The Sclerotinia sclerotiorum-inducible promoter pBnGH17D7 in Brassica napus: isolation, characterization, and application in host-induced gene silencing

Abstract Sclerotinia stem rot (SSR), caused by Sclerotinia sclerotiorum, is among the most devastating diseases in Brassica napus worldwide. Conventional breeding for SSR resistance in Brassica species is challenging due to the limited availability of resistant germplasm. Therefore, genetic engineering is an attractive approach for developing SSR-resistant Brassica crops. Compared with the constitutive promoter, an S. sclerotiorum-inducible promoter would avoid ectopic expression of defense genes that may cause plant growth deficits. In this study, we generated a S. sclerotiorum-inducible promoter. pBnGH17D7, from the promoter of B. napus glycosyl hydrolase 17 gene (pBnGH17). Specifically, 5'-deletion and promoter activity analyses in transgenic Arabidopsis thaliana plants defined a 189 bp region of pBnGH17 which was indispensable for S. sclerotiorum-induced response. Compared with pBnGH17, pBnGH17D7 showed a similar response upon S. sclerotiorum infection, but lower activity in plant tissues in the absence of S. sclerotiorum infection. Moreover, we revealed that the transcription factor BnTGA7 directly binds to the TGACG motif in pBnGH17D7 to activate BnGH17. Ultimately, pBnGH17D7 was exploited for engineering Sclerotinia-resistant B. napus via host-induced gene silencing. It induces high expression of siRNAs against the S. sclerotiorum pathogenic factor gene specifically during infection, leading to increased resistance.


Introduction
Due to the world's growing population and changing climate, powerful tools are needed to engineer desirable agronomic traits in crops that increase productivity. Compared with conventional breeding, genetic engineering enables the introduction, removal, or modification of desired genes in a specific crop with minimal modifications to the crop genome (Dong and Ronald, 2019). Three different types of promoters are typically employed to control transgene expression in plant genetic engineering: constitutive promoters, inducible promoters, and tissue-specific promoters. In plants, constitutive promoters such as the Cauliflower mosaic virus (CaMV) 35S and maize ubiquitin promoters are most commonly used (Benfey and Chua, 1990;Cornejo et al., 1993). However, constitutive promoters drive transgene expression throughout all stages of plant development, in most plant tissues, and under all conditions, making it challenging to control specific temporal and spatial expression of transgenes (Holtorf et al., 1995;Sunilkumar et al., 2002). In addition, constitutive promoters trigger a continuously high level of transgene expression, which often leads to unnecessary nutrition consumption and plant growth deficits (Hull et al., 2000;Pino et al., 2007). In contrast to constitutive promoters, transgenes driven by tissue-specific promoters could achieve optimal effectiveness (Koellhoffer et al., 2015;Li et al., 2019). Inducible promoters that regulate transgene expression in a desired temporal and/or spatial manner lessen the incidence of unexpected adverse effects on plant growth. Therefore, tissue-specific and inducible promoters are the preferred route in plant genetic engineering.
Plants deploy a wide range of immune defense strategies against pathogens that can be exploited to confer disease resistance through genetic engineering. However, genetic immunity to disease often comes with the cost of reduced plant growth and reproduction (Ning et al., 2017;Guo et al., 2018). Yield penalties caused by enhanced disease resistance have been described in several crop species. The wheat resistance (R) gene Wsm1 conferred plant resistance to Wheat streak mosaic virus, but caused a relative 11-28% reduction in yield, even in unstressed conditions (Sharp et al., 2002). A similar observation was made with powdery mildew-resistant barley, in which the mlo R gene-mediated resistance to Bgh (powdery mildew) reduced grain yield by 4% (Jørgensen, 1992). In Arabidopis thaliana, non-expresser of pathogenesis-related gene 1 (AtNPR1), a master immune regulatory gene, was a prime gene to be employed in disease resistance engineering, in that AtNPR1 conferred resistance to diverse pathogens in different plant species (Cao et al., 1998;Lin et al., 2004;Quilis et al., 2008;Wally et al., 2009). However, constitutive overexpression of AtNPR1 resulted in reduced plant height and yield loss in rice, limiting its potential application in broad-spectrum resistance (Quilis et al., 2008). Thus, immune responses during genetic engineering should be precisely regulated to mitigate the cost of resistance. This can be achieved by inducing the expression of defense genes at particular times or in specific plant tissues (Karasov et al., 2017). It is believed that expression of defense genes under the control of pathogen-inducible promoters could maintain the balance between plant growth and disease resistance during genetic engineering.
To date, a number of natural pathogen-inducible promoters have been isolated and identified in plants, including: the Magnaporthe grisea-inducible promoters OsR2329, OsR2184, and OsPBZ1 in rice (Sasaki et al., 2007); the Phytophthora sojae-inducible promoter PcCMPG1 in Petroselinum crispum (Kirsch et al., 2001); the Bgh-and Rhynchosporium secalis-inducible promoter HvGER4c in barley and wheat (Himmelbach et al., 2010); the Uncinula necator-inducible promoter VpSTS in grapevine (Xu et al., 2010); the Xanthomonas axonopodisinducible promoter NtpPPP1 in Citrus sinensis Osbeck (Zou et al., 2014); and the Erwinia amylovora-inducible promoters Ntstr246C, Ntsgd24, and StPgst1 in pear and apple (Malnoy et al., 2003(Malnoy et al., , 2006. However, research on pathogen-inducible promoters in Brassica crops is limited. As a major oil crop worldwide, oilseed rape (Brassica napus) provides vegetable oil for humans and edible fodder for animals. However, the growth of oilseed rape is constantly threatened by Sclerotinia stem rot (SSR). This disease, caused by the broad-host-range fungal pathogen Sclerotinia sclerotiorum, leads to severe reduction in seed yield and quality worldwide. SSR resistance in B. napus is a quantitative trait, determined by multiple minor quantitative trait loci (QTLs) (Wu et al., 2016a). However, none of the QTLs has been cloned, limiting their utilization in SSR resistance breeding. Reverse genetic analysis of SSR resistance has been conducted, and several defense genes have been identified (Ding et al., 2021).
Genetic engineering for SSR resistance using the identified defense genes is a promising strategy for controlling SSR. However, overexpression of defense genes may cause plant growth deficits (Ning et al., 2017). Therefore, the regulation of defense gene expression precisely controlled by S. sclerotiorum-inducible promoters is an optimal strategy to generate SSR-resistant varieties with stable B. napus yields. While two synthetic promoters containing pathogen-related cis-acting elements have been described that respond to S. sclerotiorum (Shokouhifar et al., 2011a, b). To date, S. sclerotiorum-inducible promoters in plants have not been identified or explored.
Here, we identified an S. sclerotiorum-inducible promoter derived from B. napus and highlighted its potential application in agriculture. The crucial promoter region and the core cis-elements that respond to S. sclerotiorum were investigated by 5ʹ-deletion analysis and site-directed mutagenesis. The transcription factor (TF) that directly binds to this crucial promoter region was verified by yeast one-hybrid (Y1H) assay, dual-luciferase assay (dual-LUC), and EMSA. Finally, this promoter was used to engineer SSR-resistant B. napus and tested for its specificity. This research highlights the potential of a S. sclerotiorum-inducible promoter to facilitate precise genetic engineering of SSR-resistant B. napus and potentially other crops.

Plant materials, abiotic stress and phytohormone treatments, and S. sclerotiorum inoculation
Brassica napus line J9712 was kindly provided by Professor Yongming Zhou (Huazhong Agricultural University, Wuhan, Hubei, China). Plants were grown in nutrient solution in the greenhouse for 4 weeks, and then the entire seeding plants were subjected to various treatments, including osmotic, cold, heat, and salt stresses, as well as to hydrogen peroxide (H 2 O 2 ) and hormone treatments with salicylic acid (SA), abscisic acid (ABA), methyl jasmonate (MeJA), and ethephon (ETH), and S. sclerotiorum inoculation. For the osmotic stress and salt stress treatments, seedlings were transferred to nutrient solutions containing 15% polyethylene glycol 6000 (PEG6000) and 200 mM NaCl, respectively, and sampled at 6 h after treatment, as described by Li et al. (2021a, b). For the cold and heat treatments, seedlings were transferred to growth chambers with light intensity of ~300 µmol⋅m −2 ⋅s −1 at 4 °C and 42 °C, respectively, and sampled 1 h and 6 h post-treatment (Li et al., 2021b). For the H 2 O 2 and hormone treatments, seedlings were sprayed with 100 µM H 2 O 2 (Mierek-Adamska et al., 2019), 1 mM SA (Wang et al., 2014a), 100 µM MeJA (Wang et al., 2014a), 100 µM ABA (Li et al., 2021b), and 100 µM ETH (Xue et al., 2020), and sampled at 3 h and 6 h post-treatments. For S. sclerotiorum inoculation, the S. sclerotiorum isolate SS-1 was cultured on potato dextrose agar (PDA; Becton, Dickinson and Company, Franklin Lakes, NJ, USA), as described by Wu et al. (2013). Agar plugs of 7 mm in diameter with S. sclerotiorum were used for detached leaf inoculation of unfolded leaves of B. napus line J9712, as described by Wu et al. (2021). Mock-inoculated leaves were treated with 7 mm diameter agar plugs. Tissues extending 10 mm beyond the inoculation site on the leaves were harvested at 3, 6, and 12 h after S. sclerotiorum or mock inoculation and stored at -80 °C. Three biological replicates were performed, and five plants were used for each biological replicate.
To construct the promoter::GUS (β-glucuronidase) expression vector, promoter fragments were cloned into pBI101 at EcoRI and BamHI restriction enzyme sites via homologous recombination (ClonExpress II One Step Cloning Kit, Vazyme, Nanjing, China).

A. thaliana transformation and treatments
Promoter::GUS recombinant plasmids were introduced into Agrobacterium tumefaciens strain GV3101 by electroporation. Arabidopsis thaliana wild-type plants Columbia-0 (Col-0) were used for transformation via the floral dipping method (Zhang et al., 2006). Seeds of the T 0 generation were selected on Murashige and Skoog (MS) medium supplemented with kanamycin (50 mg l -1 ), and the positive plants were further verified by PCR. T 1 and T 2 transgenic plants were grown in nutrient soil and, after verification by PCR, they were transplanted for hormone treatments, S. sclerotiorum inoculation, and the GUS staining assay. All A. thaliana plants were grown in growth chambers under a 16 h light/8 h dark photoperiod (~300 µmol⋅m −2 ⋅s −1 ) at 22 °C during the day and 20 °C at night, and 60% relative humidity.
For S. sclerotiorum inoculation, 4-week-old unfolded leaves were detached from A. thaliana plants, and placed on agar for leaf inoculation. Mycelial agar plugs (2 mm in diameter) punched from the margin of a 2-day-old cultures of S. sclerotiorum grown on PDA were used as the inoculum and were closely appended to the adaxial surface of leaves, according to Wu et al. (2013). Mock-inoculated leaves were treated with 2 mm diameter agar plugs. The inoculated and mock-inoculated leaves were covered with plastic film to maintain moisture at 22 °C. At 12 h and 24 h post-inoculation (hpi), the inoculated leaves were collected for GUS staining and quantiative real-time PCR (qRT-PCR) analysis, respectively. For H 2 O 2 and hormone treatments, 4-week-old A. thaliana plants were sprayed with 1 mM SA (Kovacs et al., 2015), 200 µM MeJA (Jiang et al., 2014), 7 mM ETH (Qiu et al., 2015), and 100 μM H 2 O 2 . At 6 h and 12 h post-treatment, A. thaliana leaves were collected for GUS staining and qRT-PCR analysis, respectively. GUS staining was performed on four independent transgenic lines for each treatment. For qRT-PCR analysis, three independent biological replicates were performed, each with three technical replicates.

Total RNA extraction and qRT-PCR analysis
Total RNA from different tissues and organs of A. thaliana and B. napus was extracted according to the TRIzol method using an RNAiso reagent kit (Vazyme) according to the manufacturer's instructions. The total RNA was reverse transcribed into first-strand cDNA with a HiScript II 1st Strand cDNA Synthesis Kit (Vazyme). qRT-PCR was carried out using AceQ Universal SYBR qPCR Master Mix (Vazyme) in an ABI Step One Plus real-time PCR system (Applied Biosystems Inc., Foster City, CA, USA). The relative expression of each gene was calculated using the 2 -△△Ct method (Livak and Schmittgen, 2001). BnUBC9 (BnaC08g12720D) and BnUBC10 (BnaA10g06670D) in B. napus (Wu et al., 2016b), AtEF-1α (At5g60390) and AtUBQ10 (At5g53300) in A. thaliana (Yang et al., 2020, Zhu et al., 2013, and SsActin (SS1G_08733) and Sstub1 (SS1G_04652)  in S. sclerotiorum were used as reference genes. The qRT-PCR primers used are shown in Supplementary Table S1. All qRT-PCR experiments were performed with three biological replicates, each with three technical replicates.

Y1H assay
A Y1H assay was performed as described by Ou et al. (2011). An A. thaliana TF library containing 1589 GAL4-AD-fused TFs was provided by Professor Lijia Qu (Peking University, Beijing, China). The pBnGH17 D7 subfragment was amplified by PCR, cloned into a pHisi-1 vector, and then transformed into yeast strain YM4271 as bait. The A. thaliana TF pooled library strains and the bait clone were grown in SD-Leu and SD-His medium overnight, respectively. The pooled library strains and bait clones were mixed at equal volumes (20 μl per well) and transferred to new 2 ml 96-well plates with yeast extract peptone dextrose medium. After 1 d growth with shaking at 200 rpm at 30 °C, the mating products were diluted 10-fold with water and then the diluted mating products (10 μl per well) were plated on screening plates [SD-Leu -His+15 mM 3-aminotriazole (3-AT)] and subsequently grown for 3 d.
The point-to-point Y1H assay was carried out according to Yang et al. (2011). Briefly, the pBnGH17 D7 subfragment was cloned into the pHIS2 vector as bait. The full-length coding sequences (CDSs) of the TGA TF genes BnTGA7 (BnaA07g33790D) and BnTGA3 (BnaC05g17700D) were amplified by PCR and cloned into the pGADT7 vector as prey. Two plasmids were co-transformed into yeast strain Y187. Transformed clones were cultured on SD/-His/-Leu/-Trp selective medium containing 50 mM 3-AT for 3 d at 30 °C. The p53HIS2 and pGAD-Rec2-53 vectors were used as positive controls, and the pHIS2 and pGAD-Rec2-53 vectors were used as negative controls.

Dual-luciferase reporter gene assay
The full-length CDSs of BnTGA7 and BnTGA3 were cloned into the pGREEN II 62-SK vector to generate effector constructs. The pBnGH17 D7 subfragment was inserted ahead of the firefly luciferase (LUC) gene in the pGREEN II 0800-LUC vector to generate a reporter construct. Then, recombinant effector and reporter constructs and the empty vector pGREEN II 62-SK were introduced into A. tumefaciens strain GV3101 (with the helper PSoup-P19 plasmid) by electroporation and used to infect Nicotiana benthamiana leaves in the light (16 h/day) at 25 °C for 2 d by Agrobacterium-mediated infiltration to induce transient gene expression (Hellens et al., 2005). The activities of the firefly luciferase (LUC) and Renilla luciferase (REN) were determined with a Dual-Luciferase Reporter Kit (Vazyme) according to the manufacturer's instructions and detected with a microplate reader (Tecan Spark, Tecan Trading AG, Zurich, Switzerland).

EMSA
The full-length CDS of BnTGA7 was cloned into the pGEX6P-1 vector to generate the glutathione S-transferase (GST)-BnTGA7 fusion protein, which was expressed in Escherichia coli strain BL21. Expression and purification of the GST-BnTGA7 protein were performed according to the manufacturer's instructions (Transgen, Beijing, China). A 30 bp DNA sequence containing the TGACG sequence was synthesized by Beijing Qingke Biotechnology and labeled with an EMSA Probe Biotin Labeling Kit (Beyotime, Shanghai, China). EMSA was performed using a chemiluminescent EMSA kit according to the manufacturer's instructions (Beyotime). In brief, GST-BnTGA7 and the labeled probe were incubated at 25 °C for 20 min in a reaction system containing EMSA/gel-shift binding buffer. For the cold competition, 200-fold unlabeled probe was added to the reaction mixtures. These reaction mixtures were loaded on an 8% native PAGE gel; after transferring to a nylon membrane, cross-linking, and blocking the proteins, the signals were detected with a chemiluminescent EMSA kit.
Host-induced gene silencing (HIGS) of the S. sclerotiorum endo-polygalacturonase gene (SsPG1) driven by pBnGH17D7 pBnGH17 D7 was cloned into the intron-containing hairpin vector pMDC83-ihpRNAi at the PmeI and SpeI sites to replace the CaMV 35S promoter via homologous recombination with the ClonExpress II One Step Cloning Kit (Vazyme). Then, the 381 bp CDS fragment (296-676 bp) of SsPG1 was cloned into the vector in sense and antisense orientations (sense-intron-antisense cassette) via homologous recombination as described by Wu et al. (2021). The primers are shown in Supplementary Table S1.
The HIGS construct was transformed into A. tumefaciens strain GV3101 by electroporation and then transformed into B. napus line J9712 via A. tumefaciens-mediated hypocotyl transformation as described by Liu et al. (2021a). Positive transgenic B. napus plants were selected by PCR with specific primers (Supplementary Table S1). All transgenic B. napus plants used in this study were from the T 2 generation.
The resistance of transgenic B. napus plants to S. sclerotiorum was assessed by detached leaf, cotyledon, and stem inoculation, according to Wu et al. (2021). Approximately 15 plants in each of the three replicates were assessed for each line. For cotyledon inoculation, both the J9712 and T 2 transgenic lines were grown in growth chambers with light intensity of 300 µmol⋅m −2 ⋅s −1 under a 16 h light/8 h dark photoperiod at 24 °C and 60% relative humidity. For detached leaf and stem inoculation, all plants were grown in the experimental field at Yangzhou University, Jiangsu, China. The field experiment was conducted using a randomized complete block design with three replications.
Small RNA sequencing was performed to determine the expression of target gene-specific siRNAs in HIGS transgenic plants. Sclerotinia sclerotiorum-and mock-inoculated leaves of two independent transgenic T 1 lines (three plants for each treatment) were randomly selected and mixed for small RNA sequencing. Small RNA sequencing and analyses were performed as described by Wu et al. (2021).

Determination of polygalacturonase (PG) activity
To quantify PG activity in S. sclerotiorum-inoculated leaves, tissues extending 10 mm beyond the inoculation site on leaves were harvested at 24 hpi and then ground into powder with liquid nitrogen. Up to 0.1 g of each sample was used to determine the enzyme activity with the Polygalacturonase assay Kit (Solarbio, Beijing, China), according to the manufacturer's instructions. The absorbance was recorded at 540 nm using a Tecan Infinite microplate reader (Spark M200, Tecan Austria GmbH, Grodig, Austria). Enzyme activity was defined as the decomposition of polygalacturonic acid per g of sample per hour at 40 °C, pH 6.0 to produce 1 μmol of galacturonic acid.

Statistical analysis
Significance analysis was performed with Student's t-test for comparing two independent groups (*P<0.05 and **P<0.01) by SPSS 19.0 (IBM SPSS Statistics, New York, NY, USA).

Screening of the S. sclerotiorum-inducible promoter in B. napus
To determine the candidate genes induced by S. sclerotiorum, we performed transcriptomic analyses of B. napus before and after S. sclerotiorum infection using RNA sequencing (Wu et al., 2016b). Among the differentially expressed genes between the S. sclerotiorum-challenged and mock-inoculated samples, the six most strongly induced genes, comprising one cysteine-rich secretory protein-, antigen 5-, and pathogenesis-related 1 protein (CAP)-encoding gene (BnaC01g04530D), two BnGH17 genes (BnaC01g21880D and BnaA01g17540D), and three legume lectin (BnLLP) genes (BnaA05g24230D, BnaCnng78710D, and BnaC01g36130D) (Fig. 1A), were examined for their expression patterns in diverse tissues and under various stress conditions. Based on the transcriptomic database of diverse tissues of the B. napus cultivar 'ZS11' (http://yanglab.hzau.edu.cn/BnTIR) (Liu et al., 2021b), BnaC01g04530D was eliminated from further analysis due to its ultrahigh expression level in roots [the transcripts per million mapped reads (TPM) value was 7558; Supplementary Fig. S1]. The three BnLLP genes were highly expressed in the root and silique walls and moderately expressed in the leaf ( Supplementary Fig. S1), while the expression levels of the two BnGH17 genes were extremely low in most of the tissues, except in roots ( Fig. 1B; Supplementary Fig. S1). The expression patterns of BnGH17 (BnaC01g21880D) under various stress conditions and with hormone treatments were determined, and BnGH17 was not significantly induced by osmotic, cold, heat, or salt stress, or by H 2 O 2, SA, MeJA, ABA, or ETH treatment ( Fig. 1C; Supplementary Fig. S2). Consequently, the BnGH17 promoter was selected as the potential S. sclerotiorum-inducible promoter for further analysis.
To test its activity in vivo, pBnGH17 was fused with GUS and transformed into wild-type A. thaliana plants. Histochemical GUS staining was carried out in different tissues of A. thaliana, including 10-day-old seedlings, rosette leaves, mature roots, inflorescences, siliques with developing seeds, and stems. GUS activity was slightly detected in seedlings and mature roots, but not in other tissues ( Fig. 2A-F) during different growth stages. Relative expression of the GUS gene, determined by qRT-PCR, was consistent with the observed GUS staining pattern ( Fig. 2G; Supplementary Fig. S3A).
In addition, the GUS staining pattern was visible in pBnGH17::GUS transgenic A. thaliana leaves treated with S. sclerotiorum at 12 hpi, and was much stronger at 24 hpi (Fig.  2H). In contrast, GUS activity was not detectable in the leaves of the mock-inoculated controls (Fig. 2H). In this case, the relative expression of GUS detected by qRT-PCR was also consistent with the GUS staining results: GUS expression increased by nearly 200-fold at 12 hpi, and then by 400-fold at 24 hpi as compared with the mock-inoculated controls ( Fig.  2I; Supplementary Fig. S3B).
The qRT-PCR results showed that the expression of BnGH17 was not induced by H 2 O 2 , SA, MeJA, or ETH treatments (Fig. 1C), even though several hormone-responsive elements were identified in the promoter region (Table 1). To further confirm this result, we examined GUS activity in pBnGH17::GUS transgenic A. thaliana after H 2 O 2 , SA, MeJA, and ETH treatments. Unsurprisingly, GUS staining was not detected at 6 h and 12 h after the H 2 O 2 , SA, and ETH treatments. While it was not detected 6 h after the MeJA treatment, weak GUS staining was detected 12 h later (Fig. 2J), consistent with GUS expression determined by qRT-PCR ( Fig.  2K; Supplementary Fig. S3C). Collectively, our analysis supported the view that pBnGH17 is a S. sclerotiorum-inducible promoter containing an S. sclerotiorum-inducible fragment.

Deletion analysis of pBnGH17 in transgenic A. thaliana
To define the cis-regulatory sequences enabling response to S. sclerotiorum infection, a series of 5ʹ deletions of pBnGH17 were generated (Fig. 3A). Each of these fragments was further fused with GUS and transformed into A. thaliana. Histochemical GUS staining was performed with the leaves of transgenic A. thaliana inoculated with and/or without S. sclerotiorum. Initially, we generated three sequences, D1 (-500 to -1), D2 (-848 to -1), and D3 (-1260 to -1). Unexpectedly, GUS activity was not detectable in the transgenic A. thaliana plants after S. sclerotiorum inoculation (Fig. 3B), indicating that the fragment in the -1784 to -1261 region contains a compulsory molecular component for triggering GUS expression after S. sclerotiorum infection.
To further narrow down the region of pBnGH17 that is responsible for the S. sclerotiorum infection, we generated another four deletions and linked them to putative core promoter regions, named D4-D7 (Fig. 3A). Interestingly, the D4 and D7 constructs in transgenic A. thaliana exhibited GUS activity after S. sclerotiorum infection as strong as the full-length pBnGH17 construct in transgenic A. thaliana (Fig. 3B). However, the GUS staining pattern did not develop in the plants harboring D5 and D6 after S. sclerotiorum infection (Fig. 3B), indicating that the 189 bp promoter region located between positions -1615 and -1427 in pBnGH17 is fundamentally essential for S. sclerotiorum infection responsiveness.
We next compared the GUS activity pattern between fulllength pBnGH17 and D7 in different plant tissues. Consistent with plants harboring full-length pBnGH17, GUS activity of the D7 construct was undetectable in the rosette leaves (Fig.  3D), inflorescences (Fig. 3F), siliques with developing seeds (Fig. 3G), and stems (Fig. 3H) of transgenic A. thaliana under normal growth conditions. In addition, the D7 construct also abolished GUS activity that was shown in seedlings and mature roots in full-length pBnGH17 transgenic plants at the growth stage without stress (Fig. 3C, E). Hence, pBnGH17 D7 was assumed to be an ideal inducible promoter to drive transgene expression during S. sclerotiorum infection.

BnTGA7 interacts with the TGACG motif in pBnGH17
To identify the putative transcription regulators that interact with pBnGH17 to mediate S. sclerotiorum-inducible gene expression, we screened a library of 1589 GAL4-fused A. thaliana TFs in a mating-based Y1H assay with the -1615 to -1427 region of pBnGH17 (pBnGH17 -1615 to -1427 ) as bait (Ou et al., 2011). At1g77920 (AtTGA7) and At1g22070 (AtTGA3) were isolated in this high-throughput A. thaliana TF screening system. We then cloned the full-length CDS of BnTGA7 and BnTGA3 in B. napus. Using a point-to-point Y1H assay, we confirmed that both BnTGA3 (BnaC05g17700D) and BnTGA7 (BnaA07g33790D) interacted with pBnGH17 -1615 to -1427 (Fig. 4A).
We further employed a dual-LUC reporter system in N. benthamiana to examine the transcriptional activity of BnTGA3 and BnTGA7 in vivo. The pBnGH17 D7 promoter-driven firefly luciferase (LUC) reporter and 35S promoter-driven Renilla luciferase (REN, the internal control) were co-introduced into the plasmid as the reporter. Plasmids with or without BnTGA3/BnTGA7 were used as the effector. The LUC:REN ratio, reflecting the transcriptional activity of the pBnGH17 D7 promoter, was monitored after both the effector and reporter were transiently co-expressed in N. benthamiana. The coexpression of pBnGH17 D7 ::LUC/p35S::REN and BnTGA7, but not BnTGA3, remarkably increased the LUC:REN ratio (Fig. 4B), suggesting that BnTGA7 activated pBnGH17 D7 promoter-driven transcription in vivo.
TGA7, belonging to the TGA TF family, regulated the expression of defense genes (such as pathogenesis-related 1, PR-1) by directly binding to the TGACG motif in A. thaliana (Shearer et al., 2009). As one putative TGACG motif was identified within pBnGH17 -1615 to -1427 , we subsequently tested if BnTGA7 directly binds to the TGACG motif of pBnGH17 Core promoter element -61 Kadonaga et al. (1986) through EMSA in vitro. The GST-BnTGA7 recombinant protein was capable of binding to probes containing the TGACG motif, whereas the GST protein alone was not functional (Fig.  4C). The intensity of the binding signals decreased after the addition of unlabeled wild-type competitors (Fig. 4C). When the labeled probe was replaced with a mutated probe, the binding was completely abolished (Fig. 4C). Moreover, the change in expression of BnTGA7 in the leaves of J9712 was detected after S. sclerotiorum inoculation. The transcript level of BnTGA7 was slightly induced at 3 hpi, reached a peak at 6 hpi, and finally recovered at 12 hpi ( Supplementary Fig. S4A, B). The expression of BnGH17 was slightly induced at 3 hpi and kept increasing until 12 hpi ( Supplementary Fig. S4C, D), suggesting that BnTGA7 functions upstream of BnGH17 and activates the expression of BnGH17 in response to S. sclerotiorum.
To further study the contribution of the identified cis-element, the TGACG motif, on S. sclerotiorum-inducible gene expression, the TGACG motif of pBnGH17 D7 was mutated to AGGGG, and the mutated promoter fragment pBnGH17 D7-Mut was fused with the GUS gene (Fig. 4D). GUS activities were completely abolished in transgenic A. thaliana lines carrying the pBnGH17 D7-Mut ::GUS fusion after S. sclerotiorum infection (Fig. 4D), further supporting our conclusion that the TGACG motif is the core motif for S. sclerotiorum response (Fig. 4D).
Taken together, the results of the Y1H assay, dual-LUC assay, EMSA, and site-directed mutagenesis of the TF-binding site suggested that BnTGA7 directly binds to the TGACG motif of pBnGH17 and activates pBnGH17-driven transcription after S. sclerotiorum infection.
Application of pBnGH17 D7 to engineer Sclerotiniaresistant B. napus PG is one of the most crucial cell wall-degrading enzymes in S. sclerotiorum pathogenicity (Amselem et al., 2011). HIGS has been shown to induce gene silencing in pathogens by in planta expression of dsRNA or hairpin RNAs (hpRNAs) homologous to essential and/or pathogenicity genes of pathogens, conferring engineered plant protection from infection (Nowara et al., 2010). It was recently revealed that HIGS of a pathogenic factor gene (endo-polygalacturonase gene, SsPG1) of S. sclerotiorum could be an effective strategy for controlling Sclerotinia rot in B. napus . Therefore, we further investigated whether pBnGH17 D7 is able to induce the expression of siRNAs in HIGS transgenic plants exclusively upon S. sclerotiorum infection. The HIGS target sequence of SsPG1 was inserted into the intron-containing hairpin vector pMDC83-ihpRNAi to generate the HIGS construct, which is composed of the hygromycin phosphotransferase selection marker gene, the pBnGH17 D7 promoter, a spacer sequence (PDK intron), and the nopaline synthase terminator (Fig. 5A).
The HIGS construct was transformed into the B. napus line J9712, and five independent T 0 -positive transgenic plants were obtained. To determine the expression of complementary siRNAs in HIGS transgenic plants, leaves of two transgenic T 1 lines (RNAi-4 and RNAi-7) were harvested 24 h after S. sclerotiorum or mock inoculation and Positive controls, p53HIS2 and pGAD-Rec2-53. Negative controls, pHIS2 and pGAD-Rec2-53. (B) Dual-luciferase reporter assay of the interaction between BnTGA3/BnTGA7 and pBnGH17 D7 in N. benthamiana leaves. The LUC/REN value of the control was set as 1 for calibration. The error bars indicate the SD. Statistical significance was determined by Student's t-test (**P<0.01). (C) EMSA of the specific binding of recombinant BnTGA7 protein to the TGACG motif of pBnGH17 D7 . Underlining signifies the TGACG motif sequence, and asterisks represent the mutated base in the TGACG motif. GST, GST-BnTGA7, labeled probe, labeled mutant probe, and 200-fold unlabeled probe were present (+) or absent (-) in each reaction. (D) Histochemical GUS staining of two independent transgenic A. thaliana lines (T 1 ) harboring the mutated promoter fragment pBnGH17 D7-Mut ::GUS fusion at 24 h post-inoculation with S. sclerotiorum (S.s). In pBnGH17 D7-Mut , the TGACG motif was mutated to AGGGG. Col-0 (WT) was the negative control. Transgenic plants harboring the pBnGH17 D7 ::GUS fusion were the positive control. mixed for small RNA sequencing. In the mock inoculation sample, only 31 siRNAs derived from the HIGS construct were detected, accounting for 0.0003% of all small RNAs detected in this library (Fig. 5B). However, after inoculation with S. sclerotiorum, the abundance of the specific siRNAs (54 341) in the HIGS transgenic plants was vastly elevated, accounting for 0.5234% of the total small RNAs detected in this library (Fig. 5B). The most abundant siRNAs were 21-24 nt in length ( Supplementary Fig. S5) and were distributed across the target gene region (Fig. 5B). Thus, our data supported that pBnGH17 D7 triggered high expression of specific siRNAs in HIGS transgenic plants only during S. sclerotiorum infection. Next, we evaluated the SSR resistance of the HIGS transgenic lines (T 2 ) with cotyledon inoculation. Noticeable watersoaked lesions initially appeared on the adaxial surface of the cotyledons of J9712 at ~24 hpi and quickly extended to the abaxial surface; symptoms become even more severe after 48 hpi (Fig. 5C). In contrast, the fungus-induced water-soaked lesions appeared only on the adaxial surface at 48 hpi in most of the cotyledons of HIGS transgenic lines (Fig. 5C). In addition, 31.1% and 21.7% of the cotyledons of transgenic lines RNAi-4 and RNAi-7 were not successfully infected by S. sclerotiorum. The average lesion area on cotyledons of the transgenic lines was reduced by 51.8-58.2% compared with those of J9712 (Fig. 5D). The relative expression levels of SsPG1 on the S. sclerotiorum-inoculated cotyledons of RNAi-4 and RNAi-7 were reduced by 81.2% and 78.1%, respectively, at 48 hpi compared with that in J9712 ( Fig. 5E; Supplementary Fig.  S6A), indicative of an enhanced resistance of these HIGS transgenic lines to S. sclerotiorum caused by target gene silencing.
To investigate whether the S. sclerotiorum-resistant phenotype occurs in other plant tissues, we repeated the experiments in detached leaves and stems. At 48 hpi, the lesion area on leaves of transgenic lines RNAi-4 and RNAi-7 was reduced by 26.2% and 20.1%, respectively, compared with those on the J9712 leaves at 48 hpi (Fig. 5F, G). At 7 dpi, the lesion lengths on stems of transgenic lines RNAi-4 and RNAi-7 were reduced by 25.2% and 23.1%, respectively, compared with those on the J9712 stems at 7 dpi ( Supplementary Fig. S7A, B). The relative expression levels of SsPG1 in transgenic lines RNAi-4 and RNAi-7 were reduced by 74.6% and 80.8%, respectively, at 24 hpi compared with that in J9712 ( Fig. 5H; Supplementary Fig.  S6B). Furthermore, the decreased expression of SsPG1 in S. sclerotiorum at 24 hpi resulted in lower PG activity on the leaves of HIGS transgenic plants during infection than in the J9712 plants (Fig. 5I).
We critically compared the agronomic traits between the HIGS transgenic and wild-type plants. Our data indicated that the crop yield and the quality of the HIGS transgenic plants was not significantly influenced (Supplementary Table  S2). It is hence conceivable that transgene expression driven by pBnGH17 D7 is induced after S. sclerotiorum infection, thereby preventing unnecessary negative impacts on plant growth and development.

pBnGH17 D7 activity is highly induced by S. sclerotiorum
In the past two decades, QTL mapping and genome-wide association studies have uncovered the genetic architecture of SSR resistance in oilseed rape (Ding et al., 2021), and a considerable number of SSR resistance QTLs have been identified. However, none of them has been subjected to fine-mapping or map-based cloning, which may be attributable to the difficulty of identifying resistance phenotypes of complex plantmicrobe-environment interactions. This dilemma has limited the utilization of resistance QTLs in SSR resistance breeding. Thus, genetic engineering for resistance to S. sclerotiorum is a promising strategy for controlling SSR. To this end, the S. sclerotiorum-inducible promoter is valuable for driving defense gene expression in response to S. sclerotiorum infection with high specificity.
To date, tissue-specific promoters and abiotic stress-inducible promoters have been identified in B. napus, including the seed-specific promoter Napin (Sohrabi et al., 2015), the antherspecific promoter Sta 44 (Hong et al., 1997), the flower-specific promoters FSP046 and FSP061 (Li et al., 2019), and the cold-inducible promoter BN115 (Sangwan et al., 2001). However, to the best of our knowledge, S. sclerotiorum-inducible promoters in Brassica have not been reported.
Collectively, our data revealed that pBnGH17, especially pBnGH17 D7 , is an S. sclerotiorum-inducible promoter. pBnGH17 and pBnGH17 D7 activity are induced by S. sclerotiorum. While BnGH17 was expressed at a low level in most plant tissues (only root-specific expression) under normal conditions, expression was highly induced upon S. sclerotiorum infection, but not by other abiotic stresses or hormone treatments ( Fig. 1; Supplementary Fig. S2). In addition, we confirmed that the activity of the BnGH17 promoter in transgenic A. thaliana was consistent with the BnGH17 expression pattern in B. napus ( Fig. 2; Supplementary Fig. S3). Although the activity of full-length pBnGH17 was observed in the roots of transgenic A. thaliana ( Fig. 2A, C), the promoter deletion pBnGH17 D7 almost completely abolished this activity (Fig. 3C, E). These data suggested that pBnGH17, especially pBnGH17 D7 , is potentially useful for SSR resistance breeding The potential application of pBnGH17 D7 for genetic engineering of resistance to S. sclerotiorum In the past few decades, the molecular mechanism of resistance to SSR in B. napus has been systematically investigated. To date, several S. sclerotiorum resistance genes have been identified through functional genomic analysis, including BnMPK3 (Wang et al., 2019); BnMPK4 ; BnMPK6 (Wang et al., 2020b); BnWRKY33 (Wang et al., 2014b;; BnWRKY15 ; BnWRKY70 (Sun et al., 2018); BnMKK4, BnWRKY28, and BnVQ12 (Zhang et al., 2021, Preprint); BnNPR1 (Wang et al., 2020a); BnMED16 (Hu et al., 2021); and BnCCR2 (Liu et al., 2021a). Overexpression or knockout these defense-related genes might incredibly enhance resistance to S. sclerotiorum in B. napus X. Wang et al., 2020;Wang et al., 2020a;Zhang et al., 2021, Preprint); however, DNA mutations and alterations in the expression of these defense-related genes often negatively influenced plant growth and yield (Ning et al., 2017). Unfortunately, yield penalties caused by enhanced SSR resistance, to date have not received much consideration, despite some evidence provided in studies on A. thaliana, rice and N. benthamiana. To name a few, overexpression of constitutively active BnMKK4 DD induced hypersensitive cell death in N. benthamiana leaves (Zhang et al., 2021, Preprint); constitutive expression of active AtMPK3 resulted in a dwarf phenotype in A. thaliana (Genot et al., 2017); and overexpression of the A. thaliana gene AtNPR1 in rice led to height reduction and yield loss (Quilis et al., 2008). Therefore, strictly controlling defense-related gene expression is gaining more and more attention. In this study, our data demonstrated that pBnGH17 D7 is useful to precisely regulate the expression of defense-related genes in a S. sclerotiorum-inducible manner, while not incurring the negative effects inherent to the use of constitutive promoters.
Compared with pBnGH17, pBnGH17 D7 is a desirable S. sclerotiorum-inducible promoter for the genetic engineering of SSR-resistant crops. The activities of pBnGH17 were at low levels in the majority of plant tissues, but were relatively high in roots (Fig 2A, C). However, pBnGH17 D7 exhibited very low activity in all tissues under normal conditions (Fig. 3C-H), and thus may have minimal adverse impacts on B. napus growth and development. When pBnGH17 D7 was employed to engineer HIGS-based Sclerotinia-resistant B. napus, pBnGH17 D7triggered expression of high levels of specific siRNAs only occured after S. sclerotiorum infection ( Fig. 5B; Supplementary  Fig. S5), thereby preventing siRNA expression in all tissues and throughout all stages of plant development. These results demonstrated the potential utility of pBnGH17 D7 for minimizing the yield penalty associated with enhanced SSR resistance in B. napus.

BnTGA7 regulates the expression of BnGH17 in response to S. sclerotiorum infection
In this study, we identified TFs that can interact with pBnGH17 D7 with the high-throughput Arabidopsis TF screening system (Ou et al., 2011). We clarified that BnTGA7 directly binds to the TGACG motif of pBnGH17 and activates the expression of BnGH17 (Fig. 4). Moreover, BnTGA7 was significantly up-regulated after S. sclerotiorum infection ( Supplementary Fig. S4A, B), suggesting the potential role of BnTGA7 in SSR resistance. TGA family members have been implicated as important plant defense regulators possibly through physical interaction with the known master immune regulator NPR1 (Kesarwani et al., 2007). To date, seven of 10 A. thaliana TGA TFs (TGA1-TGA7) have demonstrated interplay with NPR1 (Després et al., 2000;Kesarwani et al., 2007;Shearer et al., 2009). TGA1 and TGA4 (group I) control basal resistance independent of NPR1, despite their physical interaction (Shearer et al., 2012); TGA2, TGA5, and TGA6 (group II) play crucial but redundant roles in systemic acquired resistance (Y. Kesarwani et al., 2007); and TGA3 and TGA7 (group III) are involved in basal resistance and the regulation of PR1 gene expression (Kesarwani et al., 2007;Chen et al., 2019). BnGH17, a member of the PR-2 protein group, encodes β-1,3-endoglucanase (Doxey et al., 2007). Our results suggested a role for BnTGA7 in activating BnGH17 expression in response to S. sclerotiorum infection in B. napus. These findings advanced our understanding of the TGA family in regulating PR-2 gene expression in the defense response. It was recently shown that overexpression of BnNPR1 enhanced resistance to S. sclerotiorum in B. napus (Wang et al., 2020a), suggesting that BnNPR1 plays a positive regulatory role in defense responses to S. sclerotiorum infection. Whether activation of BnGH17 by BnTGA7 depends on NPR1-mediated enhancement of DNA binding activity remains to be determined.
In summary, we established a new strategy for enhancing Sclerotinia resistance with minimal adverse effects in B. napus. The S. sclerotiorum-inducible promoter pBnGH17, which harbors the TGACG-motif between positions -1432 and -1427, is essential for the S. sclerotiorum response (Fig. 6). BnTGA7 directly binds to the TGACG motif in pBnGH17 and activates transcriptional expression of BnGH17 (Fig. 6). Furthermore, pBnGH17 D7 was successfully utilized to drive SSR resistance in B. napus via the HIGS technique. As S. sclerotiorum is a typical necrotrophic fungal pathogen with a wide range of hosts, this work will also offer reference for engineering SSR resistance into other important economic crops such as soybean, sunflower, and peanut.

Supplementary data
The following supplementary data are available at JXB online. Fig. S1. Tissue-specific expression pattern of BnaC01g04530D, BnaA05g24230D, BnaC01g36130D, and BnaCnng78710D. Fig. S2. The expression patterns of BnGH17 under various stress conditions. Fig. S3. GUS expression analysis in pBnGH17:GUS transgenic A. thaliana plants. Fig. S4. Expression analysis of BnTGA7 and BnGH17 in leaves of B. napus after inoculation with S. sclerotiorum. Fig. S5. Length distribution of target gene-specific siRNAs. Fig. S6. The expression levels of SsPG1 in S. sclerotioruminfected B. napus cotyledons and leaves. Fig. S7. Assessment of the disease resistance of HIGS transgenic B. napus (T 2 ) to S. sclerotiorum by the stem inoculation method in the field. Table S1. Primers used for qRT-PCR, gene cloning, and vector construction in this study. Table S2. Agronomic traits of HIGS transgenic lines in the T 2 generation.