Introduction

During the courses of growth and development, plants are exposed to various biotic and abiotic stresses, such as pathogen infection and drought stress. Plants have evolved through a series of mechanisms to adapt to these environmental challenges at physiological and molecular levels (Duque et al. 2013; Gupta and Huang 2014; AbuQamar et al. 2017).

Under the drought stress condition, the plant adaptive mechanisms include diverse physiological alterations such as the retention of water, root modification, and morphological changes in stomatal behavior and leaves (Shulaev et al. 2008; Cao and Li 2010; Gechev and Hille 2012). The physiological changes are determined by molecular responses, eventually resulting in the adaptation to diverse stress challenges (Rejeb et al. 2014; Ramegowda and Senthilkumar 2015; Sham et al. 2015). One of the most important mechanisms in keeping the environmental acclimatization is the activation of numerous stress-related genes, such as genes belonging to the transcription factor families of WRKY, myeloblastosis (MYB), and apetala2/ethylene response factor (AP2/ERF), which confer to the metabolic adjustments of plants under both biotic and abiotic stresses ultimately (Pandey et al. 2015; Sham et al. 2014, 2017). The identification of stress-related genes is one of the most important strategies for understanding the complex molecular adaptation mechanisms of plants under stress conditions. Transcription factors can dramatically influence the stress resistance of plants by regulating the expression of a series of downstream genes involved in stress responses (Singh et al. 2002; Sudha and Ravishankar 2002; Harb et al. 2010).

Ethylene-responsive transcription factors (ERFs) are DNA-binding proteins belonging to the AP2/ERF family. These plant-specific transcription factors have been proven to be involved in plant growth and development, environmental adaptation, and stress resistance (Dietz et al. 2010). The subsets of ethylene responses are modulated by different ERFs (Fujimoto et al. 2000). To date, ERFs are involved in abiotic stress-induced responses in many plants, such as rice (Pegoraro et al. 2013; Santos et al. 2013), maize (Nguyen et al. 2009), tobacco (Zhang et al. 2009, 2016), and tomato (Pan et al. 2012). ERF proteins are also involved in defense responses to pathogen attacks. (Oñate-sánchez and Singh 2002; Zhang et al. 2004; Zuo et al. 2010).

We previously cloned a full-length cDNA of LeERF-1, a B3 subfamily member of AP2/ERF family, from the medicinal plant L. erythrorhizon, and reported that LeERF-1 might be involved in light- and ethylene-regulated biosynthesis of secondary metabolites, shikonin and its derivatives, in L. erythrorhizon (Zhang et al. 2011).

Stress factors, such as low temperature, drought stress, and salt stress can alter the gene expression profile and influence the production of secondary metabolites (Rao and Ravishankar 2002; Ishita et al. 2010). Plants synthesize a diverse range of active secondary metabolites to protect themselves against a wide variety stresses (Dixon 2001; Iriti and Faoro 2009; Murcia et al. 2016). Therefore, efforts investigating the role of LeERF-1 on the molecular adaptation mechanisms of stresses are of fundamental importance to bridge the gap between stress response and secondary metabolism. However, the main function of LeERF-1 in stress response remains unclear.

In the current study, we cloned and characterized the promoter sequence of LeERF-1, and further clarified the role of LeERF-1 in response to drought and pathogen stresses in transgenic Arabidopsis. This new understanding of the regulatory role of LeERF-1 on stress resistance would allow a better manipulation and engineering of LeERF-1 to enhance plant adaption to abiotic and biotic stresses.

Materials and methods

Cloning and sequence analysis of the LeERF-1 promoter

The promoter of LeERF-1 was isolated from the genomic DNA of L. erythrorhizon using a thermal asymmetric interlaced (TAIL)-PCR (Liu et al. 1995; Liu and Whittier 1995; Siebert et al. 1995). PCR was performed using three LeERF-1 gene-specific primers ERF-GSP1, ERF-GSP2, and ERF-GSP3 (Suppl. Table S1) with the random primer AP1 in the TaKaRa Genome Walking kit [TaKaRa Biotechnology (Dalian) Co., Ltd.]. The plant cis-acting regulatory DNA elements (PLACE) database (http://www.dna.affrc.go.jp/PLACE/) were used to identify the putative functional cis-acting elements of the LeERF-1 promoter (Higo et al. 1999).

Plant materials and growth conditions

The seeds of Arabidopsis thaliana (Col-0 ecotype) transformed with the expression vector pBI121–LeERF-1–eGFP (Zhang et al. 2011) were surface-sterilized with 70% ethanol, washed five times with sterile water, and placed on 1/2 MS medium. The overexpression of the LeERF-1 transgenic plants (35S:LeERF-1) was confirmed by both PCR analysis with the gene-specific primers (Suppl. Table S1) and subcellular localization of LeERF-1 protein as we previously reported (Zhang et al. 2011). The wild-type (WT) and 35S:LeERF-1 plants were grown at 23 °C with 16-h-light/8-h-dark cycle with 80% relative air humidity. Plants producing 100% kanamycin-resistant progenies in the T3 or T4 generation were considered homozygotes and were selected for transgenic analysis. The DNA of the transgenic plants above were used as template for PCR amplification of the inserted target sequence in the pBI121–LeERF-1 vector using the primer pair 35S-F/GFP-R (Suppl. Table S1). Transgenic Arabidopsis lines conformed by RT-PCR were randomly selected and used for further studies (Suppl. Fig. S1).

Drought stress treatments

Three randomly selected transgenic Arabidopsis lines (OE-9, OE-29, and OE-57) were stratified under 4 °C for 2 days on 1/2 MS medium supplemented with or without 5% PEG and were germinated under normal growth condition (23 °C with 16-h-light/8-h-dark cycle and relative air humidity was about 80%). After 3 weeks, these were used for phenotypic observation and drought-resistant selection.

Seeds were germinated on a 1/2 MS containing 5% PEG or without PEG for 2 weeks to measure the root lengths of the seedlings. Four-week-old leaves were punched into small discs and incubated in serial concentrations of PEG solution (5%, 10%, and 15%, respectively) with the aqueous solution treatment as control. Phenotypic changes were observed after 72 h.

Leaves of OE-9, OE-29, OE-57, and WT grown in soil pots were detached and placed on filter papers at the ambient environment for 0, 10, 30, 60, and 120 min, respectively. The stomatal opening of guard cells in isolated epidermal tissues was captured using a confocal laser-scanning fluorescence microscope (U-RFL-T, Olympus, Japan).

The values are represented by means ± SD (n = 3), and the bars with different letters are significantly different at P < 0.05 (lower letters) or P < 0.01 (capital letters), respectively.

Pathogen infection

Four-week-old plants grown on soil under normal growth condition were used for pathogen infection. Pseudomonas syringae pv. tomato DC3000 (Pst. DC3000) was selected as the infection strain (Buell et al. 2003; Xin and He 2013; Bao et al. 2014; Chae Woo et al. 2014). This strain was cultured in liquid NYGB medium containing 50 µg/mL rifampicin and placed in a rotary shaker at 200 rpm for 8–9 h at 28 °C. The sample was collected by centrifugation, and re-suspended in 10 mM MgCl2 containing 0.02% Silwet L-77 to a concentration of OD600 of about 0.1. The inoculation of the bacteria-containing Arabidopsis leaf tissues was performed similar to the method described by Cao et al. (2005). After the inoculation, the plants were kept at 100% humidity for 24 h and then raised in a growth room at 25 °C under the 16-h-light/8-h-dark condition.

The phenotypic observation was performed after 3 days of inoculation. The disease was evaluated based on the size of lesion and the bacterial population in leaf tissues (Cao et al. 2005).

To determine the bacterial growth, leaves for each inoculated plant were collected and placed in a 1.5-mL centrifuge tube after 6 days of inoculation. Samples were vigorously shaken in the test tube with 1 mL of 10 mM MgCl2. Moreover, to count the number of spores, the leaves were removed, and the spore-containing suspension was centrifuged at 5000g for 5 min. The spores were re-suspended and counted with a blood cell counter (Bertoni and Mills 1987; Oh et al. 2005).

Statistical analysis

Statistical analyses were performed using the SPSS 17.0 software (IBM, IL, USA). One-way ANOVA and significant difference method (LSD) were used for the comparison between WT and LeERF-1 transgenic lines. The values are represented by means ± SD, and the bars with different letters are significantly different at P < 0.05 (lower letters) or P < 0.01 (capital letters), respectively.

Results

Cloning and sequence analysis of the LeERF-1 promoter region

To understand the expression regulation of LeERF-1, the promoter fragment of LeERF-1 was isolated from the genomic DNA of L. erythrorhizon via genome-walking method (Suppl. Fig. S2), and 2615 bp of the PCR product (about 5000 bp) was obtained by walking sequencing from one end (GenBank Accession Number KX768421) after excluding the overlap with the cDNA sequence of LeERF-1. The putative cis-acting elements of the obtained promoter sequence were then analyzed. Results showed that the promoter sequence contains multiple potential regulatory motifs corresponding to several known cis-acting elements which are related to tissue-specific expression, phytohormones (abscisic acid, jasmonate, ethylene, salicylate, and gibberellic acid) regulation, secondary metabolism, and biotic (disease resistance) and abiotic stress (drought, cold, and salt) responses. In addition, seven light regulation elements, CIACADIANLELHC, DRECRTCOREAT, EBOXBNNAPA, IBOXCORE, INRNTPSADB, SORLIP2AT, and TBOXATGAPB, were also found in the promoter region (Suppl. Table S2).

Overexpression of LeERF-1 improves drought resistance

Based on the above bioinformatics analysis of the promoter sequence, we further tested whether the heterologous expression of LeERF-1 can actually affect the capacity of the transgenic Arabidopsis to tolerate drought stress.

We first checked the growth performance of germinated seedlings on a 1/2 MS medium with 5% PEG. On 1/2 MS medium without PEG, no difference was observed on the seed germination (Suppl. Fig. S3) or seedling phenotypes (Suppl. Fig. S4) between WT and three transgenic Arabidopsis lines. However, all germinated seedlings on 1/2 MS medium with 5% PEG showed visual drought-associated phenotype symptoms, such as leaf rolling and wilting, after 3 weeks. The seedlings of WT showed no greening and failed to survive after 3 weeks, whereas transgenic lines remained healthy and displayed higher survival ratio than WT, indicating that LeERF-1 transgenic Arabidopsis exhibited higher resistance to drought stress than WT (Fig. 1a).

Fig. 1
figure 1

The phenotype comparison between WT and three LeERF-1 transgenic Arabidopsis lines under simulated drought stress. a The phenotype of germinated seedlings of WT and three transgenic Arabidopsis lines on 1/2 MS medium with 5% PEG for 3 weeks. b The comparison of root length between WT and transgenic Arabidopsis lines on 1/2 MS medium with 5% PEG for 2 weeks. c The statistical comparison of root lengths between WT and LeERF-1 transgenic lines under simulated drought stress. The values are means ± SD, and the bars with different letters are significantly different at P < 0.05 (lower letters) or P < 0.01 (capital letters), respectively

Full size image

Root elongation assays indicated that the seedlings of WT exhibited an apparently suppressed phenotype in terms of root length when germinated on 1/2 MS medium with 5% PEG (Fig. 1b). The root length of WT was significantly lower than that of the overexpressed transgenic seedlings (P < 0.01) (Fig. 1c). The longer root length of LeERF-1 transgenic plants was consistent with drought resistance phenotype as compared with WT.

We then checked the drought resistance by immersing the leaf discs prepared from 4-week-old plants for 72 h in water or in 5%, 10%, and 15% PEG solution. With the different concentration of PEG, the leaf discs from WT exhibited more serious bleaching and necrosis when compared with those of the transgenic lines which appeared green and healthy (Fig. 2). These results suggest that the constitutive overexpression of LeERF-1 leads to the accelerated leaf drought resistance.

Fig. 2
figure 2

The phenotype comparison of leaf discs between WT and transgenic lines immersed into aqueous solution and 5%, 10%, and 15% PEG stress treatments for 72 h

Full size image

Considering the closure of stomata was correlated with the useful strategy for improving the efficient water use under water-limiting conditions (Tezara et al. 1999; Merlot et al. 2002), we conducted the examination of stomatal apertures under drought stress condition. The results showed no difference in stomatal pore size of plants before drought stress treatment. Drought treatment of guard cells of the isolated epidermal tissues induced the alterations of stomatal closure in WT and in all lines of LeERF-1 transgenic Arabidopsis (Fig. 3a). Less stomatal opening of the guard cells was observed with the extension of drought stress time in each detected line. At each stress time points of 30, 60, and 120 min, LeERF-1 transgenic Arabidopsis had significantly lower stomatal apertures compared to that of WT (P < 0.01) (Fig. 3b). The percentage of the stomatal aperture stomata in LeERF-1 transgenic Arabidopsis leaves was from 68.18 to 76.61% at 120 min stress point, compared with WT.

Fig. 3
figure 3

The stomatal opening assay of LeERF-1-overexpressing transgenic plants in soil pots. Leaves were detached and placed on filter papers at ambient environment for 0, 30, 60, and 120 min, respectively. The stomatal opening of guard cells in isolated epidermal tissues was imaged using a confocal laser-scanning fluorescence microscope (a). The data were statistically analyzed (b), in which the values are means ± SD (n = 3) and the bars with capital letters indicate significant differences at P < 0.01 (least significant difference)

Full size image

Transgenic Arabidopsis of LeERF-1 confers improved resistance to pathogens

The inoculation of the Arabidopsis leaf tissues with Pst. DC3000 was performed to determine whether the overexpression of LeERF-1 affects disease responses to bacterial pathogens. The leaves from WT exhibited more serious chlorotic symptoms of wilting and yellow spots after Pst. DC3000 inoculation for 3 days compared to the LeERF-1 transgenic Arabidopsis which did not exhibit visible macroscopic signs of infection after 3 days. After 6 days of bacterial inoculation, the leaves of WT appeared completely wilted; in contrast, LeERF-1 transgenic Arabidopsis lines exhibited fewer chlorotic symptoms compared with WT at 6 days after inoculation (Fig. 4a). The incidence rates of leaves from LeERF-1 transgenic Arabidopsis were significantly lower than that of WT infected with Pst. DC3000 strains after 3 and 6 days of inoculation (P < 0.01) (Fig. 4b). This result indicates a heterologous overexpression of LeERF-1 enhanced resistance to bacterial pathogen Pst. DC3000.

Fig. 4
figure 4

The pathogen infection assay of LeERF-1-overexpressing transgenic plants. a The chlorotic symptoms of WT and LeERF-1 transgenic Arabidopsis plants infected with P. syringae DC3000 (Pst.DC3000) strains after 6 days of inoculation. b The incidence rates of leaves from WT and LeERF-1 transgenic Arabidopsis plants infected with Pst DC3000 strains at 3 and 6 days after inoculation. c The bacterial population from the leaves of WT and transgenic Arabidopsis lines infected with Pst. DC3000 strains at 6 days after inoculation. The values are means ± SD (n = 3), and the bars with different letters are significantly different at P < 0.05 (lower letters) or P < 0.01 (capital letters), respectively

Full size image

Furthermore, we speculated that these chlorotic symptoms might reflect bacterial multiplication inside the leaf tissues. To confirm this speculation, the bacterial population on the leaves of the WT and LeERF-1 transgenic Arabidopsis lines was measured. After treatment with Pst. DC3000 for 6 days, the bacterial growth in both the WT and LeERF-1 transgenic Arabidopsis lines increased significantly. Consistent with the visible symptoms, LeERF-1 transgenic Arabidopsis lines exhibited significantly lower levels of bacterial proliferation than that of WT plants after 6 days of inoculation (P < 0.01) (Fig. 4c).

Therefore, these results indicated that the heterologous overexpression of LeERF-1 may synergistically contribute to the plant immune system for resisting the attack of bacterial pathogen Pst. DC3000.

Discussion

AP2/ERF is a large family of transcription factors in the ethylene signaling transduction pathway in plants. Based on the amino acid sequence similarity of the AP2/ERF DNA-binding domain, these AP2/ERF transcription factors in Arabidopsis are classified into five groups: AP2 subfamily, DREB subfamily, ERF subfamily, RAV subfamily and one very specific gene, At4g13040. The ERF subfamily and DREB subfamily can be further divided into several subgroups (Gutterson and Reuber 2004; Thamilarasan et al. 2014; Dossa et al. 2016). Accumulating evidence have suggested that a number of ERF members participate in the regulation of biotic and abiotic stress responses (Oñate-Sánchez et al. 2007; Pan et al. 2012; Zhang et al. 2012, 2016).

Although the AP2/ERF family has been widely studied in various plants, the current study is the first report on the role of LeERF-1, a B3 subfamily member of AP2/ERF from the medicinal plant L. erythrorhizon, based on its pathogen and drought stress responses in transgenic Arabidopsis.

Our previous study reported that the light-regulated LeERF-1 was possibly involved in the biosynthesis of shikonin and its derivatives, the secondary metabolites of L. erythrorhizon. To test whether LeERF-1 can also confer stress responses and to verify the presence of stress response cis-elements in this region, we first cloned and characterized the promoter sequence of LeERF-1. The PLACE analysis showed that the promoter of LeERF-1 contains different potential regulatory motifs which are corresponding to the diverse stress responses. We presumed that the expression of LeERF-1 may be driven by multiple stress-related regulators.

With the heterologous overexpression in A. thaliana, we investigated the role of LeERF-1 on the molecular adaptation mechanisms to drought stress. Results showed that the LeERF-1 transgenic Arabidopsis seedlings displayed healthy phenotype, higher survival ratio, lower root inhibition rate, slower water loss, and smaller stomatal apertures under the drought stresses, compared with WT. The drought resistance phenotype of LeERF-1 transgenic plants was consistent with slower water loss in leaf discs and stomatal opening of the guard cells as compared with the WT.

Given the growth and development dependence of plants on the tight regulation of water uptake by root growth and stomata closure under drought stress condition, we hypothesized that the proper regulation of LeERF-1 might be necessary for proper root elongation to absorb water. In addition, the overexpression of LeERF-1 might alter the proper stomatal behavior to conserve water by reducing water loss. However, this hypothesis needs to be further identified.

In summary, these results indicated that the heterologous expression of LeERF-1 enables the Arabidopsis to have more adaptive abilities for growth under drought stress condition, i.e., the LeERF-1 confers enhanced drought stress resistance in plant.

The regulation of AP2/ERF transcription factors in plant physiological and developmental processes have been extensively documented (Dietz et al. 2010; Zhang et al. 2012; Cheng et al. 2013). The ERF subfamily factors are mainly involved in response to biotic stresses such as pathogenesis (Oñatesánchez and Singh 2002; Zhang et al. 2004, 2009; Dossa et al. 2016).

Until recently, systematic profiling of genomics, transcriptomics, proteomics, and metabolomics can be a potential solution to the pathogenesis of human pandemic influenza viruses and plant diseases (Abuqamar et al. 2016; Tisoncikgo et al. 2016). Since the pathogen Pst. DC3000 has been used as a model for plant–pathogen interactions (Buell et al. 2003; Xin and He 2013; Bao et al. 2014; Chae Woo et al. 2014), we also analyzed the response of transgenic Arabidopsis seedlings overexpressing LeERF-1 under Pst. DC3000 infection.

In our current study, the seedlings of LeERF-1 transgenic Arabidopsis showed fewer chlorotic symptoms, lower incidence rates, and lower levels of bacterial proliferation on leaves after inoculation of bacterial pathogen, compared with that of WT. Results suggested that LeERF-1 also plays various important roles in regulating plant responses to pathogen attacks. Our studies of LeERF-1 function on the plant resistance to bacterial pathogen will also provide a foundation for genetic manipulation and breeding programs in the L. erythrorhizon and in other plants.

The regulation of ERFs in plant growth, development, and response to different environmental stress factors have been extensively documented (Dietz et al. 2010; Zhang et al. 2012; Cheng et al. 2013; Dossa et al. 2016). On the other hand, the diverse stress factors can alter the expression profile of stress-related genes and influence the production of secondary metabolites to protect themselves against a wide variety stresses (Dixon 2001; Rao and Ravishankar 2002; Iriti and Faoro 2009). As an ERF subfamily member, LeERF-1 has been involved in the ethylene-regulated biosynthesis of shikonin and its derivatives in L. erythrorhizon (Zhang et al. 2011).

Therefore, our study on the role of LeERF-1 under diverse stress conditions will help us further understand the relationship between the molecular adaptation mechanisms of stresses and secondary metabolism in the medicinal plant L. erythrorhizon in the future. We speculated that the accumulation of LeERF-1 transcripts enhances disease and stress resistance, which in turn correlated with the induction of shikonin production for more adaptive abilities in the growth of L. erythrorhizon. Further research on the exact function of LeERF-1 in L. erythrorhizon experiencing stresses will provide direct evidence for the induction of stress responses such as the specialized metabolite production of shikonin.

In conclusion, the heterologous overexpression of LeERF-1 gene reveals its regulating role in response of pathogen attack and drought stress. Our results showed that LeERF-1 is an attractive engineering target gene conferring stress resistance. This work might help to provide a useful tool to understand the complicated relationship between stress-related genes and the biosynthesis of secondary metabolites in L. erythrorhizon in the future.

Author contribution statement

RJF, GHL, JLQ and YHY designed the research and wrote the paper; RJF, AQL, RNT, WJZ, ALZ, FYW and YHL performed the experiments and analyzed the data; XMW, YJP and RWY discussed the results.