Skip to main content
NCBI home page
Applied and Environmental Microbiology logoLink to Applied and Environmental Microbiology
. 2003 Feb;69(2):1143–1153. doi: 10.1128/AEM.69.2.1143-1153.2003

High-Performance Liquid Chromatography Analyses of Pyoverdin Siderophores Differentiate among Phytopathogenic Fluorescent Pseudomonas Species

Alain Bultreys 1,*, Isabelle Gheysen 1, Bernard Wathelet 2, Henri Maraite 3, Edmond de Hoffmann 4
PMCID: PMC143633  PMID: 12571041

Abstract

The relationship of pyoverdins produced by 41 pathovars of Pseudomonas syringae and by phytopathogenic Pseudomonas species was investigated. A high-performance liquid chromatography method for analyzing the culture medium proved to be superior to isoelectric focusing for detecting pyoverdin production, for differentiating slightly different pyoverdins, and for differentiating atypical from typical Fe(III)-chelated pyoverdins. Nonfluorescent strains were found in Pseudomonas amygdali, Pseudomonas meliae, Pseudomonas fuscovaginae, and P. syringae. Pseudomonas agarici and Pseudomonas marginalis produced typical pyoverdins. Among the arginine dihydrolase-negative fluorescent Pseudomonas species, spectral, amino acid, and mass spectrometry analyses underscored for the first time the clear similarities among the pyoverdins produced by related species. Within this group, the oxidase-negative species Pseudomonas viridiflava and Pseudomonas ficuserectae and the pathovars of P. syringae produced the same atypical pyoverdin, whereas the oxidase-positive species Pseudomonas cichorii produced a similar atypical pyoverdin that contained a glycine instead of a serine. The more distantly related species Pseudomonas asplenii and Pseudomonas fuscovaginae both produced a less similar atypical pyoverdin. The spectral characteristics of Fe(III)-chelated atypical pyoverdins at pH 7.0 were related to the presence of two β-hydroxyaspartic acids as iron ligands, whereas in typical pyoverdins one of the ligands is always ornithine based. The peptide chain influenced the chelation of iron more in atypical pyoverdins. Our results demonstrated that there is relative pyoverdin conservation in the amino acids involved in iron chelation and that there is faster evolution of the other amino acids, highlighting the usefulness of pyoverdins in systematics and in identification.

Fluorescent Pseudomonas species produce Fe(III)-chelating siderophores called pyoverdins that fluorescence under UV light. The pyoverdins have the following common feature: a constant quinoline chromophore, responsible for the color of the molecule, is bound to a peptide chain and to a dicarboxylic acid or to a dicarboxylic amide (4). Pyoverdins could be useful in systematics and in identification because of the variation found in the peptide part of the molecules (5, 8, 14). Siderovars regroup strains that produce pyoverdins with the same peptide chain (27). One siderovar is generally detected by isoelectric focusing electrophoresis (IEF) studies for well-defined species (1, 26, 28). Conversely, there are different siderovars in certain species, such as Pseudomonas fluorescens (about 19 siderovars) or Pseudomonas putida (about 13 siderovars) (14). The peptide chain of one of the four pyoverdins described for Pseudomonas aeruginosa (27, 31) is synthesized by a nonribosomal pathway by means of the peptide synthetase PvdD (25). DNA hybridization studies have shown that genes comparable to pvdD are present in other fluorescent Pseudomonas species (25, 30), but no clear evolutionary relationship has been found yet among the peptide chains of the pyoverdins described for different species (14).

Fluorescent Pseudomonas species are differentiated on the basis of phenotypic traits (29). More than 40 pyoverdin peptide chain compositions have been identified in the group containing the arginine dihydrolase-positive, saprophytic or opportunistic animal-pathogenic, fluorescent Pseudomonas species (14), but only one composition has been found in the group containing the arginine dihydrolase-negative, phytopathogenic, fluorescent Pseudomonas species (7, 8, 10, 19). The latter group includes Pseudomonas syringae, Pseudomonas viridiflava, and Pseudomonas cichorii (29). Surprisingly, strains of P. viridiflava and of distantly related pathovars of P. syringae produce identical atypical pyoverdins (7, 8). A constant feature of the pyoverdins is the presence of three iron-binding ligands; one ligand is located in a catechol moiety in the chromophore, and the other two are located in the peptide chain and are hydroxamic acids derived from ornithine or β-hydroxyaspartic acid (β-OH-Asp) (4). The atypical feature of the pyoverdin of P. syringae and P. viridiflava is the implication of two β-OH-Asp residues and no derivatives of ornithine in the chelation of iron (7, 10). This feature influences the spectral characteristics of the Fe(III)-chelated atypical pyoverdin (7, 8).

Little information is available concerning the relationship of the pyoverdins produced by pathovars of P. syringae and by the other, diversely related, phytopathogenic Pseudomonas species. The taxonomic positions of P. syringae pv. coronafaciens, P. syringae pv. glycinea, P. syringae pv. phaseolicola, P. syringae pv. savastanoi, P. syringae pv. cannabina, and P. syringae pv. tremae have been discussed recently (12, 15, 16, 32, 43). Recent studies (16, 17, 18, 34, 35, 36, 37, 41) have brought Pseudomonas meliae, Pseudomonas ficuserectae, and Pseudomonas amygdali nearer the systematic group containing the phytopathogenic fluorescent Pseudomonas species and Pseudomonas marginalis, Pseudomonas agarici, Pseudomonas asplenii, and Pseudomonas fuscovaginae nearer the systematic group containing the saprophytic fluorescent Pseudomonas species.

In this study, a new high-performance liquid chromatography (HPLC) method proved to be the most accurate method for studying pyoverdin production in the culture media of strains of the phytopathogenic species mentioned above. We confirmed that P. ficuserectae, P. viridiflava, and 38 pathovars of P. syringae produce identical atypical pyoverdins. For the first time, there appeared to be clear similarities between different atypical pyoverdins produced by related species (P. syringae and P. cichorii). Also, P. fuscovaginae and P. asplenii both produced a different atypical pyoverdin. Our results provide insight into the evolution of pyoverdins and highlight the usefulness of pyoverdins in systematics and in identification.

MATERIALS AND METHODS

Bacterial strains and culture conditions.

The characteristics of strains used in this study are described in Table 1. Strain UPB 1023 was isolated from rice showing typical bacterial sheath rot symptoms in the People's Republic of China and was tentatively identified as P. fuscovaginae on the basis of some phenotypic characteristics. Nevertheless, in later tests this strain appeared to differ from P. fuscovaginae in other characteristics and to be nonpathogenic on rice (Maraite, unpublished data). It was included in the study to compare its pyoverdin profile with the pyoverdin profiles of other representatives of P. fuscovaginae from various geographical areas. A collection of 170 fluorescent and nonfluorescent oxidase-negative strains isolated from cherry, plum, and pear orchards in Belgium was investigated for pyoverdin production. Some of these strains were affiliated with P. syringae pv. syringae or P. syringae pv. morsprunorum by using PCR tests that detected the syrD or cfl genes involved in phytotoxin production (3, 6), but 41 strains were negative in these tests (Bultreys and Gheysen, unpublished data). All precultures were grown at 28°C on medium 2 agar (6). Solid-liquid cultures were grown unshaken at 20°C in petri dishes containing 10 ml of liquid medium and one block of the corresponding agar medium (7).

TABLE 1.

Characteristics of strains used in this study

Straina Geno- speciesb Host or origin Country Fluorescence on agar mediumc
Type of (dihydro) pyoverdin detected by HPLC in liquid mediumd Measured mol wt of the dominant pyoverdin
Source or reference
GASN medium King's B medium Method usede Mol wt (AMU)f
P. syringae pv. syringae strains 1
    LMG 1247T Lilac England + + Pa A 1,175g LMGh
    PsP1 Pear Belgium + + Pa A 6
P. syringae pv. aptata strains 1
    LMG 5059T Sugar beet United States + + Pa A 1,175g LMG
    UPB 110 Sugar beet Belgium − − DPa A 1,177g 6
P. syringae pv. pisi LMG 5079T 1 Pea New Zealand + + Pa A 3′ 1,175 LMG
P. syringae pv. atrofaciens LMG 5095T 1 Wheat New Zealand + + Pa A 2 1,175 LMG
P. syringae pv. papulans LMG 5076T 1 Apple Canada + + Pa A 2 1,175 LMG
P. syringae pv. aceris LMG 2106T 1 Unknown Unknown + + Pa A 3P-M′ 1,175 LMG
P. syringae pv. panici LMG 2367T 1 Unknown Unknown + + Pa A 3′ 1,175 LMG
P. syringae pv. japonica LMG 5068T 1 Barley Japan + + Pa A 2 1,175 LMG
P. syringae pv. dysoxyli LMG 5062T 1 Dysoxylum spectabile New Zealand + + Pa A 2 1,175 LMG
P. syringae pv. lapsa LMG 2206T 1 Zea sp. Unknown + + Pa A 2 1,175 LMG
P. syringae pv. glycinea strains 2
    LMG 5515 Soybean Canada + + Pa A 2 1,175 LMG
    LMG 5553 Soybean Yugoslavia + + Pa A 2 1,175 LMG
P. syringae pv. mori LMG 5074T 2 White mulberry Hungary + + Pa A 2 1,175 LMG
P. syringae pv. morsprunorum strains 2
    LMG 5075T Plum Unknown + + Pa A 1,175g LMG
    LMG 2222 Sweet cherry England + + Pa A 1,175g LMG
    PmC36 Sweet cherry Belgium − − ND 6
P. syringae pv. savastanoi LMG 2209T 2 Common olive Yugoslavia + + Pa A 3 1,175 LMG
P. syringae pv. phaseolicola LMG 2245T 2 Bean Canada + + Pa A 2 1,175 LMG
P. syringae pv. tabaci LMG 5393T 2 Tobacco Hungary + + Pa A 2 1,175 LMG
P. syringae pv. sesami LMG 2289T 2 Sesame Yugoslavia − − ND LMG
P. syringae pv. myricae LMG 5668T 2 Myrica rubra Japan + + Pa A 2 1,175 LMG
P. syringae pv. eriobotryae LMG 2184T 2 Eriobotrya japonica United States + + Pa A 3 1,175 LMG
P. syringae pv. ciccaronei LMG 5541T 2 Ceratonia siliqua Italy + + Pa A 2 1,175 LMG
P. syringae pv. mellea LMG 5072T 2 Tobacco Japan + + Pa A 3′ 1,175 LMG
P. syringae pv. ulmi LMG 2349T 2 Elm Yugoslavia + + Pa A 2 1,175 LMG
P. syringae pv. tomato LMG 5093T 3 Tomato England + + Pa A 1,175g LMG
P. syringae pv. maculicola LMG 5295 3 Radish United States + + Pa A 3′ 1,175 LMG
P. syringae pv. antirhini LMG 5057T 3 Snapdragon England + + Pa A 2 1,175 LMG
P. syringae pv. apii LMG 2132T 3 Celery United States − + Pa A 3K′ 1,175 LMG
P. syringae pv. berberidis LMG 2147 3 Barberry New Zealand + + Pa A 2 1,175 LMG
P. syringae pv. lachrymans LMG 5070T 3 Cucumber United States +/− +/− Pa A 3K 1,175 LMG
P. syringae pv. passiflorae LMG 5185T 3 Passiflora edulis New Zealand + + Pa A 3 1,175 LMG
P. syringae pv. persicae LMG 5184T 3 Peach France − − ND LMG
P. syringae pv. viburni LMG 2351T 3 Arrowwood United States + + Pa A 3′ 1,175 LMG
P. syringae pv. delphinii LMG 5381T 3 Larkspur New Zealand + + Pa A 2 1,175 LMG
P. syringae pv. ribicola LMG 2276T 3 Currant Unknown + + Pa A 3′ 1,175 LMG
P. syringae pv. coronafaciens strains 4
    LMG 2330 Unknown Unknown + + Pa A 3P-M′ 1,175 LMG
    LMG 5060 Oats England + + Pa A 3 1,175 LMG
    LMG 13190 Oats Unknown + + Pa A 3′ 1,175 LMG
P. syringae pv. garcae strains 4
    LMG 5064T Coffee Brazil − − ND LMG
    LMG 5065 Coffee Kenya + + Pa A 2 1,175 LMG
P. syringae pv. oryzae LMG 10912T 4 Rice Japan + + Pa A 3 1,175 LMG
P. syringae pv. primulae LMG 2252T 6 Primrose United States + + Pa A 3′ 1,175 LMG
P. syringae pv. tagetis LMG 5090T 7 Marigold Zimbabwe + + Pa A 2 1,175 LMG
P. syringae pv. helianthi LMG 5067T 7 Mirasol Mexico + + Pa A 3′ 1,175 LMG
P. syringae pv. theae LMG 5092T 8 Tea Japan + + Pa A 2 1,175 LMG
P. syringae pv. cannabina LMG 5096T 9 Marijuana Hungary − − ND LMG
P. viridiflava strains 6
    LMG 2352T Bean Switzerland + + Pa A 1,175g LMG
    LMG 2353 Pear England + + Pa A 3 1,175 LMG
    LMG 6480 Witloof Belgium + + Pa A 3 1,175 LMG
P. amygdali LMG 2123T 2 Almond Crete, Greece − − ND LMG
P. meliae LMG 2220T 2 Melia azedarach Japan − − ND LMG
P. ficuserectae LMG 5694T 2 Ficus erecta Japan +/− +/− Pa A 4 1,175 LMG
P. cichorii strains
    LMG 2162T Endive Germany + + Pa B 1 1,145 LMG
    LMG 5034 Dahlia sp. Zimbabwe + + Pa B 1 1,145 LMG
    LMG 8401 Lettuce Belgium + + Pa B 1 1,145 LMG/PICK>
P. asplenii strains
    LMG 2137 Fern Unknown + + Pa C 1 1,368 LMG
    LMG 5147 Unknown Unknown + + Pa C LMG
P. fuscovaginae strains
    LMG 2158T Rice Japan + + Pa C LMG
    UPB 264b Rice Burundi + + Pa C 1 1,368 6
    UPB 407 Leersia hexandra Burundi − − ND H. Maraite
    UPB 526 Wheat Mexico + + Pa C H. Maraite
    UPB 594 Wheat Mexico + + Pa C H. Maraite
    UPB 735 Rice Philippines + + Pa C H. Maraite
    UPB 736 Rice Madagascar + + Pa C H. Maraite
    UPB 737 Sorghum Burundi + + Pa C H. Maraite
    UPB 1013 Wheat Nepal + + Pa C H. Maraite
    UPB 1023 Rice People's Republic of China + + Pt 1 1,059 H. Maraite
P. agarici LMG 2112T Agaricus bisporus New Zealand + + Pt LMG
P. marginalis LMG 14572 Dahlia Unknown + + Pt LMG
P. marginalis pv. marginalis LMG 5177 Bean Unknown + + Pt LMG
P. fluorescens LMG 1794T Water England + + Pt 1,212g LMG
P. chlororaphis LMG 5004T Contaminated plate Unknown + + Pt LMG
P. putida LMG 2257T Soil United States + + Pt LMG
Open in a new tab
a

T = pathotype or type strain.

b

Genospecies of the pathotype strain of the pathovar. Data were obtained from reference 16 and were obtained by DNA-DNA hybridization.

c

Data from this study. +, positive; −, negative; +/−, weak fluorescence.

d

Data from this study. Pa A, atypical pyoverdin Pa A; DPa A, atypical dihydropyoverdin DPa A related to the atypical pyoverdin Pa A (8); Pa B, atypical pyoverdin Pa B; Pa C, atypical pyoverdin Pa C; Pt, typical pyoverdins; ND, no pyoverdin detected.

e

Data from this study. Method 1, production of large volume in GASN medium, purification by ion-exchange chromatography methods (7), and ESI MS following direct infusion; method 2, production of small volume in GASN medium, accelerated purification procedure on C18 resin, and ESI MS following direct infusion; method 3, same as method 2 but prepurification on C18 resin and ESI MS (positive ion, MS) performed in line by HPLC-MS by using the isocratic HPLC program; method 3′, same as method 3 except analysis by MS and MS-MS; method 3K, same as method 3 but production in King's B medium; method 3K′, same as method 3K except analysis by MS and MS-MS; 3P-M′, same as method 3′ but production in PRO-M medium; method 4, production of small volume in King's B medium, concentration on C18 resin, and ESI MS performed in line by HPLC-MS and HPLC-MS-MS by using the gradient HPLC program.

f

Data from this study unless otherwise specified. Molecular weight of the dominant Fe(III)-chelated pyoverdin or dihydropyoverdin.

g

Data from reference 8.

h

LMG, Laboratorium voor Microbiologie van Gent, Belgian Coordinated Collections of Microorganisms, Ghent, Belgium.

Determination of the types of the pyoverdins produced by fluorescent Pseudomonas species.

Production of fluorescent pigments was investigated at 28°C on GASN (7) and King's B (21) agar media. Detection was carried out by using UV light (wavelength, 360 nm). Solid-liquid cultures were used to investigate pyoverdin production in liquid medium. GASN medium was used, but King's B, CSGA (24), and PRO-M (7) media were used for poor pyoverdin producers. After chelation to Fe(III), the types of pyoverdins were determined visually (8) and by spectrophotometry by using either spectrophotometric analysis of the culture media (8) or the HPLC method described below.

Detection and comparison of pyoverdin production by HPLC.

Solid-liquid cultures were grown for 3 days. The liquid fractions were transferred to individual tubes, and 40 μl of an FeCl3 solution (1 M) was added to each tube. Each culture was shaken for 20 min, and the bacteria were removed by centrifugation (22 min, 10,000 × g) and filtration through a 0.2-μm-pore-size membrane filter. The HPLC analyses were carried out by using the filtered culture media adjusted to pH 5.0 to 5.3. For fine pyoverdin comparisons only, the pH was temporarily adjusted to pH 7.0, and pyoverdin production was estimated by measuring the absorbance at 403 nm in order to determine the injection volume for each strain. The aim was to compare, by using HPLC, the retention times (RT) of peaks with comparable heights. The HPLC analyses were performed with Nucleosyl C18 columns and a Waters 2190 system. The difference in RT (ΔRT) (wavelength, 403 nm) between the principal pyoverdins of an analyzed strain and a reference strain (either P. syringae pv. syringae B301D or P. fuscovaginae UPB 264b) was calculated. Within a sample set, the culture medium of the reference strain was injected for a maximum of three analyses before or after each investigated strain. When the heights of the peaks were found to differ too much, the analyses were carried out again after readjustment of the injected volumes. Pyoverdin detection was performed by using the spectra of the molecules between 200 and 500 nm obtained with a Waters PDA 996 photodiode array detector. Two HPLC programs were used, in which solution A was a 17 mM NaOH-acetic acid buffer (pH 5.3) and solution B was acetonitrile. HPLC program 1 was as follows (flow rate, 1 ml/min): 100% solution A, 1 min; from 100% solution A to 97% solution A, 2 min; 97% solution A, 9 min; and from 97% solution A to 30% solution A, 25 min. HPLC program 2 (flow rate, 0.9 ml/min) was as follows: 100% solution A, 8 min; from 100% solution A to 98% solution A, 2 min; 98% solution A, 10 min; from 98% solution A to 95% solution A, 5 min; from 95% solution A to 30% solution A, 15 min; and 30% solution A, 5 min. A comparison of pyoverdin production by P. cichorii LMG 2162, P. fluorescens LMG 1794, P. putida LMG 2257, and the pathovars of P. syringae (except P. syringae pv. sesami, P. syringae pv. apii, P. syringae pv. lachrymans, P. syringae pv. persicae, P. syringae pv. coronafaciens, and P. syringae pv. cannabina) was initially carried out in GASN medium by using HPLC program 1. HPLC program 2 and GASN medium were then generally used. The comparisons were carried out in King's B medium for P. syringae pv. apii LMG 2132 and P. syringae pv. lachrymans LMG 5070; in King's B and PRO-M media for P. syringae pv. coronafaciens LMG 2330; and in CSGA medium for P. ficuserectae LMG 5694. Nonfluorescent strains were tested on all media.

Comparison of pyoverdin production by IEF.

The unchelated and the Fe(III)-chelated pyoverdins were detected by IEF as previously described (8).

Purification of the atypical pyoverdins detected by HPLC.

Large quantities of the different atypical Fe(III)-chelated pyoverdins detected by HPLC, including the pyoverdins of P. cichorii LMG 2162, LMG 5034, and LMG 8401, P. asplenii LMG 2137, and P. fuscovaginae UPB 264b, were purified. The dominant typical pyoverdin of P. fuscovaginae strain UPB 1023, whose identity was uncertain, was also purified because of its peculiar features as a P. fuscovaginae-produced pyoverdin. Purified pyoverdins of the P. cichorii strains and of P. fuscovaginae UPB 1023 were obtained by using the procedures used to produce and purify the principal pyoverdin of P. syringae (7). The procedures were slightly different for P. fuscovaginae UPB 264b and P. asplenii LMG 2137. After addition of FeCl3 and elimination of the bacteria, the liquid media adjusted to pH 5.0 were passed through a 20-cm-long C8 column or a 10-cm-long C18 column made up in a 50 mM NaOH-acetic acid buffer (pH 5.0). The dominant product was eluted in fractions containing 50, 60, and 75% methanol in water. After evaporation, these fractions were dissolved in a 30 mM NaOH-formic acid buffer (pH 4.2) and combined. The pyoverdins of both strains were purified by repeated passage through a type CM C25 Sephadex cation-exchange chromatography column eluted with a 30 mM NaOH-formic acid buffer (pH 4.2).

Amino acid and MS analyses of purified pyoverdins.

The amino acid analyses were carried out as previously described (7). HI hydrolyses were carried out to detect the presence of ornithine. The molecular masses of the purified Fe(III)-chelated molecules were determined by mass spectrometry (MS) by using electrospray ionization (ESI) and an Ion Trap Finnigan MAT LCQ mass spectrometer as previously described (8). The molecules were compared further by tandem MS (MS-MS).

Spectral analyses.

The spectral analyses of purified Fe(III)-chelated pyoverdins were carried out as previously described (8) at pH 3.0, 3.5, 4.0, 4.5, 5.0, 5.5, 6.0, and 7.0 for P. cichorii LMG 8401 and P. asplenii LMG 2137. The Fe(III)-chelated typical pyoverdins of P. fuscovaginae UPB 1023 and P. fluorescens LMG 1794 (7) were compared at pH 7.0 and 3.0.

Preparation of the pyoverdins produced by pathovars of P. syringae and related species for MS analyses.

In order to validate the HPLC method of comparing pyoverdin production, the Fe(III)-chelated pyoverdins of the strains that appeared to produce the known atypical pyoverdin of P. syringae (8) were prepared for MS analyses. An accelerated purification procedure was developed. For each strain, four solid-liquid cultures were started in GASN medium and incubated for 48 h. A 1,000-mg Chromabond C18 column (Macherey-Nagel) was made up in 50 mM NaOH-acetic acid buffer (pH 5.0). After addition of 160 μl of FeCl3 (1 M) to the culture medium, elimination of the bacteria, and adjustment of the pH to a value near 5.0, 10 ml of the culture medium was passed through the C18 column. The column was washed with 6 ml of ultrapure water, which partially released the pyoverdin. The last 2 ml of water was collected and mixed with 2 ml of 100 mM NaOH-acetic acid buffer (pH 5.0). The C18 column was then washed with 6 ml of a water-methanol solution (1:1, vol/vol) and with 6 ml of methanol. The procedure was started again by using the water fraction collected during the first passage. The collected fraction consisting of 2 ml of water was vacuum evaporated before the MS analyses by using direct infusion. An additional desalting procedure was carried out if necessary with a C18 column. When necessary, another prepurification procedure was carried out. GASN, King's B, PRO-M, or CSGA medium cultures that had been incubated for 48 to 72 h were used, depending on fluorescence detection under UV light. A 15- to 100-ml portion of culture medium was passed through the 1,000-mg C18 column. The column was washed with 3 ml of ultrapure water, and a 9-ml fraction of ultrapure water was collected and evaporated before HPLC-MS analyses. A 6-ml water-methanol (1:1, vol/vol) fraction was also collected and evaporated prior to possible HPLC-MS analyses.

Comparison of the atypical pyoverdins produced by pathovars of P. syringae and related species by MS and by HPLC-MS.

The purified fractions were analyzed by MS following direct infusion. ESI MS analyses in the positive-ion mode or in the negative-ion mode were carried out, as were MS-MS analyses of the molecular ions of the pyoverdin of P. syringae ([M+H]+ of m/z 1176 or [M-H]− of m/z 1174) (8). The prepurified fractions collected in water from the C18 columns were analyzed by HPLC-MS by using isocratic HPLC conditions (ratio of H2O-acetic acid [pH 4.0] to acetonitrile, 97:3 [vol/vol]). Either ESI MS analyses were carried out in line in the positive-ion mode or both ESI MS analyses in the positive-ion mode and MS-MS analyses of the m/z 1176 [M+H]+ ion were carried out in line alternately. For P. ficuserectae LMG 5694, the fraction collected in water-methanol (1:1, vol/vol) was analyzed by HPLC-MS and by HPLC-MS-MS by using a gradient HPLC program identical to HPLC program 2 described above but with an H2O-acetic acid solution (pH 4.0) used as solution A.

Use of detection of pyoverdin production by spectrophotometry and by HPLC in a field study.

A total of 170 field isolates from Belgian orchards were investigated for pyoverdin production. Spectrophotometric detection of atypical pyoverdin in the culture medium was carried out as previously described (8). The pyoverdins were further identified by HPLC analyses of the GASN culture media. HPLC program 2 and the control strains P. syringae pv. syringae PsP2 and P. syringae pv. syringae B301D were used.

Detection of toxic lipodepsipeptide production by strains of P. fuscovaginae and P. asplenii.

A biological test that demonstrates the ability of P. syringae strains to produce toxic lipodepsipeptides was used; P. fuscovaginae UPB 264b and LMG 2158 gave positive results in this test previously (6). The bacteria were incubated at 28°C for 48 h before spraying of Rhodotorula pilimanae MUCL 30397, and the petri dishes were incubated at 20°C until inhibition zones were observed. Two repetitions consisting of two plates were carried out.

RESULTS

Determination of the types of the pyoverdins produced by the fluorescent Pseudomonas species investigated.

The results obtained in this study are shown in Table 1. Except for P. syringae pv. apii LMG 2132 (8), comparable results were obtained by using GASN and King's B agar media to detect fluorescence under UV light on solid culture media. The visual and spectrophotometric analyses of the liquid culture media were generally informative enough to draw conclusions about the types of pyoverdins produced. However, the final information for the very poor pyoverdin producers or for the organisms that did not produce pyoverdin was obtained by using HPLC. As the pH of the HPLC buffer used was pH 5.3, the spectral characteristics of the molecules obtained in line allowed the types of pyoverdins produced to be determined (Fig. 1). The RT data allowed discrimination between pyoverdins with different peptide chains produced by different species (Fig. 1). P. amygdali LMG 2123, P. meliae LMG 2220, and P. fuscovaginae UPB 407, as well as strains of P. syringae pv. morsprunorum, P. syringae pv. sesami, P. syringae pv. persicae, P. syringae pv. garcae, and P. syringae pv. cannabina (Table 1), were confirmed to be organisms that do not produce pyoverdin under the conditions used in this study. The strains of the phytopathogenic species P. agarici and P. marginalis produced typical pyoverdins, as did the strains of the saprophytic species P. putida, P. fluorescens, and Pseudomonas chlororaphis tested. Except for the two P. marginalis strains, these strains produced different typical pyoverdins (data not shown). P. ficuserectae LMG 5694 induced very weak fluorescence on solid media, and an atypical pyoverdin was detected by HPLC in liquid medium. This observation was not made when the spectrophotometric test was used because production was too weak. The phytopathogenic species P. syringae, P. viridiflava, and P. cichorii produced atypical pyoverdins that were indistinguishable as determined by the visual and spectrophotometric tests. They appeared very close together and very close to the pyoverdin of P. ficuserectae LMG 5694 when HPLC was used. In contrast to P. fuscovaginae UPB 1023, a strain whose identity was uncertain and which produced a typical pyoverdin, P. asplenii LMG 2137 and P. fuscovaginae UPB 264b gave very similar results. The HPLC analyses indicated that these organisms produced an atypical pyoverdin that differed from the pyoverdin of P. syringae. However, the type of pyoverdin could not be easily ascertained by the spectrophotometric tests because of production of a metabolite that absorbed at around 400 nm and interfered with the analyses; the broad charge transfer bands of typical Fe(III)-chelated pyoverdins were, however, absent. In the visual tests, the darkening of the culture medium that accompanied the decrease in the pH from pH 7.0 to 4.0 when atypical pyoverdin producers, such as P. syringae and P. cichorii, were tested (8) was not as pronounced in the P. asplenii and P. fuscovaginae strains. A persistent orange color of the culture medium at acidic pH values was a characteristic of the latter strains.

FIG. 1.

FIG. 1.

Open in a new tab

Determination by HPLC of the types of the pyoverdins produced by P. fuscovaginae UPB 1023 (A) and P. syringae pv. coronafaciens LMG 13190 (B). For each strain, the panel on the left shows the results of a global HPLC analysis (detection at 403 nm), and the panel on the right shows the spectral characteristics, analyzed in line, of the Fe(III)-chelated dominant pyoverdin. Strain UPB 1023 produced a typical pyoverdin with an RT of 22.100 min (A), and P. syringae pv. coronafaciens LMG 13190 produced an atypical pyoverdin with an RT of 18.374 min (B). The absorbance was measured each 2.4 nm. AU, absorbance units.

Fine comparisons of pyoverdin production by HPLC.

The fine comparisons were restricted to the species that produce atypical pyoverdins. As the pyoverdin profiles obtained were complex, the dominant pyoverdins of the reference strains were used in the comparisons (Fig. 2). Modifications of the HPLC programs enabled the results in the zones of interest to be refined. Thus, the results obtained with HPLC program 1 were discriminatory enough with regard to the ΔRT to distinguish P. cichorii LMG 2162 from the reference strain P. syringae pv. syringae B301D (ΔRT, about 0.160 min), but the results obtained with HPLC program 2 were even more discriminatory (Fig. 2) and showed that the ΔRT was then 0.310 ± 0.038 min for three repetitions of the comparison, whereas the ΔRT for P. syringae pv. syringae PsP1 was only 0.046 ± 0.024 min. To obtain accurate analysis results, it was essential to follow the procedures described above for comparison within a sample set. The fluorescent pathovars of P. syringae tested, the P. viridiflava strains from various host plants, and P. ficuserectae LMG 5694 produced the same atypical pyoverdin, Pa A (Table 1). The P. cichorii strains from various geographical areas and host plants produced another atypical pyoverdin, Pa B, which is closely related to Pa A (Fig. 2B). All of the strains of P. asplenii and P. fuscovaginae from various geographical areas except strain UPB 1023 from the People's Republic of China produced a third atypical pyoverdin, Pa C. Strain UPB 1023 produced a typical pyoverdin (Fig. 1A) that had a ΔRT of about 6 min compared to the atypical pyoverdin of the other P. fuscovaginae strains tested.

FIG. 2.

FIG. 2.

Open in a new tab

Comparisons of the RT of dominant Fe(III)-chelated pyoverdins obtained by using HPLC program 2. (A) Reference strain P. syringae pv. syringae B301D with an observed RT of 18.783 min for its dominant pyoverdin; (B) test strain P. cichorii LMG 2162 with an observed RT of 19.075 min for its pyoverdin; (C) test strain P. syringae pv. syringae PsP1 with an observed RT of 18.758 min for its pyoverdin. AU, absorbance units.

Comparison of pyoverdin production by IEF.

The IEF profiles of the unchelated and Fe(III)-chelated pyoverdins of the P. fuscovaginae and P. asplenii strains tested except strain UPB 1023 were comparable and different from the IEF profiles of the P. syringae strains (Fig. 3A). The IEF profiles of the unchelated and Fe(III)-chelated pyoverdins of the P. syringae, P. viridiflava, and P. cichorii strains were very similar (Fig. 3) despite the differences detected by HPLC. P. ficuserectae LMG 5694 produced too few pyoverdins to be analyzed.

FIG. 3.

FIG. 3.

Open in a new tab

IEF patterns obtained from filtered 3-day GASN medium culture supernatants (40 μl) containing unchelated pyoverdins detected under UV light (wavelength, 360 nm) (A) and Fe(III)-chelated pyoverdins detected by their natural color (B). (A) Lane 1, P. asplenii LMG 2137; lane 2, P. putida LMG 2257; lane 3, P. cichorii LMG 2162; lane 4, P. cichorii LMG 8401; lane 5, P. syringae pv. morsprunorum LMG 2222. (B) Lane 1, P. syringae pv. theae LMG 5092; lane 2, P. cichorii LMG 2162; lane 3, P. syringae pv. tabaci LMG 5393; lane 4, P. syringae pv. savastanoi LMG 2209; lane 5, P. syringae pv. syringae B301D; lane 6, P. syringae pv. coronafaciens LMG 5060.

Amino acid and MS analyses of the purified pyoverdins.

The amino acid compositions and the MS and MS-MS data for the purified principal Fe(III)-chelated pyoverdins of P. cichorii LMG 2162, LMG 5034, and LMG 8401 were identical. The atypical pyoverdin (Pa B) contained one glycine, one serine, two threonines, one lysine, and two β-OH-Asp; no ornithine was detected after HI hydrolysis. The molecular mass of the Fe(III)-chelated molecule as determined by ESI was 1,145 atomic mass units (AMU), which was 30 AMU less than the molecular mass of the Fe(III)-chelated pyoverdin of P. syringae (Pa A) (8). The Pa B pyoverdin of P. cichorii apparently differed from the Pa A pyoverdin of P. syringae only by replacement of one serine by one glycine, which is 30 AMU smaller (Table 2). The amino acid compositions and the MS and MS-MS data for the purified principal pyoverdins of P. fuscovaginae UPB 264b and P. asplenii LMG 2137 were identical and were clearly different from those of the Fe(III)-chelated pyoverdin of P. cichorii. The peptide composition of this atypical pyoverdin (Pa C) was partially determined (Table 2). Two β-OH-Asp residues and no ornithine were found after the HCl and HI hydrolyses. Other amino acids detected were threonine, glycine, alanine, lysine, and an undetermined unusual amino acid. The serine that was found in the Pa B pyoverdin of P. cichorii was absent in Pa C, and alanine and the undetermined amino acid that were present in Pa C were absent in Pa B. The molecular mass of the Fe(III)-chelated Pa C as determined by ESI was 1,368 AMU, which was 223 AMU more than the molecular mass of the Pa B pyoverdin of P. cichorii. The Fe(III)-chelated pyoverdin of strain UPB 1023 had a molecular mass as determined by ESI of only 1,059 AMU, which was 309 AMU less than the molecular mass of the pyoverdin of P. fuscovaginae UPB 264b and was rather low for an Fe(III)-chelated pyoverdin. One β-OH-Asp residue and one ornithine were found after the HCl and HI hydrolyses. The results were in total agreement with the results of the HPLC analyses but not with the results of the IEF analyses, which did not distinguish between pyoverdins Pa A and Pa B (Fig. 3). P. cichorii comprised one siderovar, while P. asplenii and P. fuscovaginae comprised another (Table 1).

TABLE 2.

Atypical pyoverdins encountered in this study

Pyoverdin Amino acid composition Mol wt of Fe(III)-chelated pyoverdin (AMU) RT (min)a
Pa A 2 β-OH-Asp, 1 Lys, 2 Thr, 2 Ser 1,175 18.78
Pa B 2 β-OH-Asp, 1 Lys, 2 Thr, 1 Ser, 1 Gly 1,145 19.07
Pa C 2 β-OH-Asp, 1 Lys, 2 Thr, 1 Ala, 2 Gly, undetermined 1,368 28.04
Open in a new tab
a

RT with HPLC program 2. These RT are representative, and the values could vary slightly from one sample set to another and with the quantity of pyoverdin injected.

Spectral analyses of the purified pyoverdin of P. cichorii.

The spectral characteristics of the Fe(III)-chelated pyoverdin of P. cichorii LMG 8401 were identical to the spectral characteristics of the Fe(III)-chelated pyoverdin of P. syringae (8) (Fig. 4). The molecule looked like a typical Fe(III)-chelated pyoverdin (4) at pH 3.0, with a maximum at 399.5 nm and two broad charge transfer bands at about 470 and 550 nm (Fig. 4A). It looked like an atypical Fe(III)-chelated pyoverdin (8) at pH 5.5, 6.0, and 7.0, with observed maxima near 408 nm and none of the broad charge transfer bands (Fig. 4). The maxima were near 405.5 nm at pH 5.0, near 404 nm at pH 4.5, near 401 nm at pH 4.0, and near 400 nm at pH 3.5, and the second charge transfer band around 470 nm was only very weakly apparent starting at pH 4.0 (Fig. 4B). Isosbestic points were detected between the spectra at different pH values (Fig. 4B). The modifications could be observed visually by a strong darkening of the solutions at lower pH values. This was observed with P. cichorii strains in the visual test for determining the pyoverdins produced.

FIG. 4.

FIG. 4.

Open in a new tab

Absorption spectra of the purified Fe(III)-chelated pyoverdin of P. cichorii LMG 8401. (A) Absorption spectra at pH 7.0 (dark line) and pH 3.0 (light line). (B) Absorption spectra at pH 5.5 (lower ɛ at 640 nm), 5.0, 4.5, 4.0, and 3.5 (higher ɛ at 640 nm).

Spectral analyses of the purified pyoverdins of P. asplenii and P. fuscovaginae.

Very weak modifications of the spectral characteristics of the Fe(III)-chelated pyoverdin of P. asplenii LMG 2137 were observed starting at pH 4.5, and isosbestic points were detected between spectra from pH 4.5 to 3.0 (Fig. 5). The modifications were not identical to those observed with the pyoverdin of P. cichorii (Fig. 4B and 5B). A rather stable maximum near 407 nm was observed from pH 7 to 4.5 for the pyoverdin of P. asplenii. It moved slightly to 406 nm at pH 4.0 and to 405.5 nm at pH 3.5, and then it moved more abruptly to 402.5 nm at pH 3.0. The spectral characteristics at pH 3.0 were somewhat comparable to the spectral characteristics of the pyoverdin of P. cichorii near pH 4.5 to 4.0, with only one apparent broad charge transfer band observed at about 550 nm and a maximum between 404 and 401 nm (Fig. 4 and 5). The spectral characteristics of the Fe(III)-chelated pyoverdin of P. fuscovaginae UPB 264b at pH 7.0 and 3.0 were identical (data not shown). The modifications could be observed visually by a very weak darkening of the solutions at lower pH values. This probably explained the weak darkening of the culture medium observed in the visual test for determining the pyoverdins produced. Figure 6 shows the spectral characteristics of the Fe(III)-chelated typical pyoverdins of P. fuscovaginae UPB 1023 and P. fluorescens LMG 1794. Surprisingly for a typical pyoverdin, the spectral characteristics of the pyoverdin of P. fuscovaginae UPB 1023 varied somewhat with the pH, perhaps in relation to the low molecular mass of this typical pyoverdin.

FIG. 5.

FIG. 5.

Open in a new tab

Absorption spectra of the purified Fe(III)-chelated pyoverdin of P. asplenii LMG 2137. (A) Absorption spectra at pH 7.0 (dark line) and pH 3.0 (light line). (B) Absorption spectra at pH 4.5 (lower ɛ at 640 nm), 4.0, 3.5, and 3.0 (higher ɛ at 640 nm).

FIG. 6.

FIG. 6.

Open in a new tab

Absorption spectra of the purified Fe(III)-chelated pyoverdins of P. fuscovaginae UPB 1023 (A) and P. fluorescens LMG 1794 (B) at pH 7.0 (dark lines) and pH 3.0 (light lines).

Comparison of the atypical pyoverdins produced by pathovars of P. syringae and related species by MS and HPLC-MS.

The 45 fluorescent strains tested belonging to 38 pathovars of P. syringae, including P. syringae pv. coronafaciens, P. syringae pv. savastanoi, P. syringae pv. phaseolicola, and P. syringae pv. glycinea, and to the related species P. viridiflava and P. ficuserectae produced the same atypical pyoverdin, Pa A, whose Fe(III)-chelated form has a molecular mass of 1,175 AMU (Table 1). For only five strains, GASN culture medium was replaced by PRO-M or King's B culture medium to produce pyoverdin (Table 1). The results obtained were in total agreement with the results of the HPLC analyses, and they validated this comparison procedure.

Use of detection of pyoverdin production in a field study.

Detection of production of atypical pyoverdin Pa A confirmed the identities of 87 fluorescent strains of P. syringae pv. syringae and 11 fluorescent strains of P. syringae pv. morsprunorum. The analyses also confirmed that 31 nonfluorescent strains of P. syringae pv. morsprunorum apparently do not produce pyoverdin. The 41 strains that do not produce toxin were fluorescent. They were confirmed to be members of the siderovar of P. syringae on the basis of detection by HPLC of atypical pyoverdin Pa A. The spectrophotometric tests detected the atypical pyoverdins lacking the two broad charge transfer bands of typical pyoverdins; the maxima of the culture media were observed between 403 and 408 nm.

Detection of toxic lipodepsipeptide production by strains of P. fuscovaginae and P. asplenii.

Inhibition zones ranging from 11.50 to 24.37 mm wide were observed for all the strains of P. fuscovaginae and P. asplenii listed in Table 1 except strain UPB 1023 from the People's Republic of China, which induced no inhibition zones. The latter strain was the only strain that produced a typical pyoverdin. This confirmed that this strain does not belong to P. fuscovaginae. P. fuscovaginae UPB 407 from Burundi produced 20.75-mm inhibition zones despite the fact that it did not produce pyoverdin. The tests confirmed the relatedness of P. fuscovaginae and P. asplenii.

DISCUSSION

The HPLC method of analyzing pyoverdin production presented in this paper proved to be a very powerful method for rapidly determining the diversity of the pyoverdins produced by different strains. It was superior to the generally used IEF technique (1, 19, 22, 27, 28) in three ways: it enabled the detection of lower pyoverdin production, such as that in P. ficuserectae; it enabled the distinction of typical pyoverdins from atypical pyoverdins (Fig. 1); and it was more reliable. Indeed, the HPLC technique allowed, for the first time, detection of the production by related species (P. syringae and P. cichorii) of atypical pyoverdins that differ by only one amino acid (Fig. 2). In contrast, both IEF tests performed on unchelated or on Fe(III)-chelated pyoverdins were unable to differentiate these closely related pyoverdins (Fig. 3). Fe(III) incorporation experiments are often carried out to confirm the results of IEF analyses (1, 19, 27, 28). However, the same membrane receptor can incorporate different pyoverdins if they have some similarities (2, 31, 33, 39), and bacteria can possess additional membrane receptors to incorporate heterogeneous pyoverdins (11, 23). With P. syringae and P. cichorii, growth stimulation tests indicated that the two similar pyoverdins could be used by the two species (Bultreys and Gheysen, unpublished). The HPLC technique of analyzing the culture medium can be optimized by choosing the correct HPLC column and HPLC program, and it could probably be adapted to analysis of any pyoverdin. It seems probable that comparable small differences between pyoverdins will still be found by using this method or a more complex HPLC-MS method for analyzing prepurified pyoverdins (20), even perhaps in species that are considered now to contain one siderovar.

The HPLC method of analyzing pyoverdin production in the culture medium showed greater diagnostic potential than the visual, spectrophotometric, and IEF-based methods previously described (8), the specificities of which were further determined in this study. The usefulness of the HPLC tests was demonstrated in the analyses of the non-toxin-producing fluorescent Belgian field isolates that could be directly affiliated with the siderovar of P. syringae. Also, our further characterization of strain UPB 1023, which was based on both pyoverdin (Table 1) and toxic lipodepsipeptide production tests, confirmed that this strain does not belong to P. fuscovaginae.

The 41 pathovars of P. syringae listed in Table 1 include representatives of genospecies 1, 2, 3, 4, 6, 7, 8, and 9 that were defined in P. syringae and related species, as well as related representatives of genospecies 2 and 6 (16). Genospecies 5 contains only P. syringae pv. tremae, which is known to be an organism that does not produce pyoverdin (16). The conclusions of this study confirm the results of previous studies that established that one dominant pyoverdin-producing siderovar is present in P. syringae and P. viridiflava (7, 8, 19), apart from a related dihydropyoverdin-producing siderovar (8). Additional information is that the related species P. ficuserectae belongs to the same pyoverdin-producing siderovar, as do P. syringae pv. coronafaciens, P. syringae pv. glycinea, P. syringae pv. phaseolicola, and P. syringae pv. savastanoi, which have been proposed to be elevated to species level or included as pathovars in new species in recent years (12, 15, 16). The pathotype strain of P. syringae pv. cannabina, which was the sole pathovar classified in genospecies 9 (16), was one of the non-pyoverdin-producing strains encountered in this study. Nonfluorescent strains were, however, also encountered in P. syringae pv. morsprunorum and P. syringae pv. garcae and in P. fuscovaginae, although other representatives of these pathovars and species produced the expected pyoverdins. Also, P. syringae pv. persicae and P. syringae pv. sesami were nonfluorescent. This study, therefore, provided no phenotypic data that could help differentiate one of the genospecies from P. syringae.

The arginine dihydrolase-negative fluorescent Pseudomonas species P. syringae, P. viridiflava, and P. cichorii (29) and the related species P. ficuserectae produced atypical pyoverdins (Table 1). In contrast, the arginine dihydrolase-positive fluorescent species P. fluorescens, P. putida, and P. chlororaphis (29) and the related species P. marginalis and P. agarici produced typical pyoverdins (Table 1). The differences are related to the amino acids involved in iron chelation: there are two β-OH-Asp residues in atypical pyoverdins, as shown in this study, and at least one ornithine derivative in typical pyoverdins (14). Moreover, the results of this study showed that the oxidase-negative species P. syringae, P. viridiflava, and P. ficuserectae can be differentiated by the rest of the peptide chain from the related species P. cichorii, which is oxidase positive. The pyoverdin-based classification is thus in total agreement with the existing classification (29) and with the recent taxonomic studies. It seems, therefore, that the peptide chain and particularly the amino acids involved in iron chelation could be useful markers in phylogenic studies. The iron binding ligands are synthesized on the peptide synthetase from only Orn or Asp by specific enzymatic processes (38, 40), and this could explain the relative conservation of these amino acids. This conservation is also apparent in the arginine dihydrolase-positive fluorescent Pseudomonas strains (14). The four typical pyoverdins of P. aeruginosa contain only Orn-based ligands. In 19 typical pyoverdins of P. fluorescens, both pyoverdins containing only Orn-based ligands and pyoverdins containing one Orn-based ligand and one Asp-based ligand were present. In 13 typical pyoverdins of P. putida, one Orn-based ligand and one Asp-based ligand were always present.

The cases of P. asplenii and P. fuscovaginae are noteworthy. Recent studies indicated that these organisms are very similar and that they are relatively closely related to species belonging to the taxonomic group containing the arginine dihydrolase-positive fluorescent Pseudomonas strains (17, 35, 36, 41). The analyses of the production of pyoverdin (Table 1) and toxic lipodepsipeptides confirmed the relatedness of these two species. However, it was shown in this study that these species have only Asp-based ligands in their pyoverdins, which indicates that there is surprising relatedness to species such as P. syringae and P. cichorii. However, this is not so surprising since P. fuscovaginae strains produce the toxic lipodepsipeptide syringotoxin, which is also produced by P. syringae pv. syringae strains from Citrus (13). Also, it was shown in this study that the rest of the peptide chain of the pyoverdins of P. asplenii and P. fuscovaginae differs from the peptide chains in the pyoverdins of P. syringae and P. cichorii, and this indicates that there is a greater distance between these species. Again, it is a sign of faster evolution of the amino acids that are not involved in iron chelation in pyoverdins. These amino acids can be more easily replaced by diverse amino acids since one modification in a pyoverdin implies that there is only one modification in one activation domain of one of the enzymatic modules of the peptide synthetase. The results of the amino acid and MS analyses strongly suggest that such a modification probably occurred in the pyoverdins of P. syringae and P. cichorii, with replacement of a serine by a glycine or vice versa. Differences of one amino acid between typical pyoverdins were observed in P. aeruginosa and in P. fluorescens (31, 39). In both cases, the pyoverdins differed essentially in terms of the presence of one additional amino acid in one of the molecules. These small differences probably reflect the ways in which pyoverdins evolve. Indeed, given the specificity for a membrane receptor that always has to recognize its pyoverdin, it seems probable that the pyoverdins evolve predominantly by replacement, loss, or acquisition of one amino acid. As shown with structurally related typical pyoverdins (2, 31, 33, 39), the membrane receptor could probably still incorporate a slightly modified pyoverdin.

The pyoverdin diversity in P. fluorescens or in P. putida is far greater (14) than the pyoverdin diversity observed in this study among P. syringae, P. viridiflava, P. ficuserectae, and P. cichorii. The genetic heterogeneity of P. fluorescens and P. putida could be an explanation (9). Another explanation has to do with the spectral analyses. The spectral characteristics of the Fe(III)-chelated atypical pyoverdins are influenced more by the composition of the peptide chains (Fig. 4 and 5) than the spectral characteristics of the Fe(III)-chelated typical pyoverdins are (Fig. 6) (4). As the spectral characteristics in the visible spectrum reflect the interactions between the iron and the chromophore, the influence of the peptide chain on the way that iron is bound to the chromophore is more important in atypical pyoverdins than in typical pyoverdins. This could restrict the possibility of modifications in the peptide chains of atypical pyoverdins. At a time when it is increasingly difficult to find phenotypic characteristics to corroborate the results of complex taxonomic and phylogenic studies (42), the two evolutionary pathways apparent in pyoverdins could help workers classify and identify the fluorescent Pseudomonas species.

Acknowledgments

We thank R. Rozenberg for his expert handling of the mass spectrometer. We are grateful to B. de Ryckel for the use of a Shimadzu spectrophotometer, and we thank the Centre de Recherches Agronomiques de Gembloux for its support of this project.

This work was supported by the Ministère des Classes Moyennes et de l'Agriculture de Belgique, by the Unité de Chimie Biologique et Industrielle of the Faculté Universitaire des Sciences Agronomiques de Gembloux for the amino acid analysis, and by the Belgian National Fund for Scientific Research (FNRS) for the mass spectrometry.

REFERENCES

  • 1.Achouak, W., L. Sutra, T. Heulin, J.-M. Meyer, N. Fromin, S. Degreave, R. Christen, and L. Gardan. 2000. Description of Pseudomonas brassicacearum sp. nov. and Pseudomonas thivervalensis sp. nov., root-associated bacteria isolated from Arabidopsis thaliana and Brassica napus. Int. J. Syst. E vol. Microbiol. 50:9-18. [DOI] [PubMed] [Google Scholar]
  • 2.Barelmann, I., K. Taraz, H. Budzikiewicz, V. Geoffroy, and J.-M. Meyer. 2002. The structures of the pyoverdins from two Pseudomonas fluorescens strains accepted mutually by their respective producers. Z. Naturforsch. Sect. C 57:9-16. [DOI] [PubMed] [Google Scholar]
  • 3.Bereswill, S., P. Bugert, B. Völksch, M. Ullrich, C. L. Bender, and K. Geider. 1994. Identification and relatedness of coronatine-producing Pseudomonas syringae pathovars by PCR analysis and sequence determination of the amplification products. Appl. Environ. Microbiol. 60:2924-2930. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Budzikiewicz, H. 1993. Secondary metabolites from fluorescent pseudomonads. FEMS Microbiol. Rev. 104:209-228. [DOI] [PubMed] [Google Scholar]
  • 5.Budzikiewicz, H. 1997. Siderophores of fluorescent pseudomonads. Z. Naturforsch. Sect. C 52:713-720. [PubMed] [Google Scholar]
  • 6.Bultreys, A., and I. Gheysen. 1999. Biological and molecular detection of toxic lipodepsipeptide-producing Pseudomonas syringae strains and PCR identification in plants. Appl. Environ. Microbiol. 65:1904-1909. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Bultreys, A., and I. Gheysen. 2000. Production and comparison of peptide siderophores from strains of distantly related pathovars of Pseudomonas syringae and Pseudomonas viridiflava LMG 2352. Appl. Environ. Microbiol. 66:325-331. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Bultreys, A., I. Gheysen, H. Maraite, and E. de Hoffmann. 2001. Characterization of fluorescent and nonfluorescent peptide siderophores produced by Pseudomonas syringae strains and their potential use in strain identification. Appl. Environ. Microbiol. 67:1718-1727. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Christensen, H., M. Boye, L. K. Poulsen, and O. F. Rasmussen. 1994. Analysis of fluorescent pseudomonads based on 23S ribosomal DNA sequences. Appl. Environ. Microbiol. 60:2196-2199. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Cody, Y. S., and D. C. Gross. 1987. Characterization of pyoverdinpss, the fluorescent siderophore produced by Pseudomonas syringae pv. syringae. Appl. Environ. Microbiol. 53:928-934. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Crosa, J. H. 1997. Signal transduction and transcriptional and posttranscriptional control of iron-regulated genes in bacteria. Microbiol. Mol. Biol. Rev. 61:319-336. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.De Vos, P., M. Goor, M. Gillis, and J. De Ley. 1985. Ribosomal ribonucleic acid cistron similarities of phytopathogenic Pseudomonas species. Int. J. Syst. Bacteriol. 35:169-184. [Google Scholar]
  • 13.Flamand, M. C., S. Pelsser, E. Ewbank, and H. Maraite. 1996. Production of syringotoxin and other bioactive peptides by Pseudomonas fuscovaginae. Physiol. Mol. Plant Pathol. 48:217-231. [Google Scholar]
  • 14.Fuchs, R., and H. Budzikiewicz. 2001. Structural studies of pyoverdins by mass spectrometry. Curr. Org. Chem. 5:265-288. [Google Scholar]
  • 15.Gardan, L., C. Bollet, M. Abu Ghorrah, F. Grimont, and P. A. D. Grimont. 1992. DNA relatedness among the pathovar strains of Pseudomonas syringae subsp. savastanoi Janse (1982) and proposal of Pseudomonas savastanoi sp. nov. Int. J. Syst. Bacteriol. 42:606-612. [Google Scholar]
  • 16.Gardan, L., H. Shafik, S. Belouin, R. Broch, F. Grimont, and P. A. D. Grimont. 1999. DNA relatedness among the pathovars of Pseudomonas syringae and description of Pseudomonas tremae sp. nov. and Pseudomonas cannabina sp. nov. (ex Sutic and Dowson 1959). Int. J. Syst. Bacteriol. 49:469-478. [DOI] [PubMed] [Google Scholar]
  • 17.Grimont, P. A. D., M. Vancanneyt, M. Lefevre, K. Vandemeulebroecke, L. Vauterin, R. Brosch, K. Kersters, and F. Grimont. 1996. Ability of biolog and biotype-100 systems to reveal the taxonomic diversity of the pseudomonads. Syst. Appl. Microbiol. 19:510-527. [Google Scholar]
  • 18.Janse, J. D., J. H. J. Derks, B. E. Spit, and W. R. Van Der Tuin. 1992. Classification of fluorescent soft rot Pseudomonas bacteria, including P. marginalis strains, using whole cell fatty acid analysis. Syst. Appl. Microbiol. 15:538-553. [Google Scholar]
  • 19.Jülich, M., K. Taraz, H. Budzikiewicz, V. Geoffroy, J.-M. Meyer, and L. Gardan. 2001. The structure of the pyoverdin isolated from various Pseudomonas syringae pathovars. Z. Naturforsch. Sect. C 56:687-694. [DOI] [PubMed] [Google Scholar]
  • 20.Kilz, S., C. Lenz, R. Fuchs, and H. Budzikiewicz. 1999. A fast screening method for the identification of siderophores from fluorescent Pseudomonas spp. by liquid chromatography/electrospray mass spectrometry. J. Mass Spectrom. 34:281-290. [DOI] [PubMed] [Google Scholar]
  • 21.King, E. O., M. K. Ward, and D. E. Raney. 1954. Two simple media for the demonstration of pyocyanin and fluorescein. J. Lab. Clin. Med. 44:301-307. [PubMed] [Google Scholar]
  • 22.Koedam, N., E. Wittouck, A. Gaballa, A. Gillis, M. Höfte, and P. Cornelis. 1994. Detection and differentiation of microbial siderophores by isoelectric focusing and chrome azurol S overlay. BioMetals 7:287-291. [DOI] [PubMed] [Google Scholar]
  • 23.Leoni, L., N. Orsi, V. de Lorenzo, and P. Visca. 2000. Functional analysis of PvdS, an iron starvation sigma factor of Pseudomonas aeruginosa. J. Bacteriol. 182:1481-1491. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Luisetti, J., J.-P. Prunier, and L. Gardan. 1972. Un milieu pour la mise en évidence de la production d'un pigment fluorescent par Pseudomonas morsprunorum f. sp. Persicae. Ann. Phytopathol. 4:295-296. [Google Scholar]
  • 25.Merriman, T. R., M. E. Merriman, and I. L. Lamont. 1995. Nucleotide sequence of pvdD, a pyoverdine biosynthetic gene from Pseudomonas aeruginosa: PvdD has similarity to peptide synthetases. J. Bacteriol. 177:252-258. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Meyer, J. M. 2000. Pyoverdines: pigments, siderophores and potential taxonomic markers of fluorescent Pseudomonas species. Arch. Microbiol. 174:135-142. [DOI] [PubMed] [Google Scholar]
  • 27.Meyer, J. M., A. Stintzi, D. De Vos, P. Cornelis, R. Tappe, K. Taraz, and H. Budzikiewicz. 1997. Use of siderophores to type pseudomonads, the three Pseudomonas aeruginosa pyoverdin systems. Microbiology 143:35-43. [DOI] [PubMed] [Google Scholar]
  • 28.Munsch, P., V. A. Geoffroy, T. Alatossava, and J.-M. Meyer. 2000. Application of siderotyping for characterization of Pseudomonas tolaasii and “Pseudomonas reactans” isolates associated with brown blotch disease of cultivated mushrooms. Appl. Environ. Microbiol. 66:4834-4841. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Palleroni, N. J. 1984. Family I. Pseudomonadaceae, p. 141-199. In N. R. Krieg and J. G. Holt (ed.), Bergey's manual of systematic bacteriology, vol. 1. Williams & Wilkins Co., Baltimore, Md.
  • 30.Rombel, I. T., and I. L. Lamont. 1992. DNA homology between siderophore genes from fluorescent pseudomonads. J. Gen. Microbiol. 138:181-187. [DOI] [PubMed] [Google Scholar]
  • 31.Ruangviriyachai, C., D. Fernndez, R. Fuchs, J.-M. Meyer, and H. Budzikiewicz. 2001. A new pyoverdin from Pseudomonas aeruginosa R′. Z. Naturforsch. Sect. C 56:933-938. [DOI] [PubMed] [Google Scholar]
  • 32.Schaad, N. W., A. K. Vidaver, G. H. Lacy, K. Rudolph, and J. B. Jones. 2000. Evaluation of proposed amended names of several pseudomomads and xanthomonads and recommendations. Phytopathology 90:208-213. [DOI] [PubMed] [Google Scholar]
  • 33.Schlegel, K., R. Fuchs, M. Schäfer, K. Taraz, H. Budzikiewicz, V. Geoffroy, and J.-M. Meyer. 2001. The pyoverdins of Pseudomonas sp. 96-312 and 96-318. Z. Naturforsch. Sect. C 56:680-686. [DOI] [PubMed] [Google Scholar]
  • 34.Sneath, P., M. Stevens, and M. J. Sackin. 1981. Numerical taxonomy of Pseudomonas based on published results substrate utilization. Antonie Leeuwenhoek 47:423-448. [DOI] [PubMed] [Google Scholar]
  • 35.Vancanneyt, M., S. Witt, W. R. Abraham, K. Kersters, and H. L. Fredrickson. 1996. Fatty acid content in whole-cell hydrolysates and phospholipid fractions of pseudomonads: a taxonomic evaluation. Syst. Appl. Microbiol. 19:528-540. [Google Scholar]
  • 36.Vancanneyt, M., U. Torck, D. Dewettinck, M. Vaerewijck, and K. Kersters. 1996. Grouping of pseudomonads by SDS-PAGE of whole-cell proteins. Syst. Appl. Microbiol. 19:556-568. [Google Scholar]
  • 37.Van Zyl, E., and P. L. Steyn. 1990. Differentiation of phytopathogenic Pseudomonas and Xanthomonas species and pathovars by numerical taxonomy and protein gel electrophoregrams. Syst. Appl. Microbiol. 13:60-71. [Google Scholar]
  • 38.Visca, P., A. Ciervo, and N. Orsi. 1994. Cloning and nucleotide sequence of the pvdA gene encoding the pyoverdin biosynthetic enzyme l-ornithine N5-oxygenase in Pseudomonas aeruginosa. J. Bacteriol. 176:1128-1140. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Weber, M., K. Taraz, H. Budzikiewicz, V. Geoffroy, and J.-M. Meyer. 2000. The structure of a pyoverdine from Pseudomonas sp. CFML 96.188 and its relation to other pyoverdins with a cyclic C-terminus. BioMetals 13:301-309. [DOI] [PubMed] [Google Scholar]
  • 40.Wilson, M. J., B. J. McMorran, and I. L. Lamont. 2001. Analysis of promoters recognized by PvdS, an extracytoplasmic-function sigma factor protein from Pseudomonas aeruginosa. J. Bacteriol. 183:2151-2155. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Yamamoto, S., H. Kasai, D. L. Arnold, R. W. Jackson, A. Vivian, and S Harayama. 2000. Phylogeny of the genus Pseudomonas: intrageneric structure reconstructed from the nucleotide sequences of gyrB and rpoD genes. Microbiology 146:2385-2394. [DOI] [PubMed] [Google Scholar]
  • 42.Young, J. M. 2000. Recent developments in systematics and their implications for plant pathogenic bacteria, p. 135-163. In F. G. Priest and M. Goodfellow (ed.), Applied microbial systematics. Kluwer Academic Publishers, Dordrecht, The Netherlands.
  • 43.Young, J. M., M. Whenua, G. S. Saddler, Y. Takikawa, S. H. DeBoer, L. Vauterin, L. Gardan, R. I. Gvozdyak, and D. E. Stead. 1996. Names of plant pathogenic bacteria 1864-1995. Rev. Plant Pathol. 75:722-763. [Google Scholar]

Articles from Applied and Environmental Microbiology are provided here courtesy of American Society for Microbiology (ASM)

RESOURCES