Introduction

Wheat is the most widely grown food crop in the world. Wheat powdery mildew, caused by the fungal pathogen Blumeria graminis f. sp. tritici (Bgt), is a serious wheat disease both in China and worldwide. The disease usually results in a 13 to 34 % loss in yield and can be as severe as 50 % when the flag leaves are severely infected during the heading and filling stages (Griffey et al. 1993). Therefore, an in-depth pathological study of disease-resistant varieties of wheat has important implications in food production.

GeneFishing (http://www.seegene.com/), established by Hwang et al. (2003), is a technique used to identify differences in gene expression based on polymerase chain reaction (PCR). The annealing control primer (ACP) is composed of a tripartite structure with a polydeoxyinosine [poly(dI)] linker between the 3′ end target core sequence and the 5′ end non-target universal sequence, which was designed to improve the specificity of PCR amplification (Hwang et al. 2003). The technology has been widely used to screen differently expressed genes in plants and animals (Ding et al. 2012; Effendy et al. 2013; Takemura et al. 2013).

Auxin, a common growth-stimulating phytohormone, regulates a wide variety of growth and developmental processes in plants (Lee et al. 2013). An increasing number of experiments reveal that auxin is involved in responses to biotic and abiotic stresses (Llorente et al. 2008; Domingo et al. 2009; Lee et al. 2009; González-Lamothe et al. 2012; Qi et al. 2012; Müller et al. 2014). Auxin-repressed protein (ARP) genes, including dormancy-related proteins (Stafstrom et al. 1998) and glycine-rich protein 1 (Shimizu et al. 2006), downregulate auxin. Several ARP genes have been isolated from plant species, such as strawberry (Reddy and Poovaiah 1990), apple (Lee et al. 1993), black locust (Park and Han 2003), tobacco (Steiner et al. 2003), cucumber (Shimizu et al. 2006), Elaeagnus umbellate Thunb (Kim et al. 2007), tree peony (Huang et al. 2008), peanut (Guimarães et al. 2010), tea plant, Chenopodiaceae, Brassica rapa (Lee et al. 2013), and pear (Shi et al. 2013). Abiotic stresses induced expression of these ARP genes in hot pepper (Hwang et al. 2005), Arabidopsis (Mlynárová et al. 2007), and B. rapa (Lee et al. 2013). Biotic stresses induced expression of these ARP genes in peanuts (Guimarães et al. 2010; Morgante et al. 2013). However, little is known about the expression of ARP genes under biotic stresses in wheat.

This study aimed to identify wheat powdery mildew-resistance genes from the hybrid wheat germplasm SN6306, which is highly resistant to powdery mildew and originated from the intergeneric hybridization between Yannong 15 and Elytrigia intermedium (Host) Nevski. In this work, wheat TaARP1 was isolated and its expression was characterized after infection with wheat powdery mildew. Furthermore, Barley Stripe Mosaic Virus-based virus-induced gene silencing (BSMV-VIGS) was used to downregulate the expression of wheat (Triticum aestivum L.) endogenous TaARP1 gene to elucidate its function in the response to Bgt. TaARP1 was upregulated by powdery mildew, and BSMV-VIGS showed that it is a potential resistance gene. Results facilitate the understanding of the function of TaARP1 in the response of wheat to biotic stress.

Materials and Methods

Plant Materials and Growth Conditions

Wheat alien disomic addition line germplasm SN6306 (2n = 44), which originated from the intergeneric hybridization between wheat cultivar Yannong 15 (2n = 42) and E. intermedium (Host) Nevski (2n = 42) and is highly resistant to powdery mildew, was stored in our laboratory. Seeds were transferred to a mixture of substrate and vermiculite (1:1) after being soaked in water for 24 h at 24 °C. Plants were grown in a phytotron set under long-day conditions (16 h light and 8 h dark) at 24 °C during the day and 18 °C at night with 75 % relative humidity and approximately 12,000 lx (white fluorescent lamps). Chinese Bgt isolate E09 was used for the pathogen inoculation test by the method described previously (Zhang et al. 2012).

DNA and RNA Isolation

Genomic DNA was extracted from SN6306 leaves using cetyltrimethylammonium bromide (CTAB) (Saghai-Maroof et al. 1984). Total RNA was isolated from the control and the powdery mildew-infected plants at 12 h using EasyPure™ Plant RNA Kit (TransGen, Beijing, China) and dissolved in RNase-free water according to the protocol of the manufacturer. Concentration and purity (A260/A280) of extracted DNA and RNA were subsequently measured by NanoPhotometer P360 (Implen GmbH, München, Germany). Integrity of the genomic DNA and total RNA was evaluated by electrophoresis on 0.8 % agarose gel at 180 V and stained with ethidium bromide. First-strand complementary cDNA (cDNA) was synthesized from 3 μg total RNA with dT-ACP1 based on the protocol described in the DEG™ GeneFishing Kit (Seegene, Seoul, Korea).

GeneFishing and Cloning

Eighty different ACP primers were used to identify the differentially expressed genes in the final volume of 20 μL containing 7 μL of diluted first-strand cDNA, 1 μL of dT-ACP2 (10 μM), 2 μL of 5 μM arbitrary ACP, and 10 μL of 2× MasterMix® (Seegene). The PCR program was divided into three stages. The first stage was one cycle of a low specific reaction at 94 °C for 5 min, followed by 50 °C for 3 min and 72 °C for 1 min. The second stage was 40 cycles of a more specific reaction at 94 °C for 40 s, followed by 65 °C for 40 s and 72 °C for 40 s. The final stage was an extension at 72 °C for 5 min. Amplified PCR products were separated using 2 % agarose gel stained with ethidium bromide. The gel was screened by Tanon 3500 Gel image system (Tanon, Shanghai, China). Differentially expressed bands were extracted from the agarose gel using a gel extraction kit (Biomiga). pMD18-T TA cloning vector (TaKaRa, Biotech, Dalian, China) was used for subcloning of extracted cDNAs using T4DNA ligase, and transformed in DH5a Escherichia coli competent cells. The sequencing of plasmids, identified to be positively inserted, was accomplished by Biosun (Shanghai, China). Homology of all inserted sequences was analyzed by running the BLAST programs in the National Center for Biotechnology Information (NCBI) (http://blast.ncbi.nlm.nih.gov/Blast.cgi) and the Triticeae Full-Length CDS Database (TriFLDB) (http://trifldb.psc.riken.jp/ver.2.0/blast.pl).

Isolation and Analysis of Full-Length cDNA and DNA of TaARP1

Results of the basic local alignment show that the primer (ARPF: GGATCTTATCGGTGGGTAG, ARPR: GCTTGCTCTGAGGCTGTTA) was designed to obtain the full-length cDNA and DNA of TaARP1. Analysis of the sequence was performed as described previously (Shi et al. 2013; Zhang et al. 2012).

Quantitative RT-PCR Analysis of TaARP1

Quantitative PCR was performed using SYBR Green Real Master Mix (Tiangen Biotech) according to protocol by the manufacturer. CFX96™ Real-Time System (Bio-Rad, Hercules, CA) was used under the following conditions: 95 °C for 90 s, 40 cycles of 94 °C for 10 s, 63 °C for 15 s, and 72 °C for 15 s. Melting curves were obtained at 65 to 95 °C with the fluorescence continuously monitored. The primers of TaARP1 used in quantitative reverse transcription PCR (qRT-PCR) were TaARP1F: AGCAACCACCGCTAAATAAGAG and TaARP1R: AAGCTCAAGACAAACCACATGC. Taβ-actin expression was used as the internal standard (F: CCGGCATTGTCCACATGAA, R: CCAAAAGGAAAAGCTGAACCG). Then, qRT-PCR were performed in triplicate and repeated. Fold changes were calculated by 2−∆∆C T (Livak and Schmittgen 2001). RNA from leaves of the control and the powdery mildew-infected plants at 12, 24, 48, and 72 h were used for qRT-PCR. RNA isolation and synthesis of the first strand were processed as described above.

BSMV-Mediated Gene Silencing of TaARP1 in Wheat

A partial sequence (262 bp) of TaARP1 was cloned and ligated to the vector of BSMV pCa-γbLIC with the primer (VF: AACCACCACCACCGTTTGCTGCTCTTT, VR: AAGGAAGTTTAAGCTCTGAGGCTGTTATTGTCTTAC; the adaptors are underlined) according to Yuan et al. (2011), with tobacco (Nicotiana benthamiana) as an intermediate host. This method is divided into two phases. First, the Agrobacterium-containing vectors were introduced into tobacco leaves to produce the virus, and the NbPDS gene of tobacco was used as a reference gene in the experiments for a photobleaching phenotype (pCaBS-α, pCaBS-β, and pCaBS-γb:NbPDS). Second, the leaves were harvested, ground, and soaked in 20 mM Na-phosphate buffer (pH 7.2) containing 1 % celite. The extract was mechanically inoculated into the leaves of wheat SN6306 during the two-leaf stage. Meanwhile, TaPDS gene of wheat was used as the reference gene for a photobleaching phenotype (pCaBS-α, pCaBS-β, and pCaBS-γb:TaPDS). Then, the expression level of the endogenous TaARP1 gene in leaves was tested by semi-quantitative RT-PCR using gene-specific primers (ARPF: GGATCTTATCGGTGGGTAG, ARPR: GCTTGCTCTG AGGCTGTTA) that anneal outside the region targeted for silencing. Wheat Taβ-Actin gene was used as the internal standard. Microscopic analysis of Bgt-inoculated leaves that exhibited symptoms of powdery mildew after BSMV-induced gene silencing of TaARP1 was conducted according to Yuan et al. (2011). Staining with Coomassie brilliant blue (CBB) according to the reported protocol (Lee et al. 2009) was employed before observation with light microscopes.

Results

Isolation and Analysis of Candidate Partial cDNAs from GeneFishing

GeneFishing was performed on the control and powdery mildew-infected plants using 80 different primers at 12 h. Partial amplification results are shown in Fig. 1. Differentially expressed bands between the control and the powdery mildew-infected plants were extracted and cloned using a TA cloning vector for sequencing. Table 1 shows 23 partial cDNAs, approximately 100 to 350 nt, obtained from SN6306 inoculated with powdery mildew. Of the 23 partial cDNA sequences, 7 (30 %) were potentially associated with disease defense (clones 69B1, 74B7, 36116, 33B2, 23B7, 30B1, and 23B5), 1 (4 %) was predicted as uncharacterized protein (clone 1B1), and the rest were related to basic metabolism (11, 48 %), transcription (2, 9 %), and signal transduction (2, 9 %).

Fig. 1
figure 1

Partial GeneFishing results of SN6306 cDNA amplified with the ACP primer pairs. M Takara DL2000 marker, A control of SN6306, B powdery mildew-infected SN6306 at 12 h, A69, A74, and A79 different ACP arbitrary primers. Arrows indicate differential amplified cDNA bands

Full size image
Table 1 Sequence similarities and characterization of differentially expressed transcripts
Full size table

Isolation and Characterization of TaARP1 Gene

A 168-nt cDNA sequence of the seven partial cDNAs presumed to encode for disease defense-related proteins was obtained using GeneFishing with clone number 69B1 (Table 1). The full-length sequence and the powdery mildew resistance of clone 69B1 were a prior consideration. Running BLASTN in TriFLDB revealed that a gene called tplb0007b16 is 95 % identical with the partial cDNA of interest. Specific primer pairs were designed to obtain the full-length sequences of the cDNA and DNA. This powdery mildew-induced gene was named TaARP1, which is 93 % identical with tplb0007b16 in TriFLDB. Full-length cDNA 751 bp and DNA 1072 bp of TaARP1 were obtained with a putative open reading frame (ORF) of 348 bp encoding for 115 amino acids and containing two introns (Fig. 2a). TaARP1 does not have a signal peptide (SignalP 4.1 Server) and thus may have a cytoplasmic function. The auxin-repressed conserved domain was detected in the protein (Fig. 2b) using BLASTX in NCBI.

Fig. 2
figure 2

The sequence and structure information of TaARP1 from wheat. a Nucleotide sequence and deduced amino acid sequence of TaARP1. b Putative conserved domain of the TaARP1 protein

Full size image

Multiple Alignment and Phylogenetic Analysis of ARP Proteins

Multiple alignments and phylogenetic analyses of the amino acid sequence of TaARP1 with 14 ARPs from other species derived from NCBI were conducted. The multiple alignments showed that TaARP1 sequence is highly identical with other ARP proteins, particularly in the area of the two domains (Fig. 3a). The resulting phylogenetic tree shows that TaARP1 in wheat is similar to ARP proteins in other monocotyledons, such as Hordeum vulgare subsp. vulgare, Aegilops tauschii, and Brachypodium distachyon (Fig. 3b).

Fig. 3
figure 3

Multiple alignments (top) and phylogenetic analyses (bottom) of the amino acid sequence of TaARP1 with 14 ARPs from other species. Identical amino acid sequences are highlighted in black and sequence similarities in grey. Domains 1 and 2 represent the regions on N- and C-terminal ends. The minimum evolution tree was constructed in DNAMAN Version 4.0. The accession numbers of ARPs in GenBank are listed as follows: TaARP1 (KJ513671), AtARP (EMT13445), HvARP (BAJ89735), BdARP (XP_003562172), SiARP (XP_004982572), ZmARP (NP_001152232), OsARP1 (AAL78369), EuARP1 (AAC62104), SvARP (AAS75891), AtARP1 (NP_564305), BrARP1 (AAO32054), MjARP (AAN16890), RpARP (AAG33924), FaARP (AAA73872), and PpARP1 (AGG19165)

Full size image

Expression Patterns of TaARP1 Genes in Response to Powdery Mildew

To evaluate the mechanism of TaARP1 in the response to powdery mildew, its expression patterns were analyzed. qRT-PCR showed that the expression level of TaARP1 was upregulated by more than ten times compared to non-inoculated controls, especially at 12 h after inoculation (Fig. 4), and the expression pattern was the same as that with GeneFishing.

Fig. 4
figure 4

Relative expression level of TaARP1 induced by powdery mildew at different stages (0, 12, 24, 48, and 72 h after infection). Non-inoculated wheat (0 h) was used as the control. hai hours after inoculation

Full size image

TaARP1 Gene Silencing Allowed Powdery Mildew Growth

To further investigate the mechanism of TaARP1 in the resistance to powdery mildew, BSMV-based gene silencing was performed. The NbPDS gene of tobacco was used as a reference gene in the experiments, according to Yuan et al. (2011). Large areas of photobleaching occurred in leaves infected with BSMV:NbPDS of N. benthamiana (Fig. 5a). This demonstrated that NbPDS was silenced; thus, BSMV was successfully produced in N. benthamiana.

Fig. 5
figure 5

Phenotype of the silencing of endogenous phytoene desaturase gene (PDS) by Agrobacterium-mediated BSMV-VIGS. a The silencing of endogenous NbPDS gene in leaves of N. benthamiana infected with BSMV:NbPDS (a) and TaPDS gene in leaves of SN6306 infected with BSMV:TaPDS (b) showed the phenomenon of photobleaching compared with untreated leaves (MOCK) and the empty vector (BSMV:00). The leaves of each treatment were shown in the insert panel of a

Full size image

Wheat TaPDS was another reference gene for a photobleaching phenotype in the study. Figure 6a shows photobleaching of the leaf of wheat germplasm line SN6306 infected with BSMV:TaPDS (Fig. 5b), indicating that BSMV has been successfully produced.

Fig. 6
figure 6

The phenotype of auxin-repressed protein gene (TaARP1) silencing in wheat. a Gene silencing phenotypes on wheat leaves 10 days after powdery mildew conidia dusting infected with BSMV:00 or BSMV:TaARP1. b Effects of downregulation of TaARP1 on powdery mildew pathogenesis and confirmation by semi-quantitative RT-PCR of the relative transcript levels of endogenesis TaARP1 in leaves infected with the BSMV:00 or BSMV:TaARP1 with the gene-specific oligonucleotide primers. Amplified Taβ-Actin gene served as an internal control. c Microscope observations of powdery mildew colonies and mycelia development at 10 days after conidial dusting appearing on leaf segments infected with BSMV:00 or BSMV:TaARP1. The bar is 200 μm

Full size image

Powdery mildew was inoculated into the leaves with silenced TaARP1. Ten days after conidia dusting, a microscopic analysis of Bgt-inoculated wheat leaves, with BSMV-silenced TaARP1, was conducted. The leaves with BSMV:TaARP1 were notably susceptible to powdery mildew infections than those infected with BSMV:00 (Fig. 6a). Semi-quantitative RT-PCR was performed to confirm silencing of TaARP1 in BSMV:TaARP1 leaves (Fig. 6b). Ten days after conidial dusting, development of outstretched mycelia and colonies on leaf segments of SN6306 infected with BSMV:TaARP1 was confirmed by staining with Coomassie brilliant blue (CBB) compared with BSMV:00 (Fig. 6c).

Discussion

Several methods are used to screen differentially expressed genes. Methods include suppression subtractive hybridization (SSH) (Diatchenko et al. 1996), GeneFishing (Hwang et al. 2003), microarray (DeRisi et al. 1996), two-dimensional gel electrophoresis (2-DE) (Wilkins et al. 1996), and RNA-Seq (Trapnell et al. 2010). In this study, GeneFishing was applied to find differentially expressed genes because of favorable features such as easy execution and its wide range of PCR products (Hwang et al. 2003). Twenty-three partial cDNAs were obtained in the response to the infection of powdery mildew using 80 ACP primers from wheat SN6306, as these sequences showed significant similarity to H. vulgare or T. aestivum. Several acquired cDNAs showed similarity to genes associated with plant defense, such as TGACG MOTIF-BINDING FACTOR (clone 36113) (Foley and Singh 2004), ethylene-responsive element binding protein 2 (clone 27B5) (Park et al. 2001), Bowman–Birk type trypsin inhibitor like (clone 74B7) (Chilosi et al. 2000), and auxin-repressed 12.5 kDa protein-like isoform 2 (clone 69B1) (Guimarães et al. 2010).

The best studied auxin, indole-3-acetic acid (IAA), is a key plant growth regulator that also impacts plant–pathogen interactions. Treatment with higher IAA can enhance disease caused by Botrytis cinerea or Pseudomonas syringae in wild-type Arabidopsis (González-Lamothe et al. 2012). Overexpression of the YUCCA1 auxin biosynthesis gene, which elevated endogenous auxin levels, led to enhanced susceptibility to the bacterial plant pathogen P. syringae strain DC3000 in Arabidopsis through a mechanism that is independent of the suppression of SA-mediated defense (Mutka et al. 2013).

Auxin-repressed protein (ARP) gene was characterized as a repressor of plant growth and an activator of disease resistance based on genetic complementation, gene silencing, and overexpression analyses (Zhao et al. 2014). The dual role of ARP1 in plant growth and disease resistance is related to transcriptional regulation of inhibiting the expression of AUXIN RESPONSE FACTOR gene ARF8, and recruiting the NPR1 gene, which is essential for the salicylic-acid-mediated defense, respectively (Zhao et al. 2014). The ARP1/GERI1 recessive mutant tobacco-repressed expression of defense response genes is consistent with impaired resistance to diseases caused by viral, bacterial, and oomycete pathogens (Zhao et al. 2014). AsARPs in Arachis stenosperma significantly responded to the infection of Meloidogyne arenaria (Morgante et al. 2013).

In the present study, a differential expressed gene TaARP1 fragment was isolated by GeneFishing method, and the full-length gene was cloned. Real-time quantitative reverse transcription PCR and Barley Stripe Mosaic Virus-mediated gene silencing analyses showed that it markedly responded to the infection of the fungal pathogen B. graminis f. sp. tritici (Bgt) in the hybrid wheat alien disomic addition line germplasm SN6306. qRT-PCR revealed that powdery mildew upregulated TaARP1 by three to ten times in the early stage, implying that TaARP1 participates in the plant defense reaction. Interestingly, this gene was slightly expressed and not induced by the fungal pathogen B. graminis (data not shown). It is of great interest to illustrate the molecular mechanism of TaARP1 gene in the regulation of the resistance to the fungal pathogen. And the underlying reasons of the different expression patterns of TaARP1 gene in the powdery mildew-resistant hybrid wheat alien disomic addition line germplasm SN6306 and its susceptible parent wheat cultivar Yannong 15 needed thorough investigation.

BSMV, a tripartite genome consisting of RNAs α, β, and γ, is a positive-sense RNA virus with a broad experimental host range (Jackson et al. 2009). BSMV-mediated gene silencing is a useful and time-saving tool in the identification of gene function in the grass family (Yuan et al. 2011). Leaves whose TaARP1 was silenced were notably susceptible to powdery mildew infections than leaves in the control. This result further proved TaARP1 function in the plant response to powdery mildew.

In summary, a total of 23 DEGs upon inoculation of leaves with Bgt were isolated from wheat alien disomic addition line germplasm SN6306 with high resistance to powdery mildew by using GeneFishing™ technology. The gene sequence and expression pattern of TaARP1 were further studied. The mechanism of TaARP1 in the response to Bgt was primarily characterized by BSMV-induced gene silencing. Results of the present study may present candidate genes that are useful for breeding powdery mildew-resistant wheat cultivars. To determine the mechanism of TaARP, an overexpression vector of TaARP was constructed and used to transform wheat. Further work will clarify the molecular mechanisms of protein interactions associated with resistance to powdery mildew infection.

The sequence described in this article was submitted to the GenBank database under accession number KJ513671.