Introduction

Sugarcane is one of the important commercial crops cultivated in more than 100 countries for the production of sugar, ethanol, and other by-products (http://www.fao.org/faostat/en/#data/QC). Colletotrichum falcatum, a hemibiotrophic fungal pathogen that causes red rot, poses a serious threat to sugarcane production in many tropical and subtropical countries (Singh and Lal 2000). The disease has been responsible for a considerable reduction of cane yield (30–100%) and sucrose recovery (25–75%) (Viswanathan 2012). Complex polyploidy and absence of whole genome information of sugarcane have hampered the efforts of breeding for achieving durable red rot resistance (Sundar et al. 2015; Viswanathan 2017). Disease management practices involving fungicidal applications were not found to be effective at the grand growth phase of the crop. Nevertheless, the application of few biotic and abiotic resistance inducers has been successfully demonstrated to induce systemic resistance against red rot of sugarcane (Ashwin et al. 2017a).

Biotic resistance inducers or pathogen-derived signatures predominantly encompass cell structure-associated biomolecules and secreted proteins (Zhang and Zhou 2010; Böhm et al. 2014). Of these, the well-conserved, essential pathogen signatures, which are often recognized by a broad range of plants are called as “pathogen-associated molecular patterns (PAMPs),” whereas the evolution of a less conserved or novel species-specific signatures that suppress host defense mechanisms are known as “effectors”. Generally, effectors counterattack the PAMP-triggered immunity (PTI), which would get activated immediately after perception of the pathogen (Boller and Felix 2009). While most of the effectors are actively secreted, PAMPs are not typically actively secreted. On the other hand, without the real-time confrontation of the pathogen, the extraneous application of these PAMPs and certain effectors (based on host genotype) can induce host defense similar to resistance inducers/elicitors (Boyd et al. 2013; Anil et al. 2014). The active resistance, thus induced by the process of priming may be systemic, durable, and capable of acting against subsequent attacks of pathogens (Conrath et al. 2006, 2015). Hence, the strategy of employing effective resistance inducers and identification of new pathogen signatures, which can sensitize the host, either by enhancing or suppressing host defense would aid in disease management as well as deciphering disease resistance or suppression-associated genes, respectively.

Hithero, a multitude of host resistance-inducing secreted PAMPs/elicitors has been identified from a wide range of pathosystems, while a handful of secreted PAMPs have been functionally demonstrated to induce disease resistance upon priming (Walters et al. 2013; Burketova et al. 2015; Hamid and Wong 2017). For instance, priming with harpin, cerato-platanin, and elicitin class of PAMPs have been reported to enhance disease resistance by inducing the production of plethora of defense-related compounds. Further, overexpression of the PAMPs like harpins, flagellins, cerato-platanins, and elicitins has also been reported to enhance a broad spectrum of disease resistance against bacterial and fungal pathogens in model plants (Shao et al. 2008; Takakura et al. 2008; Yang et al. 2009; Du et al. 2015).

Priming of sugarcane with benzothiadiazole (BTH), an analogue of salicylic acid and Cf elicitor, a glycoprotein derived from the cell wall of C. falcatum has been shown to induce the production of various defense-related compounds and suppress red rot severity (Ramesh Sundar et al. 2008, 2009; Ashwin et al. 2017a). The application of these inducers has been reported to upregulate the expression of many defense-related transcription factors and resistance gene analogs (RGAs) (Muthiah et al. 2013; Selvaraj et al. 2014). However, the factors such as cost-effectiveness and the complexity in mass production of BTH and Cf elicitor, respectively, have restricted the employment of these inducers for large scale field applications. Recently, a study on comparative secretomic analysis of C. falcatum cultures that differed in virulence has identified CfEPL1 (eliciting plant response-like) protein, a potential PAMP that induces systemic resistance in sugarcane (Ashwin et al. 2017b). Besides, the study has identified few pathogenicity-related proteins and highlighted the importance of mining the secreted proteins of C. falcatum.

On the whole, the hurdles confronted with BTH and Cf elicitor combined with the significance of elucidating functional information of pathogen signatures have necessitated scouting for identification of defense triggering protein(s) from C. falcatum secretome. Therefore, in this study, we systematically screened the secretome of C. falcatum and identified a novel protein that possesses defense elicitation property. Besides, this study has also characterized the functional importance of putative domains of the identified novel protein through a series of bioassays in sugarcane and Nicotiana tabacum.

Materials and methods

Fungal culture and plant materials

A standard high virulent isolate of C. falcatum, Cf671 (Microbial Type Culture Collection and Gene Bank, Chandigarh, India, accession number 12142), was used for all the experiments in this study. It was cultured on oat meal agar plates and incubated at 28 °C with 14-:10-h light and dark cycles (light intensity 100 μmol/m2/s) (Model MIR554, Panasonic Healthcare Co., Ltd., Gunma, Japan). The plates were subcultured once in every fortnight for maintenance of the culture. Eight-month-old sugarcane cultivars, namely, CoC 671 (red rot susceptible) and Co 93009 (red rot resistant), grown under natural climatic conditions at the institutional experimental farm (ICAR—Sugarcane Breeding Institute, Coimbatore, India) were used in this study. For certain bioassays, 2-month-old greenhouse grown N. tabacum plants were used. To establish sugarcane cell suspension cultures, friable embryogenic calli generated from meristematic tissues of CoC 671 were used (Ramesh Sundar and Vidhyasekaran 2003).

Protein extraction, fractionation, and purification

For extraction of secreted proteins from C. falcatum, mycelial discs of around 5 mm diameter were inoculated in ten 500-mL flasks each containing 150 mL of oat meal broth and incubated at 28 °C for 10 days without agitation with similar light and dark cycles as stated above. After 10 days of incubation, the mycelial mats were harvested and the spent medium was clarified by centrifugation. The supernatant was then serially filtered through 0.8- and 0.45-μm membrane filters (Supor® PES, Pall Corporation, MI, USA) to ensure removal of spores, mycelial filaments, and particulates of oat meal broth, if any. To extract total secreted proteins from this filtrate in its native form, ammonium sulphate ((NH4)2SO4) precipitation method was used as described by Ashwin et al. (2017b). Briefly, ((NH4)2SO4) salt was slowly added to the filtered spent medium, until it reaches a 90% saturation at 4 °C, while constantly mixing with a magnetic stirrer for 20 h. Precipitated proteins were then recovered using centrifugation at 12,000 rpm for 30 min and resuspended in a buffer (Buffer A) that composed of 10 mM tris, 40 mM sodium chloride (NaCl), 1 mM dithiothreitol, and 10% glycerol with a pH of 7.5. In order to desalt the residual (NH4)2SO4 from the protein solution, they were washed with 20 volumes of buffer A using 10 kDa Amicon ultra-15 centrifugal ultrafilters (Merck Millipore, County Cork, Ireland).

To fractionate the total secreted proteins based on size, ultrafilters with different cutoff ranges were used. First, the desalted protein solution was filtered using 50-kDa Amicon ultra-15 centrifugal ultrafilters (Merck Millipore, County Cork, Ireland). While the retentate containing proteins greater than 50 kDa was collected and marked as > 50 kDa fraction, the filtrate was subjected to further fractionation with a 30-kDa filter. The resultant retentate representing the proteins ranging from 30 to 50 kDa was marked as 30–50 kDa fraction, and the remaining filtrate was just concentrated and marked as 10–30 kDa fraction. Besides these three fractions, total secreted proteins from the filtered spent medium were directly fractionated by sequential precipitation with different saturation levels of ((NH4)2SO4), namely, 0–30, 30–60, and 60–90% (Dixon and Webb 1962; Wingfield 2016). These protein fractions were immediately desalted with buffer A as described above and used for bioassay-based screenings.

Anion exchange chromatography (AEC) using Bio-scale mini macro-prep high Q cartridge (Catalog no. 732-4122, Bio-Rad Laboratories, Inc., CA, USA) was employed for further separation of active protein fraction in BioLogic™ LP liquid chromatography system (Bio-Rad, Laboratories, Inc., CA, USA). Here, the proteins were eluted with a linear gradient of 0.04 to 1 M NaCl in the same buffer. Based on the chromatogram, individual tubes corresponding to a single peak were pooled and marked as a single fraction. Prior to bioassay, these pooled peak-specific fractions were desalted and concentrated as described above. The active protein fraction was further purified by size exclusion chromatography (SEC) by employing Agilent Bio SEC 5 μm, 150 Å column in Agilent 1260 infinity HPLC system (Agilent Technologies Inc., Waldbronn, Germany). At each stage and whenever required, protein concentrations were quantified by Bradford method, using BSA as reference standards. Similarly, the quality of the extracted and fractionated proteins was ascertained with 1-DE (SDS-PAGE).

Bioassays on sugarcane suspension culture

After elicitation with fractionated or purified proteins, changes in the extracellular pH of CoC 671 suspension cells were monitored at different time intervals until 3 h post-elicitation. Similarly, the accumulation of hydrogen peroxide (H2O2) was quantified at different time intervals as described by Bindschedler et al. (2001). In brief, 300 μL of suspension cells was drawn at different time points and centrifuged at 12,000 rpm for 15 s. Then, 100 μL of supernatant was immediately added to 1 mL of freshly prepared xylenol orange reagent and incubated at room temperature for 30 min. After incubation, the absorbance of the mixture was measured at 560 nm. Absorbance values of the test samples were then plotted against the standards prepared with different concentrations of H2O2. BTH, a well demonstrated resistance-inducing agent in sugarcane, was used as a positive control for studying defense elicitation. The level of H2O2 accumulation was detected at 48 h post-elicitation by using DAB staining method as described by Ashwin et al. (2017b). For DAB staining, 10 μL/mL of 3,3′-diaminobenzidine (DAB) tetrahydrochloride hydrate (Sigma-Aldrich Co. LLC, MO, USA) (1 mg/mL) solution, pH 3.8, was added immediately after the elicitation of suspension cells with fractionated or purified proteins and incubated under dark conditions for 48 h. The development of the brown precipitate, as a result of reaction between DAB and H2O2, was documented both macroscopically and microscopically. For all DAB staining assays, ascorbic acid, the quenching agent of H2O2, was separately amended with their respective positive control to verify whether the browning of the cells was exclusively due to H2O2 production.

For certain experiments, viability of suspension cells was determined using Evans blue as described by Davies et al. (2006). After 48 h of elicitation, 0.05% (w/v) of Evans blue was added with 1 mL of suspension cells and incubated for 15 min at room temperature. Unbound dye was removed by extensive washing of the cells with Milli-Q grade water and the dye that was bound to the dead cells was solubilized by grinding them in 50% methanol containing 1% SDS and incubating the homogenate at 50 °C for 30 min. After incubation, the absorbance of the supernatant was measured at 600 nm. Percentage of cell death was calculated by considering the relative absorbance values of heat-treated cells as 100% death and the mock-treated cells as 0% death. Three replications were used for all these bioassays, whereas only the representative images were presented for DAB staining.

Bioassays on N. tabacum

To evaluate the defense elicitation properties of fractionated or purified proteins on a nonhost model system, 50 μL of protein solutions was infiltrated on the adaxial side of N. tabacum leaves using a needleless syringe. After 24 and 48 h of infiltration, the spots were macroscopically examined for the development of hypersensitive responses (HR), if any. To detect the level of H2O2 accumulation in response to perception of proteins, the infiltrated portion of the leaves was excised (3 cm in diameter) and stained with DAB as described by Chen et al. (2012). In brief, after 24 h of infiltration, the leaf discs were excised and stained with DAB solution (1 mg/mL) at room temperature for 8 h under dark condition. Stained discs were then bleached in 95% ethanol for 20 min and mounted in 50% glycerol for observation under light microscope. Meanwhile, to detect callose deposition, the leaf discs were completely cleared with ethanol-acetic acid (3:1, v/v) solution for 24 h, followed by a wash with 50% ethanol and three consecutive washes with 70 mM dipotassium hydrogen phosphate buffer (pH 9). The discs were then stained with 0.05% aniline blue prepared in 70 mM dipotassium hydrogen phosphate buffer (pH 9) for 2 h. After a brief rinsing with distilled water, the discs were mounted in 50% glycerol and observed under UV illumination using an epifluorescence microscope. Each of these experiments was repeated three times and only the representative images were presented here.

MS/MS analysis and database search

The purified active protein fraction was electrophoresed (SDS-PAGE) on a 20-cm-long gel in Ettan DALT six electrophoresis unit (GE Healthcare Bio-Sciences AB, Uppsala, Sweden) and the band that represented a single protein was manually excised and subjected to MALDI TOF/TOF analysis as described by Barnabas et al. (2016). After destaining, the spot was alternatively washed with 50 mM ammonium bicarbonate (NH4HCO3) and 50% acetonitrile for four times and subjected to in-gel digestion with sequencing-grade modified trypsin (Promega Corporation, Madison, USA) for 12 h. Digested peptides were then extracted with 50% acetonitrile containing 0.1% trifluoroacetic acid (TFA). For MALDI TOF/TOF analysis, equal volumes of digested peptides and α-cyano-4-hydroxycinnamic acid (CHCA) solution (5 mg/mL of CHCA in 70% acetonitrile with TFA 0.1%) were mixed and spotted in triplicates onto a standard 386-well stainless steel MALDI target plate in 4800 Plus MALDI TOF/TOF Analyzer (AB Sciex, CA, USA) and the spectra were collected in a data dependent mode. After a survey scan (MS), MS/MS was performed for the 10 most abundant ions and all the spectra were exported for post-MS analysis.

For protein identification, the above-obtained MS/MS spectra were searched against two different C. falcatum-specific protein databases, viz., transcriptome shotgun assembly (Bioproject PRJNA272832) (Prasanth et al. 2017) and whole genome shotgun assembly (Bioproject PRJNA272959) of Cf671 isolate using Peaks studio software (Version 8; Bioinformatics Solution Inc., ON, Canada), which was supported with de novo, Peaks DB, PTM, and Spider algorithms. The set search parameters were 1+ for peptide charge state, 50 ppm for peptide mass tolerance, 0.3 Da for fragment mass tolerance, and 1 and 3 for maximum allowed missed cleavage and PTM predictions, respectively. Besides, carbamidomethylation (C) and oxidation (M) were set for fixed and variable modifications, respectively. Average local confidence and − 10lgP threshold for fetching peptide spectral matches (PSMs) were set to ≥ 50, to eliminate false positive identifications. False discovery rate (FDR) of PSMs was also estimated by decoy fusion method during Peaks DB search.

In silico analysis

The identified protein was annotated using BLASTP search and the parameters like theoretical molecular weight and pI were predicted using ProtParam tool (ExPASy, Swiss Institute of Bioinformatics). Multiple sequence alignment was performed with Clustal Omega (http://www.ebi.ac.uk/Tools/msa/clustalo/). For validation of secretory nature of the protein and signal peptide domain prediction, SignalP v 4.1(www.cbs.dtu.dk/services/SignalP/) and TargetP v 1.1 (www.cbs.dtu.dk/services/TargetP/) were used. Besides, the probability of being an effector or possessing effector properties was predicted using EffectorP v 1.0 (http://effectorp.csiro.au/) (Sperschneider et al. 2016).

Cloning, expression, and purification

For functional characterization, different domain deletional variants of the purified protein sequences were constructed in pET28a expression vector (Novagen® Merck Millipore, Darmstadt, Germany) with 6X His and T7 fusion tags at N-terminal end and expressed in the Escherichia coli strain Rosetta™ 2(DE3)pLysS (Novagen® Merck Millipore, Darmstadt, Germany). Briefly, the truncated sequences of the gene that encode the purified protein were amplified with restriction site linked sequence-specific primers (Table S1a in the supplementary material). These inserts and pET28a vector were separately double digested with EcoRI and SalI (ThermoFisher Scientific Inc., Waltham, USA) and ligated with T4 DNA Ligase (ThermoFisher Scientific Inc., Waltham, USA). After ligation, they were transformed into Rosetta™ 2(DE3)pLysS cells and grown on kanamycin-containing Luria Bertani (LB) agar plates. The veracity of the positive clones was verified by colony PCR, restriction digestion, and Sanger sequencing. To express the protein of interest, these positive transformants were allowed to grow up to 0.5 OD and induced with 0.5 mM isopropyl β-d-1-thiogalactopyranoside (IPTG) at 30 °C for 4 h. After which, the cells were pelleted down by centrifugation at 6000 rpm for 5 min and resuspended in 0.1 volume of extraction buffer (50 mM Tris, 200 mM NaCl, 10 mM β-mercaptoethanol, 10% glycerol). The suspension was then subjected to three cycles of freeze-thaw shocks and briefly sonicated at 10 s pulse on, 10 s pulse off for 1 min at 30% amplitude for cell lysis. The lysates were then centrifuged at 13,000 rpm for 10 min at 4 °C and the supernatant that contained the recombinant proteins was purified using a Ni-Sepharose 6 Fast Flow column (GE Healthcare Bio-Sciences, AB, Uppsala, USA), a low pressure chromatography system (BioLogic™ LP System, Bio-Rad, Laboratories, Inc., CA, USA). Here, the recombinant his-tag proteins were eluted against a linear gradient of extraction buffer containing 500 mM imidazole from 0 to 500 mM for 30 min. Fractions containing recombinant proteins were immediately desalted with buffer A as described above and purified further by size exclusion chromatography using Sephacryl S100 HR column.

Western blot

The purity of recombinant proteins was verified by immunoblotting using T7 tag-specific antibody as described by Ashwin et al. (2017b). In brief, purified proteins were electrophoresed and transferred into an activated polyvinylidene fluoride (PVDF) membrane (0.2 μm) by semi-dry transfer method using Trans-Blot® SD Semi-Dry Transfer Cell (Bio-Rad, Laboratories, Inc., CA, USA). After blocking, the membrane was incubated with T7 tag-specific primary antibody (T7 Tag Monoclonal antibody, Merck Millipore, USA) followed by the secondary antibody (Goat Anti-Mouse IgG-Alkaline Phosphatase antibody, Sigma-Aldrich Co. LLC, MO, USA) and developed with the substrate 5-bromo-4-chloro-3-indolyl phosphate (BCIP) in conjunction with Nitro blue tetrazolium (NBT). For priming and bioassays, fusion tags of purified recombinant proteins were removed using Thrombin CleanCleave™ Kit (Sigma-Aldrich, Co. LLC, MO, USA) by centrifugation recovery method as per manufacturer’s instructions.

Defense priming and detached leaf bioassays

To evaluate the priming efficacy of recombinant proteins, 8-month-old CoC 671 canes were foliar sprayed (25 mL/cane) twice with optimum concentration of recombinant proteins at 10-day interval using a handheld sprayer. After 2 weeks, primed and unprimed (mock-treated) canes were inoculated with Cf671 spore suspension by nodal swabbing method as described by Viswanathan et al. (1998). In brief, the leaf sheaths around the sixth node (from the meristematic cane top) were removed and the pathogen was inoculated by wrapping a thin wad of absorbent cotton, which was drenched with 2 mL of spore suspension (106 spores/mL). In order to maintain the moisture to facilitate germination and penetration of the pathogen into the root primordia and leaf scar, the inoculated nodal portion was completely covered with Parafilm® “M” film (Sigma-Aldrich, Co. LLC, MO, USA). For pathogen quantitation and gene expression analysis, the stalk samples of primed and unprimed canes were collected at different time points (0, 12, 24, 48, 72, 120, and 600 h post-inoculation (hpi)) and stored at − 80 °C until further processing.

On the other hand, after 2 weeks of foliar spray, matured third leaves of primed and unprimed canes were detached and subjected to detached leaf bioassay as described by Ashwin et al. (2017b). Briefly, the bottom portion of the detached leaves was excised and washed and minor injuries (few pinpricks in approximately 5 mm2 area) were made in the lamina on both sides of the midrib to breach the thick cuticle layer enriched with waxes. After that, 50 μL of Cf671 spores (106 spores/mL) was inoculated on the injured surface and incubated in an illuminated chamber (Model E41L2, Percival Scientific Inc., IA, USA) maintained at 28 °C with 60% humidity and 14:10 h light-dark cycling conditions. For co-infiltration experiment, Cf671 spore suspension (106 spores/mL) was prepared in the recombinant protein solution and inoculated on unprimed detached leaves as described above. After 72 h, the leaves were assessed both macroscopically and microscopically for lesion development and infection structures, respectively. While the brownish yellow lesion length was measured using a ruler, the number of appressorium per square millimeter area was counted from 10 different spots (250 μm2) per leaf to minimize errors. Each set of treatments comprised of five independent replications and the experiment was repeated three times for reproducibility.

Statistical analysis

Quantitative datasets representing lesion length and number of appressorium were subjected to Kolmogorov-Smirnov test, Shapiro-Wilk’s test, and Levene’s test to check the assumptions of normality and homogeneity of variance. They were then statistically analyzed using one-way analysis of variance (ANOVA). The differences between the treatments were analyzed by post hoc Tukey’s test at the significance of p ≤ 0.05 using the software IBM SPSS statistics 21.0 (SPSS, Chicago, USA).

C. falcatum biomass quantitation and gene expression analyses

For in planta pathogen biomass quantitation, genomic DNA from the nodal portion (covering approximately 2 cm above and below the site of pathogen challenge) of the stalk samples collected at different time points (0, 12, 24, 48, 72, 120, and 600 hpi) were extracted using CTAB method. The biomass of C. falcatum was then quantified using the constitutively expressing gene, eukaryotic elongation factor (CfeEF1α) by absolute quantification method using qPCR (Diguta et al. 2010). A standard curve was generated from CfeEF1α amplification from tenfold dilutions of C. falcatum DNA with a linear relationship (R2 value) of 0.989. Quantitation cycle (Cq) values of the test samples were then substituted in the standard curve equation to determine the quantity of pathogen biomass.

For gene expression analyses, total RNA was extracted from the nodal region as described above using TRI Reagent® (Sigma-Aldrich Co. LLC, MO, USA) and then converted into cDNA using RevertAid™ H minus First Strand cDNA Synthesis Kit (ThermoFisher Scientific, Waltham, USA). Among the constitutive genes of sugarcane, viz., ScEF1α, ScGAPDH, and Sc25S rRNA, the elongation factor ScEF1α was found to be the most stable reference gene by the tool NormFinder Excel add-in v 0.953 (https://moma.dk/normfinder-software) for qRT-PCR analysis. Similarly, C. falcatum elongation factor (CfeEF1α) was found to be the most stable reference gene, among CfeEF1α and Cf actin genes. Therefore, the expression of target genes from sugarcane and C. falcatum was normalized with the expression of ScEF1α and CfeEF1α, respectively. Three biological replicates with two technical duplicates each were used for all qPCR analyses. All these analyses were performed in StepOnePlus™ Real-Time PCR (Applied Biosystems, CA, USA) using Power SYBR Green PCR Master Mix (Applied Biosystems, CA, USA). The details of primer sequences, amplicon length, amplification efficiency, and annealing temperature were listed (Table S1b in the supplementary material). Melt curve analysis was also performed to verify the amplification specificity of target genes. Relative fold expression and respective standard deviation of target genes were determined by 2-ΔΔCT method.

The nucleic acid sequences of all the genetic products like full length genes, truncated genes, and the genes employed for expression analyses were submitted in GenBank and their respective accession numbers were listed in Table S1 in the supplementary material.

Results

Fractionation and bioassays for screening of elicitor-active protein fraction(s)

To identify a C. falcatum protein that could induce or trigger host defense mechanism, a systematic top-down approach was adopted as described in the schematic diagram (Fig. S1 in the supplementary material). Initially, the extracted secreted proteins were fractionated with different saturation levels of (NH4)2SO4 (0–30, 30–60, and 60–90%) and different cutoff ranges of centrifugal ultrafilters (10–30, 30–50, and > 50 kDa) for step by step screening of elicitor-active protein containing fraction(s). Fractionation quality of all these six fractions was subsequently verified by SDS-PAGE (Fig. 1).

Fig. 1
figure 1

Fractionation of secreted proteins of in vitro cultured C. falcatum using ammonium sulphate ((NH4)2SO4) precipitation and size-based ultrafilters

Full size image

As part of the preliminary screening of elicitor-active fraction(s), these six fractions were subjected to a series of bioassays that include quantification and detection of H2O2, the primary messenger of plant defense in sugarcane suspension cells, and observation for the development of HR in N. tabacum. After elicitation of CoC 671 suspension cells with fractionated proteins (250 μg/mL), the level of H2O2 was quantified at different time points. Among the six fractions, a steep increase in H2O2 production was observed with 10–30 kDa and 60–90% (NH4)2SO4 fractions at 30 min post-elicitation, which gradually reduced to reach a plateau after 120 min (Fig. 2a). However, the 30–60% (NH4)2SO4 fraction elicited only half of the level of H2O2 elicited by the above two fractions. In contrast, the 30–50 kDa, > 50 kDa, and 0–30% fractions elicited the least quantity of H2O2 at 30 min and attained the basal level of mock control after 120 min.

Fig. 2
figure 2

Preliminary screening of (NH4)2SO4 and ultrafilter fractions of secreted proteins using a series of bioassays. a Temporal accumulation profile of H2O2 in sugarcane cv. CoC 671 suspension cells after elicitation with fractionated proteins. b Detection of H2O2 production in DAB spiked sugarcane cv. CoC 671 suspension cells at 48 h after elicitation with fractionated proteins. c Screening of fractionated proteins for HR induction in Nicotiana tabacum leaves at 24 h post-infiltration. BTH was used as a positive control to estimate and detect H2O2 levels in a and b, respectively. Zero to 30, 30–60, and 60–90% represent the fractions of (NH4)2SO4 precipitation, whereas 10–30, 30–50, and > 50 kDa represent the fractions of size-based ultrafiltration. For mock control, assay buffer (buffer A) was used

Full size image

When the same batch of suspension cells was stained with DAB at 48 h post-elicitation, similar patterns of H2O2 accumulation level were observed visually (Fig. 2b). When the protein fractions (250 μg/mL) were infiltrated on N. tabacum leaves (a standard experiment to establish defense elicitation properties), only 10–30 kDa, 30–60%, and 60–90% (NH4)2SO4 fractions triggered rapid HR development (Fig. 2c), which were in line with the results of H2O2 assays. A critical observation of SDS-PAGE profiles of these three fractions (10–30 kDa, 30–60%, and 60–90%) revealed a similarity in protein banding pattern in the range of 10–25 kDa. Therefore, the 10–30-kDa fraction that had the least number of total proteins while representing the common proteins (10–25 kDa range) among the three was selected for further fractionation and screening. Since the results of HR assays in the tobacco leaves were correlated with the results of H2O2 assays in sugarcane suspension cells, only HR assay was employed for further screening of elicitor-active protein containing fractions.

Anion exchange chromatography-based separation of the elicitor-active 10–30-kDa ultrafilter fraction resulted in six major peaks (Fig. 3a). The fractions representing these peaks (I, II, III, IV, V, and VI) were separately collected, desalted, and analyzed through SDS-PAGE (Fig. 3b). HR assay of these fractions showed that the fractions “IV” and “V” have triggered HR development (Fig. 3c). However, the extent of HR triggered by fraction “V” was relatively far higher than fraction “IV” and equal to the 10–30-kDa fraction in all three replications. Therefore, the fraction “V” that contained two prominent protein bands was selected for further separation and purification using size exclusion chromatography. The HPLC chromatogram profile of this size-based separation vis-a-vis purification showed two distinct prominent peaks (Fig. 4a). The purity of this separated protein was subsequently verified by SDS-PAGE (Fig. 4b). Evaluation of these two purified proteins on N. tabacum leaves revealed that the protein represented by fraction “I” was responsible for the development of HR (Fig. 4c). Altogether, this systematic molecular sieving process has led to the discovery of an approximately 14-kDa secreted protein that induced defense responses in sugarcane and triggered HR in N. tabacum.

Fig. 3
figure 3

Separation of 10–30 kDa ultrafilter fraction by anion exchange chromatography (AEC) and subsequent screening for HR induction. a Separation of 10–30 kDa ultrafilter fraction containing proteins by AEC using Bio-Scale Mini Macro-Prep High Q Cartridge (Bio-Rad, Laboratories, Inc., California, USA). b SDS-PAGE profile of desalted AEC fractions. c Screening of AEC fractions for HR induction in N. tabacum leaves at 24 h post-infiltration

Full size image
Fig. 4
figure 4

Purification of HR-inducing protein from AEC fraction “V” by size exclusion chromatography (SEC) and subsequent screening for HR induction. a Purification of HR-inducing protein from the elicitor-active AEC fraction “V” by SEC using Agilent Bio SEC-5 μm, 150 Å column. b SDS-PAGE profile of SEC fractions. c Identification of HR-inducing protein fraction through HR assay on N. tabacum leaves at 24 h post-infiltration

Full size image

MS/MS-based identification and in silico analysis of the HR-inducing protein

To identify the HR-inducing protein, the protein spot was subjected to MALDI TOF/TOF MS and the collected spectra were searched against C. falcatum Cf671-specific databases using Peaks 8 studio tool. The search resulted in a unanimous hit of a hypothetical protein with 10 PSMs that had a coverage of 58% and FDR of ≥ 0.0% (Table S2 in the supplementary material). An attempt to annotate this full length hypothetical protein (132 amino acids) using protein BLAST revealed that this protein was novel with not more than 40% similarity with other proteins (Fig. S2a in the supplementary material). The search has indicated that this novel protein was recently identified as one of the high abundant proteins expressed by the virulent light-cultured C. falcatum and provisionally named as second hypothetical protein (HYP2) by our group (Ashwin et al. 2017b). Notably, the next closest hits of this protein were irregularly distributed in all three major divisions of the fungal kingdom, viz., Ascomycota, Basidiomycota, and Mucoromycota (Fig. S2b in the supplementary material). Multiple sequence alignment of few closest hits has indicated that this novel protein contains a novel motif of conserved pattern of six cysteine residues (Fig. S2c in the supplementary material), which suggest that this protein could belong to a novel family of small secreted proteins (SSPs). Owing to this novelty and its functional properties, henceforth, we designate this protein as C. falcatum plant defense-inducing protein 1 (CfPDIP1). Further, gene isolation and analysis indicated that the genomic DNA sequence (NCBI accession code KY178273.1) of this CfPDIP1 gene contains a 5′ untranslated region (1–12 bases) and one single intron of 178 bp size, which starts from 235th and the 412th base position (Fig. S3a in the supplementary material).

Signal peptide analysis of the identified CfPDIP1 protein with SignalP and TargetP tools confirmed its secretory nature and predicted the presence of a signal peptide with the cleavage site between 21st and 22nd residues in the N-terminal region (Figs. S3b and S3c in the supplementary material). The graphical representation of PSM coverage on full length CfPDIP1 hit, as shown by Peaks 8 studio tool, also indicated that they were devoid of this predicted signal peptide region (Fig. 5a). Since the CfPDIP1 protein was identified as one of the high abundant proteins expressed by the virulent C. falcatum pathotype by Ashwin et al. (2017b), the CfPDIP1 protein sequences with and without signal peptide were analyzed with EffectorP tool (Table S3 in the supplementary material). This analysis predicted that both forms of CfPDIP1 proteins could possibly act as effectors during host-pathogen interactions. Altogether, these in silico analyses substantiated the MS-based identification of the HR-inducing protein as novel CfPDIP1 protein and its theoretical properties were found to be corroborated with the observed attributes.

Fig. 5
figure 5

PSM coverage of the identified HR-inducing protein (CfPDIP1) and expression profiling of CfPDIP1 transcript during compatible and incompatible interaction with the sugarcane cultivars CoC 671 and Co 93009, respectively. a PSM coverage of the HR-inducing protein identified by MALDI TOF/TOF MS using Peaks studio 8.0 tool along with the prediction of fixed, variable, and other post-translational modifications. b Temporal quantification of in planta biomass of C. falcatum by absolute quantification method. Superscript a: Y-axis values were broken (vertical bars) to make the small increments in pathogen biomass of Co 93009 is visible. c Temporal expression pattern of CfPDIP1 transcript by comparative CT method. Superscript b: Numerical values in X-axis indicate hours post-inoculation except the in vitro axenic culture of C. falcatum. “0 h” indicates mock inoculated control. Error bars indicate standard deviation of three biological replicates with two technical replicates each

Full size image

Transcriptional profiling of CfPDIP1 expression during compatible and incompatible interactions

CfPDIP1 protein, by virtue of its potential in triggering host defense responses and its predicted effector role, the transcriptional abundance of CfPDIP1, was profiled in planta at different time points during compatible and incompatible interactions with sugarcane. CoC 671 (red rot susceptible) and Co 93009 (red rot resistant) sugarcane cultivars were used to study compatible and incompatible interactions, respectively. Profiling of in planta pathogen biomass showed that the rate of pathogen colonization was relatively far higher in the compatible interaction than the incompatible one (Fig. 5b). Specifically, a distinct elevation in the rate of pathogen colonization was observed since 48 hpi in the compatible interaction. In contrast, in incompatible interaction, the colonization rate was restricted with a relatively far less quantity of pathogen biomass, which was completely vanished at 600 hpi (25 days post-inoculation). Relative expression profiling of CfPDIP1 with respect to pathogen biomass showed that the expression of CfPDIP1 was increased few fold in the incompatible interaction as compared to the compatible interaction between 24 and 72 hpi (Fig. 5c). However, in compatible interaction, the expression level was relatively less but observed throughout the period of colonization. Therefore, the significant differences in the expression level of CfPDIP1 at the early stage of colonization during compatible and incompatible interactions necessitated an in-depth functional characterization.

Putative domain-based dissection and production of recombinant CfPDIP1 proteins

To characterize the functional properties of the novel CfPDIP1 protein, the full length CfPDIP1 protein sequence was putatively dissected into three distinct domains based on the signal peptide and the recently identified truncated isoform of this protein (Ashwin et al. 2017b) (Fig. 6a). Different sets of primers were designed to amplify and clone three different domain deletional variants of CfPDIP1 in the pET28a expression vector, which resulted in the expression of 6X His and T7 fusion tags at the N-terminal end of the target proteins (Fig. 6b). These three domain deletional variants represented full length CfPDIP1 with signal peptide domain (CfPDIP1FL), full length CfPDIP1 without signal peptide domain (CfPDIP1ΔN1-21), and a truncated CfPDIP1 that putatively represents the isoform or the main domain (CfPDIP1ΔN1-45). Interestingly, in addition to the first two variants, the third variant, CfPDIP1ΔN1-45, was also predicted to possess the properties of an effector by EffectorP tool (Table S3 in the supplementary material).

Fig. 6
figure 6

Putative domain-based dissection, in vitro expression and purification of CfPDIP1 protein for functional characterization. a Full length protein sequence of CfPDIP1. Residues in different colors indicate putatively distinct domains. b Dissection of CfPDIP1 protein to generate different domain deletional constructs for in vitro expression and subsequent functional characterization. Lines begin and end with red arrows indicate the respective set of primers used for amplifying distinct constructs of CfPDIP1. c SDS-PAGE profiles of in vitro expressed and purified recombinant CfPDIP1 proteins observed under Coomassie brilliant blue (CBB) staining and western blotting with T7 tag-specific antibody (Merck Millipore). Except the full length CfPDIP1 protein (CfPDIP1FL), domain deletional recombinant CfPDIP1 proteins, viz., CfPDIP1ΔN1-21 and CfPDIP1ΔN1-45, exhibit dimeric and tetrameric forms

Full size image

SDS-PAGE analysis of these purified recombinant proteins showed more than one band, in addition to the expected molecular weight range in CfPDIP1ΔN1-21 and CfPDIP1ΔN1-45 fractions (Fig. 6c). Therefore, the purity of these recombinant proteins was assessed by western blot with T7 tag-specific antibody. The blotting profile revealed that the nonspecific bands observed in the purified CfPDIP1ΔN1-21 and CfPDIP1ΔN1-45 fractions were their respective dimeric and tetrameric forms. Interestingly, the recombinant CfPDIP1FL protein that had the signal peptide at the N-terminal end did not form any dimeric or tetrameric forms as evidenced in the sensitive western blot experiment.

Dose optimization of recombinant CfPDIP1 proteins for bioassays and defense priming

To determine the optimum concentration of recombinant CfPDIP1 proteins for bioassays and priming efficacy evaluations, CoC 671 suspension cells were treated with different concentrations of purified native CfPDIP1 protein and the parameters such as H2O2 accumulation and the concomitant relative percentage of cell viability were measured at 48 h post-treatment (Fig. S4 in the supplementary material). Results indicated that H2O2 induction was inversely correlated with the relative percentage of cell viability. Hence, 50 μg/mL, the minimum CfPDIP1 concentration that induced higher amount of H2O2 production (on an average of 15 μM H2O2) with minimal reduction of cell viability (< 20%), was considered as the optimum dosage concentration and used accordingly for subsequent experiments on functional characterization of recombinant CfPDIP1 proteins.

Functional characterization of recombinant domain deletional variants of CfPDIP1

To ascertain the functional role of CfPDIP1 and to characterize the significance of different domains of CfPDIP1, the recombinant domain deletional variants of CfPDIP1 proteins were subjected to various bioassays using CoC 671 suspension cells and N. tabacum leaves. The purified recombinant CfEPL1ΔN1-92 protein, a PAMP of C. falcatum which was demonstrated to induce systemic resistance in sugarcane by Ashwin et al. (2017b), was used as a positive control in all the experiments carried out hereinafter. The recombinant CfPDIP1-elicited H2O2 accumulation profile showed that CfPDIP1ΔN1-21 followed by CfPDIP1ΔN1-45 induced maximum levels of H2O2 on CoC 671 suspension cells, which was comparable to the positive controls, native CfPDIP1, and CfEPL1ΔN1-92 proteins (Fig. 7a). However, the quantity of H2O2 induced by the full length CfPDIP1 (CfPDIP1FL) was relatively far less than the other two CfPDIP1 proteins. An extracellular alkalinization assay also showed a relatively similar trend of defense elicitation, wherein CfPDIP1ΔN1-21 resulted in rapid alkalinization followed by CfPDIP1ΔN1-45 (Fig. 7b). Specifically, in both the assays, the level of defense response induced by CfPDIP1ΔN1-21 was slightly higher than the purified native CfPDIP1 protein but apparently lesser than CfEPL1ΔN1-92 protein. Besides, histochemical detection of the degree of H2O2 accumulation in suspension cells at 48 h after elicitation also showed a similar trend of defense elicitation by the recombinant domain deletional CfPDIP1 variants (Fig. 7c). Precisely, among the recombinant CfPDIP1 proteins, the maximum amount of brown precipitation was observed with CfPDIP1ΔN1-21, which was comparable with the native CfPDIP1 and CfEPL1ΔN1-92 proteins.

Fig. 7
figure 7

Elicitation of sugarcane suspension cells (cv. CoC 671) with recombinant CfPDIP1 proteins. a Temporal accumulation profile of H2O2 in CoC 671 suspension cells after elicitation with recombinant CfPDIP1 proteins. b Induction of extracellular alkalinization by recombinant CfPDIP1 proteins. c Microscopic observation of H2O2 accumulation in CoC 671 suspension cells by DAB staining at 48 h after elicitation with recombinant CfPDIP1 proteins. Bar indicates 100 μm. For mock control, assay buffer was used. Native CfPDIP1 indicates purified native CfPDIP1 protein fraction. Purified recombinant CfEPL1ΔN1–92 protein (a PAMP) of C. falcatum was used as a positive control

Full size image

In contrary to the results of sugarcane suspension cell bioassays, which showed that CfPDIP1ΔN1-21 distinctively induced maximum degree of defense, HR assay on N. tabacum leaves showed that all the three recombinant CfPDIP1 proteins rapidly triggered HR (Fig. 8a). In fact, the magnitude and rate of HR development induced by these three recombinant proteins were on par with the native CfPDIP1 and CfEPL1ΔN1-92 proteins. Further, DAB staining of these infiltrated leaves at 24 hpi showed a distinctively higher production of H2O2 than the mock control (Fig. 8b). However, callose deposition assay revealed that the amount of callose deposits induced per unit area has significant variations among the three recombinant proteins (Fig. 8c). Particularly, the number of callose deposits observed in CfPDIP1ΔN1-21 was higher than the other two variants and native CfPDIP1 proteins, even though it was lesser than the positive control (CfEPL1ΔN1-92) in all the three replications (data not shown). Since CfPDIP1ΔN1-21 was found to trigger maximum defense responses among the three recombinant CfPDIP1 proteins in most of the bioassays, it was used for further defense priming efficacy-related experiments.

Fig. 8
figure 8

Induction of defense responses by recombinant CfPDIP1 proteins on N. tabacum leaves. a Detection of HR induced by recombinant CfPDIP1 proteins in N. tabacum leaves at 24 and 48 h post-infiltration. b Microscopic observation of H2O2 accumulation in N. tabacum leaves by DAB staining at 24 h after infiltration with recombinant CfPDIP1 proteins. c Microscopic observation of recombinant CfPDIP1 proteins induced callose depositions (bright white spots) in N. tabacum leaves by aniline blue staining method at 24 h post infiltration. Bars indicate 100 μm. For mock control, assay buffer was used. Native CfPDIP1 indicates purified native CfPDIP1 protein fraction. Purified recombinant CfEPL1ΔN1-92 protein (a PAMP) of C. falcatum was used as a positive control

Full size image

Effect of priming and co-infiltration of CfPDIP1∆N1-21 on sugarcane leaves

To characterize the functional role of CfPDIP1∆N1-21 during sugarcane-C. falcatum interaction, CfPDIP1∆N1-21 primed and unprimed CoC 671 leaves were evaluated by detached leaf assay method by inoculating C. falcatum (Cf671) spores. Similarly, to assess the overexpression effect of CfPDIP1∆N1-21 during host-pathogen interaction, unprimed CoC 671 leaves were co-infiltrated with CfPDIP1∆N1-21 mixed Cf671 spores, in order to mimic overexpression. The results of these experiments showed that CfPDIP1∆N1-21 priming has reduced the size of disease (lesion) development on par with the PAMP, CfEPL1ΔN1-92 (Fig. 9a and Table S4 in the supplementary material). However, the imitation of overexpression of CfPDIP1∆N1-21 enhanced the rate of disease development, when compared to the unprimed pathogen inoculated control. Microscopic observations showed that the overexpression effect of CfPDIP1∆N1-21 has intensified the number of appressorial structures (a specialized infection structure that facilitates penetration into the host cells), which in turn increased the rate of acervuli formations (an asexual fruiting body producing spores) (Fig. 9b and Table S4 in the supplementary material). On the other hand, a relatively smaller number of appressoria and very few pre-emergence stage acervuli were observed on CfPDIP1∆N1-21 primed leaves. Overall, the priming and co-infiltration assays indicated that the CfPDIP1∆N1-21 protein could indeed possess a distinct functional property, in addition to the properties of an elicitor.

Fig. 9
figure 9

Co-infiltration and priming effects of CfPDIP1∆N1-21 on sugarcane leaves by detached leaf assay method. a Macroscopic observation of leaf bioassay at 72 h post-inoculation. Brown and yellow lesion indicates the extent of pathogen colonization and their consequent senescence effect, respectively. b Microscopic observation of leaf bioassay at 72 h post-inoculation. Bar indicates 50 μm. Red arrows indicate emerged and emerging acervuli structures whereas dark brown spots indicate appressorial structures of C. falcatum. Purified recombinant CfEPL1ΔN1-92 protein (a PAMP) of C. falcatum was used as a positive control

Full size image

Up-regulation of defense-related genes and concomitant suppression of disease severity in CfPDIP1∆N1-21 primed canes

To evaluate the efficacy of CfPDIP1∆N1-21 priming against red rot at molecular level, the biomass of C. falcatum and the concomitant expression of few important candidate defense-related genes were profiled at different time points (0, 12, 24, 48, 72, 120, 600 hpi) of host-pathogen interaction using qPCR. The pathogen biomass quantitation profiles which otherwise indicate the degree of pathogen colonization and disease severity showed a significant reduction in the amount of pathogen biomass in CfPDIP1∆N1-21 primed canes, since 72 hpi (Fig. 10a). Interestingly, the degree of disease suppressive effect observed with CfPDIP1∆N1-21 priming was on par with the priming with the PAMP, CfEPL1ΔN1-92.

Fig. 10
figure 10

Quantitation of C. falcatum biomass and relative expression analysis of potential candidate defense-related genes of CfPDIP1∆N1-21 primed sugarcane stalks. “0 h” indicates primed mock inoculated control. Purified recombinant CfEPL1ΔN1-92 (a PAMP) primed canes were used as a positive control

Full size image

Further, the expression of five potential candidate defense-related genes or putative defense markers, viz., nonexpressor of pathogenesis-related protein 1 (ScNPR1), isochorismate synthase (ScICS), β-1,3-glucanase D (ScPR2), chitinase VII (ScPR3), and defensin 5 genes were profiled (Fig. 10b–f). Interestingly, the expression profiles showed that all these five genes were up-regulated in response to CfPDIP1∆N1-21 priming at 0 hpi (pre-inoculation stage). In other words, the basal expression of these genes was induced in response to priming, even before pathogen inoculation. More specifically, the expression of ScICS and ScNPR1, the master regulators of major defense pathways, was significantly up-regulated at the earlier time points (Fig. 10b, c). Similarly, the expression of ScPR2, ScPR3, and defensin 5 genes remained to be up-regulated until 48 hpi in CfPDIP1∆N1-21 primed samples (Fig. 10d–f). Overall, the expression pattern of most of these defense genes in CfPDIP1∆N1-21 was apparently similar to CfEPL1ΔN1-92 primed samples.

Discussion

Identifying the molecular determinants of plant-pathogen interactions, especially, PAMPs/elicitors, effectors, and R genes, is the key to elucidate the mechanisms of pathogenicity and disease resistance, which would aid in developing sustainable crop protection and improvement programs (Wirthmueller et al. 2013; Ashwin et al. 2017c). In this context, mining of secreted proteins through proteomic approaches is gaining more significance as it represents the functional biomolecules of living system (Ashwin et al. 2017c). Recently, we have analyzed the secretome profiles of light-cultured (high virulent) and dark-cultured (less virulent) C. falcatum by using two dimensional electrophoresis (2DE) coupled with MALDI TOF/TOF and found that CfEPL1 a well-conserved PAMP protein was highly abundant in the less virulent C. falcatum, whereas many pathogenicity-related proteins were highly abundant in the high virulent C. falcatum (Ashwin et al. 2017b). We further reported that the major cerato-platanin domain (CfEPL1ΔN1-92) of CfEPL1 protein was essential to induce systemic resistance in sugarcane.

In the present study, we systematically screened the in vitro secretome of C. falcatum (high virulent) using top-down approach, a strategy similar to the principle of molecular sieving to identify the protein(s) that could sensitize and trigger plant defense responses. A similar kind of top-down strategy was adopted by many researchers for more than a decade. For instance, the proteinaceous elicitors such as cerato-platanins (Pazzagli et al. 1999), PevD1 (Wang et al. 2012), and MoHrip1 (Chen et al. 2012) were purified from the culture filtrates of fungal pathogens, whereas the elicitors such as Sm1 (Djonovic et al. 2006) and AsES (Chalfoun et al. 2013) were purified from the culture filtrates of symbiotic plant microbes and saprophytes, respectively. This top-down strategy is in contrast to our previous study (Ashwin et al. 2017b), wherein the PAMP, CfEPL1 was characterized using a bottom-up approach.

H2O2, the predominant nonradical of reactive oxygen species (ROS), is regarded as one of the important secondary messengers and sometimes as a primary messenger of defense signaling in plants against any stress and in response to the perception of various PAMPs (elicitors) and effectors (Møller and Sweetlove 2010). ROS production in turn triggers a complex network of defense signaling pathways to act against the perceived invasion through extracellular/apoplastic alkalinization, phytoalexin production, reinforcement of structural barriers like callose deposition in cell wall, programmed cell death (HR), etc. (Barna et al. 2012; Lehmann et al. 2015). Hence, in this study, the quantitative and qualitative analyses of H2O2 production and the observation of HR induction were considered as the indices of defense elicitation in the bioassays. Since our goal has been to shortlist the number of proteins to be screened in each step, the 10–30-kDa fraction that had less number of proteins but similar to other two active fractions on that range was selected for further screening. In subsequent fractionation and separation of 10–30 kDa fraction using AEC and SEC, a single protein of approximately 14 kDa protein that could trigger HR in N. tabacum leaves was purified.

Identification of this purified HR-inducing protein using MALDI TOF/TOF and subsequent in silico analysis revealed that it is a novel hypothetical protein, discovered recently by us as one of the high abundant proteins in the secretome profile of high virulent C. falcatum with a provisional name of HYP2 (Ashwin et al. 2017b). Besides, in that previous study, we have reported a truncated isoform of this CfPDIP1 protein of approximately 10 kDa. However, an in-depth transcriptional expression analysis indicated that the isoform variation was only at the protein level and not due to alternative splicing as supposed to be due to unknown reasons. Hence, in the present study, the coverage of the peptide spectral matches (PSMs) was analyzed against the full length CfPDIP1 protein sequence derived from the C. falcatum database and found that the sequence of purified HR-inducing protein matched with the full length CfPDIP1 protein, excluding the predicted signal peptide region. Particularly, the PSMs have covered the truncated portion (21st–45th amino acids) of the second isoform. Besides, the observed molecular weight of the HR-inducing protein closely matched with the high abundant CfPDIP1 protein than the truncated variant of CfPDIP1. With all these analyses, we have established that the HR-inducing protein identified in this study was the high abundant full length (except signal peptide) isoform of CfPDIP1 protein. Since the signal peptides of most of the classically secreted proteins get cleaved while passing through the membrane, the PSMs of HR-inducing protein did not cover the signal peptide region of the full length CfPDIP1 protein (Peberdy 1994; Emanuelsson et al. 2007).

EffectorP tool predicted that all the three variants (domain deletional) of CfPDIP1, viz., CfPDIP1FL (full length with signal peptide), CfPDIP1ΔN1-21 (full length without signal peptide), and CfPDIP1ΔN1-45 (the truncated isoform), have greater probability to be classified as effectors. However, transcriptional profiling showed that the expression of CfPDIP1 in the compatible interaction was few folds lesser than the incompatible interaction, and the expression was constitutive throughout the phases of colonization. In general, Colletotrichum spp. follow three different models of hemibiotrophic colonization during their interaction with different hosts. Among the three, C. falcatum follows the Colletotrichum graminicola lifestyle model of intracellular hemibiotrophy, wherein which the biotrophy persists at the progressively colonizing cells, while the necrotrophy confined at the center of colonization (Crouch et al. 2014). The different phases of this lifestyle comprehend the initial infection phase that encompasses germination, appressorial formation, and penetration (0–24 h), a biotrophic phase that represents the formation of primary hyphae (24–48 h), and a necrotrophic phase that includes the brief transition from primary to secondary hypha and subsequent colonization of nearby cells (> 48 h) (O’Connell et al. 2012; Gan et al. 2012). Therefore, the ubiquitous expression of CfPDIP1 has suggested that it could be an essential biomolecule of C. falcatum and might be required throughout its life span under both in vitro and in planta conditions, which substantiated the properties of a PAMP rather than the properties of an active secreted effector. However, the spike in the expression level of CfPDIP1 at earlier stages of host-pathogen interaction than the axenic culture indicated that the CfPDIP1 protein may play a potential role in the dynamic host-pathogen interaction. Many filamentous fungi, including the Colletotrichum spp., do secrete a range of effector-like proteins even before the contact with plant surface or throughout the life cycle (O’Connell et al. 2012; Kleemann et al. 2012; Lanver et al. 2014; Guzman-Guzman et al. 2017). These kinds of proteins act like the hydrophobins or repellants and protect the hyphal structure from the attacks of the plant defense molecules during colonization (Ruocco et al. 2015; Lo presti et al. 2015). Altogether, the expression analyses have substantiated the elicitor properties of CfPDIP1 protein and necessitated an elaborate functional characterization.

In vitro expression and purification of three different domain deletional variants of CfPDIP1 showed that a certain quantity of the purified recombinant CfPDIP1ΔN1-21 and CfPDIP1ΔN1-45 proteins tends to assemble into dimeric and tetrameric forms. This assembling property has continued even after purification of their respective monomeric forms using SEC. Further, the increment of the concentration of reducing agents did not reduce them to their monomeric forms (data not shown). A similar phenomenon was also reported earlier with the recombinant CfEPL1 proteins of C. falcatum (Ashwin et al. 2017b). On the contrary, CfPDIP1FL protein did not form any dimeric or tetrameric forms. This was probably because of the presence of 21 residue signal peptide at the N-terminal end that might have prevented the formation such structures. Therefore, based on the results of dose optimization experiment, 50-μg/mL concentration of the purified recombinant proteins that had both dimeric and tetrameric forms was used for subsequent bioassays. For dose optimization, the maximum generation of H2O2 with minimum amount of cell death was used as the criteria because generally the degree of ROS production has been directly proportional to and associated with cell death (Wang et al. 2016; Chen et al. 2012, 2014). Moreover, it is recommended to use the minimal concentration of elicitors for host defense priming, as higher concentrations may have some detrimental effects on plant metabolism, which would also impact on fitness cost (Thakur and Sohal 2013; Martinez-Medina et al. 2016).

Functional bioassay of purified recombinant CfPDIP1 proteins on CoC 671 suspension cells indicated that only CfPDIP1ΔN1-21 followed by CfPDIP1ΔN1-45 triggered rapid extracellular alkalinization and maximum amount of H2O2 production. In contrast, CfPDIP1FL elicited only the least amount of defense responses. These observations suggested that the presence of the signal peptide in CfPDIP1FL might have either affected its function or hindered its perception as an elicitor in sugarcane. Unlike suspension cell assays, the bioassays on N. tabacum leaves such as HR assay and DAB staining of H2O2 did not show significant difference in the efficacy of defense elicitation among the CfPDIP1 variants. Only the callose deposition assay showed that the number of callose deposits in CfPDIP1ΔN1-21 was slightly higher than the other two variants. Here, the observed differential responses with respect to different hosts reinstate the fact that the perception of an elicitor and subsequent induction of defense, either PTI or effector-triggered immunity (ETI), are genotype-specific and may vary from plant to plant (Boller and Keen 1999; Thomma et al. 2011). Altogether, the functional characterization of the three domain deletional CfPDIP1 proteins indicated that the signal peptide region (1–21 residues) was dispensable, but the intermittent domain (22–45 residues) and the main domain (46–132 residues) were essential for induction of maximum defense responses in sugarcane. Therefore, the variant CfPDIP1ΔN1-21 that comprises of both the intermittent as well as the main domains was used for further defense priming assays.

Disease suppressive effect observed with CfPDIP1∆N1-21 priming on sugarcane leaves has substantiated the earlier reports stating that without the real-time confrontation of a pathogen, perception of PAMPs and certain effectors can activate host defense and promote resistance, which would help to overcome or suppress subsequent attack from the pathogen (Wiesel et al. 2014; Burketova et al. 2015). Even though the co-infiltration of CfPDIP1∆N1-21 on unprimed leaves enhanced disease susceptibility, it cannot be attributed to the properties of an effector, because of its constitutive and higher expression during incompatible interaction. Similar effector-like properties have been reported with the secreted hydrophobins of many filamentous fungi. These cysteine-rich, constitutively secreted proteins are essential for attachment to different surfaces and play a major role in host-pathogen interactions (Guzman-Guzman et al. 2017). Presumably, CfPDIP1 may also have similar structural adhesion or formation related roles during the plant-pathogen interaction, in addition to the properties of being an elicitor. Nevertheless, further experiments like ectopic expression, deletion, and complementation of the CfPDIP1 gene only can decisively confirm whether CfPDIP1 may play the role of effector or not.

Quantitative assessment of pathogen biomass and transcriptional profiling of potential candidate defense markers (ScNPR1, ScICS, ScPR2, ScPR3, and defensin 5) indicated that CfPDIP1ΔN1-21 priming has indeed activated the host defense pathways even before the pathogen inoculation and resulted in the reduction in pathogen biomass in the distal stalk tissues. This otherwise indicated that the foliar priming has activated systemic resistance in stalk tissues and suppressed disease severity as in the case of the PAMP, CfEPL1ΔN1-92 during sugarcane-C. falcatum interaction (Ashwin et al. 2017b). Since the expression of ScNPR1, the master regulator of major defense-related pathways, is associated with the downstream expression of PR proteins and other defense-related proteins, the defense markers ScPR2 (β-1,3-glucanase), ScPR3 (chitinase VII), and defensin 5 might have got up-regulated in succession with ScNPR1 expression (Pajerowska-Mukhtar et al. 2013; Ali et al. 2017).

Comprehensively, in this study, we report the purification and functional characterization of a novel secreted protein, CfPDIP1, which could act as an elicitor upon priming and might play an important role during the sugarcane-C. falcatum interaction. To our knowledge, this is the first study demonstrating the elicitation property of a novel class of protein from Colletotrichum spp. Since the novel CfPDIP1 protein was found to induce host resistance in a red rot susceptible sugarcane cultivar, CoC 671, the prospect of similar or even higher degree of defense elicitation effect in red rot resistant and field tolerant sugarcane cultivars is huge. On a futuristic perspective, the identification of the interacting partners of CfPDIP1 during in planta interaction would help to delineate the infection strategy of C. falcatum and would decipher the mechanism of PTI in sugarcane.