Cryptic carbapenem antibiotic production genes are widespread in Erwinia carotovora: facile trans activation by the carR transcriptional regulator

Abstract

Few strains of Erwinia carotovora subsp. carotovora (Ecc) make carbapenem antibiotics. Strain GS101 makes the basic carbapenem molecule, 1-carbapen-2-em-3-carboxylic acid (Car). The production of this antibiotic has been shown to be cell density dependent, requiring the accumulation of the small diffusible molecule N-(3-oxohexanoyl)-L-homoserine lactone (OHHL) in the growth medium. When the concentration of this inducer rises above a threshold level, OHHL is proposed to interact with the transcriptional activator of the carbapenem cluster (CarR) and induce carbapenem biosynthesis. The introduction of the GS101 carR gene into an Ecc strain (SCRI 193) which is naturally carbapenem-negative resulted in the production of Car. This suggested that strain SCRI 193 contained functional cryptic carbapenem biosynthetic genes, but lacked a functional carR homologue. The distribution of trans-activatable antibiotic genes was assayed in Erwinia strains from a culture collection and was found to be common in a large proportion of Ecc strains. Significantly, amongst the Ecc strains identified, a larger proportion contained trans-activatable cryptic genes than produced antibiotics constitutively. Southern hybridization of the chromosomal DNA of cryptic Ecc strains confirmed the presence of both the car biosynthetic cluster and the regulatory genes. Identification of homologues of the transcriptional activator carR suggests that the cause of the silencing of the carbapenem biosynthetic cluster in these strains is not the deletion of carR. In an attempt to identify the cause of the silencing in the Ecc strain SCRI 193 the carR homologue from this strain was cloned and sequenced. The SCRI 193 CarR homologue was 94% identical to the GS101 CarR and contained 14 amino acid substitutions. Both homologues could be expressed from their native promoters and ribosome-binding sites using an in vitro prokaryotic transcription and translation assay, and when the SCRI 193 carR homologue was cloned in multicopy plasmids and reintroduced into SCRI 193, antibiotic production was observed. This suggested that the mutation causing the silencing of the biosynthetic cluster in SCRI 193 was leaky and the cryptic Car phenotype could be suppressed by multiple copies of the apparently mutant transcriptional activator.