Growth phenotypes of Pseudomonas aeruginosa lasR mutants adapted to the airways of cystic fibrosis patients

Abstract

The opportunistic pathogen Pseudomonas aeruginosa undergoes genetic change during chronic airway infection of cystic fibrosis (CF) patients. One common change is a mutation inactivating lasR, which encodes a transcriptional regulator that responds to a homoserine lactone signal to activate expression of acute virulence factors. Colonies of lasR mutants visibly accumulated the iridescent intercellular signal 4-hydroxy-2-heptylquinoline. Using this colony phenotype, we identified P. aeruginosa lasR mutants that emerged in the airway of a CF patient early during chronic infection, and during growth in the laboratory on a rich medium. The lasR loss-of-function mutations in these strains conferred a growth advantage with particular carbon and nitrogen sources, including amino acids, in part due to increased expression of the catabolic pathway regulator CbrB. This growth phenotype could contribute to selection of lasR mutants both on rich medium and within the CF airway, supporting a key role for bacterial metabolic adaptation during chronic infection. Inactivation of lasR also resulted in increased beta-lactamase activity that increased tolerance to ceftazidime, a widely used beta-lactam antibiotic. Loss of LasR function may represent a marker of an early stage in chronic infection of the CF airway with clinical implications for antibiotic resistance and disease progression.

Fig. 1

Inactivation of lasR confers a growth advantage with amino acids. A. Growth yields with phenylalanine (PHE) after 48 h of growth of engineered mutants with lasR inactivated (black shaded bars); spontaneous lasR mutants that emerged in vivo, or, in the case of ENV25L1, in vitro (grey shaded bars), and the parental strains with wild-type lasR (unshaded bars). Parentally related pairs of strains from individual CF patients are grouped together. Values are the average cfu of two cultures, and error bars show standard deviation. Cultures were inoculated with approximately 2 × 10⁵ cfu ml⁻¹ of cells grown with succinate. B. Growth yields of CF416, compared with the engineered lasR mutant CF416lasR::Gm, after growth for 48 h with phenylalanine (PHE), isoleucine (ILE) and tyrosine (TYR); and growth for 24 h with succinate (SUC) and succinate in the presence of fluorophenylalanine (SUC/F-PHE). Values are the average cfu of three to five cultures, and error bars show standard deviation. Cultures were inoculated with approximately 2 × 10⁵ cfu ml⁻¹ of cells grown on LB agar, except for cultures with succinate which were inoculated with succinate-grown cells. C. Growth with phenylalanine (PHE) of CF416 and CF1213R, and the associated isogenic lasR mutant strains CF416L1 and CF1213 respectively. Cultures were inoculated with cells grown on LB agar. Results are representative of at least three independent experiments.

Fig. 2

Inactivation of lasR results in visible accumulation of HHQ. A. Growth on LB agar, photographed after incubation overnight at 37°C followed by 7 days at 22°C (strain names are shown schematically above). The surface of the six lasR mutant colonies has a metallic sheen. B. Colonies of three strains growing on LB agar, photographed after incubation overnight at 37°C followed by 3 days at 22°C, before (Left) and after (Right) flooding with crystal violet stain (0.04%) covered and obscured CF416, and floated the lysed surface of colonies of the lasR mutants CF416L1 and CF1213. White bar = 5 mm. C. Analysis of HAQs extracted from two fractions of a lawn of cells of PA14 and PA14lasR::Gm, one released by flooding the lawn with liquid (floating material), and one by subsequently suspending the cells themselves (agar surface). D. HHQ (100 mg in 10 ml methanol) confers a metallic sheen (arrow) to a lawn of cells of a pqsApqsH mutant lacking HAQs and incapable of converting HHQ to PQS. Methanol alone had no effect (data not shown). E. Methyl anthranilate fumes generated a circular zone (arrow) in which the iridescent metallic sheen in a lawn of cells of CF416lasR::Gm was suppressed. Methyl anthranilate was added to a filter disk on the lid of an inverted Petri dish, and the agar surface was photographed after incubation overnight at 37°C followed by 3 days at 22°C.

Fig. 3

Colony phenotypes can be used to identify spontaneous lasR mutants that emerge in vitro. A. A sector (arrow), purified as spontaneous mutant strain CF416L10, emerging from a streak of CF416 cells whose growth on LB agar is inhibited (concave edge) by a perpendicular streak of CF416 cells. The photograph was taken after incubation overnight at 37°C followed by 5 days at 22°C. B. An iridescent sector (arrow) emerging from a colony of CF416 growing on LB agar, photographed after incubation overnight at 37°C followed by 3 weeks at 22°C. C. Colonies growing on LB agar, photographed after incubation overnight at 37°C followed by 3 days at 22°C. White bar = 5 mm. D. (Upper) Streaks of cells growing on protease indicator agar, photographed after incubation at 37°C for 24 h. The zones of clearing (dark halo) reflect digestion of skim milk in the agar; the centre of the CF416L1 streak exhibits lysis. (Lower) Twitching motility zones (below the agar) surrounding the central spot of surface growth, photographed after incubation for 3 days at 22°C. E. Growth of a horizontal streak of CF416 cells (lower arrow) creates a surrounding zone (upper arrow) in a lawn of CF416L1 cells where metallic sheen and autolysis is suppressed. The two strains were added to LB agar concurrently, and photographed after incubation overnight at 37°C.

Fig. 4

LasR amino acid residues altered by spontaneous missense mutations. The LasR domain organization is shown, and is based on comparison with TraR; delineated below is the extent of the primary dimerization domain and the DNA recognition helix, shown above are the wild-type amino acid residues altered in spontaneous mutants, those in bold being conserved in the LuxR-homologues LasR, RhlR and QscR from P. aeruginosa, and TraR from Agrobacterium tumefaciens (Chugani et al., 2001; Vannini et al., 2002; Zhang et al., 2002). Shown for comparison are the two residues (A67 and P117) altered in strains CF1323 and CF1213 (Table 1) respectively.

Fig. 5

Inactivation of lasR confers enhanced utilization of nitrogen sources that is partly dependent on the catabolic regulator CbrB. A. Growth yields with alanine as sole source of carbon and nitrogen (ALA[ C+N]) after 24 h of growth. Values are the average cfu of two to three cultures, and error bars show standard deviation. Cultures were inoculated with approximately 2 × 10⁵ cfu ml⁻¹ of cells grown with succinate. B. Growth yields with succinate/lactamide (SUC/LACT) after 24 h of growth. Values are the average cfu of three cultures, and error bars show standard deviation. Cultures were inoculated with approximately 2 × 10⁵ cfu ml⁻¹ of cells grown with succinate. C. Growth advantage with succinate/lactamide (SUC/LACT) medium displayed by the lasR mutant strains CF416L1, CF1642 and CF5296 as compared with the parental strain CF416. Cultures were inoculated with cells grown with succinate. Results are representative of at least three independent experiments. D. Real-time PCR analysis of transcription of the cbrB gene in the lasR mutants CF5296 and CF416lasR::Gm as compared with the parental strain CF416. Values are the averages of two independent experiments, and error bars show standard deviation. E. Growth yields with succinate/lactamide (SUC/LACT) after 24 h of growth of strains with wild-type lasR carrying a plasmid expressing cbrB (pCbrB) or the empty vector (pVect). Values are the average cfu of three cultures, and error bars show standard deviation. Cultures were inoculated with approximately 2 × 10⁵ cfu ml⁻¹ of cells grown on LB agar. Plasmids were maintained in liquid cultures with 50 mg ml⁻¹ gentamicin (for isolate 3-0.8) or 150 µg ml⁻¹ carbenicillin (for isolate 6-1).

Fig. 6

Inactivation of lasR confers increased β-lactamase activity and β-lactam tolerance. A. Ceftazidime-resistant colonies emerged in a lawn of CF416 or CF416L1 cells on LB agar with 20 µg ml⁻¹ of ceftazidime (Petri dish upper and lower halves), but only the lasR mutant lawns yielded small partially resistant colonies whose growth is inhibited by the addition of 8 µg ml⁻¹ of the β-lactamase inhibitor tazobactam (Petri dish lower halves). Photographs were taken after incubation overnight at 37°C followed by 3 days at 22°C. B. β-Lactamase activity in culture supernatants, measured as a change in optical density at 490 nm (ΔA₄₉₀) after addition of the chromogenic substrate nitrocefin. Cases with no change are marked with an asterisk. Values are the average of three technical replicates, and error bars show standard deviations. Equivalent results were obtained using a qualitative whole-culture assay. C. β-Lactamase activity in whole cultures of CF416 and CF416lasR::Gm. Values are the averages of three cultures, and error bars show standard error of the mean. One unit of β-lactamase is defined as 1 µmol of nitrocefin hydrolysed per min per mg of total protein. D. Reduced killing by ceftazidime of the lasR mutant CF416L1 relative to CF416 and CF416L1R. Values are the average cfu of two cultures, and error bars show standard deviations.