Functional characterization of Lilium lancifolium cold-responsive Zinc Finger Homeodomain (ZFHD) gene in abscisic acid and osmotic stress tolerance

Plant materials

The bulbs of wild tiger lily (Lilium lancifolium) with a size of 3–5 cm collected from Heilongjiang province, China, were used in this study. The bulbs were washed with tap water, sterilized with carbendazim, air-dried, covered with a 1-cm layer of moist vermiculite, and stored at 4 °C for 40 days. After vernalization, the bulbs were cultured in the science and technology experimental greenhouse of Beijing Forestry University (116.3° E, 40.0° N).

Arabidopsis thaliana ecotype Columbia-0 (Col-0) was used for the ectopic expression study; tobacco Nicotiana benthamiana (NT78) was used for the subcellular localization study. The seeds of tobacco and Arabidopsis were sown in plastic containers filled with the sterile substrate (peat:vermiculite = 1:1), and cultured in a growth chamber with a light/dark regime 16/8 h (100 µmol·m⁻²·s⁻¹ ), at 22/16 °C, and 65% relative humidity. The containers were covered with plastic wrap under growth chamber condition until seed germination.

Gene cloning and sequence analysis

RNAisomate RNA Easyspin isolation system (Aidlab Biotech, China) was used for total RNA extraction. The RNA quantity was examined by NanoDrop 2000 (Thermo Scientific, Waltham, MA, USA). The First-strand synthesis of cDNA was produced using PC54-TRUEscript RT kit (+gDNA Erase) (Aidlab Biotech, China) using 1 µg total RNA.

The complete cDNA sequence of LlZFHD4 was obtained from tiger lily’s RNA-seq data published before (Yong et al., 2018; Yong, Zhang & Lyu, 2019b). Open reading frame (ORF) of LlZFHD4 was amplified using PC09-2×Taq PCR MasterMix (Aidlab Biotech, China) with the forward primer 5′-ATGGATCTCCCCATTTATC-3′ and the reverse primer 5′-TCAGTAGTCACTCTGCAATG-3′. The PCR reaction system and procedure were performed according to the manufacturer’s protocol. The PCR products were ligated into the pEASYT1-Blunt cloning vector (TransGen Biotech, Beijing, China) and sequenced by the Beijing TransGen Biotech Company.

Amino acid sequence alignment analyses were performed by BLASTN in the National Center for Biotechnology Information (NCBI) and DNAMAN (version 7). The unrooted phylogenetic trees were constructed in MEGA5 software using the neighbor-joining method with 1,000 replications. The online database ExPASy (http://expasy.org/tools/protparam.html) was used for the theoretical molecular weight and isoelectric point calculation.

Promoter cloning and sequence analysis

DNeasy Plant Mini Kit (Qiagen, Valencia, CA, USA) and Genome Walker Kit (Clontech, CA, USA) were used for genomic DNA extraction from mature leaves of tiger lily and LlZFHD4 promoter cloning, respectively (Yong, Zhang & Lyu, 2019a, 2019b). The promoter region of LlZFHD4 was amplified by nested PCR methods according to the user manual of Genome Walker Kit (Clontech, CA, USA) with the primary PCR primer GSP1: GGGTGGCTGCTAGGGTCAAACAGTAAT and the secondary PCR primer GSP2: AAGGTGGTTACAATGGGAAGAAGATG. Using touchdown PCR method, the primary PCR reaction procedure was as follows: pre-denaturation at 94 °C for 1 min; seven cycles of denaturation at 94 °C for 25 s, 65–59 °C for 30 s (decrease 1 °C per cycle), 72 °C for 3 min; 32 cycles of denaturation at 94 °C for 25 s, 57 °C for 30 s, 72 °C for 30 s; and extension at 72 °C for 10 min. The 1/50 dilution of the primary PCR product was used as a template in the secondary PCR. The reaction procedure was as follows: pre-denaturation at 94 °C for 5 min, 35 cycles of denaturation at 94 °C for 40 s, 58 °C for 40 s, 72 °C for 3 min, and extension at 72 °C for 30 s. The prediction of conserved cis-element motifs was performed on PLACE (http://www.dna.affrc.go.jp/PLACE/signalscan.html) databases.

Quantitative real-time PCR (qRT-PCR) analysis

Primer Premier 5.0 was used for primer design: primer annealing temperature 50–65 °C, primer length 18–24 bp, and amplification length 80–250 bp. The primers of LlZFHD4 and AtRD29A, AtRD29B, AtRD20, AtLEA14, AtGolS1, AtAPX2 from Arabidopsis were listed in Table S1. The SYBR^® qPCR mix (Takara, Dalian, China) and Bio-Rad/CFX Connect^™ Real-Time PCR Detection System (Bio-Rad, Irvine, CA, USA) were used in the qRT-PCR experiment. The reaction procedure was as follows: pre-denaturation at 95 °C for 30 s, 40 cycles of denaturation at 95 °C for 5 s, annealing temperature for 30 s, and extension at 72 °C for 30 s. Relative mRNA content was calculated using the 2^−ΔΔCt method. Each sample was amplified in biological and technical triplicate. Tiger lily LlTIP1 (Wang et al., 2014) and Arabidopsis Atactin (NM_112764) were used as internal reference genes (Table S1).

Subcellular distribution of LlZFHD4

The ORF of LlZFHD4 without stop codon was amplified with the forward primer 5′-CATTTACGAACGATACTCGAG(XhoI)ATGGATCTCCCCATTTATC-3′ and the reverse primer 5′-CACCATCACTAGTACGTCGAC(SalI)GTAGTCACTCTGCAATG-3′. The amplified product was inserted into the XhoI and SalI site of the pBI121-GFP vector driven by a CaMV 35S promoter according to the ClonExpress II kit’s user manual (Vazyme, Nanjing, China). The confirmed recombinant vector pBI121-LlZFHD4-GFP was transformed into Agrobacterium strain GV3101 competent cells by the freeze-thaw method; and introduced in the tobacco leaf epidermal cells by the infiltration method with GV3101 cells. After 32 h incubation, the GFP fluorescence signals were detected in agroinfiltrated tobacco leaves using Leica TCS SP8 Confocal Laser Scanning Platform (Yong, Zhang & Lyu, 2019a, 2019b).

Yeast transcription activation assay of LlZFHD4

Using a ClonExpress II kit, the entire coding region (1–540 bp) and N-terminus (1–270 bp) and C-terminus (271–540 bp) of the LlZFHD4 cDNA region were amplified (primers are shown in Table S1), and cloned into the EcoRI and BamHI site of pGBKT7 vector MCS region (Clontech, CA, USA), resulting in recombinant vector pGBKT7-LlZFHD4-A (1–180 aa), pGBKT7-LlZFHD4-N (1–90 aa) and pGBKT7-LlZFHD4-C (91–180 aa). The confirmed recombinant vectors were introduced into the Y2HGold yeast strain (Huayueyang, Beijing, China) according to our previous study (Yong, Zhang & Lyu, 2019a, 2019b). The 1/10 and 1/100 dilution transformed yeast cells were incubated on plates containing the appropriate SD selection medium (Cao et al., 2017) at 30 °C. Yeast colony formation was photographed after three days.

Generation of LlZFHD4 transgenic Arabidopsis

The entire coding region of LlZFHD4 was inserted into the pBI121 vector driven by CaMV 35S for ectopic expression study. The CaMV35S promoter region in the pBI121-GUS vector was replaced by the LlZFHD4 promoter region for promoter activity analysis. After being mobilized to Agrobacterium strain GV3101, the recombinant vectors were transformed into wide-type (WT) Arabidopsis with GV3101 cells through the floral dip method. The MS medium containing 50 mg/L kanamycin was used to screen the progeny transgenic plant seeds. T3-generation of LlZFHD4 transgenic lines were harvested after qRT-PCR confirmation; T2-generation of LlZFHD4 promoter transgenic lines were collected after PCR confirmation with promoter-specific primers: forward primer 5′-GCTTGATATCGAATTCG-3′ and reverse primer 5′-TCTCAATTTAGGATCCT-3′.

Stress treatments

Seedlings during the blooming period were used for stress experiments (Fig. 2). Mature leaves, bulbs, bulb roots, stem roots, stems, and flower petals were sampled, respectively. All tissues and organs were directly frozen in liquid nitrogen and stored at −80 °C for gene expression analysis.

Cold, dehydration, salinity, and ABA treatments were performed on tiger lily by treating eight-week-old seedlings with 4 °C, 16.1 % PEG6000 (−0.5 MPa), 100mM NaCl, and 100µM exogenous ABA for 0, 1, 3, 6, 12, 24 h before sampling for expression analysis of LlZFHD4 (Yong, Zhang & Lyu, 2019a, 2019b). Similarly, three-week-old LlZFHD4 promoter transformant Arabidopsis plants were also treated with the above conditions for GUS gene expression assay. Three-week-old LlZFHD4 transgenic Arabidopsis and WT seedlings were harvested under normal conditions for expression analysis of AtRD29A, AtRD29B, AtRD20, AtLEA14, AtGolS1, and AtAPX2 genes (Table S1). All experiments were repeated three times in biological triplicates.

Stress tolerance assays in LlZFHD4 transgenic Arabidopsis

Three-week-old T3-generation transgenic Arabidopsis lines and WT seedlings were pretreated under 4 °C for 3 h, and then treated under –4, –6, or –8 °C for 12 h as cold treatment (Yong, Zhang & Lyu, 2019a, 2019b). The water intake of seedlings was withheld for 30 days as drought treatment (Yong, Zhang & Lyu, 2019b). After stress treatments, the stress-treated Arabidopsis seedlings were recovered in the growth chamber under normal growth conditions for seven days. The survival rate of Arabidopsis seedlings was scored. The relative electrolyte leakage and soluble sugar content were determined before (control) and after 3 h 4 °C and 16.1% PEG6000 (−0.5 MPa) treatments by the thermal conductivity measurement method and the anthrone method, according to Zhang et al. (2015). The measuring method of water loss rate was described in previous study of Cao et al. (2007).

Seeds of selected T3-generation homozygous transgenic lines and WT were sowed on MS medium containing 2 µM ABA or 50 mM NaCl for seven days (Yong, Zhang & Lyu, 2019a); and then the germination was scored and photographed.

Gene isolation and sequence analysis of LlZFHD4

LlZFHD4 gene comprises 543 bp open reading frame (ORF) corresponding to a protein of 180 amino acids with a calculated molecular mass of 20.08 kDa and a pI of 8.91 (Fig. S1). The LlZFHD4 contained a DNA-binding homeodomain in the C-terminus, and a conserved zinc finger domain in the N-terminus; two segments Ia and Ib were located in the zinc finger domain (Fig. 3A). Amino acid sequence alignment results showed that LlZFHD4 shared 67%, 47%, 60%, 59%, 54%, 53% identities with AtZHD4 and AtZHD3 from Arabidopsis, PdZHD4 from Phoenix dactylifera, EgZHD4-like from Elaeis guineensis, OsZHD3 from rice, GmZHD4-like from Glycine max (Fig. 3A). A phylogenetic tree was constructed based on the known amino acid sequences of ZFHD genes in the model plant (Arabidopsis) and crop (rice), showing that the LlZFHD4 sequence clustered closely with Arabidopsis AtZFHD4 and rice OsZHD4 shared 67% and 55% identities, respectively (Fig. 3B).

LlZFHD4 is a nucleus-localized transcriptional activator

The LlZFHD4 ORF without stop codon was connected with the N-terminal of the GFP gene driven by a CaMV 35S promoter. The 35S-LlZFHD4-GFP fusion protein was observed in the subcellular localization of LlZFHD4. A nuclear localization signal was detected in 35S-LlZFHD4-GFP transformed tobacco leaf epidermal cell; in contrast, a ubiquitous distribution signal in the whole cell was detected in control (35S-GFP) (Fig. 4A), suggesting LlZFHD4 is a nucleus-localized protein.

To investigate whether the LlZFHD4 has transactivation activity, recombinant pGBKT7 vector LlZFHD4-A (entire coding region), LlZFHD4-N (N-terminal region), LlZFHD4-C (C-terminal region), and pGBDKT7 vector (control) were introduced into the Y2HGold yeast strain. The LlZFHD4-A and LlZFHD4-N contained yeast strain grew well on the SD/-Trp/-His/-Ade medium, and the colonies are blue on SD/-Trp/-His-x-α-gal medium. These results indicate the LlZFHD4 has transcriptional activity, and the deletion of the C-terminal region (from the position of 91 to 180 aa) did not affect the activation (Fig. 4B), which means that LlZFHD4 is a transcription activator with transactivation in the N-terminus.

Expression patterns of LlZFHD4 under stresses

Under normal conditions, LlZFHD4 was expressed in all detected organs of tiger lily, including mature leaf, bulb, bulb root, stem root, stem, and flower petal. The transcript level of LlZFHD4 was shown to be the highest in flower petal followed by bulb, while it was low in leaf and stem (Fig. 5A). Under ABA (100 μM) treatment, the expression of LlZFHD4 was highly induced within 2 h with a threefold to fourfold increase, and then peaked at 24 h (Fig. 5B). Similarly, salt (NaCl 100 mM) treatment also induced the expression of LlZFHD4 within 2 h showing a twofold to threefold increase (Fig. 5E). However, treatment of plants with cold (4 °C) and drought (16.1% PEG6000) stresses could not up-regulate the expression of LlZFHD4 until 24 h with a fivefold to sixfold increase (Figs. 5C, 5D). These results showed that LlZFHD4 is cold, drought, and salt-responsive (Table S2 lists all the raw data of Fig. 5).

Promoter analysis of LlZFHD4 in response to stresses

A 789 bp upstream of the ATG start codon of the LlZFHD4 gene was cloned and used as the LlZFHD4 promoter sequence (Fig. S2). Putative cis-acting regulatory elements annotated as stress- or hormone-responsive elements (T-Box, BoxI. and ARF elements) were identified (Table 1). Three binding sites (MYC, DRE, and W-box) of MYC, DREB, and WRKY TFs were also found in the promoter region of LlZFHD4. The expression of the GUS gene driven by the LlZFHD4 promoter was detected by qRT-PCR in transgenic Arabidopsis seedlings. The qRT-PCR results showed that treatment of transgenic Arabidopsis with cold (4 °C), drought (16.1% PEG6000), salt (NaCl 100 mM), and ABA (100 μM) could induce GUS gene with a maximal transcript level at 12 h, leading to a five fold to eightfold increase (Fig. 6); suggesting the promoter activity of LlZFHD4 can be induced by these stresses in some degree (Table S3 lists all the raw data of Fig. 6).

Overexpressing LlZFHD4 alters the abiotic stress tolerance of transgenic Arabidopsis

Two T2 generations LlZFHD4 transgenic Arabidopsis lines, Line 6 and Line 7 (L6, L7), with relatively high LlZFHD4 transcript levels, were chosen by qRT-PCR (Fig. S3). The T3 generation of L6 and L7 were used for subsequence stress tolerance analysis. Under normal growth conditions, L6, L7, and WT Arabidopsis seedlings all grew well. No difference in plant morphology between L6, L7, and WT plants was noticed (Figs. 7A, 7B). However, under cold and water stress treatments, the less damaging effects were observed on L6 and L7, compared to WT. After exposing to freezing temperatures (especially under −6 °C) for 12 h or withholding water for 30 days, L6 and L7 seedlings displayed better growth status with larger leaf area, and significantly higher survival rate as compared to WT plants (Figs. 7A, 7B). Additionally, some essential physiological parameters, including water-loss rate, relative electrolyte leakage, and soluble sugar content, were measured before (control) and after 3 h 4 °C and 16.1% PEG6000 (−0.5 MPa) treatments. We found that L6 and L7 plants showed lower water-loss rates and electrolyte leakage amounts, higher soluble sugar levels than WT plants (Figs. 7C, 7D, 7E). These results indicate the LlZFHD4 transgenic plants are more tolerant to cold and water stresses than WT plants.

Under salt treatment, however, lower germination ratios measured by radicle protrusion rate were observed in L6 and L7, especially in L7, than in WT on the MS agar plates supplemented with 50 mM NaCl (Figs. 8A, 8B). We also found L6 and L7, especially L6, displayed lower germination ratios than WT under ABA (2 µM) treatment measured by both radicle protrusion and cotyledon greening (Figs. 8A, 8B). These results indicate the LlZFHD4 transgenic plants might be less tolerant to salinity and more hypersensitive to ABA than WT plants.

Overexpressing LlZFHD4 increases the expression of stress- and ABA-responsive genes

To further explore the molecular mechanism underlying LlZFHD4 in response to osmotic stresses, we assessed the expression levels of some known stress- and ABA-responsive genes in LlZFHD4 transgenic Arabidopsis under normal growth conditions. The results showed that the transcripts of AtRD29A, AtRD20, AtGolS1, AtLEA14, AtAPX2, and AtRD29B genes (Table S1) accumulated significantly higher in L6 and L7 than WT seedlings (Fig. 9). More importantly, the expression level of AtRD29A, AtRD29B, and AtAPX2 in L6 and L7 was even more than 20 folds of that in WT (Fig. 9), implying LlZFHD4 may confer cold and water stress tolerances by effectively up-regulating stress- and ABA-responsive genes.