Transcriptome approach to understand the potential mechanisms of resistant and susceptible alfalfa (Medicago sativa L.) cultivars in response to aphid feeding

TU Xiong-bing, ZHAO Hai-long , ZHANG Ze-hua

1 State Key Laboratory for Biology of Plant Diseases and Insect Pests, Institute of Plant Protection, Chinese Academy of Agricultural Sciences, Beijing 100193, P.R.China

2 College of Plant Protection, Shenyang Agricultural University, Shenyang 110161, P.R.China

Abstract Plant breeding for resistance to agricultural pests is an essential element in the development of integrated crop management systems, however, the molecular and genetic mechanisms underlying resistance are poorly understood. In this pilot study,we conducted a transcriptomic analysis of a resistant (R) and susceptible (S) variety of alfalfa, with (+A) or without (-A)aphids (totally four treatments). We used the resistant cultivar Zhongmu 1 and the susceptible cultivar Soca. A total of 3 549 mRNAs were differentially expressed, of which 1 738 up-regulated and 1 307 down-regulated genes were identified in S+A/S-A plants, while 543 up-regulated and 331 down-regulated genes were identified in the R+A/R-A plants. KEGG analysis mapped 112 and 546 differentially expressed genes to 8 and 17 substantially enriched pathways for Zhongmu 1 and Soca, respectively. Six shared pathways were linked to plant resistance traits, including phenylpropanoid biosynthesis associated with salicylic acid synthesis, and linoleic acid metabolism associated with both jasmonic acid and flavonoid biosynthesis. Ultimately, we proposed a preliminary regulatory mechanism of alfalfa cultivar resistance response to aphids feeding based on transcriptome analyses and published documents.

Keywords: alfalfa cultivar resistance, aphids, transcriptomic, molecular mechanism

1. lntroduction

Plant resistance breeding to minimize the impact of insect herbivory on yield persistence and productivity has long been an important component of integrated pest management(IPM) (Dedryver et al. 2010). Resistance provides cheap,sustainable, and environmentally safe insect pest control,while minimizing the use of insecticides (Kozjak and Megli? 2012). For plant resistance breeding, traditional plant breeding techniques were used to select resistance lines,or exotic resistance genes were inserted into previously susceptible crop plants by genetic engineering (Lu et al.2012). In this study, we conducted a transcriptomic analysis of aphid-resistant and aphid-susceptible alfalfa varieties to understand the molecular and genetic factors involved in plant resistance to an aphid.

Aphids are major insect pests of alfalfa (Medicago sativa L.) (Fabales: Fabaceae), consuming sap from phloem tissue, removing plant nutrients, and causing decreased growth, low yield, and plant death (Xu 2008).They also vector numerous alfalfa viruses (Ng and Perry 2004). Aphid management is challenging because aphids are highly mobile, with short generation times and high reproductive rates, which allows them to quickly colonise and damage plants. Consequently, large quantities of insecticides are applied for their control, which adds to the cost of food production, and can cause both short and longterm ecosystem damage, including non-target impacts on beneficial insects (predators, parasitoids, and pollinators).Moreover, the overuse of aphicides can lead to high levels of insecticide resistance in aphid populations which further complicates control (Liu et al. 2012).

Mechanisms contributing to insect resistance in legumes include structural defences, secondary metabolites and anti-nutritional compounds (Edwards and Singh 2006;Golawska et al. 2006). However, the underlying genetic basis of resistance is still not well understood (Dogimont et al. 2010). In this study, we compared aphid-resistant(Zhongmu 1) and aphid-susceptible (Soca) alfalfa cultivars at the transcriptome (RNA-seq) level (Carolanv et al.2011), each with and without aphids, to better understand the genetic and molecular mechanisms underlying plant resistance.

2. Materials and methods

2.1. Sample preparation

Two alfalfa cultivars were selected based on the results of field experiments evaluating plant resistance to aphids (Tu et al. 2017). The cultivars were Zhongmu 1 (origin China)which is classified as aphid-resistant (R), and Soca (origin Hungary) classified as an aphid-susceptible cultivar (S).Seeds from each cultivar were planted into pots containing field-collected soil, after which the pots were placed outside under ambient conditions to germinate and grow. Plants were watered 2 to 3 times per week as required. To exclude aphids, the plants were placed in a cage covered in a fine mesh cloth. After about 35 d, when the plants had reached 50% budding stage, one of the plants from each cultivar was selected and 30 Therioaphis trifolii (Monell) (spotted alfalfa aphid) placed onto the leaves and left for a further 72 h under ambient conditions. The control treatments had no aphids and were maintained under the same conditions.There were four treatments: Zhongmu 1 plus (R+A) or minus(R-A) aphids and Soca plus (S+A) or minus (S-A) aphids with only one replicate per treatment. Aphid populations on Soca were well established at the end of the three days, with about 22 live aphids, compared to Zhongmu 1 where only about nine live aphids were present. At the conclusion of the 72 h, the top 3-4 new leaves were removed from each treatment and placed in separate plastic bags. Any aphid on the vegetation was removed before bagging. Samples were frozen within 20 min at -80°C for omics analysis.

2.2. RNA extraction, library construction and sequencing

Total RNA was isolated from all four treatments using the RNA Plant Mini Kit with column DNase digestion (Qiagen,Hilden, Germany) and following the manufacturer’s instructions. RNA degradation and contamination was detected on 1% agarose gels. RNA concentration was measured using Qubit RNA Assay Kit in Qubit 2.0 Flurometer(Life Technologies, Carlsbad, CA, USA). Additionally, RNA integrity was assessed using the RNA Nano 6000 Assay Kit of the Bioanalyzer 2100 System (Agilent Technologies,Santa Clara, CA, USA).

A total of 3 μg RNA was used per sample as input material for the RNA preparations. Finally, three samples with RNA integrity number (RIN) values＞8 were used for construction of the libraries. Sequencing libraries were generated using NEBNext Ultra? RNA Library Prep Kit for Illumina (NEB,USA) following manufacturer’s recommendations and index codes were added to attribute sequences to each sample.Briefly, mRNA was purified from total RNA using poly-T oligo-attached magnetic beads. Fragmentation was carried out using divalent cations under elevated temperature in NEBNext first strand synthesis reaction buffer. First strand cDNA was synthesized using random hexamer primer and M-MuLV reverse transcriptase (RNase H). Subsequently,second strand cDNA synthesis was performed using DNA polymerase I and RNase H. Remaining overhangs were converted into blunt ends via exonuclease/polymerase activities. After adenylation of 3′ ends of DNA fragments,NEBNext adaptor with hairpin loop structure was ligated to prepare for hybridization. In order to select cDNA fragments of preferential 150-200 bp in length, the library fragments were purified with AMPure XP System (Beckman Coulter, Beverly, USA). Then 3 μL USER enzyme (NEB,USA) was used with size-selected, adaptor-ligated cDNA at 37°C for 15 min followed by 5 min at 95°C before PCR.Then PCR was performed with Phusion High-Fidelity DNA polymerase, universal PCR primers and index (X) primer.At last, PCR products were purified (AMPure XP System,Beckman Coulter, Beverly, USA) and library quality was assessed on the Agilent Bioanalyzer 2100 System.

The clustering of the index-coded samples was performed on a cBot Cluster Generation System using TruSeq PE Cluster Kit v3-cBot-HS (Illumia) according to the manufacturer’s instructions. After cluster generation,the library preparations were sequenced on an Illumina Hiseq 2500 platform and 125 bp paired-end reads were generated.

2.3. Sequence reads, mapping, assembly and annotation

Raw data (raw reads) of fastq format were firstly processed through in-house Perl scripts. In this step, clean data (clean reads) were obtained by removing reads containing adapter,ploy-N and low quality reads from the raw data. At the same time, Q20, Q30, GC-content and sequence duplication level of the clean data were calculated. All the downstream analyses were based on clean data with high quality.

The left files (read1 files) from all samples were pooled into one big left.fq file, and right files (read2 files) into one big right.fq file. Transcriptome assembly was accomplished based on the left.fq and right.fq using Trinity with min_kmer_cov set to 2 by default and all other parameters set default(Grabherr et al. 2011).

Gene function was annotated based on the following seven databases: NR (NCBI non-redundant protein sequences),NT(NCBI non-redundant nucleotide sequences), Pfam(protein family), KOG/COG (clusters of orthologous groups of proteins), Swiss-Prot (a manually annotated and reviewed protein sequence database), KO (KEGG Ortholog database)and GO (Gene Ontology), using BLAST with a cutoff e-value of 10-5.

2.4. Differential expression gene analysis

Prior to differential gene expression analysis, for each sequenced library, the read counts were adjusted using the edge R Program package through one scaling normalized factor. Differential expression gene analysis of two samples was performed using the DEGseq R package (Anders and Huber 2010). P-value was adjusted using Q-value(Storey and Tibshirani 2003). Q-value＜0.005 and |log2(fold change)|＞1 were set as the threshold for significantly differential expression.

2.5. GO and KEGG enrichment analysis of differentially expressed transcripts

GO enrichment analysis of the differentially expressed genes (DEGs) was implemented by the GO seq R packages based Wallenius non-central hyper-geometric distribution (Young et al. 2010), which can adjust for gene length bias in DEGs.

KEGG is a database resource for understanding highlevel functions and utilities of the biological system (Kanehisa et al. 2008), whether the cell, the organism or the ecosystem derived from molecular-level information, especially largescale molecular datasets generated by genome sequencing and other high-throughput experimental technologies (http://www.genome.jp/kegg/). We used KOBAS Software to test the statistical enrichment of differential expression genes in KEGG pathways (Mao et al. 2005).

2.6. Validation by real-time quantitative RT-PCR

Sequences of four important DEGs including nuclear envelope pore membrane protein POM 121B (c91095_g1),transmembrane protein (c88000_g1), ankyrin repeat plantlike protein (c99263_g1), and BURP domain-containing protein (c20427_g1) were obtained from the transcriptome datasets. PCR primers were designed for each of these four genes and were used to determine if the genes were expressed in the EST pool of both R+A and R-A cultivars.PCR conditions were as follow: 95°C for 3 min; 40 cycles of 94°C for 20 s, 55°C for 20 s, and 72°C for 20 s; with a final extension at 72°C for 5 min. Three biological replicates were used per treatment (Hegedus and Rimmer 2005). The expression ratios were calculated using the 2-ΔΔCTmethod(Yu et al. 2007).

2.7. Data archiving

Alfalfa cultivars transcriptome datasets are available at NCBI project PRJNA326161 with accession number SRP076762,and SRA with accession numbers: SRS1511728,SRS1511730, SRS1511732 and SRS1511733.

3. Results

3.1. Omics analysis

Two alfalfa cultivars, one resistant, the other susceptible,each with or without aphid infestation were sequenced individually, generating about 43-54 million clean reads, and 6.5-8.2 G clean bases for each library (Table 1). To identify the molecular mechanism underlying these transcriptomic profiles, we aligned unigene sequences to protein databases, including NR, Swiss-Prot, KEGG, and COG(e-value＜0.00001) by BLASTx, and nucleotide database NT (e-value＜0.00001) by BLASTn, retrieving proteins with the highest sequence similarity to the given unigenes along with their functional annotations. Of the 184 892 unigenes,we found 135 534 that were annotated (Table 1).

3.2. DEGs between resistant and susceptible alfalfa cultivars

Following exposure to aphids, a total of 543 and 331 up- and down-regulated transcripts, respectively were observed in the aphid resistant cultivar Zhongmu 1 (R+A) compared to the unexposed control (R-A). For the aphid susceptible cultivar Soca, a total of 1 738 and 1 307 up- and downregulated transcripts were found when aphid exposed plants(S+A) were compared to the control (S-A) (FDR≤0.001 and |log2ratio|≥1) (Table 2). Most of these transcripts were expressed within a 1- to 5-fold difference (Table 2).However, the number of DEGs found in Soca was about 5 times higher than in Zhongmu 1 (Table 2). Only 370 transcripts were expressed in both cultivars (Appendix A).To identify the DEG categories of the 370 shared transcripts,we compared their up- or down-regulated expression in the four treatments. Of the 370 shared transcripts, 235 transcripts were annotated to plants and divided into six clusters (Fig. 1; Appendix B):

(I) Active defence by the resistant (R) and the susceptible(S) cultivar, with (+A) or without aphids (-A), including 40S ribosomal protein S15-like, 60S ribosomal protein L11 transcripts. In this cluster, all transcripts of R+A and S+A were up-regulated, while the transcripts of R-A and S-A were down-regulated.

(II) Passive defence by the resistant cultivar and active defence by the susceptible cultivar, including cytochrome P450 family ABA 8-hydroxylase peroxidase transcripts. In this cluster, transcripts of R-A and S+A were up-regulated,while transcripts of R+A and S-A were down-regulated.

(III) Active defence by the susceptible cultivar, including galactinol-raffinose galactosyltransferase, alcohol dehydrogenase-like protein. In this group, transcripts of S+A were up-regulated, while transcripts of S-A were down-regulated.

(IV) Passive defence by resistant and susceptible cultivars, including fasciclin-like arabinogalactan protein and adenosylhomocysteinase. In this group, all transcripts of R-A and S-A were up-regulated, while transcripts of R+A and S+A were down-regulated.

(V) Active defence by resistant cultivar and passive defence of susceptible cultivar, including NB-ARC domain disease resistance protein and pathogenesis-related protein 1a. In this group, transcripts of R+A and S-A were up-regulated, while transcripts of R-A and S+A were downregulated.

(VI) Inherent plant cultivar differences, including kunitztype trypsin inhibitor-like 2 protein, gibberellin-regulated protein. In this group, all transcripts of R+A and R-A were up-regulated, while transcripts of S+A and S-A were downregulated.

3.3. GO and pathway analysis

The GO classification of 3 549 transcripts that were significantly differentially expressed between R and S alfalfa cultivars (≥1-fold change, FDR≤0.001) are shown in Appendix C. With Blast2GO, 614 differentially expressed transcripts were assigned to 41 GO classes in the R group(Fig. 2-A), while 2 382 differentially expressed transcripts were assigned to 58 GO classes in the S group (Fig. 2-B).The majority of these genes were assigned to categories such as biological process, cellular component, molecularfunction, response to biological regulation, cellular process,metabolic process, single-organism process, binding,catalytic activity, organelle part, cell and cell part (Appendix B). To investigate their biological functions, 874 DEGs of the R group were mapped to 171 pathways, while 3 045 DEGs of the S group were mapped to 300 pathways in the KEGG database. After exposure to aphids, 112 and 546 DEGs from the R+A cultivar and the S+A cultivar respectively, were assigned to reference pathways in KEGG. Only 8 and 17 biological pathways were significantly enriched (P＜0.05) in the R+A group and the S+A group,respectively (Table 3). Of these, six common pathways were highly enriched between them: ribosome (KEGG:ko03010), phenylpropanoid biosynthesis (KEGG: ko00940),phenylalanine metabolism (KEGG: ko00360), linoleic acid metabolism (KEGG: ko00591), alpha-linolenic acid metabolism (KEGG: ko00592), and flavonoid biosynthesis(KEGG: ko00941) (Table 3).

Table 1 Summary of RNA-seq metrics from alfalfa cultivar transcriptomes from resistant cv. Zhongmu 1 and susceptible cv. Soca,exposed or unexposed to aphids1)

Table 2 Differentially expressed genes (DEGs) between aphid-resistant Zhongmu 1 and susceptible Soca alfalfa cultivars

Fig. 1 Hierarchical clustering of differentially expressed genes (DEGs) between aphid-resistant alfalfa cultivar Zhongmu 1 and aphid-susceptible cultivar Soca. Each line in the figure represents a gene, with the columns representing aphid-exposed Zhongmu 1(R+A), unexposed Zhongmu 1 (R-A), aphid-exposed Soca (S+A) and unexposed Soca (S-A). Red bands indicate up-regulated genes while blue bands indicate down-regulated genes.

To identify the functional genes potentially related to resistance in these biological pathways, we analyzed 19 KEGG pathways (Table 3). Results indicated ubiquitous genes including ATP synthase, cytochrome c oxidase, E3 ubiquitin-protein, calphotin-like protein, ribosomal protein,and myosin protein were up-regulated after aphid infestation.Conversely, enzymes involved in plant growth and stress response such as beta-amylase, transmembrane protein,and LRR receptor-like kinase were down-regulated following aphid infestation. We also found that epidermal structure resistant genes, including chorion protein S36-like, RR1 cuticle protein, keratin type II cytoskeletal 1-like, cuticle protein 16.5, and isoform B-like, and genes related to stress tolerance, including heat shock protein 83 and apolipoprotein D, were both up-regulated in the aphid-resistant and-susceptible plants (Appendix B). Also, genes involved in phenylpropanoid biosynthesis (ko00940) and phenylalanine metabolism (ko00360) pathways (e.g., salicylic acid biosynthesis and metabolism) were up-regulated in both cultivars. Genes for linoleic acid metabolism (ko00591)and alpha-linolenic acid metabolism (ko00592) were downregulated in the S cultivar. However, the immune response after aphid infestation was more active in the R cultivar than in the S cultivar, and when attacked by aphids, the R cultivar up-regulated synthesis of linoleic acid as part of the defence response to aphid feeding.

Table 3 Significantly enriched KEGG pathways in response to aphid exposure for aphid resistant cv. Zhongmu 1 and susceptible cv. Soca

3.4. Validation by real-time quantitative RT-PCR

To verify the RNA sequencing results, four important DEGs including nuclear envelope pore membrane protein POM 121B, transmembrane protein, ankyrin repeat plant-like protein, and BURP domain-containing protein were used for relative quantitative analysis. Results showed the relative expression of nuclear envelope pore membrane protein(NEPMP) POM 121B (F=1 412.30, P＜0.0001, Fig. 3), and BURP domain-containing protein (F=11.88, P=0.0261, Fig. 3)were higher in the R+A cultivar than that in the R-A cultivar.But for the transmembrane protein (TP) (F=7.39, P=0.0531)and ankyrin repeat plant-like protein (ARPLP) (F=55.47,P=0.0017), the relative expression in the R+A cultivar was lower than that in the R-A cultivar (Fig. 3). The relative quantitative expression trends of these four specific DEGs were the same as the omics RNA sequencing results (Fig. 3).

4. Discussion

4.1. Plant-induced resistance to aphid damage

Over the last 450 million years, land plants have evolved a vast array of anti-herbivore mechanisms, including diverse constitutive and induced defences (Whitman and Ananthakrishnan 2009). Constitutive defences are those present before herbivore attack, and include spines, thorns,hairs, trichomes, wax, hard tissues, silicates, and many types of toxins or repellent chemicals, all of which are inherent plant features (Paiva 2000). Induced defences are those increased by the plant following herbivore attack (Whitman and Ananthakrishnan 2009). In this paper, we found that several induced defence genes including nuclear envelope pore membrane protein POM 121B, chorion protein S36-like, RR1 cuticle protein, keratin type II cytoskeletal 1-like, cuticle protein 16.5 and isoform B-like, were upregulated in both resistant (Zhongmu 1) and susceptible(Soca) alfalfa lines that were subject to aphid infestation,indicating that induced defences played an important role in responding to aphid attack (Maleck and Dietrich 1999;Stotz et al. 1999; Walling 2000). Furthermore, we found six shared pathways in both the resistant alfalfa variety(Zhongmu 1) and the susceptible variety (Soca), including ribosome pathway (ko03010) (related to stress resistance substances synthesis (Gao et al. 2007)), phenylpropanoid biosynthesis (ko00940), phenylalanine metabolism(ko00360) (related to salicylic acid (SA) synthesis (Leon-Reyes et al. 2010)), linoleic acid metabolism (ko00591)(related to jasmonic acid (JA) synthesis (Caarls et al. 2015)and flavonoid biosynthesis (ko00941)), and alpha-linolenic acid metabolism (ko00592) (related to jasmonic acid (JA)synthesis (Caarls et al. 2015) and flavonoid biosynthesis(ko00941)). Flavonoids represent an important secondary defence metabolic pathway associated with plant defences(Simmonds 2003; Jones and Dangl 2006). The different responses as mentioned above indicate that both aphidresistant and -susceptible alfalfa cultivars can regulate gene expression in SA, JA, and flavonoid biosynthesis pathways to induce expression of defensive genes and proteins (e.g.,polyphenol oxidase, protease inhibitor), and to enhance defence capacity (Prince et al. 2014; Tzin et al. 2015; Kloth et al. 2016; Züst and Agrawal 2016).

Fig. 3 The relative quantitative expression trends of four specific differentially expressed genes (DEGs) using 2-ΔΔCT and log2FPKM ratio with [(R+A)/(R-A)] methods. R+A represents aphid-exposed Zhongmu 1, while R-A represents aphid-unexposed Zhongmu 1. The 2-ΔΔCT method was used to compare four specific DEGs in the R+A vs. R-A cultivars to illustrate the relative quantitative expression (Yu et al. 2007),while the log2FPKM ratio showed the omics RNA sequencing results. NEPMB, nuclear envelope pore membrane protein POM 121B; BURP, BURP domain-containing protein; TP,transmembrane protein; ARPLP, ankyrin repeat plant-like protein.

4.2. Plant defence signal recognition and transformation after pest infestation

Plants can produce secondary metabolism compounds to defend against pest attack, but the mechanism for recognizing pest attack and inducing a response is less clear (Whitman and Ananthakrishnan 2009). Previous studies have shown that the presence of resistance genes expressed in the leaves of a resistant tomato line induced an isolate-specific defence response to feeding on the plant’s phloem by the potato aphid Macrosiphum euphorbia(Rossi et al. 1998), but how plants recognize defence singal have not been revealed. Subsequently, it is reported that nucleotide-binding sites leucine-rich repeat (NBS-LRR)family genes may be involed in plant recognizing pest attack.For example, Mi-1.2, was cloned from Solanum peruvianum,and has been classified as a member of the NBS-LRR family,Class II family of disease and nematode resistance genes(Martin et al. 2003), with resistance facilitated by encoding a protein to recognize aphid damage (Smith and Boyko 2007). In this paper, we also found a R gene, LRR receptorlike kinase, which may be involved in alfalfa defence signal recognition. Compared to other R gene products,the structure domain of the new found alfalfa R gene was still under detection (Martin et al. 1993; Song et al. 1995;Salmeron et al. 1996). Although these R gene products induce resistance to a range of taxonomically unrelated pathogens, their common structural features suggest common mechanistic features in the recognition of different pathogen types and elicitors (Parker et al. 1997). Thus, it may be expected that the R gene products share common functional domains with the receptors for the specific and nonspecific elicitors (Brechenmacher et al. 2015).

Further evidence of plant triggered defence response is through recognition of secretions from the insect mouth(Felton and Tumlinson 2008; Mithofer and Boland 2008).A single compound, N-(17-hydroxylinolenoyl)-L-glutamine(volicitin), was isolated from Spodoptera exigua mouthparts,which was a fatty acid-amino acid conjugates (FAC) in Lepidoptera larval saliva (Halitschke et al. 2001; Alborn et al.2003). It is constituted by two functional groups, including plant derived linolenic acid and glutamine. Glutamine is derived from the insect, and the combination of these two groups leads to the synthesis of ‘volicitin’ in insect mouthparts. In this study, we found 7 and 21 related genes in α-linolenic acid synthesis in R+A vs. R-A, and S+A vs.S-A group, respectively. We speculate that the linolenic acid synthesis response to aphids indicates a common host plant response to insect attack. Thus, these genes may relate to FACs formation in alfalfa stress response (Hermsmeier et al. 2001).

4.3. Function of DEGs

Analysis of DEGs and biological pathways showed that 235 transcripts were differently expressed and may have various roles in alfalfa cultivar resistance (Appendix D).PCR analysis of four specific DEGs in resistant Zhongmu 1 with and without aphids showed a similar expression trend for the RNA sequence (Fig. 3); results indicated that the transcriptome approach would be helpful for us to reveal the alfalfa resistance mechanism (Liu et al. 2016). In this paper, we clustered all DEGs into six classes. Five of the classes were associated with plant responses to aphid attack with both active and passive defence of both resistant and sensitive cultivars. The sixth class was related to inherent cultivar differences. This result was crucial for our future work, as it helps identify key genes related to aphid resistance in an important forage crop, and offers not only the opportunity to enhance selective breeding but also provides the opportunity to incorporate key resistance genes into alfalfa cultivars that have valuable agronomic characteristics (e.g., high yielding, tolerant to grazing) but are otherwise constrained by poor resistance to aphids(Smith and Clement 2012).

5. Conclusion

In this paper, genes associated with phenylpropanoid biosynthesis (ko00940) and phenylalanine metabolism(ko00360) pathways were up-regulated in both resistant and susceptible alfalfa cultivars. As both pathways are involved in salicylic acid biosynthesis and metabolism, this indicates that the expression of salicylic acid plays an important role in the plant response to aphid feeding. Genes of linoleic acid metabolism (ko00591) and alpha-linolenic acid metabolism(ko00592) were down-regulated in the aphid susceptible cultivar Soca, which showed that the immune response after aphid infestation was more active than the equivalent control and that when under attack, linoleic acid synthesis may involve in the defence response to aphid feeding.

Acknowledgements

Thanks to Prof. Mark McNeill (AgResearch Limited, New Zealand) and other two anonymous reviewers for their helpful suggestions. This research was funded by the earmarked fund for China Agriculture Research System(CARS-34-07).

Appendicesassociated with this paper can be available on http://www.ChinaAgriSci.com/V2/En/appendix.htm