Genetics of biosynthesis and structure of the capsular exopolysaccharide from the Asian pear pathogen Erwinia pyrifoliae

Won-Sik Kim1, Martin Schollmeyer1, Manfred Nimtz2, Victor Wray2 and Klaus Geider1

Max-Planck-Institut für Zellbiologie, Rosenhof, 68526 Ladenburg, Germany1
GBF, Gesellschaft für Biotechnologische Forschung mbH, Mascheroder Weg 1, 38124 Braunschweig, Germany2

Author for correspondence: Klaus Geider. Tel: +49 6203 106 117. Fax: +49 6203 106 122. e-mail: kgeider{at}zellbio.mpg.de


   ABSTRACT
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Erwinia pyrifoliae is a novel bacterial pathogen, which causes Asian pear blight and is related to Erwinia amylovora, the causative agent of fire blight. E. pyrifoliae produces exopolysaccharide (EPS) related to amylovoran in its sugar composition and sugar linkages. This was shown by degradation of the EPS with a viral depolymerase, and by methylation analysis and ESI/MS. The structure of the repeating units was confirmed by 1H-NMR spectra. The EPS of E. pyrifoliae carried side chains, which were mainly terminated by acetyl and pyruvyl residues as found previously for amylovoran. On the other hand, a second side chain with glucose found for up to 65% of the repeating units of amylovoran was completely absent. The nucleotide sequences of five genes of the cps cluster of E. pyrifoliae encoding proteins for EPS synthesis were characterized and displayed a high homology with the corresponding ams genes. Similar functions of the gene products are assumed. As for ams mutants of E. amylovora, a cpsB mutant of E. pyrifoliae did not synthesize EPS and did not produce ooze on slices of immature pears or symptoms on pear seedlings. The cps mutant was complemented for EPS synthesis and virulence on pear slices with a gene cluster of E. amylovora that included amsB.

Keywords: Asian pear blight, fire blight, amylovoran, sugar linkage, cps genes

Abbreviations: CPC, cetylpyridinium chloride; EPS, exopolysaccharide; HR, hypersensitive reaction

The GenBank accession number for the sequence reported in this paper is AJ300463.


   INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Erwinia pyrifoliae was isolated from diseased Asian pear fruit trees with symptoms similar to fire blight (Rhim et al., 1999 ). The causative pathogen produces mucoid, slightly yellow colonies on semi-selective agar, whereas Erwinia amylovora forms intensely yellow coloured colonies. Biochemical and molecular data including DNA–DNA hybridization and Biotype 100 assays classified E. pyrifoliae as a novel species distinct from E. amylovora and other erwinias (Kim et al., 1999 ). E. amylovora and E. pyrifoliae therefore belong to different species, causing fire blight and Asian pear blight, respectively. Their host range partially overlaps (Kim et al., 2001 ) and includes Nashi and European pears as hosts. A recent survey in the literature for fire blight host plants revealed more than 200 species (Jock et al., 2000 ) where symptoms were observed and E. amylovora was isolated. E. pyrifoliae has only been isolated from Nashi pears (Rhim et al., 1999 ), and artificial inoculations rarely showed symptoms on apples and never on cotoneaster (Kim et al., 2001 ). Strains of E. amylovora which do not infect fruit trees were isolated in North America from raspberry. One reason for host preferences could be a different structure of the capsular exopolysaccharide (EPS). Indeed, raspberry strains were described that lack the glucose residue at the branched galactose (Maes et al., 2001 ). On the other hand, this property is not unique for E. amylovora rubus strains, as at least one strain in our collection synthesized this glucose residue (M. Schollmeyer, M. Nimtz & K. Geider, unpublished).

The production of capsular EPS is a common feature of many Gram-negative bacteria (Sutherland, 1988 ). For plant pathogens, EPS capsules and slime are important for protection of the pathogens against recognition by plant defence mechanisms, to bind water, keeping the bacteria moist, and for retention of nutrients and ions released from the damaged plant cells (Leigh & Coplin, 1992 ). E. amylovora produces the acidic EPS amylovoran, and the gene cluster encoding amylovoran biosynthesis (ams) was cloned and sequenced (Bugert & Geider, 1995 ). Mutation analysis of the synthesis genes showed EPS to be an important virulence factor for the pathogen (Steinberger & Beer, 1988 ; Bellemann & Geider, 1992 ; Bernhard et al., 1993 ). Erwinia stewartii (Pantoea stewartii subsp. stewartii) produces stewartan, and EPS-deficient mutants of E. stewartii failed to cause wilt and water-soaking on corn (Dolph et al., 1988 ). Amylovoran and stewartan consist of highly polymerized repeating units (Jumel et al., 1997 ). The sugar backbone of a repeating unit of amylovoran comprises only galactose residues (Nimtz et al., 1996a ), whereas in stewartan, glucose substitutes a galactose residue of amylovoran (Nimtz et al., 1996b ). As will be discussed in Results, the major side chain of amylovoran consists of a glucuronic acid residue attached to the backbone of galactose and is terminated by a galactose carrying pyruvyl and acetyl residues. The similar side chain of stewartan is terminated by a glucose residue and carries a further glucose residue as a second side chain of the repeating units. For amylovoran, the number of the repeating units carrying this glucose residue is dependent on the growth conditions of the bacteria, as preparations from suspension cultures showed a low incidence (approx. 10%) of the monosaccharide residue (Nimtz et al., 1996a ), whereas more than half of the repeating units are branched in amylovoran from agar-grown cells.

E. pyrifoliae produces EPS in minimal medium, ooze on slices of immature pears and mucoid colonies on MM2 agar similar to E. amylovora (Rhim et al., 1999 ). In this study, we have investigated the various biochemical properties of the EPS, analysed genes for its synthesis and assessed its importance for virulence.


   METHODS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Bacterial strains and plasmids.
These are listed in Table 1. Wild-type strains were isolated from diseased trees in Korean Nashi pear orchards in the area of Chuncheon (Rhim et al., 1999 ).


View this table:
[in this window]
[in a new window]
 
Table 1. Bacterial strains and plasmids used in the assays

 
EPS isolation, determination of its concentration and degradation.
To determine EPS synthesis of E. pyrifoliae, cultures of Ep1/96 were grown in minimal medium (MM2A) for 2 days at 28 °C. After centrifugation, 50 µl cetylpyridinium chloride (CPC; 50 mg ml-1) was added to 1 ml supernatant diluted in water, and the EPS concentration was determined by measuring turbidity at 600 nm (Bellemann et al., 1994 ).

For isolation, EPS was prepared from E. pyrifoliae strain Ep1/96 grown on a cellophane disk, which was placed on MM2A agar (Kim & Geider, 2000 ), for 3 days at 28 °C. The slimy cells were suspended in 3 ml water, vortexed and centrifuged. The supernatant was freeze-dried for storage of the EPS. Amylovoran from strain Ea1/79 was prepared by the same method.

For degradation, EPS (100 µg ml-1) in 100 mM NaCl/10 mM sodium acetate (pH 4·7) was incubated with 175 ng EPS depolymerase from E. amylovora phage {phi}Ea1h (Kim & Geider, 2000 ) per ml assay mixture at 28 °C. The remaining high-molecular-mass EPS fraction was measured by the CPC method (Bellemann et al., 1994 ) every 30 min.

Lectin staining of capsular EPS.
Ep1/96 or mutants were grown for 2 days on MM2A agar and cells from a colony were transferred onto a glass slide. They were suspended in 5 µl FITC-labelled lectin from Abrus precatorius (Sigma), which specifically binds to the galactose residue of the amylovoran side chain. The cells were visualized in the fluorescence microscope Zeiss Axiovert 405 in bright field and UV with an oil immersion lens at 1000-fold total magnification and the filter combination BP450–490/FT510/LP520 (excitation filter/dichroic/emission filter) as described previously (Bellemann et al., 1994 ).

Structural characterization of the capsular polysaccharides by methylation analysis, ESI-MS and NMR.
The monosaccharide composition of the native polysaccharide was analysed and the linkage arrangement investigated by methylation analysis as described previously (Nimtz et al., 1996a ).

For mass spectra, the repeating units, obtained by enzymic depolymerization of EPS, were dissolved in methanol/water (50/50, v/v) at a concentration of approximately 5 pmol µl-1, and 3 µl of the solution was introduced into gold-coated nanospray glass capillaries. The tip of the capillary was placed orthogonally in front of the entrance hole of a QTof II mass spectrometer (Micromass) equipped with a nanospray ion source, and a voltage of -800 V was applied. For collision-induced dissociation experiments, parent ions were selectively transmitted from the quadrupole mass analyser into the collision cell. Argon was used as the collision gas and the kinetic energy was set at approximately +25 eV. The resulting daughter ions then were separated by an orthogonal time-of-flight mass analyser.

For 1H NMR spectroscopic analysis, the oligosaccharides were repeatedly lyophilized against D2O (Fluka, >99·95 atom% D) at pD 7 and ambient temperature. 1H NMR spectra at 600 MHz were recorded at 300 K on a Bruker AVANCE DMX 600 NMR spectrometer incorporating a gradient unit. A 1·3 s presaturation pulse was employed prior to the experimental pulse sequence in order to suppress the signal of the residual HOD resonance. All 1D and 2D COSY 1H spectra were recorded using standard Bruker software.

Cloning, sequencing and sequence analysis.
Several primer pairs (17–20 oligonucleotides) from E. amylovora used for sequencing of the ams region (Bugert et al., 1995 ) were applied to PCR reactions with E. pyrifoliae (Table 2). PCR products were separated on a 0·8% agarose gel and DNA fragments of the size expected according to the corresponding ams genes were eluted with a QIAEXII gel extraction kit (QIAGEN). For sequencing analysis, the PCR fragments were cloned into the vector pGEM-T (Promega) and sequenced with an automatic sequencer (ALFexpress; Amersham Pharmacia Biotech). The data were analysed with the sequence analysis programs Clone manager v.5/Align Plus 4 (Scientific & Educational Software). Database searches were performed on the internet with the BLASTP+BEAUTY program (Worley et al., 1995 ). The nucleotide sequence data were deposited in the EMBL Nucleotide Sequence Database (accession number AJ300463).


View this table:
[in this window]
[in a new window]
 
Table 2. PCR primers applied for cloning of DNA from cps genes of E. pyrifoliae

 
Mutagenesis of the cps gene in E. pyrifoliae.
The cpsB gene of E. pyrifoliae was mutated by disruption with plasmid pfdB14Z' carrying a 0·7 kb DNA fragment with an internal section of the encoding region. At first, this DNA fragment was amplified by PCR with primers 124/B1 and 125/C2c (Table 2) and cloned into pGEM-T (Promega). Plasmids with insertions were screened after transformation of Escherichia coli DH5{alpha} on agar with Ap and X-Gal for the formation of white colonies. Plasmid pGEM-B1 was digested with restriction enzymes SalI and SphI and the insert transferred into plasmid pfdB14Z', a suicide plasmid in cells without gene 2 protein. After propagation in E. coli JM83-2 with expression of fd gene 2, the plasmid was introduced into E. pyrifoliae strain Ep1/96 by electroporation to disrupt cpsB by a single cross-over in homologous recombination with the plasmid. Cm-resistant colonies were assayed for their inability to synthesize EPS on MM2A agar plates. For genetic complementation, plasmid pEA109 was transferred into the mutant by electroporation and selected with tetracycline.

Pathogenicity assays on immature pear slices or pear seedlings.
Immature pear fruits were harvested in early summer and stored at 4 °C. Pear slices were inoculated with freshly grown bacteria from agar with a toothpick and incubated in an air-sealed Petri dish for at least 6 days at 28 °C before the symptoms were assessed.

In order to grow seedlings from European pears (Pyrus communis cv. ‘Kirchensaller Mostbirne’) and from Asian pears (Pyrus pyrifolia cv. ‘Nashi’), seeds were vernalized for 5 days in ice water at 4 °C and kept in sand at 10 °C for 2–4 weeks. Germinated seedlings were potted and incubated in light for 4–5 weeks at 22 °C. For virulence assays, secondary leaves of the plants were cut at the tip and inoculated with a pipette tip dipped in a suspension of bacteria. Symptoms were monitored after further incubation of the seedlings for 2–3 weeks.


   RESULTS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Analysis of EPS from E. pyrifoliae with a viral depolymerase and by lectin staining
Isolated EPS from E. pyrifoliae strain Ep1/96 was first characterized by degradation with amylovoran depolymerase (amylovoran lyase) from bacteriophage {phi}Ea1h. Purified His-tagged amylovoran depolymerase was applied for cleavage of preparations of EPS from E. pyrifoliae and E. amylovora. The degradation kinetics of the two EPS species showed a similar slope, indicating a high similarity of the sugars at the sites of enzymic cleavage (Fig. 1).



View larger version (14K):
[in this window]
[in a new window]
 
Fig. 1. Degradation experiments with viral EPS depolymerase. EPS from E. amylovora E9 ({bullet}) and E. pyrifoliae Ep1/96 ({circ}) were digested with purified EPS-depolymerase of bacteriophage {phi}Ea1 (Kim & Geider, 2000 ) cleaving in the backbone behind the branched galactose of amylovoran (Nimtz et al., 1996a ).

 
For further biochemical characterization of the sugar structure of E. pyrifoliae EPS, capsulated cells were stained with FITC-labelled lectin from A. precatorius. The lectin recognizes galactose (Tomita et al., 1972 ) and binds to the terminal galactose residue of the side chain of amylovoran (Bellemann et al., 1994 ). At least 90% of the E. pyrifoliae capsules were stained with the FITC-labelled lectin, indicating the termination of the side chain with galactose (Table 3), a similar result to that observed for amylovoran capsules of E. amylovora (Bellemann et al., 1994 ).


View this table:
[in this window]
[in a new window]
 
Table 3. Plant assays and lectin staining of wild-type and EPS-mutant strains of E. amylovora and E. pyrifoliae

 
The chemical structure of EPS from E. pyrifoliae
The carbohydrate linkage arrangement of the intact polysaccharides was determined by methylation analysis. For the polysaccharide produced by E. pyrifoliae, the derivatives characteristic of 3-, 6-, 4,6-di- and 3,4-disubstituted galactose residues were detected and, after an additional reduction step using NaBD4, 4-substituted glucuronic acid could be detected as the respective glucitol derivative modified with two 2 D atoms at C-6, suggesting a structure (Fig. 2) similar to amylovoran (Nimtz et al., 1996a ). However, no trace of a 3,4,6-trisubstituted galactose derivative, indicating a partial O-6 glucosylation of the central galactose residue described for amylovoran, could be detected. This was in marked contrast to the results from an amylovoran preparation, reanalysed as a standard, which produced a clear signal of the 3,4,6-trisubstituted galactose derivative, and gave considerable amounts of a terminal glucose residue, indicating a relatively high incidence of the additional glucose moiety at the central galactose.



View larger version (15K):
[in this window]
[in a new window]
 
Fig. 2. Structure of E. pyrifoliae EPS (A) compared to those of the repeating units of amylovoran (B). *, Residue present in approximately 65% of repeating units of amylovoran. The cells of strains Ep1/96 and Ea1/79 were grown on MM2A minimal agar for isolation of EPS.

 
After enzymic depolymerization of the polysaccharides from cells grown on MM2A agar with cellophane disks, the resulting oligosaccharides were analysed by negative-ion-mode nanospray mass spectrometry as depicted in Fig. 3. Doubly charged molecular ions (M-2H)2- at m/z 455·1, 476·1 and 497·1 in a ratio of approximately 1:15:100 were detected for the E. pyrifoliae repeating unit (Fig. 3B) corresponding to molecular masses of 912·2, 954·2 and 996·2 Da, compatible with the repeating pentasaccharide already described for wild-type amylovoran (Nimtz et al., 1996a ) modified by pyruvate and 0–2 acetyl groups (Fig. 2). The standard amylovoran preparation (Fig. 3A) showed identical signals in a ratio of approximately 1:11:40, suggesting a slightly lower degree of acetylation. Additionally, a weak signal at m/z 526·1, also doubly charged, was detected, which can be explained by the exchange of one acetyl group for a succinyl group. Interestingly, however, the most intense doubly charged signals at m/z 557·2, 578·2 and 607·2 indicate the predominant presence of a repeating unit bearing similar substitution patterns of pyruvyl and acetyl and/or succinyl groups, but carrying an additional hexose residue in most repeating units of amylovoran, confirming the results obtained by methylation analysis. All detected doubly charged molecular ions were subjected to ESI-MS/MS analyses. The resulting daughter ion spectra (not shown) were compatible with the published amylovoran monosaccharide sequence, and confirmed the identity and linkage positions of the acyl residues.



View larger version (19K):
[in this window]
[in a new window]
 
Fig. 3. ESI/MS profiles of EPS from E. amylovora and E. pyrifoliae: negative-ion-mode ESI spectra of the enzymically depolymerized EPSs isolated from E. amylovora (A) and E. pyrifoliae (B). The doubly charged molecular ions at m/z 455·1, 476·1 and 497·1 (M-2H)2-, corresponding to molecular masses of 912·2, 954·2 and 996·2, respectively (Hex4 HexA Pyr Ac0–2) indicate the presence of the pentasaccharide repeating unit modified by one pyruvate residue and 0–2 acetyl groups. In the amylovoran preparation, another series of intense doubly charged signals at m/z 557·2 and 578·2 was detected, corresponding to the repeating unit enlarged by an additional hexose residue (Hex5 HexA Pyr Ac1–2). Further relatively weak signals at m/z 526·2 and 607·2 can be explained by the exchange of one acetyl to a succinyl residue of the penta- or hexasaccharide, respectively (Hex4–-5 HexA Pyr Ac Suc).

 
In order to verify these results, the depolymerized samples were also analysed by 1D and 2D COSY 1H NMR spectroscopy and compared to the NMR data obtained from the original wild-type amylovoran preparation analysed by us (Nimtz et al., 1996a ). The 1D spectrum of the ‘old’ amylovoran wild-type repeating unit shown in Fig. 4(B), whose signals have been assigned previously (Nimtz et al., 1996a ) was very similar to the spectrum of the E. pyrifoliae repeating unit (Fig. 4A). Therefore, the basic structures of both repeating units must be identical. The absence of signals typical of a glucose residue linked to O-6 of the core galactose moiety in the E. pyrifoliae repeating unit clearly demonstrated that this structural motif is not present in the polysaccharide, whereas only 10–20% of the repeating units of amylovoran from Ea1/79 cells grown in suspension culture (Nimtz et al., 1996a ) bore this monosaccharide residue, as was apparent from the signal of its anomeric proton (marked by an arrow in Fig. 4B). Some differences existed in the substituent pattern of the terminal 4,6-pyruvylated galactose moiety of the side branch of the repeating unit between the two polysaccharides: E. pyrifoliae EPS had a larger proportion of acetylated substituents than amylovoran, as is apparent from the signals in the region 5·5 to 5·7 p.p.m. (Fig. 4) of the anomeric protons of the respective monosaccharide unit with two (linked to O-2 and O-3, 5·6–5·7 p.p.m.), one (linked to O-2 or O-3, 5·55–5·6 p.p.m.) and no additional acetyl group (5·51 p.p.m.). Interestingly, in repeating units of amylovoran prepared from Ea1/79 cells grown on minimal agar (Fig. 4C), the presence of approximately 65% of the structure containing the additional O-6-linked glucose residue was observed (Fig. 4C), confirming the results obtained by mass spectrometry. The introduction of this additional monosaccharide unit caused small changes of the chemical shift of the protons of the various monosaccharide residues, resulting in a splitting of their signals in a ratio of approximately 2:1 (Fig. 4C, marked by asterisks).



View larger version (24K):
[in this window]
[in a new window]
 
Fig. 4. Comparison of the low-field region of the 600 MHz 1H 1D NMR spectrum of the capsular EPS from Erwinia pyrifoliae (A) with those from E. amylovora grown under various conditions (B and C, see text). The signals have been assigned previously (Nimtz et al., 1996a ). The signals of H-1 of the additional glucose residue linked to the core {alpha}-galactose unit in E. amylovora are marked with an arrow. The introduction of this unit causes a doubling of various other signals of the repeating unit; these are denoted by an asterisk and correspond to H-1 of the terminal galactose (5·5–5·7 p.p.m.), H-2 of the same unit with two acetate units (5·38 p.p.m.) and one acetate unit at C-2 (5·15 p.p.m.), and H-1 of the core 3,4-galactose unit (5·05 p.p.m. with glucose and 5·03 p.p.m. without glucose).

 
Cloning and sequence analysis of cps genes from E. pyrifoliae
Assuming a high similarity of the EPS-encoding cps genes of E. pyrifoliae to the ams genes of E. amylovora, we applied several primer pairs derived from the ams region in PCR amplification assays (Table 2). PCR products from E. pyrifoliae were obtained with primers derived from amsG, H, A, B, C and E. No PCR signals were obtained with primer pairs derived from amsD and amsF. For sequence analysis, DNA with the cps genes was amplified and cloned into pGEM-T. The scheme for cps sequencing is presented in Fig. 5. Fragments with cpsB (691 bp) and cpsC (710 bp) from E. pyrifoliae were cloned first and the adjacent genes were investigated. To fill the gap between the 3' end of cpsB and 5' end of cpsC, the 1·2 kb PCR product obtained with primers CPS1L and CPS2Rc, which have been designed for specific PCR detection of E. pyrifoliae (Kim et al., 2001 ), was cloned, and the sequence covering the gap was determined (Fig. 5). The complete cpsA gene was cloned with primers derived from the 5' end of cpsB and the 3' region of amsH. The intact cpsG gene was amplified by a combination of primers from the sequence upstream of cpsG and from the amsH 5' region. Based on the nucleotide sequences of cpsG and the end of cpsI, we designed primers to amplify a gap between cpsG and cpsA with genes cpsH, cpsI and cloned a 1·7 kb PCR fragment. The total nucleotide sequence of 7280 bp with the cps genes G, H, I, A, B and part of cpsC was confirmed for both strands.



View larger version (14K):
[in this window]
[in a new window]
 
Fig. 5. Scheme of the early cps region of E. pyrifoliae and its comparison to the corresponding ams genes of E. amylovora. The lines indicate the size and position of the cloned PCR products. The numbers refer to identity of proteins (*, degree of identity of the partial sequence of cpsC with the corresponding region of amsC). x, Site of gene disruption in the cpsB mutant.

 
Five complete ORFs from the start of the cps cluster with cpsG and an incomplete ORF with part of cpsC are oriented in the same direction (Fig. 5). The intergenic space between cpsI and cpsA was identical with the corresponding ams region of E. amylovora (TTGGGA-TTTTCC), including a putative ribosome-binding site (RBS) (underlined). The sizes of cps genes are the same as the corresponding ams genes except cpsH (380 amino acids), which encodes a protein three amino acids larger than AmsH (377 amino acids).

Alignment of the amino acid sequences deduced from cpsG, H, I, A, B, C with the program BLASTP+BEAUTY (Worley et al., 1995 ) showed a high homology for proteins encoded by the corresponding E. amylovora ams genes. CpsG has 95% identity with AmsG. More than 90% identity was also found for CpsH/AmsH, CpsI/AmsI and CpsB/AmsB. The amino acid sequence of the analysed part of CpsC is 94% identical with the corresponding part of AmsC of E. amylovora.

Construction, virulence and complementation of an E. pyrifoliae EPS mutant
To verify that EPS is also a virulence factor of E. pyrifoliae, the chromosomal cpsB gene was mutated by disruption with a 0·7 kb internal DNA fragment from cpsB. After transformation of Ep1/96 with pfdB14Z'-B1, 120 colonies were picked from StI agar with Cm onto MM2A agar and seven non-mucoid colonies were obtained. Cointegrate formation of pfdB14Z'-B1 with cpsB was confirmed by PCR analysis for these mutants (data not shown). One mutant, Ep1/96-mB1, was assayed for EPS production with CPC. The growth rate of the mutant and the wild-type strain Ep1/96 was identical, but no EPS was detected, similar to the E. amylovora EPS mutant Ea1/79-D32, which is deficient in amsB (Fig. 6). In the lectin staining assay, no EPS capsules were observed for Ep1/96-mB1. Consequently, cpsB is required for EPS synthesis by E. pyrifoliae.



View larger version (15K):
[in this window]
[in a new window]
 
Fig. 6. EPS synthesis of E. amylovora and E. pyrifoliae wild-type and mutant strains. The turbidity values in the CPC assay (Bellemann et al., 1994 ) were normalized to a cell density of 1 (OD600). The E. pyrifoliae mutant in cpsB Ep1/96-mB1 was complemented with the ams gene cluster (H–L) of E. amylovora cloned in the low-copy-number plasmid pEA109.

 
The EPS-deficient mutant Ep1/96-mB1 was tested for virulence on seedlings of Asian pear (Pyrus pyrifolia) and European pear (Pyrus communis) and on slices of immature pears for ooze production. No exudate was observed on pear slices nor any symptoms on pear seedlings, in contrast to the wild-type E. pyrifoliae strain Ep1/96 (Fig. 7, Table 3). The induction of a hypersensitive reaction (HR) in tissue of non-host plants is an important virulence factor of phytopathogenic bacteria. The EPS mutant Ep1/96-mB1 showed HR after infiltration of a bacterial suspension (108 c.f.u. ml-1) into tobacco leaves (Table 3). The EPS deficiency, caused by the mutation in cpsB, is thus responsible for the lack of symptom formation.



View larger version (70K):
[in this window]
[in a new window]
 
Fig. 7. Virulence assays of E. pyrifoliae and E. amylovora and their EPS mutants on Nashi pear seedlings.

 
For mutant complementation, plasmid pEA109 (Bernhard et al., 1993 ; Bugert & Geider, 1995 ) carrying an insert of the E. amylovora ams region with genes amsH, I, A, B, C, D, E, F, J, K and L was mobilized into strain Ep1/96-mB. The resulting strain Ep1/96-mB1(pEA109) was mucoid on MM2A agar plates, similar to the wild-type Ep1/96, and produced EPS on MM2A agar and in suspension cultures (Fig. 6), and ooze on slices of immature pears. The isolated EPS was characterized by methylation analysis. The signal for 2-Gal, indicating the branched sugar with the side chain and the 1,6-linked glucose residue (Fig. 2B), was entirely absent. None of the genes on plasmid pEA109 apparently encodes a sugar transferase for attachment of the glucose side chain. On the other hand, the complementation of the mutant Ep1/96-mB1 by plasmid pEA109 with amsB confirmed that the genes amsB and cpsB are equivalent in their function.


   DISCUSSION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
EPS is thought to modulate pathogen recognition by plant defence reactions (Leigh & Coplin, 1992 ). Amylovoran-deficient mutants of E. amylovora are non-pathogenic (Bellemann & Geider, 1992 ; Bernhard et al., 1993 ; Bugert & Geider, 1995 ) and levan-deficient mutants were found retarded in symptom formation on apple shoots (Geier & Geider, 1993 ). For a fast formation of an EPS shield, many pathogens secrete levansucrase, and the formation and regulation of this enzyme has been extensively investigated for E. amylovora (Geier & Geider, 1993 ; Zhang & Geider, 1997 ; Du & Geider, 2002 ). E. pyrifoliae does not synthesize levan (Rhim et al., 1999 ), which might add to the host specificity. On the other hand, many E. amylovora strains with low levan synthesis were isolated from fire blight host plants (Bereswill et al., 1997 ). The Asian pear pathogen thus only relies on its capsular EPS, and structural differences to amylovoran could contribute to a different host range. Similarity of EPS from E. amylovora and E. pyrifoliae was shown by degradation with viral EPS depolymerase (Kim & Geider, 2000 ), which cleaves amylovoran between two galactose residues at their ß-1->3 linkage.

For EPS of E. pyrifoliae, the molecular mass of the cleavage products showed the complete absence of a glucose residue linked to O-6 of the central galactose residue as found for repeating units of amylovoran. The data with ESI/MS were confirmed by methylation analysis and 1D and 2D 1H NMR spectroscopy. A higher proportion of acetylation at the terminal pyruvylated galactose of the side branch of E. pyrifoliae was found as a minor difference compared to the E. amylovora EPS originally analysed by us (Nimtz et al., 1996a ). Interestingly, in the amylovoran repeating unit prepared from cells grown on minimal agar, there was a higher degree of acetylation than in the case of amylovoran from suspension cells and additionally about 10% of the acetyl groups were found to be exchanged for succinyl residues, which were not detected in EPS from E. pyrifoliae nor from E. amylovora grown as suspension cells. However, all the acyl substituents were found to be linked exclusively to the terminal galactose residue of the sugar side-chain. Moreover, whereas in EPS preparations from suspension cells only 10% of an additional glucose residue linked to O-6 of the core galactose of the repeating unit was found (Nimtz et al., 1996a ), in the amylovoran preparation from agar-grown cells, approximately 65% of the repeating unit carried this moiety (Figs 2, 3 and 4). Therefore, the occurrence of this additional glucose residue and the nature of the acyl substituents in the sugar side-chain are apparently also dependent on the culture conditions used for the growth of the bacteria.

The high degree of similarity of the chemical structures of the two EPS species may not justify naming E. pyrifoliae EPS ‘pyrifolan’ by analogy to amylovoran. Their relatedness was also evident from the homology between ams and cps genes encoding proteins for biosynthesis of EPS. The corresponding genes were more than 90% homologous to each other. CpsG is assumed to transfer UDP-galactose to the lipid carrier (C. Langlotz & K. Geider, unpublished). CpsA has 92% identity with AmsA, which is a tyrosine kinase associated with production of EPS and virulence of E. amylovora (Ilan et al., 1999 ). CpsB is assumed to be a glycosyl transferase, which transfers galactose from UDP-galactose to galactose attached to the lipid-carrier. AmsI is an acid phosphatase (Bugert & Geider, 1997 ). AmsH may be involved in transport of the repeating unit (Bugert & Geider, 1995 ) (Table 4, Figs 2 and 5). The cps mutant Ep1/96mB1 was successfully complemented with the plasmid pEA109 carrying the intact amsB gene. The gene products CpsB and AmsB are therefore exchangeable for synthesis of the repeating units. On the other hand, the glucose residue at the branched galactose was not observed in the repeating units obtained from the pEA109-complemented mutant (M. Schollmeyer, M. Nimtz & K. Geider, unpublished). It can be concluded that the ams genes on pEA109 do not encode for the responsible sugar transferase. In addition to some differences in the acetylation pattern and substitution by succinyl groups, this glucose residue is a major difference for EPS of E. pyrifoliae and amylovoran of E. amylovora. Amylovoran produced in E. stewartii cps mutants carrying pEA109 is not acetylated, in contrast to stewartan synthesized in E. amylovora ams mutants with pES2144, a cosmid with the cps genes of Pantoea (Erwinia) stewartii (Bernhard et al., 1996 ). Genes for acetylation should reside apart from the ams region inserted in plasmid pEA109.


View this table:
[in this window]
[in a new window]
 
Table 4. Characteristic features of ORFs and predicted proteins from cps genes of E. pyrifoliae

 
The restricted host range of E. pyrifoliae, presently confined to Nashi pears and European pears (Kim et al., 2001 ) could be influenced by properties of the two major pathogenicity factors, associated with HR- and EPS-encoding genes. The elicitor HrpN differs between E. pyrifoliae and E. amylovora (S. Jock & K. Geider, unpublished) and the missing glucose in the side chain of E. pyrifoliae EPS could also affect the host range of the Korean pear pathogen compared to the fire blight pathogen.


   REFERENCES
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Bellemann, P. & Geider, K. (1992). Localization of transposon insertions in pathogenicity mutants of Erwinia amylovora and their biochemical characterization. J Gen Microbiol 138, 931-940.[Medline]

Bellemann, P., Bereswill, S., Berger, S. & Geider, K. (1994). Visualization of capsule formation by Erwinia amylovora and assays to determine amylovoran synthesis. Int J Biol Macromol 16, 290-296.[Medline]

Bereswill, S., Jock, S., Aldridge, P., Janse, J. D. & Geider, K. (1997). Molecular characterization of natural Erwinia amylovora strains deficient in levan synthesis. Physiol Mol Plant Pathol 51, 215-225.

Bernhard, F., Coplin, D. L. & Geider, K. (1993). A gene cluster involved in amylovoran synthesis of Erwinia amylovora: characterization and its relationship to cps genes in Erwinia stewartii. Mol Gen Genet 239, 158-168.[Medline]

Bernhard, F., Schullerus, D., Bellemann, P., Nimtz, M., Coplin, D. L. & Geider, K. (1996). Genetic transfer of amylovoran and stewartan synthesis between Erwinia amylovora and Erwinia stewartii. Microbiology 142, 1087-1096.[Abstract]

Bugert, P. & Geider, K. (1995). Molecular analysis of the ams operon required for exopolysaccharide synthesis of Erwinia amylovora. Mol Microbiol 15, 917-933.[Medline]

Bugert, P. & Geider, K. (1997). Characterization of the amsI gene product as a low molecular weight acid phosphatase controlling exopolysaccharide synthesis of Erwinia amylovora. FEBS Lett 400, 252-256.[Medline]

Coplin, D. L. & Majerczak, D. R. (1990). Extracellular polysaccharide genes in Erwinia stewartii: directed mutagenesis and complementation analysis. Mol Plant-Microbe Interact 3, 286-292.

Dolph, P. J., Majerczak, D. R. & Coplin, D. L. (1988). Characterization of a gene cluster for exopolysaccharide biosynthesis and virulence in Erwinia stewartii. J Bacteriol 170, 865-871.[Medline]

Du, Z. & Geider, K. (2002). Characterization of an activator gene upstream of lsc, involved in levan synthesis of Erwinia amylovora. Physiol Mol Plant Pathol 60, 9-17.

Falkenstein, H., Bellemann, P., Walter, S., Zeller, W. & Geider, K. (1988). Identification of Erwinia amylovora, the fireblight pathogen, by colony hybridization with DNA from plasmid pEA29. Appl Environ Microbiol 54, 2798-2802.

Geier, G. & Geider, K. (1993). Characterization and influence on virulence of the levansucrase gene from the fire blight pathogen Erwinia amylovora. Physiol Mol Plant Pathol 42, 387-404.

Geider, K., Hohmeyer, C., Haas, R. & Meyer, T. F. (1985). A plasmid cloning system utilizing replication functions of the filamentous bacteriophage fd. Gene 33, 341-349.[Medline]

Geider, K., Baldes, R., Bellemann, P., Metzger, M. & Schwartz, T. (1995). Mutual adaptation of bacteriophage fd, pfd plasmids and their host strains. Microbiol Res 150, 337-346.[Medline]

Ilan, O., Bloch, Y., Frankel, G., Ullrich, H., Geider, K. & Rosenshine, I. (1999). Protein tyrosine kinases associated with production of exopolysaccharide and virulence of animal and plant bacterial pathogens. EMBO J 18, 3241-3248.[Abstract/Free Full Text]

Jock, S., Rodoni, B., Gillings, M., Kim, W.-S., Copes, C., Merriman, P. & Geider, K. (2000). Screening of ornamental plants from the Botanic Gardens of Melbourne and Adelaide for the occurrence of Erwinia amylovora. Australas Plant Pathol 29, 120-128.

Jumel, K., Geider, K. & Harding, S. E. (1997). The solution molecular weight and shape of the bacterial exopolysaccharides amylovoran and stewartan. Int J Biol Macromol 29, 251-258.

Kim, W.-S. & Geider, K. (2000). Characterization of a viral EPS-depolymerase, a potential tool for control of fire blight. Phytopathology 90, 1263-1268.

Kim, W.-S., Gardan, L., Rhim, S.-L. & Geider, K. (1999). Erwinia pyrifoliae sp. nov., a novel pathogen that affects Asian pear trees (Pyrus pyrifolia Nakai). Int J Syst Bacteriol 49, 899-906.[Abstract]

Kim, W.-S., Jock, S., Paulin, J.-P., Rhim, S.-L. & Geider, K. (2001). Molecular detection and differentiation of Erwinia pyrifoliae and host range analysis of the Asian pear pathogen. Plant Dis 85, 1183-1188.

Leigh, J. A. & Coplin, D. L. (1992). Exopolysaccharides in plant-bacterial interactions. Annu Rev Microbiol 46, 307-346.[Medline]

Maes, M., Orye, K., Bobev, S. & 7 other authors (2001). Influence of amylovoran production on virulence of Erwinia amylovora and different amylovoran structure in E. amylovora isolates from Rubus. Eur J Plant Pathol 107, 839–844.

Nimtz, M., Mort, A., Domke, T., Wray, V., Zhang, Y., Qiu, F., Coplin, D. & Geider, K. (1996a). Structure of amylovoran, the capsular exopolysaccharide from the fire blight pathogen Erwinia amylovora. Carbohydr Res 287, 59-76.[Medline]

Nimtz, M., Mort, A., Wray, V., Domke, T., Zhang, Y., Coplin, D. L. & Geider, K. (1996b). Structure of stewartan, the capsular exopolysaccharide from the corn pathogen Erwinia stewartii. Carbohydr Res 288, 189-201.[Medline]

Rhim, S.-L., Völksch, B., Gardon, L., Paulin, J. P., Langlotz, C., Kim, W.-S. & Geider, K. (1999). Erwinia pyrifoliae, an Erwinia species different from Erwinia amylovora, causes a necrotic disease of Asian pear trees. Plant Pathol 48, 514-520.

Steinberger, E. M. & Beer, S. V. (1988). Creation and complementation of pathogenicity mutants of Erwinia amylovora. Mol Plant-Microbe Interact 1, 135-144.

Sutherland, I. W. (1988). Bacterial surface polysaccharides: structure and function. Int Rev Cytol 113, 187-231.[Medline]

Tomita, M., Kurokawa, T., Onozaki, K., Ichiki, N., Osawa, T. & Ukita, N. (1972). Purification of galactose-binding phytoagglutinins and phytotoxin by affinity column chromatography using Sepharose. Experientia 28, 84-85.[Medline]

Worley, K. C., Wiese, B. A. & Smith, R. F. (1995). BEAUTY: an enhanced BLAST-based search tool that intergrates multiple biological information resources into sequence similarity search results. Genome Res 5, 173-184.[Abstract]

Zhang, Y. & Geider, K. (1997). Differentiation of Erwinia amylovora strains by pulsed-field gel electrophoresis. Appl Environ Microbiol 63, 4421-4426.[Abstract]

Received 5 July 2002; revised 9 September 2002; accepted 12 September 2002.



This Article
Abstract
Full Text (PDF)
Alert me when this article is cited
Alert me if a correction is posted
Citation Map
Services
Email this article to a friend
Similar articles in this journal
Similar articles in PubMed
Alert me to new issues of the journal
Download to citation manager
Google Scholar
Articles by Kim, W.-S.
Articles by Geider, K.
Articles citing this Article
PubMed
PubMed Citation
Articles by Kim, W.-S.
Articles by Geider, K.
Agricola
Articles by Kim, W.-S.
Articles by Geider, K.


HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
INT J SYST EVOL MICROBIOL MICROBIOLOGY J GEN VIROL
J MED MICROBIOL ALL SGM JOURNALS
Copyright © 2002 Society for General Microbiology.