MfPIF1 of Resurrection Plant Myrothamnus flabellifolia Plays a Positive Regulatory Role in Responding to Drought and Salinity Stresses in Arabidopsis.

Phytochrome-interacting factors (PIFs), a subfamily of basic helix-loop-helix (bHLH) transcription factors (TFs), play critical roles in regulating plant growth and development. The resurrection plant Myrothamnus flabellifolia possesses a noteworthy tolerance to desiccation, but no PIFs related to the response to abiotic stress have been functionally studied. In this study, a dehydration-inducible PIF gene, MfPIF1, was cloned and characterized. Subcellular localization assay revealed that MfPIF1 is localized predominantly in the nucleus. Overexpression of MfPIF1 in Arabidopsis thaliana led to enhanced drought and salinity tolerance, which was attributed to higher contents of chlorophyll, proline (Pro), soluble protein, and soluble sugar, activities of antioxidant enzymes as well as lower water loss rate, malondialdehyde (MDA) content, and reactive oxygen species (ROS) accumulation in transgenic lines compared with control plants. Moreover, MfPIF1 decreased stomatal aperture after drought and abscisic acid (ABA) treatment, and increased expression of both ABA biosynthesis and ABA-responsive genes including NCED3, P5CS, and RD29A. Overall, these results indicated that MfPIF1 may act as a positive regulator to drought and salinity responses, and therefore could be considered as a potential gene for plant genetic improvement of drought and salinity tolerance.


Introduction
Various abiotic stress responses are induced by the signal of environmental stresses, for instance, drought, salinity, and high irradiance when plants suffer from adverse circumstances. These responses are controlled by a wide range of sophisticated mechanisms. The interactions of many mechanisms maintain membrane stability, balance osmotic pressure, and reduce damages of active oxygen [1,2], or participate in signal transduction and transcriptional regulation [3], which are partially induced by gibberellin (GA), abscisic acid (ABA), ethylene (ET), jasmonates acid (JA), and salicylic acid (SA) [4,5]. The WRKY, ZFP, bHLH, MYB, NAC, bZIP and DREB transcription factors (TFs) from various families Resurrection plants can keep alive without suffering permanent injury, even if their vegetative organs confine in massive dehydration, and the ability to tolerate the nearly complete desiccation state of their vegetative organs is called desiccation tolerance (DT) [37]. Myrothamnus flabellifolia Welw. (Myrothamnaceae), a short shrub from southern Africa, is known as the only wooden resurrection plant for its DT features [38]. To survive in the extremely dry mountain environment, the unique fan leaves of M. flabellifolius can fold and roll up tightly when plant tissue is dehydrated, which makes the plants turn quickly into a long-term desiccant state and rehydrate rapidly after contact with water [39,40]. In the meantime, the primary problem faced by photosynthesis under drought stress is that photosynthesis agencies can provide the possibility of generating toxic reactive oxygen species (ROS) [41]. Resurrected plants inhibit the production of ROS by reducing the interaction of light and chlorophyll, and they can also be removed by antioxidants. As a homoiochlorophyllous resurrection plant, M. flabellifolia can maintain low-intensity photosynthesis without destroying the photosynthetic mechanism, thereby resisting severe drought stress [42]. Despite a number of physiological and biochemical subjects in M. flabellifolia being studied, the molecular mechanisms of extreme tolerance to desiccation and the ability to revitalize still remain unknown. A previous study showed that numerous diverse TFs were involved in desiccation tolerance in M. flabellifolia through transcriptome sequencing techniques, among them MfPIF1, which is upregulated immediately at the early stage of dehydration [43]. However, its functions involved in dehydration response have not been further investigated. In the present work, we report that the heterologous overexpression of MfPIF1 enhances both drought and salinity stress tolerance and ABA sensitivity in Arabidopsis. Considering all these findings, we propose that MfPIF1 plays a positive regulatory role in resisting drought and salinity stresses in transgenic plants, which may give a reference to molecular breeding to endow plants with tolerance to abiotic stress.

Isolation and Characterization of MfPIF1
The cDNA sequence of MfPIF1 was cloned from M. flabellifolia by PCR amplification. The length of the obtained nucleotide sequence is 1188 bp, which possesses an open reading frame (ORF) encoding 395 amino acids. The protein has a calculated isoelectric point of 5.40 and a predicted molecular mass of 43.76 kDa. A putative bipartite nuclear localization sequence (NLS) "KRSRAAEVHNLSERRRR" at 300 aa was found in the MfPIF1 protein ( Figure 1a). SMART analysis demonstrated that MfPIF1 contains a typical bHLH domain. Moreover, multiple sequence alignment between MfPIF1 and five homologous bHLH proteins indicated that MfPIF1 has an APB domain at its N-terminus and a basic region followed by an HLH domain, and these two domains of MfPIF1 had high consistence with those from other plant species. Notably, we also found a putative APA domain in MfPIF1. Unlike the PIF1s of other plants, a conserved amino acid residue at position 169 was changed from asparagine to serine (N to S), which is unique for M. flabellifolia among the sequences analyzed ( Figure 1a). The subsequent phylogenetic analysis revealed that the MfPIF1 was most humongous to grape VvPIF1 which were grouped to a monophyletic clade (Figure 1b). Active phytochrome B-binding (APB) and active phytochrome A-binding (APA) domains are marked as bold lines, conserved amino acid residues are marked by dots, and the altered amino acid residue at position 169 is indicated by an arrow. The basic region is indicated by a white box, and helix-loophelix (HLH) conserved domain is labeled with black boxes linked by a curve. (b) Phylogenetic tree constructed using neighbor-joining method. MfPIF1 is indicated by a triangle. The GenBank accession numbers are listed in Supplementary Table S1.

MfPIF1 is Localized in the Nucleus of Cells
To confirm the above conjecture, we transfected the 35S::MfPIF1-YFP into tobacco leaf epidermal cells for instantaneous expression. Scanning confocal microscopic analysis showed that the fluorescence was detected in the whole cell in 35S::YFP control. On the contrary, the intense yellow fluorescence was nearly exclusively observed in nucleus of 35S::MfPIF1-YFP transformed cell, which proved that MfPIF1 is located in the nucleus ( Figure 2). . Identical and similar amino acids are shaded in black and gray, respectively. The putative NLS is marked by a dashed line. Active phytochrome B-binding (APB) and active phytochrome A-binding (APA) domains are marked as bold lines, conserved amino acid residues are marked by dots, and the altered amino acid residue at position 169 is indicated by an arrow. The basic region is indicated by a white box, and helix-loop-helix (HLH) conserved domain is labeled with black boxes linked by a curve. (b) Phylogenetic tree constructed using neighbor-joining method. MfPIF1 is indicated by a triangle. The GenBank accession numbers are listed in Supplementary Table S1.

MfPIF1 is Localized in the Nucleus of Cells
To confirm the above conjecture, we transfected the 35S::MfPIF1-YFP into tobacco leaf epidermal cells for instantaneous expression. Scanning confocal microscopic analysis showed that the fluorescence was detected in the whole cell in 35S::YFP control. On the contrary, the intense yellow fluorescence was nearly exclusively observed in nucleus of 35S::MfPIF1-YFP transformed cell, which proved that MfPIF1 is located in the nucleus ( Figure 2).

Overexpressing MfPIF1 Enhanced Tolerance to Drought and Salt
To explore the potential function of MfPIF1 in responding to abiotic stress, heterogeneous expression of the MfPIF1 gene in Arabidopsis was performed by constructing a binary vector pGSA1403-MfPIF1. T1 transgenic Arabidopsis lines were acquired from kanamycin resistance screening, and the T3 transgenic lines, Line D, Line N, and Line Q, were propagated subsequently. The qRT-PCR results confirmed that expression level of MfPIF1 could not be detected in wild type (WT), but it was found in all three transgenic lines investigated, with the Line N exhibiting significantly higher expression than the other two lines (Figure 3a). This experiment showed that MfPIF1 was overexpressed in the measured transgenic Arabidopsis, and these three transgenic lines can be selected for subsequent experiments.

Overexpressing MfPIF1 Enhanced Tolerance to Drought and Salt
To explore the potential function of MfPIF1 in responding to abiotic stress, heterogeneous expression of the MfPIF1 gene in Arabidopsis was performed by constructing a binary vector pGSA1403-MfPIF1. T 1 transgenic Arabidopsis lines were acquired from kanamycin resistance screening, and the T 3 transgenic lines, Line D, Line N, and Line Q, were propagated subsequently. The qRT-PCR results confirmed that expression level of MfPIF1 could not be detected in wild type (WT), but it was found in all three transgenic lines investigated, with the Line N exhibiting significantly higher expression than the other two lines (Figure 3a). This experiment showed that MfPIF1 was overexpressed in the measured transgenic Arabidopsis, and these three transgenic lines can be selected for subsequent experiments.
In order to investigate if MfPIF1 is related to drought and salinity stress tolerance, WT and three T 3 transgenic lines were exposed to the corresponding stress treatments at the seedling stage and adult stage. At seedling stage, no obvious difference between the transgenic and WT plants could be observed when growing in routine conditions ( Figure 3b). After mannitol and salt treatments, significantly longer roots were observed in transgenic lines. This difference was more pronounced under treatments with rather lower concentrations of mannitol (200 mM) and NaCl (50 mM) (Figure 3b-d). In a coordinated manner, a clearly large leaf area was also observed in transgenic lines ( Figure 3b). screening, and the T3 transgenic lines, Line D, Line N, and Line Q, were propagated subsequently. The qRT-PCR results confirmed that expression level of MfPIF1 could not be detected in wild type (WT), but it was found in all three transgenic lines investigated, with the Line N exhibiting significantly higher expression than the other two lines (Figure 3a). This experiment showed that MfPIF1 was overexpressed in the measured transgenic Arabidopsis, and these three transgenic lines can be selected for subsequent experiments. For adult stage test, the transgenic and WT plants growing for 4 weeks were cultivated in soil upon drought or 300 mM NaCl treatment to further investigate drought and salt tolerance. No obvious morphological differences were found between the transgenic and WT plants in normal environment, drought for five days, and salt for three days (Figure 4). Ten days after withholding (DAW) watering, slightly withered leaves could be found and the leaf chlorophyll content of MfPIF1 transgenic lines was 1.34-1.40 times higher than that of WT (Figure 4a  Negative effects of salinity stress on plant growth became visible after seven days with NaCl treatment (Figure 4b). In the meantime, the leaf chlorophyll content of MfPIF1 transgenic lines was 1.21-1.31 times higher than that of WT leaves after treatment with salt for 7 days (Figure 4c).When plants were exposed to salinity stress for 12 days, more serious salt injuries were observed on WT plants, whereas transgenic plants were affected to a slight extent of etiolated and wilting symptoms. After 18 days, almost all leaves of WT were withered; however, a considerable portion of leaves of the transgenic plants stayed green and all three lines flowered ( Figure 4b).
We measured the water loss rate of rosette leaves at different time points of dehydration. According to Figure 4d, the water loss rates of transgenic plants were remarkably lower than WT in all time points except for 0 h. This result indicated that MfPIF1 transgenic plants probably had enhanced water-retaining capacity and lost water much more slowly than the WT. Malondialdehyde (MDA), a significant indicator of membrane-lipid peroxidation, can lead to severe damage for the cell membranes. Though the MDA content was elevated in either the WT or transgenic plants under drought and salt treatments, the contents of MDA were significantly lower in all three transgenic lines than WT plants (Figure 4e). Some osmotic adjustment substances can prevent plant cells from dehydrating and improve the tolerance to environmental stress. Thus, we compared contents of Negative effects of salinity stress on plant growth became visible after seven days with NaCl treatment (Figure 4b). In the meantime, the leaf chlorophyll content of MfPIF1 transgenic lines was 1.21-1.31 times higher than that of WT leaves after treatment with salt for 7 days (Figure 4c).When plants were exposed to salinity stress for 12 days, more serious salt injuries were observed on WT plants, whereas transgenic plants were affected to a slight extent of etiolated and wilting symptoms. After 18 days, almost all leaves of WT were withered; however, a considerable portion of leaves of the transgenic plants stayed green and all three lines flowered ( Figure 4b).
We measured the water loss rate of rosette leaves at different time points of dehydration. According to Figure 4d, the water loss rates of transgenic plants were remarkably lower than WT in all time points except for 0 h. This result indicated that MfPIF1 transgenic plants probably had enhanced water-retaining capacity and lost water much more slowly than the WT. Malondialdehyde (MDA), a significant indicator of membrane-lipid peroxidation, can lead to severe damage for the cell membranes. Though the MDA content was elevated in either the WT or transgenic plants under drought and salt treatments, the contents of MDA were significantly lower in all three transgenic lines than WT plants ( Figure 4e). Some osmotic adjustment substances can prevent plant cells from dehydrating and improve the tolerance to environmental stress. Thus, we compared contents of several osmolytes, proline, soluble protein and soluble sugar, among three transgenic lines and WT plants with or without drought and salt stresses. As shown in Figure 4f-h, WT and transgenic lines exhibited similar contents of all three osmolytes before treatment. Both stresses increased accumulations of osmolytes in WT and the transgenic lines. Nevertheless, in comparison to WT, all transgenic lines presented remarkably higher contents of proline, as well as soluble protein and soluble sugar.

Effect of MfPIF1 Overexpression on Antioxidant Metabolism in Arabidopsis under Drought and Salinity Stresses
As is well known, the cellular oxidative damage aggravates with the level of lipid peroxide rising when plants suffered from abiotic stresses, which is evoked by a large excess of reactive oxidative species (ROS), for instance, hydrogen peroxide (H 2 O 2 ) and superoxide anion radical (O 2 − ), existing in plant cells. Therefore, we used the histochemical staining of DAB and NBT to detect cellular ROS levels under drought and salinity stresses. As shown in Figure 5, the larger leaf areas of WT plants were stained in deeper color than those of transgenic lines (Figure 5a,b), indicating that the transgenic lines underwent less cellular oxidative damage under both stresses. In accordance with these results, less H 2 O 2 content was detected in three transgenic lines after the drought and salt treatments, which also showed higher anti-superoxide anion activity (Figure 5c,d). several osmolytes, proline, soluble protein and soluble sugar, among three transgenic lines and WT plants with or without drought and salt stresses. As shown in Figure 4 (f-h), WT and transgenic lines exhibited similar contents of all three osmolytes before treatment. Both stresses increased accumulations of osmolytes in WT and the transgenic lines. Nevertheless, in comparison to WT, all transgenic lines presented remarkably higher contents of proline, as well as soluble protein and soluble sugar.

Effect of MfPIF1 Overexpression on Antioxidant Metabolism in Arabidopsis under Drought and Salinity Stresses
As is well known, the cellular oxidative damage aggravates with the level of lipid peroxide rising when plants suffered from abiotic stresses, which is evoked by a large excess of reactive oxidative species (ROS), for instance, hydrogen peroxide (H2O2) and superoxide anion radical (O2 − ), existing in plant cells. Therefore, we used the histochemical staining of DAB and NBT to detect cellular ROS levels under drought and salinity stresses. As shown in Figure 5, the larger leaf areas of WT plants were stained in deeper color than those of transgenic lines (Figure 5a,b), indicating that the transgenic lines underwent less cellular oxidative damage under both stresses. In accordance with these results, less H2O2 content was detected in three transgenic lines after the drought and salt treatments, which also showed higher anti-superoxide anion activity (Figure 5c,d). Moreover, we measured antioxidant enzyme activities such as superoxide dismutase (SOD), peroxidase (POD), and catalase (CAT), which are considered as key enzymes in ROS scavenging and are of great relevance to drought and salt tolerance. Consistent with results of ROS level measurement, the activities of SOD, POD, and CAT were significantly enhanced in WT and the transgenic plants upon drought and salt treatments. However, transgenic plants presented apparent stronger activities than those of WT (Figure 5e-g). These results combined proved that overexpression of MfPIF1 increased capacities of scavenging ROS under stressful conditions, and hence decreased cellular oxidative damage. Moreover, we measured antioxidant enzyme activities such as superoxide dismutase (SOD), peroxidase (POD), and catalase (CAT), which are considered as key enzymes in ROS scavenging and are of great relevance to drought and salt tolerance. Consistent with results of ROS level measurement, the activities of SOD, POD, and CAT were significantly enhanced in WT and the transgenic plants upon drought and salt treatments. However, transgenic plants presented apparent stronger activities than those of WT (Figure 5e-g). These results combined proved that overexpression of MfPIF1 increased capacities of scavenging ROS under stressful conditions, and hence decreased cellular oxidative damage.

MfPIF1 Overexpression Promoted Stomatal Closure Induced by Drought and ABA
The response of stomatal movement mediated by ABA plays a central role in transpiration upon drought stress. Hence, we assessed the stomatal closure of leaves under treatments of 300 mM mannitol and 20 µM ABA. Under normal conditions, most of the stomata were opening in all the plants (Figure 6a), and the ratios of stomatal aperture were not significantly different between transgenic and WT plants (Figure 6b). After treatment by mannitol and ABA, the stomatal apertures of three transgenic lines were reduced to 0.25-0.26 and 0.10-0.12, respectively, which were remarkably lower than those of WT plants (0.32 and 0.21) (Figure 6b). These results indicate that the expression of MfPIF1 promotes stomatal closure in response to mannitol and ABA, which perhaps contributed to the reduced transpiration and decreased water loss rate.

MfPIF1 Overexpression Promoted Stomatal Closure Induced by Drought and ABA
The response of stomatal movement mediated by ABA plays a central role in transpiration upon drought stress. Hence, we assessed the stomatal closure of leaves under treatments of 300 mM mannitol and 20 µM ABA. Under normal conditions, most of the stomata were opening in all the plants (Figure 6a), and the ratios of stomatal aperture were not significantly different between transgenic and WT plants (Figure 6b). After treatment by mannitol and ABA, the stomatal apertures of three transgenic lines were reduced to 0.25-0.26 and 0.10-0.12, respectively, which were remarkably lower than those of WT plants (0.32 and 0.21) (Figure 6b). These results indicate that the expression of MfPIF1 promotes stomatal closure in response to mannitol and ABA, which perhaps contributed to the reduced transpiration and decreased water loss rate.

Overexpression of MfPIF1 Up-Regulates Expression Levels of ABA-Responsive Genes
To further explore the potential molecular mechanisms for enhanced drought and salinity tolerance in MfPIF1-overexpressing lines, we measured the expression quantity of NCED3, P5CS, and RD29A in Arabidopsis plants by using qRT-PCR, for one day and four days of artificially simulated drought treatment (10% PEG-6000), or salt treatment (300 mM NaCl). NCED3 is related to ABA biosynthesis, and all three genes are responsive to ABA, drought, and salinity stresses. Additionally, P5CS is involved in proline biosynthesis. As shown in Figure 7, the similar transcription levels of NCED3 were observed between WT and transgenic Arabidopsis under normal conditions. After treatments, the transcription level in WT rose slightly, whereas those in all three transgenic lines increased more rapidly and greatly than the WT, in which Line D responded to salinity stress more slowly than another two transgenic lines (Figure 7a). This might be partially due to the lower gene expression of MfPIF1 in Line D (Figure 3a). There were higher expression levels of P5CS and RD29A in transgenic lines than those in WT before drought and salt treatments. Being exposed to drought treatment, the transcription levels of P5CS and RD29A in MfPIF1-overexpressing lines were notably upregulated while slightly increased in WT plants. Under salinity stress, expression levels of P5CS and RD29A exhibited similar tendency with those observed in NCED3. These results suggested that MfPIF1 positively regulated ABA-responsive gene expression in Arabidopsis.

Overexpression of MfPIF1 Up-Regulates Expression Levels of ABA-Responsive Genes
To further explore the potential molecular mechanisms for enhanced drought and salinity tolerance in MfPIF1-overexpressing lines, we measured the expression quantity of NCED3, P5CS, and RD29A in Arabidopsis plants by using qRT-PCR, for one day and four days of artificially simulated drought treatment (10% PEG-6000), or salt treatment (300 mM NaCl). NCED3 is related to ABA biosynthesis, and all three genes are responsive to ABA, drought, and salinity stresses. Additionally, P5CS is involved in proline biosynthesis. As shown in Figure 7, the similar transcription levels of NCED3 were observed between WT and transgenic Arabidopsis under normal conditions. After treatments, the transcription level in WT rose slightly, whereas those in all three transgenic lines increased more rapidly and greatly than the WT, in which Line D responded to salinity stress more slowly than another two transgenic lines (Figure 7a). This might be partially due to the lower gene expression of MfPIF1 in Line D (Figure 3a). There were higher expression levels of P5CS and RD29A in transgenic lines than those in WT before drought and salt treatments. Being exposed to drought treatment, the transcription levels of P5CS and RD29A in MfPIF1-overexpressing lines were notably upregulated while slightly increased in WT plants. Under salinity stress, expression levels of P5CS and RD29A exhibited similar tendency with those observed in NCED3. These results suggested that MfPIF1 positively regulated ABA-responsive gene expression in Arabidopsis.

Discussion
PIFs promote crosstalk among multiple different transcriptional pathways [44]. In recent years, an increasing amount of researches have identified PIFs as key members in the transcriptional pathways underlying abiotic stress, besides the well-known pathways in response to light. The current research reported the characterization of a PIF gene MfPIF1 in M. flabellifolius. Sequence analysis suggested that MfPIF1 contains a conserved bHLH domain, a critical APB domain, and a predicted APA domain (Figure 1a) and showed the highest homology to grape VvPIF1 (Figure 1b). It is localized in the nucleus, suggesting that it may function as a transcription factor to play a basic role in transcription activation or repression.
To investigate whether MfPIF1 works in the regulation of abiotic stress response, transgenic Arabidopsis plants overexpressing MfPIF1 were generated. According to our results, MfPIF1 transgenic plants showed better growth status upon drought or salt treatments (Figures 3 and 4), demonstrating that MfPIF1 acts as a positive regulator to the drought and salinity response in Arabidopsis. Plants adapt to adverse environments through morphological, physiological, metabolic, and molecular alterations [45]. Development of a deep root system is a water deficit adaptation strategy that allows the plants to obtain water from the soil to meet the requirement of transpiration [46,47]. In this research, longer primary roots were observed in transgenic seedlings, which may contribute to better water absorption capacity (Figure 3b). Decreased chlorophyll content in droughtstressed or salt-stressed plants is attributed to the inhibition of chlorophyll synthesis [48,49]. MDA as an indicator of lipid peroxidation caused by reactive oxygen species (ROS) could be used for investigating drought and salinity stress tolerance in plants [50]. Hence, the higher chlorophyll content and lower MDA content of MfPIF1 transgenic plants after stress treatment further confirmed the enhancement of drought and salinity tolerance (Figure 4c,d).
It is well accepted that drought and salinity, commonly presented as osmotic stress, can exert serious injuries in plants. To minimize the damages from osmotic stress, proline, soluble protein, and soluble sugar contents can increase and improve the water holding capacity of leaves by facilitating osmoregulation [51][52][53]. Our studies showed that MfPIF1 transgenic lines improved the contents of proline, soluble protein, and soluble sugars under drought or salt conditions (Figure 4f-h) which proved that MfPIF1 may contribute to accumulation of osmolytes to regulate their osmotic potential.

Discussion
PIFs promote crosstalk among multiple different transcriptional pathways [44]. In recent years, an increasing amount of researches have identified PIFs as key members in the transcriptional pathways underlying abiotic stress, besides the well-known pathways in response to light. The current research reported the characterization of a PIF gene MfPIF1 in M. flabellifolius. Sequence analysis suggested that MfPIF1 contains a conserved bHLH domain, a critical APB domain, and a predicted APA domain ( Figure 1a) and showed the highest homology to grape VvPIF1 (Figure 1b). It is localized in the nucleus, suggesting that it may function as a transcription factor to play a basic role in transcription activation or repression.
To investigate whether MfPIF1 works in the regulation of abiotic stress response, transgenic Arabidopsis plants overexpressing MfPIF1 were generated. According to our results, MfPIF1 transgenic plants showed better growth status upon drought or salt treatments (Figures 3 and 4), demonstrating that MfPIF1 acts as a positive regulator to the drought and salinity response in Arabidopsis. Plants adapt to adverse environments through morphological, physiological, metabolic, and molecular alterations [45]. Development of a deep root system is a water deficit adaptation strategy that allows the plants to obtain water from the soil to meet the requirement of transpiration [46,47]. In this research, longer primary roots were observed in transgenic seedlings, which may contribute to better water absorption capacity (Figure 3b). Decreased chlorophyll content in drought-stressed or salt-stressed plants is attributed to the inhibition of chlorophyll synthesis [48,49]. MDA as an indicator of lipid peroxidation caused by reactive oxygen species (ROS) could be used for investigating drought and salinity stress tolerance in plants [50]. Hence, the higher chlorophyll content and lower MDA content of MfPIF1 transgenic plants after stress treatment further confirmed the enhancement of drought and salinity tolerance (Figure 4c,d).
It is well accepted that drought and salinity, commonly presented as osmotic stress, can exert serious injuries in plants. To minimize the damages from osmotic stress, proline, soluble protein, and soluble sugar contents can increase and improve the water holding capacity of leaves by facilitating osmoregulation [51][52][53]. Our studies showed that MfPIF1 transgenic lines improved the contents of proline, soluble protein, and soluble sugars under drought or salt conditions (Figure 4f-h) which proved that MfPIF1 may contribute to accumulation of osmolytes to regulate their osmotic potential.
Metabolism processes in plant cells are susceptible to the negative effects of drought and salinity stress conditions. Unbalanced metabolism will lead to oxidation stress in cells by both forcing ROS generation and accelerating ROS accumulation, such as H 2 O 2 and O 2 − , leading to oxidation of cell components, affecting normal metabolism activities and destroying organelle integrity [54,55]. From our results, the WT plants showed fairly higher degree of DAB and NBT staining than MfPIF1 transgenic lines upon stress treatment (Figure 5a,b). In accordance with these results, the remarkably lower content of H 2 O 2 and higher anti-superoxide anion activity were found in transgenic lines (Figure 5c,d). Plants have developed efficient ROS scavenging systems by increasing various antioxidation enzyme activities, for instance, SOD, POD, and CAT [56,57]. In our study, all three measured antioxidation enzymes exhibited significantly higher activities in transgenic lines when exposed to drought and salt conditions than those in WT (Figure 5e-g), suggesting that overexpression of MfPIF1 resulted in a strengthened ROS scavenging system and prevented plants from the severe stress-induced oxidative damage. Phytohormone ABA plays a complicated role in modulating multiple biochemical processes including seed germination, plant growing and development, along with biotic and abiotic stress responses. Under drought and salt adverse circumstances, plants could promote ABA biosynthesis transferring signals in several complicated defense mechanisms, in which the more representative are stomatal movement and relative gene expression [58,59]. As shown in Figures 4d and 6a,b, transgenic lines presented decreased water loss rates and enhanced stomatal closure induced by drought and exogenous ABA. These results indicated that heterologous expression of MfPIF1 plants increased efficiency of stomata closure to reduce water loss through transpiration under water-limiting conditions.
There are two main pathways regulating the expression of penetration stress-response genes: one is ABA-dependent, and another is an ABA-independent mechanism; meanwhile, crosstalk also occurs between both pathways [60,61]. It has been reported that the increase of ABA content of M. flabellifolia during the dehydration process confirmed that ABA can participate in the defense reaction of resurrection plants [62]. Increasing experiments illustrated that PIFs are involved in drought or salinity stress [24,63], among which several transcription networks in abiotic stress responses are regulated by ABA, and these regulatory networks partially overlap with light signaling [25][26][27]. Our data provided evidence that MfPIF1 increased sensitivity to stomatal movement induced by ABA, indicating that it may function in stress-response regulation through ABA-dependent pathways. Long-term cell dehydration will cause serious damage to plant cells, which is closely related to the oxidative damage caused by the accumulation of ROS during the dehydration process of M. flabellifolia. Especially under strong solar radiation, the ROS homeostasis is extremely susceptible. M. flabellifolia retains most of the chlorophyll and maintains the integrity of thylakoid membranes during dehydration, and even if definite photosynthesis is kept, it can still minimize the damage caused by photooxidation and cope with dehydration-induced ROS damage by mobilizing the free radical scavenging system, as well as activating antioxidant enzymes and protective metabolites [64]. In addition, resurrection plants accumulate osmolytes including various low molecular weight proteins, sugars, and compatible solutes, for instance, trehalose and sucrose are considered as important osmoregulators of membrane and cytoplasm [65]. The latest research indicates that homoiochlorophyllous plants Craterostigma pumilum, Selaginella tamariscina, and Boea hygrometrica all show the same morphological and physiological adaptation characteristics during dehydration and hydration [66][67][68][69]. Based on these physiological analysis of the DT adaptation characteristics of resurrection plants, we speculate that if these resurrection plants became adapted to high light intensity environments that are also prone to drought/salinity, these two regulatory networks may become connected by having MfPIF1 upstream or downstream of ABA signaling pathways.
Rate-limiting enzyme gene NCED3 regulates ABA biosynthesis under stress conditions [70]. In the present research, NCED3 expressive level was remarkably higher in three transgenic lines after exposure to drought and salinity stress (Figure 7a), suggesting that more endogenous ABA might be synthesized in transgenic plants under abiotic stress. Furthermore, two typical genes responding to ABA-dependent abiotic stress, P5CS, and RD29A [71,72], are significantly upregulated in transgenic Arabidopsis upon drought and salinity stresses, and some transgenic lines have higher expression levels even under normal conditions (Figure 7b,c). As a significant response to abiotic stress, plants can increase proline content to maintain the osmotic balance. P5CS plays a distinct role in the control of proline biosynthesis and contributes to proline accumulation during abiotic stress [73,74]. RD29A expression can be affected by drought and salinity stresses via ABA-dependent signal transduction pathways, which might be due to the existence of ABRE and DRE motifs in its promoter [75]. Lyall et al. performed gene expression analysis in the monocot resurrection plant Xerophya humilis during vegetative desiccation, and speculated that the expression of the ABI3 regulator in leaf tissues is activated by a desiccation-responsive pathway. Based on the above research, we hypothesize that MfPIF1 may respond to drought stress by directly activating these ABRE-binding factors [76]. Altogether, our results presented evidence that MfPIF1 enhanced tolerance of Arabidopsis to drought and salinity stresses by participating in ABA biosynthesis and ABA-dependent stress-responding pathway.

Plant Materials and Growth Conditions
The M. flabellifolia, provided by the Department of Plant Science, University of California, Davis, was grown in the greenhouse of Landscape Architecture Department, Sichuan Agricultural University, Chengdu, Sichuan, China. Experimental plants were grown at a neutrophilous day condition (12 h light/12 h dark) at 22 • C/18 • C, 60% relative air humidity and sufficient light in potter pots.
Wild type (WT) Arabidopsis ecotype Columbia and transgenic lines were cultivated in a mixed media of soil and vermiculite (1:1, v/v) in plastic pots under 75% relative humidity and long day (16 h light/8 h dark) treatment at 24 • C/22 • C for 4 weeks before treatments. Seeds of transgenic Arabidopsis and WT were sterilized by a 1:1 diluted bleach solution for 5 min, then washed three times using sterilized deionized water. Seeds were subsequently placed on 1/2-strength Murashige and Skoog (MS) medium with 0.7% (w/v) agar and 2% (w/v) sucrose and adjusted pH to 5.8-6.0. After being vernalized at 4 • C for two days and growing in an illuminating incubator for about 10 days, young seedlings were transplanted to pots in a growth chamber.

Cloning and Bioinformatic Analysis of MfPIF1
General RNA was isolated from M. flabellifolia leaves using Plant Total RNA Isolation Kit (TINAGENE Co., Beijing, China). Synthesis of the first strand cDNA was then prepared with Reverse Transcriptase M-MLV (RNaseH-) (Takara Bio, Dalian, China) with the instruction of kits. The coding DNA sequence (CDS) of MfPIF1 was amplified by Phanta Max Super-Fidelity DNA Polymerase (Vazyme Biotech Co., Nanjing, China) with a pair of special primers with SmaI or SpeI restriction site (Supplementary Table S2). Purified PCR products were cloned to a pEasy-T1 Simple vector (TransGen Biotech, Beijing, China) and the construct of the pEasy-T1-MfPIF1 transfer vector plasmid was transformed into strain DH5α E. coli, and confirmed by TsingKe Biotech Co., Beijing, China.
The ORF of the nucleic acid sequence was extrapolated based on NCBI ORFfinder (https: //www.ncbi.nlm.nih.gov/orffinder/). Isoelectric point (pI) and molecular weight of the MfPIF1 protein were determined by ExPASy (https://web.expasy.org/compute_pi/). Conserved domains of the deduced protein sequence were analyzed with SMART (http://smart.embl-heidelberg.de/). Comparison of amino acid sequences of MfPIF1 with other close homologs from different species was done using NCBI software BLASTP (https://blast.ncbi.nlm.nih.gov/Blast.cgi). Multiple alignments between MfPIF1 and its homologous proteins were performed using DNAMAN v. 9.0 software. A phylogenetic tree was obtained through software MEGA v. 7.0 [77] based on the neighbor-joining approach, and the bootstrap test was replicated 1000 times. The prediction of protein secondary structure was studied with Jpred 4 (http://www.compbio.dundee.ac.uk/jpred/index.html).

Subcellular Localization of MfPIF1
The full-length ORF without the termination codons of MfPIF1 was amplified using primers with homologous arm sequences (Supplementary Table S2). The confirmed product was double-digested using HindIII and BamHI, followed by inserting into the pHB-YFP vector to construct a fusion expression vector MfPIF1-YFP driven by a CaMV (Cauliflower Mosaic virus) 35S promoter. Sequence-verified constructs 35S::MfPIF1-YFP and 35S::YFP were introduced into the Agrobacterium tumefaciens (GV3101) cells using the freezing-thawing method. Leaves of wild-type tobacco growing to 4 weeks old (Nicotiana benthamiana) were injected with suspensions including either the fusion construct or the control vector (YFP alone). All the transformed tobaccos were cultured at 22 • C in the dark for 16 h and then cultured for two days before observing the expressed of YFP by a laser confocal scanning microscope (Nikon, Tokyo, Japan).

Vector Construction and Generation of Transgenic Lines
To produce 35S::MfPIF1 lines, the encoding region of MfPIF1 was PCR-amplified with primers containing either a SmaI or SpeI restriction site (Supplementary Table S2), and the amplicon was linked into the same enzyme recognition sites of the plant binary vector pGSA-1403, resulting in the construct pGSA1403-MfPIF1 under the control of the CaMV 35S promoter. After that, the recombinant plasmid 35S::pGSA1403-MfPIF1 was transformed into the A. tumefaciens LBA4404, and Arabidopsis plants were transformed using the floral-dip transformation method [78]. The first generation T 0 seeds of MfPIF1 transgenic plants were collected and selected by 1/2 MS culture medium containing kanamycin (50 µg/mL). Kanamycin-resistance seedlings were transplanted to pots with soil for the following studies. Positive transgenic plants were detected with PCR with gene-specific primers as described above. Three homozygous T 3 positive lines were selected for further tolerance studies and other experiments.

Expression Analysis of MfPIF1 and ABA-Responsive Genes
Leaves from transgenic and WT seedlings growing to 4 weeks old, incubated under normal conditions, for one day and four days under simulating drought stress with PEG-6000 (10%) or salinity stress with 300 mM NaCl were used for expression analysis. General RNA from different lines was extracted with Plant RNA Kit (Omega Bio-tek, Norcross, GA, United States), which was also purified with RNase-Free DNase I (Omega Bio-tek) and reverse transcribed into cDNA by use of Uscript II (Innovagene biotech, Hunan, China).
The qRT-PCR amplification was performed in 25 µL reaction mixture (innovagene biotech) including 12.5 µL 2 × Taq SYBR Green qPCR Mix, 0.5 µL of 10 µM of each primer, 4 µL 5-fold diluted cDNA, and 7.5 µL of Nuclease-free H 2 O, which was performed using the real-time PCR instrument CFX Connect (Bio-Rad, Hercules, CA, USA). Amplification conditions were devised of 94 • C for 3 min, 42 cycles of degeneration at 94 • C for 8 s, annealing/extension at 60 • C for 60 s. The relative expression quantity of the target gene was evaluated based on the method of 2 −∆∆CT [79]. Finally, the results were normalized by an internal reference gene AtActin2 for quantitative analysis of relative genes. Each RT-qPCR experiment was reproduced at least three times. The gene-specific amplification primers were listed in Supplementary Table S2.

Assays of Drought and Salinity Stress Tolerance
For seedling stage stress assays, sterilized seeds of transgenic and WT plants were placed on 1/2 MS solid medium containing varying concentrations of mannitol (0, 200, 250 or 300 mM) and NaCl (0, 50, 100 or 150 mM). Culture dishes were settled vertically in the light incubator in a cycle for 16 h light (24 • C) and 8 h dark (22 • C). After growing 9 days, the taproot length of each line (15 seedlings every petri dish, and three dishes in total) was measured. Each experiment was performed in three replicates.
In order to explore the tolerance for mature plants to drought and salinity stress, the same amount of soil was added into pots in same size and well watered. Approximately 50 seeds for WT plants and three T 3 transgenic lines were vernalized at 4 • C for two days and evenly sown into pots with regular cultivation (watering same amount of water regularly). Four weeks later, each pot was treated with drought or salt. In the case of drought treatment, ample water was irrigated in the tray before drought stress treatment to make the soil water content reach saturated water capacity, and after the seedlings recovered for 24 h, plants were subjected to drought stress through stopping watering over 15 days and then rewatered. For salt treatment, plants were treated by a salt solution with 300 mM NaCl twice at 3-day intervals. During the stress treatment, the real salt concentration of soil increased gradually. All pots were photographed every two or three days, and three days after re-watering, and also before the drought or salt stress. Samples of 4-week-old plants for physiological index measurements under stress conditions were obtained through drying treatment over 10 days and salt (300 mM NaCl) over 7 days. These assessments were conducted in triplicate.

Estimation of Water Loss Rate
To figure out the change of water loss rate, 0.5 g rosette leaves from WT plants and three T 3 transgenic lines grown for four weeks were excised and promptly weighed. The leaves in the same status from each line were then laid on an experiment table and weighed at setting time points (0, 1, 2, 3, 4, 5, 6, and 7 h), with three replicates for each line. Water loss percentage can be obtained according to the fresh weights of isolated leaves before dehydration.

Physiological Measurements
Chlorophyll was extracted from leaf tissue in 95% ethanol as previously described [80]. Proline was measured following the modified method of acidic ninhydrin reaction as stated earlier [81]. The soluble protein and soluble sugar content were determined using TP quantitative assay kit (Nanjing Jiancheng, Nanjing, China) and the plant soluble sugar content test kit (Nanjing Jiancheng), respectively, following the user manual instructions. The accumulation of hydrogen peroxide (H 2 O 2 ) and superoxide anion radical (O 2 − ) in leaves were visualized by histochemical staining with 3,3 -diaminobenzidine (DAB) and nitroblue tetrazolium (NBT) separately [82]. Leaves were then decolorized in 95% ethanol before recorded by photograph. The H 2 O 2 level and anti-superoxide anion activity (an indicator of O 2 − content) were analyzed by hydrogen peroxide assay kit (Nanjing Jiancheng) and inhibition and produce superoxide anion assay kit (Nanjing Jiancheng), respectively, according to the operating instructions. The enzyme liquid was extracted for combined determination of superoxide dismutase (SOD), peroxidase (POD), and catalase (CAT) activities as well as malondialdehyde (MDA) content. Detailed descriptions of these assays were elaborated by Zheng et al. [83] and Du and Bramlage [84]. Three replicates were executed for these experiments.

Analysis of Stomatal Aperture Responsive to Drought and ABA Treatment
To measure stomatal movement induced by drought and ABA, rosette leaves of WT plants and three T 3 transgenic lines growing for 4 weeks were floated on a solution to induce stomatal opening (50 mM KCl, 0.1 mM CaCl 2 , and 10 mM MES, pH 6.15) and placed under light for 2.5 h. Then, we transferred these leaves into the opening solution with 0, and 300 mM mannitol, or 20 µM ABA and they were incubated in a growth chamber for a further 2 h. Stomata on the leaf lower epidermal layers were immediately observed and photographed by means of an optical microscopy (DP80, Olympus, Japan), and the stomatal aperture (width to length ratio) from 60 stomata of each line was measured. All experiments were repeated in triplicate.

Statistical Analysis
Data from this study were analyzed by Student's t-test in SPSS v. 23.0. The measured values were expressed as the mean ± standard deviation (SD) of three replicates, and differences were viewed as to be significant at * p < 0.05 and ** p < 0.01.
Supplementary Materials: Supplementary materials can be found at http://www.mdpi.com/1422-0067/21/8/3011/ s1. Table S1: The GenBank accession numbers of MfPIF1 and some highly humongous PIFs used to construct phylogenetic tree in Figure 1; Table S2: List of primers used in this study.

Conflicts of Interest:
The authors declare no conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results.