Raman difference spectroscopic studies of dithiobenzoyl substrate and product analogs binding to the enzyme dehalogenase: π‐electron polarization is prevented by the CO to CS substitution

The enzyme 4‐chlorobenzoyl coenzyme A (CoA) dehalogenase catalyzes the transformation of 4‐chlorobenzoyl‐CoA to 4‐hydroxybenzoyl‐CoA. High‐quality Raman spectra of substrate or product or their analogs in the enzyme's active site can be obtained using Raman difference spectroscopy. Here, data are presented for the substrate and product analog, 4‐chlorobenzoyldithio‐CoA and 4‐hydroxybenzoyldithio‐CoA, respectively, where the benzoyl CO group has been replaced by CS. The rationale behind this substitution is to explore the consequences of perturbing enzyme active site–ligand CO interactions. Interpretation of the Raman data for the complexes was placed on a firm footing by undertaking density functional theory calculations on the model compounds S ‐ethyl 4‐chlorobenzoate dithioester and S ‐ethyl 4‐hydroxybenzoate dithioester. Based on the calculations and on data for the CoA moiety, essentially all the Raman bands in the spectra of the substrate and product dithio analogs could be assigned with confidence. The calculations show that the dithioester phenyl moiety is non‐planar with the CS bond typically being twisted 30° of the plane of the phenyl ring. The Raman difference data for the dehalogenase bound dithio‐based substrate and product show that only minor spectral changes occur upon binding and this indicates that the conformation about the dithioesterer bonds is essentially unchanged by binding. The results for the product analog are in strong contrast to those for the complex involving the natural product, 4‐hydroxybenzoyl‐CoA, where binding brings about a complete rearrangement of the benzoyl's normal modes. The reorganization is due to a marked polarization of the benzoyl's π‐;electrons, which, in turn, is brought about by electron pulling forces acting at the carbonyl and electron pushing forces near the benzoyl 4‐position in the active site. However, π‐electron polarization, and the accompanying normal mode rearrangement, do not happen for the dithio analog because the CS substituted group does not transmit the electron pulling forces, that act on the CO in the natural substrate, to the phenyl ring. The spectroscopic results explain why the dithio‐based substrate reacts with dehalogenase 200 times more slowly than the thiolester substrate. For the latter, electron pull at the CO is used to deplete electron density at the benzoyl's 4‐position and to facilitate the substitution of the 4‐Cl atom by the 4‐OH group. Copyright © 2000 John Wiley & Sons, Ltd.