Proton magnetic resonance studies of carbonic anhydrase. II. Group controlling catalytic activity

The seven resonances observed in the histidine region of the proton magnetic resonance (pmr) spectrum of human carbonic anhydrase B and reported in the preceding paper are studied in the presence of sulfonamide, azide, cyanide, and chloride inhibitors and in metal-free, cadmium substituted, cobalt substituted, and carboxymethylated forms of the enzyme. Results indicate that the two resonances that move-downfield with increasing pH and the two that do not move with pH reflect residues located at the active site. The first two resonances are assigned to the same titratable histidine whose pK value of 8.24 corresponds to that of the group controlling catalytic activity. Addition of anions or sulfonamides, removal of zinc, or substitution of cadmium for zinc at the active site, procedures known to abolish enzymatic activity, prevent titration of this residue. Partial inhibition of carbonic anhydrase by chloride slectively increases the pK value of the group controlling catalytic activity and of the histidine with pK equals 8.24. Experiments with metal-free and cadmium carbonic anhydrases and comparisons with model systems suggest that this histidine is bound to the metal ion at high pH; at low pH this complex appears to dissociate as protons compete with the metal for the imidazole group. It is proposed that ionization of the group controlling catalytic activity represents loss of the pyrrole proton of this neutral ligand when it binds to Zn(II), forming an imidazolate anion and juxtaposing a strong base and a powerful Lewis acid at the active site. When bound to zinc as an anion, this histidine can act as a general base catalyst in the hydration of carbon dioxide and be replaced as a metal ligand by an oxygen of the substrate in the course of the reaction. The histidine-metal complex is thought to exist in a strained configuration in the active enzyme so that its imidazole-metal bond is readily broken on addition of substrates or inhibitors. This model is consistent with the available data on the enzyme and is discussed in relation to alternative proposals.