Introduction

Flavonoids, a large group of plant secondary metabolites, have abundant biological functions, such as defense against phytopathogens and stress protection. They are found primarily in fruit, grains, bark, vegetables, roots, flowers, tea, wine and stems (Sankari et al. 2014). The basic structure of flavonoids consists of 15 carbon atoms, two benzene rings joined by a linear three-carbon chain. More than 6000 naturally occurring flavonoids have been identified in various plants (Czemmel et al. 2012; Perez-Cano and Castell 2016). These flavonoids can be classified into six classes: flavonols, flavones, isoflavones, flavanones, flavanols and anthocyanidins. Flavonoids have diverse biological functions, including protection against ultraviolet (UV) radiation and phytopathogens, signaling during nodulation, male fertility, auxin transport, as well as the coloration of flowers as a visual signal that attract spollinators (Ferreyra et al. 2012). Moreover, they provide abundant resources for agents that promote and maintain health (Sankari et al. 2014).

Flavonoids are synthesized through the phenylpropanoid pathway, transforming phenylalanine into 4-coumaroyl-CoA, which finally enters the flavonoid biosynthesis pathway (Ferreyra et al. 2012; Li et al. 2017). Flavanone 3-hydroxylase (F3H) is the third enzyme of the central flavonoid biosynthetic pathway, catalyzing the 3-hydroxylation of (2S)-flavanones such as naringenin to dihydroflavonols. For example, F3H serves as the intermediate for the biosynthesis of flavan-3-ols in Reaumuria soongorica (Liu et al. 2013a). The high expression of CtF3H in quinochalcone-type safflower line is associated with the accumulation of quinochalcones and flavonols (Tu et al. 2016). In Artemisia annua, the F3H transcripts are found to be accumulated in the cultivar with higher level of flavonoids (Xiong et al. 2016). Furthermore, flavonoids and their related key enzymes play an important role in response to diverse biotic and abiotic stresses. Flavonols are increased in Ligustrum vulgare under salinity or UV radiation stress and have a significant antioxidant function in plant photoprotection (Agati et al. 2011). In a desert plant Reaumuria soongorica, the F3H gene (i.e., RsF3H) involves in the flavonoid biosynthesis pathway in response to UV-B radiation and drought stress (Liu et al. 2013a). Overexpression of a tea F3H gene (i.e., CsF3H) increases the content of flavan-3-ols in tobacco and confers tolerance to salt stress and fungus Alternaria solani infection (Mahajan and Yadav 2014). Overexpression of a tomato F3H-like gene (i.e., SlF3HL) stimulates flavonoid biosynthesis and improves chilling tolerance in tobacco (Meng et al. 2015). However, there is still no report about the roles of F3H in bryophytes to stress responses.

In Antarctica continent, high salt, low temperatures, extreme dryness and high radiation are the major constraints to plant growth. The moss flora constitutes the predominant terrestrial vegetation in ice-free coastal regions of Antarctica (Skotnicki et al. 2004). Mosses have evolved a variety of mechanisms to cope with the extreme environment of Antarctica. For example, Antarctic cryptogams resist the effects of UV radiation using efficient systems that repair damages by synthesizing compounds such as UV-B-absorbing pigments and anthocyanins (Singh et al. 2010). The moss Bryum argenteum can tolerate the high-intensity light by producing abundant xanthophyll cycle pigments (Schroeter et al. 2012). Furthermore, several cell wall-bound insoluble phenylpropanoids function in relatively passive UV-screening processes that increase the tolerance to UV radiation (Clarke and Robinson 2008). However, the underlying mechanisms that enable Antarctic mosses to survive under extreme conditions have not been characterized as completely as the corresponding processes of vascular plants such as Arabidopsis thaliana and rice.

To clarify the mechanism regulating stress acclimation, we previously conducted a transcript-level investigation of an Antarctic moss (i.e., Pohlia nutans) exposed to cold stress using Illumina high-throughput sequencing technology (Liu et al. 2013b). We also found that several receptor-like kinases (RLK) from Pohlia nutans serve in high salinity (Wang et al. 2016) and ROS scavenging (Zhang et al. 2014), and provide protection against chilling-stress tolerance (Liu et al. 2017), involving in the adaptation of Antarctic moss to the polar environment. In this paper, we cloned a novel F3H gene (i.e., PnF3H) from the Antarctic moss P. nutans and studied its role in plant salt stress resistance. We proposed that PnF3H may serve as a flavonoid biosynthetic enzyme, conferring the salinity tolerance in the Antarctic moss P. nutans.

Materials and methods

Plant materials and stress treatments

We collected the Antarctic moss Pohlia nutans from the terrane near the Great Wall Station on King George Island during the 31th CHINARE in March 2014. Mosses were cultivated in flowerpots containing a mixture of Base Substrate (Klasmann–Deilmann, Geeste, Germany) and local soil (1:1) (Liu et al. 2017). They were incubated at 16 °C and 70% relative humidity, under 70 μmol photons m− 2 s− 1 light. Mosses cultivated under above conditions were used as controls in all experiments. For cold treatments, mosses were kept in 4 °C for 0, 1, 3, 6, 12, or 24 h. For NaCl and ABA treatment, mosses were sprayed with 200 mM NaCl or 50 μM ABA, respectively, and incubated for 0, 0.5, 1, 3, 6, or 12 h. For the mock osmotic stress and UV-B radiation stress, mosses were treated with 20% (w/v) polyethylene glycol 6000 solution or transferred under an 8 W UV-B lamp (302 nm, 70 μW cm− 2 irradiance), respectively, for 0, 0.5, 1, 3, 6, or 12 h. The samples were immediately frozen in liquid nitrogen after various stress treatments.

Reverse transcription-PCR and real-time PCR analyses

Total RNA was extracted using CTAB method (Zhang et al. 2014). The CTAB buffer contained 2% cetyltrimethylammonium bromide, 1% polyvinylpyrrolidone K-30, 100 mM Tris–HCl (pH8.0), 20 mM EDTA, 1.4 M NaCl, 0.5 g L− 1 spermidine, and 2% β-mercaptoethanol (added just before use). Moss gametophyte tissues or A. thaliana seeds 2 days after germination with NaCl and ABA treatment were used for total RNA isolation. cDNA was synthesized from 0.5 μg total RNA using the 5× All-In-One RT MasterMix Kit (Abm, Canada). Real-time PCR assays were conducted using the Bestar SybrGreen qPCR Master Mix (DBI Bioscience, Shanghai, China). The PCR reaction consists of a 2 min denaturation at 95 °C, followed by 40 cycles of 95 °C/20 s, 58 °C/20 s, and 72 °C/20 s, completed by an extension step of 10 min at 72 °C. All reactions were completed in triplicate. Data were analyzed using the 2− ΔΔCt method (Livak and Schmittgen 2001).

Bioinformatic analysis of PnF3H sequence

The Arabidopsis F3H protein (GenBank Acc. Q9S818) was used to blast against the P. nutans transcriptome (Liu et al. 2013b). Then, a flavanone 3-hydroxylase protein (i.e., PnF3H) with 356 Amino acids were identified from P. nutans. The ClustalW software was used for multiple sequences alignments between PnF3H and other plant F3Hs. To construct the phylogenetic tree, a neighbor-joining (NJ) method was adopted within ClustalX and MEGA4.0 software (Kumar et al. 2008). Meanwhile, the 3-D structure analysis of PnF3H was performed using the SWISS-MODEL program (Biasini et al. 2014), while PyMol software was used to visualize and manipulate structural models.

Subcellular localization of PnF3H

To construct the p35S::PnF3H::green fluorescent protein (GFP) plasmid, the full-length sequence of PnF3H was amplified and inserted into a modified pBI221 vector. The gene-specific PCR primers were listed in Table S1. Subcellular localization was performed using a polyethylene glycol 4000-mediated protoplast transformation method (Liu et al. 2017). Briefly, 4 × 104 mesophyll protoplasts isolated from 4-week-old A. thaliana Col-0 seedlings were transformed with 10 µg p35S::PnF3H::GFP/pBI221 vector. The GFP signal and chlorophyll autofluorescence were detected by a confocal laser-scanning microscope (LSM700, Carl Zeiss).

Construction of PnF3H transgenic Arabidopsis

PnF3H was amplified using the gene-specific primers listed in Table S1. The PCR procedure comprised of 3 min denaturation at 95 °C, followed by 34 cycles of 94 °C for 30 s, 57 °C for 30 s, 72 °C for 60 s, with a final reaction of 72 °C for 10 min. The PCR product was inserted into Xba I and Kpn I sites of a modified pROK2 vector containing a CaMV35S promoter. Then, the CaMV35S::PnF3H/pROK2 vector was introduced into Arabidopsis Col-0 plants using an Agrobacterium tumefaciens-mediated floral dip method (Zhang et al. 2006). Homozygous transgenic segregates were selected on 1/2 Murashige and Skoog (MS) medium supplemented with 50 μg mL− 1 kanamycin. At last, three PnF3H-overexpressing transgenic Arabidopsis lines (i.e., AtOE1, AtOE2, and AtOE3) from the T3 generation were selected for phenotypic analysis. The expression levels of PnF3H in AtOE plants were detected by RT-PCR.

Arabidopsis treatment and phenotype assay

For seed germination assay, surface-sterilized seeds were plated on 1/2 MS agar medium supplemented with 0, 100 or 125 mM NaCl, with 0, 0.5, or 0.75 μM ABA. Seeds were held in darkness at 4 °C for 2 days to break dormancy, and subsequently were cultured under a 16-h light/8-h dark photoperiod at 22 °C for 4–6 days. Seed germination rates were calculated by counting the proportion of the WT plants and the AtOE lines bearing open cotyledons (Zhang et al. 2014; Wang et al. 2016). For root length assays, seeds were sown on 1/2 MS agar medium with NaCl, ABA and H2O2, and held in darkness at 4 °C for 2 days to break dormancy, and then kept the plates upright in 16-h light/8-h dark photoperiod at 22 °C for 3 days (Yu et al. 2017).

Measurement of the ROS level and the activities of antioxidant enzymes

ROS level was detected by DAB staining as described previously (Wang et al. 2016). 2-week-old wild type and transgenic plants were treatment with 200 mM NaCl for 24 h or not. Then leaves were detached and infiltrated in 1 mg mL− 1 DAB with 10 mM sodium phosphate buffer (pH 3.8) overnight in dark. These stained leaves were decolorized by boiling in 95% ethanol solution. Finally, the staining samples were photographed. For antioxidant enzymes activities, 2-week-old Arabidopsis seedlings were treated by 200 mM NaCl for 24 h. Whole seedlings (0.5 g) were ground in an ice-cold mortar in the 5 mL homogenization buffer (50 mM potassium phosphate, pH 7.4). Supernatants were used to determine the activities of ROS-scavenging enzymes (Meng et al. 2015; Chen et al. 2017). Finally, the absorbance value of reaction mixtures was measured at 560 and 240 nm, respectively.

Measurement of ABA by LC–ESI–MS/MS system

Plant materials (100 mg fresh weight) were frozen in liquid nitrogen, ground into powder, and extracted with 1 mL methanol containing 20% water at 4 °C for 12 h. The extract was centrifuged at 12,000×g under 4 °C for 15 min. The supernatant was collected and then evaporated to dryness under nitrogen gas stream, reconstituted in 100 mL acetonitrile containing 5% water. The solution was centrifuged and the supernatant were analyzed using an LC–ESI–MS/MS system as previously described (Fu et al. 2012; Chen et al. 2013). The analytical conditions were as follows, HPLC: column, Waters ACQUITY UPLC HSS T3 C18 (1.8 µm, 2.1 mm × 100 mm); solvent system, water (0.04% acetic acid):acetonitrile (0.04% acetic acid); gradient program, 100:0 V/V at 0 min, 5:95 V/V at 11.0 min, 5:95 V/V at 12.0 min, 95:5 V/V at 12.1 min, 95:5 V/V at 15.0 min; flow rate, 0.40 mL min−1; temperature, 40 °C; injection volume: 5 μL. The effluent was alternatively connected to an ESI-triple quadrupole-linear ion trap (Q TRAP)-MS.

Naringenin-feeding experiments and flavonoid metabolism profiling by HPLC analysis

Surface-sterilized seeds (3 mg) of wild-type and transgenic Arabidopsis were cultured in 100 mL of liquid MS medium (Murashige and Skoog plant salt mixture, Gamborg’s B5 vitamin salt mixture, 1% sucrose, 0.05% MES, pH 5.7) under long-day conditions for 2 weeks with or without 200 μM naringenin. Then seedlings were washed with distilled water three times, frozen in liquid nitrogen (Yun et al. 2008). Frozen Arabidopsis seedlings were freeze-dried for 48 h and were fully ground to powder. The dried powder (0.5 g) was mixed with 1 mL extraction solvent (70% HPLC-grade methanol with 0.1% HPLC-grade formic acid) and then sonicated in an ultrasonic water bath at room temperature for 1 h. The samples were placed at 4 °C overnight. The following day, the resulting extract was centrifuged at 12,000 rpm for 20 min at 4 °C. The supernatant was then filtered through Millipore filter (0.45 µm) prior to injection in HPLC. For HPLC analysis, 10 μL of the filtrate was performed with reversed-phase HPLC (Shimadzu LC-20A, Japan) coupled to an PDA detector on a Agilent C18 column (4.6 × 250 mm, 5 µm), mobile phases were H2O (A) and acetonitrile (B), and the elution program was: 0–30 min, 30% B-80% B; 30–35 min, 80% B; 35–45 min, 30% B. The flow rate of the mobile phase was 0.8 mL min−1, the temperature of the column was set at 30 °C, the elution was monitored by the absorption profile at 280 nm (Mahajan and Yadav 2014; Xiong et al. 2016).

Results

Multiple sequence alignment and phylogenetic analysis

The full-length of PnF3H cDNA from Pohlia nutans was 1072 bp, encoding a 356-amino-acid protein with a predicted molecular weight of 39.10 kDa and an isoelectric point of 6.22. Multiple sequence alignment analysis indicated that PnF3H is homologous with other F3Hs from species such as Physcomitrella patens subsp. patens (Pp1s287_51), Arabidopsis thaliana (AT3G51240), Elaeis guineensis (XP_0109225), Erythranthe guttata (XP_0128321), Medicago truncatula (XP_0134491) (Fig. 1a). Like other members of 2OG-FeII_Oxy dioxygenase family, PnF3H contained the conserved domains: the highly conserved ferrous iron binding site HXDX55H (labeled red box) and the 2-oxoglutarate (2-ODD) binding domain RXS (Arg301 and Ser303) (Fig. 1a). Homology modeling of PnF3H was carried out using the SWISS-MODEL program, and RXS motif and ferrous iron binding site were presented in Fig. 1h. Phylogenic analysis showed that PnF3H was clustered more closely with F3Hs from other species other than with anthocyanidin synthase (ANS) or flavonol synthase (FLS) (Fig. 1b).

Fig. 1
figure 1

PnF3H encodes an F3H protein and induced by various abiotic stresses. a Sequence alignment of PnF3H with other F3Hs. The positions of the five conserved motifs were labeled with bold lines. Red frames indicated the conserved motifs required for ferrous iron (HXD) and 2-oxoglutarate binding (RXS); Asterisks (*) indicated the conserved amino acid residues in F3Hs. b Phylogenetic tree was constructed using the neighbor-joining method. (c–g) PnF3H is induced by various abiotic stresses. c 200 mM NaCl treatment; d Cold treatment (4 °C); e 20% PEG6000 treatment for simulated drought stress. f UV-B radiation; g 50 μM abscisic acid (ABA) treatment. Vertical bar indicate mean ± SE of three replicates of the sample. Asterisks (*) indicate significant differences of means between the control group and the treatment group (p < 0.05). h 3-D structural model of PnF3H. The α-helix, β-sheet and random coils were colored with salmon, cyan, and bright blue, respectively. The conserved ferrous iron binding site HXDX55H (His233, Asp235, and His289) and the 2-oxoglutarate (2-ODD) binding domain RXS (Arg299 and Ser301) were labeled by arrow. i PnF3H protein targets to the cytosol. Arabidopsis protoplasts were transformed with p35S::GFP/pBI221 vectors (I) and p35S::PnF3H::GFP/ pBI221 (II). The confocal images were taken after 12 h of transfection. Green fluorescence of PnF3H-GFP or GFP proteins, bright field of protoplast, red fluorescence of chloroplasts and merged images were captured. Bar 10 μm. (Color figure online)

Full size image

PnF3H transcription level of multiple abiotic stresses

The PnF3H transcript profiles under various abiotic stress conditions were uncovered by Real-time PCR analysis. The PnF3H expression levels increased by 2.75-fold at 0.5 h after the 200 mM NaCl treatment and remained high up to 12 h after treatment (Fig. 1c). Similarly, the PnF3H expression levels increased by 2.49-fold at 1 h and then slightly decreased to 1.47-fold at 24 h with cold treatment (Fig. 1d). For the simulated drought stress (i.e., 20% polyethylene glycol 6000), the PnF3H expression levels enhanced by 2.39-fold at 3 h and the transcript abundance remained high up to 12 h (Fig. 1e). In the UV-B radiation, the PnF3H expression levels increased and reached the maximum level of 4.57-fold at 3 h (Fig. 1f). After exogenous ABA treatment, the PnF3H expression levels increased quickly and reached the maximum level of 2.57-fold at 0.5 h (Fig. 1g). Thus, PnF3H transcripts were induced by various stress and may play a positive role in stress tolerance.

Subcellular localization of PnF3H in Arabidopsis mesophyll protoplasts

To investigate the protein subcellular localization, PnF3H-GFP fusion proteins were transiently expressed in Arabidopsis mesophyll protoplasts. In p35S::GFP/pBI221-carrying (i.e., vector control) protoplasts, GFP fluorescence were distributed throughout the protoplasts. In p35S::PnF3H::GFP-containing protoplasts, GFP signals were mainly distributed to the cytosol (Fig. 1i). When green fluorescence, bright field, and red fluorescence images were merged, distinct yellow signals were not detected in chloroplasts. These results suggested that PnF3H is a cytosol protein.

PnF3H enhanced the salinity tolerance and the antioxidant activity of transgenic Arabidopsis

Three homozygous PnF3H overexpressing lines (i.e., AtOE1, AtOE2, and AtOE3) were constructed and the gene expression levels of PnF3H were detected by RT-PCR method. For phenotype analysis, seeds from the wild-type plants and AtOE lines were sown on 1/2 MS agar medium supplemented with or without NaCl. In the absence of exogenous NaCl, there were no differences in the germination rates between the AtOE lines and the WT plants. However, following the 100 mM NaCl treatment, the germination rates of AtOE1, AtOE2, and AtOE3 plants were 56.3, 45.0, and 40.0%, respectively, which were higher than the 25.0% germination rate for the WT plants (Fig. 2a, c). In the presence of 125 mM NaCl, the germination rates of AtOE lines were also higher than that of the WT plants (Fig. 2c). For root growth assay, under normal conditions, no differences were observed between the WT plants and the AtOE lines (Fig. 2b). In the presence of 125 mM NaCl, the AtOE lines were more vigorous than the WT plants, forming longer primary roots.

Fig. 2
figure 2

PnF3H contributed to the salt- and oxidative- stress tolerance in transgenic Arabidopsis. a The seed germination rate of the AtOE lines was significantly higher than that of the WT plants under salt stress (6 days’ seed germination). b Primary root length of Arabidopsis seedlings under salt- or oxidative- stress. c Statistical analysis of seed germination rates as shown in (a). d Statistical analysis of primary root length of Arabidopsis seedlings under salt- or oxidative- stress in (b). e ROS levels of Arabidopsis by DAB staining without salt treatment, and ROS levels of Arabidopsis by DAB staining with 200 mM NaCl treatment. f The expression levels of several stress resistance genes were analyzed by real-time PCR. g SOD and h CAT activities of 2-week-old Arabidopsis seedlings with 200 mM NaCl treatment. Data were presented as means ± SE, and asterisks (*) indicated significant differences of means between the AtOE lines and the WT plants (P < 0.05)

Full size image

Salt tolerance is usually related to oxidative stress responsive signaling pathway (Golldack et al. 2014). PnF3H overexpressing lines showed superior growth status comparing with the WT line in the presence of 1.0 mM H2O2. The root length of the transgenic plants was approximately twofold longer than that of the WT plants at 1/2 MS medium containing 1 mM H2O2 (Fig. 2d). DAB staining showed that signals in the AtOE lines were weaker than those in the WT plants (Fig. 2e), suggesting that the transgenic plants exhibited lower H2O2 level under NaCl treatment. Additionally, the activities of superoxide dismutase (SOD) and catalase (CAT) in the AtOE lines were higher than that in the WT plants after NaCl treatment (Fig. 2g, h).

To reveal the possible molecular mechanisms underlying the improvement of stress resistance in transgenic Arabidopsis, 2-day-old germinating seeds were subjected to 200 mM NaCl treatment. The expression levels of several stress responsive genes (i.e., AtSOS3, AtP5CS1 and AtHKT1) in the AtOE lines and the WT plants were detected by real-time PCR. Under 200 mM NaCl treatment, the expression levels of those genes were up-regulated in the AtOE lines than that in the WT plants. Moreover, the expression levels of antioxidant enzymes (i.e., AtCAT1 and AtAPX1) in PnF3H overexpression lines were also higher than that in the WT plants (Fig. 2f).

PnF3H decreased the ABA sensitivity in transgenic Arabidopsis

ABA is a vital stress response-associated phytohormone that responses to biotic and abiotic stresses (Verma et al. 2016). For germination assays, results showed that the AtOE lines were insensitive to the provision of exogenous ABA. In the absence of exogenous ABA, there were no obvious differences in the germination rates between the AtOE lines and the WT plants. Under ABA treatment, the germination of the AtOE lines and the WT plants were both significantly inhibited, while the germination rates of the AtOE lines were markedly higher than that of the WT plants (Fig. 3a). The germination rates of the AtOE lines, which were germinated 3 days after sowing at 0.5 μM ABA, were more than 75.0%, while the germination rates of the WT plants were approximately 31.2% (Fig. 3b). The germination rates after treating plants with 0.75 μM ABA were 37.5, 43.7, 52.5% for AtOE1, AtOE2, and AtOE3 plants, respectively, which were higher than that of the WT plants (18.7%) (Fig. 3b). For early root growth assay, under normal conditions, there were indistinguishable phenotypes between the WT plants and the AtOE lines (Fig. 3c). However, in the presence of 0.5 μM ABA or 0.75 μM ABA, the AtOE lines were more vigorous than the WT plants, forming longer primary roots (Fig. 3c).

Fig. 3
figure 3

PnF3H reduced the ABA sensitivity in transgenic Arabidopsis. a Seed germination in the AtOE lines were significantly higher than that in the WT plants under ABA treatment (3 days’ seed germination). b Statistical analysis of the seed germination shown in (a). c PnF3H enhanced the primary root growth after ABA treatment. d Statistical analysis of the root length shown in (c). e Endogenous ABA content of 2-week-old Arabidopsis seedlings. f The expression levels of ABA-responsive genes were determined by real-time PCR. Tublin was used as an internal control. Vertical bars are means ± SE, and asterisks (*) indicate significant differences of means between the AtOE lines and the WT plants at p < 0.05

Full size image

To clarify the molecular mechanism mediating the ABA effects, germinated seeds were sprayed with 50 μM ABA or deionized water (control samples) 2 days after germination. Real-time PCR analysis showed that the transcription levels of ABA-responsive genes such as AtABI3, AtABI4, AtABI5, and AtABF3 were substantially down-regulated in the AtOE lines when compared the WT plants, with respect to plants at 50 μM ABA treatment (Fig. 3f). Furthermore, by measuring the content of endogenous ABA in 2-week old plants, results showed that the ABA levels in the AtOE lines were lower than that in the WT plants (Fig. 3e). Thus, PnF3H inhibited the expression of ABA pathway which were related to the seed germination and seedling growth in transgenic plants.

Determination of PnF3H activity in transgenic plants

Plants were cultured in liquid MS medium in the absence or presence of 200 μM naringenin for 2 weeks respectively (Yun et al. 2008). Without naringenin treatment, there was no obvious difference in phenotypes between the AtOE plants and the WT plants (Fig. 4a). However, under 200 μM naringenin treatment, the root length of the AtOE lines were longer than that of the WT plants (Fig. 4a). Thus, overexpressing PnF3H could alleviate the inhibitive effects of naringenin on plant root growth. Moreover, flavonoids in the AtOE lines appeared new peaks under naringenin treatment when compared to the WT plants. PnF3H might play a key role in catalyzing naringenin to new products (Fig. 4 b and c, peaks P1 and P2).

Fig. 4
figure 4

Naringenin-feeding experiments and flavonoid metabolism profiling by HPLC analysis. a Surface-sterilized Arabidopsis seeds were cultured in liquid MS medium for 2 weeks with or without 200 μM naringenin (Sigma-Aldrich, USA). AtOE lines have growth superiority and alleviate the inhibition of naringenin on plant growth. b HPLC profile of total flavonoids of 2-week old Arabidopsis without 200 μM naringenin in liquid MS medium. c HPLC profile of total flavonoids of 2-week old Arabidopsis with 200 μM naringenin in liquid MS medium and 10 mM naringenin standard (Sigma-Aldrich, USA) monitored at 280 nm. d Hypothetical model of PnF3H function in abiotic stress

Full size image

Discussion

Flavonoids are a large family of plant phenolic pigments. Flavonoid biosynthesis pathway has been basically revealed through enzymologic studies in the model plant Arabidopsis thaliana (Saito et al. 2013). However, the secondary metabolic pathways in plants are extremely complex and each plant species has obviously distinctive profiles of flavonoids (Tu et al. 2016). Therefore, it is essential to identify species-specific genes in the flavonoid biosynthesis pathway. In this study, we carried out a sequence analysis of a flavanone 3-hydroxylase gene (i.e., PnF3H) from the Antarctic moss P. nutans and revealed its roles in abiotic stresses. Bioinformatics analysis suggested that PnF3H was highly homologous with F3Hs from other plants (Fig. 1a). F3H is classified under FeII and 2-ODD (Schofield and Zhang 1999). PnF3H also contained two conserved motifs: HxDxnH (His233, Asp235, and His289) for binding FeII and RxS (Arg299 and Ser301) for binding 2-OG (2-oxoglutarate). Consistent with the diverse physiological functions, flavonoids are found in most plant cell compartments, including the cytosol, vacuole, chloroplast, nucleus, endoplasmic reticulum and the extracellular space (Zhao and Dixon 2010). Subcellular localization analysis showed that PnF3H located in cytosol (Fig. 1i).

Flavonoid biosynthesis enzyme, particularly flavanone 3-hydroxylase have been recently suggested as playing antioxidant functions in the responses of plants to a wide range of abiotic stresses. In the desert plant, Reaumuria soongorica, real-time PCR analysis showed that there was a rapid increase in gene expression of RsF3H under UV-B radiation and drought stress (Liu et al. 2013a). F3H is considered to be an important enzyme of flavonoid pathway leading to accumulation of flavan-3-ols in tea. Expression analysis revealed the upregulation of F3H transcripts (i.e., CsF3H) under salt stress (Mahajan and Yadav 2014). As show in Fig. 1, the transcript levels of PnF3H were also significantly increased under various treatments (i.e., high salinity, cold, 20% PEG6000, UV-B radiation or ABA).

Salt stress is one of the major abiotic factors (Hediye Sekmen et al. 2007). Flavonoids are involved in the restriction of salt-induced oxidative damages (Hamada AbdElgawad et al. 2016). For example, over-expression of a tea flavanone 3-hydroxylase gene (i.e., CsF3H) increased the content of flavan-3-ols and enhanced salt tolerance in transgenic tobacco (Mahajan and Yadav 2014). High salinity can also affect the growth of vegetation in Antarctica, as many species grow near the Antarctic coast. In this study, we found that overexpression of PnF3H improved plant tolerance to salinity stress with higher seeds germination rates and longer primary roots (Fig. 2). P5CS and HKT1 are salt tolerance genes (Zhu 2001). P5CS encodes ∆1-pyrroline-5-carboxylate synthetase, under the control of a stress-induced promoter AIPC (ABA-inducible promoter complex), confers tolerance to salt stress (Guerzoni et al. 2014). HKT1 encodes high-affinity K+ transporter that facilitates Na+ exclusion from root xylem. HKT1 and its functional variation are highly associated with salt tolerance (Mickelbart et al. 2015). In this study, the expression levels of AtP5CS1 and AtHKT1 were markedly higher in the AtOE lines than that in the WT plants (Fig. 2).

Abiotic stress usually leaded to accumulation of reactive oxygen species (ROS) and oxidative damage, plants resisted abiotic stress by scavenging the generation of reactive oxygen species (Liu et al. 2013a). Overexpression of anthocyanidin reductase from rosa rugosa (i.e., RrANR) in tobacco increased transcript expression and activity of several ROS-related and stress-responsive genes under chilling stress (Luo et al. 2016). The overexpression of a flavanone 3-hydroxylase from Camellia sinensis (i.e., CsF3H) enhanced the salt stress tolerance of the transgenic tobacco, and the result was accompanied with higher transcript expression level and activity of the antioxidant enzymes (Mahajan and Yadav 2014). In this study, the AtOE lines showed an obviously enhanced oxidative tolerance when compared to the WT plants (Fig. 3b). The result of DAB staining showed that the transgenic Arabidopsis plants had a lower H2O2 level (Fig. 2e). The AtOE lines also enhanced the gene expression levels and enzyme activities of ROS-scavenging enzymes under salt stress (Fig. 2f–h). Thus, PnF3H can contribute to reduce ROS level under abiotic stress.

Phytohormone ABA has been proposed to act as a pivotal regulator in regulating plant processes to resisting abiotic stresses, seed maturation, seed dormancy and plant development (Keskin et al. 2010). In tobacco, overexpression of RrANR enhances tolerance to abiotic stress via modulating ABA signaling (Luo et al. 2016). In this study, PnF3H transgenic Arabidopsis exhibited insensitivity to ABA in terms of seed germination and early root growth (Fig. 3a, c). Furthermore, the levels of endogenous ABA in the AtOE lines were also lower than that in the WT plants (Fig. 3e). Previously, molecular studies suggested that ABA signaling pathway might be conserved across land plant evolution (from mosses to higher plants) (Chater et al. 2013). For example, PpABI3 regulated ABA-responsible gene expression by interacting with the nuclear factor PpNF-Y in P. patens (Yotsui et al. 2013). In this study, the expression levels of genes (AtABI3, AtABI4, AtABI5 and AtABF3) were decreased in transgenic Arabidopsis plants compared with the WT plants (Fig. 3f), which may be explained the reduced ABA sensitivity in transgenic Arabidopsis. Overall, these results suggested that PnF3H may serve as a flavonoid biosynthetic enzyme involving in regulating the salt tolerance and ABA sensitivity in P. nutans.

Naringenin is a precursor compound of flavonoids and the key substrate for flavonoids metabolism pathway (Yun et al. 2008; Du et al. 2010). Naringenin is taken up by Arabidopsis roots and can be catalyzed by flavanone-3-hydroxylase (F3H) into new flavonoids (Shirley et al. 1995). 1-N-naphthylphthalamic acid (NPA) is an auxin transport inhibitor. Naringenin is similar to NPA and can also inhibit auxin transport and then Arabidopsis root growth and gravitropism were inhibited (Brown et al. 2001; Yun et al. 2008). Overexpression flavone synthase (FNS-I) from Petroselinum crispum have no significant differences with the WT Arabidopsis. But, under naringenin-feeding condition, the FNS-I transgenic lines could accumulate substantial amounts of apigenin and alter the synthesis of flavones metabolize in transgenic Arabidopsis. In this study, naringenin-feeding experiment showed that AtOE lines also can alleviate the inhibitive effects of naringenin on plant growth and change flavonoid metabolism spectrum (Fig. 4). These results indicate that PnF3H has catalytic function in Arabidopsis and converts naringenin into new products.

Based on overall observations, the roles of PnF3H in enhancing salt stress tolerance and decreasing ABA sensitivity were proposed in Fig. 4d. Our proposition is that PnF3H down-regulates ABA signaling and further regulates the ABA-related stress responsive pathway and increases the level of ROS-scavenging activity, thereby explaining the positive effect of PnF3H on the salinity tolerance of Arabidopsis.