SEARCH

SEARCH BY CITATION

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

Burkholderia glumae BGR1 produces a broad-host range phytotoxin, called toxoflavin, which is a key pathogenicity factor in rice grain rot and wilt in many field crops. Our molecular and genetic analyses of toxoflavin-deficient mutants demonstrated that gene clusters for toxoflavin production consist of four transcriptional units. The toxoflavin biosynthesis genes were composed of five genes, toxA to toxE, as Suzuki et al. (2004) reported previously. Genes toxF to toxI, which are responsible for toxoflavin transport, were polycistronic and similar to the genes for resistance-nodulation-division (RND) efflux systems. Using Tn3-gusA reporter fusions, we found that ToxR, a LysR-type regulator, regulates both the toxABCDE and toxFGHI operons in the presence of toxoflavin as a coinducer. In addition, the expression of both operons required a transcriptional activator, ToxJ, whose expression is regulated by quorum sensing. TofI, a LuxI homologue, was responsible for the biosynthesis of both N-hexanoyl homoserine lactone and N-octanoyl homoserine lactone (C8-HSL). C8-HSL and its cognate receptor TofR, a LuxR homologue, activated toxJ expression. This is the first report that quorum sensing is involved in pathogenicity by the regulation of phytotoxin biosynthesis and its transport in plant pathogenic bacteria.


Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

Burkholderia glumae causes rice grain rot and seedling rot in rice nursery boxes (Cottyn et al., 1996; Suzuki et al.,  1998).  Rice  grain  rot  occurs  at  the flowering stage when temperature and moisture are high and causes yield losses of up to 34% (Yoshimura et al., 1987; Tsushima et al., 1996). The bacterium produces bright yellow pigments identified as toxoflavin {1,6-dimethylpyrimido[5,4-e]-1, 2 , 4-triazine-5, 7(1H,6H)-dione: molecular weight = 193} and fervenulin, which are essential in the pathogenicity of rice seedling rot and grain rot (Sato et al., 1989; Iiyama et al., 1995; Nagamatsu, 2001). We previously reported that B. glumae BGR1 causes bacterial wilt in many field crops and produces the toxoflavin responsible for major symptom development (Jeong et al., 2003). Toxoflavin production is dependent on growth temperature, which is maximal at 37°C, and no detectable amount is produced at 25–28°C (Matsuda and Sato, 1988).

Plant pathogenic bacteria produce phytotoxins that are toxic to plant cells, which influences symptom development (Durbin, 1991). Pseudomonas species produce various phytotoxic compounds (Gross, 1991; Walton and Panaccione, 1993). Of these, the mode of action, biosynthesis and regulation of coronatine, syringomycin, tabtoxin and phaseolotoxin are relatively well characterized (Gross, 1991). In many cases, phytotoxins from Pseudomonas syringae are not required for pathogenicity, but function as virulence factors affecting disease severity (Gross, 1991; Walton and Panaccione, 1993). However, toxoflavin is a critical virulence factor in B. glumae, probably owing to its mode of action in plants. Toxoflavin is an active electron carrier between NADH and oxygen, and it can produce hydrogen peroxide and bypass the cytochrome system (Latuasan and Berends, 1961). This may explain why toxoflavin has antibacterial, antifungal and herbicidal activities and is toxic to mice, causing haematuria, diarrhoea and lachrymation (Nagamatsu, 2001).

The tox operon responsible for toxoflavin biosynthesis is polycistronic and consists of five genes (toxA to E) (Shingu and Yoneyama, 2004; Suzuki et al., 2004). Shingu and Yoneyama (2004) suggested that the LysR-type regulator ToxR regulates the expression of the toxABCDE operon, and they speculated that toxoflavin is synthesized in part via a biosynthetic pathway common to the synthesis of riboflavin, starting with GTP as the precursor (Suzuki et al., 2004). However, it is not clear whether ToxR regulates the toxoflavin biosynthesis genes. If ToxR is a LysR-type regulator, ToxR likely requires a coinducer. Currently, it is not known how B. glumae cells transport toxoflavin or protect themselves against it.

Quorum sensing is a mechanism by which bacteria regulate a set of genes in a cell-density dependent manner (Fuqua et al., 1996). The communication is mediated by diffusible signal molecules, N-acyl homoserine lactones (HSLs), that differ with respect to the length, saturation and substitutions of the side-chain (Miller and Bassler, 2001; Whitehead et al., 2001). Quorum sensing in Burkholderia cepacia is associated with siderophore ornibactin biosynthesis, extracellular protease, swarming and biofilm formation (Lewenza and Sokol, 2001; Aguilar et al., 2003).

In this study, we identified operons for toxoflavin biosynthesis and transport, and demonstrated that ToxR, a LysR-type activator, requires toxoflavin as a coinducer to activate the expression of both operons. Furthermore, we showed that the expression of the toxoflavin biosynthesis and transport genes is regulated by another transcriptional activator, ToxJ, whose expression is activated by quorum sensing. We propose a working model for the circuits by  which B. glumae cells regulate toxoflavin biosynthesis and transport. This is the first report that quorum sensing is involved in phytotoxin production and hence pathogenicity in plant pathogenic bacteria.

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

Organization of the toxoflavin biosynthesis and transport genes

To identify the DNA region responsible for toxoflavin biosynthesis, a genomic library of B. glumae BGR1 was screened by colony hybridization using probe DNA amplified from a known toxA gene. Three overlapping clones were obtained: pBGT6, pBGT7 and pBGT17. After mutagenizing the three clones with Tn3-gusA, each mutation was marker-exchanged into B. glumae BGR1. We identified 10 Tn3-gusA insertions that interfere with toxoflavin production, defining the essential region for toxoflavin production (Fig. 1).

image

Figure 1. Organization of toxABCDE, toxFGHI, and two regulatory genes, toxJ and toxR. Open arrows indicate the positions and orientations of the genes for toxoflavin transport and biosynthesis. Vertical bars in the maps indicate the positions and orientations of the Tn3-gusA insertions, and the major phenotypes of the mutants are represented below the restriction map. Vertical bars with closed circles indicate positions of Ω cassette insertions. B, BamHI; E, EcoRI; H, HindIII.

Download figure to PowerPoint

DNA sequence analyses of the 28.4 kb DNA region revealed 11 potential open reading frames: toxA, B, C, D, E, F, G, H, I, J and R (Fig. 1). Each gene is preceded by a conserved ribosome binding site (data not shown). ToxA, B, C, D, E and R were almost identical to the previously reported Tox proteins of B. glumae (Suzuki et al., 2004).

Sequence analysis of the 9.9 kb EcoRI-HindIII fragment from pBGT17 suggested that this region encodes a transporter system for toxoflavin. ToxF was 19.3 kDa in size and showed weak similarity to the multidrug efflux transporter of the major facilitator superfamily (MFS) from Prochlorococcus marinus ssp. pastoris. ToxG (possibly 40.1 kDa in size) showed 40% identity and 58% positives to the resistance-nodulation-division (RND) efflux membrane fusion protein (MFP) precursor from Pseudomonas aeruginosa PAO1. The toxH gene encoded a 110 kDa protein with 54% identity and 68% positives to the RND efflux transporter from P. aeruginosa PAO1. ToxI was predicted to be 53.9 kDa and most similar to the OprM protein of P. aeruginosa. The toxR gene was 906 bp in size, and its protein showed high similarity to various LysR-type regulators. ToxR exhibited 33% identity and 49% positives to TtgS protein from Pseudomonas putida, which is involved in the transcriptional expression of the hydrocarbon efflux pump TtgDEF, and 32% identity and 47% positives to MexT protein from P. aeruginosa PA01, which is involved in regulating the MexEF-OprN multidrug efflux system (data not shown). ToxJ shares similarities with the putative helix-turn-helix DNA binding domain near the carboxy-terminus of various transcriptional regulators (data not shown). However, ToxJ did not have any invariant amino acids typical of the LuxR family.

Tn3-gusA insertions in the toxABCD genes abolished toxoflavin, reumycin and fervenulin production (Fig. 2). The toxE mutant BGK840 produced much less of all three toxins compared with the wild type (Fig. 2). Toxoflavin production by mutants carrying Tn3-gusA insertions in the toxFGH genes was not detected, and the mutants produced much less reumycin (Fig. 2). By contrast, the mutants produced as much fervenulin as did the wild type (Fig. 2). Mutations in toxFGH genes resulted in slower growth in vitro compared with that of the wild type (data not shown). The growth of the mutants was arrested by adding 10 µM toxoflavin, whereas the growth of the wild type was not influenced by the added toxin (data not shown). Along with the similarity between the ToxFGHI and RND efflux systems, mutational analyses, the slow-growth, and growth-arrest of the mutants supported the postulate that the toxFGHI operon is not involved in the biosynthesis of the three toxins, but is important for transporting toxoflavin and reumycin. The toxI::Tn3-gusA70 mutation did not affect the production of any toxin (Fig. 2). Tn3-gusA and Ω fragment insertions in toxR or toxJ abolished the production of all three toxins (Figs 1 and 2).

image

Figure 2. Toxoflavin production by tox mutants. The toxoflavin produced is shown as seen by the naked eyes and under UV light at 365 nm. The identity of the phytotoxins shown was confirmed by LC-MS of the purified toxins and comparison with a synthetic standard (S: F, fervenulin; R, reumycin; T, toxoflavin). Mutations in toxABCD, toxJ, and toxR resulted in no phytotoxin production. A toxE mutant produced reduced amounts of toxins. Mutations in toxFGH conferred as much fervenulin production as in the wild-type BGR1. A mutation in toxI did not affect phytotoxin production. The origin (o) and solvent front (f) on TLC are marked.

Download figure to PowerPoint

We used reverse transcription polymerase chain reaction (RT-PCR) to determine if the toxFGHI genes are polycistronic. As shown in Fig. 3, we used six sets of primers to amplify toxF-G, G, G-H, H, H-I and I. RT-PCR followed by Southern hybridization indicated that toxFGHI genes are transcribed as a single transcript (Fig. 3). Interestingly, an 805 bp DNA region between toxJ and toxI and another 553 bp downstream from toxE showed 47% and 65% identity to the putative transposase from B. cepacia, respectively (Fig. 1). The DNA sequences of the toxA, B, C, D, E, F, G, H, I, J and R regions of B. glumae BGR1 have been deposited in GenBank under Accession No. AY641455.

image

Figure 3. Transcriptional unit in the toxFGHI operon as determined by RT-PCR. RT-PCR products were confirmed by Southern hybridization analysis. A. Black arrows indicate the extension and transcription directions of the toxABCDE and toxFGHI operons. An arrow below the open arrows represents the product of the RT reaction. The short thick bars below the RT arrow indicate the 6 PCR products from the corresponding RT reactions. The expected sizes of the PCR products are indicated below the label. B. Agarose gel analysis (upper part) and Southern analysis (lower part) of the RT-PCR products of the toxFGHI operon. Southern hybridization was performed using the toxFGHI operon region (10.5 kb EcoRI fragment of pBGT17) as probes. Lanes 1–3, 4–6, 7–9, 10–12, 13–15 and 16–18 correspond to the products of PCR1, PCR2, PCR3, PCR4, PCR5 and PCR6, respectively. Lanes 1, 4, 7, 10, 13 and 16: PCR products from the DNA template as positive controls; lanes 2, 5, 8, 11, 14 and 17: PCR products from the RNA template as negative controls; lanes 3, 6, 9, 12, 15 and 18: RT-PCR products.

Download figure to PowerPoint

Virulence of toxoflavin-deficient mutants

The ability of mutants to induce bacterial grain rot in rice was determined by inoculating rice panicles with the wild type and toxin-deficient mutants during the flowering stage. When rice panicles were inoculated with the wild type and toxA, E, F, I, R and J mutants, the toxin-deficient mutants did not show any visible symptoms (disease index 0.1), whereas the wild type caused severe symptoms (disease index 4.2) (Fig. 4). The toxE mutant BGK840, which produced small amounts of toxins, exhibited some symptoms (disease index 1.5) (Fig. 4). The toxF mutant BGK34 producing fervenulin and small amounts of reumycin induced slight bacterial grain rot (disease index 1.5) (Fig. 4). This indicated that bacterial grain rot in rice caused by B. glumae is highly dependent on the ability of the bacterium to produce toxin.

image

Figure 4. The pathogenicity of phytotoxin-deficient mutants. Panicles of ‘Milyang 23′ were inoculated with B. glumae strains, as described in Experimental procedures. The wild-type strain BGR1 produced severe symptoms, resulting in empty heads of grain. The panicles inoculated with toxA, toxJ and toxR mutants were nearly asymptomatic, while those inoculated with toxE or toxF mutants showed significantly reduced symptoms. The panicles inoculated with a toxI mutant developed severe symptoms of grain rot. The photograph was taken 7 days after inoculation. The disease index of the tested rice plants was described in Experimental procedures.

Download figure to PowerPoint

Quorum signal-deficient mutants failed to produce phytotoxins in B. glumae BGR1

We reasoned that toxoflavin production at the late exponential growth phase might be dependent on the cell density of the culture. This led us to test whether the wild-type strain BGR1 produces autoinducers (AIs) using the biosensor strain Chromobacterium violaceum CV026. The bioassay indicated that the strain produces AIs corresponding to the synthetic N-hexanoyl homoserine lactone (C6-HSL) and N-octanoyl HSL (C8-HSL), as determined on thin layer chromatography overlaid with the biosensor cells (Fig. 5A). To confirm their chemical structures, the purified compounds were subjected to ESI-MS analysis. The two compounds displayed molecular ion (M + H) peaks at m/z 200 and m/z 228, respectively, and a fragmentation ion at m/z 102, which is characteristic of the homoserine lactone ring. Moreover, their ESI-MS spectra were identical to that of the synthetic C6-HSL and C8-HSL standards, which showed that the two compounds are C6-HSL and C8-HSL, respectively (data not shown).

image

Figure 5. A quorum sensing signal is required for phytotoxin production in B. glumae BGR1. A bioassay of HSLs using a Chromobacterium violaceum biosensor on TLC (A). The BGS2 mutant did not produce any HSL signal molecule as determined on TLC assays, but this was complemented by introducing the tofI clone, pBGA43. The identity of each HSL shown in (A) was confirmed by comparison with a synthetic standard (S1: C4, C4-HSL; C6, C6-HSL; C8, C8-HSL). Phytotoxin production was restored by introducing pBGA43 into BGS2 and by adding C8-HSL exogenously (B). The identity of each phytotoxin shown in (B) was confirmed by comparison with a synthetic standard (S2: F, fervenulin; R, reumycin; T, toxoflavin). The origin (o) and solvent front (f) on TLC are marked. Phytotoxin production by the BGS2 mutant in response to C4-, C6-, C8-, C10- and C12-HSL (C).

Download figure to PowerPoint

To clone the C6- and C8-HSL synthase gene, a cosmid genomic library of BGR1 was mobilized into P. fluorescens 1855.344, which does not produce any N-acyl HSL. After screening transconjugants on plates containing Agrobacterium indicator cells, one cosmid clone, pBGA18, which conferred AI production to P. fluorescens 1855.344, was isolated (Fig. 6). The plasmid contained a 25 kb insert, and the smallest subclone conferring C6- and C8-HSL production was isolated as a 3.5 kb EcoRI-SacI fragment (Fig. 6). The DNA sequence analysis identified the tofI and tofR genes responsible for C6- and C8-HSL synthesis, respectively (Fig. 6). The tofI and tofR genes were separated by 796 bp and transcribed separately (Fig. 6). The tofI gene was 612 bp in size and encoded a 22.4 kDa protein homologous to members of the LuxI family (data not shown). The tofR gene was 720 bp in size and encoded a 26.6 kDa protein homologous to members of the LuxR family (data not shown). The DNA sequences of the cloned tofRI region of B. glumae BGR1 have been deposited in GenBank under Accession No. AY641454.

image

Figure 6. Organization of the tofR and tofI genes. Open arrows indicate the positions and orientations of the genes for C6- and C8-HSL production. Vertical bars in the maps indicate the positions and orientations of the Tn3-gusA insertions, and the major phenotypes of the mutants are represented below the restriction map. Vertical bars with closed circles indicate the positions of Ω cassette insertions. B, BamHI; E, EcoRI; H, HindIII; K, KpnI; S, SacI; Sf, SfiI; V, EcoRV. Enzyme sites from the vector are shown in parentheses.

Download figure to PowerPoint

To determine if AI production affects toxin production by the bacterium, the tofR::Ω mutant BGS1 and the tofI::Ω mutant BGS2 were constructed (Table 1). Mutant BGS2 produced neither AIs nor phytotoxins (Fig. 5A and B). When pBGA43 carrying the tofI gene was mobilized into BGS2, the transconjugants recovered production of both AIs and phytotoxins (Fig. 5A and B). When various AIs were added to the BGS2 culture at different concentrations, C8-HSL conferred phytotoxin production on the mutant at 100 nM, while a much higher concentration (1000 nM) of C10-HSL was required to achieve the same level of phytotoxin production (Fig. 5C). C4-, C6- and C12-HSL failed to confer phytotoxin production on the BGS2 mutant, even at 1000 nM (Fig. 5C).

Table 1.  Bacterial strains and plasmids.
Strain or plasmidCharacteristicsSource or reference
  1. AmpR, ampicillin resistance; CmR, chloramphenicol resistance; KmR, kanamycin resistance; NalR, nalidixic acid resistance; RifR, rifampicin resistance; SpR, spectinomycin resistance; TetR, tetracycline resistance.

Escherichia coli
 DH5αFΦ80 dlac ZΔM15Δ(lac ZYA-argF)U169 end A1 recA1 hsd R17 (rKmK+) deo R thi- 1 sup E44λgyr A96 rel A1Gibco BRL
 S17-1Tra+, recA, SpRSimon et al. (1983)
 C2110polA, NalRStachel et al. (1985)
 HB101FmcrB mrr hsd S20(rBmB) rec A13 leu B6 ara-14 pro A2 lac Y1 gal K2 xyl-5 mtl-1 rps L20(SmR) sup E44λGibco BRL
Agrobacterium tumefaciens
 NT1(pDCI41E33)Autoinducer indicator strainCook et al. (1997)
Chromobacterium violaceum
 CV026Autoinducer indicator strainMcClean et al. (1997)
Burkholderia glumae
 BGR1Wild type, RifRJeong et al. (2003)
 BGS1BGR1 tofR::ΩThis study
 BGS2BGR1 tofI::ΩThis study
 BGS4BGR1 toxR::ΩThis study
 BGS5BGR1 toxJ::ΩThis study
 BGK18/S1K18/S2K18/S4K18BGR1 toxJ::Tn3-gusA18/BGS1 toxJ::Tn3-gusA18/BGS2 toxJ::Tn3- gusA18/ BGS4 toxJ::Tn3-gusA18This study
 BGK43/S1K43/S2K43/S5K43BGR1 toxR::Tn3-gusA43/BGS1 toxR::Tn3-gusA43/BGS2 toxR::Tn3- gusA43/BGS5 toxR::Tn3-gusA43This study
 BGK59/S4K59/S5K59BGR1 toxA::Tn3-gusA59/BGS4 toxA::Tn3-gusA59/BGS5 toxA::Tn3-gusA59This study
 BGK536/S4K536/S5K536BGR1 toxB::Tn3-gusA536/BGS4 toxB::Tn3-gusA536/BGS5 toxB::Tn3-gusA536This study
 BGK421/S4K421/S5K421BGR1 toxC::Tn3-gusA421/BGS4 toxC::Tn3-gusA421/BGS5 toxC::Tn3-gusA421This study
 BGK465/S4K465/S5K465BGR1 toxD::Tn3-gusA465/BGS4 toxD::Tn3-gusA465/BGS5 toxD::Tn3-gusA465This study
 BGK840/S4K840/S5K840BGR1 toxE::Tn3-gusA840/BGS4 toxE::Tn3-gusA840/BGS5 toxE::Tn3-gusA840This study
 BGK34/S4K34/S5K34BGR1 toxF::Tn3-gusA34/BGS4 toxF::Tn3-gusA34/BGS5 toxF::Tn3-gusA34This study
 BGK858/S4K858/S5K858BGR1 toxG::Tn3-gusA858/BGS4 toxG::Tn3-gusA858/BGS5 toxG::Tn3-gusA858This study
 BGK20/S4K20/S5K20BGR1 toxH::Tn3-gusA20/BGS4 toxH::Tn3-gusA20/BGS5 toxH::Tn3-gusA20This study
 BGK70/S4K70/S5K70BGR1 toxI::Tn3-gusA70/BGS4 toxI::Tn3-gusA70/BGS5 toxI::Tn3-gusA70This study
Plasmids
 pBluescript II SK(+)Cloning vehicle; phagemid, pUC derivative, AmpRStratagene
 pLAFR3Tra, Mob+, RK2 replicon, TetRStaskawicz et al. (1987)
 pLAFR6As pLAFR3 but without lacZα, contains multilinker of pUC18 flanked by synthetic trp terminators, TetRHuynh et al. (1989)
 pRK2013Helper plasmid; Tra+, ColE1 replicon, KmRFigurski and Helinski (1979)
 pHoKmGusPromoterless β-glucuronidase gene, KmR, AmpR,Bonas et al. (1989)
 pSSheCmRStachel et al. (1985)
 pHP45ΩΩ cassette, SpR, SmRPrentki and Krisch (1984)
 pBGT625.2 kb DNA fragment from strain BGR1 cloned into pLAFR3This study
 pBGT728.4 kb DNA fragment from strain BGR1 cloned into pLAFR3This study
 pBGT1728.5 kb DNA fragment from strain BGR1 cloned into pLAFR3This study
 pBGT312.0 kb Ω cassette was inserted into HincII site within toxR sequence and cloned into pLAFR3This study
 pBGT352.0 kb Ω cassette was inserted into HindIII site within toxJ sequence and cloned into pLAFR3This study
 pBGT381.5 kb EcoRI-EcoRV fragment including toxJ from pBGT20 cloned into pLAFR3This study
 pBGT401.1 kb Nar I-SmaI fragment including toxR from pBGT20 cloned into pLAFR3This study
 pBGA1825.3 kb DNA fragment from strain BGR1 cloned into pLAFR3This study
 pBGA37R7.9 kb BamHI-SacI fragment including tofRI region from pBGA18 cloned into pBluescript II SK(+)This study
 pBGA38R7.9 kb BamHI-HindII fragment from pBGA37R cloned into pLAFR3This study
 pBGA430.85 kb EcoRV-SfiI fragment including tofI from pBGA18 cloned into pLAFR3This study
 pBGA47R2.0 kb Ω cassette was inserted into BglII site within tofR sequence cloned into pLAFR3This study
 pBGA482.0 kb Ω cassette was inserted into SnaBI site within tofI sequence cloned into pLAFR3This study
 pBGA541.3 kb EcoRI-KpnI fragment including tofR from pBGA18 cloned into pLAFR3This study
 pBG46R10.7 kb KpnI-XhoI fragment from pBGA18::Tn3-gusA 46 cloned into pLAFR6, tofR::Tn3-gusA46 fusion onlyThis study
 pBG294I10.9 kb KpnI-XhoI fragment from pBGA18::Tn3-gusA 294 cloned into pLAFR6, tofI::Tn3-gusA294 fusion onlyThis study

As the AI-deficient mutants failed to produce phytotoxins, rice panicles were inoculated with the mutants to determine their virulence. The disease caused by mutant BGS2 was much less severe than that produced by the wild type (data not shown). When BGS2 carried pBGA43 in trans, the disease was as severe as with the wild type (data not shown). These results indicated that quorum sensing regulates phytotoxin production and thereby the pathogenicity of B. glumae.

TofR and C8-HSL complex activates tofI expression

To determine if TofR regulates tofI expression, a tofI::Tn3-gusA294 fusion was cloned into pLAFR6 (resulting in pBG294I), and then introduced into the tofR::Ω mutant BGS1 and the tofI::Ω mutant BGS2. When the β-glucuronidase activity was measured in each strain, tofI expression was abolished in the absence of tofR or C8-HSL (Fig. 7A). C6-HSL did not induce tofI expression, even in the presence of tofR, indicating that tofI expression is positively regulated by TofR and C8-HSL complex (Fig. 7A). The presence of two lux box consensus sequences in the upstream region of tofI supported the postulate that TofR regulates tofI expression.

image

Figure 7. Effect of TofR and C8-HSL on tofI and toxJ expression. TofR and C8-HSL complex regulates tofI expression (A). □, BGS1(pBG294I, plasmid-borne tofI::Tn3-gusA294) with C6-HSL; ▪, BGS1(pBG294I) with C8-HSL; ○, BGS2(pBG294I) with C6-HSL; •, BGS2(pBG294I) with C8-HSL. TofR and C8-HSL complex regulates toxJ expression (B). The expression of toxJ was abolished by the knock-out of tofR and tofI; while in S2K18, toxJ expression was increased to the level of BGK18 by adding 100 nM C8-HSL, but not C6-HSL. The 100 nM C8-HSL was added to cultures grown in the medium. β-Glucuronidase activity was measured as described in Experimental procedures. One unit of β-glucuronidase was defined as 1 nmol of 4-methyllumbelliferone released per bacterium per minute. All values are the means ± SDs of values from triplicate experiments.

Download figure to PowerPoint

Expression of toxJ is activated by TofR and C8-HSL complex

As two regulatory proteins, ToxJ and ToxR, and quorum sensing are involved in regulating toxin production in B. glumae, we postulated that TofR and C8-HSL complex regulates the expression of one of the two regulatory genes. A lux box-like sequence is present upstream from toxJ, but not toxR, suggesting that TofR regulates toxJ expression. The toxJ::Tn3-gusA18 and toxR::Tn3-gusA43 fusions were constructed in wild-type BGR1, tofR::Ω mutant BGS1, and tofI::Ω mutant BGS2 backgrounds to determine how quorum sensing influences toxJ and toxR expression. The expression of toxJ was increased fivefold in the presence of tofR compared with the expression in the absence of tofR (Fig. 7B). To confirm that toxJ expression is upregulated by quorum sensing, we added exogenous C8-HSL to S2K18 mutant (toxJ::Tn3-gusA18 and tofI::Ω); toxJ expression was increased by adding 100 nM C8-HSL, indicating that TofR and C8-HSL complex activated toxJ expression (Fig. 7B). No significant variation in toxR expression was observed in the tofI or tofR mutant (data not shown). The palindromic nucleotides CTG and CAG separated by 10 bp, which are essential for promoter function in P. aeruginosa lasB (Rust et al., 1996), were found in the two putative lux boxes of tofI (data not shown).

ToxR requires toxoflavin as a coinducer to activate expression of the two tox operons

Because two regulatory genes, toxJ and toxR, are involved in regulating toxin production in B. glumae, we postulated that ToxR might regulate toxJ expression, ToxJ might regulate toxR expression, or both ToxR and ToxJ might be required to activate expression of the biosynthetic genes. To determine which one of these hypotheses is correct, Tn3-gusA fusions in toxJ and toxR were constructed in toxR::Ω mutant BGS4 and toxJ::Ω mutant BGS5 backgrounds, respectively. When the β-glucuronidase activity of each construct was measured, the expression of toxJ was not increased in the presence of toxR compared with the expression in the absence of toxR (data not shown). No significant variation in toxR expression was observed in the absence of toxJ (data not shown). Therefore, we concluded that both ToxR and ToxJ are required to activate expression of the biosynthesis and transporter genes. To assess how ToxR affects the expression of the two tox operons, Tn3-gusA fusions in the toxABCDE and toxFGHI operons were constructed in the wild-type strain BGR1 and in the toxR::Ω mutant BGS4. When the β-glucuronidase activity of each construct was measured, no expression of either operon was observed in the absence of toxR(Fig. 8A and Table 2). The expression of the two operons required both ToxR and 1 µM toxoflavin (Fig. 8A and Table 2). This indicated that ToxR requires toxoflavin as a coinducer to activate the expression of both operons. The presence of the T-N11-A sequence in the upstream region of the toxABCDE and toxFGHI operons supports the hypothesis that ToxR regulates the expression of both operons (data not shown).

image

Figure 8. The expression of the toxABCDE operon requires ToxR and toxoflavin (A) and ToxJ (B). The toxABCDE operon was not expressed in the absence of toxR or toxoflavin, while the expression was increased when toxR and 1 µM of toxoflavin were provided (A). The toxABCDE operon was not expressed in the presence of toxR and 1 µM of toxoflavin without toxJ (B). All values are the means ± SDs of values from triplicate experiments.

Download figure to PowerPoint

Table 2.  Expression of toxFGHI::Tn3-gusA fusions in toxR and toxJ backgrounds.
Fusions CoinducerSpecific activity of β-glucuronidase
Wild-type straintoxR::ΩtoxJ::Ω
+++
  1. Cultures were grown at 37°C in LB medium in the presence or absence of toxoflavin (1 µM) as an inducing compound.

  2. β-Glucuronidase activity was measured as described in Experimental procedures.

  3. One unit of β-glucuronidase was defined as 1 nmol of 4-methyllumbelliferone released per bacterium per minute.

  4. Values are means ± SDs of three replicates.

toxF::Tn3-gusA341411 ± 341531 ± 451.7 ± 0.11.9 ± 0.10.2 ± 0.1150 ± 7
toxG::Tn3-gusA8581139 ± 221604 ± 521.6 ± 0.11.8 ± 0.10.2 ± 0.1139 ± 9
toxH::Tn3-gusA201369 ± 241660 ± 700.6 ± 0.10.7 ± 0.10.2 ± 0.1126 ± 5
toxI::Tn3-gusA70  31 ± 2  86 ± 50.2 ± 0.10.2 ± 0.10.2 ± 0.1 3.9 ± 0.2

Expression of the two tox operons is dependent on ToxJ

Given that the BGS5 (toxJ::Ω) mutant was defective in toxoflavin production, we examined whether ToxJ functions as an activator for the expression of the toxABCDE and toxFGHI operons. To address this question, Tn3-gusA fusions in the toxABCDE and toxFGHI operons were constructed in the wild-type strain BGR1 and in the toxJ::Ω mutant BGS5. The expression of the toxABCDE operon absolutely required ToxR and 1 µM toxoflavin, and the expression was activated with ToxJ compared with the expression level in the absence of toxJ (Fig. 8A and B). The expression of the toxFGHI operon required ToxR and 1 µM toxoflavin, while its expression was less affected by ToxJ than was the expression of the toxABCDE operon (Table 2). These results indicate that both tox operons require not only ToxR and toxoflavin, but also ToxJ.

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

This study investigated how toxoflavin biosynthesis and transport genes are organized and how toxoflavin production is regulated in B. glumae. We found that the toxoflavin biosynthesis genes of B. glumae BGR1 are nearly identical to previously reported genes and confirmed that toxABCDE is polycistronic, as reported previously (Suzuki et al., 2004). We identified the toxoflavin transport system and found that fervenulin is probably transported by another system. Fervenulin is the tautomeric isomer of toxoflavin and has a methyl group on the N8 portion of the fervenulin backbone, which suggests that an additional biosynthesis gene is needed. Therefore, it is plausible that fervenulin uses a different transport system owing its chemical difference from toxoflavin.

Toxoflavin transporters appear to be RND superfamily transporters. RND-type multidrug transporters have broad substrate specificity and protect bacterial cells from the actions of antibiotics on both sides of the cytoplasmic membrane (Nikaido, 1996; Grkovic et al., 2002). The genetic organization of toxoflavin efflux systems is similar to those of previously reported RND-type transporters, but differs from the other systems in that there is an additional protein, ToxF. RND-type transporters are located in the inner membrane and function in concert with accessory proteins belonging to the MFP family and the outer membrane protein (OMP) family (Wandersman, 1992; Nikaido, 1996; Grkovic et al., 2002). ToxF appears to be an integral membrane protein; however, it is not clear how ToxF associates with typical RND-type three-component transporter systems. In B. glumae, it is very likely that toxH encodes an RND-type transporter protein, toxG encodes MFP, and toxI encodes OMP. In other Gram-negative bacteria, individual outer membrane channels are often unique for each transporter and are expressed in the same gene cluster, such as OprJ, OprM and OprN in P. aeruginosa or MtrE in Neisseria gonorrhoeae (Poole et al., 1996; Morita et al., 2001; Poole, 2001; Shafer et al., 2001). Interestingly, mutations in toxI did not affect toxoflavin transport, indicating that B. glumae may not need a functioning OMP channel to pump solute molecules directly into the medium, circumventing the outer membrane barrier. The slower growth of the toxFGH mutants in vitro versus the growth of the wild type is a good indication that toxoflavin biosynthesized inside the cells is toxic, resulting from the inability of the mutants to pump it out. Growth arrest caused by added toxoflavin to the toxFGH mutants is another indication that ToxFGH is important for pumping out the toxin, as reported in other (Grkovic et al., 2002).

It has been proposed that ToxR regulates the toxABCDE operon, because disruption of toxR abolished the ability to produce toxoflavin and resulted in loss of toxA mRNA expression (Shingu and Yoneyama, 2004). However, it has not been proved that ToxR regulates the toxABCDE operon genetically. ToxR is a LysR-type regulatory protein, and these proteins activate target genes in the presence of coinducers (Schell, 1993). We demonstrated that ToxR and the toxoflavin complex regulates the expression of both the toxABCDE and toxFGHI operons. It was thought that toxoflavin is a coinducer for ToxR. The fact that the toxJ mutant produces a very low level of toxoflavin led us to believe that basal levels of ToxR and toxoflavin are present in the system; however, it is not quite ready to start toxoflavin biosynthesis. The bacterium wants to ensure that toxoflavin production is regulated in a cell-density dependent manner via another regulator, ToxJ. It is very unlikely that other gene products are involved in toxoflavin modification by behaving as a coinducer. If exogenous toxoflavin underwent detoxification, we would not see the growth arrest of the toxFGH mutants.

The toxE mutant BGK840 probably produces small amounts of toxins as compared to the wild type, owing to the high similarity of toxE with the deaminase for riboflavin biosynthesis. This implies that toxoflavin is synthesized in part via a biosynthetic pathway common to that of riboflavin (Suzuki et al., 2004). The expression of toxE was very high in the BGK840, whereas that of toxABCD was very low without toxoflavin. This suggests that the BGK840 produces sufficient toxoflavin to act as a coinducer of ToxR.

As toxoflavin was produced when B. glumae cells reached the late exponential phase, it is very likely that quorum sensing regulates toxoflavin production. By cloning tofR and tofI, generating proper tofR and tofI mutants, and analysing the signals chemically, we clearly demonstrated that C8-HSL is a key signal triggering toxoflavin biosynthesis. TofR responded to C10-HSL, but did not respond to C6-HSL, suggesting its different biological roles in B. glumae. The presence of putative lux boxes in the tofI promoter region is consistent with our observation that TofR and C8-HSL complex is essential for activating tofI transcription. The minimum concentration (100 nM) of C8-HSL needed to confer toxoflavin production to the tofI mutant BGS2 was higher than that reported for other biological systems (Hwang et al., 1995). This indicates that B. glumae cells do not start producing toxoflavin until the cells reach high densities.

We showed that quorum sensing regulates toxoflavin production; however, it was not clear whether TofR and C8-HSL complex regulates the expression of the toxABCDE and toxFGHI operons directly. If TofR and C8-HSL complex mediates the regulation of both operons indirectly, perhaps the complex regulates toxJ expression, given that ToxJ is a transcriptional regulator. It is unlikely that ToxJ is a LuxR-type regulator because it does not require N-acyl HSL to be active (unpubl. data). The putative DNA-binding region at the C-terminus of ToxJ was similar to that of other transcriptional regulators, but there were none of the conserved amino acid residues for N-acyl HSL binding that are found in LuxR-type regulators (Slock et al., 1990). The presence of a possible binding box for TofR and C8-HSL complex was consistent with our finding that TofR and C8-HSL complex activates toxJ expression. Based on the finding that the expression of the two tox operons depends on ToxJ, we concluded that quorum sensing regulates toxJ expression and then ToxJ activates both operons, along with ToxR and toxoflavin complex.

Combining the genetic organization of the two operons and the circuits regulating toxoflavin biosynthesis and transport, we devised a working model for how B. glumae cells regulate toxoflavin production as the most important virulence factor (Fig. 9). The bacterium probably has multilevel regulation for toxoflavin biosynthesis and transport in order to save energy. Phytotoxin production consumes large amounts of energy, so that B. glumae cells must ensure that they reach certain cell densities before they start to produce it. The B. glumae cells use toxoflavin as a coinducer to amplify toxoflavin biosynthesis and transport with ToxR as a very efficient way of producing highly toxic phytotoxins. This might explain why bacterial rot of rice is very serious and dangerous if it occurs under optimum environmental conditions. Quorum sensing systems are critical in the pathogenicity of P. aeruginosa (Passador et al., 1993). In B. pseudomallei, the causative agent of melioidosis of humans and animals, a quorum sensing system plays a key role in virulence by modulating the production of MprA protease (Valade et al., 2004). To our knowledge, this is the first and only study to demonstrate that quorum sensing is involved in the pathogenicity of plant pathogenic bacteria by regulating the phytotoxin production indirectly. It is actually activated by the transcriptional activators ToxJ and ToxR and toxoflavin complex.

image

Figure 9. The toxoflavin production and regulation circuits in B. glumae BGR1. The five toxoflavin biosynthesis genes (toxABCDE) and the four genes (toxFGHI) responsible for its transport are polycistronic. The LysR-type regulatory protein ToxR activates both operons and requires toxoflavin (double hexagon shape) as a coinducer. B. glumae BGR1 produces C8-HSL and its cognate receptor TofR. At a critical autoinducer concentration, the TofR protein (diamond shape) binds to C8-HSL (pentagon shape) and activates toxJ transcription. ToxJ in turn activates the transcription of the toxABCDE and toxFGHI operons in a cascade fashion. Open rectangles indicate lux box-like sequences, and closed rectangles indicate LysR-type regulatory protein binding sites. The oval represents a bacterial cell.

Download figure to PowerPoint

Experimental procedures

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

Bacterial strains, plasmids, media and growth conditions

The bacterial strains and plasmids used in this study are listed in Table 1. All the strains were cultured in Luria–Bertani (LB) medium. The B. glumae strain BGR1 and Escherichia coli were cultured at 37°C. Antibiotics were used at the following concentrations: ampicillin, 100 µg ml−1; chloramphenicol, 20 µg ml−1; gentamicin, 50 µg ml−1; kanamycin, 50 µg ml−1; nalidixic acid, 20 µg ml−1; rifampicin, 50 µg ml−1; spectinomycin, 25 µg ml−1; and tetracycline, 10 µg ml−1. Two complementary HSL biosensors were used: A. tumefaciens strain NT1 (pDCI41E33) and C. violaceum strain CV026. Agrobacterium and Chromobacterium strains were grown at 28°C. The Agrobacterium indicator strain was grown in AB minimal mannitol liquid medium (Chilton et al., 1974). 5-Bromo-4-chloro-3-indoyl-β-d-galactopyranoside (Xgal) was used at 40 µg ml−1 when necessary.

Recombinant DNA techniques

Chromosomal DNA from B. glumae strain BGR1 was isolated using the method of Sambrook et al. (1989). Small-scale plasmid DNA isolation from E. coli was performed using QIAprep® Spin (Qiagen) according to the supplier's instructions, or by the alkaline lysis method (Sambrook et al., 1989). Small-scale cosmid DNA was isolated from E. coli and B. glumae using the alkaline lysis method. Large-scale plasmid and cosmid DNA preparations were performed by alkaline lysis followed by cesium chloride density gradient centrifugation (Sambrook et al., 1989). The restriction enzyme digestions were performed as recommended by the suppliers (Takara). The gel electrophoresis was performed in agarose gels (0.7%[w/v]), and Southern transfers were performed on Hybond-N nylon membranes (Amersham Biosciences, Uppsala, Sweden) as described by the manufacturer. For colony hybridization, all the procedures used to prepare the probe DNA and for hybridization were as described by the manufacturer (Amersham Biosciences, Uppsala, Sweden). To construct a genomic library of strain BGR1, 20–30 kb DNA inserts were prepared by partial digestion of total genomic DNA with Sau3AI size fractionated in sucrose gradients, and ligated into the BamHI site of pLAFR3. The ligated DNA was packaged into bacteriophage λ, as described by the manufacturer (Promega), and then transfected into E. coli HB101. All the pLAFR3 derivatives were mobilized into B. glumae BGR1 strains using triparental mating (Figurski and Helinski, 1979).

To  amplify  the toxA  DNA  probe,  PCR  was  performed using a PTC-150 MiniCycler (MJ research) under the following  conditions:  an  initial  96°C  for  2 min,  followed  by 30 cycles of 96°C for 1 min, 50°C for 1 min and 72°C for 1 min. The following PCR primers were used: TOXA1 (5′-ATGAG TACGACAGCTCGA-3′) and TOXA2 (5′-TCAAGGCTTG CAGACCAG-3′). All the other basic molecular techniques used were as described by Sambrook et al. (1989).

Transposon mutagenesis

pBGT6, pBGT7 and pBGT17 carrying all the essential biosynthesis genes of toxoflavin and pBGA18 with AI activity were mutagenized with Tn3-gusA, as described by Bonas et al. (1989). The insertion site and orientation of Tn3-gusA in each mutant were mapped using restriction enzyme digestion analysis and direct sequencing of the plasmid using the primer Tn3gus (5′-CCGGTCATCTGAGACCATTAAAAGA-3′), which allows sequencing out of the Tn3-gusA. The mutagenized plasmids that carried Tn3-gusA insertions were introduced individually into the parent strain BGR1 by conjugation and marker-exchanged into strain BGR1, as described previously (Fellay et al., 1989). All marker-exchanges were confirmed by Southern hybridization analysis.

Construction of Ω cassette insertion mutants

The Ω cassette from pHP45Ω was inserted into the unique BglII or SnaBI site within the tofR or tofI sequence in pBGA37R to give pBGA47R and pBGA48, respectively. Similarly, the Ω cassette was inserted into the unique HincII or HindIII site within the toxR or toxJ sequence to generate pBGT31 or pBGT35, respectively. These plasmids were introduced into B. glumae BGR1 to generate the tofR::Ω, tofI::Ω, toxR::Ω and toxJ::Ω mutant strains (BGS1, BGS2, BGS4 and BGS5, respectively) using marker-exchange mutagenesis. We then introduced Tn3-gusA fusions that had been constructed in the toxA, toxB, toxC, toxD, toxE, toxF, toxG, toxH, toxI, toxJ and toxR genes into BGS1, BGS2, BGS4, BGS5 and wild-type strain BGR1 using marker-exchange, after which we measured the β-glucuronidase activity.

β-Glucuronidase assay

The β-glucuronidase enzyme assay was performed as described previously, with some modifications (Jefferson et al., 1987). All B. glumae BGR1 derivatives were grown in LB medium, centrifuged, resuspended in GUS extraction buffer, and lysed using sonication with a VCX-400 sonicator (Sonics and Materials). The extract was used in the β-glucuronidase enzyme assay with 4-methylumbelliferyl glucuronide as the substrate. The fluorescence was measured at 365 nm excitation and 460 nm emission in a TKO100 fluorometer (Hoefer Scientific Instruments). One unit of β-glucuronidase was defined as 1 nm of released 4-methylumbelliferone per bacterium per minute.

DNA sequencing and data analysis

The DNA inserted in pBGT17 and pBGA18 was digested with the appropriate restriction enzymes and subcloned into pBluescript II SK(+) before sequencing. Universal and reverse primers were used for the primary reactions, and synthetic primers were then used to sequence both strands. The DNA sequence data were analysed using the programs blast at the National Center for Biotechnology Institute (Gish and States, 1993), MEGALIGN (DNASTAR), and GENETYX-WIN (Software Development).

RT-PCR analysis

Strain BGR1 was grown in LB medium to the exponential growth phase (12 h after inoculation); total RNA was isolated as described (Aiba et al., 1981); and the RNA samples were treated with RQ1 DNase (Promega) to remove any contaminating DNA. RT-PCR analysis was performed as follows. Two micrograms of total RNA from B. glumae BGR1 were reverse transcribed into cDNA using AMV reverse transcriptase XL, as described by the manufacturer (Takara). The reverse transcription reaction was performed at 50°C for 1 h, followed by 5 min at 75°C. PCR reactions were performed using a PTC-150 MiniCycler (MJ research) under the following conditions: an initial 96°C for 2 min, followed by 40 cycles of 96°C for 1 min, 50°C for 1 min and 72°C for 1 min. The following primer was used for RT reactions: RT1 (5′-CCGTCTTCA GCGTGTCCTCG-3′ from the toxI sequence). The following PCR primers were used: PCR1 (5′-CCGTCTTCAGCGTG TCCTCG-3′ and 5′-TCGCCGCCTGCACGATG-3′), PCR2 (5′-TCAGCGCGATCTCGATCAGC-3′ and 5′-CCTGGTGC CCCTGGTGTTC-3′), PCR3 (5′-CGAGGTTCTGGCCGCTC AG-3′ and 5′-ACCTCTTCATCCGCCGCC-3′), PCR4 (5′-CTTGCCCTGGGTGGTGCTC-3′ and 5′-GGCGATATGGTC GCGGTGG-3′), PCR5 (5′-AGCTGC CCCACGTATTTCTGC-3′ and 5′-CGGTGGTGTTCGGCGG-3′), PCR6 (5′-CCGTGC CACAGGAACAGG-3′ and 5′-GCCCTGGTGCTGACGCC-3′). Southern hybridization and DNA sequencing were used to confirm all of the RT-PCR products (Fig. 3B). As a positive control, pBGT17 DNA was used as a PCR template. As a negative control, PCR reactions with the same primer sets were performed using RNA samples that had not been reverse transcribed.

Autoinducer assays

Samples, in 1–4 µl volumes, were applied to C18 reversed-phase TLC plates (Merck), and chromatograms were developed with methanol/water (60:40, v/v). After development, the solvent was evaporated, and the dried plates were overlaid with soft agar containing bacterial indicators. A 1 ml overnight culture of the bacterial indicators was used to inoculate 50 ml of AB for Agrobacterium indicator strain and LB for Chromobacterium indicator strain. After the agar solidified, the overlaid plates were incubated at 28°C in a closed plastic container. For plate assays, the medium containing biosensor strains was poured into culture plates. When the agar had solidified, the strains to be tested for AI production were spotted directly onto the surface of the plate. The plates were incubated at 28°C overnight.

Synthesis and purification of N-acyl homoserine lactones and phytotoxins

The HSLs used in this study –N-butanoyl-l-homoserine lactone (C4-HSL), N-hexanoyl-l-homoserine lactone (C6-HSL), N-octanoyl-l-homoserine lactone (C8-HSL), N-decanoyl-l-homoserine lactone (C10-HSL), N-dodecanoyl-l-homoserine lactone (C12-HSL) – were synthesized as described by Zhang et al. (1993). Analytical TLC was performed on Merck silica gel 60 F254 precoated plates. 1H NMR spectra were taken on Vraian Gemini 300, Inova 400, or Inova 500 MHz spectrometers. 1H NMR shifts were referenced to tetramethylsilane. Stock solutions at 10 mM in acetonitrile (HPLC grade) were diluted into the growth medium to give the stated concentrations.

The supernatant (4 l) from stationary-phase cultures of B. glumae BGR1 was extracted with ethyl acetate (1:1). The ethyl acetate was removed by rotary evaporation, and the residue was reconstituted in 0.5 ml of methanol. For a given sample, a portion of the concentrated extract was diluted in ethyl acetate, and 1 µl was spotted onto a reporter TLC plate along with a set of standards. The remainder of the concentrated sample was applied to a second TLC plate as a series of two to four spots along the baseline, and the two plates were chromatographed in the same chamber. Following the development of the reporter TLC plate with CV026 strain, the C18 matrix in the regions of the preparative plate corresponding to the compound to be analysed were scraped off, combined, and extracted three times with 2 ml volumes of methanol. The combined extracts were clarified by centrifugation and passed through a fine sintered glass filter. The filtrates were taken to dryness, and the residue was resolved in 0.5 ml of methanol and applied to a C18 reverse-phase column (Sep-Pak® Vac C18 Cartridges). The fractions were eluted with a linear gradient of methanol in water (20–100%). We  collected  10  fractions,  and  assayed  these  fractions using the CV026 biosensor strain. The active fractions were analysed using electrospray ionization mass spectrometry (ESI-MS) (JMS-LC mate, JEOL).

The phytotoxins used (fervenulin, reumycin and toxoflavin) were synthesized as described previously (Yoneda et al., 1971; Nagamatsu, 2001). Stock solutions at 10 mM in methanol (HPLC grade) were diluted into the growth medium to give the stated concentrations.

The phytotoxin productivity assay was performed as described previously, with some modifications (Iiyama et al., 1995). Each strain was inoculated into 5 ml of LB broth and incubated at 37°C for 24 h with shaking. To determine the phytotoxin productivity of the strains, 5 ml of chloroform was added to the cell-free supernatants with centrifugation at 13 000× g to extract the phytotoxins. After evaporation, the chloroform residue was dissolved in 5 ml of aqueous 80% methanol, and the absorbance at 260 nm was measured. The concentration of phytotoxin produced by each strain was determined by comparison with the standard curve of the purified phytotoxin at different concentrations. The chloroform extract of LB was used as a blank control in the spectrophotometer to substract any background absorbance. For analytical TLC, the chloroform extracts were concentrated in vacuo using a rotary evaporator at below 40°C. The residue was dissolved in 10 µl of methanol and then developed on a silica gel 60 TLC plate (Merck) with chloroform/methanol (95:5, v/v). The spots were detected under UV light (365 nm).

Plant inoculation

Rice plants (Oryza sativa cv. Milyang 23) were grown in a greenhouse, inoculated at the flowering stage with a bacterial suspension (108−109 cfu ml−1) using an atomizer (Binks Wren airbrush; Binks, Glendale Heights, IL), and kept in a greenhouse. The disease in the rice plants was evaluated daily for 10 days, as described by Iiyama et al. (1995), using the following scale: 0 = healthy panicle, 1 = panicle 0–20% discoloured, 2 = panicle 20–40% discoloured, 3 = panicle 40–60% discoloured, 4 = panicle 60–80% discoloured, 5 = panicle 80–100% discoloured. Disease severity was determined using: Disease degree = Σ(number of samples per score × score)/the total number of panicles. Pathogenicity assays were repeated three times with three replications.

Acknowledgements

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

This work was supported by Grant No. CG1412 from the Crop Functional Genomics Center of the 21st Century Frontier Research Program, which is funded by Ministry of Science and Technology of the Republic of Korea. J.K., J.-G.K., Y.K., J.Y.J. and S.K. are recipients of graduate fellowships from the Ministry of Education as part of the Brain Korea 21 Project.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  • Aguilar, C., Friscina, A., Devescovi, G., Kojic, M., and Venturi, V. (2003) Identification of quorum-sensing-regulated genes of Burkholderia cepacia. J Bacteriol 185: 64566462.
  • Aiba, H., Adhya, S., and De Crombrugghe, B. (1981) Evidence for two functional gal promoters in intact Escherichia coli cells. J Biol Chem 256: 1190511910.
  • Bonas, U., Stall, R.E., and Staskawicz, B.J. (1989) Genetic and structural characterization of the avirulence gene avrBs3 from Xanthomonas campestris pv. vesicatoria Mol General Gene 218: 127136.
  • Chilton, M.D., Currier, T.C., Farrand, S.K., Bendich, A.J., Gordon, M.P., and Nester, E.W. (1974) Agrobacterium tumefaciens DNA and PS8 bacteriophage DNA not detected in crown gall tumors. Proc Natl Acad Sci USA 71: 36723676.
  • Cook, D.M., Li, P.-L., Ruchaud, F., Padden, S., and Farrand, S.K. (1997) Ti plasmid conjugation is independent of vir : reconstitution of the tra functions from pTiC58 as a binary system. J Bacteriol 179: 12911297.
  • Cottyn, B., Cerez, M.T., Van Outryve, M.F., Barroga, F., Swings, J., and Mew, T.W. (1996) Bacterial disease of rice I. Pathogenic bacteria associated with sheath rot complex and grain discoloration of rice in the Philippines. Plant Dis 80: 429437.
  • Durbin, R.D. (1991) Bacterial phytoxins: mechanism of action. Experientia 47: 776783.
  • Fellay, R., Krisch, H.M., Prentki, P., and Frey, J. (1989) Omegon-Km: a transposable element designed for in vivo insertional mutagenesis and cloning of genes in gram-negative bacteria. Gene 76: 215226.
  • Figurski, D.H., and Helinski, D.R. (1979) Replication of an origin-containing derivative of plasmid RK2 dependent on a plasmid function provided in trans. Proc Natl Acad Sci USA 76: 16481652.
  • Fuqua, C., Winans, S.C., and Greenberg, E.P. (1996) Census and consensus in bacterial ecosystems: the LuxR-LuxI family of quorum-sensing transcriptional regulators. Annu Rev Microbiol 50: 727751.
  • Gish, W., and States, D.J. (1993) Identification of protein coding regions by database similarity search. Nat Genet 3: 266272.
  • Grkovic, S., Brown, M.H., and Skurray, R.A. (2002) Regulation of bacterial drug export systems. Microbiol Mol Biol Rev 66: 671701.
  • Gross, D.C. (1991) Molecular and genetic analysis of toxin production by pathovars of Pseudomonas syringae. Annu Rev Phytopathol 29: 247278.
  • Huynh, T.V., Dahlbeck, D., and Staskawicz, B.J. (1989) Bacterial blight of soybean: regulation of a pathogen gene determining host cultivar specificity. Science 245: 13741377.
  • Hwang, I., Cook, D.M., and Farrand, S.K. (1995) A new regulatory element modulates homoserine lactone-mediated autoinduction of Ti plasmid conjugal transfer. J Bacteriol 177: 449458.
  • Iiyama, K., Furuya, N., Takanami, Y., and Matsuyama, N. (1995) A role of phytotoxin in virulence of Pseudomonas glumae. Ann Phytopathol Soc Jpn 61: 470476.
  • Jefferson, R.A., Kavanagh, T.A., and Bevan, M.W. (1987) GUS fusions: β-glucuronidase as a sensitive and versatile gene fusion marker in higher plants. EMBO J 6: 39013907.
  • Jeong, Y., Kim, J., Kim, S., Kang, Y., Nagamatsu, T., and Hwang, I. (2003) Toxoflavin produced by Burkholderia glumae causing rice grain rot is responsible for inducing bacterial wilt in many field crops. Plant Dis 87: 890895.
  • Latuasan, H.E., and Berends, W. (1961) On the origin of the toxicity of toxoflavin. Biochim Biophys Acta 52: 502508.
  • Lewenza, S., and Sokol, P.A. (2001) Regulation of ornibactin biosynthesis and N-acyl-l-homoserine lactone production by CepR in Burkholderia cepacia. J Bacteriol 183: 22122218.
  • Matsuda, I., and Sato, Z. (1988) Regulation between pathogenicity and pigment productivity in the causal agent of bacterial grain rot of rice. Ann Phytopathol Soc Jpn 54: 378.
  • McClean, K.H., Winson, M.K., Fish, L., Taylor, A., Chhabra, S.R., Camara, M., et al. (1997) Quorum sensing and Chromobacterium violaceum: exploitation of violacein production and inhibition for the detection of N-acylhomoserine lactones. Microbiology 143: 37033711.
  • Miller, M.B., and Bassler, B.L. (2001) Quorum sensing in bacteria. Annu Rev Microbiol 55: 165199.
  • Morita, Y., Komori, Y., Mima, T., Kuroda, T., Mizushima, T., and Tsuchiya, T. (2001) Construction of a series of mutants lacking all of the four major mex operons for multidrug efflux pumps or possessing each one of the operons from Pseudomonas aeruginosa PAO1: MexCD-OprJ is an inducible pump. FEMS Microbiol Lett 202: 139143.
  • Nagamatsu, T. (2001) Syntheses, transformation, and biological activities of 7-azapteridine antibiotics: toxoflavin, fervenulin, reumycin and their analogs. Recent Res Devel Org Bioorg Chem 4: 97121.
  • Nikaido, H. (1996) Multidrug efflux pumps of gram-negative bacteria. J Bacteriol 178: 58535859.
  • Passador, L., Cook, J.M., Gambello, M.J., Rust, L., and Iglewski, B.H. (1993) Expression of Pseudomonas aeruginosa virulence genes requires cell-to-cell communication. Science 260: 11271130.
  • Poole, K. (2001) Multidrug efflux pumps and antimicrobial resistance in Pseudomonas aeruginosa and related organisms. J Mol Microbiol Biotechnol 3: 255264.
  • Poole, K., Tetro, K., Zhao, Q., Neshat, S., Heinrichs, D.E., and Bianco, N. (1996) Expression of the multidrug resistance operon mexA-mexB-oprM in Pseudomonas aeruginosa: mexR encodes a regulator of operon expression. Antimicrob Agents Chemother 40: 20212028.
  • Prentki, P.A., and Krisch, H.M. (1984) In vitro insertional mutagenesis with a selectable DNA fragment. Gene 29: 303313.
  • Rust, L., Pesci, E.C., and Iglewski, B.H. (1996) Analysis of the Pseudomonas aeruginosa elastase (lasB) regulatory region. J Bacteriol 178: 11341140.
  • Sambrook, J., Fritsch, E.F., and Maniatis, T. (1989) Molecular Cloning: a Laboratory Manual, 2nd edn. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press.
  • Sato, Z., Koiso, Y., Iwasaki, S., and Matsuda, I. (1989) Toxins produced by Pseudomonas glumae. Ann Phytopathol Soc Jpn 55: 356359.
  • Schell, M.A. (1993) Molecular biology of the LysR family of transcriptional regulators. Annu Rev Microbiol 47: 597626.
  • Shafer, W.M., Veal, W.L., Lee, E.H., Zarantonelli, L., Balthazar, J.T., and Rouquette, C. (2001) Genetic organization and regulation of antimicrobial efflux systems possessed by Neisseria gonorrhoeae and Neisseria meningitidis. J Mol Microbiol Biotechnol 3: 219224.
  • Shingu, Y., and Yoneyama, K. (2004) Eessential regulator gene toxR for toxoflavin biosynthesis of Burkholderia glumae. J General Plant Pathol 70: 108114.
  • Simon, R., Priefer, U., and Pühler, A. (1983) A broad host range mobilization system for in vivo genetic engineering: transposon   mutagenesis   in   gram-negative   bacteria. Bio/Technol 1: 784791.
  • Slock, J., Kolibachuk, D., and Greenberg, E.P. (1990) Critical regions of the Vibrio fischeri LuxR protein defined by mutational analysis. J Bacteriol 172: 39743979.
  • Stachel, S.E., An, G., Flores, C., and Nester, E.W. (1985) A Tn3 lacZ transposon for the random generation of β-galactosidase gene fusions: application to the analysis of gene expression in Agrobacterium. EMBO J 4: 891898.
  • Staskawicz, B., Dahlbeck, D., Keen, N., and Napoli, C. (1987) Molecular characterization of cloned avirulence genes from race 0 and race 1 of Pseudomonas syringae pv. glycinea. J Bacteriol 169: 57895794.
  • Suzuki, F., Sawada, H., Azegami, K., and Tsuchiya, K. (2004) Molecular characterization of the tox operon involved in toxoflavin biosynthesis of Burkholderia glumae. J General Plant Pathol 70: 97107.
  • Suzuki, F., Zhu, Y., Sawada, H., and Matsuda, I. (1998) Identification of proteins involved in toxin production by Pseudomonas glumae. Ann Phytopathol Soc Jpn 64: 7579.
  • Tsushima, S., Naito, H., and Koitabashi, M. (1996) Population dynamics of Pseudomonas glumae, the causal agent of bacterial grain of rice, on leaf sheaths of rice plants in relation to disease development in the field. Ann Phytopathol Soc Jpn 62: 108113.
  • Valade, E., Thibault, F.M., Gauthier, Y.P., Palencia, M., Popoff, M.Y., and Vidal, D.R. (2004) The PmlI-PmlR quorum-sensing system in Burkholderia pseudomallei plays a key role in virulence and modulates production of the MprA protease. J Bacteriol 186: 22882294.
  • Walton, J.D., and Panaccione, D.G. (1993) Host-selective toxins and disease specificity; perspectives and progress. Annu Rev Phytopathol 31: 275303.
  • Wandersman, C. (1992) Secretion across the bacterial outer membrane. Trends Genet 8: 317322.
  • Whitehead, N.A., Barnard, A.M.L., Slater, H., Simpson, N.J.L., and Salmond, G.P.C. (2001) Quorum-sensing in gram-negative bacteria. FEMS Microbiol Rev 25: 365404.
  • Yoneda, F., Shinomura, K., and Nishigaki, S. (1971) A convenient synthesis of toxoflavins and toxoflavin-n-oxides. Tetrahedron Lett 13: 851854.
  • Yoshimura, D., Yoshida, K., and Otofuji, M. (1987) Effect of meteorological factors on the occurrence of rice bacterial grain rot. Proc Assoc Plant Prot Kyushu 33: 912.
  • Zhang, L.H., Murphy, P.J., Kerr, A., and Tate, M.E. (1993) Agrobacterium conjugation and gene regulation by N-acyl-l-homoserine lactones. Nature 362: 446448.