Pseudomonas aeruginosa with LasI Quorum-Sensing Deficiency during Corneal Infection

Abstract

purpose. To understand the importance of Pseudomonas aeruginosa quorum-sensing systems in the development of corneal infection, the genotypic characteristics and pathogenesis of seven ocular isolates with low-protease and acyl homoserine lactone (AHL) activity and quorum-sensing mutants of PAO1 deficient in lasI, lasR, or rhlR were investigated in the study.

methods. The possession of the quorum-sensing genes lasI, lasR, rhlI, rhlR, and the quorum-sensing controlled genes lasB, aprA, and rhlAB in the clinical isolates were determined by polymerase chain reaction and Southern blot hybridization. Elastinolytic activity, controlled by the las system, was assayed using elastin Congo red and rhamnolipid production controlled by the rhl system was assessed using agar plates containing methylene blue/cetyltrimethyl ammonium bromide. Induction of keratitis was examined in a scarified inbred BALB/c mouse model.

results. The clinical isolates Paer1 and -3 were lasI and lasR negative, and the isolates Paer2 and -4 were rhlR and rhlAB negative. The isolates Paer17, Paer26, 6294 and 6206 possessed all the genes examined. There was no rhamnolipid production in clinical isolates Paer2 and -4. The isolates Paer1 and -3 were virtually avirulent in the scarified mouse corneas. Using isogenic PAO1 mutants, strain lasI showed a markedly reduced virulence in the corneal infection model. The remainder of the clinical isolates and the lasR or rhlR mutant strains caused severe keratitis.

conclusions. These results indicate that quorum-sensing deficiency may occur naturally in clinical isolates, and the possession of lasI and hence a functional Las quorum-sensing system may be important in development of corneal infection.

The pathogen Pseudomonas aeruginosa is the most frequently isolated organism in clinical cases of ulcerative keratitis associated with contact lens wear as well as from corneal wounds. ¹ P. aeruginosa produces many extracellularproducts, including proteases (elastase, LasA protease, protease IV, and alkaline protease), toxins (exotoxin A and exoenzyme S) and hemolysins (phospholipase and rhamnolipid), which are candidates for important virulence factors in keratitis. ^{2 3 4} Expression of many of the virulence factors in P. aeruginosa appears to be controlled in a cell density–dependent fashion through a mechanism known as quorum-sensing. ⁵  

Two quorum-sensing systems, the las and rhl systems have been described in P. aeruginosa. ^{5 6} The las system consists of lasI, lasR, and an N-acyl-homoserine lactone signal molecule, N-3-oxo-dodecanoyl homoserine lactone (3O-C₁₂-HSL). ^{7 8} The lasI-encoded synthase directs the formation of the diffusible signal 3O-C₁₂-HSL, which interacts with the LasR transcriptional activator ^{9 10} to activate a number of virulence genes, including lasB, lasA, aprA, toxA, and lasI itself. ^{9 10 11 12 13 14} The second P. aeruginosa quorum-sensing system (rhl) consists of rhlI, rhlR, ^{15 16} and the signal molecule, N-butyryl homoserine lactone (C₄-HSL). ^{8 17} The rhlI product catalyzes the synthesis of the diffusible signal C₄-HSL, which, in conjunction with RhlR, activates expression of rhlAB rhamnolipid synthesis genes, rhlI, and to some extent lasB. ^{8 16 18 19} A hierarchy exists with the las quorum-sensing system being situated above the rhl system. Several additional regulatory proteins have been shown to control the las or rhl system, including QscR, ²⁰ RpoS, ²¹ GacA, ²² and Vfr. ²³ Another signal molecule, 2-heptyl-3-hydroxy-4-quinolone, referred to as the Pseudomonas quinolone signal (PQS), has been identified. ²⁴ PQS can induce both lasB and rhlI in P. aeruginosa. ^{24 25} The production and activity of this signal was also dependent on LasR and RhlR. ²⁴  

A few in vivo studies have demonstrated that the virulence of P. aeruginosa in certain models is associated with quorum sensing. Quorum-sensing–deficient mutants, mainly strains deficient in lasI, rhlI, double lasI/rhlI, or lasR, generated in standard laboratory strains have been reported to be attenuated in virulence in mouse models of pneumonia ^{26 27} and burn wound infection, ^{28 29} as well as in a rat model of chronic lung infection. ³⁰ However, the roles of quorum-sensing systems in the virulence of P. aeruginosa–related keratitis remain unclear. The only published data suggest that a lasR-deficient strain displays a virulence similar to that of the parental strain during corneal infection. ³¹ The difference between the virulence in the lasR mutant mouse in pneumonia and keratitis models points to the possible organ-specific roles of quorum sensing in the virulence of P. aeruginosa. 

Our recent study has demonstrated that a small group of ocular isolates of P. aeruginosa possess a low ability to produce elastase and quorum-sensing signals. ³² In this study, genotypic characteristics in these ocular isolates were further investigated. The role that Pseudomonas quorum sensing plays in the pathogenesis of corneal infection was evaluated in a mouse model of keratitis by using the low-elastase and acyl homoserine lactone (AHL) isolates and the laboratory quorum-sensing mutants. The results demonstrated that lack of the quorum-sensing gene lasI rendered strains weakly virulent during corneal infection. 

Methods

Bacterial Strains and Growth Conditions

The bacterial strains and plasmids used in the study are listed in Table 1 , including seven ocular isolates of P. aeruginosa with a low production of proteases and quorum-sensing signal molecules. ³² P. aeruginosa 6294, with high levels of proteases and signaling molecules, was included as an internal control. P. aeruginosa and Escherichia coli strains were routinely grown in Luria-Bertani (LB) medium. ³⁴ Antibiotics (Sigma-Aldrich Co., St. Louis, MO) were used at the following concentrations: for E. coli, ampicillin at 100 μg/mL; for P. aeruginosa, tetracycline at 50 μg/mL, gentamicin at 100 μg/mL, and mercuric chloride (Sigma-Aldrich) at 15 μg/mL in solid medium and at 7.5 μg/mL in broth. The concentrations of antibiotics for complementation studies were ampicillin at 100 μg/mL, carbenicillin at 400 μg/mL in solid medium and 300 μg/mL in broth. 

PCR for Detection of Quorum-Sensing Genes

Oligonucleotide primers, listed in Table 2 , were designed using Prime (GCG) on the Biomanager program (Australian National Genomic Information Services; available to subscribers at www.angis.org.au) and synthesized by Invitrogen (Australia Pty. Ltd., Victoria, Australia). Bacterial DNA was released by using lysis buffer according to the manufacturer’s instruction (microLYSIS; Microzone Ltd., Sussex, UK). PCR was performed in a 25-μL reaction mixture containing a half volume of PCR mix (32 mM (NH₄)₂SO₄, 125 mM Tris-HCl [pH 8.8], 0.02% Tween 20, 2 mM dNTPs, 2.5 mM MgCl₂, and DNA polymerase 0.05 U/μL; BioMix Red; BioLine GmbH, Luckenwalde, Germany), 100 pmol of each primer, and 2 μL of bacterial DNA template. Parameters for the amplification cycles were denaturation for 1 minute at 94°C, annealing of primers for 1 minute at 52°C, and primer extension for 1.5 minutes at 72°C, for 30 cycles. Agarose gel (2.0%) electrophoresis was used for examining the products after PCR. 

Southern blot was also performed for detection of quorum-sensing genes. Bacterial genomic DNA was extracted with a genomic DNA purification system (Wizard SV; Promega Corp, Madison, WI) according to the manufacturer’s instructions. Digoxygenin (DIG)-labeled DNA probes were generated from strain PAO1 using a PCR DIG probe synthesis kit (Roche Diagnostics GmbH; Mannheim, Germany). Oligonucleotide primers for amplification of quorum-sensing gene probes were same as the primers used in PCR detection (listed on Table 2 ). Bacterial genomic DNA (100 ng) was dot blotted onto a nylon membrane, and the DNA was fixed by heating the membrane in a 120°C oven for 30 minutes. After incubation in hybridization buffer (DIG Easy Hyb; Roche Diagnostics GmbH, Mannheim, Germany) at 42°C for 2 hours, the blot was hybridized in the same buffer containing 30 μL/mL of probe at 42°C overnight. After a low-stringency wash (0.5× SSC), the labeled DIG-probe-hybrid was visualized by standard colorimetric detection. DNA extracted from Pseudomonas putida 15 was used as a negative control. The hybridization values were evaluated on computer (Quantity One imaging-analysis software; BioRad, Herts, UK). 

Complementation of Mutations in P. aeruginosa

For genetic complementation, plasmids pUCPlasI, ³³ pUCPlasR, and pUCPrhlR (Beatson S, unpublished data, 2001), kindly supplied by Scott Beaston (Institute for Molecular Bioscience, University of Queensland, Australia) were introduced into PAO1 quorum-sensing mutants by conjugation. ³³  

Quorum-Sensing Controlled Phenotypes

LasB elastinolytic activity in the culture supernatants of the test strains was determined by the elastin Congo red (ECR) assay, as previously described. ³² Briefly, 0.75 mL of buffered (0.1M Tris-HCl, [pH 8.0], 1 mM CaCl₂) ECR at the concentration of 10 mg/mL were mixed with 0.25 mL of culture supernatants, which were collected by centrifugation and filtered through a 0.2-μm pore filter. The mixture was incubated in 37°C with shaking for 18 hours. Any red color in the supernatant due to the cleaving of ECR was read at 495 nm in a spectrophotometer after centrifugation of the reaction mixture at 3000g for 10 minutes. A control tube containing 0.75 mL buffered ECR and 0.25 mL LB was used. Rhamnolipid production was examined by using the methods of Kohler et al. ³⁵ In brief, 2 μL of broth culture of each test strain was spotted on a M9-based agar plate supplemented with 0.2% (wt/vol) glucose, 2 mM MgSO₄, 0.05% (wt/vol) tyrosine as the nitrogen source, 0.0005% (wt/vol) methylene blue, and 0.02% (wt/vol) cetyltrimethyl ammonium bromide. The plate was incubated first at 37°C for 24 hours and then at room temperature for 48 hours. A light blue transparent halo surrounding the colonies indicated rhamnolipid production. 

Induction of P. aeruginosa Keratitis in the Mouse Scratch Model

The method has been described in detail elsewhere. ^{36 37} All experiments conformed to the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research and were granted ethics approval from the University of New South Wales animal ethics committee. In the present study, inbred BALB/c female mice (6–10 weeks old) were used. Two 1-mm long and parallel scratches were made in the surface of one eye, penetrating the corneal epithelium, and were inoculated subsequently with 5 μL bacterial inoculum prepared in phosphate-buffered saline (PBS) at a concentration of 10⁹ colony-forming units (CFU)/mL. To determine ocular virulence, 9 to 24 eyes were infected per isolate on two occasions. Twenty-four hours after challenge, mice were lightly sedated with isofluorane in oxygen and were examined under a slit lamp biomicroscope for the progress of infection. Grading of corneal infection was recorded in accordance with a published grading scheme: grade 0, eye macroscopically identical with the uninfected contralateral control eye; grade 1, faint opacity partially covering the pupil; grade 2, dense opacity partially or fully covering the pupil; grade 3, dense opacity covering the entire anterior segment; grade 4, perforation of the cornea and/or phthisis bulbi. ³⁸ To determine the recoverable number of bacteria in infected mouse corneas, mice were killed by cervical dislocation 24 hours after challenge. Each enucleated eye was washed once and homogenized in 1 mL of sterile PBS with a sterile tissue grinder. After serial dilution, the recoverable bacteria were enumerated by viable bacterial cell counts. 

Statistical Analysis

Parametric ANOVA was performed to compare the clinical scores of mouse corneas and the number of bacteria recovered from infected corneas with different stains. 

Results

Quorum-Sensing Genes and the Related Phenotypes in the Low-Elastase Isolates

All strains tested possessed the rhlI, lasB, and aprA genes, as tested by PCR (Fig. 1) and Southern hybridization (Table 3) . The rhlAB gene was absent in strains Paer2 and -4, and, concordantly, rhamnolipid activity was not detected in these two isolates (Fig. 2) . For other quorum-sensing genes, lasI and lasR were not detected in the isolates Paer1 and -3 by PCR (Fig. 1) or Southern hybridization (Table 3) , and rhlR was absent in the isolates Paer2 and -4 (Fig. 1 ; Table 3 ). The remainder of the strains possessed all the genes tested (Fig. 1) . The quorum-sensing mutants PAO-JP1 (lasI ⁻), PAOR1 (lasR ⁻), and PDO111 (rhlR ⁻) used in the study were negative in their corresponding quorum-sensing genes (results not shown). 

Corneal Infection of Mice by Low-Elastase Isolates

Twenty-four hours after inoculation, mouse corneas were examined with a slit lamp biomicroscope to grade the severity of infection. Challenge of mouse corneas with strains Paer1 and -3 for 24 hours did not result in apparent infection in any of the mice. The strains 6206 and Paer2, -26, -4, and -17 produced corneal infection (Fig. 3) , causing corneal lesions with average scores ranging from 2.4 to 3.3. Most mice infected with strains 6206, Paer2, and Paer4 demonstrated a ring of infiltrates in the peripheral corneas compared with the extended central corneal infiltrates caused by strain 6294, which possessed a high level of proteases and quorum-sensing signaling molecules and was used as an internal control. Relatively high levels of viable bacteria were recovered in the corneas challenged with strains Paer2, -4, -17, and -26 and 6206 (Fig. 4) . However, very few viable bacterial cells (0–900 CFU/eye) were recovered from the corneas challenged with strains Paer1 and -3 (Fig. 4) . 

Corneal Infection of Mice by Quorum-Sensing Mutants

The data reported thus far indicated that the loss of virulence during corneal infection may be due to the deficiency of the las quorum-sensing genes lasI or lasR. To explore this in more detail, we compared the virulence of P. aeruginosa PAO1 and its isogenic quorum-sensing mutants deficient in genes lasI, lasR, and rhlR. First, we compared the isogenic mutants and complemented strains to PAO1 in terms of their ability to produce elastinolysis and rhamnolipid. The ability of the lasI, lasR, and rhlR mutants to produce elastinolysis was reduced (Table 4) . In addition, the rhlR mutant did not produce rhamnolipid (Table 4) . Complemented mutants had elastinolysis and rhamnolipid restored to levels comparable to those in the wild-type PAO1. In comparison with the wild-type strain PAO1, the lasI mutant showed statistically reduced virulence in the corneal infection model (Fig. 5 , P < 0.05), reduced infection rate (Table 5 ; P < 0.05), and a reduced number of bacteria colonizing the corneas of the mice (Fig. 6 ; P < 0.05). The lasR or rhlR mutants showed no statistically different level of corneal infection, although the lasR mutants had a statistically significant lower numbers of bacteria recovered from infected corneas (P < 0.05). Genetic complementation of lasI with pUCPlasI restored virulence during corneal infection (Fig. 5) , the infection rate (Table 5) , and the number of bacteria recovered from the infected corneas (Fig. 6) . 

Discussion

The study demonstrated that four P. aeruginosa strains isolated from asymptomatic contact lens wearers or the noninfectious corneal inflammatory response, contact lens–induced acute red eye (CLARE) ³⁹ were deficient in either lasI and lasR or rhlR and rhlAB. By contrast, strains isolated from MK possessed all these genes, even if they had phenotypically similar characteristics to the previously mentioned four strains, in terms of protease and AHL production. To our knowledge, this is the first report of clinical isolates carrying defective las or rhl quorum-sensing genes. Although elastase-deficient strains have been isolated from wound and tracheal infections, ⁴⁰ all those strains had detectable levels of at least lasR mRNA, indicating the presence of that gene. ⁴⁰  

The Pseudomonas strains Paer1 and -3 were avirulent in scarified mouse corneas and did not possess either the lasI or lasR genes. This result suggests that las quorum-sensing system may play an important role during corneal infection. The avirulence of strain Paer1 has been reported before, ^{36 37 41} but the factors contributing to its avirulence were not known. The avirulence of this strain is not attributable to the loss of elastase or other proteases. ⁴¹ However, strain Paer1 has been shown to be relatively noninvasive and noncytotoxic. ^{36 37}  

The importance of the las system, particularly the possession of lasI, in ocular virulence was confirmed using isogenic mutants of PAO1. The mice infected with mutants lacking lasI showed reduced virulence, and the virulence of the lasI mutant was fully restored by complementation with a functional lasI gene. P. aeruginosa quorum-sensing mutants lacking lasI or rhlI have been reported to be less virulent in a mouse model of P. aeruginosa burn infection ^{28 29} and a model of acute P. aeruginosa pulmonary infection. ²⁶ The observation that the lasR mutant did not have a statistically significant reduction in virulence (apart from reduced bacterial numbers in the eye) agrees with a previous report in which there was no significant difference reported between the 50% infective doses of PAO1 and the lasR mutant during corneal infection. ³¹ The lack of infection caused by lasI mutants could be due to direct effects of the AHL produced by the product of this gene, N-(3-oxododecanoyl)homoserine lactone, which can stimulate a significant induction of mRNAs for the cytokines interleukin (IL)-1α and IL-6 and the chemokines macrophage inflammatory protein (MIP)-2, monocyte chemotactic protein-1, MIP-1β, inducible protein 10, and T-cell activation gene 3, in addition to cyclooxygenase (Cox)-2 expression in the skin of mice. ⁴² This may be more likely than a direct effect on gene transcription of those genes regulated by the Las system, because the lasR mutants were relatively unaffected in their virulence. Alternatively, lasR but not lasI mutants may be compensated for in vivo by the presence of the RhlRI quorum-sensing system. A proteomics analysis of quorum-sensing mutants of P. aeruginosa has been conducted. ⁴³ Generally, mutations in either lasR or rhlR produced effects on the extracellular proteins released in vitro similar to those of mutants lasI or rhlI, respectively. However, lasI mutants produced much less endopeptidase (PrpL) activity compared to the lasR mutants or wild-type PAO1 and much less of the two-partner secretion exoprotein PA0041 than the lasR mutant only. ⁴³ This may indicate a role for this (PrpL) endopeptidase activity in corneal infection. Indeed, it has been shown that PrpL endopeptidase is identical with protease IV of P. aeruginosa. ⁴⁴ Mutants of P. aeruginosa that lack protease IV (PrpL endopeptidase) show much reduced corneal virulence, ⁴⁵ and P. putida containing a plasmid encoding protease IV causes increased corneal virulence. ⁴⁶  

The finding that the clinical isolates with rhlR deficiency and the rhlR mutant did not have reduced virulence indicates that RhlR is not essential for the establishment of corneal infection in a murine model. RhlR-regulated genes include those that encode for elastase, LasA protease, rhamnolipid, and pyocyanin. ¹⁸ Rhamnolipid, a rhamnose-containing glycolipid biosurfactant, has a detergent-like structure and is believed to solubilize the phospholipids of lung surfactant, making them more accessible to cleavage by phospholipase C. ⁴⁷ Rhamnolipid also inhibits the mucociliary transport and ciliary function of human respiratory epithelium. ⁴⁸ The results of the present study showed that rhamnolipid production was probably nonessential for ocular virulence, because the strains Paer2 and -4 deficient in rhamnolipid production caused corneal infection. However, in an alternative infection model based on an alfalfa seedling model of infection, mutations in rhlR caused a reduced infection frequency. ⁴⁹ The rhlR gene is controlled to some degree by the las system as part of the quorum-sensing hierarchy. ²¹ It is possible that the function of RhlR may be compensated for by the las quorum-sensing system. 

Our previous study demonstrated that the clinical ocular isolates of P. aeruginosa used in the present study produce little or no elastase activity. ³² However, the results of the present study revealed that all these strains possess lasB and aprA genes. The expression of the elastase gene (lasB) is mainly regulated at the transcriptional level by the lasI-lasR system. ^{9 11 13} The lack of or low expression of lasB in strains Paer1, -3, -2, and -4 may be attributable to the lack of las or rhl quorum-sensing genes in these strains. Clearly strains Paer17, Paer26, and 6206 possess alternative pathogenic traits. 

In conclusion, the discovery that a defect in the lasI gene in P. aeruginosa leads to reduction of the corneal infection suggests that the las system plays an important role in P. aeruginosa–induced corneal disease. This may be mediated by the direct effect of the product of LasI or by a reduction in transcription of genes controlled by the Las system. These findings make the quorum-sensing genes attractive targets for antimicrobial therapy. Strategies or agents capable of blocking the production of LasI may be useful for preventing Pseudomonas keratitis. 

 Supported by the Australian Federal Government through the Cooperative Research Centres program.

 Submitted for publication September 6, 2003; revised December 3, 2003; accepted December 11, 2003.

 Disclosure: H. Zhu, None; R. Bandara, None; T.C.R. Conibear, None; S.J. Thuruthyil, None; S.A. Rice, None; S. Kjelleberg, None; M. Givskov, None; M.D.P. Willcox, None

 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.

 

Table 1.

 

Bacterial Strains and Plasmids Used in This Study

Strain or Plasmid	Relevant Phenotype or Genotype	Source and/or Reference
P. aeruginosa
Paer1	Low protease and AHL production	CLARE 32
Paer2	Low protease and AHL production	Asymptomatic contact lens wearer 32
Paer3	Low protease and AHL production	Asymptomatic contact lens wearer 32
Paer4	Low protease and AHL production	Asymptomatic contact lens wearer 32
Paer17	Low protease and AHL production	MK 32
Paer26	Low protease and AHL production	MK 32
6206	Low protease and AHL production	MK 32
6294	High protease and AHL production	MK 32
PAO1	Wild-type prototroph
PAO-JP1	lasI −, ΔlasI derivative of PAO1, Tcr	Mutant of PAO1 8
PAOR1	lasR −, ΔlasR derivative of PAO1, Tcr	Mutant of PAO1 9
PDO111	rhlR −, ΔrhlR::Tn501 derivative of PAO1, Hgr	Mutant of PAO1 18
	endA1 hsdR17 (r k − m k +) glnV44 thi-1 recA1 gyrA relA1 Δ(lacIZYA-argF)U169 deoR [φ80dlacΔ(lacZ)M15]
E. coli DH5α			Standard strain 33
Plasmids
pUCPlasI	lasI cloned into pUCPKS with lac	Plasmid 33
pUCPlasR	lasR cloned into pUCPKS	Plasmid, this study
pUCPrhlR	rhlR cloned into pUCKS	Plasmid, this study

 CLARE, contact lens induced acute red eye ³⁹ ; MK, microbial keratitis; AHL, acyl homoserine lactone.

Table 2.

 

Oligonucleotide Primers Used in the Study

Primer	Sequence	Length of Primer (bp)	Size of Amplified Product (bp)
lasB-f	5′ ttctacccgaaggactgatac 3′	21	153
lasB-r	5′ aacacccatgatcgcaac 3′	18
aprA-f	5′ accctgtcctattcgttcc 3′	19	140
aprA-r	5′ gattgcagcgacaacttgg 3′	19
rhlAB-f	5′ tcatggaattgtcacaaccgc 3′	21	151
rhlAB-r	5′ atacggcaaaatcatggcaac 3′	21
lasI-f	5′ cgtgctcaagtgttcaagg 3′	19	295
lasI-r	5′ tacagtcggaaaagcccag 3′	19
lasR-f	5′ aagtggaaaattggagtggag 3′	21	130
lasR-r	5′ gtagttgccgacgacgatgaag 3′	19
rhlI-f	5′ ttcatcctcctttagtcttccc 3′	22	155
rhlI-r	5′ ttccagcgattcagagagc 3′	19
rhlR-f	5′ tgcattttatcgatcagggc 3′	20	133
rhlR-r	5′cacttccttttccaggacg 3′	19

 f, foreward; r, reverse.

Figure 1.

 

Detection of quorum-sensing genes in clinical isolates of P. aeruginosa by polymerase chain reaction. Agarose gel (2%) electrophoresis showing the PCR products amplified by using different quorum-sensing gene primers in each test strain.

Table 3.

 

Southern Hybridization of the P. aeruginosa genes lasI, lasR, rhlI, and rhlAB

Strain	Percentage Hybridization with Probe
		lasI	lasR	rhlAB	rhlR
Paer1	13	27	100	98
Paer2	100	98	0	0
Paer3	14	10	100	100
Paer4	98	99	4	0
Paer17	100	100	99	100
Paer26	100	100	100	100
6206	100	100	100	99
PAO1	100	100	100	100
Pp15	15	16	0	0

 Pp15, Pseudomonas putida 15, used as a negative control. Percentage hybridization with probe was calculated after Southern dot blot analysis. The results demonstrate that strains Paer1 and Paer3 lack the lasI and lasR genes and strains Paer2 and Paer4 lack the rhlAB and rhlR genes.

Figure 2.

 

Production of rhamnolipid by clinical isolates of P. aeruginosa. A transparent halo surrounding the colonies on the test plate indicates rhamnolipid production.

Figure 3.

 

Clinical score of mouse corneas infected with low-elastase isolates of P. aeruginosa for 24 hours. Box: 25th to the 75th percentiles; horizontal line: 50th percentile; vertical bar: 10th to 90th percentiles. *P < 0.05 compared with other clinical isolates.

Figure 4.

 

Number of bacteria recovered from mouse corneas 24 hours after challenge with clinical isolates of P. aeruginosa. *P < 0.05 compared with other clinical isolates.

Table 4.

 

Restoration of Elastase Activity and Rhamnolipid Production in P. aeruginosa PAO1 Quorum-Sensing Mutants

Strain	Genotype	Gene(s) on Plasmid	Elastinolysis (A495/OD660)	Rhamnolipid
PAO1	Wild-type		1.08 ± 0.08	+
PAO-JP1	lasI −		0.14 ± 0.13	+
		+lasI	1.20 ± 0.07	N
PAOR1	lasR −		0.03 ± 0.01	−
		+lasR	1.11 ± 0.26	N
PDO111	rhlR −		0.32 ± 0.09	−
		+rhlR	1.03 ± 0.16	+

 LasB elastase activities were determined by measuring the absorbance of enzyme reaction mixture at 495 nm due to red pigments released from ECR, and normalized by dividing the A₄₉₅ by the bacterial cell density (OD₆₆₀) of the culture. Values are the average of at lease two experiments (each in duplicate) ± SD. N, not determined.

Figure 5.

 

Clinical score of mouse corneas infected with P. aeruginosa PAO1 and its quorum-sensing mutants for 24 hours. Presentation of data is a described in Figure 3 . *P < 0.05 compared with wild-type PAO1.

Table 5.

 

Infection Rate in Mice Challenged with Quorum-Sensing Mutants of P. aeruginosa

Strain	Genotype	Mice Infected (%)	Occurrence of Infection (n)
PAO1	Wild type	83	20/24
PAO-JP1	lasI −	44	8/18
PAOR1	lasR −	81	17/21
PDO111	rhlR −	81	17/21
PAO-JP1	+lasI	89	8/9

 Mice were challenged for 24 hours.

Figure 6.

 

Number of bacteria recovered from the cornea of infected mice 24 hours after challenge with P. aeruginosa PAO1 and its quorum-sensing mutants. *P < 0.05 compared with wild-type PAO1.