Introduction

Dandelions, belonging to the Taraxacum genus, are widely distributed species used as food and traditional medicine for the treatment of various disease (hepatitis B, type 2 diabetes, and heart disease) (Martinez et al. 2015). Dandelions have been used as the raw materials in various products such as tea, syrup, coffee, and wine, which led to the rapid development of the dandelion industry (Lis and Olas 2019). Chlorogenic acid (CGA) and caffeic acid (CA) are the main phenolic compounds in dandelions which can scavenge harmful free radicals that are capable of attacking healthy body cells, causing them to lose their structure and function (Güneş et al. 2019; Jędrejek et al. 2017). However, concentrations of CGA and CA in Taraxacum antungense are lower than those in other plants such as Eucommia ulmoides and Lonicera japonica, restricting the industrialization of dandelion (Liu et al. 2019). Bioengineering strategies could potentially increase CGA concentrations in dandelions; however, a greater understanding of the CGA biosynthesis pathway is required.

Pathways for the CGA biosynthesis exist in many plants including Nicotiana tabacum, Solanum tuberosum, and Ocimum basilicum (Niggeweg et al. 2004; Yoshikawa et al. 2018). The common key enzymes in the first three steps of CGA biosynthesis are phenylalanine ammonia-lyase (PAL), 4-coumarate-CoA ligase (4CL) and cinnamate 4-hydroxylase (C4H), which are not only involved in CGA biosynthesis, but also participated in the biosynthesis of precursor compounds that catalyze other metabolic pathways such as anthocyanin, salvianolic acid, flavone and flavonol biosynthesis (Fatemeh et al. 2018; Gonzalez et al. 2008; Huang et al. 2019). Then, hydroxycinnamoyl-coenzyme (Co)A shikimate/quinate hydroxycinnamoyl transferase (HCT) and hydroxycinnamoyl-CoA quinate hydroxycinnamoyl transferase (HQT) both can directly catalyze CGA in vitro and in vivo (Niggeweg et al. 2004; Zhang et al. 2018a). However, only HQTs (TaHQT1 and TaHQT2) have been isolated, identified, and evaluated in T. antungense, the other key enzymes still need to be cloned and characterized (Liu et al. 2019) (Supplementary Figure S1).

Plant phenolic acids are involved in the regulation of plant disease resistance and are important plant components acting against biotic and abiotic stress (Gharibia et al. 2019). During the host defense reaction of perennial ryegrass to fungal pathogens, CGA and flavonoids increased significantly, and PAL expression levels also increased significantly (Rahman et al. 2015). Furthermore, maize PAL proteins contribute to resistance to sugarcane mosaic virus (SCMV) infection, most likely through positive regulation of salicylic acid (SA) accumulation with increased CGA concentration (Yuan et al. 2019). In T. antungense, powdery mildew, Sphaerotheca fusca (Fr.) Blum, has been found to decrease plant growth and development (Guo et al. 2019). Therefore, it is an important research topic to examine the relationship between powdery mildew infection and phenolic acid accumulation in T. antungense. Besides the isolation of CGA pathway genes, transcription factors (TFs) may play an important role in regulating plant secondary metabolism (Shi et al. 2020) and responding to powdery mildew infection.

WRKY is a large TF family in eukaryotes, especially in plants, such as Arabidopsis thaliana, Taraxacum kok-saghyz Rodin, Oryza sativa, Gossypium aridum, Salvia miltiorrhiza, Manihot esculenta and Ophiorrhiza pumila (Lin et al. 2017; Wang et al. 2014; Deng et al. 2019; Xu et al. 2020). The defining feature of WRKY TFs is their DNA-binding domain, known as WRKY domain, which contains approximately 60 amino acids including the conserved 'WRKYGQK'-motif at the N-terminus and an atypical zinc-finger motif, either C2H2 (CX4–5CX22–23HX1H) or C2HC (CX7CX23HX1C), at the C-terminus (Li et al. 2015; Deng et al. 2019). WRKY TFs are mainly related to plant stress resistance and are involved in the regulation of plant development and metabolism (Li et al. 2020; Yu et al. 2001). SA-sensitive WRKY TFs such as AtWRKY28/75, OsWRKY13, CsWRKY50, and FvWRKY42, enhance resistance to pathogens, showing that the SA pathway participates in plant disease resistance (Qiu et al. 2008; Yuan et al. 2019). WRKY TFs are involved in the regulation of secondary metabolism in medicinal plants, including in the accumulation of flavonoids, phenolic acids, and alkaloids (Cao et al. 2018). In Ophiorrhiza pumila, OpWRKY3 can increase camptothecin biosynthesis (Wang et al. 2019). AsWRKY44 represses expression of the wound-induced sesquiterpene biosynthetic gene ASS1 in Aquilaria sinensis (Sun et al. 2020). However, the function of WRKY TFs and their regulatory model has been never reported in T. antungense.

In this study, a novel full-length gene TaWRKY14 (accession number: MH513939.1) encoding for a WRKY TF from T. antungense was isolated and functionally characterized. TaWRKY14 was found to regulate the transcription of TaPAL1 by recognizing the W-box to promote CGA biosynthesis. Thus, the TaWRKY14-overexpressing transgenic line had a higher powdery mildew resistance than the control. These findings provide new insights regarding the TaWRKY14 regulatory network and provide an excellent candidate TF gene for the metabolic engineering of T. antungense.

Materials and methods

Plant material and reagents

Taraxacum antungense plants were collected and kept in our laboratory greenhouse as previously reported (Liu et al. 2018). Seeds of A. thaliana and T. antungense were sown in a soil mixture (matrix:vermiculite = 3:1) and cultivated for infiltration experiments in pots (4−5 weeks). Plants were maintained at 25 °C with a 16/8 h light/dark cycle. Tissue samples from fresh roots and leaves were collected from 3-month-old T. antungense seedlings, washed with distilled water, dried with filter paper, and immediately stored at − 80 °C for organ expression analysis. Primer synthesis and DNA sequencing were performed by GENWEIZHI Biotechnological Company, China. PowerUp™ SYBR Green Master Mix (Applied biosystems, Thermo fisher Scientific, United States) was used for quantitative reverse transcription (qRT)-PCR. Prime STARGXL DNA polymerase was purchased from Takara Company (Dalian, China) and restriction enzymes were purchased from Transgen Biotech Co., LTD (Beijing, China).

TaWRKY14 isolation and bioinformatics analysis

TransZol reagent (Transgen) was used for total RNA extraction, and the final RNA concentrations were determined using a NanoQuant plate™. Then, RNA integrity was measured by electrophoresis on 0.8% (w/v) agarose gel. cDNA was synthesized using an EasyScript® First-Strand cDNA Synthesis SuperMix Kit (Transgen). The coding DNA sequence of TaWRKY14 was amplified using gene-specific primers (from the 5′- and 3′- ends of the open reading frames, named TaWRKYF and TaWRKYR; Supplementary Table S1). PrimeSTAR® GXL DNA polymerase (Takara) was used in all PCRs, and PCR errors were identified by sequencing multiple clones derived from independent PCR reactions. Amplified cDNA fragments were cloned and sequenced. A total of 50 ng cDNA and 10 pmol of each primer were used for PCR, using a cycling program as follows: 94 °C for 5 min; 38 cycles of 94 °C for 30 s, 55 °C for 35 s, and 72 °C for 1 min 5 s; then 72 °C for 8 min. The PCR product was purified and connected to a vector (pMD19-T) that was used to transform E. coli DH5α for sequencing. Double enzyme sites (Nco I and Spe I) were added by amplifying the full-length TaWRKY14 gene using PCR.

The ortholog was searched using blast alignment for sequence analysis (https://www.ncbi.nlm.nih.gov/BLAST/). ClustalX (version 1.8) and the MEGA program (version 8.0) were used for multiple sequence alignment and molecular phylogenetic tree construction (neighbor-joining method).

Elicitor preparation and treatment

To study the effect of various phytohormones on TaWRKY14 expression, T. antungense leaves were treated with MeJA, NaCl, SA, drought conditions, and powdery mildew for different time intervals, and expression analysis was performed using qRT-PCR.

Methyl jasmonate (MeJA) was first dissolved in a small volume of ethanol and then in distilled water to a storage concentration of 100 mM. SA and NaCl (Aladdin) were directly dissolved in distilled water to a storage concentration of 100 and 500 mM, respectively, and an equivalent volume of sterilized water was used as a control. Well-grown T. antungense plants were subjected to the different treatments. All the elicitors were sterilized through 0.22-μm syringe filters (Pall Corporation) and then added to cultures at a final concentration of 100 μM. From the different treatment groups (MeJA, SA, and NaCl), different plant tissues were sampled after 0, 3, 6, 9, 12, 24, and 48 h for the subsequent study. For drought treatment, watering was ceased, and samples were harvested separately at 0, 2, 4, 6, 8, and 10 days after the onset of drought treatment (No significant difference in soil moisture content after 8 days as been showed in Figure S2). For powdery mildew treatment, S. fusca kept in our laboratory was maintained on seedlings of T. antungense. Six-week-old plant seedlings were inoculated with fresh spores of S. fusca. Then, tissue from different plant parts was harvested at different time intervals after inoculation for gene expression analysis, and those of non-inoculated plants were harvested as a control (Zhou et al. 2018).

Subcellular localization of TaWRKY14

To identify the subcellular location of TaWRKY14 in vivo, we fused the coding sequences of TaWRKY14 to the GFP reporter gene driven by the CaMV 35S promoter. The fused expression vector was used to perform a transient expression assay in the protoplast of A. thaliana.

The full-length sequences of the TaWRKY14 gene (without the stop coding ‘TAG’) was amplified and recombined into pMD19-T with one pair of restriction sites (Nde I and BamH I). The amplicons were cloned into pRI101-GFP producing pRI101-TaWRKY14-GFP. The expression constructs pRI101-TaWRKY14-GFP and pRI101-GFP vectors alone (control) were used for individually transforming into the protoplast of A. thaliana. The procedures were performed as previously described (Zhang et al. 2018a). All images were acquired and processed using a Leica TCS SP8 confocal microscope and software (Leica Microsystems GmbH, Wetzlar, Germany).

Creation of transgenic T. antungense plants

We constructed the overexpression vector pRI101-TaWRKY14 and transformed this vector into T. antungense as previously reported (Liu et al. 2019; Fig. 4a; Figure S3). Twenty-three independent transgenic lines were identified by genomic PCR using the forward primer of the cauliflower mosaic virus CaMV 35S promoter and the reverse primer of TaWRKY14 (TaWRKY14R) (Table S1; Fig S4).

Gene expression profiling

Total RNA was extracted, and cDNA synthesis was performed following the above-mentioned methods from different tissues of T. antungense, including wild-type (WT) and transgenic lines. The gene-specific primers designed for TaWRKY14, PAL1, C4H, 4CL, HCT, and HQT are listed in Table S1 and compared with Taraxacum genomic data (Lin et al. 2017). qRT-PCR was performed following the instructions of SYBR® PreMix Ex Taq (Takara Bio Inc., Dalian, China) with StepOne™. The following qRT-PCR cycling program was used: 94 °C for 10 min; and 40 cycles of 94 °C for 15 s, 58 °C for 30 s, and 72 °C for 30 s. Relative expression levels were calculated based on the 2−DDCt method with β-actin used as a reference gene. The primer sequences used for the expression analysis of various biosynthetic genes are listed in Table S1.

Quantification of CGA, CA, rutin, and luteolin with high-performance liquid chromatography (HPLC)

HPLC was used to investigate the CGA, CA, rutin, and luteolin concentrations in the materials, including the WT and transgenic lines of T. antungense. All samples were extracted using the ultrasound-assisted method and passed through a 0.22 μm syringe filter. CGA, CA, rutin, and luteolin standards were identified using HPLC (Fig. S5, Table S2) with an Agilent Eclipse XDB-C18 column (pinnacle, 150 × 4.6 mm; inner diameter, 5 μm; Restek, Bellefonte, PA, USA) and performed on an Agilent 1260 with DAD detector (Agilent Technologies, Palo Alto, CA, USA). The sample injection volume was 20 μL, and the column temperature was maintained at 30 °C. The binary elution solvent consisted of A (Methanol) and B (0.1% CH2O2), and a gradient elution procedure was used as follows: 0−10 min (10% A), 10−30 min (32–32% A), 30−38 min (32−42% A), 38−52 min (42−65% A), 52−57 min (65% A), and 57−60 min (65−10% A). The flow rate was maintained at 1.0 mL/min, and the absorbance was measured at a wavelength of 320 nm for CGA and CA detection, and 254 nm for rutin and luteolin detection. SA concentration was determined using HPLC as previously described (Dewdney et al. 2000).

Yeast one-hybrid assay

New primers were synthesized to amplify the 5′ ends of the TaPAL1 promoter (> GWHAAAA00002974, OriSeqID = utg5309). W-boxes (− 938 to − 114 bp) were also found in the TaPAL1 gene promoter (Figure S6). We performed a yeast one-hybrid assay following Kang et al. (2014) to confirm the interaction between TaWRKY14 and the W-boxes. The TaPAL1 promoter sequence (from − 1250 to − 1456 bp relative to translation start site, named proTaPAL1) was inserted into the pAbAi vector, and BstBI was used for digesting the recombinant plasmids. The linearized recombinant plasmids were transformed into the yeast strain Y1H and tested for Aureobasidin A (AbA) concentrations (100−500 ng/mL) on SD/-Ura medium. The coding sequence of full-length TaWRKY14 was fused into the pGADT7 vector and transferred into the competent cells (prepared by recombinant plasmids of pAbAi-proTaPAL1 transformed into Y1H), and growth reached 2 mm. The transformants were cultivated on SD/-Leu medium with AbA (different concentrations) for 3−5 days.

Field experiment

Sphaerotheca fusca was used for evaluating the pathogen resisting ability in WT and overexpression transgenic lines using detached leaves at the seedling stage following Zhou et al. (2018). Six-week-old seedlings were transferred to pots and moved to a greenhouse for 10 days (28 °C, 70% RH). Powdery mildew was inoculated using the spore suspension spray method, and a hemocytometer was used to calculate the spore suspension concentration. Freshly prepared spore suspensions (10 mL, 106 conidia mL−1) were sprayed evenly on the young leaves. As a control, healthy young leaves were sprayed with sterile water and moisturized for 48 h. Leaves were collected at 0, 2, 4, 6, 8, and 10 d after inoculation and used for disease index calculation. Leaf samples were frozen in liquid nitrogen and stored at − 80 ℃ prior to RNA extraction and then used for qRT-PCR. Disease index was calculated using the following equation:

$${\text{Disease index }} = {{\left( {{\text{number of diseased leaves }} \times {\text{ representative value of disease level}}} \right)} \mathord{\left/ {\vphantom {{\left( {{\text{number of diseased leaves }} \times {\text{ representative value of disease level}}} \right)} {\left( {{\text{sum of the number of investigated leaves }} \times {\text{ the highest severity of the disease representative value}}} \right) \times 100\% }}} \right. \kern-\nulldelimiterspace} {\left( {{\text{sum of the number of investigated leaves }} \times {\text{ the highest severity of the disease representative value}}} \right) \times 100\% }}$$

The representative value of disease level (RVDL) of leaves affected by S. fusca were scored as grades 0–4, i.e., 0 for the highest resistance without visible symptoms, 1 for highly resistant, 2 for moderately resistant, 3 for moderately susceptible, and 4 for highly susceptible with vast stretches of hyphae producing large amounts of conidiophore.

Measurement of physiological parameters and antioxidant enzyme activity

T. antungense plants (WT or transgenic lines; 3-days inoculation with powdery mildew or sterile water) were used for detection by malondialdehyde (MDA), proline, and chlorophyll concentrations (Karagöezler et al. 2008; Liu et al. 2018). PAL, superoxide dismutase (SOD), and peroxidase (POD) activities were assayed according to the colorimetric method (Gao et al. 2017), nitrogen blue tetrazole photoreduction method, and guaiacol method (Tian et al. 2019), respectively.

Statistical analysis

One-way analyses of variance were carried out using SPSS 22.0 (SPSS, Inc., Chicago, IL, USA) for all data sets. The values represent the means of three biological replicate plants for CGA, CA, rutin, and luteolin concentrations in the field experiment, and of three technical replicates for qRT-PCR analysis. RVDLs of the transgenic plants were deduced from three independent experiments. Mean values were compared using a least-significant difference test, and differences were considered significant at P < 0.05.

Results

Isolation and sequence analysis of the TaWRKY14 gene

The full-length cDNA encoding TaWRKY14 was 1059 bp, including the 5′- and 3′-untranslated regions, and encoded a 352 amino acid protein (Genbank accession number: MH513939.1). Multiple sequence alignment identified a single conserved 'WRKYGQK' domain in TaWRKY14 (Fig. 1b; pfam03106 or smart00774) with the online tool Pfam (https://pfam.xfam.org/search/sequence). The best matches obtained from a local alignment search within a non-redundant protein database (blastp) were with S. miltiorrhiza SmWRKY40 (98.5% identity, 98% positivity), which has not been explored before. TaWRKY14 also matched other WRKY14s with more than 50% identity (Nicotiana attenuata, Dendrobium catenatum, Arabidopsis lyrata subsp. Lyrata, and Apostasia shenzhenica) (Fig. 1a, b).

Fig. 1
figure 1

Multiple alignment, phylogenetic analysis and organ expression analysis of TaWRKY14. a Phylogenetic analysis of TaWRKY14 with related WRKY transcription factors. The phylogenetic tree was constructed using MEGA software (version 8.0) based on the neighbor-joining method. Values above the branches are bootstrap percentages (1000 replicates). The WRKY14 protein as labeled after taxon marker. b Sequence alignment of TaWRKY14 sequences in T. antungense. Multiple sequence alignment of TaWRKY14 (in this article, MH513939.1), AlWRKY14 (EFH67144.1), SmWRKY40 (AKA27880.1), AtWRKY14 (NP564359.1), AtWRKY35 (AEC09027.1), NaWRKY14 (OIS96845.1), AsWRKY14 (PKA55748.1), and DcWRKY14 (PKU72194.1). The conserved domain pfam03106 (WRKYGQK) is indicated by a red square. (Color figure online)

Full size image

TaWRKY14 expression patterns and subcellular localization

Organ expression analysis was performed to assess the expression pattern of TaWRKY14. It revealed significant differences with the highest expression in root followed by that of leave, and low expression in flower and stem (Fig. 2a). T. antungense leaves were treated with 100 μM MeJA for different time intervals (1, 3, 6, 9, 12, and 24 h), and no significant effect was observed on TaWRKY14 expression levels (Fig. 2b). Contrastingly, under NaCl (500 ng/mL) treatment, TaWRKY14 expression level was gradually increased from 1 to 9 h (3.5-fold) and then it slowly declined until 24 h (1.5-fold). Under SA treatment, TaWRKY14 expression was increased around 3-fold after 6 h and then declined. A similar trend was noticed in the presence of powdery mildew, with TaWRKY14 increasing more than 3-fold in 6 h. Finally, drought treatment was performed for 10 days, with TaWRKY14 expression level gradually increasing up to 6-fold after 6 days of treatment and then slowly decreasing on the tenth day (Fig. 2c).

Fig. 2
figure 2

Tissue expression, response to different treatments and subcellular localization patterns of TaWRKY14. a Organ expression of TaWRKY14 gene in wild T. antungense (leave, root, flower and stem, respectively). Different letters indicate significant differences in comparisons (P < 0.05). b Different treatment including NaCl (500 μM), salicylic acid (SA;100 μM), methyl jasmonate (MeJA;100 μM) and powdery mildew inoculation (for 1 h, 3 h, 6 h, 9 h, 12 h, 16 h, and 24 h). c Drought stress (0d, 2d, 4d, 6d, 8d, and 10d). d Arabidopsis protoplast was transiently transformed with constructs containing TaWRKY14-GFP, or GFP (control) under the control of the 35S promoter

Full size image

GFP fluorescence of the control distributed evenly throughout the cell, while GFP fluorescence of TaWRKY14-GFP was observed only in the nucleus (Fig. 2d).

Role of TaWRKY14 in the biosynthesis of CGA, CA, rutin and luteolin

The results of qRT-PCR analysis showed that the expression levels of TaWRKY14 genes in the TaWRKY14-transformed lines were significantly higher than those in WT (Fig. 3a  and 3b). Three lines (TaWRKY14-OE3, TaWRKY14-OE7, and TaWRKY14-OE19) with higher expression level than that of the control lines were selected to analyze CGA, CA, rutin and luteolin concentrations by HPLC. CGA concentration of the three transgenic lines was higher than that in the control (0.72 ± 0.054 mg/g FW) and it was maximum in TaWRKY14-OE7 (1.14 ± 0.141 mg/g FW). However, there were no significant differences in CA, rutin, and luteolin concentrations between WT and TaWRKY14 transgenic lines (Fig. 3c). To identify the biosynthetic genes of CGA regulated by TaWRKY14, the expression of genes encoding key enzymes in the CGA biosynthetic pathway of T. antungense (PAL, C4H, 4CL, HCT, and HQT) were investigated through qRT-PCR. Results indicated that TaPAL1 was up-regulated to various degrees in the TaWRKY14-overexpression lines (Fig. 3d).

Fig. 3
figure 3

TaWRKY14-overexpression transgenic lines generated by Agrobacterium tumefaciens-mediated transformation, identification and analysis by HPLC. a (1) Preculture (48 h); (2) callus induction for treated with recombinant Agrobacterium GV3101 carrying the WRKY14-overexpressing construct; (3) subculture and shoots induction; (4) well-rooted explants of transgenic lines. b qRT-PCR expression of TaWRKY14 in the empty vector and three independent TaWRKY14 over-expressing transgenic lines. c Determination of CGA, CA, rutin and luteolin concentration by high performance liquid chromatography (HPLC) in T. antungense transgenic lines. FW: fresh weight. d Expression of the key enzymes in the CGA biosynthetic pathway in control and TaWRKY14 overexpression transgenic lines. Error bars represent the standard deviations of three replicates. Statistical significance was assessed with Student’s t test (*: P < 0.05)

Full size image

Activation of the TaPAL1 promoter by TaWRKY14

The results obtained for TaWRKY14 on the PlantTFDB and NCBI website indicated it could bind the W-box sequence ('TTGACC/T'-motif). By analyzing the sequence of the TaPAL1 promoter (Fig S4), we identified four W-boxes (proPAL1-W-box) from − 938 to − 114 bp of the ATG codon (Fig. 4a). As TaPAL1 is related to phenolic acid biosynthesis, the TaPAL1 gene promoter was selected to carry out the yeast one-hybrid assay.

Fig. 4
figure 4

Interaction between TaWRKY14 and W-box. pAbAi-W-box was grown normally on 300 ng/mL Aureobasidin A (AbA), but pAbAi-pMutant-W-box could not grow on 300 ng/mL AbA. 400 ng/mL AbA is the minimum concentration of AbA to verify the interaction between TaWRKY14 and the W-boxes. We used the pAbAi-W-box and pAbAi-pMutant-W-box contained Y1H gold yeast preparation of competent cell and transformed pGADT7-TaWRKY14 plasmid, the co-transformed pAbAi-W-box + pGADT7-TaWRKY14 was able to grow on 400–500 ng/mL AbA, but pAbAi-pMutant-W-box + pGADT7-TaWRKY14 was unable. The difference between pAbAi-W-box and pAbAi-pMutant-W-box was that the W-box contained three ‘TTGACT’ and one ‘TTGACC’, boxes; while pAbAi-pMutant-W-box was using ‘AAAAAA’ sequence instead of ‘TTGACC’-box

Full size image

Y1H yeast cells transformed with pAbAi-proPAL1 recombinant vector were unable to grow on SD/-Ura medium (300 ng/mL AbAi). After transforming the pGADT7-TaWRKY14 vector into Y1H yeast cells (pAbAi-proPAL1 identified by PCR), they could grow on 400 ng/mL AbAi on SD/-Leu medium, which indicated that TaWRKY14 could bind the proPAL1 'TTGACC/T'-motif (Fig. 4b).

Effect of powdery mildew on gene expression and chemical composition of TaWRKY14 transgenic T. antungense

SA concentrations were not significantly different between TaWRKY14-overexpressing transgenic lines and WT before the treatment. Analysis of TaWRKY14-overexpressing transgenic lines treated with S. fusca for 12, 24, and 72 h showed that SA and CGA concentrations reached their maximum in 12 h (0.851 ± 0.0366 and 2.08 ± 0.191 mg/g FW, respectively, in TaWRKY14 transgenic T. antungense), which were significantly higher than those in the WT group. Then, concentrations decreased but were still higher than those in the 0 h treatment (Fig. 5a, b). Both the TaWRKY14 and TaPAL1 expression levels in TaWRKY14-overexpression T. antungense were higher than those in the WT group, and showed higher expression level under S. fusca treatment. Moreover, TaPAL1 gene expression level showed the same trend as that of TaWRKY14 (Fig. 5c, d). Thus, we selected TaWRKY14-OE7 lines for further study.

Fig. 5
figure 5

Possible model of TaWRKY14 (OE-3, OE-7 and OE-19) increasing CGA accumulation and mechanism of responding to powdery mildew in T. antungense. a, b Concentrations of SA and CGA in T. antungense plants treated with Powdery mildew pathogen (S. fusca (Fr.) Blum) were measured at 12, 24, and 72 h post-inoculation, respectively. c, d Expression levels of protein genes TaWRKY14 and TaPAL1 were measured by qRT-PCR, respectively

Full size image

TaWRKY14 improved the powdery mildew resistance of T. antungense

The powdery mildew disease index of T. antungense increased with treatment time. The disease index of transgenic and WT lines inoculated with S. fusca showed that TaWRKY14-overexpressing transgenic lines increased powdery mildew resistance (Fig. 6 and Table 1). On the second day, powdery mildew was found in the WT line, while it was found on the sixth day in the TaWRKY14-overexpression transgenic line. Thus, TaWRKY14 significantly improved the powdery mildew resistance.

Fig. 6
figure 6

Disease index of TaWRKY14-transformed (OE-7) and WT T. antungense plants under S. fusca (Fr.) Blum inoculation for 0, 2, 4, 6, 8, and 10 days

Full size image
Table 1 Disease index of TaWRKY14-transformed (OE-7) and WT T. antungense plants under S. fusca (Fr.) Blum inoculation for 1, 2, 4, 6, 8, and 10 days
Full size table

MDA, proline, and total chlorophyll concentrations in transgenic and WT lines

MDA concentrations in the leaves were different before and after 3-day S. fusca inoculation. In the absence of S. fusca, the WT and OE-7 transgenic lines did not show significant differences in MDA concentrations. However, after treatment with S. fusca, MDA concentrations in the WT and OE-7 lines increased by 297.9% and 77.0%, respectively (Table 2). The increase of MDA concentration in the TaWRKY14-overexpression lines was significantly lower than that in WT. Proline concentrations in all cultivars after exposure to S. fusca sharply increased by 396.6% and 67.0% in the WT and OE-7 lines, respectively. Additionally, total chlorophyll concentration in the OE-7 lines was significantly higher than that in WT after S. fusca inoculation. These results suggested that the antioxidant capacity in TaWRKY14-overexpressing transgenic lines after S. fusca inoculation was higher than that in the WT line.

Table 2 Comparison of malondialdehyde (MDA) concentration, proline concentration, and total chlorophyll concentration among TaWRKY14-transformed (OE-7) and WT T. antungense plants under S. fusca (Fr.) Blum inoculation for 3 days
Full size table

PAL, SOD, and POD enzyme activity in transgenic and WT lines

After treatment with S. fusca, PAL enzyme activity increased by 93.71% and 34.40% (Table 3) and SOD enzyme activity increased by 22.86% and 51.90% in the WT and OE-7 lines, respectively. Moreover, a similar tendency was found in POD enzyme activity. These results suggested that after S. fusca inoculation, the enzyme activity in TaWRKY14-overexpressing transgenic lines was higher than that in the WT line.

Table 3 Comparison of PAL, SOD, and POD enzyme activity among TaWRKY14-transformed (OE-7) and WT T. autungense plants under S. fusca (Fr.) Blum inoculation for 3 days
Full size table

Discussion

TaWRKY14 increased CGA biosynthesis in T. antungense

TaWRKY14 TF in T. antungense was isolated and compared with WRKY14 from different plants (such as Hevea brasiliensis and S. miltiorrhiza), with more than 50% homology (IIe family). Both TaWRKY14 and SmWRKY40 have common features such as being SA-sensitive and MeJA-insensitive, which is consistent with earlier findings. Additionally, SmWRKY40 is closer to AtWRKY14 and AtWRKY35 in putative A. thaliana orthologs (Li et al. 2015). Subcellular localization showed that TaWRKY14 localizes to the nucleus similar to AtWRKY14 and other WRKY14 members (Chi et al. 2013). Furthermore, drought stress can increase TaWRKY14 expression level, having the same tendency as that of VaWRKY14 (Zhang et al. 2018b). However, in Capsicum annuum L, CaWRKY14 decreases 0.3-fold under drought stress, which may be explained by CaWRKY14 belonging to the IId group, while TaWRKY14 belongs to the IIe group in the WRKY family (Sziderics et al. 2010). Therefore, WRKY14 TFs in different species may have different functions based on the evolutionary trends in environmental adaptation. In this article, TaWRKY14 plays an important role in SA, NaCl, powdery mildew, and drought signaling pathways, ultimately helping plants to cope with the various types of abiotic and biotic stresses.

Previous studies have reported that multiple WRKY TFs, such as AtWRKY12, AtWRKY23, and AtWRKY44, are involved in the biosynthesis of phenylpropanoid in A. thaliana (Grunewald et al. 2012). In rice leaves, OsWRKY14 appeared to play a key regulatory role in MeOH-induced Trp and Trp-derived secondary metabolite biosynthesis and bind to TDC1 promoters through W-box (Kang et al. 2014). Moreover, polyterpene rubber synthesis can be regulated by HbWRKY14 (Li et al. 2011). We found that TaWRKY14 gene expression level in different tissue has the same tendency with the main bioactive components (CGA, CA, rutin and luteolin) especially for CGA, suggesting that TaWRKY14 may be involved in the biosynthesis of these compounds. We have obtained TaWRKY14-overexpressing transgenic lines to clarify the functions of TaWRKY14 in T. antungense. The results showed that the concentrations of CA, rutin and luteolin in the WT and TaWRKY14-overexpressing transgenic lines were not significantly difference. This indicates that WRKY14 TF may have no direct correlation with the biosynthesis of CA, rutin and luteolin in T. antungense. Meanwhile, we found that overexpressing TaWRKY14 significantly increased the CGA concentration by 1.58 fold, with the maximum in the TaWRKY14-OE7 (1.14 ± 0.141 mg/g FW) line compared to the control (0.72 ± 0.054 mg/g FW). These results indicate that TaWRKY14 plays a positive role in regulating CGA biosynthesis.

We performed qRT-PCR to analyze the expression levels of genes encoding key enzymes in the CGA biosynthetic pathway in transgenic lines. Overexpressing TaWRKY14 promoted the up-regulation of TaPAL1 expression level, suggesting that TaPAL1 might be a target gene of TaWRKY14. Indeed, TTGACC/T-binding related elements were identified in the TaPAL1 promoter. Y1H assay revealed that TaWRKY14 directly binds to W-box elements in the TaPAL1 promoter region. Together, these results demonstrate that TaPAL1 is a target of TaWRKY14.

In phenylalanine biosynthesis, PAL is the first key enzyme which can be regulated by different family members of TFs (Khakdan et al. 2018). SmMYB98 increases salvianolic acid production by activating the expression of PAL1 and RAS1 (Hao et al. 2020). AaWRKY1 responds to UV-B-induced flavonoid accumulation in Artemisia annua through the PAL protein (Pandey et al. 2019). Overexpression of AtPAP1 increased anthocyanin accumulation by activating the expression of AtPAL (Chhon et al. 2020). In the present study, we demonstrated that TaWRKY14 binds to the W-box elements in the TaPAL1 gene promoter, thereby activating its expression and promoting CGA accumulation. Thus, the functions of the multiple PAL genes in Taraxacum should be further studied (Kim and Hwang 2014; Shine et al. 2016).

TaWRKY14 enhances powdery mildew resistance in T. antungense and biological evaluation of TaWRKY14-overexpression transgenic plants

Phylogenetic analysis indicated that TaWRKY14 belongs to the IIe subfamily of the WRKY TF family. WRKY IIe subfamily members such as AtWRKY22/29, are activated by the MAP kinase signaling cascade and confer resistance to both bacterial and fungal pathogens (Liu et al. 2005). AtWRKY14 has been confirmed to be involved in plant growth, leaf senescence, drought stress, and response to SA hormones (Kagale et al. 2010; Noman et al. 2019). In the current study, we found that S. fusca infection can increase TaWRKY14 expression level, and TaWRKY14-overexpressing transgenic lines had higher powdery mildew resistance.

Previous studies reported that PAL is a key enzyme in response to pathogen infection (Kim and Hwang 2014). Increased PAL gene expression level may be related with SA accumulation, as reported in Arabidopsis, tobacco, pepper, and soybean (Shine et al. 2016; Yuan et al. 2019). We found that CGA and SA concentration, and TaPAL1 and TaWRKY14 gene expression level in TaWRKY14-overexpressing transgenic lines treated with S. fusca inoculation were significantly higher than those in the WT group. A possible reason for this may be that pathogen infection led to the enhancement of the SA signaling pathway, causing TaWRKY14 to increase the expression of the PAL gene, thereby amplifying the SA signal and increase SA and CGA accumulation. As SA-responsive pathogenesis-related protein genes (PRs) have been found in plants (Tian et al. 2019), further research on the interaction between TaWRKY14 and PRs in T. antungense would be of interest.

Plants have evolved a variety of adaptive mechanisms to respond to powdery mildew (Zhou et al. 2018), including antioxidant enzyme activities and non-enzymatic antioxidant defense systems. TaWRKY14-overexpressing transgenic lines had a higher antioxidant capacity under the powdery mildew treatment (represented by MDA, proline, and total chlorophyll), which might help plants maintain membrane permeability and photosynthetic capacity. This is consistent with Liu et al. (2018), who reported that CGA is related with plant responses to stress.

We also detected the PAL, SOD, and POD enzyme activities in different transgenic lines which are important physiological indicators of powdery mildew resistance (Jing et al. 2020; Patel et al. 2020). The higher PAL, SOD, and POD enzyme activities in TaWRKY14-overexpressing transgenic lines showed higher powdery mildew resistance. However, the mechanisms of TaWRKY14 increasing SOD and POD enzyme activities through CGA and other phenolic acid compounds remain to be further studied. Moreover, TaWRKY14-overexpression lines can be used as a new resource for further quality evaluation and for industrial crop cultivation.

Conclusion

TaWRKY14 TF can increase CGA biosynthesis by regulating the expression of TaPAL1 in T. antungense, improve powdery mildew resistance, and promote the expression of antioxidant enzymes such as SOD and POD. The expression of TaPAL1 in the phenylalanine metabolic pathway can respond to powdery mildew stress, thereby increasing the synthesis of endogenous phenolic compounds in plants, which can help plants resist powdery mildew.