Studies on the Extracellular Polysaccharides (EPS) Produced in vitro by Pseudomonas phaseolicola I. Indications for a Polysaccharide Resembling Alginic Acid in Seven P. syringae Pathovars

Extracellular polysaccharides (EPS) produced by Pseudomonas syringae pv. phaseolicola are obviously composed of two main components: the long known levan consisting of fructose, and a mannuronan consisting mainly of mannuronic acid (manA), thus resembling alginic acid (alginate). The identification of manA was established by TLC utilizing different developing systems, and by cellulose acetate electrophoresis in different buffers. References were authentic uronic acids and hydrolyzed authentic alginate. A rough quantification of the “alginate” present in crude EPS was achieved with a selective colour reaction which largely excluded compounds other than uronic acids. Levan was only synthesized with sucrose as primary carbon source. When grown on several other sugars and related compounds “alginate” was the predominant component of the EPS. Additionally, rhamnose, fucose, glucose and amino sugars were found in some instances in hydrolysates of crude EPS, suggesting the release of lipopolysaccharides (LPS) from the bacterial cell walls during culture. Growth on carbon sources not related to sugars resulted in these “LPS” as the main constituent of EPS. After cultivation with sucrose, the “alginate” was restricted to the “slime” fraction of the EPS. In the “capsular” fraction, levan was predominating. A screening program revealed the capacity to synthesize the “alginate” in six additional P. syringae pathovars: pisi, lachrymans, aptata, tomato, syringae, and glycinea. All of the strains tested so far produced levan from sucrose, however, the “alginate” was formed not by all of them. There was a tendency that fresh isolates produced more “alginate” than strains subcultured for an extended time in vitro. This was also true for the total amount of EPS.     

Genes required for pathogenicity and hypersensitivity are conserved and interchangeable among pathovars of Pseudomonas syringae

Summary A group of pathogenicity genes was previously identified in Pseudomonas syringae pv. phaseolicola which controls the ability of the pathogen to cause disease on bean and to elicit the hypersensitive response on non-host plants. These genes, designated hrp, are located in a ca. 20 kb region which was referred to as the hrp cluster. Homologous sequences to DNA segments derived from this region were detected in several pathovars of P. syringae but not in symbiotic, saprophytic or other phytopathogenic bacteria. A Tn5-induced Hrp^- mutation was transferred from P. syringae pv. phaseolicola to P. syringae pv. tabaci and to three races of P. syringae pv. glycinea by marker exchange mutagenesis. The resulting progeny were phenotypically Hrp^-, i.e. no longer pathogenic on their respective hosts and unable to elicit the hypersensitive response on non-host plants. These mutants were restored to wild-type phenotype upon introduction of a recombinant plasmid carrying the corresponding wild-type locus from P. syringae pv. phaseolicola. The marker exchange mutants of P. syringae pv. glycinea psg0 and Psg5 which carry different avr genes for race specific avirulence did not elicit a hypersensitive response on incompatible soybean cultivars. It appears, therefore, that P. syringae pathovars possess common genes for pathogenicity which also control their interaction with non-host plants. Furthermore, the expression of race/cultivar specific incompatibility of P. syringae pv. glycinea requires a fully functional hrp region in addition to the avr genes which determine avirulence on single-gene differential cultivars of soybean.     

Early detection of bean infection by Pseudomonas syringae in asymptomatic leaf areas using chlorophyll fluorescence imaging

Chlorophyll fluorescence imaging has been used to analyse the response elicited in Phaseolus vulgaris after inoculation with Pseudomonas syringae pv. phaseolicola 1448A (compatible interaction) and P. syringae pv. tomato DC3000 (incompatible interaction). With the aim of modulating timing of symptom development, different cell densities were used to inoculate bean plants and the population dynamics of both bacterial strains was followed within the leaf tissue. Fluorescence quenching analysis was carried out and images of the different chlorophyll fluorescence parameters were obtained for infected as well as control plants at different timepoints post-infection. Among the different parameters analysed, we observed that non-photochemical quenching maximised the differences between the compatible and the incompatible interaction before the appearance of visual symptom. A decrease in non-photochemical quenching, evident in both infiltrated and non-infiltrated leaf areas, was observed in P. syringae pv. phaseolicola-infected plants as compared with corresponding values from controls and P. syringae pv. tomato-infected plants. No photoinhibitory damage was detected, as the maximum photosystem II quantum yield remained stable during the infection period analysed.     

Vergleich eines "glatten" und eines "rauhen" Stammes von Pseudomonas syringae pv. phaseolicola

Comparison of a “smooth” and a “rough” isolate of Pseudomonas syringae pv. phaseolicola The “smooth” (S) wild strain of Pseudomonas syringae pv. phaseolicola was compared with a “rough” (R) variant of low virulence. Both strains grew nearly equally well on a sucrose containing medium with yeast extract and casamino acids, and the strains did not differ markedly in the quantity of produced EPS (= extracellular polysaccharides). Principally the same results were obtained for high and medium concentrations of sucrose, or when sucrose was replaced by glucose or fructose. However, on glucose and fructose considerably lower quantities of EPS were produced. The biological activity of S-EPS was higher than that of R-EPS. This difference between the EPS preparations was not as marked as leaf inoculation with both bacterial isolates. After prolonged bacterial culture the EPS-production increased further, so that the differences between both strains decreased. A different EPS type was produced on the glycerol containing medium of KING B. Variations in the composition of this medium resulted in different morphology of the agar grown cultures, and the relative differences between S and R bacteria changed. When 62 different physiological tests for both bacterial strains were compared, the “rough” bacteria revealed a lowered range of positive reactions, with a few exceptions. However, it appeared unlikely that the reduced virulence of the “rough” bacteria was due to these differences. Obviously, defects in the extracellular products, but not in levan, were responsible for the reduction of virulence.     

Multiple copies of a DNA sequence from Pseudomonas syringae pathovar phaseolicola abolish thermoregulation of phaseolotoxin production

Phaseolotoxin, a phytotoxin of Pseudomonas syringae pv. phaseolicola, is produced at 18°C but not at 28°C. Here we report that a fragment (24.4 kb) cloned from the wild-type strain, which does not harbour a gene(s) involved in phaseolotoxin biosynthesis, abolishes this thermoregulation in the wild type and suppresses a Tox^? mutant at both temperatures. A subclone harbouring a 465bp fragment contains motifs that are characteristic of DNA-binding sites. In mobility shift assays we have detected a protein(s) from the wild-type and the mutant strains, grown at appropriate temperatures, that specifically binds to the fragment containing the DNA-binding motifs. We propose that the binding protein is a repressor which is ‘titrated’ by this fragment when it is present in the cell on a multiple copy plasmid, thus allowing expression of phaseolotoxin genes.     

Ornithine carbamoyltransferase genes and phaseolotoxin immunity in Pseudomonas syringae pv. phaseolicola

Two different DNA fragments encoding ornithine carbamoyltransferase (OCTase) were cloned from Pseudomonas syringae pv. phaseolicola NPS3121. These fragments did not cross-hybridize and encoded OCTases which differed with respect to their sensitivity to purified phaseolotoxin, an OCTase inhibitor produced by this phytopathogenic bacterium. Recombinant plasmids carrying these DNA fragments complemented OCTase-deficient strains of Escherichia coli and Pseudomonas aeruginosa. Extracts of the complemented E. coli strain contained OCTase enzyme activities with similar degrees of sensitivity to purified phaseolotoxin as extracts of P.s.phaseolicola grown at either 20 or 30°C. The OCTase activity detectable in extracts of P.s.phaseolicola grown at 20°C is insensitive to phaseolotoxin while that detectable in extracts of cells grown at 30°C is sensitive to the toxin. E.coli HB101 harboring recombinant plasmids carrying the gene(s) encoding the phaseolotoxin-insensitive enzyme activity exhibited resistance to purified phaseolotoxin. The results of Tn5 mutagenesis and Southern blotting and the pattern of complementation of OCTase-deficient and Tox^- mutant strains suggest that the gene(s) encoding the phaseolotoxin-insensitive OCTase is part of a gene cluster involved in phaseolotoxin production.     

Differences Between Lipopolysaccharide Compositions of Plant Pathogenic and Saprophytic Pseudomonas Species

Lipopolysaccharides (LPS) were obtained by washing cells of plant pathogenic and saprophytic Pseudomonas species with saline (fraction 1) and then with saline-EDTA (fraction 2). The cells subsequently were extracted with phenol to yield a third aqueous preparation (fraction 3). Each fraction type contained the LPS components, lipid A, heptose, 2-keto-3-deoxy sugar, and neutral and amino sugars. The neutral sugar compositions of fractions 1, 2, and 3, although similar within a species, differed between the Pseudomonas species. The LPS of two pathovars (pv.) of Pseudomonas syringae had glucose and rhamnose as major components: 13 (±3)% glucose and 87 (±3)% rhamnose for P. syringae pv. pisi and 18 (±5)% glucose and 76 (±2)% rhamnose for P. syringae pv. syringae. Fucose was present in addition to glucose and rhamnose for P. syringae pv. phaseolicola (68 [±8]% rhamnose, 14 [±1]% fucose, and 14 [±5]% glucose) and P. syringae pv. tabaci (24 [±2]% rhamnose, 54 [±3]% fucose, and 17 [±1]% glucose). The LPS from different races of P. syringae pv. pisi and P. syringae pv. phaseolicola could not be distinguished by neutral sugar composition. Three saprophytic species, P. aeruginosa, P. fluorescens, and P. putida, also produced LPS which had different proportions of rhamnose, fucose, and glucose. The LPS from three isolates of P. putida were distinct in possessing a high proportion of amino sugar and containing glucose as the major neutral sugar component (86 to 100%). The LPS fractions from plant pathogenic and saprophytic Pseudomonas species did not elicit browning or phytoalexin production in treated dark red kidney bean cotyledons or red Mexican bean leaves. Rather, chlorosis of the LPS-treated leaf tissue was observed.     

An ELISA method to identify the phytotoxic Pseudomonas syringae pv. actinidiae exopolysaccharides: A tool for rapid immunochemical detection of kiwifruit bacterial canker

The phytotoxic exopolysaccharides produced by Pseudomonas syringae pv. actinidiae, the causal agent of bacterial canker of kiwifruit, were isolated and partially identified. Their phytotoxic activity was evaluated on host and non-host plants and their role in the complex mechanisms of host-pathogen interaction was also discussed. The phytotoxic exopolysaccharides, which are natural antigens, were used to arise specific antibodies by rat immunization. The antibodies were used to develop a rapid and specific method to unambiguously detect P.s. pv. actinidiae exopolysaccharides isolated from bacterial culture and infected plant samples. Indeed, the antibodies recognized the exopolysaccharides produced by other two strains of P. s. pv. actinidiae but did not cross reacted with those isolated from P. s. pv. syringae and Pseudomonas viridiflava culture filtrates. Finally, the same antibodies significantly recognized the exopolysaccharides extracted from infected kiwi leaves.     

In vitro culture response of common bean explants to filtrate from Pseudomonas syringae pv. Phaseolicola and correlation with disease resistance

Summary The in vitro culture responses from different explants of a race-specific resistant cultivar (Red Mexican) and a racesusceptible cultivar (Palme?a) to halo-blight pathogen (Pseudomonas syringae pv. phaseolicola were studied. Two kinds of filtrate obtained from a phaseolotoxin producer wild type and a non-producer mutant of P. syringae pv. phaseolicola race-7 were used. Callus formation of Red Mexican was significantly reduced in the presence of phaseolotoxin. Bud-shoot growth was more sensitive than callus formation to other metabolites present in the pathogen filtrate, but the presence of phaseolotoxin in the media showed a positive correlation between resistance to halo blight race-7 pathogen and bud-shoot growth. Our results indicate that differential in vitro responses are influenced by the plant genotype and by the metabolite composition and concentration of the filtrate.     

Insight into Types I and II nonhost resistance using expression patterns of defense-related genes in tobacco

Plants protect themselves against pathogens using a range of response mechanisms. There are two categories of nonhost resistance: Type I, which does not result in visible cell death; and Type II, which entails localized programmed cell death (or hypersensitive response) in response to nonhost pathogens. The genes responsible for these two systems have not yet been intensively investigated at the molecular level. Using tobacco plants (Nicotiana tabacum), we compared expression of 12 defense-related genes between a Type I (Xanthomonas axonopodis pv. glycines 8ra) nonhost interaction, and two Type II (Pseudomonas syringae pv. syringae 61 and P. syringae pv. phaseolicola NPS3121) nonhost interactions, as well as those expressed during R gene-mediated resistance to Tobacco mosaic virus. In general, expression of most defense-related genes during R gene-mediated resistance was activated 48 h after challenge by TMV; the same genes were upregulated as early as 9 h after infiltration by nonhost pathogens. Surprisingly, X. axonopodis pv. glycines (Type I) elicited the same set of defense-related genes as did two pathovars of P. syringae, despite the absence of visible cell death. In two examples of Type II nonhost interactions, P. syringae pv. phaseolicola NPS3121 produced an expression profile more closely resembling that of X. axonopodis pv. glycines 8ra, than that of P. syringae pv. syringae 61. These results suggest that Type I nonhost resistance may act as a mechanism providing a more specific and active defense response against a broad range of potential pathogens.     

Polysaccharides of <Emphasis Type="Italic">Pseudomonas</Emphasis> Pathovar Strains That Infect Pea,Tomato, and Soya Bean

In order to understand the mode of action of taxonomically related Pseudomonas syringae pathovar strains that infect pea, tomato, and soya bean, we examined their extracellular polysaccharides (EPS). Maximum production of polysaccharide in shake culture of these pathogens was observed between 24 and 60 h. P. syringae pv. pisi 519, the bacterial blight pathogen of pea, produced a higher amount of polysaccharide (34.87 g/mL) at 60 h compared with 32.67 g/mL produced by P. syringae pv. glycinea NCPPB 1783, the bacterial blight pathogen of soya bean, and 30.03 g/mL produced by P. syringae pv. tomato NCPPB 269, the bacterial speck pathogen of tomato. EPS produced by P. syringae pv. pisi 519, P. syringae pv. tomato NCPPB 269, and P. syringae pv. glycinea NCPPB 1783 was characterized with infrared (FTIR), nuclear magnetic resonance (NMR), high performance thin layer chromatography, (HPTLC), and matrix-assisted laser desorption/ionization time-of-flight (MALDI-TOF) mass spectrometry. HPTLC profiles revealed the presence of glucose and glucuronic acid in all bacteria and mannose only in P. syringae pv. tomato. Molecular mass of EPS of P. syringae pv. pisi (m/z 933.8), P. syringae pv. tomato (m/z 950.4), and P. syringae pv. glycinea (m/z 933.5) was confirmed by MALDI-TOF mass spectrometry.     

Localized Changes in Peroxidase Activity Accompany Hydrogen Peroxide Generation during the Development of a Nonhost Hypersensitive Reaction in Lettuce

Peroxidase activity was characterized in lettuce (Lactuca sativa L.) leaf tissue. Changes in the activity and distribution of the enzyme were examined during the development of a nonhost hypersensitive reaction (HR) induced by Pseudomonas syringae (P. s.) pv phaseolicola and in response to an hrp mutant of the bacterium. Assays of activity in tissue extracts revealed pH optima of 4.5, 6.0, 5.5 to 6.0, and 6.0 to 6.5 for the substrates tetramethylbenzidine, guaiacol, caffeic acid, and chlorogenic acid, respectively. Inoculation with water or with wild-type or hrp mutant strains of P. s. pv phaseolicola caused an initial decline in total peroxidase activity; subsequent increases depended on the hydrogen donor used in the assay. Guaiacol peroxidase recovered more rapidly in tissues undergoing the HR, whereas changes in tetramethylbenzidine peroxidase were generally similar in the two interactions. In contrast, increases in chlorogenic acid peroxidase were significantly higher in tissues inoculated with the hrp mutant. During the HR, increased levels of Mn²⁺/2,4-dichlorophenol-stimulated NADH and NADPH oxidase activities, characteristic of certain peroxidases, were found in intercellular fluids and closely matched the accumulation of H₂O₂ in the apoplast. Histochemical analysis of peroxidase distribution by electron microscopy revealed a striking, highly localized increase in activity within the endomembrane system and cell wall at the sites of bacterial attachment. However, no clear differences in peroxidase location were observed in tissue challenged by the wild-type strain or the hrp mutant. Our results highlight the significance of the subcellular control of oxidative reactions leading to the generation of reactive oxygen species, cell wall alterations, and the HR.     

Antibacterial Activity of Tea and Coffee Wastes Against Some Plant Pathogenic Pseudomonas syringae Strains

In vitro and greenhouse trials were conducted to elucidate the potential use of extracts of tea and coffee wastes to control plant diseases caused by the bacterial pathogens, Pseudomonas syringae pv. pisi (race 1 and 2) and P. s. pv. phaseolicola (race 1). The antibacterial activity was measured as the diameter of the inhibition zone in agar and also by periodical viable cell counts in laboratory tests. The effect on the hypersensitive reaction and the potential for disease control after leaf infiltration and seed treatment were studied on bean plants in the greenhouse. Results showed that both the tea and coffee extracts possessed antibacterial activity against the three pathogens, but that the effects varied depending on the strain and the test method. Strong reduction of halo blight disease and improvement in plant growth was obtained in presence of the coffee extract. For the halo blight pathogen, P. s. pv. phaseolicola, there was a good correlation between the results from the viable cell count method and the greenhouse tests, but the results from the in vitro studies did not agree with those from greenhouse as regards the P. s. pv. pisi strains. It is suggested that the component(s) in tea and coffee responsible for controlling the bean pathogen may not be the same as that for the pea pathogens.     

Multilocus enzyme electrophoresis analysis ofPseudomonas syringae pv.phaseolicola

A multilocus enzyme electrophoresis technique was developed to detect variation in seven enzyme loci among isolates ofPseudomonas syringae pv.phaseolicola, representing three races from different geographical locations, the causal agent of the halo blight disease of beans. Cellulose acetate gel electrophoresis of seven enzymes revealed 19 electrotypes (ET) among 21Pseudomonas syringae pv.phaseolicola isolates. One of the pathovarsyringae and one of the pathovartomato isolates were represented by two different ET. The population of Turkish isolates and three races of the pathovarphaseolicola appeared to be genetically diverse.     

Impact of the exopolysaccharides Pel and Psl on the initial adhesion of Pseudomonas aeruginosa to sand

In this study, the impact of the exopolysaccharides Pel and Psl on the cell surface electron donor-electron acceptor (acid-base) properties and adhesion to quartz sand was investigated by using Pseudomonas aeruginosa PAO1 and its isogenic EPS-mutant strains Δpel, Δpsl and Δpel/Δpsl. The microbial adhesion to hydrocarbon (MATH) test and titration results showed that both Pel and Psl contribute to the surface hydrophobicity of the cell. The results of contact angle measurement, however, showed no correlation with the cell surface hydrophobicity measured by the MATH test and the titration method. Packed-bed column experiments indicated that the exopolysaccharides Pel and Psl are involved in the initial cell attachment to the sand surface and the extent of their impact is dependent on the ionic strength (IS) of the solution. Overall, the Δpel/Δpsl double mutant had the lowest adhesion coefficient to sand compared with the wild-type PAO1, the Δpel mutant and the Δpsl mutant. It is hypothesized that in addition to bacterial surface hydrophobicity and DLVO forces, other factors, eg steric repulsion caused by extracellular macromolecules, and cell surface appendages (flagella and pili) also contribute significantly to the interaction between the cell surface and a sand grain.     

Comparative analysis of the lipopolysaccharides of a rough and a smooth strain of Pseudomonas syringae pv. phaseolicola

The lipopolysaccharides (LPS) of a rough (R) and a smooth (S) strain of Pseudomonas syringae pv. phaseolicola were analysed. The S-LPS revealed markedly more rhamnose and fucose, but less glucose, than the R-LPS. The presence of 3-O-methyl-rhamnose (acofriose) in the S-LPS was confirmed by cochromatography with authentic acofriose. SDS polyacrylamide gel electrophoresis of the S-LPS demonstrated a cluster of regularly spaced high molecular weight fractions, which was almost lacking in the R-LPS. The main fatty acids of the lipid A of both LPS species were 3-OH-10:0,3-OH-12:0,2-OH-12:0, and 12:0. Two N-linked diesters were demonstrated: 3-O(12:0)-12:0 and 3-O(2-OH-12:0)-12:0. S-LPS was subjected to mild hydrolysis and the degraded polysaccharide separated into three fractions by gel permeation chromatography on a Fractogel TSK HW-50 column. Fraction I, representing nearly only the O-specific side chain, consisted of rhamnose and fucose in a molar ratio of 4:1, with 4% of the rhamnose being 3-O-methylated (acofriose). Fraction II, representing mostly core material, was composed of glucose, rhamnose, heptose, glucosamine, galactosamine, alanine, and a still unidentified amino compound, in an approximate molar ratio of 3:1:1:1:1:1:1, and KDO. Fraction III consisted of released monomers and salts. The LPS was highly phosphorylated (3.28% phosphorus in the core fraction). The thus characterized composition of the LPS O-chain seems to be unique for the pathovar phaseolicola of P. syringae, although many similarities exist to other pathovars as well as to other bacterial species.Abbreviations LPS lipopolysacchairdes - GC/MS combined gas liquid chromatography-mass spectrometry - HVE high voltage electrophoresis - KDO 2-keto-3-deoxyoctonic acid - PAGE polyacrylamide gel electrophoresis - SDS sodium dodecylsulfate P.s. pv. phaseolicola is termed P. phaseolicola in the text     

Heterologous expression of the gene for the ethylene-forming enzyme fromPseudomonas syringae in the cyanobacteriumSynechococcus

Summary Bioconversion of atmospheric carbon dioxide to ethylene was studied in a recombinant cyanobacterium. The gene for the ethylene-forming enzyme ofPseudomonas syringae pv.phaseolicola PK2 was cloned and expressed in the cyanobacteriumSynechococcus PCC7942 R2-SPc by use of a shuttle vector pUC303. The ethylene-forming activityin vivo ofSynechococcus PCC7942 R2-SPc that carried the gene for the ethylene-forming enzyme ofP. syringae pv.phaseolicola PK2 was one-fifth of that ofE. coli JM109 that harbored the same plasmid. The enzyme accounted for 0.021% by weight of the total soluble protein inSynechococcus PCC7942 R2-SPc.     

The Release of Alginate Lyase from Growing Pseudomonas syringae pathovar phaseolicola

Pseudomonas syringae pathovar phaseolicola, which produces alginate during stationary growth phase, displayed elevated extracellular alginate lyase activity during both mid-exponential and late-stationary growth phases of batch growth. Intracellular activity remained below 22% of the total activity during exponential growth, suggesting that alginate lyase has an extracellular function for this organism. Extracellular enzyme activity in continuous cultures, grown in either nutrient broth or glucose–simple salts medium, peaked at 60% of the washout rate, although nutrient broth-grown cultures displayed more than twice the activity per gram of cell mass. These results imply that growth rate, nutritional composition, or both initiate a release of alginate lyase from viable P. syringae pv. phaseolicola, which could modify its entrapping biofilm. Received: 14 April 2000 / Accepted: 11 August 2000     

Construction of a Bacterial Artificial Chromosome Library and Characterization of hrp/hrc Gene Cluster of Pseudomonas Syringae Pathovar tagetis LMG5090

Pseudomonas syringae pv. tagetis causes apical chlorosis of several plant species in the Asteraceae, including marigold. As a means to facilitate the isolation of pathogenicity genes and to characterize the genome of this bacterium, we have constructed a bacterial artificial chromosome library of P. syringae pv. tagetis strain LMG5090. The library consists of 1,536 clones with insert size ranging from 30 to 160 kb and an average size of 86 kb. Based upon colony hybridization, the BAC clone 420E23 containing the hrp/hrc gene cluster encoding the type III secretion system was identified from this library and subsequently shotgun sequenced. The hrp/hrc gene cluster of P. syringae pv. tagetis has a 23 kb sequence which contains 27 open reading frames. Comparative analysis of the hrp/hrc gene cluster of P. syringae pv. tagetis LMG5090, P. syringae pv. tomato DC3000, P. syringae pv. syringae B728a, and P. syringae pv. phaseolicola 1448A revealed that the entire hrp/hrc gene cluster of P. syringae pv. tagetis is conserved and identically arranged in all four pathovars