Evolutionary engineering of a beta-Lactamase activity on a D-Ala D-Ala transpeptidase fold

The beta-Lactamase hydrolytic activity has arisen several times from DD-transpeptidases. We have been able to replicate the evolutionary process of beta-Lactamase activity emergence on a PBP2X DD-transpeptidase. Some of the most interesting changes, like modifying the catalytic properties of an enzyme, may require several mutations in concert; therefore it is essential to explore efficiently sequence space by generating the right diversity. We designed a biased combinatorial library in which biochemical and structural information were incorporated by site directed mutagenesis on relevant residues and then subjected to random mutagenesis to allow for mutations in unforeseen positions. We isolated mutants from this library conferring 10-fold higher cefotaxime resistance levels than the background wild-type through mutations exclusively in the coding sequence. We demonstrate that only three substitutions in the DD-transpeptidase active site, two produced by the directed and one by the random mutagenesis, are sufficient to acquire this activity. The purified product of one mutant (MutE) had a 10(5)-fold increase in cefotaxime deacylation rate allowing it to hydrolyze beta-Lactams yet it has apparently conserved DD-peptidase activity. This work is the first to show a possible evolutionary intermediate between a beta-Lactamase and a DD-transpeptidase necessary for the development of antibiotic resistance.