Microbes Environ. Vol. 24, No. 4, 305–314, 2009 http://wwwsoc.nii.ac.jp/jsme2/ doi:10.1264/jsme2.ME09155

Distribution of N-Acylhomoserine Lactone-Producing Fluorescent Pseudomonads in the Phyllosphere and of (Solanum tuberosum L.)

NOBUTAKA SOMEYA1*, TOMOHIRO MOROHOSHI2, NOBUYA OKANO2, EIKO OTSU1, KAZUEI USUKI1, MITSURU SAYAMA1, HIROYUKI SEKIGUCHI1, TSUKASA IKEDA2, and SHIGEKI ISHIDA1 1National Agricultural Research Center for Hokkaido Region (NARCH), National Agriculture and Food Research Organization (NARO), 9–4 Shinsei-minami, Memuro-cho, Kasai-gun, Hokkaido 082–0081, Japan; and 2Department of Applied Chemistry, Utsunomiya University, 7–1–2 Yoto, Utsunomiya 321–8585, Japan (Received August 17, 2009—Accepted October 3, 2009—Published online October 30, 2009)

Four hundred and fifty nine isolates of fluorescent pseudomonads were obtained from the leaves and roots of potato plants. Of these, 20 leaf isolates and 28 root isolates induced violacein production in two N-acylhomoserine lactone (AHL)-reporter strains—Chromobacterium violaceum CV026 and VIR24. VIR24 is a new reporter strain for long N- acyl-chain-homoserine lactones, which can not be detected by CV026. Thin-layer chromatography revealed that the isolates produced multiple AHL molecules. We compared the 16S rRNA gene sequences of these isolates with sequences from a known database, and examined phylogenetic relationships. The AHL-producing isolates generally separated into three groups. Group I was mostly composed of leaf isolates, and group III, root isolates. Group II com- prised both leaf and root isolates. There was a correlation between the phylogenetic cluster and the AHL molecules produced and some phenotypic characteristics. Our study confirmed that AHL-producing fluorescent pseudomonads could be distinguished in the phyllosphere and rhizosphere of potato plants. Key words: fluorescent pseudomonad, quorum sensing, N-acylhomoserine lactone, Solanum tuberosum L., 16S rRNA gene

Potato (Solanum tuberosum L.) is one of the world’s most phenotypic characteristics of AHL-producing , espe- important crops. In Japan, many commercial cultivars are cially fluorescent pseudomonads present around potato susceptible to phytopathogens and require high-nutrient con- plants, have not. ditions for growth. Repeated pesticide applications and The objective of this study was to examine the composi- excessive fertilization may result in fungicide resistance, tion and production of AHL signal molecules produced by contamination, or harm to non-target organisms. We are try- fluorescent pseudomonad isolates within the total fluorescent ing to identify alternative disease management and fertiliza- pseudomonad communities present in the phyllosphere and tion strategies that are both environmentally friendly and rhizosphere of potato using a new long-acyl-chain-AHL inexpensive. One approach is the use of beneficial plant- reporter strain. We also examined the phenotypic character- related microorganisms such as biocontrol microorganisms istics such as biofilm formation, motility, produc- and plant growth-promoting microorganisms. tion and virulence to potato plants and phylogenetic relation- Fluorescent pseudomonads are globally distributed as ships of collected isolates. judged by their frequent isolation from diverse environ- ments, including the plant rhizosphere and phylloplane (24). Materials and Methods It has been reported that many isolates of pseudomonad Chemicals species from the phylloplane and roots are useful biocontrol N-Butyryl-DL-homoserine thiolactone (C4HCTL), C4HSL, N- and plant growth-promoting agents (13, 34, 35). Fluorescent butyryl-DL-homoserine lactone (C4HSL), N-(β-ketocaproyl)-L- pseudomonads are known to have several biocontrol and homoserine lactone (3OC6HSL), N-hexanoyl-DL-homoserine plant growth-promoting traits such as production, lactone (C6HSL), N-heptanoyl-DL-homoserine lactone (C7HSL), competition, production of plant growth-promoting sub- N-(3-oxooctanoyl)-L-homoserine lactone (3OC8HSL), N-octanoyl- DL DL stances and induction of resistance to host plants (3, 32). -homoserine lactone (C8HSL), N-decanoyl- -homoserine lactone (C10HSL), N-dodecanoyl-DL-homoserine lactone (C12HSL), These traits are regulated by complex factors in each bacte- and N-tetradecanoyl-DL-homoserine lactone (C14HSL) were used rium. Recently, it was shown that many of the beneficial as synthetic AHL-standards. All chemicals were purchased from traits of fluorescent pseudomonads are regulated by the Sigma-Aldrich (St. Louis, MO, USA). quorum-sensing system that functions via N-acylhomoserine Synthetic phenazine-1-carboxylic acid (PCA; Pharmeks, Mos- lactone (AHL) signal molecules (42). Although the diversity cow, Russia) was used as the control in the thin-layer chromatogra- of AHL-producing bacteria in some plant and phy (TLC) assay for the detection of PCA produced by the isolates. phylloplanes has been reported (9, 45), the distribution and Bacterial strains and culture conditions Chromobacterium violaceum ATCC 31532 produces C6HSL as * Corresponding author. E-mail: [email protected]; Tel: +81– a signal molecule for the quorum-sensing system. C. violaceum 155–62–9280; Fax: +81–155–61–2127. CV017 is a spontaneous streptomycin-resistant mutant of ATCC 306 SOMEYA et al.

31532. C. violaceum CV026, a reporter strain, is a non-C6HSL units (CFU) mL−1 of strain CV026 or VIR24 was streaked onto LB producer derived from CV017 by a mini-Tn5 insertion in the cviI agar medium. Each fluorescent pseudomonad isolate obtained from gene encoding the AHL synthase, LuxI homolog CviI (16). The potato plants was inoculated onto LB agar plates using a sterile C. violaceum type strain ATCC 12472 was used for the devel- toothpick. The tested bacterium was positioned 10 mm from the opment of a new reporter strain for long N-acyl-chain-homoserine streak of the reporter strains. The plates were incubated for 72 h at lactones. Both strains of C. violaceum were grown at 28°C in Luria- 25°C. After incubation, violacein (purple pigment) biosynthesis was Bertani (LB; Sigma-Aldrich) medium. examined at the boundary of each reporter strain with the tested iso- Fluorescent pseudomonad isolates were cultured on King’s B lates. medium (KB; Eiken Chemical, Tokyo, Japan). To examine biofilm To examine the violacein-inducing activity of each isolate, 10 formation by these isolates, a glucose-supplied basal salts medium mg of cells from the reporter strains used in the above assay was (GBS) was used, which had the following composition: (NH4)2SO4, placed in a 1.5-mL microtube. The cells were lysed by adding 100 1.1 g; K2HPO4, 2.29 g; KH2PO4, 0.9 g; MgSO4·7H2O, 5 mg; 5 mmol μL of 10% (w/v) sodium dodecylsulfate (SDS) and mixing for 10 s of glucose; and sterilized distilled water (SDW), 1 L (46). on a vortex mixer. Violacein was extracted from the cell lysate by For the potato tuber putrefaction assay, a pathogenic bacterium adding 900 μL of water-saturated butanol, vortexing for 5 s, and Pectobacterium carotovorum subsp. carotovorum NBRC 3830 was centrifuging for 5 min. Subsequently, 150 μL of each extract was used as the positive control. Strain NBRC 3830 was cultured on LB analyzed at 585 nm using a DU7400 spectrophotometer (Beckman agar medium. Coulter, CA, USA). Violacein-inducing activity was classified on the basis of A585 values, as follows: −, smaller than 0.01; ±, 0.01 to Development of a new reporter strain for long N-acyl-chain- 0.05; +, 0.05 to 0.5; ++, 0.5 to 1.0; and +++, greater than 1.0. Each homoserine lactones experiment was repeated three times. The violacein-inducing iso- The C. violaceum type strain ATCC 12472 produces several lates of fluorescent pseudomonads were selected as AHL-producers AHLs such as N-(3-hydroxydecanoyl)-L-homoserine lactone (20). and used for further analysis. An in-frame deletion mutant of the cviI gene encoding AHL synthase in ATCC 12472 was generated by homologous recombi- Extraction, detection and characterization of AHLs nation with sacB selection. DNA flanking sequences upstream and A cell suspension (1 μL) containing each AHL-producing isolate downstream of the cviI gene were amplified using genomic DNA at 108 CFU mL−1 was incubated in 5 mL of liquid LB medium on a from ATCC 12472 as the template and the following primer sets that reciprocal shaker (140 strokes min−1) for 72 h. The bacterial cells contained MluI, BglII, or SalI restriction sites (underlined): 5'-ACG- were collected by centrifugation (5,100×g for 10 min), and the CGTACTGAGCCGACTTGTCTATTTCCG-3' and 5'-AGATCT- supernatant was filtered through an 0.20-μm filter. Culture filtrates AGTACCAGTCCACCTTGTTGCAGC-3' for the upstream region, were subsequently extracted three times with an equivalent volume and 5'-AGATCTCGAAAACCGAGCTTATCCGTTCAC-3' and of ethyl acetate. After air-drying at room temperature, each extract 5'-GTCGACACCAGTACTCCCAGGAAAACTTCG-3' for the was re-dissolved in 50 μL of dimethyl sulfoxide. The sample (10 downstream region. The two PCR products were cloned into the μL, which contains extracts from 1 mL of culture filtrate) was spot- pGEM-T easy cloning vector (Promega, Madison, WI, USA) to ted onto a C18 reverse-phase TLC plate (KC18 Silica Gel 60A, construct pCV1 and pCV2. The regions upstream and downstream Whatman International, Maidstone, UK) and developed with a of the cviI gene were released from MluI-BglII-treated pCV1 and methanol-water solution (60:40, v/v). After elution, the plates were SalI-BglII-treated pCV2, respectively. Both DNA fragments were overlaid with 0.7% (w/v) agar containing 108 CFU mL−1 of inserted into the MluI-SalI-treated plasmid pGP704Sac38 (17) to the reporter strain, either CV026 or VIR24. The AHLs were char- create pGP-CV1, which contains a novel BglII site at the site of acterized on the basis of the Rf values and shapes of the spots. joining and a copy of a truncated cviI that lacks 80% of the coding Ten different AHLs were used as the standards. These were region. The 1.3-kbp Kanr cassette from PstI-digested pKPR11 (30) C4HCTL, C4HSL, 3OC6HSL, C6HSL, C7HSL, 3OC8HSL, was also inserted into the PstI site of pGP-CV1 to create pGP-CV2. C8HSL, C10HSL, C12HSL, and C14HSL. Each experiment was Chromosomal cviI in ATCC 12472 was deleted by bacterial conju- repeated three times. gation and homologous recombination (17). ATCC 12472 and E. coli S17-1 λpir harboring pGP-CV2 were conjugated. Isolates cor- 16S rRNA gene sequencing and computer analysis of the DNA responding to single-crossover events were selected on the basis of sequence kanamycin resistance; these were purified and streaked onto an LB The 16S rRNA genes of the AHL-producing isolates were first agar plate with 5% (w/v) sucrose to select homologous recombi- amplified by PCR with the Taq DNA polymerase and the 63f nants. The presence of the expected in-frame cviI deletion in ATCC (5'-CAGGCCTAACACATGCAAGTC-3') and 1387r (5'-GGGCG- 12472 was confirmed by PCR. The in-frame cviI deletion mutant GWGTGTACAAGGC-3') primers and then sequenced, as described was designated VIR24 (Viocacein Inducible Reporter no. 24). previously (36). The sequences obtained were edited, and consensus sequences were assembled using Genetyx-Mac (Genetyx, Tokyo, Plant sampling and isolation of fluorescent pseudomonads Japan). Similarity searches were performed to compare these nucle- Forty leaf and 60 root samples of potato (Solanum tuberosum L.) otide sequences with those in DDBJ-BLAST for the closest matches were collected from both commercial and experimental fields and NCBI-BLAST for the nearest matches. For the phylogenetic located in the Hokkaido, Ibaraki, Kanagawa, Fukuoka, Nagasaki, tree, the 16S rRNA gene sequences from 129 type strains of the and Kagoshima prefectures of Japan. One gram of root or leaf sam- genus were used, and sequences from 17 type strains ple was washed with 9 mL of sterile 15 mM phosphate buffer (pH of the genus Acinetobacter were used as the out-group. The 7.0), and serial dilutions were cultivated on KB medium agar con- CLUSTAL_X program was used for the phylogenetic analysis of taining 50 μg mL−1 cycloheximide; these were incubated for 3 d the sequence datasets (40), and a tree was constructed by the neigh- in the dark. After cultivation, fluorescent pseudomonad colonies bor-joining method (23, 31). were selected by ultraviolet irradiation at 365 nm using an ultra- violet lamp (MODEL UVGL-58; UVP, Upland, CA, USA). Five Phenotypic characteristics of AHL producers colonies were randomly selected from the fluorescent colonies in Various characteristics of pseudomonads, including biofilm for- each sample. mation, motility, pathogenicity to plants and animals, and produc- tion of secondary metabolites such as PCA, are reported to be under Detection of isolates that induce violacein biosynthesis in AHL the control of the quorum-sensing system that functions via AHL reporters signal molecules (42). A cell suspension containing approximately 109 colony-forming Biofilm formation by the AHL-producing isolates was monitored AHL-Producing Fluorescent Pseudomonads 307 according to a method described previously (46). Bacterial cells Table 1. Induction of violacein biosynthesis by N-acylhomoserine were pre-cultured by shaking in LB broth for 48 h at 25°C and then lactones (AHLs) in the Chromobacterium violaceum washed twice by centrifugation (9,000×g for 5 min). Subsequently, reporter strains CV026 and VIR24 using the cross-streak assay their OD595 value was adjusted to 0.11–0.13 using SDW, and they were used as inoculums for the assay. The inocula were placed in CV026 VIR24 the wells of 96-well polystylene microtitre dishes (Corning Glass Works, Corning, NY, USA) containing 100 μL of GBS. After incu- AHLsa) Conc. (M) Conc. (M) bation, the culture medium was aspirated from each well, and the 0 10−6 10−5 10−4 10−3 10−2 0 10−6 10−5 10−4 10−3 10−2 wells were air-dried. Each well was stained with 100 μL of a 1% b) −−−−+−−−−−− (w/v) crystal violet solution at room temperature for 15 min, which C4T − was followed by five thorough washes with SDW. The adhered C4 −−−−−+++−−−−−− crystal violet stain was dissolved in 200 μL of 95% (v/v) ethanol, 3OC6 −−−−−+−−−± + +++ and the biofilm formed in each well was quantified by measur- C6 −−−−+ +++ − − − ± + ++ ing the OD595 value using a GeneQuant pro spectrophotometer C7 −−−± ++ +++ − − ± + ++ +++ (Amersham Pharmacia Biotech., Uppsala, Sweden). 3OC8 −−−−−+−−± + +++ +++ To examine the motility of isolates, each isolate was incubated in C8 −−−−−+−−−+ +++ +++ liquid KB medium for 24 h at 25°C and washed twice by centrifu- C10 −−−−−−−−+ +++ +++ +++ gation (9,000×g for 5 min). Subsequently, the concentration was C12 −−−−−−−−−++ +++ +++ 7 −1 μ adjusted to 10 CFU mL using SDW. A cell suspension (1 L) C14 −−−−−−−−−± + + was inoculated onto the center of plates containing the MAKCS a) medium (41) in 0.28% (w/v) agar (for swimming motility) or 0.4% C4T, N-butyryl-DL-homoserine thiolactone; C4, N-butyryl-DL-homo- (w/v) agar (for swarming motility). The plates were incubated in the serine lactone; 3OC6, N-(β-ketocaproyl)-L-homoserine lactone; C6, dark for 24 h at 25°C. After incubation, the radial extension of the N-hexanoyl-DL-homoserine lactone; C7, N-heptanoyl-DL-homo- colonies was measured. serine lactone; 3OC8, N-(3-oxooctanoyl)-L-homoserine lactone; C8, To examine PCA production by the isolates, the PCA bio- N-octanoyl-DL-homoserine lactone; C10, N-decanoyl-DL-homoserine DL synthesis gene phzCD was first screened by PCR using the specific lactone; C12, N-dodecanoyl- -homoserine lactone; C14, N-tetra- decanoyl-DL-homoserine lactone. primers PCA2a (5'-TTGCCAAGCCTCGCTCCAAC-3') and PCA3b b) −, no induction; ±, slight; +, weak; ++, moderate; +++, strong. The (5'-CCGCGTTGTTCCTCGTTCAT-3') (29). The positive isolates violacein concentration was determined by A585, and is indicated by were incubated on KB medium agar for 72 h, and PCA was the shaded areas. extracted with an equal volume of ethyl acetate. The PCA present in the extracts was detected by TLC assays (27). To examine the influence of the isolates on the potato plants, (>10−4 M), C7HSL (>10−5 M), 3OC8HSL (>10−5 M), C8HSL leaves and tuber slices were inoculated with each isolate. The (>10−4 M), C10HSL (>10−5 M), C12HSL (>10−4 M), and detached leaves of three major cultivars (cv. Irish Cobbler, cv. May C14HSL (>10−4 M) but was not observed in the presence of Queen, and cv. Toyoshiro) of potato were placed in a plastic tray short-acyl-chain HSLs such as C4HCTL and C4HSL. This with wet filter paper to keep them moist. A bacterial cell suspension (10 μL of suspension containing 102, 104, and 106 CFU in phos- result indicated that the reporter strain VIR24 could detect phate-buffered saline) was inoculated into each leaf by the needle- long-acyl-chain HSL molecules that could not be detected by prick method. After incubation for 24 and 48 h, the induction of the CV026 strain using a simple cross-streak assay on agar necrosis was evaluated. The potato tuber slices (cv. Irish Cobbler) plates. were also placed in a plastic tray with wet filter paper to keep them moist. The bacterial cell suspension (10 μL of suspension contain- Induction of violacein biosynthesis in AHL reporters by ing 102, 104, and 106 CFU in phosphate-buffered saline) was applied potato isolates to the surfaces of potato tuber slices and incubated for 48 h at 25°C. P. carotovorum subsp. carotovorum NBRC 3830 was used as the When the KB medium was used for bacterial isolation, 3 5 5 8 positive control. The potato tuber slices were evaluated for decay approximately 10 to 10 CFU and 10 to10 CFU of bac- after 48 h. Each experiment was repeated twice. terial colonies were obtained from one gram (fresh weight) of potato leaves and roots, respectively. Fluorescent Results pseudomonads could be distinguished from other bacterial colonies when subjected to UV irradiation. Fluorescent colo- C. violaceum VIR24 is a new reporter strain of long nies could be randomly selected from culturable bacterial N-acyl-chain-homoserine lactones colonies at five isolates per leaf or root sample. A total of Both AHL-reporter strains, C. violaceum CV026 and 159 fluorescent pseudomonad isolates were collected from VIR24, produced violacein when cultured with AHL mole- leaves, while 300 were obtained from the roots. Isolates from cules. However, they differed in sensitivity to the AHL mole- leaves were designated StFLB (Solanum tuberosum-isolated cules in the simple cross-streak assay on agar plates (Table 1). Fluorescent Leaf Bacterium) and isolates from roots were For example, violacein production was induced in CV026 designated StFRB (S. tuberosum-isolated Fluorescent Rhizo- in the presence of C4HCTL (>10−2 M), C4HSL (>10−2 M), Bacterium), StBRB (S. tuberosum-isolated Brown Rhizo- 3OC6HSL (>10−2 M), C6HSL (>10−3 M), C7HSL (>10−4 M), Bacterium), and StRB (S. tuberosum-isolated RhizoBacte- 3OC8HSL (>10−2 M), and C8HSL (>10−2 M). C10HSL, rium). All of these isolates were obtained from samples taken C12HSL, and C14HSL, which are long-acyl-chain HSLs, did from different regions in Japan. not induce violacein production in CV026 (Table 1). In the Using the cross-streak assay, a total of 20 isolates (12.6%) present study, we developed a novel AHL reporter strain from leaves and 28 isolates (9.3%) from roots were found to named VIR24 that was derived from the C. violaceum type induce violacein production in the reporter strains. All the strain ATCC 12472. Violacein production in VIR24 was positive isolates induced violacein production in VIR24. observed in the presence of 3OC6HSL (>10−4 M), C6HSL However, only 4 leaf isolates and 4 root isolates induced 308 SOMEYA et al.

Table 2. Phylogenetic group, isolated location, AHL production, and violacein production in CV026 (Table 2). similar database sequences of the 16S rRNA gene of AHL- producing isolates from potato leaves and roots Characterization of AHL molecules by TLC analysis Closest Violacein TLC using strains CV026 and VIR24 revealed that bacte- Isolates Isolated induction Detected AHLs in database sequence, rial extracts contained one or more AHL molecules. Repre- (Groupa)) locationb) (CV026/ the TLC assayd) c) and % sentative results of TLC plates overlaid with each reporter VIR24) e) similarity are shown in Fig. 1. The pigment had Rf values that were in StFLB010 (I) NMH −/+ U.M. (Rf 0.24) A, 99% the same range as those of the synthetic AHL molecules used StFLB015 (I) KMH −/+ C7, 3OC6/C8, U.M. (Rf 0.24) A, 99% as standard markers (Table 2). 3OC6HSL and 3OC8HSL StFLB020 (I) SMH −/+ C8, 3OC6/C8, U.M. (Rf 0.41) A, 99% StFLB031 (I) SMH −/+ C8, 3OC6/C8, U.M. (Rf 0.42) A, 99% could not be distinguished on the basis of their Rf values StFLB034 (I) SMH −/+ C8, 3OC6/C8, U.M. (Rf 0.41) A, 99% under these assay conditions (Fig. 1A). It was also difficult StFLB041 (I) SMH −/+ C8, 3OC6/C8, U.M. (Rf 0.41) A, 99% to distinguish the C12HSL and C14HSL molecules (Fig. StFLB042 (I) SMH −/+ C8, 3OC6/C8, U.M. (Rf 0.41) A, 99% StFLB044 (I) BMH −/+ C8, 3OC6/C8, U.M. (Rf 0.41) A, 99% 1B). The TLC plate overlaid with CV026 (Fig. 1A) showed StFLB045 (I) BMH −/+ C8, 3OC6/C8, U.M. (Rf 0.24 A, 99% mainly short-acyl-chain HSL molecules, while that overlaid and 0.41) with VIR24 (Fig. 1B) showed mainly long-acyl-chain HSL StFLB046 (I) BMH −/+ C8, U.M. (Rf 0.24 and 0.41) A, 99% StFLB048 (I) BMH −/+ C8, U.M. (Rf 0.25 and 0.40) A, 99% molecules. However, some of the molecules detected did not StFLB051 (I) NMH −/+ C8, U.M. (Rf 0.43) A, 99% correspond to the 10 synthetic AHLs used, and were −/+ f StFLB136 (I) KKI C8, 3OC6/C8, U.M. (R 0.48, A, 99% regarded as undefined molecules (U.M.) (Table 2). 0.29) StFLB155 (I) FAK −/+ C8, C12/C14, 3OC6/C8, A, 99% U.M. (Rf 0.49, 0.29) Nucleotide sequence of the 16S rRNA gene from StFRB280 (I) KKI −/+ C8, C12/C14, 3OC6/C8, A, 99% AHL-producing isolates U.M. (Rf 0.48, 0.29) StFLB078 (IIA) SMK +/± C4, C6, 3OC6/C8, U.M. (Rf B, 100% We sequenced the 16S rRNA genes of 20 AHL-producing 0.62, 0.37, 0.25) isolates obtained from leaves and 28 collected from roots. StFLB079 (IIA) SMK +/± C4, C6, 3OC6/C8, U.M. (Rf B, 100% 0.62, 0.38, 0.25) The nucleotide sequences determined for the partial 16S StFLB092 (IIA) HKK +/+ C4, C6, 3OC6/C8, U.M. (Rf C, 99% rRNA gene sequences have been deposited in the DDBJ 0.62, 0.38, 0.25) database under the accession numbers AB506028– StFLB105 (IIA) NMK ±/± C4, C6, 3OC6/C8, U.M. (Rf B, 100% 0.61, 0.39, 0.26) AB506069 and AB512618–AB512623, as listed in Table 2. StFRB209 (IIA) HKK +/± C4, C6, C7, C8, U.M. (Rf D, 98% A closest match analysis showed that most of the sequences, 0.58) with the exception of StFRB209 (98%), had a similarity StFLB049 (IIB) BMH −/++ C8, U.M. (Rf 0.43) E, 100% StFRB032 (IIB) KII −/+++ C8, C10 F, 100% greater than 99% to known sequences from bacterial strains StFRB147 (IIB) HFF −/+++ C8, C10 G, 100% (Table 2). StFRB208 (IIB) HKK −/+++ C8 F, 100% Most isolates showed the highest degree of similarity to StFRB218 (IIB) SMK −/+++ C8 F, 99% StFRB219 (IIB) SMK −/+++ C8 F, 100% unidentified Pseudomonas isolates (Table 2). On the other StBRB008 (IIB) NMK +/+ C6, U.M. (Rf 0.69, 0.39) H, 100% hand, some isolates such as StFLB092, StBRB008 and +/+ f StBRB009 (IIB) NMK C6, U.M. (R 0.70, 0.39) H, 100% StBRB009 were most similar to identified Pseudomonas StRB046 (IIB) HKK +/+ C4, C6, C8, U.M. (Rf 0.58) I, 99% StFRB234 (IIIA) NMK −/++ N.D. J, 99% species, and further phylogenetic analysis supported these StFRB067 (IIIA) NMH −/+++ C8, C10 K, 100% StFRB074 (IIIB) KMH −/+++ C8, C10 K, 100% StFRB159 (IIIB) AUN −/+++ C8, C10 L, 100% StFLB016 (IIIC) KMH −/+ U.M. (Rf 0.24 and 0.41) M, 100%  moderate; +++, strong. The violacein concentration was deter- −/+ f mined by A585. StFRB097 (IIIC) BMH U.M. (R 0.20) N, 99% d) StFRB116 (IIIC) KMN −/+ 3OC6/C8, U.M. (Rf 0.21) O, 99% C4, C4HSL; C6, C6HSL, C8, C8HSL, C10, C10HSL, C12, C12HSL, StFRB117 (IIIC) KMN −/+ 3OC6/C8, U.M. (Rf 0.21) P, 99% C14, C14HSL; 3OC6, 3OC6HSL; and 3OC8, 3OC8HSL. U.M., StFRB118 (IIIC) KMN −/+ N.D. P, 100% undefined molecule corresponding to the 10 synthetic AHL mole- StFRB125 (IIIC) OUN −/+ C8, 3OC6/C8 O, 99% cules used in this study. The Rf value of each molecule is indicated. StFRB140 (IIIC) KMN −/++ C8, 3OC6/C8 O, 100% 3OC6/C8, 3OC6HSL and 3C8HSL showed the same Rf values in this StFRB142 (IIIC) KMN −/+ C8, 3OC6/C8 O, 100% assay. Therefore, these compounds could not be distinguished. C12/ StFRB163 (IIIC) OUN −/++ U.M. (Rf 0.43, 0.22) O, 100% C14, C12HSL and C14HSL showed similar Rf values in this assay. StFRB164 (IIIC) OUN −/++ U.M. (Rf 0.43, 0.22) O, 100% Therefore, these compounds could not be distinguished. N.D., StFRB165 (IIIC) OUN −/++ U.M. (Rf 0.43, 0.22) O, 100% StFRB118 and StFRB234 induced violacein production in the StFRB166 (IIIC) OUN −/++ U.M. (Rf 0.43, 0.23) O, 100% reporter strain in the cross-streak assay, but the signal molecules were −/+ f not detected in the TLC assay. StFRB167 (IIIC) OUN U.M. (R 0.43, 0.23) O, 100% e) StFRB285 (IIIC) FAK −/+ C8, 3OC6/C8 O, 99% On the basis of the 16S rRNA gene sequence (approximately 1,300 StFRB286 (IIIC) FAK −/+ C8, 3OC6/C8 O, 99% bp) and in comparison to the sequences in the database. Closest matches of the following database sequences; A, Pseudomonas sp. a) Group was determined by the phylogenetic analysis using the 16S HC001005-4 (EU364835); B, Pseudomonas syringae pv. tabaci rRNA gene sequence in Fig. 2. strain BC2367 (FJ755789); C, Pseudomonas viridiflava strain LMG b) Isolated at the following locations; NMH, Nakafushiko, Memuro, 2352T (Z76671); D, Pseudomonas syringae pv. tagetis MAFF Hokkaido; KMH, Kamifushiko, Memuro, Hokkaido; SMH, Shinsei, 302271 (AB001449); E, Pseudomonas sp. BR7-01 (EU853197); F, Memuro, Hokkaido; BMH, Bisei, Memuro, Hokkaido; SMK, Sata- Pseudomonas frederiksbergensis strain IMER-B4-8 (FJ796428); G, kori, Minamiosumi, Kagoshima; HKK, Hosoyamada, Kanoya, Pseudomonas sp. L22 (EF672047); H, Kagoshima; NMK, Nejime, Minamiosumi, Kagoshima; KKI, Kamis- subsp. aurantiaca NCIB 10068T (DQ682655); I, Pseudomonas awa, Kasumigaura, Ibaraki; FAK, Funako, Atsugi, Kanagawa; KII, aurantiaca strain B23 (EU169169); J, Pseudomonas sp. BR3-13 Kamiishizaki, Ibaraki, Ibaraki; KMN, Kazusa, Minamishimabara, (EU853185); K, Pseudomonas putida strain c58 (FJ950573); L, Nagasaki; OUN, Obama, Unzen, Nagasaki; HFF, Hakozaki, Fuku- Pseudomonas putida strain IS82 (FJ596989); M, Pseudomonas sp. oka, Fukuoka; and AUN, Aino, Unzen, Nagasaki. 13651W (EU741102); N, Pseudomonas sp. 2V1C (EU693555); O, c) Violacein induction in the Chromobacterium violaceum reporter Pseudomonas sp. s52 (EU099607); and P, Pseudomonas putida strain strains CV026 and VIR24. −, no induction; ±, slight; +, weak; ++,  fA5 (FJ947054). AHL-Producing Fluorescent Pseudomonads 309

Fig. 1. Detection of AHL molecules by bacterial isolates of fluorescent pseudomonads using the Chromobacterium violaceum reporter strains, CV026 (A) and VIR24 (B). Samples of extracts from each bacterial isolate were obtained as indicated in materials and methods. Lane M1 contains synthetic AHL molecules, N-butyryl-DL-homoserine lactone (C4), N-hexanoyl-DL-homoserine lactone (C6), N-heptanoyl-DL-homoserine lactone (C7), N-octanoyl-DL-homoserine lactone (C8), N-decanoyl-DL-homoserine lactone (C10), N-dodecanoyl-DL-homoserine lactone (C12) and N- tetradecanoyl-DL-homoserine lactone (C14). Lane M2 contains synthetic AHL molecules, N-butyryl-DL-homoserine thiolactone (C4T), N-(β- ketocaproyl)-L-homoserine lactone (3OC6), and N-(3-oxooctanoyl)-L-homoserine lactone (3OC8). Lane 1, StBRB009; 2, StBRB008; 3, StFRB116; 4, StFRB097; 5, StFRB074; 6, StFRB067; and 7, StFRB032. relationships. However, some isolates such as StRB046 and nea, as shown in Fig. 2 (6). StBRB008 and StBRB009 were isolates belonging to group IIIB, showed greatest similarity placed in the “Pseudomonas chlororaphis group” (4, 25), to identified specific Pseudomonas species, but their phylo- and produced C6 and two undefined AHL molecules. genetic clusters were closely grouped into another Pseudo- All group III isolates, except StFLB016, were from roots monas species (Table 2, Fig. 2). (Fig. 2). The group could be divided into three subgroups, IIIA, IIIB, and IIIC (Fig. 2). As group IIIA, StFRB234 was Phylogenetic analysis of the 16S rRNA gene from located into the cluster related to Pseudomonas umsongensis AHL-producing isolates isolated from agricultural soil (14). This isolate induced Based on results of a phylogenetic analysis with the 16S violacein production in a cross-streak assay, but we could rRNA gene from 129 type strains of the genus Pseudomonas, not identify the AHLs produced (Table 2). StFRB067, 48 AHL-producing isolates from potato plants were placed StFRB074, and StFRB159 in group IIIB were placed into the into three groups, I, II, and III (Fig. 2). cluster related to a grass-isolated species, Pseudomonas Group I included 15 isolates, 14 from leaf and one from graminis (5). All three of these isolates produced C8HSL and root, and all of which showed a close phylogenetic relation- C10HSL (Table 2). The IIIC isolates were grouped into the ship with Pseudomonas cichorii (Fig. 2). Most of the group I cluster related to Pseudomonas putida. They produced only isolates mainly produced C8HSL and 3OC6/C8HSL (Table one or two AHLs such as undefined AHL molecules or C8 2). and C6/C8HSL (Table 2). Group II comprised a mix of leaf isolates and root isolates. The group was divided into the two subgroups, IIA and IIB Phenotypic characteristics of AHL-producing isolates (Fig. 2). Group IIA included 4 leaf isolates, StFLB078, Biofilms formed to a slight extent in most cases, but some StFLB079, StFLB092, StFLB105, and one root isolate, isolates showed substantial biofilm formation (Table 3). StFRB209. These isolates showed a close phylogenetic rela- Groups I and III in the phylogenetic analysis (Fig. 2), were tionship with Pseudomonas ficuserectae, Pseudomonas mixed when it comes to biofilm formation. However, all syringae, and Pseudomonas viridiflava. All of group IIA three isolates, StFRB067, StFRB074, and StFRB159, except StFRB209 produced C4HSL, C6HSL and 3OC6/ belonging to group IIIB showed substantial biofilm forma- C8HSL. On the other hand, most isolates in group IIB were tion. Only one isolate in group IIIC, StFLB016, formed a from roots, and produced different AHLs in each cluster biofilm. In group II, only StFLB092 and StFRB209 belong- (Fig. 2, Table 2). Except for StBRB008 and StBRB009, all ing to the P. viridiflava cluster formed a biofilm. of group IIB produced C8HSL. StFRB147 and StFRB032 Most isolates showed both swimming and swarming in the also produced C10HSL in addition to C8HSL. StFRB032, MAKCS medium, but those belonging to group IIB mostly StFRB147, StFRB208, StFRB218, and StFRB219, showed did not show substantial motility. In the phylogenetic groups, the highest similarity to Pseudomonas frederiksbergensis IIA, IIIB, and IIIC, isolates belonging to the same cluster (1), and were grouped into the cluster closely related to P. showed similar characteristics of motility (Table 3, Fig. 2). frederiksbergensis. StRB046 produced C4, C6 and undefined Of all the isolates, only StBRB008 and StBRB009 pro- an AHL molecule but not C8HSL (Table 2). StRB046 was duced the antibiotic PCA (Table 3). We detected the grouped into the cluster related to Pseudomonas mediterra- phenazine synthesis-related phzCD genes in StBRB008 and 310 SOMEYA et al.

Fig. 2. Neighbor-joining trees of the 16S rRNA gene sequences obtained from AHL-producing fluorescent pseudomonad isolates (StFLB, Solanum tuberosum-isolated Fluorescent Leaf Bacterium; StFRB, S. tuberosum-isolated Fluorescent RhizoBacterium; StBRB, S. tuberosum- isolated Brown RhizoBacterium; StRB, S. tuberosum-isolated RhizoBacterium) from the leaf and root of potato (S. tuberosum L.). The bacterial isolates used in this study are shown in bold. From the results of a phylogenetic analysis with the 16S rRNA gene from 129 type strains of the genus Pseudomonas, 48 AHL-producing isolates from potato plants were placed into three groups, I, II (IIA and IIB), and III (IIIA, IIIB, and IIIC). The scale bar represents 0.05 substitutions per nucleotide position. Bootstrap values of 50 or more (from 100 replicates) are indicated at the nodes. Seventeen type strains of the genus Acinetobacter were used as the root organisms.

StBRB009 but did not detect any amplicons from other iso- necrosis in the three cultivars (Fig. 3A). These isolates also lates by PCR (data not shown). caused the putrefaction of tuber slices similar to the positive All of the group I isolates except StFLB051 caused leaf control P. carotovorum subsp. carotovorum (Fig. 3B, Table 3). AHL-Producing Fluorescent Pseudomonads 311

Table 3. Quorum sensing and other characteristics of AHL-produc- Group I isolates exhibited apparent relationships between ing isolates from potato leaves and roots their effects on potato plants and their phylogenetic group. In Influence of group IIA, there were no isolates which caused leaf necrosis c) isolates on Isolates b) Motility d) or tuber putrefaction. In group IIB, IIIB, and IIIC, none of a) Biofilm PCA e) (Group ) potato plants the isolates except StFRB074 caused tuber putrefaction, but Swimming Swarming 1234 a few caused leaf necrosis (Table 3). StFLB010 (I) 0.06 47.5 14.2 − ++++ StFLB015 (I) 0.22 47.5 47.5 − ++++ StFLB020 (I) 0.10 47.5 47.5 − ++++ Discussion StFLB031 (I) 0.47 47.5 47.5 − ++++ StFLB034 (I) 0.61 18.2 47.5 − ++++ C. violaceum CV026 is a well-known AHL reporter strain. StFLB041 (I) 0.03 17.8 47.5 − ++++ However, CV026 could not recognize long-acyl-chain − ++++ StFLB042 (I) 0.02 47.5 47.5 homoserine lactones (HSLs) such as C10–C14HSLs, as StFLB044 (I) 0.01 23.0 13.2 − ++++ StFLB045 (I) 0.01 15.2 9.5 − ++++ shown by the results in Table 1. Many researchers have used StFLB046 (I) 0.00 23.3 47.5 − ++++ the CV026 reporter strain to study the distribution of AHL- StFLB048 (I) 0.19 19.7 10.8 − ++++ producing microorganisms in various environments such as StFLB051 (I) 0.12 26.8 12.8 − −−−− StFLB136 (I) 0.07 13.3 47.5 − ++++ soil or surrounding plants (12, 45); however, selection using StFLB155 (I) 0.02 13.0 47.5 − ++++ only CV026 might fail to detect long-acyl-chain HSL-pro- StFRB280 (I) 0.01 47.5 47.5 − ++++ ducing microorganisms. Most reporter strains are restricted StFLB078 (IIA) 0.03 18.7 47.5 − −±−− in terms of the range of AHLs to which they can respond StFLB079 (IIA) 0.02 19.8 47.5 − −±−− StFLB092 (IIA) 0.28 31.5 17.0 − −−−− (38). Agrobacterium tumefaciens reporter strain based on the StFLB105 (IIA) 0.04 19.3 47.5 − −−−− TraI/R system can be used to detect a broad range of AHLs StFRB209 (IIA) 0.30 34.0 1.4 − −−−− and showed marked sensitivity to AHLs, including the long- − −−−− StFLB049 (IIB) 0.01 25.3 33.7 acyl-chain HSLs (33). However, the handling of reporter StFRB032 (IIB) 0.02 6.8 5.8 − −−±+ StFRB147 (IIB) 0.00 8.7 5.8 − −−−− strains such as Agrobacterium is complicated especially in StFRB208 (IIB) 0.03 10.2 5.7 − −−−− comparison with the simpler cross-streak assay on agar StFRB218 (IIB) 0.04 5.8 4.3 − −−±− plates used for reporter strains such as Chromobacterium StFRB219 (IIB) 0.02 9.3 6.7 − −−−− StBRB008 (IIB) 0.04 10.2 8.8 + −−−− (38). In the present study, we developed a novel AHL StBRB009 (IIB) 0.04 15.2 5.0 + −−−− reporter strain, C. violaceum VIR24, that was derived from StRB046 (IIB) 0.02 11.5 7.3 − +−−− the C. violaceum type strain ATCC 12472. This reporter − −−−± StFRB234 (IIIA) 0.01 15.3 6.8 could detect long-acyl-chain HSL molecules that could not StFRB067 (IIIA) 0.36 47.5 25.7 − −+++ StFRB074 (IIIB) 0.37 47.5 24.2 − ++++ be detected by CV026. Therefore, the results from the simple StFRB159 (IIIB) 0.49 47.5 26.2 − −−−− cross-streak assay using both CV026 and VIR24 suggested StFLB016 (IIIC) 1.26 47.5 28.3 − −±±+ that the isolates produce both short- and long-acyl-chain StFRB097 (IIIC) 0.17 47.5 39.2 − −±±± StFRB116 (IIIC) 0.15 47.0 47.5 − −−−− HSL molecules. StFRB117 (IIIC) 0.13 47.5 1.7 − −−−− The cross-streak assay showed that 10.4% of isolates StFRB118 (IIIC) 0.12 47.5 1.5 − −−−− induced the production of the purple pigment violacein in the StFRB125 (IIIC) 0.06 47.5 26.7 − −−−− reporter strains CV026 and VIR24. It has been reported that StFRB140 (IIIC) 0.05 47.5 24.8 − −−−− StFRB142 (IIIC) 0.04 18.8 11.2 − −−−− the AHL-producing bacteria represent 10–30% of the cultur- StFRB163 (IIIC) 0.04 47.5 47.5 − −−−− able bacteria in soil and rhizospheric environments (9, 10, StFRB164 (IIIC) 0.03 47.5 47.5 − −−−− 43, 45). In this study, AHL production was observed with − −−−− StFRB165 (IIIC) 0.04 47.5 47.5 12.6% of the leaf isolates and 9.3% of the root isolates. StFRB166 (IIIC) 0.04 47.5 47.5 − −−−− StFRB167 (IIIC) 0.04 47.5 47.5 − −−−− These proportions of AHL producers in the fluorescent StFRB285 (IIIC) 0.04 31.7 17.3 − −+++ pseudomonads surrounding plants corresponded exactly with StFRB286 (IIIC) 0.04 34.7 20.0 − −+++ the numbers reported in previous studies. a) Group was determined by the phylogenetic analysis using the 16S TLC using the C. violaceum indicator strains CV026 and rRNA gene sequence in Fig. 2. b) VIR24 revealed that bacterial extracts contained one or more Biofilm formation (the OD595 values) were determined as described AHL molecular species. The combined use of these two previously (46). The values represent the means of three replicates (The standard errors were not significant in all isolates). reporters with different sensitivities to AHLs would be useful c) Radial extension (mm) of colonies on the MAKCS medium contain- for quorum sensing. However, some of the molecules ing 0.28% (w/v) agar (swimming) or 0.4% (w/v) agar (swarming) at detected did not correspond to the 10 synthetic AHLs used, 24 h after incubation. The values represent the means of three repli- and were regarded as U.M. in this study (Table 2). It has cates (The standard errors were not significant in all isolates). been reported that molecules other than AHLs occasionally d) Phenazine-1-carboxylic acid (PCA) production, PCA was detected as described previously (27). exhibit cross-talk with quorum-sensing bacterial sensors (8). e) Influence of isolates on potato plants. 1, Putrefaction of potato tuber; Some of the isolates used in this study might have produced 2, Necrosis of potato leaf (cv. Irish Cobbler); 3, Necrosis of potato such active molecules. The genus Pseudomonas is known leaf (cv. May Queen); 4, Necrosis of potato leaf (cv. Toyoshiro). to produce multiple AHL signal molecules: C4HSL and 3- Potato tuber putrefaction was measured by inoculating 10 μL of cell 2–6 oxo-C12-HSL in Pseudomonas aeruginosa, 3-oxo-C6HSL suspension at 10 CFU on a potato tuber slice (cv. Irish Cobbler). Necrosis of leaf was evaluated on the leaves of three cultivars at 24 h in P. syringae, C6HSL and an unknown molecule in P. after bacterial inoculation chlororaphis subsp. aureofaciens, C6HSL in P. chlororaphis, 3-oxo-C12HSL in P. putida, and 3-OH-C14:1-HSL and un- 312 SOMEYA et al.

Fig. 3. Influence of AHL-producing isolates on potato plants. (A) Necrosis in the potato leaf on inoculation of AHL-producing isolates from potato leaves and roots. Ten microliters of a bacterial cell suspension containing 102, 104, and 106 CFU was inoculated on the leaves. The photo- graph was taken 24 h after the inoculation. (B) Putrefaction of the potato tuber (cv. Irish Cobbler) after inoculation with AHL-producing isolates. PC represents Pectobacterium carotovorum subsp. carotovorum, which was the positive control. defined long-acyl-chain HSLs in Pseudomonas fluorescens isolates belonged to IIB, IIIA, IIIB, and IIIC related to plant (42). In the present study, we observed that the fluorescent root- or soil-related pseudomonads such as P. chlororaphis, pseudomonad isolates from the potato plants produced vari- P. frederiksbergensis, P. umsongensis, and P. putida. There- ous AHL molecules in addition to those above. It was con- fore, it is considered that leaf-colonizing and root-colonizing sidered that AHL-producing isolates utilize these AHLs for AHL-producing fluorescent pseudomonads obviously differ- the regulation of own functions. However, AHL molecules entiate their niches around potato plants. The phyllosphere are diffusible, and not only utilized by the bacteria that pro- and rhizosphere, even in the same plant, differ greatly due to duce them but also recognizable by coexisting organisms differences in environmental conditions, which lead to (26, 32, 39, 44). Under the conditions that exist in the changes in both biotic and abiotic factors (2). Environmental phylloplane and rhizosphere, microorganisms may be able to factors are believed to be responsible for the difference in the use signal molecules produced by other organisms. The sys- distribution of specific species in the phyllosphere and tem of quorum-sensing through AHLs is known to be utilized rhizosphere. The results of the phylogenetic analysis reflect- by Gram-negative bacteria pathogenic to plants (15, 19). It ed that separate species of AHL-producing fluorescent is worth noting that AHL-producing nonpathogenic bacteria pseudomonads colonize the phyllosphere and rhizosphere of and pathogenic bacteria interact with each other via AHLs potato plants. around plants. Moreover, it has been reported that certain In the present study, we observed that phylogenetic groups environmental microorganisms have the potential to degrade have relationships with the AHLs produced and some char- AHL molecules (11, 18, 21). These AHL-degraders can con- acteristics. For example, most of the group I isolates mainly trol plant pathogens and also interfere with the functioning of produced C8HSL, and 3OC6/C8HSL (Table 2). A root iso- beneficial colonizers. In nature, the presence and consump- late, StFRB280, also produced C8HSL and 3C6/C8HSL, and tion of AHLs may occur at all times, and the circulation of C12/C14HSL. Therefore, it makes sense that StFRB280 is signal molecules in the environment may influence bacterial belonging to group I. We observed that all of these isolates traits. caused leaf necrosis and putrefaction of tuber slices. Group I Based on the results of the phylogenetic analysis, most isolates showed a close phylogenetic relationship with P. leaf isolates belonged to group I and IIA that were related to cichorii, known as a phytopathogen. The presence of leaf colonizing pseudomonads, including phytopathogens Pseudomonas species that are pathogenic to potato plants such as P. cichorii, P. syringae, P. ficuserectae, Pseudo- has not been reported in Japan. However, P. cichorii and monas amygdali, Pseudomonas avellanae, and P. viridiflava. Pseudomonas marginalis are reported as opportunistic These relative species belonged to the “P. syringae group” pathogens that can cause potato tuber rot, leaf spots and tuber based on the 16S rRNA sequence (4). In contrast, most root eye necrosis in the UK (37). Although all of the isolates were AHL-Producing Fluorescent Pseudomonads 313 obtained from healthy potato plants, group I isolates in the of microorganisms on plant surfaces. Annu. Rev. Phytopathol. present study might have the potential to infect any plant 38:145–180. 3. Antoun, H., and D. Prévost. 2005. Ecology of plant growth promoting species including potato plants as minor pathogens. This is rhizobacteria, p. 1–38. In Z.A. Siddiqui (ed.), PGPR: Biocontrol and not surprising because Morris et al. (22) showed that plant Biofertilization, Springer, Dordrecht, The Netherlands. pathogenic P. syringae survived in a wide range of habitats 4. Anzai, Y., H. Kim, J.Y. Park, H. Wakabayashi, and H. Oyaizu. 2000. where there were no host plants. Furthermore, it is well Phylogenetic affiliation of the pseudomonads based on 16S rRNA sequence. Int. J. Syst. Evol. Microbiol. 50:1563–1589. known that P. syringae utilize the quorum-sensing system 5. Behrendt, U., A. Ulrich, P. Schumann, W. Erler, J. Burghardt, and W. for exopolysaccharide production, motility, and virulence Seyfarth. 1999. A taxonomic study of bacteria isolated from grasses: a and that this system functions via AHL molecules (28). The proposed new species Pseudomonas graminis sp. nov. Int. J. Syst. group I isolates certainly showed greater motility than the Bacteriol. 49:297–308. 6. Catara, V., L. Sutra, A. Morineau, W. Achouak, R. Christen, and L. isolates that formed other clusters. Relationships between the Gardan. 2002. Phenotypic and genomic evidence for the revision of phylogenetic group, AHL variety and characteristics were Pseudomonas corrugata and proposal of Pseudomonas mediterranea observed in group II. Especially, short-acyl-chain HSLs such sp. nov. Int. J. Syst. Evol. Microbiol. 52:1749–1758. as C4HSL and C6HSL, were observed in specific clusters 7. Chin-A-Woeng, T.F.C., G.V. Bloemberg, and B.J.J. Lugtenberg. 2003. Phenazines and their role in biocontrol by Pseudomonas bacte- in group II. In group II, two isolates (StBRB008 and ria. New Phytol. 157:503–523. StBRB009) produced PCA. It is well known that many P. 8. Degrassi, G., C. Aguilar, M. Bosco, S. Zahariev, S. Pongor, and chlororaphis isolates produce phenazine antibiotics such as V. Venturi. 2002. Plant growth-promoting Pseudomonas putida PCA, and PCA-producing fluorescent pseudomonads can WCS358 produces and secrets four cyclic dipeptides: cross-talk with control various plant pathogens (7). In P. chlororaphis, the quorum sensing bacterial sensors. Curr. Microbiol. 45:250–254. 9. d’Angelo-Picard, C., D. Faure, I. Penot, and Y. Dessaux. 2005. Diver- biosynthesis of phenazine antibiotics is regulated by the sity of N-acyl homoserine lactone-producing and -degrading bacteria quorum-sensing system via C6HSL (42). StBRB008 and in soil and tobacco rhizosphere. Environ. Microbiol. 7:1796–1808. StBRB009 also produced C6HSL, therefore, PCA produc- 10. Elasri, M., S. Delorme, P. Lemanceau, G. Stewart, B. Laue, E. tion in the two isolates may be regulated by C6HSL. Glickmann, P.M. Oger, and Y. Dessaux. 2001. Acyl-homoserine lactone production is more common among plant-associated The quorum-sensing system that functions via AHLs regu- Pseudomonas spp. than among soilborne Pseudomonas spp. Appl. lates many of the traits of fluorescent pseudomonads that are Environ. Microbiol. 67:1198–1209. beneficial to plants (42). Many researchers have attempted to 11. Jafra, S., J. Przysowa, R. Czajkowski, A. Michta, P. Garbeva, and use the functions of this system to control plant diseases or J.M. Van der Wolf. 2006. Detection and characterization of bacteria from the potato rhizosphere degrading N-acyl-homoserine lactone. promote plant growth. However, many such attempts have Can. J. Microbiol. 52:1006–1015. failed to due to the difficulties encountered in regulating 12. Khmel, I.A., M.A. Veselova, A.Z. Metlitskaya, S. Klein, V.A. microbial functions under agricultural conditions. In the Lipasova, A.V. Mayatskaya, and L.S. Chernin. 2002. Synthesis of present study, we demonstrated that certain fluorescent signaling N-acyl-homoserine-lactones participating in quorum sens- ing regulation in rhizospheric and soil-borne bacteria Pseudomonas pseudomonads present in the phyllosphere and rhizosphere and Xanthomonas. Russ. J. Genet. 38:467–469. of potato plants possess quorum-sensing systems that func- 13. Kloepper, J.W., J. Leong, M. Teintze, and M.N. Schroth. 1980. tion via AHLs. These isolates showed multiple functions Enhanced plant growth by siderophores produced by plant growth- such as antibiotic production, biofilm formation, and motil- promoting rhizobacteria. Nature 286:885–886. 14. Kwon, S.W., J.S. Kim, I.C. Park, S.H. Yoon, D.H. Park, C.K. Lim, ity. We will attempt to apply these beneficial features to agri- and S.J. Go. 2003. Pseudomonas koreensis sp. nov., Pseudomonas cultural use in future studies. umsongensis sp. nov. and Pseudomonas jinjuensis sp. nov., novel species from farm in Korea. Int. J. Syst. Evol. Microbiol. 53:21– 27. Acknowledgements 15. Licciardello, G., I. Bertani, L. Steindler, P. Bella, V. Venturi, and V. We would like to thank the following people for their coopera- Catara. 2007. Pseudomonas corrugata contains a conserved N-acyl tion in collecting plant samples: Y. Ogawa, Pest Control Office in homoserine lactone quorum sensing system; its role in tomato patho- Nagasaki Prefecture; T. Ogawa, Nagasaki Agricultural and Forestry genicity and tobacco hypersensitivity response. FEMS Microbiol. Ecol. 61:222–234. Technical Development Center, Japan; Y. Kodama, Kagoshima 16. McClean, K.H., M.K. Winson, L. Fish, et al. 1997. Quorum sensing Prefectural Institute for Agricultural Development, Japan; S. and Chromobacterium violaceum: exploitation of violacein produc- Numata, Plant Biotechnology Institute-Ibaraki Agricultural Center, tion and inhibition for the detection of N-acylhomoserine lactones. Japan; and H. Shinohara, Tokyo University of Agriculture, Japan. Microbiology 143:3703–3711. We thank P. Williams, University of Nottingham, UK for providing 17. Morohoshi, T., T. Inaba, N. Kato, K. Kanai, and T. Ikeda. 2004. C. violaceum strain CV026. We also thank the following people Identification of quorum-sensing signal molecules and the LuxRI for technical assistance: M. Mori, S. Tsuda, N. Takakuwa, Y. Hasa, homologs in fish pathogen Edwardsiella tarda. J. Biosci. Bioeng. S. Takahashi, H. Ogawa, N. Taira, T. Yamada, A. Morizumi, Y. 98:274–281. Shimada, M. Kikuchi and H. Tokuji at the National Agricultural 18. Morohoshi, T., A. Ebata, S. Nakazawa, N. Kato, and T. Ikeda. 2005. Research Center for Hokkaido Region, Japan. We are grateful to T. N-acyl homoserine lactone-producing or -degrading bacteria isolated Nishizawa, College of Agriculture, Ibaraki University, Japan and H. from the intestinal microbial flora of ayu fish (Plecoglossus altivelis). Sawada, National Institute of Agrobiological Sciences, Japan for Microbes Environ. 20:264–268. 19. Morohoshi, T., Y. Nakamura, G. Yamazaki, A. Ishida, N. Kato, and helpful discussions. T. Ikeda. 2007. The plant pathogen Pantoea ananatis produces References N-acylhomoserine lactone and causes center rot disease of onion by quorum sensing. J. Bacteriol. 189:8333–8338. 1. Andersen, S.M., K. Johnsen, J. Sørensen, P. Nielsen, and C.S. 20. Morohoshi, T., M. Kato, K. Fukamachi, N. Kato, and T. Ikeda. Jacobsen. 2000. Pseudomonas frederiksbergensis sp. nov., isolated 2008. N-acylhomoserine lactone regulates violacein production in from soil at a coal gasification site. Int. J. Syst. Evol. Microbiol. Chromobacterium violaceum type strain ATCC 12472. FEMS 50:1957–1964. Microbiol. Lett. 279:124–130. 2. Andrews, J.H., and R.F. Harris. 2000. The ecology and biogeography 21. Morohoshi, T., N. Someya, and T. Ikeda. 2009. Novel N-acyl- 314 SOMEYA et al.

homoserine lactone-degrading bacteria isolated from the leaf surface 34. Someya, N., K. Tsuchiya, T. Yoshida, M.T. Noguchi, and H. Sawada. of Solanum tuberosum and their quorum-quenching properties. 2007. Combined application of Pseudomonas fluorescens strain Biosci. Biotechnol. Biochem. 73:2124–2127. LRB3W1 with a low dosage of benomyl for control of cabbage yel- 22. Morris, C.E., L.L. Kinkel, K. Xiao, P. Prior, and D.C. Sands. 2007. lows caused by f. sp. conglutinans. Biocontrol Surprising niche for the plant pathogen Pseudomonas syringae. Sci. Technol. 17:21–31. Infect. Genet. Evol. 7:84–92. 35. Someya, N., K. Tsuchiya, S. Sugisawa, M.T. Noguchi, and T. 23. Nonaka, K., H. Ohta, Y. Sato, and K. Hosokawa. 2008. Utilization Yoshida. 2008. Growth promotion of lettuce (Lactuca sativa L.) by a of phenylpropanoids by Pseudomonas putida soil isolates and its rhizobacterium Pseudomonas fluorescens strain LRB3W1 under iron- probable taxonomic significance. Microbes Environ. 23:360–354. limiting condition. Environ. Control Biol. 46:139–146. 24. Palleroni, N.J. 2005. Genus I. Pseudomonas Migula 1894, 237AL, 36. Someya, N., and K. Akutsu. 2009. Indigenous bacteria may interfere p. 323–379. In D.J. Brenner, N.R. Krieg, J.T. Staley, and G.M. with the biocontrol of plant diseases. Naturwissenschaften 96:743– Garrity (ed.), Bergey’s Manual of Systematic Bacteriology, 2nd ed., 747. Part B The . Springer, New York, USA. 37. Stead, D. 1999. Bacterial diseases of potato: relevance to in vitro 25. Peix, A., A. Valverde, R. Rivas, et al. 2007. Reclassification of potato seed production. Potato Res. 42:449–456. Pseudomonas aurantiaca as a synonym of Pseudomonas chloro- 38. Steindler, L., and V. Venturi. 2007. Detection of quorum-sensing N- raphis and proposal of three subspecies, P. chlororaphis subsp. acyl homoserine lactone signal molecules by bacterial biosensors. chlororaphis subsp. nov., P. chlororaphis subsp. aureofaciens subsp. FEMS Microbiol. Lett. 266:1–9. nov., comb. nov. and P. chlororaphis subsp. aurantiaca subsp. nov., 39. Subramoni, S., and V. Venturi. 2009. LuxR-family ‘solos’: bachelor comb. nov. Int. J. Syst. Evol. Microbiol. 57:1286–1290. sensors/regulators of signalling molecules. Microbiology 155:1377– 26. Pierson, E.A., D.W. Wood, J.A. Cannon, F.M. Blachere, and L.S. 1385. Pierson III. 1998. Interpopulation signaling via N-acyl-homoserine 40. Thompson, J.D., T.J. Gibson, F. Plewniak, F. Jeanmougin, and D.G. lactones among bacteria in the wheat rhizosphere. Mol. Plant- Higgins. 1997. The CLUSTAL_X windows interface: flexible Microbe Interact. 11:1078–1084. strategies for multiple sequence alignment aided by quality analysis 27. Pierson III, L.S., and L.S. Thomashow. 1992. Cloning and hetero- tools. Nucleic Acids Res. 25:4876–4882. logous expression of the phenazine biosynthetic locus from Pseudo- 41. Tomihama, T., Y. Nishi, and K. Arai. 2006. The effects of monas aureofaciens 30–84. Mol. Plant-Microbe Interact. 5:330–339. exopolysaccharide production and flagellum on motility, biofilm 28. Quiñones, B., G. Dulla, and S.E. Lindow. 2005. Quorum sensing formation, survival and virulence of Pseudomonas syringae pv. theae regulates exopolysaccharide production, motility, and virulence in (In Japanese). Jpn. J. Phytopathol. 72:3–13. Pseudomonas syringae. Mol. Plant-Microbe Interact. 18:682–693. 42. Venturi, V. 2006. Regulation of quorum sensing in Pseudomonas. 29. Raaijmakers, J.M., D.M. Weller, and L.S. Thomashow. 1997. FEMS Microbiol. Rev. 30:274–291. Frequency of antibiotic-producing Pseudomonas spp. in natural envi- 43. Veselova, M., M. Kholmeckaya, S. Klein, et al. 2003. Production of ronments. Appl. Environ. Microbiol. 63:881–887. N-acylhomoserine lactone signal molecules by Gram-negative soil- 30. Reece, K.S., and G.J. Phillips. 1995. New plasmids carrying anti- borne and plant-associated bacteria. Folia Microbiol. 48:794–798. biotic-resistance cassettes. Gene 165:141–142. 44. von Rad, U., I. Klein, P.I. Dobrev, J. Kottova, E. Zazimalova, A. 31. Saitou, N., and M. Nei. 1987. The neighbor-joining method: a new Fekete, A. Hartmann, P. Schmitt-Kopplin, and J. Durner. 2008. method for reconstructing phylogenetic trees. Mol. Biol. Evol. 4:406– Response of Arabidopsis thaliana to N-hexanoyl-DL-homoserine- 425. lactone, a bacterial quorum sensing molecule produced in the 32. Schuhegger, R., A. Ihring, S. Gantner, et al. 2006. Induction of rhizosphere. Planta 229:73–85. systemic resistance in tomato by N-acyl-L-homoserine lactone- 45. Yoshida, S., L.L. Kinkel, H. Shinohara, N. Numajiri, S. Hiradate, M. producing rhizosphere bacteria. Plant Cell Environ. 29:909–918. Koitabashi, K. Suyama, H. Negishi, and S. Tsushima. 2006. Produc- 33. Shaw, P.D., G. Ping, S.L. Daly, C. Cha, J.E. Cronan Jr, K.L. Rinehart, tion of quorum-sensing-related signal molecules by epiphytic bacteria and S.K. Farrand. 1997. Detecting and characterizing N-acyl- inhabiting wheat heads. Can. J. Microbiol. 52:411–418. homoserine lactone signal molecules by thin-layer chromatography. 46. Yoshida, S., N. Ogawa, T. Fujii, and S. Tsushima. 2009. Enhanced Proc. Natl. Acad. Sci. USA 94:6036–6041. biofilm formation and 3-chlorobenzoate degrading activity by the bacterial consortium of Burkholderia sp. NK8 and Pseudomonas aeruginosa PAO1. J. Appl. Microbiol. 106:790–800.