Importance of the region around lysine 196 for catalytic activity of adenylyl cyclase from Escherichia coli.

Abstract

Escherichia coli adenylyl cyclase contains no sequence that corresponds to the previously defined ATP/GTP binding consensus (A,G)XXXXGK(S,T). Using a search for lysine residues located adjacent to glycine residues, three regions that were possible candidates for part of the ATP binding site were identified. These were the residues located at positions 59, 90, and 196. A plasmid vector capable of overexpressing the cya gene under the control of the lambda PL promoter was mutated at these three loci to convert those lysine residues to methionine. Assays for catalytic activity of the mutated hyperexpressed proteins revealed that only the mutation at position 196 led to loss of activity. Photoaffinity labeling experiments using 8-azido-ATP provided evidence that the loss of activity was associated with a loss of the capability of the enzyme to bind ATP. A further series of replacement mutations in the hyperexpression vector was created at position 196. Assays of the adenylyl cyclase activity of the mutated proteins showed that replacement of lysine 196 by arginine led to minimal change in the activity. Replacements by histidine, glutamine, or glutamic acid resulted in approximately 10-20-fold reductions in the activity; replacements by methionine, isoleucine, or aspartic acid resulted in total loss of activity. When the mutated forms of the cya gene were expressed under the control of the cya promoter, the activity of the wild-type protein was higher than that of all the mutants, including the arginine replacement mutant. All of the mutants that retained activity also retained the capability of adenylyl cyclase to be stimulated by either inorganic orthophosphate or GTP. A helical wheel analysis of the region of adenylyl cyclase around lysine 196 revealed a structure compatible with an amphipathic helix with one face enriched with basic amino acid residues. Assays for adenylyl cyclase activity of a series of replacement mutations of residues on the hydrophilic face of the helix (R188I, R192I, G195I) as well as on the hydrophobic face (R197I) indicated that the R188I, G195I, and K196I replacement mutants were inactive, and R192I was approximately 30% as active as the wild-type, while the R197I mutant was equivalent to the wild-type control. A model is suggested for a unique binding motif in E. coli adenylyl cyclase in which there is a repetition of 3 basic residues on one face of a helix where there is an interaction with the three phosphate groups of ATP.