Early responses of tobacco suspension cells to rhizobacterial elicitors of induced systemic resistance

Colonization of roots by selected strains of fluorescent Pseudomonas spp. can trigger induced systemic resistance (ISR) against foliar pathogens in a plant species-specific manner. It has been suggested that early responses in cell suspension cultures in response to rhizobacterial elicitors, such as generation of active oxygen species (AOS) and extracellular medium alkalinization (MA), are linked to the development of ISR in whole plants. Perception of flagellin was demonstrated to elicit ISR in Arabidopsis, and bacterial lipopolysaccharides (LPS) have been shown to elicit several defense responses and to act as bacterial determinants of ISR in various plant species. In the present study, the LPS-containing cell walls, the pyoverdine siderophores, and the flagella of Pseudomonas putida WCS358, P. fluorescens WCS374, and P. fluorescens WCS417, which are all known to act as elicitors of ISR in selected plant species, were tested for their effects on the production of AOS, MA, elevation of cytoplasmic Ca(2+) ([Ca(2+)](cyt)), and defense-related gene expression in tobacco suspension cells. The LPS of all three strains, the siderophore of WCS374, and the flagella of WCS358 induced a single, transient, early burst of AOS, whereas the siderophores of WCS358 and WCS417 and the flagella of WCS374 and WCS417 did not. None of the compounds caused cell death. Once stimulated by the active compounds, the cells became refractory to further stimulation by any of the active elicitors, but not to the elicitor cryptogein from the oomycete Phytophthora cryptogea, indicating that signaling upon perception of the different rhizobacterial compounds rapidly converges into a common response pathway. Of all compounds tested, only the siderophores of WCS358 and WCS417 did not induce MA; the flagella of WCS374 and WCS417, although not active as elicitors of AOS, did induce MA. These results were corroborated by using preparations from relevant bacterial mutants. The active rhizobacterial elicitors led to a rapid increase in [Ca(2+)](cyt), peaking at 6 min, whereas the inactive siderophores of WCS358 and WCS417 elicited a single spike at 1 min. Elicitation of the cells by cell-wall LPS of WCS358 or the siderophore of WCS374 induced a weak, transient expression of several defense-related genes, including PAL and GST. The spectrum of early responses of the suspension cells was not matched by the expression of ISR in whole tobacco plants against Erwinia carotovora pv. carotovora. Of the live bacterial strains, only WCS358 elicited significant ISR, but application of the LPS or the siderophore of all three strains also elicited ISR. Notably, the absence of elicitation of AOS and MA in suspension-cultured cells but induction of ISR in whole plants by the siderophore of WCS358, which was lost upon treatment with the siderophore-minus mutant of WCS358, indicates that the early responses in suspension cells are not predictive of the ability to induce ISR in whole plants. Possible explanations for these discrepancies are discussed.     

Mutational changes in physiochemical cell surface properties of plant-growth-stimulating Pseudomonas spp. do not influence the attachment properties of the cells.

Bacteriophage-resistant mutant strains of the root-colonizing Pseudomonas strains WCS358 and WCS374 lack the O-antigenic side chain of the lipopolysaccharide, as was shown by the loss of the typical lipopolysaccharide ladder pattern after analysis by sodium dodecyl sulfate-polyacrylamide gel electrophoresis. These strains differed from their parent strains in cell surface hydrophobicity and in cell surface charge. The observed variation in these physicochemical characteristics could be explained by the differences in sugar composition. The mutant strains had no altered properties of adherence to sterile potato roots compared with their parental strains, nor were differences observed in the firm adhesion to hydrophilic, lipophilic, negatively charged, or positively charged artificial surfaces. These results show that neither physicochemical cell surface properties nor the presence of the O-antigenic side chain plays a major role in the firm adhesion of these bacterial cells to solid surfaces, including potato roots.     

A novel cell surface polysaccharide in Pseudomonas putida WCS358, which shares characteristics with Escherichia coli K antigens, is not involved in root colonization.

Previously we have shown that flagella and the O-specific polysaccharide of lipopolysaccharide play a role in colonization of the potato root by plant growth-promoting Pseudomonas strains WCS374 and WCS358. In this paper, we describe a novel cell surface-exposed structure in Pseudomonas putida WCS358 examined with a specific monoclonal antibody. This cell surface structure appeared to be a polysaccharide, which was accessible to the monoclonal antibody at the outer cell surface. Further study revealed that it does not contain 2-keto-3-deoxyoctonate, heptose, or lipid A, indicating that it is not a second type of lipopolysaccharide. Instead, the polysaccharide shared some characteristics with K antigen described for Escherichia coli. From a series of 49 different soil bacteria tested, only one other potato plant growth-promoting Pseudomonas strain reacted positively with the monoclonal antibody. Mutant cells lacking the novel antigen were efficiently isolated by an enrichment method involving magnetic antibodies. Mutant strains defective in the novel antigen contained normal lipopolysaccharide. One of these mutants was affected in neither its ability to adhere to sterile potato root pieces nor its ability to colonize potato roots. We conclude that the bacterial cell surface of P. putida WCS358 contains at least two different polysaccharide structures. These are the O-specific polysaccharide of lipopolysaccharide, which is relevant for potato root colonization, and the novel polysaccharide, which is not involved in adhesion to or colonization of the potato root.     

Siderophore-mediated uptake of Fe3+ by the plant growth-stimulating Pseudomonas putida strain WCS358 and by other rhizosphere microorganisms.

Under iron-limited conditions, Pseudomonas putida WCS358 produces a siderophore, pseudobactin 358, which is essential for the plant growth-stimulating ability of this strain. Cells of strain WCS358, provided that they have been grown under Fe3+ limitation, take up 55Fe3+ from the 55Fe3+-labeled pseudobactin 358 complex with Km and Vmax values of 0.23 microM and 0.14 nmol/mg of cell dry weight per min, respectively. Uptake experiments with cells treated with various metabolic inhibitors showed that this Fe3+ uptake process was dependent on the proton motive force. Furthermore, strain WCS358 was shown to be able to take up Fe3+ complexed to the siderophore of another plant-beneficial P. fluorescens strain, WCS374. The tested pathogenic rhizobacteria and rhizofungi were neither able to grow on Fe3+-deficient medium in the presence of pseudobactin 358 nor able to take up 55Fe3+ from 55Fe3+-pseudobactin 358. The same applies for three cyanide-producing Pseudomonas strains which are supposed to be representatives of the minor pathogens. These results indicate that the extraordinary ability of strain WCS358 to compete efficiently for Fe3+ is based on the fact that the pathogenic and deleterious rhizosphere microorganisms, in contrast to strain WCS358 itself, are not able to take up Fe3+ from Fe3+-pseudobactin 358 complexes.     

Suppression of bacterial wilt in Eucalyptus urophylla by fluorescent Pseudomonas spp. in China

Bacterial wilt caused by Ralstonia solanacearum race 1, biovar III has become a severe problem in Eucalyptus plantations in south China. The disease mainly attacks young eucalypt trees, and no effective control measures are available yet. To explore possibilities to develop biological control of the disease, strains of fluorescent Pseudomonas spp. that are effective in suppressing plant diseases by known mechanisms, were tested for their potential to control bacterial wilt in Eucalyptus. Pseudomonas putida WCS358r, Pseudomonas fluorescens WCS374r, P. fluorescens WCS417r, and Pseudomonas aeruginosa 7NSK2 antagonize R. solanacearum in vitro by siderophore-mediated competition for iron, whereas inhibition of pathogen growth by P. fluorescens CHA0r is antibiosis-based. No correlations were found between antagonistic activities of these Pseudomonas spp. in vitro and biocontrol of bacterial wilt in Eucalyptus in vivo. None of the strains suppressed disease when mixed together with the pathogen through the soil or when seeds or seedlings were treated with the strains one to four weeks before transfer into soil infested with R. solanacearum. However, when the seedlings were dipped with their roots in a bacterial suspension before transplanting into infested soil, P. fluorescens WCS417r significantly suppressed bacterial wilt. P. putida WCS358r was marginally effective, whereas its siderophore-minus mutant had no effect at all, indicating that siderophore-mediated competition for iron can contribute but is not effective enough to suppress bacterial wilt in Eucalyptus. A derivative of P. putida WCS358r, constitutively producing 2,4-diacetylphloroglucinol (WCS358::phl) reduced disease. Combined treatment with P. fluorescens WCS417r and P. putida WCS358::phl did not improve suppression of bacterial wilt.     

Isolation and analysis of genes involved in siderophore biosynthesis in plant-growth-stimulating Pseudomonas putida WCS358

The plant-growth-stimulating Pseudomonas putida WCS358 was mutagenized with transposon Tn5. The resulting mutant colony bank was screened for mutants defective in the biosynthesis of the fluorescent siderophore. A total of 28 mutants, divided into six different classes, were isolated that were nonfluorescent or defective in iron acquisition or both. These different types of mutants together with the probable overall structure of the siderophore, i.e., a small peptide chain attached to a fluorescing group, suggest a biosynthetic pathway in which the synthesis of the fluorescing group is preceded by the synthesis of the peptide part. A gene colony bank of P. putida WCS358 was constructed with the broad-host-range cosmid vector pLAFR1. This genomic library, established in Escherichia coli, was mobilized into the 28 individual mutants, screening for transconjugants restored in fluorescence or growth under iron-limiting conditions or both. A total of 13 cosmids were found to complement 13 distinct mutants. The complementation analysis revealed that at least five gene clusters, with a minimum of seven genes, are needed for siderophore biosynthesis. Some of these genes seem to be arranged in an operon-like structure.     

Multiple outer membrane receptors for uptake of ferric pseudobactins inPseudomonas putida WCS358

Under iron limitationPseudomonas putida WCS358 produces a fluorescent siderophore, pseudobactin 358, which, after complexing iron, is transported back into the cell via the specific outer membrane receptor PupA. In addition, this strain has the capacity to take up iron via a large variety of siderophores produced by other fluorescent pseudomonads. Putative receptor genes for such siderophores were identified in the chromosome of strain WCS358 by PCR using primers matching two domains conserved in four ferric pseudobactin receptors, including PupA. Eleven amplification products within the expected size range were obtained. Sequence analysis confirmed that the products were derived from genes encoding outer membrane receptors. Two complete receptor genes were isolated from a genomic library ofP. putida WCS358. Both protein products are involved in the transport of a limited number of specific ferric pseudobactins. These results indicate that the ability ofP. putida WCS358 to exploit many different heterologous pseudobactins is related to the presence of multiple outer membrane receptor proteins.     

Multiple outer membrane receptors for uptake of ferric pseudobactins inPseudomonas putida WCS358

Under iron limitationPseudomonas putida WCS358 produces a fluorescent siderophore, pseudobactin 358, which, after complexing iron, is transported back into the cell via the specific outer membrane receptor PupA. In addition, this strain has the capacity to take up iron via a large variety of siderophores produced by other fluorescent pseudomonads. Putative receptor genes for such siderophores were identified in the chromosome of strain WCS358 by PCR using primers matching two domains conserved in four ferric pseudobactin receptors, including PupA. Eleven amplification products within the expected size range were obtained. Sequence analysis confirmed that the products were derived from genes encoding outer membrane receptors. Two complete receptor genes were isolated from a genomic library ofP. putida WCS358. Both protein products are involved in the transport of a limited number of specific ferric pseudobactins. These results indicate that the ability ofP. putida WCS358 to exploit many different heterologous pseudobactins is related to the presence of multiple outer membrane receptor proteins.     

Siderophores and outer membrane proteins of antagonistic, plant-growth-stimulating, root-colonizing Pseudomonas spp

As an approach to understanding the molecular basis of the reduction in plant yield depression by root-colonizing Pseudomonas spp. and especially of the role of the bacterial cell surfaces in this process, we characterized 30 plant-root-colonizing Pseudomonas spp. with respect to siderophore production, antagonistic activity, plasmid content, and sodium dodecyl sulphate-polyacrylamide gel electrophoresis patterns of their cell envelope proteins. The results showed that all strains produce hydroxamate-type siderophores which, because of the correlation with Fe3+ limitation, are thought to be the major factor responsible for antagonistic activity. Siderophore-negative mutants of two strains had a strongly decreased antagonistic activity. Five strains maintained their antagonistic activity under conditions of iron excess. Analysis of cell envelope protein patterns of cells grown in excess Fe3+ showed that most strains differed from each other, although two classes of similar or identical strains were found. In one case such a class was subdivided on the basis of the patterns of proteins derepressed by iron limitation. Small plasmids were not detected in any of the strains, and only one of the four tested strains contained a large plasmid. Therefore, it is unlikely that the Fe3+ uptake system of the antagonistic strains is usually plasmid encoded.     

Antagonistic Effect of Nonpathogenic Fusarium oxysporum Fo47 and Pseudobactin 358 upon Pathogenic Fusarium oxysporum f. sp. dianthi

Pseudobactin production by Pseudomonas putida WCS358 significantly improves biological control of fusarium wilt caused by nonpathogenic Fusarium oxysporum Fo47b10 (P. Lemanceau, P. A. H. M. Bakker, W. J. de Kogel, C. Alabouvette, and B. Schippers, Appl. Environ. Microbiol. 58:2978-2982, 1992). The antagonistic effect of Fo47b10 and purified pseudobactin 358 was studied by using an in vitro bioassay. This bioassay allows studies on interactions among nonpathogenic F. oxysporum Fo47b10, pathogenic F. oxysporum f. sp. dianthi WCS816, and purified pseudobactin 358, the fluorescent siderophore produced by P. putida WCS358. Both nonpathogenic and pathogenic F. oxysporum reduced each other's growth when grown together. However, in these coinoculation experiments, pathogenic F. oxysporum WCS816 was relatively more inhibited in its growth than nonpathogenic F. oxysporum Fo47b10. The antagonism of nonpathogenic F. oxysporum against pathogenic F. oxysporum strongly depends on the ratio of nonpathogenic to pathogenic F. oxysporum densities: the higher this ratio, the stronger the antagonism. This fungal antagonism appears to be mainly associated with the competition for glucose. Pseudobactin 358 reduced the growth of both F. oxysporum strains, whereas ferric pseudobactin 358 did not; antagonism by pseudobactin 358 was then related to competition for iron. However, the pathogenic F. oxysporum strain was more sensitive to this antagonism than the nonpathogenic strain. Pseudobactin 358 reduced the efficiency of glucose metabolism by the fungi. These results suggest that pseudobactin 358 increases the intensity of the antagonism of nonpathogenic F. oxysporum Fo47b10 against pathogenic F. oxysporum WCS816 by making WCS816 more sensitive to the glucose competition by Fo47b10.     

Cloning and characterisation of the rpoS gene from plant growth-promoting Pseudomonas putida WCS358: RpoS is not involved in siderophore and homoserine lactone production

The rpoS gene which encodes a stationary phase sigma factor has been identified and characterised from the rhizosphere-colonising plant growth-promoting Pseudomonas putida strain WCS358. The predicted protein sequence has extensive homologies with the RpoS proteins form other bacteria, in particular with the RpoS sigma factors of the fluorescent pseudomonads. A genomic transposon insertion in the rpoS gene was constructed, these mutants were analysed for their ability to produce siderophore (iron-transport agent) and the autoinducer quorum-sensing molecules called homoserine lactones (AHL). It was determined that RpoS was not involved in the regulation of siderophore and AHL production, synthesis of these molecules is important for gene expression at stationary phase. P. putida WCS358 produces at least three different AHL molecules.     

Colonization of the <Emphasis Type="Italic">Arabidopsis</Emphasis> rhizosphere by fluorescent <Emphasis Type="Italic">Pseudomonas</Emphasis> spp. activates a root-specific,ethylene-responsive <Emphasis Type="Italic">PR-5</Emphasis> gene in the vascular bundle

Plants of which the roots are colonized by selected strains of non-pathogenic, fluorescent Pseudomonas spp. develop an enhanced defensive capacity against a broad spectrum of foliar pathogens. In Arabidopsis thaliana, this rhizobacteria-induced systemic resistance (ISR) functions independently of salicylic acid but requires responsiveness to jasmonic acid and ethylene. In contrast to pathogen-induced systemic acquired resistance (SAR), ISR is not associated with systemic changes in the expression of genes encoding pathogenesis-related (PR) proteins. To identify genes that are specifically expressed in response to colonization of the roots by ISR-inducing Pseudomonas fluorescens WCS417r bacteria, we screened a collection of Arabidopsis enhancer trap and gene trap lines containing a transposable element of the Ac/Ds system and the GUS reporter gene. We identified an enhancer trap line (WET121) that specifically showed GUS activity in the root vascular bundle upon colonization of the roots by WCS417r. Fluorescent Pseudomonas spp. strains P. fluorescens WCS374r and P. putida WCS358r triggered a similar expression pattern, whereas ISR-non-inducing Escherichia coli bacteria did not. Exogenous application of the ethylene precursor 1-aminocyclopropane-1-carboxylate (ACC) mimicked the rhizobacteria-induced GUS expression pattern in the root vascular bundle, whereas methyl jasmonic acid and salicylic acid did not, indicating that the Ds element in WET121 is inserted in the vicinity of an ethylene-responsive gene. Analysis of the expression of the genes in the close vicinity of the Ds element revealed AtTLP1 as the gene responsible for the in cis activation of the GUS reporter gene in the root vascular bundle. AtTLP1 encodes a thaumatin-like protein that belongs to the PR-5 family of PR proteins, some of which possess antimicrobial properties. AtTLP1 knockout mutant plants showed normal levels of WCS417r-mediated ISR against the bacterial leaf pathogen Pseudomonas syringae pv. tomato DC3000, suggesting that expression of AtTLP1 in the roots is not required for systemic expression of ISR in the leaves. Together, these results indicate that induction of AtTLP1 is a local response of Arabidopsis roots to colonization by non-pathogenic fluorescent Pseudomonas spp. and is unlikely to play a role in systemic resistance.     

Characterization and taxonomic significance of lipopolysaccharides of Leptospira interrogans serovar hardjo

Lipopolysaccharides (LPSs) from Leptospira interrogans serovar hardjo (reference strain hardjoprajitno and strain hardjobovis) were prepared by the hot phenol-water procedure. High yields of LPSs were found in the phenol phase. Gel electrophoresis of the phenol phase LPSs showed similar patterns for all strains in contrast to the different patterns found in the water phase LPSs. Sugar composition was also similar among all strains with rhamnose as the predominant sugar. Mannosamine was detected by high performance thin layer and gas-liquid chromatography. 2-Keto-3-deoxyoctonic acid (KDO) was comparable with authentic KDO by paper chromatography. Periodate oxidation at near neutral pH with or without prior hydrolysis showed that most of the KDO was substituted. The fatty acid composition of strain hardjobovis LPS was slightly different from that of the reference strain hardjoprajitno. Myristic and 3-hydroxymyristic acid were not detected in any of the LPS preparations. In conjunction with genetic and other data, the two strains are sufficiently different to be regarded as members of two separate species sharing common antigens. There is sufficient evidence to rename the hardjoprajitno strain type L. interrogans hardjo-p, and the hardjobovis strain type L. borgpeterseni hardjo-b.     

The sss colonization gene of the tomato-Fusarium oxysporum f. sp. radicis-lycopersici biocontrol strain Pseudomonas fluorescens WCS365 can improve root colonization of other wild-type pseudomonas spp.bacteria

We show that the disease tomato foot and root rot caused by the pathogenic fungus Fusarium oxysporum f. sp. radicis-lycopersici can be controlled by inoculation of seeds with cells of the efficient root colonizer Pseudomonas fluorescens WCS365, indicating that strain WCS365 is a biocontrol strain. The mechanism for disease suppression most likely is induced systemic resistance. P. fluorescens strain WCS365 and P. chlororaphis strain PCL1391, which acts through the production of the antibiotic phenazine-1-carboxamide, were differentially labeled using genes encoding autofluorescent proteins. Inoculation of seeds with a 1:1 mixture of these strains showed that, at the upper part of the root, the two cell types were present as microcolonies of either one or both cell types. Microcolonies at the lower root part were predominantly of one cell type. Mixed inoculation tended to improve biocontrol in comparison with single inoculations. In contrast to what was observed previously for strain PCL1391, mutations in various colonization genes, including sss, did not consistently decrease the biocontrol ability of strain WCS365. Multiple copies of the sss colonization gene in WCS365 improved neither colonization nor biocontrol by this strain. However, introduction of the sss-containing DNA fragment into the poor colonizer P. fluorescens WCS307 and into the good colonizer P. fluorescens F113 increased the competitive tomato root tip colonization ability of the latter strains 16- to 40-fold and 8- to 16-fold, respectively. These results show that improvement of the colonization ability of wild-type Pseudomonas strains by genetic engineering is a realistic goal.     

Phosphorylcholine decoration of lipopolysaccharide differentiates commensal Neisseriae from pathogenic strains: identification of licA-type genes in commensal Neisseriae

Phosphorylcholine (ChoP) is a potential candidate for a plurispecific vaccine, because it is present on surface components of many mucosal organisms, including Haemophilus influenzae, Streptococcus pneumoniae and Pseudomonas aeruginosa. In addition, ChoP has been detected on pili of Neisseria meningitidis and Neisseria gonorrhoeae. In this study, we demonstrate the presence of the phosphorylcholine epitope on the lipopolysaccharides (LPSs) of several species of commensal Neisseriae (Cn), a property that differentiates commensal from the pathogenic strains of Neisseriae. In an extended survey of 78 strains, we confirmed the exclusive expression of the ChoP epitope on pili of pathogenic Neisseriae. Despite the presence of pili on Cn, which are homologous to Class II pili of N. meningitidis, they did not react with anti-ChoP antibody. This observation was further supported by the fact that 14C-labelled choline was incorporated only in the LPSs of Cn. Analysis of the LPS of N. lactamica strain NL4 revealed two distinct and interconvertible molecular species of LPS with high and low levels of reactivity with anti-ChoP antibody. In addition, on/off phase variation gave rise to frequent modulation in the levels of antibody reactivity. A concurrent modulation was also observed in the binding of C-reactive protein, CRP, a ChoP-binding reactant that is implicated in bacterial clearance. Genetic analysis showed the presence of a gene in several Cn spp. with significant sequence identity to H. influenzae licA. This gene encodes choline kinase and is also involved in phase variation of the LPS-associated ChoP in H. influenzae. In contrast, licA-like genes were not identified in the pathogenic Neisseria strains tested. They are absent from N. meningitidis strain Z2491 genome database. These data suggest that the genetic basis for ChoP incorporation in Cn LPS resembles that in H. influenzae spp. and may be distinct from that generating the ChoP epitope on pili of pathogenic Neisseriae. Further, the modulation of ChoP expression on Cn LPS, and corresponding modulation of CRP binding, has the potential to confer the property of immune avoidance and thus of persistence on mucosa.