Mutation Report for: E197Q_human-BCHE

The peripheral anionic site (PAS) of human butyrylcholinesterase is involved in the mechanism of substrate activation by positively charged substrates and ligands. Two substrate binding loci, D70 in the PAS and W82 in the active site, are connected by the Omega loop. To determine whether the Omega loop plays a role in the signal transduction between the PAS and the active site, residues involved in stabilization of the loop, N83, K339 and W430, were mutated. Mutations N83A and N83Q caused loss of substrate activation, suggesting that N83 which interacts with the D70 backbone may be an element of the transducing system. The K339M and W430A mutant enzymes retained substrate activation. Residues W82, E197, and A328 in the active site gorge have been reported to be involved in substrate activation. At butyrylthiocholine concentrations greater then 2 mM, W82A showed apparent substrate activation. Mutations E197Q and E197G strongly reduced substrate activation, while mutation E197D caused a moderate effect, suggesting that the carboxylate of residue E197 is involved in substrate activation. Mutations A328F and A328Y showed no substrate activation, whereas A328G retained substrate activation. Substrate activation can result from an allosteric effect due to binding of the second substrate molecule on the PAS. Mutation W430A was of special interest because this residue hydrogen bonds to W82 and Y332. W430A had strongly reduced affinity for tetramethylammonium. The bimolecular rate constant for reaction with diisopropyl fluorophosphate was reduced 10000-fold, indicating severe alteration in the binding area in W430A. The kcat values for butyrylthiocholine, o-nitrophenyl butyrate, and succinyldithiocholine were lower. This suggested that the mutation had caused misfolding of the active site gorge without altering the Omega loop conformation/dynamics. W430 as well as W231 and W82 appear to form the wall of the active site gorge. Mutation of any of these tryptophans disrupts the architecture of the active site.

Because of deficiencies in the present treatments for organophosphorus anticholinesterase poisoning, we are attempting to develop a catalytic scavenger that can be administered as prophylactic protection. Currently known enzymes are inadequate for this purpose because they have weak binding and slow turnover, so we are trying to make an appropriate enzyme by protein engineering techniques. One butyrylcholinesterase mutant, G117H, has the desired type of activity but reacts much too slowly. This communication describes an attempt to determine the reason for the slow reaction so that a more efficient enzyme might be designed. The results indicate that the mutation at residue 117 has resulted in a distortion of the transition state of the reaction of organophosphorus compounds with the active site serine. This information will be used to develop other mutants that avoid transition state stabilization sites.

Organophosphorus acid anhydride (OP) "nerve agents" are rapid, stoichiometric, and essentially irreversible inhibitors of serine hydrolases. By placing a His near the oxyanion hole of human butyrylcholinesterase (BChE), we made an esterase (G117H) that catalyzed the hydrolysis of several OP, including sarin and VX [Millard et al. (1995) Biochemistry 34, 15925-15930]. G117H was limited, however, because it was irreversibly inhibited by pinacolyl methylphosphonofluoridate (soman); soman is among the most toxic synthetic poisons known. This limitation of G117H has been overcome by a new BChE (G117H/E197Q) that combines two engineered features: spontaneous dephosphonylation and slow aging (dealkylation). G117H/E197Q was compared with the single mutants BChE G117H and E197Q. Each retained cholinesterase activity with butyrylthiocholine as substrate, although kcat/Km decreased 11-, 11- or 110-fold for purified G117H, E197Q, or G117H/E197Q, respectively, as compared with wild-type BChE. Only G117H/E197Q catalyzed soman hydrolysis; all four soman stereoisomers as well as sarin and VX were substrates. Phosphonylation and dephosphonylation reactions were stereospecific. Double mutant thermodynamic cycles suggested that the effects of the His and Gln substitutions on phosphonylation were additive for PSCR or PRCR soman, but were cooperative for the PSCS stereoisomer. Dephosphonylation limited overall OP hydrolysis with apparent rate constants of 0.006, 0.077, and 0.128 min-1 for the PR/SCR, PSCS, and PRCS soman stereoisomers, respectively, at pH 7.5, 25 degrees C. We conclude that synergistic protein design converted an archetypal "irreversible inhibitor" into a slow substrate for the target enzyme.

Organophosphate-inhibited cholinesterases can be reactivated by nucleophilic compounds. Sometimes phosphylated (phosphorylated or phosphonylated) cholinesterases become progressively refractory to reactivation; this can result from different reactions. The most frequent process, termed 'aging', involves the dealkylation of an alkoxy group on the phosphyl moiety through a carbocation mechanism. In attempting to determine the amino acid residues involved in the aging of butyrylcholinesterase (BuChE), the human BuChE gene was mutated at several positions corresponding to residues located at the rim of the active site gorge and in the vicinity of the active site. Mutant enzymes were expressed in Chinese hamster ovary cells. Wild-type BuChE and mutants were inhibited by di-isopropylfluorophosphate at pH 8.0 and 25 degrees C. Di-isopropyl-phosphorylated enzymes were incubated with the nucleophilic oxime 2-pyridine aldoxime methiodide and their reactivatability was determined. Reactivatability was expressed by the first-order rate constant of aging and/or the half-life of aging (t12). The t12 was found to be of the order of 60 min for wild-type BuChE. Mutations on Glu-197 increased t12 60-fold. Mutation W82A increased t12 13-fold. Mutation D70G increased t12 8-fold. Mutations in the vicinity of the active site serine residue had either moderate or no effect on aging; t12 was doubled for F329C and F329A, increased only 4-fold for the double mutant A328G+F329S, and no change was observed for the A328G mutant, indicating that the isopropoxy chain to be dealkylated does not directly interact with Ala-328 and Phe-329. These results were interpreted by molecular modelling of di-isopropylphosphorylated wild-type and mutant enzymes. Molecular dynamics simulations indicated that the isopropyl chain that is lost interacted with Trp-82, suggesting that Trp-82 has a role in stabilizing the carbonium ion that is released in the dealkylation step. This study emphasized the important role of the Glu-197 carboxylate in stabilizing the developing carbocation, and the allosteric control of the dealkylation reaction by Asp-70. Indeed, although Asp-70 does not interact with the phosphoryl moiety, mutation D70G affects the rate of aging. This indirect control was interpreted in terms of change in the conformational state of Trp-82 owing to internal motions of the Omega loop (Cys-65-Cys-92) in the mutant enzyme.