Molecular basis for the elicitation of plants innate ... · Molecular basis of elicitation of plant defenses cruciferous crops that is of worldwide importance. Xcc can also infect

Molecular basis of elicitation of plant defenses

THE ELICITATION OF PLANT INNATE IMMUNITY BY LIPOOLIGOSACCHARIDE OF XANTHOMONAS CAMPESTRIS

Alba Silipo*, Antonio Molinaro*†, Luisa Sturiale‡, J. Maxwell Dow§, Gitte Erbs¶, Rosa Lanzetta*, Mari-Anne Newman¶, Michelangelo Parrilli *

From *Dipartimento di Chimica Organica e Biochimica, Università di Napoli Federico II, Napoli, Italy. ‡Istituto per la Chimica e la Tecnologia dei Materiali Polimerici - ICTMP – CNR, Catania,

Italy, §BIOMERIT Research Centre, Department of Microbiology, BioSciences Institute, National University of Ireland, Cork, Ireland, ¶Section for Plant Pathology, Royal Veterinary and

Agricultural University (KVL), Copenhagen, Denmark.

†To whom correspondence should be addressed. E-mail: [email protected]. Prof. Antonio Molinaro. Dipartimento di Chimica Organica e Biochimica, Università di Napoli, Complesso Universitario Monte

Sant' Angelo, Via Cintia 4, 80126 Napoli, Italy, Tel. +39-081-674123, Fax: +39-081-674393.

Running title: Molecular basis of elicitation of plant defenses

1

JBC Papers in Press. Published on July 27, 2005 as Manuscript M506254200

Copyright 2005 by The American Society for Biochemistry and Molecular Biology, Inc.

by guest on June 11, 2020http://w

ww

.jbc.org/D

ownloaded from

mailto:[email protected]

http://www.jbc.org/


SUMMARY

Lipopolysaccharides (LPSs) and lipooligosaccharides (LOSs) are major components of the cell surface of Gram negative bacteria with diverse roles in bacterial pathogenesis of animals and plants that include elicitation of host defenses. Little is known about the mechanisms of perception of these molecules by plants and about the associated signal transduction pathways that trigger plant immunity. Here we address the issue of the molecular basis of elicitation of plant defenses through the structural determination of the LOS of the plant pathogen Xanthomonas campestris pv. campestris strain 8004 and examination of the effects of LOS and fragments obtained by chemical treatments on the immune response in Arabidopsis thaliana. The structure shows a strong accumulation of negatively charged groups in the lipid A-inner core region and has a number of novel features, including a galacturonyl phosphate attached at 3-deoxy-D-manno-oct-2-ulosonic acid (Kdo) residue and a unique phosphoramide group in the inner core region. Intact LOS and the lipid A and core oligosaccharides derived from it were all able to induce the defense-related genes PR1 and PR2 in Arabidopsis and to prevent the hypersensitive response (HR) caused by avirulent bacteria. Although LOS induced defense-related gene transcription in two temporal phases, the core oligosaccharide induced only the earlier phase and lipid A only the later phase. These findings suggest that plant cells can recognize lipid A and core oligosaccharide structures within LOS to trigger defensive cellular responses and that this may occur via two distinct recognition events.

INTRODUCTION Lipopolysaccharides (LPSs) are ubiquitous and

vital components of the cell surface of Gram-negative bacteria (1,2). They are amphiphilic macromolecules composed of a hydrophilic hetero-polysaccharide (comprising the core oligosaccharide and O-specific polysaccharide or O-chain) covalently linked to a lipophilic moiety termed lipid A, which anchors these macromolecules to the outer membrane. LPSs not possessing the O-chain are termed rough LPSs or lipo-oligosaccharides (LOSs).

In animal and insect cells, innate immune defenses are triggered by the perception of pathogen associated molecular patterns (PAMPs), conserved and generally indispensable microbial structures including LPSs. The recognition of PAMPs by these cells is often mediated by LRR (Leucine Rich Repeat) proteins such as Toll in Drosophila and the Toll-like receptors (TLR) in mammals (3-6). Recognition of LPSs occurs through the lipid A moiety which is responsible for most of the biological effects of LPS in animals. Lipid A toxicity in animals strongly depends on its structure and is also influenced by the covalently linked core region, which possesses immunogenic properties (1,2).

LPSs apparently have diverse roles in bacterial pathogenesis of plants. As major components of the outer membrane, LPSs are involved in the protection of bacterial cell, contributing to reduce the membrane permeability and thus allowing growth of bacterium in the unfavorable conditions of the plant environment (7). In contrast, LPSs can be recognized by plants to elicit or potentiate plant defense-related responses as part of a group of general elicitors that include flagellin and periplasmic oligosaccharides (8,9). One of the most widely studied effects of LPSs on plant cells is their ability to prevent the hypersensitive response (HR) induced in plants by avirulent bacteria. HR is a programmed cell death response, triggered by live bacteria, that is often associated with plant host resistance.

In comparison with animal and human cells, little is known about the mechanisms of LPS perception by plants and cognate signal transduction pathway. Recent findings have suggested that the lipid A moiety may be at least partially responsible for LPS perception by Arabidopsis thaliana leading to a rapid burst of NO, a hallmark of innate immunity in animals (9). Using synthetic O-antigen polysaccharides (oligorhamnans) it has been shown that the O-chain of LPS is recognized by Arabidopsis, and that this recognition leads to elicitation of a specific gene transcription response associated with defense (10). These observations, together with the established influence of the core region on lipid A toxicity in animals, indicate that the elucidation of the structure and of the biological activity of both lipid A and core region are of high importance for a better understanding of LPS action in plants.

Xanthomonas campestris pv. campestris (Xcc) is the causative agent of black rot, a disease of

2


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

Molecular basis of elicitation of plant defenses cruciferous crops that is of worldwide importance. Xcc can also infect non-crop crucifers such as the model plant Arabidopsis thaliana. Both LPS and LOS have been described in Xcc, with LOS being the predominant form in some strains (7). Here we have examined the molecular basis for the induction by Xcc LOS of defense responses in Arabidopsis thaliana. Our approach has been to determine the complete structure of purified Xcc LOS and to examine its activity in the prevention of HR caused by avirulent bacteria, and the induction of defense-related genes. We show that Xcc LOS is a unique molecule with a high negative charge density and a phosphoramide group, never found before in such molecules. Through comparison of the activity of intact Xcc LOS with chemically obtained (and structurally defined) fragments, we provide evidence that LOS acts to elicit defense responses in A. thaliana via independent recognition of the lipid A and the core oligosaccharide moieties respectively.

EXPERIMENTAL PROCEDURES

Bacterial growth and LOS extraction - Strain 8004 of Xanthomonas campestris pv. campestris was grown in 100 litres of peptone-yeast extract-glycerol medium (NYGB) in a New Brunswick SSR 7 fermenter and harvested as described previously (11). Freeze-dried cells were extracted three times with a mixture of aqueous 90% phenol/chloroform/petroleum ether (2:5:8 v/v/v) (12). After removal of the organic solvents under vacuum, the LOS fraction was precipitated from phenol with water, washed first with aqueous 80% phenol, and then three times with cold acetone, and lyophilized with a yield of about 4.3% of the dry mass. For detection of LOS, fractions were analyzed by SDS-polyacrylamide gel electrophoresis on 12% gels, which were stained with silver nitrate.

Chemical degradation of LOS for structural analysis - For isolation of OS1 and OS2, LOS (40 mg) was dissolved in tetrahydrofurane and treated with anhydrous hydrazine (2 ml), stirred at 37°C for 90 min, cooled, poured into ice-cold acetone (20 ml), and allowed to precipitate. The precipitate was then centrifuged (3000 × g, 30 min), washed twice with ice-cold acetone, dried, dissolved in water and lyophilized (oligosaccharide OS2, 32 mg, 80% of the LOS). An aliquot of OS2 product (20 mg) was de-N-acylated with 4 M KOH as described (13). Salts were removed by gel permeation chromatography with Sephadex G-10 (Pharmacia) column (50 × 1.5 cm) to yield the

resulting oligosaccharide OS1 (16 mg, 40% of the LOS).

Chemical degradation of LOS for plant tests - Free lipid A was obtained by hydrolysis of the LOS (40 mg) with 10mM sodium acetate buffer pH 4.4, (100°C, 3h). The solution was extracted three times with CHCl3/MeOH/H2O (100:100:30 v/v/v) and centrifuged (4°C, 5000 g, 15 min). The organic phase contained the lipid A and the water phase contained the core oligosaccharide.

General Analytical Methods. - Determination of sugars residues and of their absolute configuration, GLC and GLC-MS were all carried out as described (14, 15). Monosaccharides were identified as acetylated O-methyl glycosides derivatives. After methanolysis (2M HCl/MeOH, 85°, 24 h) and acetylation with acetic anhydride in pyridine (85°, 30 min) the sample was analyzed by GLC-MS. Linkage analysis was carried out by methylation of the complete core region as described (16). The sample was hydrolyzed with 4 M trifluoroacetic acid (100°C, 4 h), carbonyl-reduced with NaBD4, carboxy-methylated, carboxyl-reduced, acetylated and analyzed by GLC-MS. Total fatty acid content was obtained by acid hydrolysis (17). LOS was first treated with HCl 4M (4h, 100°C) and then neutralized with NaOH 5M (30 min, 100°C). Fatty acids were then extracted in CHCl3, methylated with diazomethane and analyzed by GLC-MS. The ester bound fatty acids were selectively released by base-catalyzed hydrolysis with NaOH 0.5M/MeOH (1:1 v/v, 85°, 2h), then the product was acidified, extracted in CHCl3, methylated with diazomethane and analyzed by GLC-MS. The absolute configuration of fatty acids was determined as described (18).

NMR spectroscopy - Structural assignments of OS1, 1D and 2D 1H-NMR spectra were recorded on a solution of 5 mg in 0.6 ml of D2O, at 300 K, at pD 7 (uncorrected value). For structural assignment of OS2, spectra were recorded on a solution of 1% deuterated SDS (Sodium Dodecyl Sulphate) with 5 µl of 32% NH4OH at 298 K at pD 9.5 (uncorrected value). 1H and 13C NMR spectra were measured on Bruker DRX-400 (OS1) and Varian INOVA 500 (OS2). Spectra were calibrated with internal acetone [δH 2.225, δC 31.45]. 31P NMR experiments were carried out using a Bruker DRX-400 spectrometer, aqueous 85% phosphoric acid was used as external reference (0.00 ppm). Nuclear Overhauser enhancement spectroscopy (NOESY) and rotating frame Overhauser enhancement spectroscopy (ROESY) were measured using data sets (t1 × t2)

3


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

Molecular basis of elicitation of plant defenses of 4096 × 512 points, and 16 scans were acquired. A mixing time of 200 ms was used. Double quantum-filtered phase-sensitive correlation spectroscopy (DQF-COSY) experiment was performed with 0.258 s acquisition time, using data sets of 4096 x 1024 points, and 64 scans were acquired. Total correlation spectroscopy experiments (TOCSY) were performed with a spinlock time of 120 ms, using data sets (t1 × t2) of 4096 × 512 points, and 16 scans were acquired. In all homonuclear experiments the data matrix was zero-filled in the F1 dimension to give a matrix of 4096 x 2048 points and was resolution enhanced in both dimensions by a shifted sine-bell function before Fourier transformation. Coupling constants were determined on a first order basis from DQF-COSY (19, 20). Heteronuclear single quantum coherence (HSQC) and heteronuclear multiple bond correlation (HMBC) experiments were measured in the 1H-detected mode via single quantum coherence with proton decoupling in the 13C domain, using data sets of 2048 × 512 points, and 64 scans were acquired for each t1 value. Experiments were carried out in the phase-sensitive mode according to the method of States et al. (21). The heteronuclear experiments were performed using pulse field gradient programs. 1H,13C HMBC was optimized for 6 Hz coupling constant, and 1H,31P HSQC for 8 Hz coupling constant. In all heteronuclear experiments the data matrix was extended to 2048 × 1024 points using forward linear prediction extrapolation (22).

Mass Spectrometry - Ions formed by the pulsed laser beam were accelerated through 24kV. Each spectrum is the result of approximately 200 laser shots. Mass spectra were assigned with a two-point external calibration. MALDI preparation of such highly anionic samples was set up with substantial modifications of the thin-layer procedure described by Körner et al. for the analyses of enzymatic pectin digests (23, 24). A small amount of the intact LOS was first suspended in a methanol-water 1/1 solution containing 5 mM ethylendiaminetetraacetic acid (EDTA) and allowed to dissolve by a brief ultrasonication. Few microliters of the obtained mixture were then desalted on a small piece of Parafilm® with some grains of cation exchange beads (Dowex 50WX8-200, Sigma–Aldrich), previously converted in the ammonium form. 0.3 µl of this sample solution were finally deposited, together with the same volume of 20 mM dibasic ammonium citrate, in a thin layer of homogeneous matrix film obtained from a solution which components are 2,4,6-trihydroxyacetophenone

(THAP) 200 mg/mL in methanol, and nitrocellulose (Trans-blot membrane, BioRad) 15 mg/mL in acetone/propan-2-ol (1:1 v/v), mixed respectively in a 4:1 v/v ratio. Plant tests - LOS and derived fractions were tested for their ability to prevent the HR in Arabidopsis thaliana accession Columbia (Col-O) caused by Pseudomonas syringae pv. tomato strain DC3000 carrying the avirulence genes avrRpm1 or avrRps4 (Pst avrRpm1, Pst avrRps4) (25, 26). Fractions dissolved in water (50 µg ml-1) were infiltrated into 6-week-old leaves of Col-O. A control leaf was inoculated with water. The plants were placed in a growth cabinet at 25 °C with 16 hours light. After 20 hours these leaves were inoculated with Pst avrRPM1 or Pst avrRps4 (107 cfu ml-1) in both the area treated with one of the compounds and in the water treated area. Development or suppression of HR were followed over 24 hours in the LOS, core oligosaccharide, lipid A pre-treated and control (water-treated) areas as described previously (27). All experiments were repeated at least three times. Growth of the virulent bacteria Pst DC3000 was followed in the plant tissue in the LOS, core oligosaccharide, lipid A pre-treated and control (water-treated) areas as described (28). The mean and standard deviation of three separate measurements at each time point are given. Analysis of PR1 and PR2 gene expression - LOS and derived fractions were dissolved in water (50 µg ml-1) and infiltrated into 6 weeks old leaves of Col-O. A further set of leaves was inoculated with water. The plants were placed in a growth cabinet at 25°C with 16 hours light. The leaves were harvested 4, 12, 20 and 24 hours after inoculation. The changes in PR1 and PR2 gene expression were followed using quantitative realtime RT-PCR analysis exactly as described (10). The PR-1 and PR-2 primer pairs (PR-1 Forward: 5’-GTGGGTTAGCGAGAAGGCTA-3’; PR-1 Reverse: 5’-ACTTTGGCACATCCGAGTCT-3’; PR-2 Forward: 5’-TCAAGGAAGGTTCAGGGATG-3’; PR-2 Reverse: 5’-AGATTCACGAGCAAGGGAGA-3’) were designed using the gene sequence for PR1 assigned AT2g14610 and PR2 assigned AT3G57260 in the Arabidopsis Genome Initiative3 and the primer3 program http://www-genome.wi.mit.edu/cgibin/primer/primer3_www.cgi as described (10).

RESULTS Compositional analysis of Xcc LOS and structural characterization of the fully de-acylated LOS

4


ww

.jbc.org/D

ownloaded from

http://www-genome.wi.mit.edu/cgibin/primer/primer3_www.cgi



http://www.jbc.org/

Molecular basis of elicitation of plant defenses fraction - Fatty acids composition, compositional and linkage analysis of carbohydrates obtained by GLC-MS are reported in Table 1. All monosaccharides are in D configuration and fatty acids with R configuration. The approach to define the primary structure of Xcc LOS was to employ two chemical degradations followed by compositional and 2D NMR and MS analyses of the obtained compounds.

The LOS fraction was completely de-acylated by anhydrous hydrazine and hot KOH. The compositional analysis (Table 1) of the product obtained OS1 revealed the presence of t-Manp, t-Glcp, 3-Manp 3-Glcp, 4-Manp, 5-Kdop, 6-GlcpN; no uronic acids were detected.

The NMR analysis (DQF-COSY, TOCSY, NOESY, 31P-1H and 13C-1H HSQC, 13C-1H HMBC) of OS1 showed the existence of a mixture of two oligosaccharides differing in the length of sugar backbone. The anomeric region of the 1H-NMR spectrum showed six signals (A-F’, Fig. 1, Table 2 and Supplemental Data Fig.1) and their area suggested for two of them, B and C, to be in 1:2 ratio with respect to the others. In the NMR spectra, the two α- and β-GlcN residues (A and E, respectively) of the lipid A backbone were clearly identified since the anomeric signals at 5.65 and 5.00 ppm were correlated to the expected carbon signals at 92.0 and 100.7 ppm, respectively, and their H-2 signals, at 3.35 and 3.04 ppm, were correlated to the nitrogen-bearing carbon signals at 55.7 and 59.7 ppm, respectively. In addition, the high multiplicity of the anomeric signal of A indicated the phosphorylation of the anomeric position. Residues B, H-1 at 5.28 ppm, C, H-1 at 5.13 ppm, and D H-1 at 5.09 possessed α-manno-configurations, as shown by both 3JH1,H2 and 3JH2,H3 values around 2-3 Hz. The H-1 signal of and F/F’ at 4.58 ppm was assigned to glucose residues β-configuration as shown by the strong intra-residual NOE contacts measured among H-1, H-3 and H-5 of F/F’. Because of the absence of the anomeric proton signal, the spin system of Kdo G (Fig. 1) was identified starting from the diasterotopic H-3 methylene protons found in a shielded region at 1.87 and 2.08 ppm (H-3ax and H-3eq, respectively). The α-configuration at C-2 was attributed by the chemical shift values of H-3eq and by the values of 3JH-7, H-8a and 3JH-7, H-8b (29).

The down-field shift of carbon resonances allowed to locate the positions of glycosylation at O-6 of A and E, at O-4 of D, at O-3 of B and F and at O-5 of G, in full accordance with the methylation data. Residues C and F’ were terminal sugars.

The interresidual NOE contacts of the NOESY spectrum (Fig. 2a) and the scalar correlations present in the HMBC spectrum (Fig. 2b and Supplemental Data Fig. 2) were used to infer the sequence of residues in the oligosaccharide chain and to confirm the glycosylation positions. The linkage between glucosamine residues A and E of the lipid A backbone was validated by the inter-residual NOE contact of H-1 E (5.00 ppm) with H-6a,b A (4.26/3.90 ppm). The weak down-field shift of C-6 E (63.7 ppm) was in agreement with the α-(2→6) ketosidic linkage of Kdo G with residue E of β-GlcN. The NOE correlation of H-5 G with H-1 D together with the HMBC correlation H-1 D/C-5 G proved the linkage between residues of Kdo G and α-Man D. Residue D was in turn substituted at O-4 by residue F/F’, according to the NOE connectivity of H-4 D with H-1 F/F’ and the HMBC long range correlation C-4 D/H-1 F/F’. Residue F’ was a terminal glucose, whereas residue F was further substituted at O-3 by a non-stoichiometric disaccharide formed by the two residues of mannose B and C (Fig.1). NOE contacts between H-1 C and H-3 B and the HMBC correlations H-1 C/C-3 B were diagnostic of the presence of a α-(1→3) mannose disaccharide which was linked at O-3 of β-glucose F by residue B, as confirmed by NOE contact of H-1 B with H-3 F and the long range correlation H-1 B/C-3 F.

The 31P-1H HSQC spectrum showed three cross peaks, whose 31P chemical shifts were in accordance with the presence of three phosphate groups. Two of these signals, at 3.11 and 4.52 ppm, that correlated with the proton signals at 5.65 and 3.72 ppm, respectively, were identified as H-1 of α-GlcN A and H-4 of β-GlcN E composing the lipid A backbone. The third phosphate group was linked at O-3 of α-Man D, as shown by the cross peak between the 31P signal at 3.45 ppm and H-4 D at 4.56 ppm.

Thus, these above results can be summarized as indicating the presence of a mixture of two oligosaccharides, differing in the length of the sugar backbone because of the presence of a terminal α-(1→3) disaccharide of mannose linked in non-stoichiometric amount at O-3 of β-D-Glc F (Fig. 1). Structural characterization of the de-O-acylated LOS fraction - In order to detect the presence of alkaline labile substituents (i.e., phosphate residues) likely lost by harsh alkaline treatment, the LOS was only de-O-acylated by mild hydrazinolysis. The obtained product (OS2) was subjected to a complete chemical and NMR investigations which revealed the presence of a

5


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

Molecular basis of elicitation of plant defenses complex mixture of oligosaccharides. The NMR spectra were recorded on a solution of 1% deuterated SDS (Sodium Dodecyl Sulphate) with 5 µl of 32% NH4OH (298 K, pD 9.5).

In full accordance with the GLC-MS analysis of OS2 (Table 1), the 1H NMR spectrum (Fig. 3a) showed two additional spin systems with respect to OS1, that were identified as two residues of α-GalA, M and H. The chemical shifts and multiplicity of their H-1 signals suggested that they were both coupled to phosphate signals. The assignment of the other spin systems already detected in the previous analysis was straightforward (Table 3).

In the 31P NMR spectrum (Fig. 3b) five main signals could be recognized in three different regions of the spectrum. Two signals were found in the chemical shift region typical of monophosphate di-ester group, at -1.9 and -0.9 ppm, whereas two other signals, at 2.94 and 5.23 ppm, were recognized as monophosphate monoester groups. The last and low-field 31P signal present in the spectrum, a doublet at 16.2 ppm, was recognized as a phosphoramide group signal, in accordance with published data (30, 31).

An extensive 2D NMR analysis lead to the identification of two pairs of oligosaccharides, differing in the phosphorylation pattern of residue D (4-α-Man), represented by two alternative and remarkably different spin systems: D-X and D-Y (H-1 signal at 4.97 and 5.24 ppm, respectively, Fig. 3a). The location of these phosphorus-containing groups was deduced from the 31P-1H HSQC spectrum (Fig. 4). The phosphate group signals at 2.94 and 5.23 ppm correlated to proton resonances at 5.26 and 3.67 ppm, attributed to H-1 A and H-4 E, respectively. Thus, as expected, two phosphate groups belong to the di-acylated lipid A and were linked to O-1 of α-GlcN A and to O-4 of β-GlcN E. The cross peaks in the region of monophosphate di-ester group correlated to resonances in the anomeric region in F2 dimension. The first signal, at -0.9 ppm, correlated to two protons at 5.49 ppm and 4.50 ppm, H-1 H and H-4 G, respectively; the other at 31P signal at -1.9 ppm also showed a double correlation to two proton signals resonating at 5.54 ppm and at 4.33 ppm, recognized as H-1 M and H-3 D-X. Thus, two phosphodiester bridges were identified, both bearing α-galacturonyl-phosphate groups. The first was attached at position O-4 of Kdo G, while the second α-galacturonyl-phosphate, present in minor amount, was located at position O-3 of α-Man D-X (Fig. 4).

The location of the phosphoramide group remained to be established. The fifth cross peak, the phosphoramide signal at 16.2 ppm, displayed strong correlation to two protons at 4.59 and 4.49 ppm, i.e., H-3 and H-2 signals of the alternative α-mannose spin system, residue D-Y, and minor correlations with H-1 and H-4 D-Y were also present, attributable to 4JP,H correlations. These correlations validated the existence of a five membered ring phosphoramide (Fig. 4) attached by O-2 and O-3 of the α-mannose residue D-Y. This assumption was also supported by the downfield chemical shift values of H-2 and H-3 D-Y, at 4.59 and 4.49 ppm (Table 3).

From the above data, a mixture of two pairs of oligosaccharides, X and Y, was identified in OS2 product. They carried, in non-stoichiometric amounts, a disaccharide of mannose and differed in the phosphorylation pattern of α-mannose D. In particular, oligosaccharide X carried a α-galacturonyl phosphate at O-3 of α-mannose (D-X) while oligosaccharide Y carried a phosphoramide ring substituting the position O-2 and O-3 of α-mannose (D-Y). The NMR data were strictly confirmed by mass spectrometry.

An aliquot of intact LOS was analyzed by negative ion MALDI-TOF (Figure 5). Although MALDI is recognized as a soft ionization technique, it is known that, together with the desorption and the ionization of intact LOSs, a prompt and regio-specific cleavage of the labile α-Kdo linkage inevitably occurs. This very informative in-source fragmentation of lipid A from core oligosaccharide is dependent upon the laser power settings (24, 32 ). The MALDI-TOF mass spectrum comprised three families of ions; at the lowest masses (m/z <1500), four peaks (P-S) were attributable to the core oligosaccharide. In the region of m/z between 1500-2000, two groups of ion peaks (L and L1) could be ascribed to the lipid A, and at m/z > 2000, four groups of ions (LP-LS) were attributable to the intact LOS.

The lipid A fragment, as expected, showed a remarkable heterogeneity due to the fatty acid variability. The main group of ions (L) was found around m/z 1644.4, indicating the presence of a hexa-acylated species carrying two GlcN, two phosphate groups and acyl chains of different length (∆m/z = 14) distributed on the disaccharide backbone. These ions were identified as lipid A molecules carrying (R)-12:0 (3-OH) and/or (R)-13:0 (3-OH) in amide linkage, different ester-linked 3-hydroxy fatty acids [(R)-10:0 (3-OH), (R)-11:0 (3-OH), (R)-12:0 (3-OH), (R)-13:0 (3-OH)] and two secondary fatty acids, 11:0 and/or

6


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

Molecular basis of elicitation of plant defenses 10:0. The less abundant ion group L1 around m/z 1767.7 (∆m/z = 123) were attributable to the presence of an additional residue of 2-aminoethyl phosphate. On the basis of further degradation of lipid A and subsequent MS analyses (33), the lipid A structure from Xcc LPS appeared to be identical to those of a Xanthomonas LPSs previously studied (34). Thus, it was characterized by a remarkable variability in the fatty acids length and possessed an uncommon symmetric [3+3] distribution of acyl chains on disaccharide backbone with the acyloxyacyl groups exclusively ester linked (34).

The prominent core oligosaccharide ions were represented by those of moieties P and Q (Figure 5). Species Q at m/z 1054.8 was consistent with an oligosaccharide possessing one Kdo, two hexoses, two phosphate groups and two hexuronic acids; species P at m/z 879.3 (∆m/z 175.5 with respect to Q) was lacking a uronic acid. Species R and S (m/z 1203.2 and 1378.8) showed a ∆m/z = 324 with respect to P and Q diagnostic of the presence of two additional hexoses.

Furthermore, it was possible to assign the mass peaks of the four groups of ions corresponding to the intact LOS (species LP-LQ, Figure 5). The group of ions termed LQ identified LOS molecules composed by a bis-phosphorylated hexa-acylated lipid A carrying primary and secondary fatty acids, one Kdo, two hexoses, two phosphate and two hexuronic acid residues; species LP differed by the absence of a hexuronic acid unit. Species composing the ion group LS carried two additional hexoses with respect to LQ (∆m/z = 324), as the ion group LR with respect to LP (Figure 5).

Although NMR indicated the presence of a five-membered ring phosphoramide, no ion peaks consistent with the presence of a phosphoramide ring were visible in the MS spectrum (18 Da less with respect to ion peaks which contain the phosphoramide group). We hypothesized that this apparent discrepancy was due to closure of the five-membered ring only in the alkaline medium in which the product was dissolved for the NMR measurements. This was verified by a MALDI-TOF mass spectrum of the product subjected to NMR measurements (not shown) that showed molecular ions differing by 18 Da from the LP-LS ions of Fig. 5. Overall therefore there was a strict correlation between the NMR data and the MS measurements. On the basis of compositional analysis, NMR and MALDI MS measurements the complete structure of the LOS from Xcc was determined (Figure 6).

The effects of LOS and its derivatives on prevention of the hypersensitive response in Arabidopsis thaliana - In many different plants, pre-treatment with LPS can prevent the hypersensitive response (HR) induced by avirulent bacteria, a phenomenon that has been termed localized induced resistance (LIR) (35). Xcc LOS triggered a similar effect in the model plant Arabidopsis thaliana; pre-treatment of leaves with purified LOS prevented HR caused by subsequently inoculated avirulent bacteria (Figure 7a-b), and reduced the growth of virulent bacteria (Figure 7c). LOS pre-treatment prevented HR that is genetically determined by either the RPM1/avrRpm1 or the RPS4/avrRps4 interaction, which are triggered by different plant signal transduction pathways (36). It is clear from these findings and work on other plants (tobacco, pepper, turnip) that the effects of LOS treatment are associated with an enhanced resistance of the plant tissue to pathogenic bacteria, which we presume to occur through LOS-dependent induction or potentiation of plant defense responses (27, 35, 37). The mechanism(s) by which LOS prevents the HR are however as yet unknown.

In order to examine the structural basis for the LOS induction of the LIR response, we investigated the activity of the components of LOS. Both diphosphoryl lipid A and the core oligosaccharide moieties of LOS derived by mild acid treatment were able to suppress the HR (Table 4) (both products were identified by MALDI-TOF and NMR as described (34)). Since lipid A and core oligosaccharide moieties are structurally quite different, our findings provide evidence for the existence of multiple elicitors or potentiators of plant defenses within the Xcc LOS. The finding that lipid A alone can trigger defense responses in A. thaliana is consonant with recent observations that lipid A can generate a rapid burst of NO production in cells of the same plant (9).

It is well established that the phosphorylation pattern of lipid A can affect its biological activity in mammalian systems; the binding of lipid A to mammalian cell surface receptors is believed to require electrostatic interactions involving the phosphate groups (38, 39). To test whether such ionic interactions also participate in the plant recognition of Xcc LOS, we assayed the LIR inducing activity of a derivative that had been treated with HF to remove all of the phosphate groups together with the phosphodiester-linked galacturonyl residues and the phosphoramide group, that is, the molecule only maintained the

7


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

Molecular basis of elicitation of plant defenses negative charge on the Kdo residue (the product was identified by MALDI-TOF and NMR). This treatment completely destroyed the ability to induce LIR (Table 4), suggesting a role for ionic interactions in the binding of not only lipid A but also the core oligosaccharide to putative plant receptors.

The effects of LOS and its derivatives on induction of the defense–related genes PR1 and PR2 - It is established that LPS from Burkholderia cepacia can induce accumulation of transcripts of a number of defense-related genes in A. thaliana leaves (9). This together with the above findings prompted us to examine the effects of LOS, lipid A and the core oligosaccharides on the induction of the defense-related genes PR1 and PR2. Xcc LOS caused a substantial increase in accumulation of transcripts for both PR1 and PR2 in A. thaliana. Transcript levels in response to Xcc LOS showed an early but transient accumulation at 12h and a later, more substantial accumulation after 20h (Table 5). Both the core oligosaccharide and lipid A preparations induced the accumulation of the defense gene transcripts, but the kinetics of the response were quite different in the two cases. The core oligosaccharide induced the more rapid response but did not induce the larger response seen after 20h, whereas lipid A produced only a relatively weak response at 12h, but induced a substantial response after 20h. These patterns of PR1 and PR2 induction, compared with that induced by LOS, strongly suggest that LOS is recognized via independent mechanisms involving the core and lipid A moieties, respectively.

CONCLUSION

To our knowledge, this is the first complete structure for a core oligosaccharide from a Xanthomonas spp. to be reported. The LOS comprised a mixture of species differing in the acylation, phosphorylation pattern and in the non-stoichiometric presence of a terminal disaccharide of mannose. It is very likely that this mannose disaccharide is previously assembled and later transferred during LOS biosynthesis; it may either constitute a primer for the attachment of the O-polysaccharide chain to the core, or, conversely, hamper this linkage. This is the first time that a phosphoramide group has been described as a component of bacterial lipopolysaccharide, and the second time it has been found in Nature as a component of a biomolecule. It was previously described as a constituent of the capsular polysaccharide of Campylobacter jejuni (31). Interestingly, the phosphoramide moiety is highly

sensitive to the acid or alkaline conditions commonly used to work-up LOSs and LPSs molecules (13), therefore, it cannot be excluded that this chemical group is more widely present. As in many LPSs, the inner core region of Xcc LOS carries anionic substituents, which in this case are located on Kdo and α-mannose residues; the Kdo residue carries a α-galacturonyl-phosphate substituent at its position O-4 whereas the adjacent α-mannose is substituted at O-3 by a second α-galacturonyl phosphate or, alternatively, by a phosphoramide group (Figure 6). The presence of negatively charged substituents in close proximity of lipid A-core region is functionally important for intermolecular associations by cross-linking of divalent cations (1). This electrostatic interaction contributes to enhance the stability of the external bacterial membrane with the formation of a strong, rigid and protective barrier (1). Innate immune defenses in animal and insect cells are triggered by the perception of pathogen associated molecular patterns (PAMPs), conserved and vital components of the microbial metabolism including LPSs. The recognition of PAMPs by these cells is mediated by LRR (Leucine Rich Repeat) proteins such as Toll in Drosophila and the Toll-like receptors (TLR) in mammals (3-6). The endotoxic properties of LPS principally reside in the lipid A, the primary immunostimulator center of Gram-negative bacteria. It acts as strong stimulator of the innate immune system via the induction of inflammatory mediators released by host cells. The investigation of the role and the mechanisms of action of LPS in plant-bacteria interaction are still at an early stage. Plants have evolved the capacity to recognize LPS, flagellin and other general elicitors that are pathogen surface molecules able to trigger cellular defenses and thus deemed as PAMPs (10, 40, 41). There is increasing evidence that the activation of plant innate response upon recognition of PAMPs partly resembles the mechanisms of activation of the innate responses in mammalian and insect systems. However, it is not clear whether mechanistic parallels for LPS perception by plants, animals and insects exist, as have been shown for flagellin perception.

Our work demonstrates that pathogen perception by plant cells has parallels to the innate immune systems in animal and insects in that similar components for recognition and signal transduction are employed. We have investigated one aspect of the recognition of LPS by plants by a

8


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

Molecular basis of elicitation of plant defenses complete structural definition of those components within a LPS capable of inducing plant defenses. Specifically we have shown that both the lipid A moiety and the core oligosaccharide are recognized by plant cells to trigger the LIR response and the expression of PR1 and PR2 in A. thaliana. The effect of lipid A alone on the induction of defense responses provides a ready explanation for the ability of LPSs from many different bacteria to induce LIR, since the structure of lipid A has a highly conserved chemical architecture within bacterial LPS. These findings together with recent observations that LPS from a number of bacterial sources and isolated lipid A can induce NO production in A. thaliana (9), indicate that lipid A acts as a PAMP recognized by plant cells and triggers the activation of immune response genes, as in mammals and insects. On the other hand, we have also demonstrated that core oligosaccharide induces plant defense and that this occurs by a mechanism that is independent from that of lipid A. We infer from the pattern of defense gene induction by LOS that plant perception of this molecule involves independent recognition of the lipid A and the core oligosaccharide moieties respectively. Moreover, the demonstration that de-phosphorylated LOS molecule, which retains a single negative charge on inner core, does not induce any defense response in A. thaliana suggests a key role for the charged phosphate, phosphoramide and galacturonic acid residues.

Although core oligosaccharide from Xcc is active in triggering LIR (and inducing defense

gene expression), those of other bacteria such as Escherichia coli and Ralstonia solanacearum are not (42). This specific effect of the Xcc core oligosaccharide could be due to its unique structural features.

An understanding of the effects that LPS has on plants and how these are triggered has implications both for the control of bacterial diseases through general non-specific mechanisms and also in biocontrol of bacterial and fungal diseases by beneficial bacteria, which in some cases is believed to occur as a consequence of recognition of LPS. The challenge ahead is to identify the plant components involved in LPS recognition and subsequent signal transduction. The appreciation that structurally distinct components within LPS may trigger the same plant responses (possibly via different receptors) has substantial implications for the design of genetic or biochemical screens to identify such receptors.

ACKNOWLEDGMENTS

The work at Università di Napoli is financially supported by a grant from MIUR, Rome (Progetti di Ricerca di Interesse Nazionale, 2004) (to M.P.). The work at Biomerit Research Centre is supported by the Science Foundation of Ireland through an Investigator award (03/IN3/B373). The work at KVL is supported by the Danish Agricultural and Veterinary Research Council. We also acknowledge financial support from Carlsbergfondet, Copenhagen, Denmark.

.

9


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/


REFERENCES

1. Raetz, C. R. H., Whitfield, C. (2002) Annu. Rev. Biochem. 71, 635-700.

2. Alexander, C., Rietschel, E. Th. (2001) J. Endotoxin Res. 7, 167-202.

3. Miyake, K. (2004) Trends in Microbiol. 12, 186-192.

4. Medzhitov, R. (2001) Nature Rev. Immunol. 1, 135-145.

5. Akira, S., Takeda, K., Kaisho, T. (2001) Nature Immunol. 2, 675-680.

6. Triantafilou, M., Triantafilou, K. (2002) Trends Immunol. 23, 301-304.

7. Dow, J. M., Osbourn, A. E., Wilson, T. J. G., Daniels, M. J. (1995) Mol. Plant-Microbe Interact. 8, 768-777.

8. Conrath, U., Thulke, O., Katz, V., Schwindling, S., Kohler, A. (2001) Eur. J. Plant Pathol. 107, 113-119.

9. Zeidler, D., Zahringer, U., Gerber, I., Dubery, I., Hartung, T., Bors, W., Hutzler, P., Durner, J. (2004) Proc. Natl. Acad. Sci. U.S.A. 101, 15811-15816.

10. Bedini, E., De Castro, C., Erbs, G., Mangoni, L., Dow, J. M., Newman, M. -A., Parrilli, M., Unverzagt, C. (2005) J. Am. Chem. Soc. 127, 2414-2416.

11. Molinaro, A., Silipo, A., Lanzetta, R., Newman, M. A., Dow, J. M., Parrilli, M. (2003) Carbohydr. Res. 338, 277–281.

12. Galanos, C., Luderitz, O., Westphal, O. (1969) Eur J Biochem. 9, 245-249.

13. Holst, O. (2000) Methods in Molecular Biology, Bacterial Toxins: Methods and Protocols, Humana Press Inc, Totowa, NJ.

14. Leontein, K., Lönngren, J. (1978) Methods Carbohydr. Chem. 62, 359-362.

15. Molinaro, A., De Castro, C., Lanzetta, R., Evidente, A., Parrilli, M., Holst, O. (2002) J. Biol. Chem. 277, 10058-10063.

16. Hakomori, S. (1964) J. Biochem. (Tokyo) 55, 205-208.

17. Silipo, A., Lanzetta, R., Garozzo, D., Lo Cantore, P., Iacobellis, N. S., Molinaro, A., Parrilli, M., Evidente A. (2002) Eur. J. Biochem., 269, 2498-2505.

18. Rietschel, E. T. (1976) Eur. J. Biochem. 64, 423-428.

19. Piantini, U., Sørensen, O. W., Ernst, R. R. (1982) J. Am. Chem. Soc. 104, 6800-6801.

20. Rance, M., Sørensen, O. W., Bodenhausen, G., Wagner, G., Ernst, R. R., Wüthrich, K. (1983) Biochem. Biophys. Res. Commun. 117, 479-485.

21. States, D. J., Haberkorn, R. A., Ruben, D. J. (1982) J. Magn. Reson. 48, 286-292.

22. Stern, A. S., Li, K. –B., Hoch, J. C. (2002) J. Am. Chem. Soc. 124, 1982-1993.

10


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/


23. Körner, R., Limberg, G., Mikkelsen, J.D., Roepstorff, P. (1998) J. Mass Spectrom. 33, 836-842.

24. Sturiale, L., Garozzo, D. Silipo, A., Lanzetta, R., Parrilli, M., Molinaro, A., (2005) Rapid. Commun. Mass Spectrom. in press.

25. Ritter, E., Debener, T., Barone, A., Salamini, F., Gebhardt, C. (1991) Mol. Gen. Genet. 227, 81-85.

26. Hinsch, M., Staskawicz, B. (1996) Mol. Plant.Microbe Interact. 9, 55-61.

27. Newman, M. -A. , von Roepenack-Lahaye, E., Parr, A., Daniels, M. J., Dow, J. M. (2002) Plant J. 29, 487-495.

28. Newman, M. -A., Daniels, M. J., Dow, J. M. (1995) Mol. Plant-Microbe Interact. 8, 778-780.

29. Birnbaum, G.I., Roy, R., Brisson, J.R., Jennings, H.J. (1987) J. Carbohydr. Chem., 6, 17–39.

30. Gilard, V., Martino, R., Malet-Martino, M., Niemeyer, U., Pohl, J. (1999) J. Med. Chem. 42, 2542-2560.

31. Szymanski, C. M., St. Michael, F., Jarrell, H. C., Li, J., Gilbert, M., Larocque, S., Vinogradov, E., Brisson, J. -R. (2003) J. Biol. Chem. 278, 24509-24520.

32. Gibson, B. W., Engstrom, J. J., John, Hines, C. M. W., Falick A. M. (1997) J. Am. Soc. Mass Spectrom. 8, 645-658.

33. Silipo, A., Lanzetta, R., Amoresano, A., Parrilli, M., Molinaro, A. (2002) J. Lipid Res., 43, 2188-2195.

34. Silipo, A., Molinaro, A., Lanzetta, R., Parrilli, M., Lindner, B., Holst O. (2004) Eur. J. Org. Chem. 2004, 1336-1343.

35. Erbs, G., Newman, M.A. (2003) Mol. Plant Pathol. 4, 421-425.

36. Aarts, N., Metz, M., Holub, E., Staskawicz, B. J., Daniels, M. J. & Parker, J. E. (1998) Proc. Natl. Acad. Sci. U.S.A. 95, 10306-10311.

37. Dow, J. M., Newman, M.-A., von Roepenack, E. (2000) Annu. Rev. Phytopathol. 38, 241-261.

38. Schromm, A. B., Brandenburg, K., Loppnow, H., Moran, A.P., Koch, M. H. J., Rietschel, E. Th., Seydel, U. (2000) Eur. J. Biochem. 267, 2008-2013.

39. Schromm, A. B., Brandenburg, K., Loppnow, H., Zahringer, U., Rietschel, E.Th., Carroll, S. F., Koch, M. H. J., Kusumoto, S., Seydel, U. (1998) J. Immunol. 161, 5464-5471.

40. Nurnberger, T., Brunner, F. (2002) Curr. Opin. Plant Pathol. 5,1-7.

41. Borregaard, N., Elsbach, P., Ganz, T., Garred, P., Svejgaard, A. (2002) Immunol. Today 21, 68-70.

42. Newman, M.-A., Daniels, M. J., & Dow, J. M. Mol. (1997) Plant-Microbe Interact. 10, 926–928.

11


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/


FOOTNOTES

The abbreviations used are: DQF-COSY, double quantum filtered correlation spectroscopy; GLC-MS, gas liquid chromatography-mass spectrometry; GlcN, 2-amino-2-deoxy-glucose; HMBC, heteronuclear multiple bond correlation; HR, hypersensitive response; HSQC, heteronuclear single quantum coherence; Kdo, 3-deoxy-D-manno-oct-2-ulosonic acid; LIR, localized induced resistance; LOS, lipooligosaccharide; LPS, lipopolysaccharide; MALDI, matrix-assisted laser desorption ionization; NOE, nuclear Overhauser effect; NOESY, nuclear Overhauser enhancement spectroscopy; PAMP, pathogen associated molecular patterns; ROESY, rotating frame Overhauser enhancement spectroscopy; SDS, sodium dodecyl sulphate; TLR, Toll-like receptor; TOCSY, total correlation spectroscopy; TOF, time-of-flight, Xcc, Xanthomonas campestris pv. campestris.

12


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/


FIGURE LEGENDS

Figure 1. The 1H NMR spectrum and the primary structure of OS1 obtained by full de-acylation of LOS from Xcc. Abbreviations, P is phosphate group.

Figure 2. Overlap of anomeric sections of a) TOCSY (black) and NOESY (grey) spectra and b) HMBC (black) and HSQC (grey) spectra of OS1 product in which interresidual correlations are shown. Capital letters are as denoted in Fig. 1. The whole spectra are also visible in Supplemental Data.

Figure 3. a) section of the 1H NMR spectrum of OS2 obtained by de-O-acylation of LOS. Capital letters refer to each identified spin system as described in the text and denoted as in Table 2. b) The 31P NMR spectrum. At 16.2 ppm the phosphoramide 31P signal is visible.

Figure 4. The 31P-1H HSQC spectrum of OS2 and the inner core oligosaccharide structure of OS2 as deduced by NMR data. Residue D is linked either by a galacturonyl-phosphate substituent (D-X) or by a phosphoramide ring (D-Y).

Figure 5. The most significant ion peaks of the MALDI-TOF mass spectrum of intact LOS from Xanthomonas campestris pv. campestris. The core oligosaccharides ions are present as multiple ion peaks due to neutral loss of CO2 (∆m/z = 44) from the acidic sugar Kdo.

Figure 6. The complete structure of the LOS from Xanthomonas campestris pv. campestris. The dotted lines indicate non stoichiometric substitutions. Dotted methyl groups on fatty acids are present as possible single substitution.

Figure 7. a, b: Prevention of the hypersensitive response (HR) in Arabidopsis thaliana accession Colombia (Col-O) leaves. The leaves were examined both macroscopically and microscopically after tryphan blue staining (insets). In the leaf not pre-treated with LOS before Pst/AvrRpm1 inoculation macroscopic symptoms of HR were observed and the cells took up the trypan blue stain, indicative of dying cells (right leaf). In the LOS pre-inoculated leaf HR was prevented and no cell death occurred (left leaf). c: Growth of the virulent Pst in Col-O leaves with (●) or without pre-treatment with LOS (o) before bacterial inoculation.

13


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

Molecular basis of elicitation of plant defenses Table 1. Monosaccharides and fatty acids components of LOS, OS1 and OS2. All monosaccharides are in D configuration and fatty acids with R configuration. In the Table are also present the diagnostic ion fragments observable by GLC/MS analysis of carbohydrate (as partially methylated alditol acetate derivatives) and of fatty acids (as O-methyl ester derivatives).

Assignment Characteristic ion

peaks (m/z) LOS OS1 OS2

6-substituted-GlcNp 117, 159, 189, 233 x x x

5-substituted-Kdop 72, 146, 206, 394 x x x

4-substituted-Manp 118, 233 x x x

terminal-GalpA 118, 162, 163, 207 x ― x

terminal-Glcp 118, 161, 162, 205 x x x

3-substituted-Glcp 118, 161, 234, 277 x x x

3-substituted-Manp 118, 161, 234, 277 x x x

terminal-Manp 118, 161, 162, 205 x x x

10:0 (3-OH) 103, 184, 201 x ― ―

Iso and iso-ante 11:0 (3-OH) 103, 198, 215 x ― ―

12:0 (3-OH) 103, 212, 229 x ― x

Iso and iso-ante 13:0 (3-OH) 103, 226, 243 x ― x

10:0 74, 87, 155, 186 x ― ―

11:0 74, 87, 169, 200 x ― ―

Legend: x = present; ― = absent

14


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/


Table 2. 1H, 13C (italic) and 31P (bold) NMR chemical shifts (ppm) of deacylated core-lipid A backbone (OS1) of LOS from Xanthomonas campestris pv. campestris. Chemical shifts are relative to internal acetone and external aq. 85% phosphoric acid.

Chemical shift δ (1H/13C/31P)

Unit 1 2 3 4 5 6 7 8 A 5.65 3.35 3.91 3.80 4.19 4.26/3.90

6-α-GlcN I 92.0 55.7 71.5 71.0 73.8 70.8 3.11

B 5.28 4.20 3.92 3.79 3.77 3.70/3.90

3-α-Man 102.5 70.0 79.5 70.5 74.0 61.2 C 5.13 4.07 3.88 3.66 3.77 3.83/3.70

t-α-Man 103.7 71.4 70.0 68.3 75.0 61.2 D 5.09 4.18 4.56 4.07 4.13 3.84-3.93

4-α-Man 101.8 76.8 74.5 76.5 75.6 61.0 3.45

E 5.00 3.04 3.85 3.72 3.71 3.58/3.69

6-β-GlcN II 100.7 59.7 73.9 75.0 75.0 63.7 4.52

F 4.58 3.33 3.64 3.38 3.47 3.72/3.91

3-β-Glc 103.6 74.6 77.3 71.1 76.8 62.1 F’ 4.58 3.33 3.50 3.38 3.47 3.72/3.91

t-β-Glc 103.6 74.6 75.9 71.1 76.8 62.1 G 1.87/2.08 4.18 4.21 3.72 3.83 3.74/3.93

5-α-Kdo 174.9 100.3 36.0 67.0 76.6 71.0 70.6 64.7

15


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

Molecular basis of elicitation of plant defenses Table 3. 1H, 13C (italic) and 31P (bold) chemical shifts (ppm) of de-O-acylated core-lipid A backbone (OS2) of LOS from Xanthomonas campestris pv. campestris. The sample was solved in 1% deuterated SDS solution with 5 µl 32% NH4OH (pD 9.5). Chemical shifts are relative to internal acetone and external aq. 85% phosphoric acid.

Chemical shift δ (1H/13C/31P) Unit 1 2 3 4 5 6 7 8 M 5.54 3.69 3.81 4.20 4.49 -

t-α-GalA 95.3 68.2 69.3 69.6 71.9 174.0 -1.9

H 5.49 3.76 3.82 4.17 4.34 -

t-α-GalA 95.7 67.5 69.5 70.7 73.3 174.0 -0.9

A 5.26 3.66 3.82 3.58 4.31 4.00/3.76

6-α-GlcN 93.1 54.1 71.3 68.9 73.4 68.0 2.94

B 5.11 4.05 3.77 3.60 3.70 3.58

3-α-Man 101.3 70.2 77.9 66.1 73.0 62.4 C 4.97 4.24 3.71 3.48 3.59 3.50/3.60

t-α-Man 102.4 69.4 70.2 67.1 73.4 62.4 F 4.416 3.16 3.50 3.20 3.32 3.73

3-β-Glc 102.4 73.8 76.0 70.3 76.0 60.8 F’ 4.416 3.16 3.50 3.20 3.32 3.73

t-β-Glc 102.4 73.8 76.0 70.3 76.0 60.8 G 2.08/1.81 4.50 4.32 3.85 4.04 3.68

5-α-Kdo 174.1 100.9 34.4 70.9 73.8 73.1 70.3 63.3 -0.9

E 4.45 3.73 3.66 3.67 3.54 3.73/3.61 6-β-GlcN 102.4 55.2 73.5 74.3 74.0 61.1

5.23 D-X 4.97 3.91 4.33 4.17 3.83 3.73/3.53

4-α-Man 100.5 68.6 73.7 74.3 73.0 62.4 -1.9

D-Y 5.24 4.59 4.49 4.00 3.69 3.90/3.70 4-α-Man 97.6 76.1 77.5 75.1 73.5 61.9

16.2 16.2

16


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

Molecular basis of elicitation of plant defenses Table 4. Ability of Xcc LOS, Xcc core and Xcc lipid A to suppress the Hypersensitive Response (HR)

caused by Pst AvrRPM1 in Arabidopsis thaliana.

Treatment Xcc LOS Xcc core Xcc lipidA Dephosphorylated

LOS H2O

HR

―

―

―

X

X

X : present; ― : 100% absent HR

Table 5. Ability of Xcc LOS, Xcc core and Xcc lipid A to induce PR-1and PR-2 gene expression in

Arabidopsis thaliana

Treatment Xcc LOS Xcc core Xcc lipidA

Time after treatment PR1 PR2 PR1 PR2 PR1 PR2

4 hours +1.8ns +3.0*** +33.7*** +11.2*** +2.5ns +1.1***

12 hours +130.9*** +53.4*** +143.9*** +40.9*** +10.9*** +9.6***

20 hours +5.4*** +1.2ns +16.8*** +6.1*** + 8.1*** +11.5***

24 hours +1192.0*** +56.1*** +11.2*** +11.5*** +448.0*** +57.8***

+: fold up regulated compared to water treated tissue, after normalization to 18SrRNA; ns: not significant; *** = P<0,001

17


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/


igure 1.

F/F’ D G E Aβ-D-Glc(1-4)-α-D-Man(1-5)-α-D-Kdo(2-6)-β-D-GlcN(1-6)-α-D-GlcN

3R 3P 4P 1P C B

F’ R= H; F R= α-D-Man(1-3)-α-D-Man(1-

A

6.0 5.0 4.0 3.0 2.0 ppm

B G

F/F’

C

DE

F/F’ D G E Aβ-D-Glc(1-4)-α-D-Man(1-5)-α-D-Kdo(2-6)-β-D-GlcN(1-6)-α-D-GlcN

3R 3P 4P 1P C B

F’ R= H; F R= α-D-Man(1-3)-α-D-Man(1-

A

6.0 5.0 4.0 3.0 2.0 ppm

B G

F/F’

C

DE

A

6.0 5.0 4.0 3.0 2.0 ppm

B G

F/F’

C

DE

F

18


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/


4.74.95.15.35.55.7 ppm

70

80

90

100

ppm

4.74.95.15.35.55.7 ppm

3.1

3.3

3.5

3.7

3.9

4.1

4.3

4.5

B1-F3

C1-B3

E1-A6D1-G5

F1-D4

B1-F3C1-B3

D1-G5

E1-A6

F1-D4

a) NOESY and TOCSY

b) HSQC and HMBC

A1B1

C1D1 E1

F1

4.74.95.15.35.55.7 ppm

70

80

90

100

ppm

4.74.95.15.35.55.7 ppm

3.1

3.3

3.5

3.7

3.9

4.1

4.3

4.5

B1-F3

C1-B3

E1-A6D1-G5

F1-D4

B1-F3C1-B3

D1-G5

E1-A6

F1-D4

a) NOESY and TOCSY

b) HSQC and HMBC

A1B1

C1D1 E1

F1

Figure 2

19


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/


16 14 12 10 8 6 4 2 0 -2 ppm

b

M

5.5 5.0 4.5 4.0 3.5 ppm

A

H

B

D-YC

E F/F’

a

D-X

16 14 12 10 8 6 4 2 0 -2 ppm

b

16 14 12 10 8 6 4 2 0 -2 ppm16 14 12 10 8 6 4 2 0 -2 ppm

b

M

5.5 5.0 4.5 4.0 3.5 ppm

A

H

B

D-YC

E F/F’

a

D-X

M

5.5 5.0 4.5 4.0 3.5 ppm

A

H

B

D-YC

E F/F’

a

D-X

Figure 3.

20


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/


O

O

P

HO

O

HOOH

COO HHO

OR=

OP

O

NH 2OR,R '=

OO

O

HOH2C R '

OCOO H

OH

O

CH 2OH

R

OO

PO OH

HO

OH

COO HOH

O

R'= OH

H

M

D-X or D-Y

G

- 2

- 6

2

6

10

14

18

ppm

H -1 H H -4 G

H -1 M H -3 D - X

H -4 E

5.6 5.2 4.8 4.4 4.0

H - 3 D - YH -2 D -Y

H -1 D -Y

H -1 A

H -4 D -Y

H - 2 H

H -2 M

O

O

P

HO

O

HOOH

COO HHO

OR=

OP

O

NH 2OR,R '=

OO

O

HOH2C R '

OCOO H

OH

O

CH 2OH

R

OO

PO OH

HO

OH

COO HOH

O

R'= OH

H

M

D-X or D-Y

G

O

O

P

HO

O

HOOH

COO HHO

OR=

OP

O

NH 2OR,R '=

OO

O

HOH2C R '

O

O

P

HO

O

HOOH

COO HHO

OR=

OP

O

NH 2OR,R '=

OO

O

HOH2C R '

OCOO H

OH

O

CH 2OH

R

OO

PO OH

HO

OH

COO HOH

O

R'= OH

H

M

D-X or D-Y

G

- 2

- 6

2

6

10

14

18

ppm

H -1 H H -4 G

H -1 M H -3 D - X

H -4 E

5.6 5.2 4.8 4.4 4.0

H - 3 D - YH -2 D -Y

H -1 D -Y

H -1 A

H -4 D -Y

H - 2 H

H -2 M- 2

- 6

2

6

10

14

18

ppm

H -1 H H -4 G

H -1 M H -3 D - X

H -4 E

5.6 5.2 4.8 4.4 4.0

H - 3 D - YH -2 D -Y

H -1 D -Y

H -1 A

H -4 D -Y

H - 2 H

H -2 M

Figure 4.

21


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/


2000

4000

6000

8000

(m/z)

879.31054.8

1203.21378.8

1644.4

1767.7

2698.3

2847.6 3036.5

01000 1500 2000 2500 3000

lipid A

2Hex2Hex

2522.7

PQ

R S LQ

LPLSLR

L

L1

core oligosaccharideCounts

2000

4000

6000

8000

(m/z)

879.31054.8

1203.21378.8

1644.4

1767.7

2698.3

2847.6 3036.5

01000 1500 2000 2500 3000

lipid A

Lipo-oligosaccharide2Hex2Hex

2522.7

PQ

R S LQ

LPLSLR

L

L1


2000

4000

6000

8000

(m/z)

879.31054.8

1203.21378.8

1644.4

1767.7

2698.3

2847.6 3036.5

01000 1500 2000 2500 3000

lipid A

2Hex2Hex

2522.7

PQ

R S LQ

LPLSLR

L

L1


2000

4000

6000

8000

(m/z)

879.31054.8

1203.21378.8

1644.4

1767.7

2698.3

2847.6 3036.5

01000 1500 2000 2500 3000

lipid A

Lipo-oligosaccharide2Hex2Hex

2522.7

PQ

R S LQ

LPLSLR

L

L1


Figure 5.

22


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/


OP

O

HO

HO

OO

O

HOH2CO

OOH

HO

OCOOH

OH

O

CH2OH

HOH2C

R

P

OH

OHO

O

OHOO

OHHOH2CO

HOHO

OHHOH2C

O

NH

O

OHO

ONH

O

OO OO

(CH2)

(CH2)

O

(CH2)

OO

(CH2)

OHO HO

O

O

PO

OH

HO

OH

COOHOH

O

O

PO

NH2

OH

R=

HO

O

OH

HO

OOH

COOH

O

O

P

HO

1-2

1 or 3 1-2

1 or 3

Figure 6.

23


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/


Figure 7.

24


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/


Supplemental Data, Figure 1. The TOCSY spectrum of OS1 product. The main TOCSY correlations of

the anomeric proton of each carbohydrate spin system are shown. Capital letters are as denoted in Table

2.

ppm

2.02.53.03.54.04.55.05.5 ppm

2.0

2.5

3.0

3.5

4.0

4.5

5.0

5.5

TOCSY

A1 A2A4 A3A5A6

B2B1

C1

F1/F’1

B3 B4C2 C3 C4

E2E6E4E5

D1

E3E6

D3 D2D4D5

F/F’2F/F’3-F/F’6

ppm

2.02.53.03.54.04.55.05.5 ppm

2.0

2.5

3.0

3.5

4.0

4.5

5.0

5.5

TOCSY

A1 A2A4 A3A5A6

B2B1

C1

F1/F’1

B3 B4C2 C3 C4

E2E6E4E5

D1

E3E6

D3 D2D4D5

F/F’2F/F’3-F/F’6

E1

ppm

2.02.53.03.54.04.55.05.5 ppm

2.0

2.5

3.0

3.5

4.0

4.5

5.0

5.5

TOCSY

A1 A2A4 A3A5A6

B2B1

C1

F1/F’1

B3 B4C2 C3 C4

E2E6E4E5

D1

E3E6

D3 D2D4D5

F/F’2F/F’3-F/F’6

ppm

2.02.53.03.54.04.55.05.5 ppm

2.0

2.5

3.0

3.5

4.0

4.5

5.0

5.5

TOCSY

A1 A2A4 A3A5A6

B2B1

C1

F1/F’1

B3 B4C2 C3 C4

E2E6E4E5

D1

E3E6

D3 D2D4D5

F/F’2F/F’3-F/F’6

E1

25


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/


Supplemental Data, Figure 2. Section of the HMBC spectrum of OS1 product. The relevant scalar 3JC,H

correlations indicated in the spectrum are those necessary to achieve primary structure of oligosaccharides

in OS1 product. Capital letters are as denoted in Table 3.

HMBC

1.52.02.53.03.54.04.55.05.5

35

40

45

50

55

60

65

70

75

80

85

90

95

100

105

G3-G2

G3-G5

E1-A6D1-G5

F1-D4 C1-B3

B1-F3C1-B3

E1-A6G3-G4

G3-G5F1-D4

D1-G5

ppm

ppm

HMBC

1.52.02.53.03.54.04.55.05.5

35

40

45

50

55

60

65

70

75

80

85

90

95

100

105

G3-G2

G3-G5

E1-A6D1-G5

F1-D4 C1-B3

B1-F3C1-B3

E1-A6G3-G4

G3-G5F1-D4

D1-G5

ppm

ppm

26


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

Mari-Anne Newmann and Michelangelo ParrilliAlba Silipo, Antonio Molinaro, Luisa Sturiale, J. Maxwell Dow, Gitte Erbs, Rosa Lanzetta,

campestrisThe elicitation of plant innate imunity by lipooligosaccharide of xanthomonas

published online July 27, 2005J. Biol. Chem.

10.1074/jbc.M506254200Access the most updated version of this article at doi:

Alerts:

When a correction for this article is posted•

When this article is cited•

to choose from all of JBC's e-mail alertsClick here

Supplemental material:

http://www.jbc.org/content/suppl/2005/08/02/M506254200.DC1


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/lookup/doi/10.1074/jbc.M506254200

http://www.jbc.org/cgi/alerts?alertType=citedby&addAlert=cited_by&cited_by_criteria_resid=jbc;M506254200v1&saveAlert=no&return-type=article&return_url=http://www.jbc.org/content/early/2005/07/27/jbc.M506254200.citation

http://www.jbc.org/cgi/alerts?alertType=correction&addAlert=correction&correction_criteria_value=early/2005/07/27/jbc&saveAlert=no&return-type=article&return_url=http://www.jbc.org/content/early/2005/07/27/jbc.M506254200.citation

http://www.jbc.org/cgi/alerts/etoc

http://www.jbc.org/content/suppl/2005/08/02/M506254200.DC1

http://www.jbc.org/

Molecular basis for the elicitation of plants innate ... · Molecular basis of elicitation of plant defenses cruciferous crops that is of worldwide importance. Xcc can also infect

Documents