Novel bacteria degrading N-acylhomoserine lactones and their use as quenchers of quorum-sensing-regulated functions of plant-pathogenic bacteria

Stéphane Uroz1, Cathy D'Angelo-Picard1, Aurélien Carlier1, Miena Elasri1, Carine Sicot1, Annik Petit1, Phil Oger1,2, Denis Faure1 and Yves Dessaux1

1 Interactions plantes et micro-organismes de la rhizosphère, Institut des Sciences du Végétal, CNRS, avenue de la Terrasse, 91198 Gif-sur-Yvette Cedex, France
2 Laboratoire de Sciences de la Terre, Ecole Normale Supérieure, 43 allée d'Italie, 6364 Lyon Cedex, France

Correspondence
Yves Dessaux
dessaux{at}isv.cnrs-gif.fr


   ABSTRACT
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Bacteria degrading the quorum-sensing (QS) signal molecule N-hexanoylhomoserine lactone were isolated from a tobacco rhizosphere. Twenty-five isolates degrading this homoserine lactone fell into six groups according to their genomic REP-PCR and rrs PCR-RFLP profiles. Representative strains from each group were identified as members of the genera Pseudomonas, Comamonas, Variovorax and Rhodococcus. All these isolates degraded N-acylhomoserine lactones other than the hexanoic acid derivative, albeit with different specificity and kinetics. One of these isolates, Rhodococcus erythropolis strain W2, was used to quench QS-regulated functions of other microbes. In vitro, W2 strongly interfered with violacein production by Chromobacterium violaceum, and transfer of pathogenicity in Agrobacterium tumefaciens. In planta, R. erythropolis W2 markedly reduced the pathogenicity of Pectobacterium carotovorum subsp. carotovorum in potato tubers. These series of results reveal the diversity of the QS-interfering bacteria in the rhizosphere and demonstrate the validity of targeting QS signal molecules to control pathogens with natural bacterial isolates.


Abbreviations: 3-oxo-C6-HSL, 3-oxo-hexanoylhomoserine lactone; 3-oxo-C8-HSL, 3-oxo-octanoylhomoserine lactone; 3-oxo-C10-HSL, 3-oxo-decanoylhomoserine lactone; C4-HSL, butanoylhomoserine lactone; C6-HSL, hexanoylhomoserine lactone; C8-HSL, octanoylhomoserine lactone; KB, King's broth; KBm, KB medium modified; LBm, LB medium, modified; N-AHSL, N-acylhomoserine lactone(s); QS, quorum-sensing; REP-PCR, repetitive DNA sequence PCR

The GenBank accession numbers for the sequences determined in this work are AF532866AF532871.


   INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Several bacteria have evolved the ability to modulate gene expression as a function of their population density, a regulatory process termed quorum-sensing (QS) (Fuqua et al., 1994). All QS systems described in Gram-negative bacteria rely upon the production of one or more mediator molecules by a bacterial population (Fuqua et al., 2001). As the microbial density increases, the concentration of the mediator molecule(s) or quoromone(s) in the environment also increases to reach a threshold value allowing its perception by the relevant microbes (Winans & Bassler, 2002). The first QS regulatory system was identified in the fish symbiont Photobacterium fischeri, which regulates the production of light in a density-dependent fashion (Eberhard et al., 1981). In this model system, the QS mediator molecule is 3-oxo-N-hexanoylhomoserine lactone (3-oxo-C6-HSL), a member of the largest family of quoromones known as N-acylhomoserine lactones (N-AHSL). All N-AHSL molecules result from the enzymic condensation of homoserine lactone with a 3-hydroxy-, a 3-oxo- or an unsubstituted fatty acid (Schaefer et al., 1996).

QS regulation occurs in numerous micro-organisms, living in diverse environments ranging from the animal gastrointestinal tract to sewage fluids, and from deep-ocean fish organs to plants (Eberhard et al., 1981; Erickson et al., 2002; Pierson et al., 1998). Some functions related to the invasiveness or the pathogenicity of several bacterial species also appear to be regulated by QS (Latifi et al., 1995; Reverchon et al., 1998; Schaefer et al., 1996; Zhang et al., 1993). All the above suggests that QS may confer a strong selective advantage upon the relevant microbes living in diverse habitats. As a consequence, strategies targeting microbial cell–cell communication systems should allow the development of valuable biological control agents (Bauer et al., 2002; Finch et al., 1998; Fray, 2002; Robson et al., 1997). Compounds inhibiting the perception of the QS signals have been described (Gram et al., 1996; Teplistki et al., 2000). However, the strategy we have developed relies upon the isolation and identification of microbes responsible for the degradation of the N-AHSL molecules. Some data related to N-AHSL degradation are available. One Gram-negative bacterium belonging to the species Variovorax paradoxus has been isolated from soil samples as a microbe degrading 3-oxo-C6-HSL (Leadbetter & Greenberg, 2000). Another bacterium, a Gram-positive Bacillus sp., has a N-AHSL hydrolase encoded by the aiiA gene (Dong et al., 2000). The aiiA gene has been cloned and introduced into tobacco to generate transgenic plants that exhibit increased resistance towards Erwinia carotovora, the pathogenicity of which is dependent upon the QS-regulated production of enzymes macerating plant cell wall (Dong et al., 2001). More recently, the aiiA gene has been detected in several strains of Bacillus sp. (Dong et al., 2002; Lee et al., 2002). Another gene related to aiiA (28 % homology at the nucleotide level) located on the large catabolic At plasmid of Agrobacterium tumefaciens may be involved in the fine tuning of the QS-dependent regulation of the transfer of another plasmid, namely the tumorigenic pTi of Agrobacterium (Zhang et al., 2002). Finally, a Ralstonia isolate degrading N-AHSL has also been identified (Lin et al., 2003).

The goals of this study were: (i) to isolate and identify bacteria that degrade N-AHSL molecules; (ii) to investigate their inactivation abilities; and (iii) to explore their potential use to antagonize QS-regulatory processes. We report here on the isolation of several novel bacterial species degrading N-AHSL, from a tobacco rhizosphere. Among these, strain W2, identified as a member of the species Rhodococcus erythropolis, showed the highest capability to inactivate a broad range of N-AHSL molecules. Strain W2 was also successfully used to interfere with at least three QS-mediated processes: violacein production by Chromobacterium, transfer of pathogenicity in Agrobacterium, and expression of pathogenicity in Pectobacterium carotovorum subsp. carotovorum (previously Erwinia carotovora; Hauben et al., 1998).


   METHODS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Bacterial strains, media and culture conditions.
Aside from strains isolated in this study, bacterial strains were Pseudomonas fluorescens strain 1855-344, Agrobacterium strains R10 and C58.00 (all from our collection), Pectobacterium carotovorum subsp. carotovorum strain Pcc797 (kindly provided by Louis Gardan, INRA Angers, France), and the biosensors Chromobacterium violaceum CV026 (McClean et al., 1997) and Agrobacterium tumefaciens NTLR4 (Cha et al., 1998). The media used were modified Luria–Bertani (LBm) and KB (KBm) rich media (Vaudequin-Dransart et al., 1995), and AT minimal medium (Petit & Tempé, 1978). The minimal medium was supplemented when necessary with mannitol (2 g l-1 final concentration), with NH4(SO4)2 (1 g l-1), or with cycloheximide (250 mg l-1). Agar (16 g l-1) was used to solidify the media. Unless otherwise stated, bacteria were grown at 25 °C, except for the biosensors, which were grown at 28 °C. N-AHSL were from various commercial sources and gifts from Professor P. Williams (University of Nottingham, UK).

Enrichment procedures for bacteria degrading or inactivating N-AHSL from a tobacco rhizosphere.
Seeds of tobacco line Nicotiana tabacum cv. Samson, free of bacterial contamination, were germinated in a soil mixture containing equal volumes of unsterilized reference soil from La-Côte-Saint-André (Isère, France) and sterile Loire River sand. The plants were grown for 3 months in a greenhouse, under long days conditions (16 h light) at 17 °C (night) and 24 °C (day). Micro-organisms were extracted from tobacco rhizosphere using a ‘Nicodenz’ (Gibco-BRL) gradient, essentially according to Lindhal & Bakken (1995). Micro-organisms recovered from 1 g soil were finally resuspended in 1 ml sterile water and serially diluted tenfold. Twenty microlitres of the original suspensions and serial dilutions were inoculated to 180 µl of the enrichment medium [AT medium supplemented with ammonium sulfate (1 g l-1) and C6-HSL (2·5 mM, i.e. 0·5 g l-1 final concentration) as sole carbon source], and incubated at 25 °C. After 48 h, 20 µl of each culture was used to inoculate 180 µl of fresh enrichment medium. The same procedure was repeated three times. At the fourth enrichment cycle, a diluted suspension was plated onto LBm and KBm media to isolate individual colonies, which were kept as frozen stocks in diluted glycerol (-80 °C).

Detection of N-AHSL and inhibitors.
N-AHSL were detected using the biosensors Chromobacterium violaceum CV026 and Agrobacterium tumefaciens NTLR4. Bioassays were performed using the plate streak assay on LBm (Hwang et al., 1994), and the silica plate analysis (C18-reverse phase), essentially as described by Shaw et al. (1997) and McClean et al. (1997). TLC analysis of N-AHSL was performed essentially according to Elasri et al. (2001). Semi-quantitative evaluation of N-AHSL concentrations was performed by comparing the diameters of the haloes generated by the biosensors with those of reference samples of N-AHSL at known concentration. For the detection of putative inhibitors, the supernatant from 5 ml culture was extracted with 5 ml ethyl acetate. The organic and aqueous phases were spotted directly onto a TLC plate and subjected to chromatography. Presence of inhibitors was investigated by reverse TLC as described previously (McClean et al., 1997).

N-AHSL degradation assay.
Individual colonies (taken from frozen stocks spread onto LB plates) were suspended in 100 µl minimum AT medium and 20 µl aliquots of these suspensions were used to inoculate 180 µl AT medium supplemented with C6-HSL (5 mg l-1) and ammonium sulfate (1 g l-1). Cultures were incubated at 25 °C for 4 days, with shaking. Disappearance of C6-HSL from the media was assessed at t0, t+2 days and t+4 days, using the silica plate assay described above. A control experiment involving non-inoculated degradation medium processed as for the inoculated media was performed at the same time as the degradation assays. For determination of the degradation spectrum, N-AHSL were used at the following final concentrations: 171 mg l-1 (C4-HSL), 5 mg l-1 (C6-HSL); 5·3 mg l-1 (3-oxo-C6-HSL and C7-HSL), 5·7 mg l-1 (C8-HSL); 6 µg l-1 (3-oxo-C8-HSL) and 1 mg l-1 (3-oxo-C10-HSL), in both AT and LBm media. Because N-AHSL are sensitive to alkaline pH (Yates et al., 2002), all degradation assays were done in AT and LBm media that were buffered to pH 6·5 by addition of 100 mM KH2PO4/K2HPO4.

Phenotypic characterization of bacterial isolates.
Gram determination was performed using a kit from Roche, according to the manufacturer's instructions. Cell morphology was observed with a Reichert light transmission microscope, at x400 and x1000 magnification, on unstained, living bacteria. Phase interference contrast and Nomarski filters were used when necessary. Trophic characteristics of the isolates were analysed using the GP2 plate Biolog system according to the manufacturer's instructions.

Molecular analysis of bacterial isolates.
The primers REP1RI and REP2-II (Versalovic et al., 1991) were used for REP-PCR analysis of the genomic DNA content. Amplifications were performed in a final volume of 25 µl essentially as described by Versalovic et al. (1991). The rrs-ribotyping was done by amplification of the 16S rRNA-encoding gene (rrs) with the universal primers 281bis and FGPS 1506'–153 (Huguet et al., 2001), and by digestion of the amplified fragments using AluI, HaeIII, MseI and Sau3A restriction enzymes, as described by Oger et al. (1998). DNA sequencing was performed by MWG-Biotech (Les Ulis, France) on amplification fragments, the minimal size of which was 1461 bp.

Inhibition of violacein production by Chromobacterium violaceum.
Wells of a microtitre plate were filled with two layers of media. The first layer consisted of 50 µl of solid LBm medium supplemented with C6-HSL at concentrations ranging from 0 to 250 ng. The second consisted of 130 µl of a cell suspension of Rhodococcus erythropolis strain W2 at variable cell density in semi-solid LBm medium (7 g agar l-1). After solidification, 20 µl of C. violaceum strain CV026 (i.e. 5x107 c.f.u.) in liquid LBm medium was added to each well. Two negative controls, one with no W2 cells added, and another with W2 cells replaced by Ps. fluorescens strain 1855-344 (unable to degrade any N-AHSL molecules), were performed on each plate. Plates were incubated at 30 °C for 24 h. Production of violacein was assessed by visual inspection of the plates and numerically documented.

Inhibition of the conjugative transfer of Agrobacterium tumefaciens Ti plasmid.
The recipient strain C58.00RS (Vaudequin-Dransart et al., 1995), devoid of both At and Ti plasmids, was maintained in liquid AT medium amended with mannitol (2 g l-1), ammonium sulfate (1 g l-1) and rifampicin (200 µg ml-1). The pTi donor strain, R10, was maintained in liquid AT medium supplemented with octopine (2 g l-1) as sole carbon and nitrogen source. Prior to conjugation, 20 µl of the donor strain R10 (108 c.f.u. ml-1 in 0·8 % NaCl) or 20 µl of a 1 : 1 mixture of strains R10 and W2 (all at 108 c.f.u. ml-1 in 0·8 % NaCl) were used to inoculate 480 µl AT medium supplemented with ammonium sulfate (1 g l-1) and octopine (2 g l-1). The media were incubated for 28 h at 25 °C. For conjugation, strain C58.00RS (serially cultivated in the presence of rifampicin) was plated at 5x107 c.f.u. onto the conjugation selection (CS) medium: solid AT medium supplemented with octopine (2 g l-1) and rifampicin (200 µg ml-1). On top of the recipient, 20 µl aliquots of the 28 h cultures were spotted. Enumeration of the transconjugants was performed both directly on the spots, and following recovery of transconjugants from the plates, resuspension, dilution in 0·8 % NaCl and plating onto the same CS medium and on LBm medium, with incubation for 4 days at 25 °C.

Inhibition of the pectinolytic activity of Pectobacterium carotovorum subsp. carotovorum.
The assay was performed on potato tubers (cv. Franceline) essentially as described by Lojkowska et al. (1995). Tubers were gently washed first with running tap water, and sterilized with sodium hypochlorite (0·12 chlorine deg.), extensively rinsed with sterile water and finally dried under sterile conditions. Strains used in this assay were Pcc797 and W2 as a quencher. Strains were cultivated overnight at 25 °C in LBm medium, suspended and diluted in sterile 0·8 % NaCl. Bacterial suspensions (pathogen alone, or with the quencher, at various cell densities) were introduced directly into the tubers using a 200 µl tip fitted on a micropipette. Tubers were incubated at 25 °C under a humid atmosphere (over 90 % humidity) for 3 days. The results of the inoculation were assessed by visual inspection after slicing the tubers, and numerically recorded.


   RESULTS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Isolation of bacteria degrading N-AHSL
Total cultivable bacteria were recovered from tobacco rhizosphere at densities ranging from 3x107 to 108 c.f.u. ml-1. The inoculated media containing C6-HSL became turbid at the first cycle (after 2 days) or at the following cycles, suggesting that bacterial growth occurred during the enrichment procedure. After the third enrichment step, the cell suspensions were diluted and plated onto both LBm and KBm media. Fifty-four individual colonies with different morphologies and colours were retained. Among them, 25 isolates indeed induced a disappearance of the N-AHSL signal after 2 and 4 days cultivation in AT medium containing this N-AHSL as sole C source, using the Chromobacterium assay (Fig. 1). They were sorted into two groups according to their efficiency in inducing the disappearance of the N-AHSL signal: (i) isolates that induced a partial disappearance of C6-HSL after 4 days and (ii) those inducing a complete disappearance of C6-HSL after 4 days. Remarkably, within this latter group, isolates W2 and W3 were able to obliterate the C6-HSL signal during short-term incubations (about 15 min; data not shown). In addition, isolate W2 was able to degrade over 95 % of C6-HSL, after a 1 day incubation in medium containing 2·5 mM of this molecule. For this strain, and for strain D1, we correlated the disappearance of C6-HSL with growth, as judged by turbidimetry (doubling time about 96 h).



View larger version (46K):
[in this window]
[in a new window]
 
Fig. 1. Detection of C6-HSL-degrading isolates. Degradation was investigated at 0, 2 and 4 days. Culture supernatants from strains to be assayed were spotted onto a silica TLC support. C6-HSL disappearance was revealed using the sensor strain Chromobacterium CV026 as described in Methods. From left to right, the vertical rows were: C (control), non-inoculated degradation medium taken at 0, 2 and 4 days; strains Awt4 to W3, various isolates exhibiting various C6-HSL degradation capabilities; and S (standard), 10 ng, 4 ng, 2 ng and 0·2 ng (detection limit) of C6-HSL (from top to bottom), spotted just before revelation with the biosensor (composite image).

 
Identification of the N-AHSL-degrading bacteria
The 25 isolates were first examined using REP-PCR analysis to identify duplicates that may have resulted from the enrichment procedure. Similar or very closely related patterns were obtained for several isolates, allowing the delineation of 15 REP groups (Table 1). One strain from each of these 15 REP groups was further analysed by ribotyping, using PCR-RFLP profiling. The generated patterns allowed the definition of six rrs RFLP groups. One representative member of each group (i.e. AT3, Cwt6, D1, K2, K6 and W2) was retained for further studies. Analysis of the nucleotide sequence of the amplified rrs genes of these six strains allowed their identification as Pseudomonas sp. (strain K2), Variovorax sp. (K6), Variovorax paradoxus (Cwt6), Comamonas sp. (D1), Comamonas testosteroni (AT3) and Rhodococcus erythropolis (W2). Morphological, biochemical and physiological data (Gram determination, Table 1; or Biolog analysis, not shown) were in agreement with the species identification from the molecular analysis. They indicate that Pseudomonas sp. strain K2 is either a Ps. putida or a Ps. fluorescens strain.


View this table:
[in this window]
[in a new window]
 
Table 1. C6-HSL-degrading isolates obtained from tobacco rhizosphere

 
N-AHSL-degrading bacteria do not inhibit the detection of the N-AHSL signal
The possibility that the disappearance of the C6-HSL signal could be due to the production by the isolates of one or more compounds inhibiting the detection of this N-AHSL by the sensor needed to be ruled out. To do so, bacterial culture supernatant was extracted with ethyl acetate. Both the organic and aqueous phases of the supernatants were analysed by reverse TLC. None of the N-AHSL-degrading strains appeared to produce compound(s) inhibiting the detection of limiting amounts of C6-HSL by the biosensor CV026 (not shown). Because the lactone moiety of N-AHSL can be readily hydrolysed at alkaline pH (Byers et al., 2002; Yates et al., 2002), all degradation assays were performed in buffered media (see Methods). Moreover, pH values were checked at t0 and at the time of sampling, and compared to the pH values of non-inoculated control medium. No change of pH was observed during growth of any bacterial clone.

The N-AHSL-degrading bacteria exhibit different substrate specificity
The N-AHSL-degrading abilities of the six characterized strains (AT3, Cwt6, D1, K2, K6 and W2) were assessed by using as substrates a series of unsubstituted and oxo-substituted N-AHSL. The detection of the remaining N-AHSL molecules was performed after 24 and 48 h in both minimal AT and rich LBm buffered media. Results revealed that the degradation properties of the various strains differed with respect to their substrate preferences and degradation kinetics (Fig. 2). For instance, only strains W2 and D1 degraded 3-oxo-C6-HSL in AT medium, while all six strains degraded this N-AHSL in LBm medium. Similarly, strains AT3 and K2 did not degrade C4-HSL, whatever the degradation medium, even after 48 h, while the other strains degraded this N-AHSL in that period of time at least in AT medium. Strains AT3 and D1 – both identified as members of the genus Comamonas – exhibited very different degradative patterns. Interestingly, strains K6 and Cwt6, identified as Variovorax sp. and V. paradoxus, respectively, did not differ much in their N-AHSL degradative preferences, which were close to those of the V. paradoxus strain VAI-C described by Leadbetter & Greenberg (2000) (data not shown).



View larger version (50K):
[in this window]
[in a new window]
 
Fig. 2. N-AHSL degradation patterns of representative isolates. N-AHSL degradation was investigated on six isolates (from top to bottom: AT3, C. testosteroni; K2, Pseudomonas sp.; K6, Variovorax sp.; Cwt6, V. paradoxus; D1, Comamonas sp.; W2, R. erythropolis) representative of the diversity of our collection of catabolic isolates. Degradation was assessed in LBm medium (L) and in Atm medium (A), after both 24 and 48 h of incubation, for the N-AHSL molecules indicated at the top of the figure. These molecules were: C4, butanoylhomoserine lactone; C6, hexanoylhomoserine lactone; OC6, 3-oxo-hexanoylhomoserine lactone; C7, heptanoylhomoserine lactone; C8, octanoylhomoserine lactone; OC8, 3-oxo-octanoylhomoserine lactone; and OC10, 3-oxo-decanoylhomoserine lactone. The colour of the squares relates to the amount of N-AHSL degraded at the relevant time point: white, 0–50 % of input N-AHSL degraded; light grey, 50–90 %; dark grey, 90–100 %. All experiments took into account the detection limit of the biosensors for each assayed N-AHSL.

 
R. erythropolis strain W2 as a quencher of the QS signal
The analysis of the degradative properties of R. erythropolis strain W2 revealed that it exhibits a broad degradation spectrum and rapid N-AHSL-degrading activity. For this reason, this strain was used for further studies to evaluate its potential to interfere with several QS-regulated functions in other bacteria. The QS-regulated functions investigated were violacein production by C. violaceum strain CV026, the transfer of A. tumefaciens Ti plasmid and P. carotovorum subsp. carotovorum pathogenicity. As a prerequisite to these experiments, we verified that strain W2 did not affect the growth of the three above-mentioned bacteria in co-culture experiments as judged from the determination of the colony number of each co-inoculated bacterium (experiments performed in LB medium, based on different bacterial colony morphology).

The ability of R. erythropolis strain W2 to interfere with violacein production is shown in Fig. 3. While Ps. fluorescens strain 1855-344, introduced at about 2x107 cells per well, did not affect violacein production, the quenching of violacein production by strain W2 was visible when 1·5x106 cells of this strain were introduced into the experimental media. These cells were able to degrade up to 15 ng C6-HSL under our conditions. When the amount of W2 cells was higher, the amount of C6-HSL degraded was also higher. For instance when 1·5x107 cells were introduced into our media, they were able to degrade 100 ng C6-HSL over 24 h, providing a degradation rate of about 400 ng C6-HSL per 109 cells per hour. Interestingly, the quantity of C6-HSL degraded showed a linear relation with the amount of W2 introduced, as deduced from the determination of R2 values (R2=0·98; data not shown). The experiment was repeated three times with comparable results.



View larger version (121K):
[in this window]
[in a new window]
 
Fig. 3. Quenching of violacein production in Chromobacterium by R. erythropolis strain W2. Violacein production by C. violaceum CV026 was investigated in the presence of increasing cell densities of the N-AHSL-degrading strain W2 (quencher). The amount of C6-HSL introduced into the wells (0–250 ng) is given at the top of the figure. From top to bottom: row 1, control without quencher; rows 2–7, increasing cell densities of W2 (approx. 1·5x106, 3x106, 4·5x106, 6x106, 7·5x106, 1·5x107 c.f.u. per well); row 8, negative control with strain W2 replaced by Ps. fluorescens strain 1855-344 introduced at 2x107 c.f.u. per well. The picture was taken after a 24 h incubation at 30 °C. The experiment was repeated three times with the same results.

 
The ability of strain W2 to interfere with Ti plasmid conjugation was assessed in crosses involving the pTi donor strain R10 and the recipient strain C58.00RS, both belonging to the species A. tumefaciens. Conjugation frequencies (per donor) were 1·5x10-3 when the conjugation was performed without the quencher strain W2, and 1·5x10-5 in the presence of the quencher strain, indicating that the introduction of strain W2 into the conjugation mixture decreased the conjugation rate of Agrobacterium cells.

The ability of R. erythropolis strain W2 to attenuate the pathogenicity of P. carotovorum subsp. carotovorum was investigated using strain Pcc797 and the host plant potato. An example of the results is shown in Fig. 4. In the absence of the quencher strain W2, P. carotovorum strain Pcc797 induced the maceration of the tissues when inoculated at concentrations ranging from 105 to 106 cells per tuber. On the other hand, neither W2 inoculation (at 105 cells per tuber, not shown, and 106 cells per tuber, Fig. 4) nor the presence of NaCl induced necrosis or maceration of the tissues. Remarkably, the co-inoculation of P. carotovorum strain Pcc797 (at about 106 cells per tuber) and R. erythropolis strain W2 (at about 106 cells) totally prevented the pathogenic strain from macerating the tissues. Attenuated virulence was also observed when 104 or 105 cells of R. erythropolis strain W2 were mixed with 106 cells of P. carotovorum strain Pcc797, per tuber. Under these conditions, a limited necrotic spot sometimes remained visible at the inoculation site (Fig. 4). The experiment was repeated three times with comparable results.



View larger version (52K):
[in this window]
[in a new window]
 
Fig. 4. Quenching of pectinolytic activities in Pectobacterium strain Pcc797 by R. erythropolis strain W2. Pectinolytic activities were assessed by visual inspection of maceration zones induced by the pathogen Pectobacterium upon inoculation of potato tuber, at two different sites, after 3 days, as indicated in Methods. A, negative control consisting of a tuber treated with 0·8 % NaCl; B, inoculation of R. erythropolis strain W2 (quencher) alone at about 106 c.f.u. per tuber; C, inoculation of Pectobacterium strain Pcc797 alone at about 106 c.f.u. per tuber; D, E, F, co-inoculation of Pectobacterium strain Pcc797 at about 106 c.f.u. per tuber, and decreasing cell density of the quencher W2 (D, 106 c.f.u. per tuber; E, 105 c.f.u. per tuber; F, 104 c.f.u. per tuber). The experiment was repeated three times with comparable results.

 

   DISCUSSION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
This work first aimed at isolating and identifying bacterial strains that degrade N-AHSL molecules. Six novel strains with this capability were obtained from tobacco rhizosphere and identified. Several indications support the view that these bacteria do degrade C6-HSL. First, they were isolated following an enrichment procedure based on the utilization of C6-HSL as sole carbon source. Second, C6-HSL disappeared from the enrichment medium, which became turbid concomitantly. Third, the six isolated strains neither inhibited the sensor strains responding to C6-HSL (as judged by reverse TLC), nor chemically degraded this molecule by increasing the pH of the growth media upon growth. In agreement with the above, strains W2 and D1 grew in liquid AT medium supplemented with C6-HSL as sole carbon source.

Out of 54 isolates analysed, only 25 were N-AHSL-degrading strains. Two explanations (at least) may be proposed to account for the occurrence of the 29 strains not degrading the QS molecules. First, they might exhibit very slow degradation that was not revealed by our experimental procedure. Second, they may have grown at the expense of some of the N-AHSL degradation products generated and released into the media by the ‘true’ degraders, as reported for the couple Arthrobacter/Variovorax (Flagan et al., 2003).

This work allowed the isolation of many more strains degrading N-AHSL than described in previous reports. Six isolates were finally identified using a polyphasic approach (ribotyping, Gram determination, morphology examination, Biolog analysis). They fell within the genera Comamonas, Pseudomonas, Rhodococcus and Variovorax. After the enrichment step, the diversity of the N-AHSL-degrading, cultivable strains isolated from the tobacco rhizosphere (which should not be extrapolated to other environments and may not be representative of the diversity prior to enrichment) may be estimated from the data reported in Table 1. The isolates degrading N-AHSL were essentially strains of Pseudomonas sp. (about 64 %), while three other genera were isolated at much lower frequencies (Comamonas 16 %;Variovorax 12 %; Rhodococcus 8 %). One of the strains, Cwt6, was identified as Variovorax paradoxus, a species that has been already described as capable of N-AHSL degradation (Leadbetter & Greenberg, 2000). N-AHSL-degrading ability does not appear to be limited to this single Variovorax species because another isolate (K6) belongs to this genus but to another, as yet unidentified, species. A major result generated by this study is the identification of N-AHSL-degrading bacteria in several phylogenetic groups where such a catabolic ability had not been described previously, i.e. the genera Pseudomonas and Comamonas. Remarkably, the most efficient N-AHSL-degrading isolate (W2) belonged to the species R. erythropolis. This strain is believed to be the first Rhodococcus strain known to degrade N-AHSL. Overall, the ability to degrade N-AHSL now appears to implicate unrelated bacterial genera belonging to the {alpha}-Proteobacteria (i.e. Agrobacterium, Zhang et al., 2002), the {beta}-Proteobacteria (Variovorax, Leadbetter & Greenberg, 2000; Ralstonia, Lin et al., 2003; and Comamonas, this work), the {gamma}-Proteobacteria (Pseudomonas, this work), the low-G+C Gram-positive bacteria (Bacillus, Dong et al., 2002; Lee et al., 2002) and the high-G+C Gram-positive bacteria (Rhodococcus, this work).

A second objective of the work was to study the specificity of the N-AHSL degradation ability of the rhizosphere isolates. This degradative ability differed noticeably from one strain to another (Fig. 2). A possibility exists, therefore, that the degradation pathways differ among isolates. Though this has not been formally demonstrated, bacteria with a broad degradation range possibly target a ‘conserved region’ of the N-AHSL molecules, for instance the lactone ring or the amide bond, while those exhibiting a narrower degradation range might rather catalyse modification(s) of a non-conserved region(s). In this study, bacteria were selected for their ability to degrade C6-HSL. The analysis of their catabolic ability clearly demonstrates that they indeed efficiently degrade this molecule (Fig. 2). Interestingly, these isolates degrade the oxo- derivative of the C6-HSL much less efficiently. This is in contrast with the work by Leadbetter & Greenberg (2000), who showed that 3-oxo-C6-HSL was utilized as a carbon source more efficiently than C6-HSL by an HSL-degrading Variovorax. This bacterium was, however, isolated from soil primarily on the basis of its ability to utilize 3-oxo-C6 HSL. We therefore speculate that the substitution status of the C3 on the acyl chain of the N-AHSL may affect the degradation rate in situ, to a certain extent. This speculation is in agreement with the observation that an aminopeptidase from Streptomyces rimosus is inhibited by a dipeptide analogue with hydroxy- and amino- substitutions close to the amide bond (Repic Lampret et al., 1999). In relation to this question, the ability of a root-associated micro-organism to degrade the QS signal may indeed constitute a trait conferring a selective advantage to this micro-organism, over other bacterial cells living in the rhizosphere. This may be true, for instance, for those microbes living in the same environment as bacteria regulating the production of antibiotic molecules via QS (Pierson et al., 1998).

The last goal of this study was to explore the potential use of QS-degrading strains to antagonize QS-regulatory processes. We chose to set up quenching experiments using R. erythropolis strain W2 as an interfering agent, because degradation assays demonstrated that it exhibits a very efficient N-AHSL-degrading activity. The quenching experiments established with no doubt that this microbe is able to interfere with three different QS-regulated functions in other microbes, i.e. production of the antibiotic pigment violacein by C. violaceaum, dissemination of the Ti plasmid by A. tumefaciens, and pathogenicity of P. carotovorum, in planta. We verified that this phenomenon was not due to a possible inhibition of the sensing of the N-AHSL signal. We also verified that the phenomenon could not be attributed to growth inhibition of any of the three ‘target’ microbes, as all three exhibited, in vitro, a similar growth pattern in the presence and in the absence of the quencher strain W2. Besides, the quenching ability of W2 was clearly compensated by increasing the concentration of N-AHSL, as documented in Fig. 3. As mentioned in the Introduction, limited data are available related to strategies for biological control based on N-AHSL degradation. One published experiment involved a pathogenic Pseudomonas strain that was rendered unable to produce N-AHSL by incorporation of several copies of the aiiA gene from Bacillus (Reimmann et al., 2002), a gene that encodes an N-AHSL lactonase (Dong et al., 2000). A second experiment involved transgenic plants expressing the same N-AHSL lactonase gene (Dong et al., 2001). Our data from natural isolates provide the first demonstration that a ‘wild-type’ microbe degrading N-AHSL may interfere with QS-regulated functions of several other micro-organisms, whether these are nonpathogenic organisms (Chromobacterium) or plant pathogens (Agrobacterium and Pectobacterium). It should be pointed out that Rhodococcus strain W2, which appears to be a potent interfering strain, was selected for its ability to degrade C6-HSL, and not because it limited or prevented growth of the pathogen(s).

In terms of biocontrol, future work may focus on the isolation of N-AHSL-degrading agents exhibiting a limited number of N-AHSL targets specific for a deleterious microbe or microbial function. Indeed, such agents could be used without harming or perturbing other QS-regulated functions in nontargeted micro-organisms. The strategy presented in this study appears to be very flexible with respect to environments, target organisms and functions. For instance, our work shows that an efficient quenching activity is not dependent upon the generation of transgenic plants. As a consequence, this modus operandi may be used to control bacterial pathogen(s) affecting plant species or cultivars recalcitrant to transformation, and, beyond phytopathology, deleterious QS-regulated functions of any microbial agents living in environments of commercial and medical interest. Furthermore, a similar approach can be proposed for QS mediator molecules other than N-AHSL, such as oligopeptides, {gamma}-butyrolactones or AI-2-like molecules, once their biological activity is understood, and once catabolic strains with the appropriate degradative ability are isolated.


   ACKNOWLEDGEMENTS
 
This work was made possible by grants from the EU (programmes ‘Biotech 4’ and ‘Ecosafe’) to Y. D. S. U. is a PhD student supported by a fellowship from the French Ministère de la Recherche et de la Technologie. The authors thank Drs Andrea Hardman and Paul Williams (Nottingham) for their interest, helpful comments and discussion on this work, and for kindly providing us with some N-AHSL samples. The authors also thank Claudine Elmerich (Gif-sur-Yvette), Bruno Smadja, Xavier Latour and Nicole Orange (Evreux) for helpful discussions and tips on assaying pathogenicity of Pectobacterium.


   REFERENCES
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Bauer, W. D. & Robinson, J. B. (2002). Disruption of bacterial quorum sensing by other organisms. Curr Opin Biotechnol 13, 234–237.[CrossRef][Medline]

Byers, J. T., Lucas, C., Salmond, G. P. & Welch, M. (2002). Nonenzymatic turnover of an Erwinia carotovora quorum-sensing signalling molecule. J Bacteriol 184, 1163–1171.[Abstract/Free Full Text]

Cha, C., Gao, P., Chen, Y. C., Shaw, P. D. & Farrand, S. K. (1998). Production of acyl-homoserine lactone quorum-sensing signals by gram-negative plant-associated bacteria. Mol Plant Microbe Interact 11, 1119–1129.[Medline]

Dong, Y. H., Xu, J. L., Li, X. Z. & Zhang, L. H. (2000). AiiA, an enzyme that inactivates the acylhomoserine lactone quorum-sensing signal and attenuates the virulence of Erwinia carotovora. Proc Natl Acad Sci 97, 3526–3531.[Abstract/Free Full Text]

Dong, Y. H., Wang, L. H., Xu, J. L., Zhang, H. B., Zhang, X. F. & Zhang, L. H. (2001). Quenching quorum-sensing-dependent bacterial infection by an N-acyl homoserine lactonase. Nature 411, 813–817.[CrossRef][Medline]

Dong, Y. H., Gusti, A. R., Zhang, Q., Xu, J. L. & Zhang, L. H. (2002). Identification of quorum-quenching N-acyl homoserine lactonases from Bacillus species. Appl Environ Microbiol 68, 1754–1759.[Abstract/Free Full Text]

Eberhard, A., Burlingame, A. L., Eberhard, C., Kenyon, G. L., Nealson, K. H. & Oppenheimer, N. J. (1981). Structural identification of autoinducer of Photobacterium fischeri luciferase. Biochemistry 20, 2444–2449.[Medline]

Elasri, M., Delorme, S., Lemanceau, P., Stewart, G., Laue, B., Glickmann, E., Oger, P. M. & Dessaux, Y. (2001). Acyl-homoserine lactone production is more common among plant-associated Pseudomonas spp. than among soilborne Pseudomonas spp. Appl Environ Microbiol 67, 1198–1209.[Abstract/Free Full Text]

Erickson, D. L., Nsereko, V. L., Morgavi, D. P., Selinger, L. B., Rode, L. M. & Beauchemin, K. A. (2002). Evidence of quorum sensing in the rumen ecosystem: detection of N-acyl homoserine lactone autoinducers in ruminal contents. Can J Microbiol 48, 374–378.[Medline]

Finch, R. G., Pritchard, D. L., Bycroft, B. W., Williams, P. & Stewart, G. S. A. B. (1998). Quorum sensing: a novel target for anti-infective therapy. J Antimicrob Chemother 42, 569–571.[Free Full Text]

Flagan, S., Ching, W. K. & Leadbetter, J. R. (2003). Arthrobacter strain VAI-A utilizes acyl-homoserine lactone inactivation products and stimulates quorum signal biodegradation by Variovorax paradoxus. Appl Environ Microbiol 69, 909–916.[Abstract/Free Full Text]

Fray, R. G. (2002). Altering plant-microbe interaction through artificially manipulating bacterial quorum sensing. Ann Bot 89, 245–253.[Abstract/Free Full Text]

Fuqua, W. C., Winans, S. C. & Greenberg, E. P. (1994). Quorum sensing in bacteria: the LuxR-LuxI family of cell density-responsive transcriptional regulators. J Bacteriol 176, 269–275.[Medline]

Fuqua, C., Parsek, M. R. & Greenberg, E. P. (2001). Regulation of gene expression by cell-to-cell communication: acyl-homoserine lactone quorum sensing. Annu Rev Genet 35, 439–468.[CrossRef][Medline]

Gram, L., de Nys, R., Maximilien, R., Givskov, M., Steinberg, P. & Kjelleberg, S. (1996). Inhibitory effects of secondary metabolites from the red alga Delisea pulchra on swarming motility of Proteus mirabilis. Appl Environ Microbiol 62, 4284–4287.[Abstract]

Hauben, L., Moore, E. R., Vauterin, L., Steenackers, M., Mergaert, J., Verdonck, L. & Swings, J. (1998). Phylogenetic position of phytopathogens within the Enterobacteriaceae. Syst Appl Microbiol 21, 384–397.[Medline]

Huguet, V., McCray Batzli, J., Zimpfer, J. F., Normand, P., Dawson, J. O. & Fernandez, M. P. (2001). Diversity and specificity of Frankia strains in nodules of sympatric Myrica gale, Alnus incana, and Shepherdia canadensis determined by rrs gene polymorphism. Appl Environ Microbiol 67, 2116–2122.[Abstract/Free Full Text]

Hwang, I., Li, P. L., Zhang, L., Piper, K. R., Cook, D. M., Tate, M. E. & Farrand, S. K. (1994). TraI, a LuxI homologue, is responsible for production of conjugation factor, the Ti plasmid N-acylhomoserine lactone autoinducer. Proc Natl Acad Sci U S A 91, 4639–4643.[Abstract]

Latifi, A., Winson, M. K., Foglino, M., Bycroft, B. W., Stewart, G. S., Lazdunski, A. & Williams, P. (1995). Multiple homologues of LuxR and LuxI control expression of virulence determinants and secondary metabolites through quorum sensing in Pseudomonas aeruginosa PAO1. Mol Microbiol 17, 333–343.[Medline]

Leadbetter, J. R. & Greenberg, E. P. (2000). Metabolism of acyl-homoserine lactone quorum-sensing signals by Variovorax paradoxus. J Bacteriol 18, 6921–6926.[CrossRef]

Lee, S. J., Park, S. Y., Lee, J. J., Yum, D. Y., Koo, B. T. & Lee, J. K. (2002). Genes encoding the N-acyl homoserine lactone-degrading enzyme are widespread in many subspecies of Bacillus thuringiensis. Appl Environ Microbiol 68, 3919–3924.[Abstract/Free Full Text]

Lin, Y. H., Xu, J. L., Hu, J., Wang, L. H., Ong, S. L., Leadbetter, J. R. & Zhang, L. H. (2003). Acyl-homoserine lactone acylase from Ralstonia strain XJ12B represents a novel and potent class of quorum-quenching enzymes. Mol Microbiol 47, 849–860.[CrossRef][Medline]

Lindhal, V. & Bakken, L. R. (1995). Evaluation of methods for extraction of bacteria from soil. FEMS Microbiol Ecol 16, 135–142.[CrossRef]

Lojkowska, E., Masclaux, C., Boccara, M., Robert-Baudouy, J. & Hugouvieux-Cotte-Pattat, N. (1995). Characterization of the pelL gene encoding a novel pectate lyase of Erwinia chrysanthemi 3937. Mol Microbiol 16, 1183–1195.[Medline]

McClean, K. H., Winson, M. K., Fish, L. & 9 other authors (1997). Quorum sensing and Chromobacterium violaceum: exploitation of violacein production and inhibition for the detection of N-acyl homoserine lactones. Microbiology 143, 3703–3711.[Abstract]

Oger, P., Dessaux, Y., Petit, A., Gardan, L., Manceau, C., Chomel, C. & Nesme, X. (1998). Validity, sensitivity and resolution limit of the PCR-RFLP analysis of the rrs (16S rRNA gene) as a tool to identify soil-borne and plant-associated bacterial populations. Genetics Selection Evolution 30, 311–321.

Petit, A. & Tempé, J. (1978). Isolation of Agrobacterium Ti plasmid regulatory mutants. Mol Gen Genet 167, 147–155.

Pierson, L. S., III, Wood, D. W. & Pierson, E. A. (1998). Homoserine lactone-mediated gene regulation in plant-associated bacteria. Annu Rev Phytopathol 36, 207–225.[CrossRef]

Reimmann, C., Ginet, N., Michel, L. & 9 other authors (2002). Genetically programmed autoinducer destruction reduces virulence gene expression and swarming motility in Pseudomonas aeruginosa PAO1. Microbiology 148, 923–932.[Abstract/Free Full Text]

Repic Lampret, B., Kidric, J., Kralj, B., Vitale, L., Pokorny, M. & Renko, M. (1999). Lapstatin, a new aminopeptidase inhibitor produced by Streptomyces rimosus, inhibits aminopeptidases. Arch Microbiol 171, 397–404.[Medline]

Reverchon, S., Bouillant, M.-L., Salmond, G. & Nasser, W. (1998). Integration of the quorum-sensing system in the regulatory networks controlling virulence factor synthesis in Erwinia chrysanthemi. Mol Microbiol 29, 1407–1418.[CrossRef][Medline]

Robson, N. D., Cox, A. R., McGowan, S. J., Bycroft, B. W. & Salmond, G. P. (1997). Bacterial N-acyl-homoserine-lactone-dependent signalling and its potential biotechnological applications. Trends Biotechnol 15, 458–464.[CrossRef][Medline]

Schaefer, A. L., Val, D. L., Hanzelka, B. L., Cronan, J. E. & Greenberg, E. P. (1996). Generation of cell-to-cell signals in quorum sensing: acyl homoserine lactone synthase activity of a purified Vibrio fischeri LuxI protein. Proc Natl Acad Sci U S A 93, 9505–9509.[Abstract/Free Full Text]

Shaw, P. D., Ping, G., Daly, S. L., Cha, C., Cronan, J. E., Jr, Rinehart, K. L. & Farrand, S. K. (1997). Detecting and characterizing N-acyl-homoserine lactone signal molecules by thin layer chromatography. Biochem J 94, 6036–6041.

Teplistki, M., Robinson, J. B. & Bauer, W. D. (2000). Plants secrete substances that mimic bacterial N-acyl homoserine lactone signal activities and affect population density-dependent behaviours in associated bacteria. Mol Plant-Microbe Interact 13, 637–648.[Medline]

Vaudequin-Dransart, V., Petit, A., Poncet, C. & 6 other authors (1995). Novel Ti plasmids in Agrobacterium strains isolated from fig tree and chrysanthemum tumors and their opinelike molecules. Mol Plant-Microbe Interact 8, 311–321.[Medline]

Versalovic, J., Koeuth, T. & Lupski, J. R. (1991). Distribution of repetitive DNA sequences in eubacteria and application to fingerprinting of bacterial genomes. Nucleic Acids Res 19, 6823–6831.[Abstract]

Winans, S. C. & Bassler, B. L. (2002). Mob psychology. J Bacteriol 184, 873–883.[Free Full Text]

Yates, E. A., Philipp, B., Buckley, C. & 8 other authors (2002). N-Acylhomoserine lactones undergo lactonolysis in a pH-, temperature-, and acyl chain length-dependent manner during growth of Yersinia pseudotuberculosis and Pseudomonas aeruginosa. Infect Immun 70, 5635–5646.[Abstract/Free Full Text]

Zhang, L., Murphy, P. J., Kerr, A. & Tate, M. E. (1993). Agrobacterium conjugation and gene regulation by N-acyl-L-homoserine lactones. Nature 362, 446–450.[CrossRef][Medline]

Zhang, H. B., Wang, L. H. & Zhang, L. H. (2002). Genetic control of quorum-sensing signal turnover in Agrobacterium tumefaciens. Proc Natl Acad Sci U S A 99, 4638–4643.[Abstract/Free Full Text]

Received 28 March 2003; revised 12 May 2003; accepted 12 May 2003.