Background

Plants have been interacting with herbivores for millions of years and have evolved a variety of defence mechanisms against herbivory, such as constitutive defences and inducible defences [1, 2]. Constitutive defences are physical barriers such as cell walls, waxy cuticles, and bark, protecting the plant from attacks [3,4,5,6]. Inducible defences include the rapid detection of herbivory by plants through specific recognition and signalling systems and the production of a range of products or secondary metabolites that are toxic, repellent or anti-digestive to herbivores [7,8,9]. Some herbivory-induced products are volatile organic compounds (VOCs) released by plants that can attract the natural enemies of herbivores, resulting in an indirect means of protection [2, 10,11,12].

Several phytohormones, including jasmonic acid (JA), salicylic acid (SA), ethylene (ET), abscisic acid (ABA), auxin, and cytokinins, are key mediators of plant defences [13,14,15,16,17,18]. JA and SA and their derivatives play a predominant role in modulating plant defences against pests and pathogens, respectively [2, 19].

The JA-dependent signalling pathway is usually activated in response to leaf-chewing herbivores, cell-content feeders and necrotrophic pathogens [20,21,22]. The SA-mediated defence pathway is primarily induced by piercing-sucking herbivores and biotrophic pathogens [21]. Hemipterans have highly modified piecing-sucking mouthparts (stylets) that follow an intercellular pathway and feed on phloem sap from sieve elements (SEs) [23]. Feeding of hemipterans, like that of whiteflies and aphids, causes minimum mechanical damage in plant cells during feeding and mainly induces SA-dependent signalling defence pathway by suppressing the JA-associated defence pathway [24].

The greenbug, Schizaphis graminum, is one of the most important and devastating cereal aphids in the world, damaging plants by feeding on phloem sap and serving as a vector for transmitting viruses, such as barley yellow dwarf virus (BYDV) [25, 26]. With global warming, the potential risk of S. graminum infestations will increase, especially in the northern hemisphere, which could increase global food insecurity and poverty by destroying economically important crops [27]. In contrast to most other aphid species, S. graminum is a phytotoxic aphid, and its feeding can rapidly induce leaf chlorosis in susceptible plants, resulting in the deterioration of plant quality and even plant death. Previous studies have demonstrated that S. graminum feeding induced SA- and JA-dependent defence pathways in sorghum (Sorghum bicolor) [28], and reactive oxygen species (ROS) levels, peroxidase (POD) and laccase activities were also increased in switchgrass (Panicum virgatum) after S. graminum feeding [29]. However, few studies have been conducted to identify the defence mechanisms in wheat in response to S. graminum feeding, and the mechanisms underlying the induction of damage by S. graminum infestation are still unclear. In the present study, we investigated the dynamic wheat responses to S. graminum feeding by integrating results from high-throughput RNA sequencing and cytological examination to uncover the mechanism underlying the induction of defence responses and damage symptoms by phytotoxic aphid feeding.

Results

Transcriptomic analysis of wheat leaves in response to S. graminum at different time points

Global transcriptomic changes in response to phytotoxic aphid feeding were examined in leaves of wheat seedlings infested with S. graminum at 2, 6, 12, 24 and 48 hpi. A total of 166.41 Gb of clean data were obtained from the 18 leaf samples, and each of these samples contained ≥7.6 Gb of data with Q30 quality scores ≥92.32% (Additional file 1). Subsequently, for each sample between 44.3 and 59.5 million reads were mapped uniquely, with roughly 5% of the total reads mapping to multiple locations (Additional file 2).

The gene expression levels were used to conduct a PCA for each of the biological replicates. Each replicate from the same group was clustered closely together, which suggested that the repeatability of each treatment was satisfactory, and the samples from different time points of S. graminum infestation were clustered far from each other and the control groups, which indicated that aphid feeding induced significant changes in gene expression (Fig. 1a).

Fig. 1
figure 1

Transcriptomic overview of a time course of S. graminum feeding on wheat leaves. a: PCA plot of global transcriptome profiles. b: Total number of transcripts that were significantly up- or downregulated in response to aphid feeding. c, d: Venn diagram illustrating the number of genes up- or downregulated by aphid feeding over the time course. P < 0.01 FDR and Log2 FC ≥ 1 or ≤ − 1

Full size image

The P value≤0.01 (false discovery rate [FDR] adjusted) and Log2-fold change (Log2FC) ≥1 or ≤ − 1 were set as thresholds for DEGs in wheat leaves at different time points. Then, these identified DEGs were used for further analysis. A total of 44,835 DEGs were identified in wheat leaves at different time points (2, 6, 12, 24 and 48 hpi) of aphid feeding (Additional files 3, 4, 5). Briefly, 9761 (9105 up- and 656 downregulated), 22,183 (13,935 up- and 8248 downregulated), 29,875 (16,214 up- and 13,661 downregulated), 32,741 (17,771 up- and 14,970 downregulated) and 30,729 (17,523 up- and 13,206 downregulated) DEGs were identified at 2, 6, 12, 24 and 48 hpi, respectively (Fig. 1b, Additional file 4). The distribution of up- and downregulated genes was calculated for each time point and are presented in a Venn diagram (Fig. 1c and d). Although a unique set of genes increased at each time point (total 74,548), the expression levels of a large number of genes (5800) were significantly upregulated at all time points. In addition, a unique set of genes was significantly downregulated at each time point (total 50,741), and only 349 genes showed decreased expression at all five time points.

Gene ontology (GO) analysis of DEGs

GO analysis was used for the functional classification of the DEGs in wheat leaves after aphid infestation. The top 30 enriched GO terms of all DEGs are shown in Additional file 6. GO analysis of DEGs induced by S. graminum feeding at early time points is shown in Fig. 2. At 2 hpi (Fig. 2a, b), within the biological process category, the upregulated DEGs were mainly enriched in metabolic process, single-organism process and phosphorus metabolic process. Within the molecular function category, the largest proportion of upregulated DEGs induced by the aphid feeding was enriched in catalytic activity and transferase activity. At 6 and 12 hpi (Fig. 2c-f), the majority of the upregulated DEGs activated by the aphid feeding were enriched in metabolic processes and single-organism processes within the biological process category, protein kinase activity and phosphotransferase activity. In the molecular function category, the upregulated DEGs were mainly enriched in catalytic activity and transferase activity.

Fig. 2
figure 2

GO enrichment analysis of the differentially expressed genes (DEGs) in wheat leaves in response to S. graminum feeding at 2, 6 and 12 hpi. a: GO enrichment analysis of upregulated DEGs at 2 hpi; b: GO enrichment analysis of downregulated DEGs at 2 hpi; c: GO enrichment analysis of upregulated DEGs at 6 hpi; d: GO enrichment analysis of downregulated DEGs at 6 hpi; e: GO enrichment analysis of upregulated DEGs at 12 hpi; f: GO enrichment analysis of downregulated DEGs at 12 hpi

Full size image

At 6 and 12 hpi (Fig. 2c-f), within the biological process category, the downregulated DEGs were mainly enriched in photosynthesis and light reaction. In the cellular component group, greater percentages of DEGs in the cellular component category were enriched in cell and cell parts. In addition, many downregulated DEGs at 6 and 12 hpi were also enriched in chloroplasts, further indicating the damage in wheat leaves caused by S. graminum feeding.

Chlorophyll content in wheat leaves after S. graminum feeding

Transcriptome analysis showed that aphid feeding negatively affected the photosynthetic processes of wheat, and the transcript levels of many light-harvesting- and photosystem-associated genes, such as ribulose-1,5-bisphosphate carboxylase, chlorophyll a-b binding proteins, ferredoxin thioredoxin reductase, and PsbP family proteins, were significantly downregulated (Table 1).

Table 1 DEGs associated with plant photosynthesis process in wheat leaves in response to S. graminum feeding
Full size table

The results in Fig. 3 suggested that the total chlorophyll content in wheat leaves at 2, 6 and 12 hpi was not significantly different from that of the control. However, the total chlorophyll content was significantly decreased to 1.49 ± 0.10 mg g− 1 FW after 24 h of aphid feeding (F5,18 = 9.447, P = 0.0001) and was further reduced to 1.07 ± 0.11 mg g− 1 FW at 48 hpi, which was significantly lower than that of the control (2.58 ± 0.18 mg g− 1 FW).

Fig. 3
figure 3

Chlorophyll content in wheat leaves after S. graminum feeding at 2, 6, 12, 24 and 48 hpi. The values are presented as the means ± SE of three biological replicates. Different letters indicate significant differences among treatments (P < 0.05, ANOVA)

Full size image

Transcript levels of genes involved in SA- and JA-dependent defence pathways in wheat leaves after S. graminum feeding

Phytohormone metabolic pathways are commonly used by plants for defence against both pests and pathogens. The transcriptome data in Table 2 showed that all six phenylalanine ammonia-lyase (PAL) genes involved in SA biosynthesis were significantly upregulated in response to S. graminum at different time points, and the expression levels of PAL gradually decreased with increased aphid feeding time (4.96 to 16.16-fold). Furthermore, PR genes that respond to SA were also significantly upregulated during all time points of aphid feeding (4.92 to 20.59-fold).

Table 2 DEGs involved in jasmonic acid and salicylic acid synthesis pathways in response to S. graminum feeding at different time points
Full size table

A greater effect on genes involved in JA metabolism was observed over time (Table 2). Three lipoxygenase (LOX) genes were significantly upregulated by S. graminum feeding at different time points (1.49 to 9.25-fold), and one lipoxygenase (LOX) was only upregulated at 2 hpi (5.89-fold). The expression levels of allene oxide cyclase (AOC) were also significantly increased at various aphid feeding time points. In contrast, the jasmonic acid-amido synthetases (JARs) were downregulated in infested plants (− 2.31 to − 3.41-fold). There were also five fatty acid desaturase (FAD) genes that had variable expression levels, with the majority being upregulated (1.36 to 6.77-fold) during infestation. However, some were also downregulated during the three later time points (− 1.20 to − 2.45-fold), which suggests that the expression of FAD genes may be fine-tuned during defence responses.

The mitogen-activated protein kinase (MAPK) cascade is a key signalling pathway of plant defence, and WRKY transcription factors (TFs) appear to be regulated by MAPKs and involved in the regulation of plant defence. The transcript levels of several MAPKs were significantly upregulated (1.01- to 3.48-fold) in response to S. graminum feeding at 12, 24 and 48 hpi. Additionally, several WRKY TFs in wheat leaves were significantly induced (1.59 to 11.14-fold) in response to aphid feeding at different time points (Table 3).

Table 3 DEGs involved in MAPK-WRKY pathways in response to S. graminum feeding at different time points
Full size table

Effects of S. graminum feeding on hydrogen peroxide (H2O2) accumulation and the activity of antioxidant enzymes in wheat leaves

As shown in Table 4, S. graminum feeding upregulated various ROS-scavenging genes, such as POD and SOD (superoxide dismutase), at 2 hpi. The expression levels of these two genes were increased at 6, 12 and 24 hpi and then gradually decreased at 48 hpi, but the transcript levels were still significantly increased compared with the control levels. The expression levels of catalase (CAT) genes showed no significant differences at 2 hpi compared to the control levels. Among them, two genes were significantly downregulated at 6, 12, 24 and 48 hpi, and the other two genes were significantly upregulated.

Table 4 DEGs associated with ROS scavenging in wheat leaves in response to S. graminum feeding at different time points
Full size table

The activities of three antioxidant enzymes, POD, SOD and CAT, were also examined in wheat leaves infested with aphids (Fig. 4). Compared with the control, the activity of POD was significantly increased after 6 h (19.01 ± 3.94 U mg− 1 protein) of S. graminum feeding and reached a peak at 48 hpi (44.44 ± 3.37 U mg− 1 protein; F5, 12 = 10.17, P = 0.001). Furthermore, the activities of SOD (55.10 ± 7.55 U mg− 1 protein; F5,12 = 5.15, P = 0.009) and CAT (0.20 ± 0.038 U mg− 1 protein; F5,12 = 7.27, P = 0.002;) were significantly upregulated after 12 h of S. graminum feeding. The increased activity of ROS scavengers induced by aphid feeding suggested that S. graminum feeding induces oxidative stress in wheat leaves.

Fig. 4
figure 4

Activity of the antioxidant enzymes POD, SOD and CAT in wheat leaves in response to S. graminum feeding. The values are presented as the means ± SE of three biological replicates. Different letters indicate significant differences among treatments (P < 0.05, ANOVA)

Full size image

To further examine the effects of aphid feeding on oxidative stress in wheat, S. graminum-infested leaves were examined after cytological staining with 3,3′-diaminobenzidine (DAB), which was used to detect the production of H2O2. As shown in Fig. 5, small and obvious brown spots were detected at 2 hpi, indicating H2O2 accumulation at the aphid feeding site. The number and size of the spots increased with increasing time aphid feeding time.

Fig. 5
figure 5

Detection of H2O2 accumulation in wheat leaves in response to S. graminum feeding at different time points using DAB staining. Images are representative of three biological replicates

Full size image

Effects of inhibition of NADPH oxidases on H2O2 accumulation and defence responses in wheat leaves

To detect the roles of plasma membrane NADPH oxidases in H2O2 accumulation induced by S. graminum feeding, wheat leaves were treated with the NADPH oxidase inhibitor diphenylene iodonium (DPI). The DAB staining results, shown in Fig. 6a, indicate that many obvious brown spots were detected at the aphid feeding sites of infested leaves, but fewer brown spots were observed after 10 μM and 25 μM DPI treatments, indicating that the H2O2 production induced by aphid feeding was inhibited by DPI. Additionally, the H2O2 contents were significantly decreased to 50.60 ± 9.51 μmol g− 1 FW and 33.93 ± 3.00 μmol g− 1 FW in wheat leaves treated with 10 μM and 25 μM DPI, respectively (F2,6 = 6.44, P = 0.032) (Fig. 6b).

Fig. 6
figure 6

Effects of pretreatment with diphenylene iodonium (DPI), an NADPH oxidase inhibitor, on the accumulation (a) and contents of H2O2 (b) induced by S. graminum feeding in wheat leaves. Images are representative of three biological replicates

Full size image

As shown in Fig. 7a and b, DPI treatment had significant effects on the expression levels of defence response genes in wheat leaves. The expression levels of the salicylic acid-related genes PAL and PR1 were significantly reduced in DPI-treated wheat leaves compared to the control (wheat leaves infested with aphids for 24 h) (F2,6 = 10.96, P = 0.01; F2,6 = 65.53, P < 0.001).

Fig. 7
figure 7

Effects of pretreatment with DPI on the expression levels of the SA-associated defence genes PAL (a) and PR1 (b) induced by S. graminum feeding in wheat leaves. The values are presented as the means ± SE of three biological replicates. Different letters indicate significant differences among treatments (P < 0.05, ANOVA)

Full size image

Discussion

Wheat transcriptomes are rapidly and significantly regulated in response to S. graminum feeding

Plants usually activate rapid responses to biotic and abiotic stress. For example, in Arabidopsis, rapid and highly localized increases in calcium was detected around the feeding sites of M. persicae after just a few minutes of feeding [30]. More than 500 genes associated with the primary metabolism and defence responses were significantly upregulated in peach leaves after 3 h of infestation by M. persicae [31]. Almost 800 genes were upregulated after 3 h of aphid feeding in maize [32]. Similarly, rapid and strong physiological responses were activated in wheat by S. graminum feeding in our study. The transcript levels of more than 9000 genes involved in various physiological processes were significantly up- or downregulated after just 2 h of aphid feeding. A previous study demonstrated that severe degenerative changes were detected in vascular cells adjacent to the stylet path of susceptible wheat plants as early as 1 h after S. graminum infestation using transmission electron microscopy [33]. Cellular damage caused by aphid feeding may be responsible for the rapid and strong physiological changes in wheat leaves.

S. graminum feeding results in damage to chlorophyll and suppression of photosynthesis

As a typical phytotoxic aphid, S. graminum feeding induced obvious chlorosis in the aphid-susceptible wheat variety [34]. Consistent with previous studies [35], the total contents of chlorophyll were significantly reduced in wheat leaves after aphid infestation. Additionally, transcriptomic analysis showed that many genes involved in photosynthesis, such as ribulose-1,5-bisphosphate carboxylase and chlorophyll A-B binding protein, were significantly downregulated after S. graminum feeding, and downregulated DEGs at early time points of aphid feeding (after 6 h) were enriched in photosynthesis and light harvesting pathways, indicating that S. graminum feeding imposed rapid damage to chlorophyll and then led to a reduction in the photosynthetic activity of the plants.

Defence signalling pathway induced in wheat following challenge with S. graminum

Piercing-sucking hemipteran insects, such as aphids and whiteflies, mainly induce SA-mediated defence signal pathways [36]. However, some studies have also demonstrated that genes involved in both the JA and SA defence response pathways, such as LOX, PIs, PAL, and PR1, are significantly upregulated by aphid feeding [37,38,39,40]. Similarly, we found that S. graminum feeding significantly increased the expression levels of several genes related to the SA and JA signalling pathways. The fold changes in the expression levels of PAL and PR1 were very high at various aphid feeding times, suggesting that a strong SA defence pathway was activated by S. graminum feeding, which might be responsible for the induction of chlorosis in wheat. However, the genes involved in the JA signalling pathway had variable expression levels, with the majority being upregulated during infestation but some also being downregulated during the later time points, which suggests that the expression of FAD genes may be fine-tuned during defence responses. Several JAR1 (jasmonic acid resistant 1) genes were also significantly downregulated by S. graminum. JAR1 is a JA-aminosynthetase that is required to activate JA in Arabidopsis [41]. These results suggested that although JA-responsive genes were upregulated, increased JA levels or the JA defence pathway may not be activated by S. graminum feeding. Zhang et al. demonstrated that the SA contents in wheat leaves after S. graminum feeding were significantly greater than those found in wheat leaves without aphid infestation, but no significant changes were detected for the JA contents in aphid-infested wheat leaves [35].

Protein phosphatases and MAPK-WRKY pathways are activated in response to S. graminum feeding

At different time points of aphid feeding, many upregulated genes were enriched in protein phosphorylation. Protein phosphatases are one of the most predominant post-translational modifications (PTMs) and play a central role in signal transduction through the phosphorylation and de-phosphorylation of proteins in eukaryotes [42]. A large body of evidence demonstrates that phosphorylation is essential for immune responses in plants [43]. For example, in Arabidopsis, more than 1170 phosphopeptides from 472 phosphoproteins were identified after treatments with flg22 or xylanase, both of which elicit immune responses in Arabidopsis cell cultures [44]. A total of 109 differentially phosphorylated residues of membrane-associated proteins on activation of the intracellular RPS2 receptor were identified by phosphoproteomic screening using an inducible expression system of the bacterial effector avrRpt2 in Arabidopsis [45]. Additionally, phosphorylation plays a central role in the progression of the signal through the MAP kinase cascade, which includes a class of protein kinases that plays a crucial signalling role in plant defence against pathogen and herbivore attacks [46]. The expression levels of several MAPKs and mitogen-activated protein kinase kinases (MAPKKs) were significantly upregulated by S. graminum feeding in wheat, indicating that MAPK signalling plays critical roles in regulating the induced defence responses of S. graminum.

The expression levels of many WRKY genes were significantly upregulated by S. graminum feeding. WRKY transcription factors comprise a large protein family sharing a DNA binding domain of approximately 60 amino acids that contains an invariable sequence WRKYGQK and a zinc-finger-like domain [47, 48]. An increasing number of WRKY TFs have been identified as substrates of MAPKs and important components in MAPK signalling pathways for the regulation of plant immunity [49,50,51]. For instance, in response to Botrytis cinerea infection, WRKY33 is phosphorylated and activated by MPK3/MPK6, which induce downstream defence responses in Arabidopsis, such as ethylene production [52]. Phospho-mimicking mutants of WRKY transcription factors are involved in the induction of RBOHB-dependent ROS burst and cell death in Nicotiana benthamiana, suggesting that WRKYs are associated with the induction of HR-cell death as MAPK substrates in plants [53]. Whether the MAPK-WRKY pathway is involved in the cell death and chlorosis caused by S. graminum is still unknown.

S. graminum feeding induces strong oxidative stress in wheat

Abiotic and biotic stresses generally induce the accumulation of ROS, such as H2O2 and O2-, and cause oxidative stress in plants. H2O2 is one of the most important ROS in plant-pathogen and plant-herbivore interactions and has an important role in signal transmission and plant defence responses [54, 55]. The production of ROS, such as hydrogen peroxide (H2O2), rapidly occurs in fern plants after just 1 h of herbivory [56, 57]. In infested barley plants, the maximum level of H2O2 was found just approximately 20 min after aphid infestation [58]. In our study, rapid H2O2 accumulation was also detected in wheat after S. graminum feeding, which suggested that aphid infestation results in strong oxidative stress. In plant cells, antioxidant enzymes, such as CAT, POD, and SOD keep ROS at low concentrations, avoiding oxidative damage while allowing them to play crucial functions in signal transduction [59]. Excess cellular levels of ROS cause damage to proteins, nucleic acids, lipids, membranes and organelles, which can lead to activation of cell death processes such as apoptosis. The transcriptomic and enzymatic results in our study showed that S. graminum feeding increased the gene expression levels and enzyme activities of ROS scavengers in wheat leaves. It is speculated that although ROS scavengers were significantly upregulated by S. graminum, H2O2 production might exceed the cellular antioxidant capacity, resulting in oxidative damage to cellular components and chlorosis in leaves.

Strong SA-dependent defence responses were triggered by S. graminum. Considerable evidence demonstrates that H2O2 can interact with plant defence hormones, such as SA [60]. For example, SA treatment can enhance H2O2 levels, and H2O2 has been proposed to function downstream of SA in plants based on the evidence that SA can participate in regulating antioxidant enzymes, such as CAT, SOD and ascorbate peroxidase (APx) [61,62,63]. In contrast, other studies report that elevated H2O2 levels could activate SA biosynthesis via stimulation of BA2H (benzoic acid 2-hydroxylase) activity in tobacco cells [64]. Furthermore, other studies have indicated that H2O2 does not function downstream of SA in regulating PR protein expression [65]. The mechanisms by which SA interacts with H2O2 production in plants remain unclear.

NADPH oxidases are involved in the regulation of S. graminum-induced H2O2 accumulation and defence responses of wheat

Plasma-membrane-localized NADPH oxidases, known as respiratory burst oxidase homologues (Rbohs), are membrane-bound enzyme complexes that are important components for H2O2 generation and plant immunity [66]. A lack of RBOH expression leads to very low levels of ROS production, resulting in the alteration of different plant responses in terms of cell death and pathogen resistance [67,68,69]. To further examine the roles of NADPH oxidases in H2O2 accumulation and defence responses induced by S. graminum feeding, wheat leaves were treated with the NADPH oxidase inhibitor DPI at different concentrations. H2O2 production and the expression levels of SA-dependent defence genes were inhibited, suggesting the involvement of NADPH oxidases in H2O2 accumulation and defence responses induced by S. graminum feeding. Phosphorylation at specific conserved residues in RBOHD is also required for ROS production in both PAMP-triggered immunity (PTI) and effector-triggered immunity (ETI) and immunity against avirulent bacteria and a virulent necrotrophic fungus [70]. Whether RBOHD phosphorylation is associated with the ROS production, defence responses and chlorosis induced by S. graminum feeding in wheat is worthy of further study.

Conclusions

In conclusion, the transcriptomic profiling of wheat performed in this study revealed dynamic physiological changes in wheat leaves in response to phytotoxic aphid S. graminum feeding. S. graminum feeding triggered rapid plant defence responses and ROS-scavenging activities in wheat plants. A cytological analysis showed that substantial H2O2 accumulated in wheat leaves in response to S. graminum feeding. Our results also demonstrated that NADPH oxidases play vital roles in the induction of H2O2 accumulation and SA-dependent defence responses triggered by S. graminum feeding. Our future studies will focus on the mechanisms of chlorosis induction by S. graminum feeding and the roles of salivary proteins of aphids in the induction of symptoms in plants.

Methods

Plants and aphids

Wheat seeds (Triticum aestivum var. Zhongmai175, Shengyuan Seed Industry Technology Co., Ltd., China) were immersed and germinated in sterilized Petri dishes with distilled water for 3–4 days at a temperature of 25 ± 1 °C and a photoperiod of 16:8 (L:D) h. After germination, healthy seedlings were selected and transferred into plastic plots containing organic soil and were grown in a climate chamber with a temperature of 20 ± 1 °C, a relative humidity of 40–60% and a photoperiod of L: D = 16 h: 8 h. All S. graminum were initiated from a single parthenogenetic female collected from a wheat field in Langfang city, Hebei Province, northern China. A clonal aphid lineage was maintained on wheat plants (Zhongmai175) under laboratory conditions at 20 ± 1 °C and 65 ± 10% relative humidity with a photoperiod of 14: 10 (L: D) h.

Wheat plant infested with aphids

At the two-leaf stage (12-day-old), twenty 3rd instar S. graminum were placed into clip cages on the second leaves of wheat plants. Aphids were gently removed from wheat leaves using a soft brush at different time points after aphid feeding. Leaf tissues at the aphid feeding sites of each plant were collected into a 1.5 ml centrifuge tube using sterilized scissors, transferred to liquid nitrogen immediately and stored at − 80 °C until use. Two leaf tissues were collected from two independent plants with the same treatment to form one independent biological replicate. Three independent biological replicates were conducted for each treatment. Plants with empty clip cages were set as control groups.

RNA extraction, library construction, and RNA sequencing

Total RNA from wheat leaves was extracted with TRIzol reagent (Invitrogen, Carlsbad, CA, USA) according to the manufacturer’s instructions. The RNA quality and quantity were examined using a DS-11 spectrophotometer (DeNovix, DE, USA), and the RNA integrity was confirmed using an RNA Nano 6000 Assay Kit with an Agilent 2100 Bioanalyser (Agilent Technologies, Santa Clara, CA, USA). Only RNA samples with an RNA Integrity Number (RIN) ≥ 7.0 were used in the subsequent analysis. Libraries were constructed using the TruSeq Stranded mRNA LT Sample Prep Kit (Illumina, San Diego, CA, USA) following the protocols provided by the manufacturer and were sequenced on the Illumina sequencing platform (Illumina HiSeq 4000), and 150-bp paired-end reads were generated. There were three independent biological replicates for each treatment.

RNA-seq data analysis

To obtain high-quality reads, the reads containing adaptor sequences, more than 10% ambiguous bases (noted as N) and low-quality bases (Qphred ≤20 bases account for more than 50% of the entire read length of the reads) were filtered. The resulting clean reads were then aligned to the wheat reference genome (IWGSC RefSeq v1.0 genome) using TopHat v2.0.12 with the default values [71, 72]. Transcript assembly was performed using Cufflinks v2.1.1 with the default parameters [73]. All transcripts were compared with gene models in the reference genome to identify novel genes expressed from previously intergenic regions (class code “u”) using Cuffcompare [74]. The gene expression levels in all the samples were calculated using fragments per kilobase of exon model per million mapped reads (FPKM), and the genes with more than 1 FPKM in at least one sample of wheat leaves were used for further analysis [75]. An adjusted P value (FDR < 0.01) and fold change (FC) ratio (|Log2 FC| ≥ 1) were used to determine the differentially expressed genes (DEGs) between aphid-infested and control leaves using DESeq R package. Principal component analysis (PCA) was performed using the DESeq2 package for clustering the samples based on gene expression patterns [76]. The heatmap of DEGs was clustered using pheatmap (version 1.0.8, http://cran.r-project.org/web/packages/pheatmap) package in R. Gene Ontology (GO) (http://geneontology.org/) enrichment analyses of the DEGs were performed to understand the biological significance of the DEGs using the GOseq R package, and GO terms were considered significantly enriched with a corrected p-value < 0.05 [77].

Changes of total chlorophyll levels in wheat leaves after aphid infestation

Total chlorophyll content in aphid-infested wheat leaves at different time points was examined according to the method of Aron [78] with slight modifications. A total of 100 mg of fresh leaf tissues was ground to a fine powder in liquid nitrogen using a mortar and pestle. Chlorophyll was extracted with 5 mL of 80% acetone (v/v) at 4 °C for overnight. The homogenate was centrifuged at 4000 g for 10 min at 4 °C, and the supernatant was used for chlorophyll assay. Chlorophyll content of leaves was detected spectrophotometrically, by reading the absorbance at 645 and 663 nm (DU800, Beckman, USA).

Detection of H2O2 accumulation in wheat leaves induced by aphid feeding

To detect H2O2 accumulation in wheat leaves induced by S. graminum feeding, 3,3′-diaminobenzidine (DAB) (Sigma, Germany) staining was performed using the protocols reported by Wang et al. [79] with some modifications. In brief, leaf segments infested with S. graminum were cut off using sterilized scissors, and then immediately immersed in 1 mg mL− 1 DAB solution (10 mmol L− 1 Na2HPO4, pH 3.8) and incubated in a dark chamber overnight at room temperature. Leaves were decolorized in 95% ethanol solution and hyalinized in saturated chloral hydrate. The stained leaves were then visualized using an Olympus BX-63 microscope (Olympus Corporation, Japan). The endogenous H2O2 content in the wheat leaves was determined according to the methods described by Ferguson et al. [80].

Determination of antioxidant enzymes in wheat leaves after aphid infestation

To examine the activities of peroxidase (POD), superoxide dismutase (SOD) and catalase (CAT) in wheat leaves at all time points after aphid infestation, a total of 200 m g of fresh leaf tissues was ground into a fine powder in liquid nitrogen using a mortar and pestle. The powder was then immediately homogenized in 1.5 mL of ice-cooled 50 mM potassium phosphate buffer (pH = 7.8) containing 0.1 mM EDTA, 1 mM phenylmethylsulfonyl fluoride (PMSF) and 1.0% (w/v) polyvinylpyrrolidone. The homogenate was centrifuged at 15,000 g for 30 min at 4 °C, and supernatant was immediately collected as crude enzyme extract for further assay. The activities of POD, SOD, CAT were determined by following the changes in absorbance at 470 nm, 560 nm and 240 nm respectively according to previously described [81].

RT-qPCR

Total RNA was extracted from leaves using TRIzol Reagent (Invitrogen) according to the manufacturer’s recommended protocols. The concentration of RNA was measured by a DS-11 Spectrophotometer (DeNovix, DE, USA). One microgram of total RNA was reverse-transcribed into first-strand cDNA with oligo dT primers using the EasyScript All-in-One First-Strand cDNA Synthesis SuperMix for RT-qPCR (TransGen Biotech) following the manufacturer’s instructions, and cDNA templates were stored at − 20 °C until use. RT-qPCR was performed using the same protocols as previously described [35]. The RT-qPCR protocol consisted of an initial heat activation step of 95 °C for 10 min, followed by 40 cycles of 95 °C for 15 s and 60 °C for 40 s. Three biological replicates were performed for each treatment, and each biological replicate consisted of three technical replicates.

Wheat seedlings treated with DPI solution

The leaves of wheat seedlings were treated with 10 μM and 25 μM diphenylene iodonium (DPI, a NADPH oxidase inhibitor) solution or deionized water (control) for 24 h and then infested with S. graminum for 48 h. Detection of H2O2 accumulation and expression levels of genes in wheat leaves were conducted as described previously.

Statistical analyses

All the data were analysed using SPSS Statistics 20.0 software (SPSS Inc., Chicago, IL., USA). Normality of distribution and homogeneity of variances were tested by Shapiro-Wilk’s test and Levene’s test respectively, and the differences among groups were examined through one-way analysis of variance (Duncan). P values less than 0.05 were considered statistically significant.