4-methylbenzoyl-CoA (4-MBA-CoA), Raman difference spectroscopy provided the position of the C[double bond]O stretching frequency (nu(C)[double bond](O)) for a total of 10 enzyme-ligand complexes. In turn, the values of the C[double bond]O frequencies could be converted to differences in effective hydrogen bonding strengths between members of the series, based on earlier model studies [Clarkson, J., Tonge, P. J., Taylor, K. L., Dunaway-Mariano, D., and Carey, P. (1997) Biochemistry 36, 10192-10199]. Catalysis in the F64P, G113A, G113S, and G113N dehalogenase mutants was very slow with k(cat) values ranging from 8 x 10(-3) to 7.6 x 10(-6) s(-1). The EAr intermediate did not accumulate to a detectable level on these enzymes during a single turnover. Catalysis in the Y65D and A112V dehalogenase mutants were almost as efficient as catalysis in wild-type dehalogenase with k(cat) values of 0.1-0.6 s(-1). In wild-type dehalogenase, 22% of the bound substrate accumulated as the EAr intermediate during a single turnover (k(obs) for EAr formation = 24 s(-(1)); in the Y65D mutant, the level of accumulation is 17% (k(obs) for EAr formation = 3 s(-1)), and in the A112V mutant, the level is 23% (k(obs) for EAr formation = 17 s(-1)). The k(obs) for EAr formation in wild-type dehalogenase and the more active dehalogenase mutants (Y65D and A112V) was taken to be an estimate of the k for EMc formation, and the k(obs) for EP formation in a single turnover was taken to be an estimate of the k for EMc formation in the severely impaired mutants (F64P, G113A, G113S, and G113N). A plot of the log k(obs) for EMc formation versus the C[double bond]O stretching frequency of bound 4-CBA-CoA (or 4-MBA-CoA) is a straight line (R(2) = 0.9584). Throughout the series, nu(C)[double bond](O) varied by 61 cm(-1), corresponding to the change in hydrogen bonding enthalpy of 67 kJ/mol. The results show that changes in polarizing forces at the benzoyl carbonyl are transmitted to the benzoyl (4) position and correlate with the rate of aromatic nucleophilic addition five chemical bonds away. Interestingly, the relationship between effective polarizing forces and reactivity seen here for dehalogenase is similar to that reported for the addition-elimination reaction involving the hydrolysis of a series of acyl serine proteases.