Recovery of active beta-lactamases from Proteus vulgaris and RTEM-1 hybrid by random mutagenesis by using a dnaQ strain of Escherichia coli

Abstract

Proteus vulgaris and RTEM-1 beta-lactamases that belong to molecular class A with 37% amino acid similarity were examined to find the relationship between amino acid residues and activity of enzymes. MICs of ampicillin were > 2,000 micrograms/ml for Escherichia coli cells producing these enzymes. We have made 18 hybrid genes by substituting the coding region of the P. vulgaris beta-lactamase gene with the equivalent portions from the RTEM-1 gene. Most of these hybrids produced inactive proteins, but a few hybrid enzymes had partial or trace activity. From one of the hybrid genes (MIC of ampicillin, 100 micrograms/ml), we recovered three kinds of active mutants which provided ampicillin MICs of 1,000 micrograms/ml by the selection of spontaneous mutations in a dnaQ strain of E. coli. In these mutants, Leu-148, Met-182, and Tyr-274 were replaced with Val, Thr, and His, respectively. These amino acids have not been identified as residues with functional roles in substrate hydrolysis. Furthermore, from these hybrid mutants, we obtained a second set of mutants which conferred ampicillin MICs of 1,500 micrograms/ml. Interestingly, the second mutations were limited to these three amino acid substitutions. These amino acid residues which do not directly interact with substrates have an effect on enzyme activity. These mutant enzymes exhibited lower K(m) values for cephaloridine than both parental enzymes.

Activity and Thermostability of GH5 Endoglucanase Chimeras from Mesophilic and Thermophilic Parents

“…Recombination of the segments between related enzymes often results in hybrids with diminished activity. For example, Hosseini-Mazinani and coworkers created 18 hybrid genes by replacing the coding region of the P. vulgaris ␤-lactamase gene with the equivalent portions from the RTEM-1 gene (41). Most of these hybrids produced inactive proteins, and a few hybrid enzymes had partial or trace activity.…”

Extreme functional sensitivity to conservative amino acid changes on enzyme exteriors 1 1Edited by J. Karn

“…Although many previous studies show that functional hybrids can be constructed from natural homologues (Schneider et al, 1981;Mas et al, 1986;Houghton et al, 1989;Malcolm et al, 1990;Zu È lli et al, 1990;Olsen et al, 1991;Vos et al, 1991;Guez-Ivanier et al, 1993;Serrano et al, 1993;Vuilleumier & Fersht, 1994;Jermann et al, 1995;Singh & Hayashi, 1995;Hosseini-Mazinani et al, 1996;Nixon et al, 1997;Crameri et al, 1998;Ostermeier et al, 1999;Lehmann et al, 2000), none suggest residue-for-residue interchangeability among sequences that differ at a third or more of their positions. One or more of three factors can account for the ability to produce active hybrids in each case.…”

“…This appears to account for the functionality of hybrid enzymes designed to replicate ancestral sequences (Jermann et al, 1995) or to match a consensus sequence (Lehmann et al, 2000) since these hybrids share 80-90 % identity with extant wild-type sequences and probably signi®cantly higher identity with ancestral sequences. Second, in most cases hybrids are constructed by splicing contiguous stretches from parent sequences (Schneider et al, 1981;Mas et al, 1986;Houghton et al, 1989;Zu È lli et al, 1990;Olsen et al, 1991;Vos et al, 1991;Guez-Ivanier et al, 1993;Vuilleumier & Fersht, 1994;Singh & Hayashi, 1995;Hosseini-Mazinani et al, 1996;Nixon et al, 1997;Crameri et al, 1998;Ostermeier et al, 1999). If, as proposed here, divergent wildtype sequences become distinct designs with their own sets of optimising features, one would clearly have improved chances of building a functional hybrid if contiguous stretches of co-optimised residues were retained.…”

“…If, as proposed here, divergent wildtype sequences become distinct designs with their own sets of optimising features, one would clearly have improved chances of building a functional hybrid if contiguous stretches of co-optimised residues were retained. Finally, some hybrids have been subjected to random mutagenesis and selection to produce functional variants and weed out non-functional ones (Hosseini-Mazinani et al, 1996;Crameri et al, 1998).…”