Logo of jbcAbout JBCASBMBSubmissionsSubscriptionsContactJBCThis Article
J Biol Chem. May 20, 2011; 286(20): 17560–17568.
Published online Mar 30, 2011. doi:  10.1074/jbc.M111.223644
PMCID: PMC3093830

The Structure of the Elicitor Cerato-platanin (CP), the First Member of the CP Fungal Protein Family, Reveals a Double ψβ-Barrel Fold and Carbohydrate Binding*An external file that holds a picture, illustration, etc.
Object name is sbox.jpg

Abstract

Cerato-platanin (CP) is a secretion protein produced by the fungal pathogen Ceratocystis platani, the causal agent of the plane canker disease and the first member of the CP family. CP is considered a pathogen-associated molecular pattern because it induces various defense responses in the host, including production of phytoalexins and cell death. Although much is known about the properties of CP and related proteins as elicitors of plant defense mechanisms, its biochemical activity and host target(s) remain elusive. Here, we present the three-dimensional structure of CP. The protein, which exhibits a remarkable pH and thermal stability, has a double ψβ-barrel fold quite similar to those found in expansins, endoglucanases, and the plant defense protein barwin. Interestingly, although CP lacks lytic activity against a variety of carbohydrates, it binds oligosaccharides. We identified the CP region responsible for binding as a shallow surface located at one side of the β-barrel. Chemical shift perturbation of the protein amide protons, induced by oligo-N-acetylglucosamines of various size, showed that all the residues involved in oligosaccharide binding are conserved among the members of the CP family. Overall, the results suggest that CP might be involved in polysaccharide recognition and that the double ψβ-barrel fold is widespread in distantly related organisms, where it is often involved in host-microbe interactions.

Keywords: Carbohydrate, Fungi, NMR, Protein Stability, Protein Structure, Cerato-platanin, Expansin, Fungal Elicitor

Introduction

Plants are often infected by microbes with diverse lifestyles, and to defend themselves, they have developed a two-layered innate immune system (1). The first line provides a basal defense against all potential pathogens and relies on the recognition either of conserved pathogen-associated molecular patterns (PAMPs),7 including flagellin, peptidoglycans, chitin, and β-glucan, some of the proteins relevant for the life cycle of the pathogen (2), or of an array of cell wall-degrading enzymes not necessarily involved in pathogenicity (3). This defense line leads to cell wall alterations, deposition of callose, induction of reactive oxygen species synthesis, activation of MAPK cascades, and accumulation of defense-related proteins (4).

Nevertheless, some pathogens may succeed in suppressing this primary defense mechanism. Thus, during evolution, plants have developed a second recognition system based on the perception of the effectors by R proteins, with subsequent activation of an effector-triggered immunity, which often ends up in a localized programmed cell death response (57).

Phytopathogenic fungi and oomycetes secrete several non-catalytic proteins (generally characterized by low molecular weight and by the presence of two to four disulfide bridges) that belong to specific protein families: elicitin, hydrophobin, NIP-1, and cerato-platanin (CP). Some members of these families have been proposed to have PAMP activity, e.g. cryptogein, INF1 elicitin, and the Pep1 and Stp1 hydrophobins (810).

The members of the CP protein family are secreted by a number of phytopathogenic fungi, animal pathogens, and non-pathogenic fungal species (supplemental Fig. S1). These proteins have been shown to act either as virulence factors and effectors, such as Snodprot1 from Phaeosphaeria nodorum, Sp1 from Leptosphaeria maculans, and MpCP1 from Moniliophthora perniciosa (1113), or either as elicitors and PAMPs, such as the proteins Sm1 and Epl1 from the soil-borne Trichoderma virens and Hypocrea atroviridis, respectively (14, 15). In particular, Sm1 elicits the production of reactive oxygen species and triggers the expression of several defense-related genes in plants, confirming the capacity of this biocontrol fungus to induce resistance, whereas Epl1 is the major protein found in the secretome of the biocontrol fungal strain H. atroviridis (1416). Moreover, Magnaporthe grisea expresses a MgSM1-CP-like protein that is able to induce a hypersensitive response in leaves and enhanced disease resistance against Botrytis cinerea and Pseudomonas syringae when transiently expressed in Arabidopsis (17).

In addition, members of the CP family play a major role in animal-fungus interaction, where they exhibit allergenic properties and induce strong immunological reactions. This is the case of the Coccidioides-specific antigen from Coccidioides immitis and the Aspf13 antigen from Aspergillum fumigatus, both causative agents of human lung diseases (18, 19), and of the immunomodulatory protein Aca1 from the medicinal fungus Antrodia camphorata (20).

The first member of the CP family is the CP protein from Ceratocystis platani, the causal agent of the canker stain disease of Platanus acerifolia (21, 22). CP plays a role in host-fungus interaction because it induces cell death, phytoalexin synthesis, and overexpression of defense-related genes (2325). For this reason, CP and the orthologous protein cerato-populin from Ceratocystis populicola, which triggers a CP-like reaction in host-microbe interaction (26, 27), have been considered PAMPs. CP can be found in the cell wall of ascospores, hyphae, and conidia, suggesting that the protein plays a role in forming the fungal cell wall (28). This last observation supports the notion that PAMPs are generally surface-associated molecules that cannot be easily altered without deeply interfering with the fungus fitness (10).

Until now, no information on the structure of CP or other CP-like proteins was available. In fact, only few examples of fungal PAMPs have been structurally characterized (10). Here, we report the NMR-derived solution structure of the first member of the CP family, the CP protein. CP has a globular fold containing two α-helices and six β-strands that form a six-stranded double ψβ-barrel, a motif remarkably similar to the those present in endoglucanases, domain I of expansins, and the defense protein barwin. Because of these similarities, the oligosaccharide-binding properties and the glycosidase activities of CP have been explored.

EXPERIMENTAL PROCEDURES

Expression and Purification of Cerato-platanin

The recombinant CP protein was expressed in Pichia pastoris and purified by reverse-phase liquid chromatography as described previously (29). The recombinant CP protein used in this work presents an additional 9-residue N-terminal stretch (EEGVSLEKR) because, under our experimental conditions, the signal peptide used for extracellular expression turned out not to be cut at the expected site. However, the protein shows the same biological activity and the same CD and 1H NMR spectra of native CP (29).

For isotopic labeling, CP was expressed in minimal medium supplemented with [15N]ammonium sulfate and, in the case of double-labeled proteins, with [13C]glycerol and [13C]methanol as described previously (30). The purified protein was estimated to be ~95% pure by SDS-PAGE, and MALDI-TOF mass spectrometry confirmed that it was 100% 13C- and 15N-isotopically labeled.

NMR Measurements

NMR samples contained 0.7 mm double-labeled [13C,15N]CP in 10 mm phosphate buffer (pH 5.8), 0.05% NaN3, and 5 or 100% D2O (Cambridge Isotope Laboratories, Inc.). Two- and three-dimensional NMR experiments were carried out at 20 °C on 500- and 600-MHz Varian Inova AS spectrometers (National Laboratory of Biosciences), a 600-MHz Varian Inova AS spectrometer (Centro Interfacoltà Misure, University of Parma), and 700- and 400-MHz Bruker AVANCE spectrometers (University of Rome “Tor Vergata”). All NMR spectrometers were equipped with a z pulsed field gradient unit and a triple resonance probe. Sequence schemes employing pulsed field gradients were used to achieve suppression of the solvent signal and spectral artifacts. Selective pulses to cover aliphatic, aromatic, or carbonyl 13C nuclear spectral zones were obtained by adiabatic pulse modulation. Quadrature detection in the indirectly detected dimensions was obtained by the hybrid States-TPPI (times proportional phase increment) method or the sensitivity-enhanced Echo-Antiecho combination. Linear prediction was applied to extend the indirect 13C-detected dimension. Direct and indirect dimensions were normally apodized using 90°-shifted squared sine-bell functions (for 13C- and 15N-edited dimensions) or Lorentzian-to-Gaussian functions (for the 1H dimension), followed by zero filling and Fourier transform. The NMR data were processed on Silicon Graphics workstations with NMRView/NMRPipe software (31, 32).

NOESY Experiments

NOE-derived distance restraints were obtained from three-dimensional 1H-15N NOESY (33) and three-dimensional 1H-13C NOESY, separately optimized for aliphatic and aromatic side chains (34) collected with 100-ms mixing times. Two-dimensional NOESY was performed to assign the peak at 11.40 ppm, which turned out to be the hydroxyl proton of Tyr-9, which is hydrogen-bonded to the side chain carboxyl of Asp-77.

Residual Dipolar Constant (RDC) Measurements

Five distinct sets of RDCs were measured: 1JHN-N, 3JHα-Cα, 1JCα-C′, 1JN-C′, and 2JHN-C′. Anisotropic data were collected by dissolving the protein directly in 14 mg/ml liquid crystalline Pf1 (Asla Lab) as the alignment medium. To decrease the signal width, we added NaCl up to 150 mm. 3JHα-Cα couplings were obtained from the 1H-coupled (F1) version of the HACACO experiment (35). 1JN-C′ and 2JHN-C′ were measured from 1H-15N heteronuclear single quantum coherence spectra with only Cα decoupling (36). 1JHN-N RDCs were measured by an in-phase/anti-phase heteronuclear single quantum coherence (HSQC) experiment (37). 1JCα-C′ RDCs were measured in a modified version of the J-module HN(COCA) experiment (38, 39).

Diffusion Experiments

Diffusion-ordered spectroscopy experiments were carried out using the BPPSTE (bipolar pulse pairs stimulated echo) method (40). The duration of the total diffusion phase-encoding gradient pulse was 2 ms, the diffusion delay was 0.05 s, and the minimum gradient strength was set to 0.3 gauss/cm. Diffusion coefficients were measured using 0.2 mm CP in 10 mm phosphate buffer (pH 5.8), and 0.02% dioxane was used as a radius standard (41).

Structure Calculation

Structure calculation was performed with XPLOR-NIH (42) using a simulated annealing protocol. Calculations were carried out over residues 1–120, excluding the first 9 N-terminal residues that do not belong to the native protein (see “Expression and Purification of Cerato-platanin” above) and that turned out to be unstructured (see supplemental Fig. S2). The simulated annealing was driven by NMR-derived constraints during the entire calculation, with force constraints and weighting factors optimized to increase the score of low energy structures. A set of 3203 interproton distance restraints was used, subdivided into three groups: strong (1.8–2.4 Å), medium (1.8–4.0 Å), and weak (1.8–5.0 Å). Because of very weak cross-peaks detected in the 15N-edited NOESY spectrum, we included also a restraint comprising 6.0–3.0 Å. Backbone dihedral restrains ([var phi] and ψ angles) were derived from chemical shift analyses using the program TALOS (43). Backbone hydrogen bonds were recognized by evaluating the spatial relationship of amide protons with potential acceptors in the initial structures produced without the use of hydrogen bond constraints. In addition, specific side chain hydrogen bonds were clearly identified using the WHAT IF server (44) with structures generated in later stages of the structural determination and were employed as additional restraints in the last cycle of minimization. Five distinct sets of backbone RDCs (84 1JHN-N, 71 3JHα-Cα, 60 1JCα-C′, 57 1JN-C′, and 52 2JHN-C′) were included. Fitting the observed RDCs to a preliminary structure, using the singular value decomposition method (45) with the PALES program (46), provided an initial guess of the magnitude and orientation of the molecular alignment tensor. Subsequent optimization during the simulated annealing process led to the following final values for the N-H normalized magnitude of the RDC tensor and rhombicity, respectively: DaNH = 5.55 Hz and R = 0.625. A complete cross-validation of RDCs was conducted by carrying out a series of simulated annealing calculations, each one lacking 10% of RDCs randomly chosen from the whole data set. The missing RDCs were back-calculated to evaluate how well each RDC could be predicted; an average Rfree of 33.6% was obtained (47). An ensemble of the 20 lower energy target function structures (from a total of 1000), in which the number of residues in disallowed regions of the Ramachandran plot is 0.0%, was chosen to represent the protein solution structure and, together with the constraint list, was deposited in the Protein Data Bank (2KQA). The programs AQUA and PROCHECK (48) were used to analyze the structures. Table 1 provides the structural statistics.

TABLE 1
Experimental restraints and structural statistics for the 20 lower energy structures of the CP protein

Glycosidase Assay

Endo-1,3-β-glucanase, polygalacturonase, and cellulase activities were determined using laminarin, polygalacturonic acid, and carboxymethylcellulose as substrates, respectively, according to previously published methods (4951). The released reducing sugars were detected using the PAHBAH procedure (52) with glucose as a standard. Enzymes (laminarinase from Trichoderma sp., pectinase from Aspergillus niger, and cellulase from Trichoderma reseei) and substrates blanks were also included. Chitinase activity was detected at 575 nm using chitin azure as a substrate. Chitinase from Trichoderma viride was used as a positive control.

N-Acetylglucosamine Binding

The binding of the β-(1,4)-tetramer of N-acetyl-d-glucosamine (Sigma) was studied after the addition of 0, 0.1, 2, 4, and 8 mm oligosaccharide to a 0.04 mm solution of 15N-labeled CP. To evaluate the binding of larger oligosaccharides, a chito-oligosaccharide mixture containing oligomers from the dimer to hexamer of β-(1,4)-linked N-acetyl-d-glucosamine (Seikagaku Biobusiness Corp.) was used. The mixture (containing 5 mg of each oligosaccharide) was dissolved in 200 μl of H2O, and aliquots of the solution were added to a 0.1 mm solution of 15N-labeled CP to obtain final concentrations of the hexamer of 0, 0.5, 1.0, 1.5, and 2.0 mm. The chemical shift perturbation in the 1H-15N HSQC spectrum upon oligosaccharide addition was evaluated by the weighted average chemical shift change (53): Δδav = ((ΔδNH2 + ΔδN2/25)/2)1/2 (ppm).

RESULTS

CP Presents a Double ψβ-Barrel Fold

The three-dimensional structure of CP was solved by triple resonance NMR methods using >3400 distance and dihedral angle restraints together with 324 RDCs in iterative cycles of structure calculations. The quality of the CP structure is highlighted by the structural statistics (Table 1). The protein region Val-1–Asn-117 is defined with high precision as reflected by the low root mean square deviations from the mean structure for the backbone atoms (±0.40 Å) and for all heavy atoms (±0.72 Å) (Fig. 1A and Table 1). A quality check using PROCHECK-NMR (48) indicated that 85.5% of the residues display dihedral angles in the allowed regions of the Ramachandran plot (Table 1).

FIGURE 1.
Three-dimensional structure of CP. A, stereo view of the superimposition of the backbone atoms of the 20 lower energy models. B, ribbon representation of the structure. C, topological representation of CP.

The solution structure of CP reveals a globular fold containing two α-helices and six β-strands forming a double ψβ-barrel (Fig. 1, B and C). The six-stranded β-barrel has a shear number of 10. The structure is consistent with CD measurements that indicate a predominant β-sheet structure and some α-helical elements (supplemental Fig. S3). In addition, the CP hydrodynamic radius, estimated by diffusion measurements, is ~19.5 Å, in agreement with the value expected for a monomeric and globular protein of that size (54). The long and superficial ψ-loop between β1 and β2 is rigid. In fact, 15N relaxation data show quite homogeneous T1 and T2 values and an average 1H-15N NOE of 0.85, indicating reduced internal motions (supplemental Fig. S2). The only slightly flexible region of the protein comprises the sequence Ser-96–Arg-101, which exhibits relatively low 1H-15N NOE and long T2 values (supplemental Fig. S2). The last 3 C-terminal residues are unstructured and highly mobile (supplemental Fig. S2). Interestingly, CD experiments showed that the protein is quite stable in a wide pH range and exhibits a somewhat high unfolding temperature of ~76 °C (supplemental Fig. S4).

Structures Related to CP

Protein sequence alignments show that CP shares significant similarities only with fungi proteins that are currently classified as members of the CP family (supplemental Fig. S1). However, a search for structurally related proteins using the Dali server (55) identified various proteins involved in polysaccharide recognition and modification (Table 2), including plant (56) and bacterial (57) expansins, endoglucanases (56, 5860), and the plant defense protein barwin (61). Although these proteins share a moderately low sequence homology with CP, they display relatively high Dali Z-scores ranging from 9.5 to 5.2 (Table 2). Superposition of CP with these proteins highlights the significant structural match especially in the β-region (Fig. 2). Other less structurally related proteins, with Z-scores in the range of 3.6 to 3.2, include the lytic transglycosylases (LTs) (62).

TABLE 2
Results of the pairwise superposition of CP with protein representatives of families bearing a double ψβ-barrel fold using the Dali server
FIGURE 2.
Comparison of the CP structure with other proteins containing a double ψβ-barrel fold. Shown is a diagram representation of the superposition of CP (green) with proteins that exhibit high Z-score values: the expansin Phlp1 (orange), the ...

CP Does Not Present Glycosidase Activity

Because CP is structurally related to endoglucanase and LT proteins, we tested whether it has glycoside hydrolase activity in in vitro assays. No cellulase, endo-1,3-β-glucanase, polygalacturonase, or chitinase activity could be detected even when CP was used at a high protein concentration relative to the positive controls (data not shown).

CP Binds Oligosaccharides

As shown in Table 2, the proteins most structurally related to CP display carbohydrate-binding properties; are involved in polysaccharide recognition; and in some cases, exhibit hydrolytic activity (56, 61). Therefore, we tested the oligosaccharide-binding capability of CP by measuring the chemical shift perturbation of the amide protons induced by oligosaccharides. Knowing that small oligosaccharides are able to bind barwin, we assayed the interaction of the tetramer of N-acetylglucosamine (GlcNAc-4) with CP. After the addition of increasing amounts of GlcNAc-4, we observed changes in the chemical shifts of a limited number of residues. The alterations were proportional to the quantity of added ligand, thus indicating a fast exchange between the apo- and holo-forms of the protein (see examples in supplemental Fig. S5A). Although the affinity of CP binding to GlcNAc-4 is weak, with an estimated Kd in the range of 10–100 μm, the weighted average chemical shift changes after the addition of >25 eq of GlcNAc-4 highlight three main regions involved in the oligosaccharide tetramer binding (Fig. 3A). These regions preferentially encompass portions of the loops: Ser-17–Ala-29 located in the β1-β2 loop, Gly-51–Ser-54 in the β2-β3 loop, and Asp-77–Gly-81 comprising the last part of the β4-strand and the β4-β5 loop. To further map the boundary of the CP oligosaccharide-binding region, 15N-isotopically labeled CP was titrated with a mixture of oligosaccharides containing dimers, trimers, tetramers, pentamers, and hexamers of N-acetylglucosamine. Fig. 3B shows that, in addition to the oligosaccharide-binding regions identified with GlcNAc-4, the α1-β6 loop, Ala-99–Val-102, and the β1-strand, carrying the conserved SYD motif, were also affected. In particular, the signal belonging to Ser-4 in the β1-strand disappeared after the addition of the oligosaccharide mixture, whereas it remained unaffected in the presence of GlcNAc-4 (supplemental Fig. S5, A and B, left panels). We interpret these results as an indication of the selective interaction of these regions with the longer GlcNAcs.

FIGURE 3.
Interaction of CP with polysaccharides. A and B, average chemical shift perturbation upon interaction with GlcNAc-4 and with a mixture of longer GlcNAcs, respectively. The results are plotted versus residue number (left panels) and mapped on the structure ...

Overall, the region involved in GlcNAc binding forms a flat and shallow groove on one face of the β-barrel, which is rich in polar (e.g. Asp-6, Asn-22, and Asp-53) and aromatic (e.g. Tyr-5 and Trp-52) residues suitable for sugar carbohydrate binding (Fig. 3B). Furthermore, all the residues involved in GlcNAc binding identified in this study are among the most conserved ones in the CP family (Fig. 3C).

DISCUSSION

Despite recent advances in the characterization of Avr plant pathogens and elicitor proteins, the limited number of structures available for these proteins has hampered the understanding of the molecular mechanism associated with the interaction of these effectors with their targets or cognate pattern recognition receptors. Indeed, to date, only a few pairs of fungal effectors/targets and PAMPs/pattern recognition receptors have been identified and structurally characterized (2).

CP Structure

Thermal unfolding measurements show that the protein is particularly stable: the unfolding temperature is up to 76 °C (supplemental Fig. S4B), although it is irreversible. On the other hand, the protein secondary structure is preserved in a wide pH range, 3–9 (supplemental Fig. S4A). This last feature, perhaps the result of the presence of two disulfide bonds resistant to pH but not to temperature, is expected to favor the efficient secretion and/or translocation of CP into the host cell.

The solution structure of the CP protein that we report here is the first one for members of the CP family. The protein presents a double ψβ-barrel fold remarkably similar to those found in plant and bacterial expansins, LTs, endoglucanases, formate dehydrogenase H, dimethyl-sulfoxide reductase, and aspartic proteinases (63) and in the plant defense protein barwin (61).

Expansins are found primarily in plants, where they play a critical role in cell enlargement and other developmental processes requiring cell wall loosening (56). On the other hand, endoglucanases are ubiquitous enzymes; catalyze the hydrolysis of cellulose to smaller oligosaccharides in numerous plants, bacteria, and fungi; and are present in 13 of 80 families of glycoside hydrolase enzymes (64). LTs cleave the β-(1,4)-glycosidic bond between N-acetylmuramic acid and GlcNAc in peptidoglycans, with the concomitant formation of a 1,6-anhydro-ring at the N-acetylmuramic acid residue.

Except for barwin, which is also the most similar protein in primary sequence (28% identity), the other proteins are larger than CP (Fig. 2). Expansins present two domains: the N-terminal domain, which constitutes the double ψβ-barrel, and the C-terminal domain, which forms an Ig-like β-sandwich. Endoglucanases have insertions in the β3-β4 and β4-β5 loops and extensions at the N and C termini. In addition, they frequently present a second domain, called the cellulose-binding domain, attached with a linker to the C terminus (65). This is not the case, however, in MeCel45A, which presents a single catalytic domain. LTs are generally multidomain proteins, showing a long insertion (~140 residues) in the β1-β2 loop, which constitutes a second domain, and an N-terminal extension. It is worth noting a significant fold similarity between CP and EXLX1, a two-domain expansin from Bacillus subtilis that binds to plant cell walls and to carbohydrates but lacks lytic activity against a variety of cell wall polysaccharides and peptidoglycan (57), functional features shared, at least in part, also by CP.

Rationale for CP Lack of Glycoside Hydrolase Activity and for Oligosaccharide-binding Capability

The absence of glycoside hydrolase activity can be rationalized by structurally comparing CP with endoglucanases and LTs. The superposition of the endoglucanase MeCel45A catalytic site with the corresponding region of CP is depicted in Fig. 4A. The enzymatic activity of MeCel45A depends on 2 residues, Asp-24 and Asp-132, which are located in the β1- and β5-strands and act as a catalytic base and acid, respectively (58). The corresponding residues in CP are Ile-8 and Asn-84, which are conserved among the members of the CP family and exhibit different chemical properties. Other MeCel45A residues potentially involved in catalysis, as suggested by their being conserved among endoglucanases, are Thr-20 and Tyr-22 (Fig. 4A), replaced in CP with the conserved residues Ser-4 and Asp-6, respectively, again, at least in the case of Tyr-22/Asp-6, with diverse chemical properties (57).

FIGURE 4.
Superposition of the active sites of the endoglucanase MeCel45A (A) and the LT EcMltA (B) with the corresponding region of CP. Residues that are important for catalysis are displayed as sticks.

Also the active site of LTs, presented here in the enzyme EcMltA, contains 2 aspartic acids as the main catalytic residues (Fig. 4B), which, as for the endoglucanase active sites, are located in β5 (Asp-308) and at the end of β4 (Asp-297). In the catalytic mechanism proposed for LTs (66), Asp-308 acts as an acid, and Asp-297, which is hydrogen-bonded to Tyr-101, acts as a base. In fact, Tyr-101 is supposed to shuttle a proton from Asp-297 to Asp-308.

This network of interactions observed in endoglucanases and LTs is only partially conserved in CP. In particular, in the CP structure, the substitution of an aspartic acid residue with Asn-84 by removing the carboxylic group might justify the disappearance of the hydrolase activity.

It is interesting to note, however, that Asp-77 and Tyr-9, both conserved in all CP family members, correspond to Asp-297 and Tyr-101 of EcMltA, respectively, whose side chains are expected to interact (Fig. 4B). The interaction between the Asp-77 and Tyr-9 side chains is clearly visible by NMR because of the reduced solvent exchange of the hydroxyl of the Tyr-9 side chain, which appeared at 11.4 ppm. In a two-dimensional NOESY spectrum, clear NOE cross-peaks between this proton and the two β-protons of Asp-77 could be observed, suggesting the presence of a hydrogen bond between them (data not shown).

Fig. 3C highlights that the residues conserved among the members of the CP family are involved in polysaccharide binding. This fact strongly supports the hypothesis that the biological function of proteins belonging to the CP family may be related to polysaccharide binding, this feature being also consistent with the location of the CP protein in the fungal cell wall (28). Overall, site-specific mutations with respect to endoglucanase and LT proteins seem to explain why CP has lost glycoside hydrolase activity while maintaining the capability to bind oligosaccharides.

The data presented in this study contribute to deeper knowledge of the role of CP in both fungus fitness and host-microbe interaction. Because CP is located in the cell wall of hyphae, conidia, and ascospores of C. platani (28), its similarity to expansins could be a sign of the involvement of the protein in cell wall remodeling and enlargement. Moreover, the ability to bind GlcNAc might imply a role of CP in host interaction.

Recognizing that chitin oligomers can act as PAMPs, eliciting basal immune responses in plants (2, 67), it has been recently reported that the virulence effector Ecp6 from the fungal pathogen Cladosporium fulvum sequesters β-(1,4)-poly-N-acetyl-d-glucosamine, released from the hyphal cell wall during infection, to prevent elicitation of host immunity (68). All considered, our data point to the possibility that, during plant-microbe interaction, CP is overexpressed and interacts with the host, acting as a typical PAMP (24, 25). The elicited plant immune responses would include also overexpression of plant cell wall-degrading enzymes, which will produce chitin oligomers (69). At this stage, we may envisage that the expected enhancement of the plant primary defense response induced by those oligosaccharides can be blocked by the CP ability to bind GlcNAc, albeit a bit more weakly compared with Ecp6.

In conclusion, we trust that the solution of CP three-dimensional structure has opened the way to unravel the multifaceted functions not only of this protein but also of the other members of the CP family. 1) Because CP oligosaccharide binding involves residues conserved throughout the CP family, it is reasonable to expect that the other family members are able to bind oligosaccharides. 2) The observation that CP and Ecp6 bind GlcNAcs with different affinities might explain why members of the same family can act either as elicitors or effectors and provide critical hints to understand the incompatibility/compatibility of some plant-pathogen interactions. In other words, it appears that we are disclosing a more widespread strategy of host immune response suppression by fungal pathogens and elicitors (68).

Supplementary Material

Supplemental Data:

*This work was supported in part by grants from the Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP) (NMR Project 99/11030-9, TAP 00/02026-7, Smolbnet 00/10266-8) and from the Ministero dell'Istruzione dell'Università e della Ricerca (MIUR) (PRIN 2005 20050717.5-002). Some of the CD and NMR experiments were carried out using equipment at the Centro Interfacoltà Misure, University of Parma.

An external file that holds a picture, illustration, etc.
Object name is sbox.jpgThe on-line version of this article (available at http://www.jbc.org) contains supplemental Figs. S1–S5 and additional references.

The atomic coordinates and structure factors (code 2KQA) have been deposited in the Protein Data Bank, Research Collaboratory for Structural Bioinformatics, Rutgers University, New Brunswick, NJ (http://www.rcsb.org/).

7The abbreviations used are:

PAMP
pathogen-associated molecular pattern
CP
cerato-platanin
RDC
residual dipolar constant
LT
lytic transglycosylase.

REFERENCES

1. de Wit P. J. G. M. (2007) Cell. Mol. Life Sci. 64, 2726–2732 [PubMed]
2. Zipfel C. (2009) Curr. Opin. Plant Biol. 12, 414–420 [PubMed]
3. Juge N. (2006) Trends Plant Sci. 11, 359–367 [PubMed]
4. Pitzschke A., Schikora A., Hirt H. (2009) Curr. Opin. Plant Biol. 12, 421–426 [PubMed]
5. Bent A. F., Mackey D. (2007) Annu. Rev. Phytopathol. 45, 399–436 [PubMed]
6. Boller T., He S. Y. (2009) Science 324, 742–744 [PMC free article] [PubMed]
7. Dodds P. N., Rathjen J. P. (2010) Nat. Rev. Genet. 11, 539–548 [PubMed]
8. Svozilová Z., Kasparovský T., Skládal P., Lochman J. (2009) Anal. Biochem. 390, 115–120 [PubMed]
9. Kim Y. T., Oh J., Kim K. H., Uhm J. Y., Lee B. M. (2010) Mol. Biol. Rep. 37, 717–727 [PMC free article] [PubMed]
10. De Wit P. J., Mehrabi R., Van den Burg H. A., Stergiopoulos I. (2009) Mol. Plant Pathol. 10, 735–747 [PubMed]
11. Wilson L. M., Idnurm A., Howlett B. J. (2002) Mol. Plant Pathol. 3, 487–493 [PubMed]
12. Jeong J. S., Mitchell T. K., Dean R. A. (2007) FEMS Microbiol. Lett. 273, 157–165 [PubMed]
13. Zaparoli G., Cabrera O. G., Medrano F. J., Tiburcio R., Lacerda G., Pereira G. G. (2009) Mycol. Res. 113, 61–72 [PubMed]
14. Djonovic S., Vargas W. A., Kolomiets M. V., Horndeski M., Wiest A., Kenerley C. M. (2007) Plant Physiol. 145, 875–889 [PMC free article] [PubMed]
15. Seidl V., Marchetti M., Schandl R., Allmaier G., Kubicek C. P. (2006) FEBS J. 273, 4346–4359 [PubMed]
16. Vargas W. A., Djonović S., Sukno S. A., Kenerley C. M. (2008) J. Biol. Chem. 283, 19804–19815 [PubMed]
17. Yang Y., Zhang H., Li G., Li W., Wang X., Song F. (2009) Plant Biotechnol. J. 7, 763–777 [PubMed]
18. Pan S., Cole G. T. (1995) Infect. Immun. 63, 3994–4002 [PMC free article] [PubMed]
19. Rementeria A., López-Molina N., Ludwig A., Vivanco A. B., Bikandi J., Pontón J., Garaizar J. (2005) Rev. Iberoam. Micol. 22, 1–23 [PubMed]
20. Hseu Y. C., Wu F. Y., Wu J. J., Chen J. Y., Chang W. H., Lu F. J., Lai Y. C., Yang H. L. (2005) Int. Immunopharmacol. 5, 1914–1925 [PubMed]
21. Pazzagli L., Cappugi G., Manao G., Camici G., Santini A., Scala A. (1999) J. Biol. Chem. 274, 24959–24964 [PubMed]
22. Pazzagli L., Pantera B., Carresi L., Zoppi C., Pertinhez T. A., Spisni A., Tegli S., Scala A., Cappugi G. (2006) Cell Biochem. Biophys. 44, 512–521 [PubMed]
23. Scala A., Pazzagli L., Comparini C., Santini A., Tegli S., Cappugi G. (2004) J. Plant Pathol. 86, 23–29
24. Fontana F., Santini A., Salvini M., Pazzagli L., Cappugi G., Scala A., Durante M., Bernardi R. (2008) J. Plant Pathol. 90, 295–306
25. Bernardi R., Baccelli I, Carresi L, Comparini C, Pazzagli L., Scala A. (2011) Forest Pathol. 41, in press
26. Pazzagli L., Zoppi C., Carresi L., Tiribilli B., Sbrana F., Schiff S., Pertinhez T. A., Scala A., Cappugi G. (2009) Biochim. Biophys. Acta 1790, 1334–1344 [PubMed]
27. Comparini C., Carresi L., Pagni E., Sbrana F., Sebastiani F., Luchi N., Santini A., Capretti P., Tiribilli B., Pazzagli L., Cappugi G., Scala A. (2009) Appl. Microbiol. Biotechnol. 84, 309–322 [PubMed]
28. Boddi S., Comparini C., Calamassi R., Pazzagli L., Cappugi G., Scala A. (2004) FEMS Microbiol. Lett. 233, 341–346 [PubMed]
29. Carresi L., Pantera B., Zoppi C., Cappugi G., Oliveira A. L., Pertinhez T. A., Spisni A., Scala A., Pazzagli L. (2006) Protein Expr. Purif. 49, 159–167 [PubMed]
30. Oliveira A. L., Pazzagli L., Pantera B., Cappugi G., Benedetti C. E., Spisni A., Pertinhez T. A. (2006) J. Biomol. NMR 36, 50. [PubMed]
31. Johnson B. A. (2004) Methods Mol. Biol. 278, 313–352 [PubMed]
32. Delaglio F., Grzesiek S., Vuister G. W., Zhu G., Pfeifer J., Bax A. (1995) J. Biomol. NMR 6, 277–293 [PubMed]
33. Zhang O., Kay L. E., Olivier J. P., Forman-Kay J. D. (1994) J. Biomol. NMR 4, 845–858 [PubMed]
34. Muhandiram D. R., Farrow N. A., Xu G., Smallcombe S. H., Kay L. E. (1993) J. Magn. Reson. 102, 314–321
35. Cicero D. O., Contessa G. M., Paci M., Bazzo R. (2006) J. Magn. Reson. 180, 222–228 [PubMed]
36. Wang Y. X., Marquardt J. L., Wingfield P., Stahl S. J., Lee-Huang S., Torchia D., Bax A. (1998) J. Am. Chem. Soc. 120, 7385–7386
37. Ottiger M., Delaglio F., Bax A. (1998) J. Magn. Reson. 131, 373–378 [PubMed]
38. McFeeters R. L., Fowler C. A., Gaponenko V. V., Byrd R. A. (2005) J. Biomol. NMR 31, 35–47 [PubMed]
39. Eliseo T., Ragona L., Catalano M., Assfalg M., Paci M., Zetta L., Molinari H., Cicero D. O. (2007) Biochemistry 46, 12557–12567 [PubMed]
40. Waldeck A. R., Kuchel P. W., Lennon A. J., Chapman B. E. (1997) Prog. Nucl. Magn. Reson. Spectrosc. 30, 39–68
41. Jones J. A., Wilkins D. K., Smith L. J., Dobson C. M. (1997) J. Biomol. NMR 10, 199–203
42. Schwieters C. D., Kuszewski J. J., Tjandra N., Clore G. M. (2003) J. Magn. Reson. 160, 65–73 [PubMed]
43. Cornilescu G., Delaglio F., Bax A. (1999) J. Biomol. NMR 13, 289–302 [PubMed]
44. Vriend G. (1990) J. Mol. Graph. 8, 52–56 [PubMed]
45. Losonczi J. A., Andrec M., Fischer M. W., Prestegard J. H. (1999) J. Magn. Reson. 138, 334–342 [PubMed]
46. Zweckstetter M., Bax A. (2000) J. Am. Chem. Soc. 122, 3791–3792
47. Clore G. M., Garrett D. S. (1999) J. Am. Chem. Soc. 121, 9008–9012
48. Laskowski R. A., Rullmann J. A., MacArthur M. W., Kaptein R., Thornton J. M. (1996) J. Biomol. NMR 8, 477–486 [PubMed]
49. Gueguen Y., Voorhorst W. G., van der Oost J., de Vos W. M. (1997) J. Biol. Chem. 272, 31258–31264 [PubMed]
50. Cervone F., Hahn M. G., De Lorenzo G., Darvill A., Albersheim P. (1989) Plant Physiol. 90, 542–548 [PMC free article] [PubMed]
51. Wood T. M., Bhat K. M. (1998) Methods Enzymol. 160, 87–112
52. York W. S., Darvill A. G., McNeil M., Stevenson T. I., Albersheim P. (1985) Methods Enzymol. 118, 3–40
53. Grzesiek S., Bax A., Clore G. M., Gronenborn A. M., Hu J. S., Kaufman J., Palmer I., Stahl S. J., Wingfield P. T. (1996) Nat. Struct. Biol. 3, 340–345 [PubMed]
54. Wilkins D. K., Grimshaw S. B., Receveur V., Dobson C. M., Jones J. A., Smith L. J. (1999) Biochemistry 38, 16424–16431 [PubMed]
55. Holm L., Park J. (2000) Bioinformatics 16, 566–567 [PubMed]
56. Sampedro J., Cosgrove D. J. (2005) Genome Biol. 6, 242. [PMC free article] [PubMed]
57. Kerff F., Amoroso A., Herman R., Sauvage E., Petrella S., Filée P., Charlier P., Joris B., Tabuchi A., Nikolaidis N., Cosgrove D. J. (2008) Proc. Natl. Acad. Sci. U.S.A. 105, 16876–16881 [PMC free article] [PubMed]
58. Davies G. J., Tolley S. P., Henrissat B., Hjort C., Schülein M. (1995) Biochemistry 34, 16210–16220 [PubMed]
59. Hirvonen M., Papageorgiou A. C. (2003) J. Mol. Biol. 329, 403–410 [PubMed]
60. Eriksson J., Malmsten M., Tiberg F., Callisen T. H., Damhus T., Johansen K. S. (2005) J. Colloid Interface Sci. 285, 94–99 [PubMed]
61. Ludvigsen S., Poulsen F. M. (1992) Biochemistry 31, 8783–8789 [PubMed]
62. Blackburn N. T., Clarke A. J. (2001) J. Mol. Evol. 52, 78–84 [PubMed]
63. Castillo R. M., Mizuguchi K., Dhanaraj V., Albert A., Blundell T. L., Murzin A. G. (1999) Structure 7, 227–236 [PubMed]
64. Henrissat B., Bairoch A. (1996) Biochem. J. 316, 695–696 [PMC free article] [PubMed]
65. Gilkes N. R., Henrissat B., Kilburn D. G., Miller R. C., Jr., Warren R. A. (1991) Microbiol. Rev. 55, 303–315 [PMC free article] [PubMed]
66. Powell A. J., Liu Z. J., Nicholas R. A., Davies C. (2006) J. Mol. Biol. 359, 122–136 [PubMed]
67. Lee C. G., Da Silva C. A., Lee J. Y., Hartl D., Elias J. A. (2008) Curr. Opin. Immunol. 20, 684–689 [PMC free article] [PubMed]
68. de Jonge R., van Esse H. P., Kombrink A., Shinya T., Desaki Y., Bours R., van der Krol S., Shibuya N., Joosten M. H., Thomma B. P. (2010) Science 329, 953–955 [PubMed]
69. Hématy K., Cherk C., Somerville S. (2009) Curr. Opin. Plant Biol. 12, 406–413 [PubMed]

Articles from The Journal of Biological Chemistry are provided here courtesy of American Society for Biochemistry and Molecular Biology

Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...

Links

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...