Identification, characterization, and regulation of a cluster of genes involved in carbapenem biosynthesis in Photorhabdus luminescens

Abstract

The luminescent entomopathogenic bacterium Photorhabdus luminescens produces several yet-uncharacterized broad-spectrum antibiotics. We report the identification and characterization of a cluster of eight genes (named cpmA to cpmH) responsible for the production of a carbapenem-like antibiotic in strain TT01 of P. luminescens. The cpm cluster differs in several crucial aspects from other car operons. The level of cpm mRNA peaks during exponential phase and is regulated by a Rap/Hor homolog identified in the P. luminescens genome. Marker-exchange mutagenesis of this gene in the entomopathogen decreased antibiotic production. The luxS-like signaling mechanism of quorum sensing also plays a role in the regulation of the cpm operon. Indeed, luxS, which is involved in the production of a newly identified autoinducer, is responsible for repression of cpm gene expression at the end of the exponential growth phase. The importance of this carbapenem production in the ecology of P. luminescens is discussed.

FIG. 1.

Comparison of the putative P. luminescens cpm operon with the car operon of E. carotovora and Serratia sp. strain ATCC 39006. (A) Map of the genes involved in carbapenem biosynthesis. (B) Amino acid sequence identity between Car and Cpm proteins of E. carotovora, Serratia sp., and P. luminescens.

FIG. 2.

Carbapenem-like activity of P. luminescens visualized by antibiosis. Three-day LB plates spotted with 5 μl of both TT01 (wild type) and mutants were inoculated with 100 μl of various β-lactam indicator strain cultures (OD₆₀₀ = 0.2) mixed in soft agar. Growth inhibition around a spot indicates production of antibiotics to which the indicator strain is sensitive. All experiments were repeated at least three times. (A) Analysis of carbapenem production and the effect of a cpmA mutation (PL2101) by antibiosis of lawns of various β-lactam indicator strains. (B) Analysis of the effect of various mutations on the ability to produce carbapenem by antibiosis of lawns of Enterobacter cloacae strains C1 (cephalosporinase producer) and C1R (cephalosporinase derepressed). 1, cpmA mutation (PL2101); 2, luxS mutation (PL2102); 3, slyA mutation (PL2103).

FIG. 3.

(A) Nucleotide sequence of the 5′ region of cpmA from P. luminescens. Promoter sequences (−35 and −10 boxes) are boxed. The Shine-Dalgarno sequence (SD) is underlined. Only the first codons of the coding sequence of cpmA are shown in boldface. (B) Primer extension analysis of cpmA transcripts in P. luminescens. Total RNA was extracted from an exponential culture of P. luminescens TT01 grown at 30°C in Schneider medium (OD₆₀₀ = 1.4). The arrow indicates the position of the extension product obtained.

FIG. 4.

(A) Primer extension analysis of cpm mRNA at various times during P. luminescens growth at 30°C in Schneider medium. Lanes: 1, early log growth (OD₆₀₀ = 0.3); 2, log growth (OD₆₀₀ = 1.0); 3, early stationary phase (OD₆₀₀ = 4.0). Equal amounts of each RNA sample (50 μg) were used. Experiments were done in triplicate. The relative cpm mRNA abundances at different times during P. luminescens growth, with the lowest mRNA level taken as 1, are represented graphically. (B) P. luminescens growth curve indicating the different times when mRNAs were extracted. 1, early log growth (OD₆₀₀ = 0.3); 2, log growth (OD₆₀₀ = 1.0); 3, early stationary phase (OD₆₀₀ = 4.0). Since P. luminescens synthesizes a variety of secondary metabolites which interfere with OD measurement at late exponential phase, correlations between CFU and OD measurements were used to define the various growth stages.

FIG. 5.

P. luminescens production of a luxS-dependent AI-2 activity. (A) AI-2 bioassay of P. luminescens culture supernatants with V. harveyi BB170 as reporter. Ten percent cell-free supernatants were mixed with the reporter strain and incubated at 30°C on a rotary shaker. Light production was determined 4 h 30 min after the BB170 cells were mixed with the various cell-free supernatants (OD₆₀₀ = 0.1). Cell-free culture fluids were prepared from P. luminescens parental (TT01) and luxS mutant (PL2102) strains grown to an OD₆₀₀ of 1.0. Schneider sterile medium and V. harveyi BB152 cell-free culture fluid were used as negative and positive controls, respectively. (B) Inhibition of growth by P. luminescens cell-free supernatants. Cell-free supernatants were prepared from a wild-type P. luminescens strain grown in Schneider medium for 5 h (exponential phase; OD₆₀₀ = 1.0) (open circles), 10 h (early stationary phase; OD₆₀₀ = 4.0) (open triangles), and 15 h (stationary phase) (open diamonds). Ten percent (vol/vol) cell-free supernatants (open symbols) or sterile Schneider medium (filled squares) was added to the V. harveyi BB170 reporter strain, and cultures were incubated at 30°C on a rotary shaker.

FIG. 6.

Analysis of the effect of luxS and slyA mutations on growth rate and cpm mRNA synthesis. (A) Growth curve comparison among TT01 (filled squares), PL2102 (luxS mutant) (open triangles), and PL2103 (slyA mutant) (open circles) strains of P. luminescens. (B) Primer extension analysis of the effect of luxS mutation on cpm mRNA synthesis. RNA samples were taken at different times from cultures of TT01 and PL2102. Lanes: 1, log growth (OD₆₀₀ = 1.0); 2, early stationary phase (OD₆₀₀ = 4.0). (C) Primer extension analysis of the effect of slyA mutation on cpm mRNA synthesis. RNA samples were taken from exponential-growth-phase cultures of TT01 and PL2103 (OD₆₀₀ = 0.9). Three independent primer extension analyses were performed. Equal amounts of each RNA sample (50 μg) were used. The relative cpm mRNA abundances measured in TT01 and mutants, with the lowest mRNA level taken as 1, are represented graphically. Error bars represent standard errors of the mean. Statistical analysis was performed with a Student t test, resulting in a significant difference (P < 0.05).