Impact of specific pbp5 mutations on expression of beta-lactam resistance in Enterococcus faecium

Abstract

We tested the impact of individual PBP 5 mutations on expression of ampicillin resistance in Enterococcus faecium using a shuttle plasmid designed to facilitate expression of cloned pbp5 in ampicillin-susceptible E. faecium D344SRF. Substitutions that had been implicated in contributing to the resistance of clinical strains conferred only modest levels of resistance when they were present as single point mutations. The levels of resistance were amplified when some mutations were present in combination. In particular, a methionine-to-alanine change at position 485 (in close proximity to the active site) combined with the insertion of a serine at position 466 (located in a loop that forms the outer edge of the active site) was associated with the highest levels of resistance to all beta-lactams. Affinity for penicillin generally correlated with beta-lactam MICs for the mutants, but these associations were not strictly proportional.

FIG. 1.

Cloning strategy to create mutants. pbp5 cloned by itself expresses resistance poorly (A). However, when it is cloned downstream of ftsW_Ef and psr from strain C68, the level of resistance expression is higher (B). So we first created a plasmid in which ftsW_Ef and psr were cloned into the shuttle vector pTCV-lac using restriction enzymes SmaI and BamHI, with the BamHI site of the insert located downstream of psr (C). We then separately amplified pbp5 from E. faecium D366 (susceptible strain) from the ribosome-binding site to downstream of the transcription terminator (with BamHI sites engineered on both ends) and cloned this into the commercially available E. coli plasmid vector pCR-XL-TOPO (Invitrogen) (D). Site-directed mutagenesis was performed with pbp5 in this plasmid. After sequence confirmation, the insert was removed by digestion with BamHI and inserted downstream of psr in the pTCV-lac construct, resulting in several variants (B) in which mutant pbp genes were located downstream of ftsW_Ef and psr. The proper orientation was confirmed by PCR amplification.

FIG. 2.

Affinities of different PBP 5 mutants for binding to penicillin. Different PBP 5 mutants with their membrane anchors deleted were produced in E. coli after truncated versions of pbp5 genes amplified from the different pCWR plasmids listed in the figure were cloned in pMC56/RBSII (Table 1). The amino acids at the indicated positions within PBP 5 are shown; + and −, presence and absence of Ser at position 466′, respectively; Penicillin (μg/ml), the concentrations of penicillin used to saturate the different PBP 5 proteins; S₅₀, the concentration of penicillin needed to saturate 50% of PBP 5.

FIG. 3.

Structure of the E. faecium PBP 5. The α-carbon atom trace of the transpeptidase domain of PBP 5 complexed with benzylpenicillin (arrow) (; Kostrewa et al., unpublished), prepared using MOLOC (6), is shown. The benzylpenicillin moiety of the acyl enzyme is shown in a small ball-and-stick representation, and the side chains of the mutated residues are highlighted by larger ball-and-stick representations.