Introduction

Cochliobolus lunatus (Nelson and Haasis) also known as Curvularia lunata (Wakker) Boed in recent times is known as one of the most aggressive and devastating plant pathogen causing disease in important food crops such as sorghum (Sorghum bicolor), wheat (Triticum aestivum), rice (Oryza sativa L), maize (Zea mays) and potato (Solanum tuberosum L.) (Agrios 2005; Iftikhar et al. 2006; Gao et al. 2012; Louis et al. 2013). Past disasters caused by Cochliobolus species include the Great Bengal famine where more than 2 million people starved to death due to reduced rice yield (Scheffer 1997), corn blight epidemics in United States of America and China in 1970s and 1990s, respectively (Ullstrup 1972; Dai et al. 1995).

Morphologically, C. lunatus produces curved conidia, a broad rounded apex cell, a distinct swollen central cell, 4–5 distinct septa and sometimes conidia also have oval stromata (Bengyella et al. 2014). C. lunatus exhibits both hemibiotrophic and necrotrophic lifestyle during interaction with potato (Louis et al. 2015). During interaction with potato leaf, conidia of C. lunatus undergo both bipolar and unipolar germination to produce minute infection hyphae that could differentiate to appressoria-like structures, and hyphae often exhibit stomatal atropism (Louis et al. 2015). Globally, increasing pathogenic variants of C. lunatus and host range (Bengyella et al. 2014) have raised the curiosity among scientist to dissect C. lunatus proteome (Xu et al. 2007; Gao et al. 2012) and secretome (Louis et al. 2014) for pathogenicity factors. Although the proteomic studies were vital in decrypting the plethora of virulence factors, little is known on host response, as well as whether C. lunatus modulates the expression of nonexpresser pathogenesis-related protein 1 (herein refer as StNPR1, which governs the expression of pathogenesis-related proteins (PrPs)) in order to outwit systemic acquired resistance (SAR).

StNPR1 is an ankyrin repeats protein which is a key activator of PrPs that can equally activate TGA transcriptional factors to further trigger PrPs expression (Cao et al. 1997; Dong 2004; Rochon et al. 2006). Understanding how C. lunatus modulates the proteome and StNPR1 in potato could allow for designing management strategies based on StNPR1 inducers for effective priming SAR, thus, reducing crop losses from brown-to-black spot disease. Potato is produced and consumed worldwide and largely consider as “food-for-the-future” (http://www.fao.org/newsroom/en/news/2008/1000816/). Since it is known that C. lunatus adopts different strategies to invade potato cultivars at the microscopic level (Louis et al. 2015), the aim of this present study was (1) to identify important functional proteins differentially modulated during infection, and (2) to investigate whether C. lunatus affects StNPR1 expression during infection. It was found that C. lunatus differentially suppressed potato proteome at late stages of infection and transiently down-regulated the expression of StNPR1 during the infection process.

Materials and methods

Plant material and growth conditions

Disease-free potato (cv. Kufri Jyoti) tubers were purchased from Central Potato Research Institute (CPRI, Shimla, India), which is Phytopthora infestans resistance near-isogenic lines. The tubers were grown in U-CON250 plant growth chamber (Danihan Labtech Co., Ltd) under 16 h photoperiod at the light intensity of 180 μmol m−2 s−1 and 80 % relative humidity (RH). The growth was maintained in a 90 cm × 70 cm × 12 cm capacity trays as microfield containing a mixture of autoclaved rice stalk vermicompost and sand (1:2 %w/w). Each plant was amended with 1 g organic fertilizer (BIOVITA™, PI industries Ltd, India) after 2 weeks of sprouting.

Pathogen inoculation

C. lunatus (GenBank accession JX907828) at the concentration of 106 conidia per ml from a 14 days old V8 agar medium plate was used. The conidial suspension was uniformly sprayed on 3-week old plants (with third fully developed terminal leaflets) using a compressed air hand sprayer till run-off. Control plants were misted with sterile water only. The experimental set-up was maintained at 20 °C and 100 % RH in the growth chamber. Infectivity of the plants was confirmed microscopically using a microscope coupled with DP7M5.0.0.5 software and an Olympus DP70 camera (Olympus BX61®, USA). The experiment was performed in replicate and repeated three times.

Protein extraction and quantification

Leaf samples were randomly collected by 24, 48, 72 and 96 h after inoculation (hai) for challenged and control plants at the same time point. One gram sample was immediately flash frozen in liquid nitrogen and crushed into fine powder. The proteins were extracted on ice in a 9 mM CaCl2 solution (containing 5 %v/v Triton‒X‒100, 0.2 % w/v polyvinylpyrrolidone and 0.2 %w/v dithiothreitol (DTT), 0.2 % L-ascorbic acid), pH 4.8 as earlier described (Waikhom et al. 2015). Proteins from three biological repeats were pooled and washed using Clean-Up Kit (Bio-Rad® laboratories, CA, USA) following the manufacturer protocol. For all downstream analysis, proteins were quantified using the Bradford dye-binding assay (Bio-Rad, CA, USA).

Expression of StNPR1 by Western blotting

Thirty-five microgram samples were profiled by SDS-PAGE (5–13 % polyacrylamide gel, Sigma) and transferred to a PVDF membrane with a Trans-Blot® Turbo™ transfer system (Bio-Rad®, CA, USA). Membranes were incubated with 1:10,000 diluted Arabidopsis thaliana (AtNPR1) primary antibody (kindly provided by Prof. Xiannian Dong, Duke University, USA) overnight at 4 °C. After a series of washing, blots were further incubated with 1:50,000 diluted HRP-conjugated secondary antibody (Bio-Rad®, CA, USA) for 1 h at room temperature. The signals were detected by chemiluminescent kit and imaged in ChemiDoc XRS (Bio-Rad®, CA, USA). The densitometry analysis was performed and may not certainly reflect the expression threshold of target protein signal and statistically analysed in R tool using one-way Anova test for the levels of significance.

Two-dimensional gel electrophoresis for proteome profiling

Cleaned proteins were suspended in ReadyPrep rehydration buffer (containing 8 M urea, 50 mM DTT, 2 % 3-[(3-cholamidopropyl)dimethylaminio]-1-propanesulphonate, 0.2 % (w/v) Biolyte® 3/10 ampholytes, and bromophenol blue trace; Bio-Rad®, CA, USA) and quantified by the dye-binding method. Isoelectric focusing (IEF) was performed in Protean®i12™ IEF CELL apparatus (Bio-Rad®, CA, USA). After preliminary testing of the immobilized pH gradient (IPG) strips of different pH ranges, we found that only pH 4–7 produced the best resolution. A 7 cm IPG dry strips with a linear pH range of 4‒7 was rehydrated passively with 180 μg protein aliquots for 16 h and run as previous described (Louis et al. 2014). Following IEF, IPG strips were equilibrated for 45 min at room temperature with equilibration buffer I (6 M urea, 2 % SDS, 0.375 M Tris–HCl pH 8.8, 20 % glycerol and 56 mM dithiothreitol (DTT)), and then transferred to equilibration buffer II (6 M urea, 2 % SDS, 0.375 M Tris–HCl pH 8.8, 20 % glycerol and 3.5 mM DTT). In the second dimension, that is 1-D, the proteins were resolved on 15 % polyacrylamide ultrathin gels (7.3 cm × 8 cm × 0.75 mm) at 80 V using Mini-Protean®TetraCell (Bio-Rad®, CA, USA). All gels were stained with 0.35 % w/v extrapure Coomassie Brilliant-blue R250 at 38 °C overnight. Gels were destained in a solution containing 7 % acetic acid and 35 % methanol till spots were visible, imaged and repeated three times per sample.

Identification of the host metabolic pathway significantly affected by infection

To determine whether the infection process alters the host proteome, images were processed in Progenesis SameSpot 4.1 (Total-Lab, USA) as follows. Following quality control of 2-D gels, only spots with pixel-to-pixel intensity ≥120 matching each other in three independent gels per sample were normalization and spots counts were performed. To have an in-depth view in variations in proteome, generated data were analysed in QlucoreOmics Explorer 2.3 software. Two group comparison (t–test) was applied for heatmap hierarchical clustering at t 141 ≤ 1.347, R ≤ 0.113, q = 0.450, p = 0.91 and filter by variance (σ/σ max = 0.053) cut-off value for control and challenge proteome at 96 hai. To identify which component of the host metabolic pathway was the most affected by the infection process, spots with fold expression (FE) ≥3.0 at Anova value P < 0.05, and judged to significantly vary at 24, 48, 72 and 96 hai compared to control were sorted, excised and submitted for identification.

Peptide spots were trypsinized and protein elution was performed as earlier described (Shevchenko et al. 2006) and peptide solution was concentrated in a vacuum centrifuge to approximately 10 μl volume. Four microlitre of trypsinized peptide solution was mixed with α-cyano-4-hydroxycinnamic on target plate of matrix-assisted laser desorption/ionization time-of-flight/time-of-flight mass spectrometry (MALDI-TOF/TOF MS) and allow to crystalize at room temperature. The analysis was performed in a MALDI-TOF 4800 proteomics analyser (Applied Biosystem, MDS SCIEX4800, USA). The study was performed in a positive reflectron mode at 25 kV, 50 ns delayed extraction time and mass spectra were acquired in the default mode and generated data were processed using DataExplorer 5.1 (Applied Biosystems®, USA). Spectra peak lists obtained were processed through in-house MASCOT search engine (v.2.3 Matrix Science, London, UK) against all new entries in the green plant MSDB, NCBInr databases and mapped to phytozome 11 (S. tuberosum v3.4 genome). Search parameters were: trypsin as enzyme, variable modifications like carbamidomethyl, fixed modification for cysteine residues, up to one cleavage, error tolerance of ±100 ppm, ±0.2 Da, oxidation of methionine and N-terminal acetylation were taken into consideration. At a MOWSE probability threshold score (P < 0.05) peptide were considered, and false-discovery rate (FDR; Elias et al. 2005) was calculated using Decoy database (http://www.matrixscience.com/help/decoy_help.html) and validated at FDR <1 %.

Gene expression analysis during infection

RNA was isolated by using TRI reagent kit according to the manufacturer’s instructions (Sigma, St. Louis, MO, USA). The quality and quantity of RNA were evaluated using a 1 % agarose gel and nanoDrop spectrophotometer (BioSpecNanodrop™, Shimadzu), respectively. cDNA was synthesized with Invitrogen kit according to the manufacturer instructions (Invitrogen™ Life Science Software, Invitrogen Inc.), and the quality and integrity of cDNA were verified by performing PCR with 26S rRNA using the primer set 26S-F: 5′-cacaatgataggaagagccgac-3′ and 26S-R: 5′-caagggaacgggcttggcagaatc-3′ (Fig. S3). Real-time quantitative polymerase chain reaction (RT-qPCR) was performed using the iScript™ One-Step RT-PCR Kit with 2X SYBR® Green buffer (Bio-Rad, USA) containing 0.4 mM of dNTP [of dATP, dCTP, dGTP, dTTP], magnesium chloride, iTaq DNA polymerase, 20 nM fluorescein, SYBR® Green I dye), stabilizers nuclease-free H2O, 10 fold diluted total cDNA, 0.3 μL of 10 μM of each primer—StNPR1-F: 5′-gctgaaatgtcagggtga-3′, StNPR1-R: 5′-cagcgtaatgaagagcgta-3′ (product size 145 bp based on GenBank accession AY615281). Amplification conditions were as follows: 95 °C for 10 min, followed by 40 cycles of 95 °C for 10 s, 54 °C for 30 s and 72 °C for 30 s. The potato β-tubulin gene Z33382 was included as internal positive control using the primer set Tu-F: 5′-tctgcaaccatgagtggtgt-3′ and Tu-R: 5′-atgttgctctcggcttcagt-3′. Negative controls (without cDNA templates) containing primers were run to check the absence of primer dimers and contamination. Also, primers (Table S1) representing genes encoding for the peptide spots that were significantly affected (at FE ≥ 3) at 24, 48, 72, and 96 hai during the infection process were validated by RT-qPCR. The primers were designed using OligoPerfect™ Designer software (Invitrogen™ Life Science Software, Invitrogen Inc.). The amplification kinetics was determined as mean of three independent experiments. The relative gene expression was computed using fold induction equation = 2−∆∆Ct as previously described (Trouvelot et al. 2008). The experiment was performed in StepOne™-StepOnePlus™ Real-Time PCR Systems (Applied Biosystem®, Life Technologies) and all data subjected to t test for the level of significance.

Results and discussion

lunatus suppresses host proteome during infection

A key area for understanding the mechanism employed to overcome potato defence is to study the changes in functional proteome since vital information on host physiological process could be derived. C. lunatus is a rapid colonizer of potato and by 24 h after inoculation (hai), we confirmed pathogenic relationship was established (Fig. 1a). As previously shown (Louis et al. 2015), the development of brown-to-black spots disease was hallmarked by minute colonizing hyphae from germinated conidia (Fig. 1a, b). Symptoms of C. lunatus infection in potato leaf consist of yellowing of leaf, water-soaked brown-to-black spots (Fig. 1b) and wilting at the advanced stage of infection. To gain insights whether C. lunatus invasion influences the host response at the proteome level, protein samples collected at 96 hai and control plant were comparatively profiled. Prominently, we found that by 96 hai, proteins within 30.91 kDa limit were suppressed, whereas proteins within 14.31 kDa were either over-expressed or induced (Fig. 1c). This result showed that C. lunatus modified host proteome during the infection process and warranted in-depth investigation into the functional proteins most affected during infection.

Fig. 1
figure 1

a Invasion of potato leaf by C. lunatus at 12 hai stained with trypan-blue-chlorazol black E-KOH as previously described (Louis et al. 2015) and observed with a bright-field Olympus BX61 microscope coupled with DP7M5.0.0.5 software. b Brown-to-black spot disease of potato at 96 hai and the close-up represents micrograph of C. lunatus at resting state. c Crude protein (35 μg) extracted from potato leaf cv. Kufri Jyoti, lanes from left to right, molecular weight markers, control plants mist with sterile water and plants challenged with C. lunatus for 96 h. The broken arrows indicated prominent differentially expressed peptides bands

Full size image

Two-dimensional gel electrophoresis (2-D) was performed to unravel the extent in protein alteration during the colonization process. Proteome maps at 24, 48, 72 and 96 hai were generated and compared against controls at the same time point. As observed using one-dimension gel electrophoresis (Fig. 1c), 2-D map revealed 97 spots in C. lunatus–challenged plants at 96 hai against 143 spots in control plants confirming an overall down-regulation in protein expression during infection at that time point (Fig. 2). Furthermore, mean changes in the number of peptide spots were as follows, 24 hai (307 spots, Fig. S1), 48 hai (274 spots, Fig. 3), 72 hai (269 spots, Fig. S2). These spots were reproducibly observed in 2-D independently run experiments. This analysis reinforces the evidence that C. lunatus suppressed potato leaf proteome during infection.

Fig. 2
figure 2

a 2-D map for control potato leaf proteome with 213 spots, b 2-D map for potato plants challenged for 96 hai with 97 spots. c Heatmap was generated in QlucoreOmics Explorer 2.3 depicting hierarchical clustering of proteome changes between control and challenged plants based on normalized relative spots volume (P < 0.05). Each line corresponds to a spot and unique spots with fold expression >2.5 are conspicuously represented

Full size image
Fig. 3
figure 3

2-D proteome map of potato leaves at 48 hai of C. lunata. Spots of interest with fold expression >3 are indicated with arrows. The immobilized pH gradient scale and Precision Plus Protein™ WesternC™ Standards are shown

Full size image

In order to determine the most prominent component of the proteome affected by C. lunatus infection, 2-D profiles were compared in Progenesis SameSpot 4.1 suite and peptide spots with mean fold expression at all time points ≥3 at Anova P < 0.05 were sort out and tagged (Figs. 2, 3, S1, S2). The expression profiles of the tagged spots are provided (Fig. 4a), and their identity following MALDI-TOF/TOF MS is shown (Table 1). Peptide spot# 1 was identified as ribulose bisphosphate carboxylase enzyme (EC 4.1.1.39; RUBISCO) which catalyses the rate determining step of photosynthesis and was down-regulated at the onset of the infection process (Fig. 4a), nonetheless, up-regulated at 96 hai. Peptide spot# 2 was identified as plastidic aldolase enzyme which catalyses the formation of triose phosphate, and whose expression was high at 48 hai. Peptide spot# 3 was identified as alcohol dehydrogenase 2 which catalyses the interconversion of aldehydes, ketones and alcohols and was differentially expressed at all time points. Peptide spot# 4 was identified as photosystem II protein, a light harvesting protein, whose expression level was high at 48 hai. Considering spot# 3 and 4 expression levels decreased steadily as from 48 hai (Fig. 4a), it is not unreasonable to propose that carbohydrate metabolism is severely affected during infection. Previously, attempt to unravel maize resistance to C. lunata using proteomics approach led to cataloguing of 100 differentially expressed spots, and identification of eight selected spots, among which two oxygen-evolving enhancer proteins were identified (Huang et al. 2009; Pechanova and Pechan 2015), akin to photosystem II protein identified herein (spot# 4). Other studies have suggested that photosystem II proteins are responsive to biotic and abiotic stresses (Sugihara et al. 2000; Wu et al. 2013). Peptide spot# 5 was identified as phosphoinositide phosphatase SAC9, whose expression was steadily up-regulated (Fig. 4a). Phosphoinositide phosphatase SAC9 is involved in modulating phosphoinositide signalling pathways during stress response (Williams et al. 2005). Overall, the strongest host response following C. lunatus infection was observed at 24 hai hallmarked by 307 peptide spots (Fig. 4b).

Fig. 4
figure 4

aView of differentially expressed peptide spots significantly and severely affected during infection in part per million (PPM) of relative density (RD). The graph represents mean of three biological replicates and bars are mean standard error, and asterisk indicates the level of significance. b A global view for mean peptide spots in potato leaf proteome during C. lunatus infection. Asterisks indicate levels of significant differences in expressed peptide spots (*P < 0.05; **P < 0.01; ***P < 0.001)

Full size image
Table 1 Details of identified proteins by MALDI-TOF–MS analysis with significant differential expression kinetics during invasion of potato by Cochliobolus lunatus
Full size table

From this study, C. lunatus infection negatively impacts on the energy pathway, that is, disrupt the potential to harness light energy during photosynthesis (spot#1 and 4) and alters carbohydrate catabolic pathway (spot# 3). Thus, this interference could deprive the host of vital energy and could explain drying and wilting observed at advance stages of infection. Given that C. lunatus produces diverse conidia (Louis et al. 2015), significant quantity of melanin and nonhost specific toxin such as methyl-5-hydroxymethylfuran-2-carboxylate (Goa et al. 2014) that could aid in niche exploitation and host pathogenicity, it would be interesting to investigate the effects of these metabolites on RUBISCO, alcohol dehydrogenase 2, plastidic aldolase and photosystem II protein 33 kDa. To validate the expression of peptide spots at the mRNA level, RT-qPCR was performed and found to be differentially expressed at all time point similar to peptide spots (Fig. 5).

Fig. 5
figure 5

Boxplots illustrate the relative mRNA expressions level for corresponding peptide spots# 1, 2, 3, 4 and 5 affected during C. lunatus infection. The differences between groups were computed, error bars refer to standard deviation between replicates and asterisk indicates the level of significance with respect to control (*P < 0.05; **P < 0.01; ***P < 0.001)

Full size image

C. lunatus transiently suppressed the expression of StNPR1 at the onset of infection

The expression kinetics for StNPR1 expression was determined by RT-qPCR and by 24 hai, C. lunatus infection significantly down-regulated StNPR1 by 8.19 fold below the constitutive level in control plant (Fig. 6a). By 48 hai and 72 hai, StNPR1 transcript decrease below the basal expression level. It was observed that at 96 hai StNPR1 transcript level increased by 1.03 fold above the constitutive threshold level.

Fig. 6
figure 6

a Expression profile of StNPR1 as determine by real-time quantitative PCR. b Western blot analysis depicting the expression of StNPR1 in potato leaf at different time points expressed as mean of relative density. The values represent mean values and bars are standard error of mean values and asterisk indicates the level of significance (*P < 0.05; **P < 0.01; ***P < 0.001)

Full size image

Western blot analysis showed StNPR1 expression decreases at 48 hai prior to an upsurge in StNPR1 level as from 72 hai (Fig. 6b). Despite the increase in StNPR1 levels by 96 hai as revealed by blotting and RT-qPCR, the overall proteome change decreased (Fig. 4b) at the same time point, indicating that C. lunatus suppressed potato leaf proteome during the formation of brown-to-black spot disease. From this result (Figs. 4b, 6b), there is no direct correlation between the overall proteome change and StNPR1 expression. By comparing StNPR1 transcript and StNPR1 protein levels (Fig. 6), an increasing trend was observed except at 24 hai where StNPR1 protein level was high and StNPR1 transcript was low. Studies have shown that strict correlation between mRNA–protein abundance in the cell is rare and often poor (Maier et al. 2009). Herein, the discrepancies could be associated with hyper-accumulation of StNPR1 when the colonization process is heightened at 24 hai (Fig. 1a) and host response is highest, hallmarked by an overall high number of peptide spots (Fig. 4b). Thus, it is not unreasonable to observe the low level of StNPR1 at 24 hai, but we propose that once mRNAs are translated following post-translational modifications, StNPR1 accumulates despite the low mRNA levels. Being a regulatory protein, StNPR1 is then progressively turned-over as from 48 hai when the host defence is progressively subdued. The makeover of local infection into global defence (or SAR) largely depends on the expression of StNPR1. A pathogen capable of transiently down-regulating StNPR1, such as C. lunatus, could rapidly restrains the fine-tuning process which leads to massive transcriptional and proteome reprogramming to turn local infection into SAR, rendering the host vulnerable to irreversible colonization.

Conclusion

Infection strategies of potato by C. lunatus include formation of appressoria-like structure to penetrate different potato genotypes, secretion of diverse effectors and metabolites, modulation of ambient microenvironment, virulence differentiation and induce-virulence variations on maize crop. C. lunatus represents a formidable challenge in agriculture because of its ability to modulate host defences and expand it host range. In this study, it is shown that the brown-to-black spot disease caused by C. lunatus is associated with suppression of proteome and transient down-regulation of StNPR1 transcript at the onset of colonization. Also, this study shows that C. lunatus chiefly mitigates protein components involved in photosynthesis, carbon fixation and light assimilation.