Introduction

Wheat (Triticum aestivum L.) is the second most important food crop after rice and has been providing nutrition to the world population over the centuries. Its annual contribution to the global economy is worth US $50 billion (Curtis and Halford 2014). With continued population growth and increasing per capita consumption, a 60% increase in wheat production is required to feed the world by 2050. However, yield levels of wheat including other major cereal crops have reached a plateau. Further, due to increasing abiotic and biotic stresses caused by changing climate, genetic homogeneity, and the need for reduced usage of input resources, the global wheat production is predicted to be lowered significantly with current crop production technologies.

Leaf rust or brown rust (Puccinia triticina Erikss.) is one of the major constraints to wheat production due to its widespread distribution, rapid evolution of new virulent races, and diverse population structures (Kolmer 2005; Bolton and others 2008; McCallum and others 2016). Wheat leaf rust causes huge economic losses due to reduced yield and quality of the produce and further, additional expenditure incurred on fungicides to control the disease, thus threatening world food security and economy (Huerta-Espino and others 2011; Khan and others 2013). Breeding resistant varieties is an effective method of controlling leaf rust disease and reducing yield losses (Dubin and Brennan 2009; Singh and others 2016). As a part of resistance breeding, to date, 76 single-resistance (R) genes and several quantitative trait loci (QTLs) associated with leaf rust resistance in wheat have been identified (Cereal rust lab, USDA 2015). R gene-conferred resistance can be quickly overcome by the rapidly evolving virulent races of a pathogen. Further, pyramiding of favorable QTLs from different sources is tedious and often suffers from linkage drag. Therefore, there is a need for devising novel strategies for developing broad-spectrum and durable resistance in wheat. Developing a novel resistance breeding strategy requires a comprehensive understanding of the molecular basis of plant–pathogen interactions. However, the information on the molecular basis of defense pathways involved in wheat and leaf rust pathogen interactions is limited.

In general, pathogen defense responses involve cellular processes like recognition of a pathogen-associated pattern or effector, mitogen-activated protein kinase (MAPK) signaling, transcriptional activation of pathogenesis-related (PR) genes, and the hypersensitive response (HR). Systemic acquired resistance (SAR) is one of the prominent resistances induced in plants. SAR is a whole-plant resistance response, that is, enhanced disease resistance in distal tissues after a localized exposure to a pathogen. SAR provides a broad-spectrum and durable resistance against plant pathogens (Conrath 2006; Fu and Dong 2013). SAR is thought to be activated by pathogens causing cell death responses, ranging from single-cell HRs to necrotic disease lesions (Kogel and Langen 2005). Several factors are implicated as regulators of PR genes during the SAR-mediated defense signaling and such signaling regulators may interact synergistically and/or antagonistically to fine tune plant defenses against pathogens. The evidence for the roles of the regulators of SAR-mediated defense signaling and responses have been very well studied in model systems such as Arabidopsis against some of its pathogens (Dodds and Rathjen 2010; Pieterse and others 2012). Salicylic acid (SA), a phytohormone, is a key regulator of the SAR against biotrophic pathogens. In Arabidopsis, NDR1 (Non-specific Disease Resistance 1, At3g20600), EDS1 (Enhanced Disease Susceptibility 1), PAD4 (Phytoalexin Deficient 4, At3g52430), SAG101 (Senescence Associated Gene 101, At5g 14930), and EDS5 (also known as SID1; At4g39030) have been shown to regulate SA metabolism (Cao and others 1997; Ishihara and others 2008; Kawamura and others 2009; Dempsey and others 2011; Bao and others 2014). However, in rice and poplar, SA metabolism and its regulation are suggested to be more intricate than in Arabidopsis (Stalman and others 2003). In rice and poplar species, although SA plays an important role in disease resistance, SA synthesis is not significantly induced during biotic stresses but the basal SA levels are higher than those found during pathogen-induced SA levels in Arabidopsis leaves (Koch and others 2000). Studies have implicated the existence of the phenylalanine ammonia-lyase (PAL) pathway in rice (Silverman and others 1995) and poplar (Ruuhola and Julkunen-Tiitto 2003), and the isochorismate (IC) pathway in Arabidopsis (Dempsey and others 2011) as the predominant pathways for SA synthesis. Thus, SA metabolism and its regulation in different species could possess both shared and divergent components and regulatory mechanisms. Further, heat shock proteins (HSPs) and heat stress transcription factors (HSFs) are differentially expressed under various stresses including diseases resulting in cellular changes (Hubert and others 2003; Thao and others 2007). HSP90, a type of HSP, has been implicated to play an important role in biotic stress responses in plants. HSP90 may associate with SGT1 and RAR1, the important signaling components of R gene-mediated defense responses of various forms in Arabidopsis, tobacco, and rice (Hubert and others 2003; Liu and others 2004; Thao and others 2007).

Sugars serve different physiological roles in plants and have also been shown to play an important role in plant defense. In plant defense, sugars act as signal molecules and supply energy for the initiation of defense responses, for example, the synthesis of pathogenesis-related (PR) proteins (Morkunas and others 2007; Bolton 2009; Lemonnier and others 2014). Role of sugars in plant immunity is termed as “Sweet priming” and “High-sugar resistance” (Birch and others 2009; Hofmann and others 2010). Upon pathogen attack, cell wall invertase genes and sugar (hexose and sucrose) transporter genes involved in internalization of sugars (hexose) have been shown to initiate defense responses (Tauzin and Giardina 2014; Sun and others 2014). For instance, in wheat, STP4 sugar transporters are activated in response to powdery mildew (Blumeria graminis DC.) (Sutton and others 2007) and in maize, SUT1 a sucrose transporter gene expression was enhanced when challenged with Colletotrichum graminicola (Ces.) (Vargas and others 2012). Overexpression of the cell wall invertase gene GRAIN INCOMPLETE FILLING 1 (GIF1) enhanced resistance to Xanthomonas oryzae pv. oryzae (Xoo) and Magnaporthe oryzae (B. Cough) in rice transgenic plants (Sun and others 2014). Further, the constitutive activation of defense-related genes like PR genes, NPR1 and WRKY45, by several folds and also higher accumulation of glucose, fructose, and sucrose was observed in the GIF1-overexpressing transgenic plants than the wild-type plants (Sun and others 2014). A durable adult-plant leaf rust resistance gene in wheat, Lr34 (= Yr18/Pm38), was shown to encode an ABC transporter (Krattinger and others 2009). Lr34/Yr18/Pm38 also confers resistance to stripe rust and powdery mildew diseases in wheat. Generally, the ABC transporters utilize the energy released from ATP hydrolysis to transport various substrates across cellular membranes (Jasinski and others 2003; Rea 2007). Expression of Lr34 in maize and sorghum resulted in enhanced resistance against rust and other fungal diseases in transgenic plants (Sucher and others 2016; Schnippenkoetter and others 2017). Moore and others (2015) demonstrated the function of Lr67, a hexose transporter gene providing multiple disease resistance in wheat. Biotrophic fungi divert assimilates for their growth through the creation of a fungal sink in the infected tissues. Modeling of fungal sink competitiveness has shown that leaf rust pathogen sporulation had a competitive priority over grain filling in wheat (Bancal and others 2012). The activity of sugar transporters involved in efflux of sugars is shown to be regulated by biotrophic pathogens in plants (Chen and others 2010; Li and others 2017). A class of efflux sugar transporters called Sugars Will Eventually be Exported Transporters (SWEETs) have been speculated to facilitate pathogen nutrition (Chen and others 2010; Eom and others 2015). In rice, five SWEET (OsSWEET11–15) genes have been shown to support Xanthomonas oryzae growth (Streubel and others 2013). Interestingly, two of the SWEET genes, OsSWEET11 and OsSWEET13, have been identified as recessive blight resistance QTLs, xa13 and xa25, respectively (Yang and others 2006; Liu and others 2011). Recently, overexpression of IbSWEET10 in sweet potato was found to enhance resistance to F. oxysporum, whereas RNA interference (RNAi) lines showed enhanced susceptibility compared to the wild-type plants (Li and others 2017). These studies suggest that sugar and related metabolite transporters play key roles in plant defense or susceptibility. Thus, understanding molecular roles of sugar transporters in plant–pathogen interactions may provide new strategies to engineer robust resistance in plants.

Moreover, the SA and sugar-mediated defense pathways are expected to provide durable and broad-spectrum resistance as different R gene signals against biotrophs converge at SA signaling (Lu and others 2016) and the sugars act as signal molecules and energy source for defense responses (Rojas and others 2014). The functions of SAR signaling and sugar transporter genes or their orthologs in response to leaf rust infection in wheat are not known. Many alien Lr genes have been introgressed into the hexaploid wheat cultivars from its wild relatives to enhance the durability and efficacy of leaf rust resistance. Lr24 located on the long arm of the 3D chromosome is one such gene derived from the wild relative Agropyron elongatum (Gupta and others 2006). Lr24 is one of the most potential genes that confer resistance to all the known pathotypes of leaf rust in the Indian subcontinent and many other parts of the world (Long and others 1994; Tomar and Menon 1998; Yuan and others 2007a, b; Sharma and others 2010; Mishra and others 2014). In this study, transcriptional reprogramming of key SAR regulatory and sugar transporter genes and candidate orthologs in two wheat near-isogenic lines (NILs) differing in Lr24 was studied during leaf rust pathogen infection to understand the possible role and mechanisms of these pathways in the early resistance response of wheat plants.

Materials and Methods

Database Search for Candidate Genes and Primer Designing

The key regulators of SAR-related genes and sugar transporter genes were selected based on review of the existing literature (Dempsey and others 2011; An and Mou 2011; Fu and Dong 2013; Moore and others 2015) and a sequence database search to study the time-course expression of two pathway genes during the compatible and incompatible interactions between wheat and the leaf rust pathogen (NCBI 2017; TAIR 2017). Sequences of these genes were downloaded from the NCBI database or by homology-based search for homologs in wheat. For finding wheat homologs of selected Arabidopsis genes, we used the BLAST tool of the Ensembl Plant database following the protocol given in the tutorial document available at the wheat training website (http://www.wheat-training.com/wp-content/uploads/TILLING/pdfs/Finding-the-wheat-homologues-of-genes-from-model-organisms.pdf). Primers were designed using the Primer Express Software v3.0.1 primer design tool and synthesized by Sigma-Aldrich® (Table 1).

Table 1 Description of genes selected for studying the SAR and sugar-mediated responses during wheat–leaf rust pathogen interactions
Full size table

Inoculation of Wheat Seedlings and Leaf Sample Collection

Two wheat near-isogenic lines (NILs), HS240 (susceptible to leaf rust) and HW2020 (BC7NIL of HS240, Gupta and others 2006), carrying the Lr24 rust resistance gene (resistant to leaf rust), and the leaf rust pathotype 77–5 (121R63–1 or THTTS) were used to study the compatible and incompatible interactions between wheat and leaf rust pathogen. Leaf rust pathotype 77–5 is the most predominant and devastating leaf rust race of the Indian subcontinent (Bhardwaj and others 2016; http://rusttracker.cimmyt.org/?page_id=11). The experiment was conducted in a randomized design. Wheat NILs were grown in a growth chamber under the controlled conditions of temperature (22–24 °C), relative humidity (80%), and light periods (14 h day with 5000 lx light; 10 h dark periods). Single spore-derived cultures of pathotype 77–5 were used as the inoculum. Fresh urediniospores were suspended in the mineral oil Soltrol (Chevron Phillips Chemical Company, US) to form 50 mg spores/ml suspension. A spore suspension of 30 μl was sprayed per plant uniformly on the leaves of 1-week-old seedlings. Mock controls, the seedlings sprayed with Soltrol oil without urediniospores, were also maintained. The seedlings were placed in a high humidity (> 90%) growth chamber after misting for 24 h under dark. Then, the seedlings were shifted to the normal growth conditions in the glass house (22–24 °C and 80% RH). The leaf samples were harvested at 0, 1, 3, 6, 12, 24, and 48 h post inoculation (hpi) from both inoculated and mock plants for RNA isolation. Harvested leaf samples were immediately frozen in liquid nitrogen and stored at − 80 °C until RNA isolation. Three independent biological replicates for each of the inoculated and control plants of both the NILs were maintained (Taylor and others 2010). The leaf rust infection on the wheat NILs was confirmed after 15 days post inoculation (dpi). The experiment was repeated twice under similar conditions.

RNA Isolation and cDNA Preparation

For total RNA isolation, 100 mg leaf tissue was homogenized in a FastPrep®-24 tissuelyzer (MP Biomedicals, USA). Total RNA was extracted using the QIAGEN RNeasy Mini Kit (Qiagen, Germany) and on-column DNase I digestion was done with the RNase-Free DNase I following the protocol given in the RNeasy Mini Handbook. The integrity, yield, and purity of total RNA was determined using 1.4% formaldehyde gel electrophoresis (Rio 2015) followed by measuring of absorbance ratios A260/280 and A260/230 in the NanoDrop 2000c® UV–Vis Spectrophotometer (Thermo Scientific, USA). cDNA was prepared from 2 µg of total RNA using a High-Capacity cDNA Reverse Transcription Kit with Oligo (dT) primer (Applied Biosystems, USA). qPCR was carried out for all the cDNA samples to ensure that cDNA yield from each RT reaction was similar (Udvardi and others 2008).

Reverse Transcription-qPCR (RT-qPCR)

Amplification efficiencies of each primer pair used in the RT-qPCR experiment were calculated using five 10-fold dilutions of template in the RT-qPCR. The selection of stably expressed reference genes across the treatment groups is crucial for relative quantification or normalization in a real-time PCR study. In our earlier experiments, expression analyses using NormFinder and Best keeper software (Andersen and others 2004; Pfaffl and others 2004) showed Ubiquitin (UBI) to be stably expressed at different time points after inoculation of leaf rust among the often used reference genes glycerol-6-phosphate dehydrogenase (G6PDH), elongation factor-1α (EF-1α), and 18S ribosomal RNA (18SrRNA). Gene expression stabilities were determined and UBI was found to be the most stably expressed reference gene. Hence, UBI (Forward primer: 5′ CCTTCACTTGGTGCTCCGTCT 3′; Reverse primer: 5′ AACGACCAGGACGACAGACACA 3′) was used for relative quantification of gene expressions. For the RT-qPCR analyses, a 20-µl reaction was set up using 1 μl of 1:10 diluted cDNA template, 10 pM of each primer, 10 µl FG-Power SYBR Green PCR Master Mix (Applied Biosystems, USA), and total volume was made up with sterile distilled water. Three technical replicates were set for each of the biological replicates at different hpi. RT-qPCR reactions were performed in Applied Biosystems’ 7500 HT Fast Real-time PCR System following the program: initial activation step at 95 °C for 7 min, followed by 40 cycles of 10 s at 95 °C and 30 s at 60 °C. Finally, slow heating at a rate of 0.1 °C/s from 60 to 95 °C was included to obtain the melting curve for each primer pair. Instrument operation, data acquisition, and processing were performed using Sequence Detection System v 1.2.2 software (Applied Biosystems, USA). The melting program at the end of the cycling program followed by agarose gel electrophoresis of RT-qPCR amplicons ensured the specificity of amplification by each primer pair. Relative quantification of gene expressions was carried out using the comparative Ct (ΔΔCt) method (Schmittgen and Livak 2008). The comparative relative expression of individual genes at different hpi is represented in log2 fold changes and that among different genes is represented in log2 changes.

Statistical Analysis

One-way analysis of variance (ANOVA) followed by Tukey’s post hoc test at P ≤ 0.05 was performed using the XLSTAT software (2017) to determine the statistical significance of temporal expressions of selected candidate genes.

Results

Disease Symptoms in Compatible and Incompatible Interactions

Inoculated seedlings were observed for leaf rust disease development and scored for symptom development at fifteen days post inoculation (dpi). The susceptible NIL displayed infection type 4, whereas resistant NIL exhibited ;– and ; infection types on the basis of a scale consisting of infection types in the range 0, ;, 1, 2, X, 3, 4 (Roelfs 1984) (Fig. S1) suggesting the two NILs have contrasting leaf rust resistance phenotypes. The mock inoculated plants did not show any infection. Similar results were observed in the repeated inoculation experiments.

Candidate Genes, Amplification Specificity, and Efficiency of Primer Pairs

Mining of NCBI and Ensembl Plant databases for the regulators of SAR and sugar-related genes resulted in identification of the following genes: TaNPR1 (gi|672798846), TaSGT1 (EF197821), TaHSP90 (gi|339765023), TaRAR1 (gi|723219603), TaHTP (gi|942473027), TaSTP13A (AT5G26340) and candidate orthologs of Arabidopsis AtNDR1 (AT3G20600), AtEDS1 (AT3G48090), AtPAD4 (AT3G52430), AtPAL (AT2G37040), and AtEDS5 (AT4G39030) (Table 1). Melt curve analysis for each gene during the RT-qPCR showed a single sharp peak (Fig. S2) and agarose gel electrophoresis showed single amplicons of the expected size in the RT-qPCR products, indicating the absence of non-specific PCR products. The amplification efficiencies of all the primer pairs used in the RT-qPCR experiments ranged between 97 and 105%, suggesting optimal reaction conditions for exponential amplification of target gene transcripts.

Differential Gene Expression Patterns During Compatible and Incompatible Interactions

All the RNA samples displayed two prominent rRNA bands, 28 and 18 s with 2:1 intensity on a formaldehyde agarose gel suggesting intact RNA. Further, A260/280 readings were > 1.8 and A260/230 readings were > 2.0 indicating lack of protein and phenol contamination in the RNA, respectively. Consistency in both purity and integrity across all RNA samples minimized variability between biological replicates. Further, qPCR of cDNA samples showed that variation in the mean range of threshold cycle (Ct) values was within ± 1 for the reference gene UBI across all samples indicating a similar cDNA yield from each RT reaction.

The relative expression profiles of SAR-related genes (TaEDS1, TaNDR1, TaPAD4, TaSGT1, TaHSP90, TaRAR1, TaEDS5, TaNPR1, and TaPAL) and sugar transporter genes (TaHTP and TaSTP13A) at 0, 1, 3, 6, 12, 24, and 48 hpi were determined to understand the transcriptional changes in these genes during compatible and incompatible interactions between leaf rust pathogen and wheat. The transcript profiles of all the candidate genes indicated differential expression patterns during compatible and incompatible interactions at different time points after inoculation. The results of RT-qPCR based time-course expression of the candidate genes are presented in the following sections.

Expression of SAR-Related Defence Genes

The analyses of expression data of SA-related genes showed two distinct patterns of expression in the two contrasting interactions between wheat and leaf rust. Genes acting upstream of SA in the SA pathway (TaEDS1, TaNDR1, TaPAD4, TaSGT1, TaHSP90, TaRAR1, and TaEDS5) showed significantly higher regulations at a later phase of leaf rust infection (48 hpi) in the compatible interaction compared to the incompatible interaction (Fig. 1A, B). Genes involved in SA biosynthesis (TaPAL) and acting downstream of SA (NPR1) in the SA pathway showed significant upregulations between 6 and 24 hpi (Fig. 1B). TaEDS1 expression levels varied from 2.5- to 13-fold in the compatible interaction compared to 1–7-fold in the incompatible interaction. TaEDS1 was significantly upregulated in susceptible NIL at 0 and 48 hpi and in general, TaEDS1 expression was greater in the susceptible NIL than in the resistant NIL at most of the time intervals after inoculation (Fig. 1A). The expression levels of TaPAD4 in the susceptible NIL remained unchanged up to 3 hpi and after that the expression gradually increased to significantly high levels at 24 and 48 hpi. However, in resistant NIL, TaPAD4 expression oscillated, that is, significantly high levels of expression were observed initially, which decreased to very low levels at 3 hpi and significantly increased at 6 hpi and again decreased after 12 hpi (Fig. 1A). Similarly, the expression of TaNDR1 was induced at low levels in either of the NILs till 24 hpi and suddenly its expression was significantly upregulated by 40-fold in the susceptible NIL at 48 hpi, whereas in the resistant NIL its expression was consistently low (Fig. 1A).

Fig. 1
figure 1figure 1

Relative expression levels of A SA pathway-related genes (TaEDS1, TaPAD4, TaNDR1, TaSGT1, TaRAR1, TaHSP90, and TaEDS5) and B sugar transporter genes (TaHTP and TaSTP13A) during compatible and incompatible interactions of wheat and leaf rust pathogen

Full size image

The levels of TaRAR1 transcripts were low at all the time intervals post inoculation in both the NILs except at 48 hpi in the susceptible NIL, where the transcripts accumulation was enhanced significantly by 36-fold. The expression pattern of TaSGT1 in both the NILs was identical to that of TaRAR1 except that TaSGT1 was induced up to 16-fold in the susceptible NIL (Fig. 1A). The TaHSP90 gene was induced to a significantly high level at 24 and 48 hpi in the susceptible NIL. However, TaHSP90 inductions were low and without much variation in both the NILs up to 12 hpi and it continued to be low in the resistant NIL at later stages (Fig. 1A). Induction of TaEDS5 expression ranged between 4- and 60-fold in the susceptible NIL, whereas in the resistant NIL expression varied between 1- and 31-fold at different time intervals. Significantly very high levels of TaEDS5 were recorded at 48 hpi in the susceptible NIL compared to the resistant NIL (Fig. 1B).

Expression levels of TaNPR1 were lower than tenfold and without significant differences in either of the NILs except at 24 hpi in resistant NIL, in which the expression was enhanced significantly up to 29-fold and there was a drop in the expression of TaNPR1 at 48 hpi in the susceptible NIL (Fig. 1B). Phenylalanine ammonia-lyase (PAL) is a key phenylpropanoid pathway enzyme involved in biosynthesis of phenolic compounds including SA. PAL expression is potentiated by SA (Zhu and others 1996). Expression of TaPAL induction was more than 80-fold at 6 and 12 hpi followed by 37-fold at 48 hpi and 15-fold at 24 hpi in resistant NIL, whereas in the susceptible NIL its expression levels were significantly low at most of the time points (Fig. 1B). The comparison of relative expression of SA-related genes at different time intervals after inoculation indicated that TaPAL and TaNPR1 showed the highest expression at the mid phase of early leaf rust infection in the resistant NIL, whereas the TaEDS1, TaEDS5, TaHSP90, TaNDR1, TaPAD4, TaRAR1, and TaSGT1 genes showed higher upregulation at a later phase of leaf rust infection in the susceptible NIL (Fig. 1A, B). Comparison of normalized expression data of these SA-related genes in compatible and incompatible interactions suggested a very high differential expression at 48 hpi due to significantly high upregulation of genes acting upstream of SA in the compatible interaction (Fig. S3a).

Expression of Sugar Transporter Genes

In the incompatible interaction, the expression of TaHTP (Lr67) increased gradually from threefold at 0 hpi to ninefold at 6 hpi and then gradually decreased after 12 hpi. High expressions were observed between 3 and 12 hpi (Fig. 1B). In contrast, an inverse expression pattern was observed in the compatible interaction. Expression levels of TaSTP13A varied from one to fourfold in the incompatible and 1- to 2.5-fold in the compatible interactions. High expression levels were observed at 6 hpi in the incompatible interaction (Fig. 1B). However, no significant differences in expression were observed between the two NILs at all time intervals. Further, the comparison of normalized expression data of two sugar transporter genes between the two NILs showed relatively higher expressions at initial stages (0 and 1 hpi) and later the expression repressed significantly from 3 to 24 hpi in the compatible interaction compared to the incompatible interaction (Fig. S3b).

Discussion

A number of resistance pathways (SAR, ISR, and plant defensin) are induced during plant defense against pathogens. Salicylic acid (SA) is an important phytohormone implicated to play a critical role in both local and systemic defense responses. Therefore, the SA signaling mechanism provides broad-spectrum and long-lasting resistance to pathogen infections throughout the plant (Gao and others 2015). Besides defense-related pathways, primary metabolism pathways like sugar metabolism and sugars have been shown to play an important role in plant innate immunity as energy sources and signaling molecules (Kano and others 2011; Aliferis and others 2014; Cabello and others 2014; Zhao and others 2015). Further, although the components of induced resistance could be conserved among plant species, their spatial and temporal regulation vary with pathosystems. Understanding of the molecular components and their regulation in cereal crops including wheat and rice is limited. The results of the present study indicate a stage-specific transcriptional reprogramming of SA-related and sugar transportation genes during compatible and incompatible interactions between leaf rust and wheat and thus, gaining an insight of SA- and sugar-mediated resistance mechanisms in wheat during leaf rust infections.

Transcriptional Changes in SA-Related Genes During Wheat–P. triticina Interactions

R gene-mediated SAR is an effective strategy of resistance in plants against a broad spectrum of biotrophic pathogens mediated by the signaling molecule, SA. The major SAR signaling components include the SA molecule, genes acting upstream of SA molecules, and the SA responsive genes acting downstream of the SA. SAR has been extensively studied in the dicotyledonous species Arabidopsis and tobacco. However, investigations of SAR in monocot species including rice and wheat are limited. Recent availability of wheat genome and transcriptome sequence data has facilitated the search for orthologs of Arabidopsis SAR components and investigation into their role and regulations in plant–pathogen interactions to gain an insight into the molecular mechanisms involved in SA-mediated resistance in wheat. Further, modulation of SA-related and signaling network genes might confer broad-spectrum resistance against biotrophic fungi like wheat rusts because they act downstream of R genes, where signals perceived by different R genes converge (Tanaka and others 2015; Lu and others 2016).

SA biosynthesis occurs via two distinct pathways, the isochorismate (IC) pathway mediated by IC synthase (ICS) and the phenylalanine ammonia-lyase (PAL) pathway. In Arabidopsis, the ICS pathway is predominant, whereas the PAL pathway is predominant in monocots like rice (Silverman and others 1995; Ruuhola and Julkunen-Tiitto 2003). Results of the current study showed significant upregulation of TaPAL1 in resistant NIL between 6 and 48 hpi compared to non-significant changes in susceptible NIL and as a consequence SA production, suggesting the role of the SA biosynthetic pathway in wheat defense against P. triticina. These results are in corroboration with other studies on plant–biotic stress interactions in rice (Duan and others 2014; Kumari and others 2016) and wheat (Sorahinobar and others 2016) suggesting TaPAL as an important component of SA-mediated resistance in wheat against leaf rust. In Arabidopsis, the genes acting upstream of SA: EDS1, NDR1, PAD4, RAR1, SGT1, and HSP90 are considered key mediators of the SA pathway for resistance against biotrophs (Kawamura and others 2009; Dempsey and others 2011; Bao and others 2014). The expression of orthologs of these genes in wheat during leaf rust infection showed a significantly high upregulation in susceptible NIL at 48 hpi, the stage reported to be involved in rapid formation of haustorial mother cells (hmc) and secondary hyphal growth in susceptible NIL compared to the resistant NIL (Wang and others 2013; Serfling and others 2016). Expression of some of these genes is similar to other studies in monocot host–pathogen interactions (Qiu and others 2007; Kumari and others 2016). However, these are not consistent with the observations made on the role of these genes in Arabidopsis (Clarke and others 2001; Makandar and others 2015), soybean (Youssef and others 2013; Wang and others 2014), and tobacco (Schornack and others 2004). Thus, results of the temporal differential expression of these SA upstream acting genes in the two wheat NILs used in this study suggest the role of these genes in SA-mediated resistance but regulated by a molecular mechanism different from those in dicotyledonae host–pathosystems. EDS5 is implicated to a play role in SA accumulation possibly by controlling transport of specific molecules across the plastid membrane (Ishihara and others 2008). Evidence suggests that EDS1, PAD4, and NDR1 are at least partially required for HR-induced EDS5 expression and therefore, EDS5 seems to function downstream of these regulators (Nawrath and others 2002). In the current study, significantly high levels of TaEDS5 transcripts were observed at 48 hpi in the susceptible NIL compared to the resistant NIL. Further, comparison of TaEDS5 expression in the two interactions suggest that TaEDS5 is co-expressed or has similar temporal expression patterns with TaEDS1, TaPAD4, and TaNDR1 in either of the NILs suggesting a cooperative or coordinated action during early leaf rust infection in wheat (Fig. 1A). Further, expression of EDS5 seems to be negatively regulated by NPR1 (a SA downstream acting SAR regulator) as suggested by elevated EDS5 transcripts in npr1 plants (Nawrath and others 2002; Ishihara and others 2008). So our results are consistent with the observations on host–pathogen interaction studies by Ishihara and others (2008) and Nawrath and others (2002).

Expression of all the SA mediators acting upstream of SA accumulation observed in this study is in contrast to the expression of NPR1, a downstream regulator of SA accumulation. We found significantly high transcript levels of TaNPR1 in the resistant NIL at 24 hpi in contrast to highly repressed expression in the susceptible NIL (Fig. 1B). Results of the TaNPR1 expression pattern in the two contrasting wheat–P. triticina interactions at different time points after the inoculation suggest that a positive role in SA-mediated resistance is consistent with the studies from both the types of angiosperms during various biotic stresses (Liu and others 2002; Lin and others 2004; Yuan and others 2007a, b; Duan and others 2014; Kumari and others 2016). These suggest NPR1 as the key regulator of SA-mediated resistance in plants. Through the ankyrin motif, NPR1 mediates a wide range of protein–protein interactions and as a consequence modulates multiple pathways which may be governed by pathogen-specific signals to drive activation of the required set of defense genes (Shah 2003). In this study, the Lr24-mediated leaf rust recognition signals could be activating TaNPR1 transcription to drive the expression of PR genes, the executers of SAR defense. Some of the pathogen effectors are known to target NPR1 to compromise SA signaling through interference with NPR1 turnover, which would impact the PR gene induction (Schellenberg and others 2010; Ustun and others 2013). Similarly, in this study TaNPR1 expression was not induced in the leaf rust-infected susceptible NIL, suggesting that a pathogen effector or pathogenicity factor might be having a repressional impact on TaNPR1 transcription, which in turn fails to activate PR genes and as a result the failed defense against leaf rust. Further, NPR1 is known to regulate SA accumulation in a negative feedback loop to avoid the uncontrolled accumulation of SA that may compromise other defense or growth pathways as suggested by compromised defense and dwarfism in the npr1 mutants (Kunkel and Brooks 2002). Hence, it is important to have feedback regulation of SA synthesis and signaling for the fine-tuning of plant defense signaling against biotrophs like leaf rust fungi.

Transcriptional Changes in Sugar Transporter Genes

Expression of a cytosolic yeast-derived invertase in transgenic tobacco plants with increased levels of sugars without SAR responses suggests that hexose sensing in the secretory pathway is required for mediating defense responses in plants (Herbers and others 1996). Further, in tomato plants, the hexose content correlated with the resistance and the expression levels of hexose sugar transporter, LeHT1; Sade and others 2013 propose the role of LeHT1 in regulating sugar content of tomato during host–pathogen interactions. In this study, expression levels of two orthologs of AtSTP13: TaSTP13 and TaHTP/Lr67 encoding two hexose transporter proteins showed upregulation in incompatible interactions. TaSTP13 showed high expression, although not significantly, in the resistant NIL at 6 hpi, that is, mid stage of the early infection period. Further, interestingly, TaHTP/Lr67 transcript levels were contrasting at all stages of early infection in the compatible interaction (Fig. 1B). These results are consistent with other studies on sugar transporters in several plants against different biotic stresses (Sade and others 2013; Lemonnier and others 2014). Moore and others (2015) concluded that Lr67 encoded HTP, whose activity is regulated by amino acid sequence differences at the critical regions, and the differences in these amino acids in HTP proteins suggest the existence of resistant and susceptible alleles of Lr67 which control broad-spectrum rust defense. In this study, results of expression analysis indicated that TaHTP (Lr67) is an effective regulator of sugar-mediated leaf rust resistance in wheat compared to TaHTP13A. Further, the upregulation of TaHTP in the incompatible interaction could be triggered directly by Lr24 or indirectly due to the need for meeting the energy needs for expression of PR genes induced by Lr24 gene.

Based on the results of stage-wise transcription reprogramming of SA-related and sugar transporter genes in the two contrasting NILs, we propose a model for their mechanism of action in the early response of wheat plants during leaf rust infection.

Model for SA and Sugar-Mediated Resistance in Wheat Against Leaf Rust (P. triticina)

Salicylic acid (SA) levels play a central role in the establishment of local resistance as well as systemic resistance (Shah 2003; Tsuda and others 2008). In the susceptible NIL, low TaPAL (a key SA biosynthesis pathway enzyme) expression possibly due to repression activity of leaf rust factors might result in low SA accumulation. The low SA levels could have a positive feedback to activate transcription of upstream regulators of SA metabolism: TaEDS1, TaNDR1, TaPAD4, TaSGT1, TaRAR1, TaHSP90, and TaEDS5. In contrast, in the resistant NIL, a very high TaPAL expression induced by Lr24 and leaf rust pathogen interactions could result in higher SA accumulation. The high levels of SA are expected to have a negative feedback on the activity of upstream regulators of SA metabolism and, therefore, repressed expression of TaEDS1, TaNDR1, TaPAD4, TaSGT1, TaRAR1, TaHSP90, and TaEDS5. Further, uncontrolled synthesis of SA may compromise other defense or growth pathways that are inhibited by SA, therefore, turnover of SA is regulated by a SA amplification loop. TaNPR1 is a key downstream regulator of SA signaling which has a negative feedback on SA synthesis to avoid the effects of uncontrolled accumulation of SA. High expression of TaNPR1 in the resistant NIL might result in a faster and effective PR protein-mediated defense. At the same time, TaNPR1 might have inhibitory effect on SA biosynthesis possibly through transcription suppression of SA upstream regulators. On the other hand, low expression of TaNPR1 in the susceptible NIL could be due to the leaf rust-mediated repression or repression by upstream regulators of SA metabolism, to inhibit the repressing activity of NPR1 on the SA amplification loop. Thus, transcriptional reprogramming of genes between the two NILs in this study suggests that expression levels of NPR1, a downstream regulator of SA signaling, might regulate SA levels which in turn could modulate the transcription and activity of upstream acting SA regulators (Fig. 2).

Fig. 2
figure 2

Proposed model for SA and sugar signaling-mediated resistance in wheat against leaf rust (P. triticina)

Full size image

Further, the results of hexose transporter genes, TaHTP and TaSTP13A, expression in this study are in accordance with the results of STP13 expression in Arabidopsis–Botrytis fungus interaction (Lemonnier and others 2014) and the model of Berger and others (2007), reporting changes in sugar metabolism during pathogen infection. Accordingly, when the leaf rust pathogen attacks a wheat plant, it initiates rapid changes, resulting in the reduction of the photosynthesis rate. It is followed by an increase in invertase expression and release of hexose which may either activate a defense response or may promote pathogen multiplication, if failed to activate the plant defense. The upregulation of HTP genes in infected resistant plants may result in internalization of larger amounts of hexose into the cells. The larger ratio of mono to disaccharides is implicated to induce the plant defense (Sade and others 2013). The enhanced amounts of internal hexose may activate phytohormone (like SA)-mediated responses like SAR or may act as energy source for defense responses (Swarbrick and others 2006). In susceptible plants, the hexose transporter expression levels may not be sufficient and hence, internalization of hexoses into the cell may not be rapid and sufficient to incite defense signaling or to meet the energy needs of defense responses and thereby results in host susceptibility (Bolouri Moghaddam and Van den Ende 2012) (Fig. 2).

In summary, the quantification of expressions of the key genes involved in these two important defense pathways suggests a stage-specific transcriptional reprogramming during the two contrasting types of wheat–leaf rust pathogen interactions. Further, considering the potential role of these two pathways in broad-spectrum and durable resistance, manipulations of key regulators of these pathways could be a novel strategy for engineering broad-spectrum and durable leaf rust resistance in wheat. Results of this study provide a probable model and clue to key regulators of the two defense pathways expected to be involved in durable and broad-spectrum resistance, and thereby their prospective application in wheat rust resistance breeding.