PAMP Activity of Cerato-Platanin during Plant Interaction: An -Omic Approach

Cerato-platanin (CP) is the founder of a fungal protein family consisting in non-catalytic secreted proteins, which work as virulence factors and/or as elicitors of defense responses and systemic resistance, thus acting as PAMPs (pathogen-associated molecular patterns). Moreover, CP has been defined an expansin-like protein showing the ability to weaken cellulose aggregates, like the canonical plant expansins do. Here, we deepen the knowledge on CP PAMP activity by the use of a multi-disciplinary approach: proteomic analysis, VOC (volatile organic compound) measurements, and gas exchange determination. The treatment of Arabidopsis with CP induces a differential profile either in protein expression or in VOC emission, as well changes in photosynthetic activity. In agreement with its role of defense activator, CP treatment induces down-expression of enzymes related to primary metabolism, such as RuBisCO, triosephosphate isomerase, and ATP-synthase, and reduces the photosynthesis rate. Conversely, CP increases expression of defense-related proteins and emission of some VOCs. Interestingly, CP exposure triggered the increase in enzymes involved in GSH metabolism and redox homeostasis (glutathione S-transferase, thioredoxin, Cys-peroxiredoxin, catalase) and in enzymes related to the “glucosinolate-myrosinase” system, which are the premise for synthesis of defence compounds, such as camalexin and some VOCs, respectively. The presented results are in agreement with the accepted role of CP as a PAMP and greatly increase the knowledge of plant primary defences induced by a purified fungal elicitor.


Introduction
Plants have the ability to detect the presence of pathogenic microorganisms coming in contact with them by means of substances produced by microbes themselves, including several proteins [1]. Plants recognize these microbe-/pathogen-associated molecular patterns (MAMPs/PAMPs), activating a defense system that is extremely effective against the potential pathogens. The first line of defense is composed of control systems that recognize several microbe elicitors, which enable plants to shift from growth and development to defense [2,3]. Among the elicitors, essential and conserved structures for pathogen survival, a novel fungal protein family composed by Cys-rich proteins recovered either on the cell wall or in the cultural filtrate of fungi is of peculiar interest. The family has been called the Cerato-platanin family (CPF, PF07249) from the name of its founder, cerato-platanin (CP), and is now composed of more than 130 members and the number is increasing on the rise [4][5][6][7]. CP is produced by

Differential Protein Expression
Proteins extracted from the CP-treated Arabidopsis leaves were separated by 2DE (2D-Electrophoresis and the differentially-expressed proteins were focalized between pI range 3-10 and a mass range of 12 to 100 kDa. About 1000 spots on each gel were reproducibility by Progenesis SameSpots (totallab, Newcastle, UK) ( Figure 1).

Differential Protein Expression
Proteins extracted from the CP-treated Arabidopsis leaves were separated by 2DE (2D-Electrophoresis and the differentially-expressed proteins were focalized between pI range 3-10 and a mass range of 12 to 100 kDa. About 1000 spots on each gel were reproducibility by Progenesis SameSpots (totallab, Newcastle, UK) ( Figure 1).

Figure 1.
Representative reference 2-DE gels of Arabidopsis control (A) and treated (B). Gels were colored by colloidal Coomassie blue staining. The Progenesis SameSpot software package was used for gels analysis. The differential proteins are identified by arrows: in red, the down-expressed spots in; in blue, the over-expressed spots in cerato-platanin (CP)-treated leaves. NL, non-linear. Tables 1 and 2 report results obtained by mass spectrometry, with information indicating the closest homolog proteins recovered in the database. Identification was performed after peptide mass fingerprinting, MASCOT research, and accessing the UniProt databank. Sometimes, more than one protein is present in one spot; these cases are reported as "mix score" and data interpretation takes into account that the protein with higher abundance can influence the spot quantitation and the mass spectrometry (MS) identification.
A comparison between "control" and "treated" enabled the identification of the 94 overall-affected spots: among these, 39 were downregulated and 55 were upregulated. As reported in literature, differentially-expressed proteins were defined when it was found in at least 1.5-fold abundance against control.
Of the 94 differentially-expressed spots, 54 proteins have been identified by MASCOT: 24 are down-expressed and 30 are over-expressed ( Figure 2).

Downregulated Proteins
Results show the downregulation of enzymes typically involved in primary metabolism, such as carbonic anhydrase, ribulose-1,5-bisphosphate carboxylase/oxygenase (RuBisCO), triosephosphate isomerase (TIM), and ATP-synthase-delta subunit, as shown in Table 1. In fact, despite control and treated samples being subjected to protamine to deplete most of the RuBisCO, many of the down-expressed spots contain peptides identifying the RuBisCO large and small chain (spots 13, 14, 18, 23, 24, 25 and 28, 30, 31, 32, 33, 44, 45, respectively). The chloroplast stem-loop binding protein, TIM, carbonic anhydrase, and ATP synthase delta-subunit (spots 12, 17, 20, 26, respectively), that are related to the carbon dioxide metabolism and photosynthesis, show 1.5-2-fold decrease in treated leaves. The peptidyl-prolyl cis-trans isomerase (spot 27) is also downregulated. Finally, the extracellular lipase 6 (EXL6, spot 42), involved in pollen development and growth, is  (B). Gels were colored by colloidal Coomassie blue staining. The Progenesis SameSpot software package was used for gels analysis. The differential proteins are identified by arrows: in red, the down-expressed spots in; in blue, the over-expressed spots in cerato-platanin (CP)-treated leaves. NL, non-linear. Tables 1 and 2 report results obtained by mass spectrometry, with information indicating the closest homolog proteins recovered in the database. Identification was performed after peptide mass fingerprinting, MASCOT research, and accessing the UniProt databank. Sometimes, more than one protein is present in one spot; these cases are reported as "mix score" and data interpretation takes into account that the protein with higher abundance can influence the spot quantitation and the mass spectrometry (MS) identification.
A comparison between "control" and "treated" enabled the identification of the 94 overall-affected spots: among these, 39 were downregulated and 55 were upregulated. As reported in literature, differentially-expressed proteins were defined when it was found in at least 1.5-fold abundance against control.
Of the 94 differentially-expressed spots, 54 proteins have been identified by MASCOT: 24 are down-expressed and 30 are over-expressed ( Figure 2).
The peptidyl-prolyl cis-trans isomerase (spot 27) is also downregulated. Finally, the extracellular lipase 6 (EXL6, spot 42), involved in pollen development and growth, is largely down-expressed in our results, thus confirming the slowdown of the primary metabolism in the defense-responding leaves. Figure 2. Level of expression of the identified spots. Data are negative when referred to the down-expressed proteins, while positive when referred to the over-expressed proteins. Numbers near the bar represents the fold ratio and the accession number (Uniprot databank). Bars represent the standard deviation of three replicates. Identification of each spot is provided in Tables 1 and 2.    Table 2 reports the over-expressed proteins identified from our gels. At a glance, the number and the fold of increase of the over-expressed proteins are twice larger than the down-expressed ones ( Figure 2).

Upregulated Proteins
One of the most over-expressed proteins is catalase-2 (spot 4), showing a concentration eight-fold higher in treated leaves than in control ones ( Figure 3). The spots related to glycerol-3-phosphate dehydrogenase (spot 8) and glyceraldehyde-3-phosphate dehydrogenase (GAPDH), either the GAPA2, chloroplastic (spot 49) and GAPC1 cytosolic (spot 51) isoforms, are largely over-expressed. Finally, spot 40 has been identified as thioredoxin H4, a thiol-disulfide oxidoreductase probably involved in the redox regulation of a number of cytosolic enzymes.
Other over-expressed spots are related to proteins involved in affecting the GSH/GSSG (the ratio of reduced/oxidized forms of glutathione). An interesting result is the over-expression of glutathione S-transferase F6 (spot 22) and of other GSH/GSSG related-enzymes, such as 2Cys-peroxiredoxin BAS1 (spot 48) and cytosolic sulfotransferase 18 (spot 29), that are considered markers of defense as much as GSH, being involved in the cysteine pathway. Myrosinase (spot 47), an enzyme able to hydrolyze glucosinolates (GL), is also over-expressed in our study on CP-exposed Arabidopsis leaves. Finally, the GDSL (Gly, Asp, Ser, Leu) esterase/lipase ESM1 (Epithiospecifier modifier 1) (spot 43 and 54 both identified with the Q9LJG3 accession number), belonging to the GDSL esterases/lipases family, shows a six-fold increase in treated samples.  Table 2 reports the over-expressed proteins identified from our gels. At a glance, the number and the fold of increase of the over-expressed proteins are twice larger than the down-expressed ones ( Figure 2).
One of the most over-expressed proteins is catalase-2 (spot 4), showing a concentration eightfold higher in treated leaves than in control ones ( Figure 3). The spots related to glycerol-3-phosphate dehydrogenase (spot 8) and glyceraldehyde-3-phosphate dehydrogenase (GAPDH), either the GAPA2, chloroplastic (spot 49) and GAPC1 cytosolic (spot 51) isoforms, are largely over-expressed. Finally, spot 40 has been identified as thioredoxin H4, a thiol-disulfide oxidoreductase probably involved in the redox regulation of a number of cytosolic enzymes.
Other over-expressed spots are related to proteins involved in affecting the GSH/GSSG (the ratio of reduced/oxidized forms of glutathione). An interesting result is the over-expression of glutathione S-transferase F6 (spot 22) and of other GSH/GSSG related-enzymes, such as 2Cys-peroxiredoxin BAS1 (spot 48) and cytosolic sulfotransferase 18 (spot 29), that are considered markers of defense as much as GSH, being involved in the cysteine pathway. Myrosinase (spot 47), an enzyme able to hydrolyze glucosinolates (GL), is also over-expressed in our study on CP-exposed Arabidopsis leaves. Finally, the GDSL (Gly, Asp, Ser, Leu) esterase/lipase ESM1 (Epithiospecifier modifier 1) (spot 43 and 54 both identified with the Q9LJG3 accession number), belonging to the GDSL esterases/lipases family, shows a six-fold increase in treated samples.  Other over-expressed proteins are related to the glycine biosynthetic pathway and photorespiration and, therefore, ultimately related to GSH metabolism and responses to biotic and abiotic stress. In fact, the spots 2, 11, and 53 (dhydrolipoyl dehydrogenase1, amino-methyltransferase, and serine hydroxymethyl-transferase1, respectively) show a two-fold increased level in comparison to control ones. Finally, we identified some other over-expressed proteins involved in defense at different degrees. Among these an eight-fold over-expressed spot is related to a pentatricopeptide (spot 5), and the 37 and 38 spots, identified for deoxyphosphosphooctonate aldolase and for lectin-like protein (Atg16530), respectively, are involved in cell wall integrity.

CO 2 Assimilation and Transpiration Rate
CO 2 assimilation and transpiration rate were determined by the use of LICOR with the aim to detect differences in photosynthesis and water between control and treated Arabidopsis leaves, either to add new knowledge on CP interaction with leaves or to increase information derived from proteomic analysis At all times considered, the net CO 2 assimilation rate was negatively affected by the CP treatment ( Figure 3). After four hours of treatment, net CO 2 assimilation in CP-infiltrated plants significantly declined by approximately 50% compared with control non-infiltrated plants and H 2 O-infiltrated plants. After 24 h the difference in CO 2 assimilation rate between CP and H 2 O-treated plants was lower, even if still significant. By contrast, no significant differences were detected between control and H 2 O-infiltrated plants to indicate that the stress induced by infiltration is smaller than that induced by CP, thus the former not affecting the validity of the experiment. The transpiration rate was negatively affected by CP treatment, but not to a significant extent. The effects on leaf water relations were less evident if compared to the decrease in photosynthetic rate, and they are largely due to the infiltration process.

VOC Accumulation
Volatiles as a consequence of treatment of Arabidopsis leaves with CP were determined to check the compounds that are induced by a purified non-catalytic fungal MAMP. Volatile organic compounds spectra from each plant were obtained by PTR-ToF-MS; peaks were present from m/z = 20 to m/z = 100. According to the procedure used by Aprea et al., 2015, the data were filtered by the elimination of signals with average intensity <1 ncps. Table 3 reports the most interesting protonated masses with the signal intensity normalized by the leaf area (expressed as normalized counts per second) for each sample. Table 4 shows the bibliographic citations in which volatile compounds have been identified by PTR-MS technologies.

Discussion
Previous studies demonstrated the role of CP as a PAMP, being involved in primary defense responses by activating the MAPK (mitogen-activated protein kinase) signaling, ROS production, camalexin synthesis, and the metabolic pathway leading to JA and ET production [5,11,12]. To best characterize the defense responses triggered by CP and, therefore, possibly by the other proteins belonging to the CP family, an -omic approach has been used in the present work. The results obtained from proteomic data, VOC analysis, and photosynthesis measurements are in agreement with each other and shed new light on the role of CP as a PAMP. In fact, to our knowledge, it is the first time that the PAMP/plant interaction is studied at the protein and volatile level.
Proteomic data are summarized in Figure 4, showing the pie-representation of the over-and down-expressed proteins resulting from our experiments: an increase of defense-related proteins at the expense of proteins involved in photosynthetic processes can be observed in CP-treated leaves, thus suggesting that the changing in basal carbon metabolism may increase the expression of defense-related genes, and promote the production of secondary compounds provided with antimicrobial activity. obtained from proteomic data, VOC analysis, and photosynthesis measurements are in agreement with each other and shed new light on the role of CP as a PAMP. In fact, to our knowledge, it is the first time that the PAMP/plant interaction is studied at the protein and volatile level. Proteomic data are summarized in Figure 4, showing the pie-representation of the over-and down-expressed proteins resulting from our experiments: an increase of defense-related proteins at the expense of proteins involved in photosynthetic processes can be observed in CP-treated leaves, thus suggesting that the changing in basal carbon metabolism may increase the expression of defense-related genes, and promote the production of secondary compounds provided with antimicrobial activity.

CP Treatment Slows Metabolism and Negatively Affects Photosynthesis
It is widely accepted that the induction of a sink metabolism during microbe interaction is one of the shifts that leads to a conversion from source to sink tissue during plant-pathogen interactions [33]. In agreement with this general observation, CP treatment induces down-expression of RuBisCO. The significance of the other downregulated proteins here identified is in agreement with the RuBisCO result: the decrease in TIM, carbonic anhydrase, ATP synthase delta-subunit, and EXL6 testify a general slowdown of the primary metabolism. The EXL6 belongs to a large subfamily of lypolitic enzymes called GDSL esterase/lipase proteins (GELPs) which have been recovered in microbes and plants and they possess important roles in morphogenesis, development, lipid metabolism, stress responses to abiotic stimuli, and, more generally, in pathogen defense [34]. Moreover, the RuBisCO downregulation has been observed in plants infected by insects and in abiotic stress responses, and it can be considered a hallmark of a metabolic strategy to sustain fitness of plants even in a stress condition [35,36]. A reduced amount of RuBisCO has also been observed after chitosan administration to Arabidopsis leaves [37]. RuBisCO downregulation also fits with the decrease of photosynthetic activity, as shown by the substantial declines in CO2 assimilation after 4 h of treatment without a significant CP-induced limitation in stomatal conductance. The decrease in leaf photosynthetic rate was also in agreement with the downregulation of the peptidyl-prolyl cis-trans isomerase which is related to the immunophilins of the thylakoid membrane of chloroplast and has roles in regulating the assembly of photosynthetic membranes [38]. Therefore, the decrease of photosynthetic activity unavoidably brings to a metabolic shift that may have the final scope of increasing the expression of defense-related genes, thus favoring the synthesis of secondary antimicrobial metabolites as below discussed.

CP Treatment Slows Metabolism and Negatively Affects Photosynthesis
It is widely accepted that the induction of a sink metabolism during microbe interaction is one of the shifts that leads to a conversion from source to sink tissue during plant-pathogen interactions [33]. In agreement with this general observation, CP treatment induces down-expression of RuBisCO. The significance of the other downregulated proteins here identified is in agreement with the RuBisCO result: the decrease in TIM, carbonic anhydrase, ATP synthase delta-subunit, and EXL6 testify a general slowdown of the primary metabolism. The EXL6 belongs to a large subfamily of lypolitic enzymes called GDSL esterase/lipase proteins (GELPs) which have been recovered in microbes and plants and they possess important roles in morphogenesis, development, lipid metabolism, stress responses to abiotic stimuli, and, more generally, in pathogen defense [34]. Moreover, the RuBisCO downregulation has been observed in plants infected by insects and in abiotic stress responses, and it can be considered a hallmark of a metabolic strategy to sustain fitness of plants even in a stress condition [35,36]. A reduced amount of RuBisCO has also been observed after chitosan administration to Arabidopsis leaves [37]. RuBisCO downregulation also fits with the decrease of photosynthetic activity, as shown by the substantial declines in CO 2 assimilation after 4 h of treatment without a significant CP-induced limitation in stomatal conductance. The decrease in leaf photosynthetic rate was also in agreement with the downregulation of the peptidyl-prolyl cis-trans isomerase which is related to the immunophilins of the thylakoid membrane of chloroplast and has roles in regulating the assembly of photosynthetic membranes [38]. Therefore, the decrease of photosynthetic activity unavoidably brings to a metabolic shift that may have the final scope of increasing the expression of defense-related genes, thus favoring the synthesis of secondary antimicrobial metabolites as below discussed.

CP Treatment Increases Expression of Defense-Related Proteins and VOC Emission
Results obtained from our experiments are consistent with the general observation that the stress-induced formation of ROS is a trademark of defense activation and they enable us to increase knowledge on the activation of defenses induced by a fungal PAMP [17]. In fact, one of the largest increases in transcripts deals with proteins related to ROS scavenging and redox homeostasis: for example, catalase, which is a highly active enzyme and does not require cellular reductants, shows an eight-fold increase in treated leaves, clearly to circumvent the ROS over-production induced by CP treatment [39]. Other largely over-expressed proteins are the glycerol-3-phosphate dehydrogenase and the glyceraldehyde-3-phosphate dehydrogenase, which are known to belong to NADPH production and to oxidative stress response [40,41]. The GDSL esterase/lipase ESM1 are also over-expressed and belong to an enzyme family whose members are involved in the regulation of morphogenesis, plant development, production of secondary metabolites, and defense response [42]. Finally, other over-expressed spots identified for the pentatricopeptide, reported to be involved in response to fungus and to chitin, and for the lectin-like protein (At3g16530), involved in defense responses to oligogalacturonides [43,44].
The spots related to proteins involved in regulating the GSH/GSSG rate show the same trend and significance: the over-expression of glutathione S-transferase F6, 2-Cys peroxiredoxin BAS1, thioredoxin H4, and cytosolic solftransferase 18 are in agreement with the recognized role of GSH in biotic stress conditions [45][46][47]. In fact, GSH controls early signaling events, expression of stress-related genes and plant defenses under biotic stress conditions, as proved by GHS1 Arabidopsis mutants showing a reduction in defense responses and low production of antimicrobial compounds, such as indole glucosinolates and camalexin, in response to a wide range of pathogens [47,48]. In this respect, other enzymes involved in the Gly biosynthetic pathway (and as a consequence in the GSH biosynthesis) are over-expressed: serine hydroxymethyl-transferase1, dihydrolipoyl dehydrogenase 1 and amino-methyltransferase, to underline the central role of GSH as a scavenger of the ROS over-production during defenses. In particular, serine hydroxymethyl-transferase 1 is directly involved in controlling cell damage during the hypersensitive defense responses [48]. Monomethyltransferases are involved not only in glycine (and consequently, GSH metabolism), but also in methylation of flavonoid compounds, such as anthocyanins and naturally-occurring stilbenes, such as resveratrol [49,50]. Moreover, a low content in GSH not only affects the redox homeostasis, but also slows down the synthesis of the antimicrobial compound camalexin, that is produced in large amount during the CP/Arabidopsis interaction [12]. In fact, camalexin, the major phytoalexin in Arabidopsis thaliana, is formed by an indole ring and a thiazole ring whose sulfur derives from Cys of the GSH molecule, thus explaining the tight relation between GSH metabolism and synthesis of antimicrobial compounds [47][48][49]. Therefore, an abundance of proteins related to GSH metabolism has to be expected, and now it is demonstrated in our experimental model using a purified PAMP as a defense inducer. The contradiction in down-expression of other glutathione-S-transferases (Q9FWR4, Q8L7C9, and Q9ZRW8) is only apparent: in fact, GST-U20 (Q8L7C9) is involved in cell elongation and flowering in response to light and its down-expression is in agreement with the shifting of the metabolism in response to pathogens; on the contrary, GST-U19 (Q9ZRW8) and GST-DHAR1 (Q9FWR4) are involved in plant defense mediated by Jasmonic acid and their down-expression fits with finding that CP primes salicylic acid and ethylene signaling pathways, but not the jasmonic acid signaling [12,[50][51][52][53].
Other over-expressed proteins detected in our study belong to the "glucosinolate-myrosinase" system, a unique defense mechanism typical of the Brassicacee family [45]. Sulfotransferase 18, myrosinase, and GDSL esterase/lipase ESM1 are involved in glucosinolate production and hydrolysis. Glucosinolates (GLs) are now considered preformed defense compounds and contribute to the protection against pathogens [45]. After cell damage, GLs are hydrolyzed by thioglucosidase enzymes ("Myrosinases"), to produce a variety of volatile products, such as thiocyanates, isothiocyanates, and nitriles, directly involved in defense, as reported by Hirschmann et al. [54].
The latter observation is fitting with results showing an increase of some VOCs in CP-treated leaves. Exposure to CP was able to increase the accumulation of dimethylsulfide, as it was found also in Silene paradoxa [55]. This VOC is enzymatically produced by plants in various environmental conditions, but its function in plant response to fungal attack is still unknown [56] The PAMP treatment induced the accumulation of another sulfur-containing compound, that of methanetiol. This molecule, together with dimethylsulfide, was found to be released by Brassica nigra upon herbivory and can be considered as a breakdown products of GLs, but, to the best of our knowledge, it was never associated to fungal infection [57].
Additionally, other kinds of VOCs are produced by the exposure of Arabidopsis plants to CP: some of them could be attributable to a PAMP-induced activation of the LOX (lipoxygenase) pathway via ROS formation, in agreement with the general observation that, due to their anti-oxidative characteristics, solubilized volatiles can also quench ROS produced during stress [20,58]. Specifically, the CP-induced upregulation of such biosynthetic pathways could have caused an accumulation of hexanal, a green leaves volatiles (GLV) C6-aldehyde that is known to be emitted by plants to counteract fungal growth [20]. In the same contest, the CP-induced accumulation of acetaldehyde could have an impact on ROS levels, as a consequence of the stress-induced activation of the LOX pathway for the biosynthesis of the GLVs [59,60]. In any case, our study is the first report about acetaldehyde production after fugal elicitation. Other CP-induced VOCs, such as the alkyl fragment, 2-butenal, isobutanal/butanone, ethyl acetate/methyl-propanoate, probably represent non-enzymatic products derived from oxidation of the polyunsaturated fatty acids upon ROS formation by CP.
Another interesting result was the CP-mediated induction of isoprene emission whereass generallys fungal infection is reported to reduce the production of such VOC [61]. Probably, the increase in the isoprene emission could be regarded as a response to the CP-mediated increase of ROS. In fact, the emission of isoprene is known to be stimulated by a wide range of environmental stresses that generate oxidative damage because of their ability in neutralizing ROS [62]. On the contrary, our VOC analysis was not able to reveal CP-mediated induction of the terpenes, as there were no signal intensities over values of m/z 100 [26]. Interestingly, signals attributable to fragments of terpenes show values of intensity lower than in control conditions or even constant over time. Therefore, even if terpenes are generally known to be produced after fungal attack and to play a role in plant defense [58], in our study the PAMP treatment seemed to downregulate their synthesis. On the other hand, it is widely accepted that environmental and biotic stress can either increase or reduce the emission rates of VOCs depending on severity, duration, and type of stress.

Plants and CP Treatment
Arabidopsis thaliana (Col-0) plants were grown in soil for five weeks in a growth chamber. Briefly, one week after germination in MS plates, Arabidopsis seedlings (one per container) were transferred into 120 mL containers filled with moistened soil and then closed off with a plastic lid (diameter 60 mm) perforated with five 8-mm holes by gently placing the seedlings in the central hole of the lid. Plants were then moved for five weeks in growth chambers with 12/12 h (day/night) photoperiod, 200 µmol¨m´2¨s´1 light intensity, 60% relative humidity, 20˝C of constant temperature and watered as required. 24 h prior to the measurements, plants were transferred in the air-conditioned room where the gas exchange were conducted. The holes in the lids were sealed with a synthetic rubber-based sealant (Terostat IX, Henkel, Düsseldorf, Germany) to suppress any potential H 2 O and CO 2 fluxes from the soil (as confirmed by using blank pots without plants).
The CP protein used in this study was obtained from the yeast Pichia pastoris. The pPIC9-cp plasmid was used for transformation to permit the recovery of the protein from the cultural filtrate [63]. A single purification step by Reverse Phase-High Performance Liquid Chromatography (RP-HPLC) was needed to obtain the pure protein in high yield (60 mg from 1 L of cultured medium). Pure heterologous CP was compared with the native one both for biological activity and structure according to [64].

Proteomic Experiments
Six to seven leaves from five-week old plants were detached and put into a moist chamber setup in petri dishes ( Figure S1). Six 10 µL drops (containing 150 µM CP or water as control) were applied on the lower surface of each leaf. Chambers were sealed and incubated under continuous light for 8 h. After incubation, drops were removed and leaves were frozen at´80˝C.

Protein Extraction
Leaves were placed into mortar containing liquid nitrogen and pulverized with a pestle immediately after treatments. 0.4 g of leaf powder was resuspended with 2 mL of ice-cold 50 mM Tris-HCl (pH 7.5), 200 mM NaCl, 1 mM EDTA, 10 mM NaF, 2 mM sodium orthovanadate, 1 mM sodium molybdate, 10% (v/v) glycerol, 0.1% Tween 20, 1 mM phenylmethylsulfonyl fluoride, 1 mM dithiothreitol, and 1ˆprotease inhibitor cocktail P9599 (Sigma-Aldrich, St. Louis, MO, USA). Than the samples were Ultra-Turrax (IKA, Staufen, Germany) treated for 15 s and centrifuged a 12,000ˆg for 10 min at 4˝C. The supernatant thus obtained was mixed with protamine sulfate (PS) at a final concentration of 0.05% to partially remove the RuBisCO enzyme [65]. The sample was maintain on ice for 30 min and centrifuged at 12,000ˆg for 10 min at 4˝C. Proteins in the supernatant were than subjected to methanol/chloroform precipitation. Briefly sample was added by methanol/chloroform/water (4:1:3 v/v), agitated vigorously and centrifuged for 10 min at 10,000ˆg at 4˝C. The upper layer was then removed without disturbing the interface and sample was further added of three volumes of methanol and centrifuged. The final pellet was dried, dissolved in 8M Urea, 4% (w/v) CHAPS, and 20mM DTT, and subjected to 2DE.

2D-Electrophoresis
2-DE replicate gels (n = 3) for each experimental condition, were performed using independent experiments. IEF was carried out on IPGs (pH 3-10 Non-Linear; 18 cm long IPG strips; GE Healthcare (Little Chalfont, UK)) through the Ettant IPGphor system (GE Healthcare). Strips were hydrated with 350 mL of 8 M Urea, 2% (w/v) CHAPS and 2% v/v carrier ampholyte, overnight at room temperature. Sample load, 600 µg per strip, was carried out by cup loading in the IPGphor Cup Loading Strip Holders, through the sample cup system at the anodic side of strips [66]. Subsequently, strips were placed in equilibration buffer (6 M urea, 75 mM Tris-HCl pH 8.8, 29.3% glycerol, 2% SDS) containing 2% (w/v) DTT for 15 min, and then in the same buffer with 2.5% iodoacetamide for 15 min. The second dimension was made on polyacrylamide linear gradient gels (9%-16%; 18 cmˆ20 cmˆ1.5 mm) at 40 mA per gel constant current. Spots were highlighted by colloidal Coomassie blue staining.

Images Analysis
Gels image were acquired by an Epson Expression 1680 Pro image scanner. Differentially-expressed spots were selected by Progenesis SameSpot analysis (Nonlinear Dynamics, Newcastle upon Tyne, UK). Two groups (control and treated leaves) were compared with each other by the one way ANOVA analysis. All spots were pre-filtered and manually verified before applying the statistical criteria (ANOVA p < 0.05 and fold > 1.5). Spot intensity, instead of normalized spot volumes, was used in statistical processing.

In-Gel Digestion and MALDI-ToF Analysis
Spots were manually cut out from gels and each sample was washed twice in 50 mM NH 4 HCO 3 /CH 3 CN 1/1 for 15 min and then dehydrated in CH 3 CN. Samples were then re-swelled in NH 4 HCO 3 with 10 mM DTT and placed for 30 min at 56˝C; after the liquid was take out and samples were incubated in the dark for 30 min at room temperature in the same volume of 55 mM IAA in 25 mM NH 4 HCO 3 . Then, gel particles were washed twice, dried, and incubated for 30 min at 37˝C in 20 µL of 20 µg¨mL´1 trypsin solution (Trypsin/Lys-C Mix Mass Spectrometry Grade, PROMEGA, Madison, WI, USA) in 25 mM NH 4 HCO 3 . An additional 10 µL of the buffer were added and incubated overnight at 37˝C.
The reaction was interrupted by 1% trifluoroacetic acid (TFA) and the supernatant was collected. Gel particles were then re-extracted with 1% TFA in 50% CH 3 CN. Supernatants were combined and analyzed on a MALDI-TOF/TOF mass spectrometer Ultraflex III (Bruker Daltonics, Bremen, Germany) by using Flex Control 3.0 as data acquisition software. The sample was mixed with the same volume of a saturated solution of a-cyano-4-hydroxycinnamic acid in 50% (v/v) CH 3 CN and 0.5% (v/v), and acquired in the reflectron mode over them, with a z-range of 860-4000, for a total of 500 shots [66].
Mass fingerprinting searching was performed in Swiss-Prot/TrEMBL databases by MASCOT (Matrix Science Ltd., London, UK, http://www.matrixscience.com) software. The taxonomy was limited to Arabidopsis. Alkylation of cysteine by carbamidomethylation was hired as fixed modification and a mass tolerance of 50 ppm was tolerable. The number of allowed missed cleavage sites was set to one.

Leaf Gas Exchange
Leaf gas exchange parameters were measured as in Bazihizina et al. [67], using the LI6400-XT gas exchange system analyzer equipped with a LI-6400-17 Whole Plant Arabidopsis (WPA) chamber and the 6400-18A external RGB light source, specifically planning to measure the whole plant gas exchange even on small rosette-type Arabidopsis plants (LICOR Inc., Lincoln, NE, USA). Four to five leaves for each plant were infiltrated with 25 µL of a 150 µM CP solution or water. All of the enclosed leaves were subjected to saturating photosynthetic photon flux density (PAR, 1000 µmol¨m´2¨s´1), 380 ppmv CO 2 (achieved by fully scrubbing CO 2 from ambient air with soda lime and replacing it with the LI-COR6400 CO 2 -injector system), 25˝C leaf temperature, and 45%-50% relative humidity. Net transpiration and photosynthetic rates were measured before any treatment and at 4 and 24 h after infiltration with water or the CP protein. Gas exchanges were also measured in non-infiltrated plants as a control. All measurements were taken on three plants from each treatment at ambient RH (60%-70%), 400 µmol¨mol´1 CO 2 concentration, 500 µmol¨s´1 flow rate, 25˝C leaf chamber temperature and 200 µmol¨m´2¨s´1 PAR. After each measurement, leaf area was measured for all plants by the image analysis of the rosette with the Easy Leaf Area software as described by Easlon and Bloom [68].

Proton Transfer Reaction-Time-of-Flight-Mass Spectrometry and VOC Determination
VOC emission was detected following the method of Taiti et al. [55]. Briefly, shoots were isolated from the system using a synthetic rubber-based sealant (Terostat IX, Henkel, Düsseldorf, Germany) placed around the base of the stem to exclude the influence of water evaporation and ambient air. Shoots of Arabidopsis thaliana were uniformly sprayed with 150 µM CP solution or with milliQ-water and, immediately, plants were transferred to a glass jar (150 mL) ( Figure S1). VOC accumulation was monitored by PTR-ToF-MS 8000 apparatus (Ionicon Analytik GmbH, Innsbruck, Austria) at different incubation times (0.5, 2, 4, 24 h at 25˘1˝C in air conditioned room). Five plants for each treatment and incubation time were evaluated, H 3 O + was used as reagent ion for the proton transfer reaction. VOCs were sampled directly from the glass jar equipped on opposite sides with two holes connected with a Teflon tubes to the PTR-ToF-MS tool and to a zero-air generator, creating a dynamic headspace. A commercial zero-air generator (Peak Scientific Instruments GmbH, Frankfurt Germany) operated at 399˝C was used for the generation of VOC-free air. Mass spectra between m/z = 20-2010 was determined with ToF acquisition of 0.1 ns for each channel, the time for analyzing each plant was about 100 s. Throughout the measurement the PTR-ToF 8000 was operating in standard mode and the settings were as follows: 2.20 mbar pressure of drift tube, 60˝C temperature, 594 V drift tube voltage, 35 V of the extraction voltage at the end of the tube (Udx), corresponding to an E/N-electric field strength per gas number density-value of 130 Td (1 Td = 10-17 V¨cm´2). The internal calibration was performed with m/z = 29.997 (NO + ), m/z = 59.049 (C 2 H 5 O 2 + ), and m/z = 180.937 (C 6 H 4 Cl 3 + ) to obtain a high mass accuracy offline, following the procedure described in [69]. PTR-ToF-MS raw data were recorded by the TofDaq™ data acquisition software (Tofwerk AG, Thun, Switzerland). All spectra were corrected by the use of Poisson correction in the DAQ settings of ToFDaq configuration options. Subsequently, the TofViewer software (version 1.4.3, Ionicon Analytik, Innsbruck, Austria) was used for data post processing. Peak quantification was performed according to the duty cycle and the signals were normalized to generate normalized count per second (ncps) values. Acquisition of 60 average spectra were used for data modeling and average signal intensity was recorded for 60 s. Identification of the m/z signals was performed by assigning the mass formulas reported and through the integration of previous knowledge of the VOCs emitted by plants.

Conclusions
Topics of this paper are above the defense induction by a purified fungal protein in interaction with the model plant Arabidopsis. The novelty of findings is mainly due to the use of CP as a model protein either for the CPF proteins or for other non-catalytic fungal elicitors. Results obtained from different experiments are in agreement with each other and enabled a high throughput analysis of the CP/plant interaction: data obtained from proteomic, volatilomic, and gas-exchange determination largely increase the knowledge about the primary defenses induced by a purified protein elicitor. Moreover, they fit well with other studies that, however, are mainly performed with entire pathogens in interaction with the plant.
The clear fall down of the photosynthetic activity drives to a metabolic shift from source to sink in Arabidopsis leaves treated with CP. This is argued either by the downregulation of proteins of the primary metabolism and inhibition of CO 2 assimilation, on one hand, and by the over-expression of enzymes involved in ROS scavenging and GSH metabolism, on the other. Moreover, results fit with some of the previously-obtained transcriptomic data. For example, the overexpression of genes corresponding to At3g16530, At1g02920, and to At2g02930, the Rossman-fold NAD(P)-binding domain containing protein, validate the overexpression of the legume lectin-like protein (Q9LK72), of the glutathione S-transferase (P42760) and of some dehydrogenases involved in defences (Q9M5K3, Q9SCX9, Q9LPW, P25858). Conversely, the proteins involved in regulating the GSH/GSSG rate, the enzymes involved in Gly biosynthetic pathway, and the "glucosinolate-myrosinase" system, have never been identified before in CP plant interaction. Finally, many of the data here obtained on VOCs are in agreement with proteomic results: in particular, as mentioned above, the emission of isoprene that is known to be stimulated by a wide range of environmental stresses and the overexpression of enzymes involved in synthesis of GLs that produce a variety of volatile products, related to plant defence.