Abstract
Phosphorus (P) is vitally important for most plant processes. However, the P available to plants is present in the soil in the form of inorganic phosphate (Pi), and is often present in only limited amounts. Water stress further reduces Pi availability. Previous studies have highlighted the important roles of members of the PHOSPHATE TRANSPORTER 1 (PHT1) family and arbuscular mycorrhizal (AM) associations for Pi acquisition by plants growing in various environments. In order to understand the Pi uptake of Lycium barbarumL., a drought-tolerant ligneous species belonging to the Solanaceae family, we cloned and characterized six L. barbarum genes encoding transporter proteins belonging to the PHT1 family, and investigated their transcriptional response to AM associations and water stress. The six cloned PHT1 genes of L. barbarum had a similar evolutionary history to that of PHT1 genes found in other Solanaceae species. Three of these genes (LbPT3, LbPT4 and LbPT5) were AM-induced; the other three genes (LbPT1, LbPT2 and LbPT7) played distinct roles in Pi acquisition, translocation and remobilization in roots and leaves. AM-induced PHT1 genes maintained their function under water stress, while moderate and severe water stress upregulated non-AM-induced PHT1 genes in roots and leaves, respectively. Moreover, although LbPT1 was upregulated in AM roots under water stress, LbPT2 and LbPT7 were inhibited in AM roots, which suggested that an AM association satisfied the demand for Pi in roots under water stress and that LbPT1 may play a role in translocating Pi from roots to shoots in this situation.
Introduction
Phosphorus (P) is vitally important for plants due to its multiple functions. For instance, it can function as the key component in signal transduction cascades, as a structural constituent of phospholipids and nucleic acids, and also as a component in energy transfer (Versaw and Harrison 2002). In spite of its abundance in soil, most of the P in soil exists in an immobilized or fixed form that is not directly available to plants (Lambers et al. 2015). The form of P that plants can utilize is inorganic phosphate (Pi) (Rausch et al. 2001). Furthermore, Pi mobility and mineralization are restricted under various environmental stresses, e.g., water stress (Sardans and Peñuelas 2004, Sobkowiak et al. 2012). Pi uptake by plants is also affected by water deficiency due to a reduction of the root extension rate (Koide 1991). To cope with the low availability of Pi in soil, plants have developed several adaptive strategies. For instance, plants can alter their allocation of resources from shoots to roots to develop better access to nutrient and water resources (Ho et al. 2005, Jaramillo et al. 2013), and form arbuscular mycorrhizal (AM) symbioses, whose far-reaching extraradical mycelia function as an extension of the plant root system (Smith and Read 2008).
The PHOSPHATE TRANSPORTER 1 (PHT1) family of proteins has been shown to play essential roles in Pi uptake by plants growing under various environmental conditions (Smith et al. 2003, Shin et al. 2004, Nussaume et al. 2011). Members of the PHT1 family have been identified in many vascular plants (González et al. 2005). The genes encoding PHT1 family members are mainly expressed in roots (González et al. 2005) and help plants to acquire Pi from soil (Nussaume et al. 2011). Previous studies and reviews (Rausch and Bucher 2002, Loth-Pereda et al. 2011, Nussaume et al. 2011, Fan et al. 2013, Chen et al. 2014) have highlighted that many PHT1 genes not only play crucial roles in Pi uptake in roots, but also perform other functions in Pi translocation and remobilization in roots and aerial parts.
Another way in which plants can acquire soil Pi is via the AM pathway, which relies on the hyphal network of AM fungi to absorb soil Pi and transfer it to the inner layer of the host root where members of the PHT1 family transport Pi into the plant (Smith et al. 2001, Smith and Read 2008). AM-induced PHT1 genes, which are exclusively expressed in the arbuscule-containing cells, have been identified in different plant species (Harrison et al. 2002, Nagy et al. 2005, Xu et al. 2007, Xie et al. 2013).
Lycium barbarum L. (Solanaceae) is a drought-tolerant, perennial ligneous shrub that is commonly grown in the northwest of China (Zhao and Zeng 1999), where water deficiency is a critical factor limiting plant growth (Yang et al. 2011). Furthermore, L. barbarum produces fruit that are highly valued for their medicinal properties by practitioners of traditional Chinese medicine (Gan et al. 2004, Luo et al. 2006). In a previous study, we showed that AM fungi were common in the roots of L. barbarum in the northwest of China (Zhang et al. 2010), which suggested that AM associations might be involved in the metabolic and physiological processes of L. barbarum in this arid region of China. Although many genes encoding PHT1 family members have been identified in annual herbaceous plants of Solanaceae (Nagy et al. 2005, Xu et al. 2007, Chen et al. 2014), the genes encoding PHT1 family members found in perennial ligneous plants of Solanaceae are poorly known.
Here, we characterized six genes encoding PHT1 family members in L. barbarum, and we investigated the Pi uptake activity of all six Pi transporter (PiT) proteins by complementation analysis in the yeast pho84 mutant (MB192), and the cellular localization of these proteins in yeast and tobacco. We investigated the response of the six PHT1 genes to Pi deficiency, AM colonization and water stress. We tested the hypothesis that the six PHT1 genes respond differently to Pi deficiency and AM colonization, and that at least two of these genes are AM-induced PHT1 genes. Moreover, we tested the hypothesis that water stress would modify the transcriptional response of PHT1 genes to AM fungi.
Materials and methods
Plants and growth conditions
Seeds of L. barbarum L. (cultivar, Ningqi No. 1) were kindly provided by Dr Yajun Wang (Ningxia Academy of Agriculture and Forestry Sciences). The seeds were surface-sterilized by immersion in 5% sodium hypochlorite for 10 min and then washed three times with sterile water. The sterilized seeds were then transferred to sterilized filter paper that had been moistened with sterilized water and were left in the dark at 28 °C to germinate.
The germinated seeds were seeded in holes (the volume of each hole was 50 ml) in a growth substrate of sterilized sand. Each hole was filled with 10 ml of half-strength Hoagland's solution (Hoagland and Arnon 1950). After 20 days, uniform seedlings were selected and transplanted to pots (one seedling per pot).
AM fungal inoculation
An AM fungal inoculum of Rhizophagus irregularis (Błaszk, Wubet, Renker & Buscot) Walker & Schüßler (BGC BJ09) was obtained from the Beijing Academy of Agriculture and Forestry Sciences, China, and consisted of sand, spores, mycelia and colonized root fragments. The number of propagules per milliliter was determined using the most probable number (MPN) method (Feldmann and Idczak 1992). The seedlings that received the inoculated treatment (referred to as the AM-treatment) were inoculated by adding 8 ml of inoculum (containing 288 propagules per milliliter) to the soil in each pot; the seedlings that received the non-inoculated treatment (referred to as the NM-treatment) were mock inoculated by adding 8 ml of sterilized inoculum to the soil of each pot with filter drains.
Cloning of L. barbarum PHT1 genes
Three pairs of degenerate primers were designed based on the conserved regions of the amino acid sequences of the available plant PiTs [SlPT (NP001234043, AAB82146, AAX85192), NtPT (AF156696, BAA86070), MtPT3 (ABM69110), MtPT5 (ABM69111), MtPT4 (AAM76744), SlPT4 (AAV97730), StPT1 (CAA67395), StPT2 (CAA67396), StPT4 (AAW51149), PhPT (ABS12068), PhPT3 (ACB37440), PhPT4 (ACB37441), PhPT5 (ACB37442)] using the consensus-degenerate hybrid oligonucleotide primer (CODEHOP) method (Rose et al. 2003) to amplify the orthologous conserved segments in L. barbarum. Two pairs of additional primers were designed based on conserved regions of mRNA sequences of plant PiTs from the Solanaceae family [SlPT2 (EF091666), CfPT2 (EF091665), SlPT2 (NM001247114) and StPT2 (X98891)]. Subsequent PCRs were performed using cDNA reverse transcribed from RNA extracted from L. barbarum roots of seedlings that received the AM-treatment or the NM-treatment. PCR products were purified using a Gel Purification Kit (Omega Bio-tek, Norcross, GA, USA), ligated into a pMD18-T simple T/A clone vector (TaKaRa Bio, Dalian, China), and transformed into competent Escherichia coli DH5α according to the standard protocol (Hanahan 1983). Positive clones were sequenced by GenScript USA Inc. (Nanjing, China).
The complete sequences of the L. barbarum PHT1 genes were obtained using the 5′ and 3′ rapid-amplification of cDNA ends (RACE) strategy (Bertioli 1997), which was performed using the SMARTer™ RACE cDNA Amplification Kit (Clontech Laboratories Inc., Mountain, CA, USA). The conserved fragments were used to design primers for 5′ RACE and 3′ RACE. Primers were also designed to verify the full length of the cDNA of the cloned PHT1 genes. The primers used are listed in Table S1 available as Supplementary Data at Tree Physiology Online.
Isolation of the promoter regions of L. barbarum PHT1 genes
The promoter regions of the PHT1 genes were obtained using high-efficiency thermal asymmetric interlaced PCR (high-tail PCR) (Liu and Chen 2007). The primers used for the high-tail PCR are listed in Table S2 available as Supplementary Data at Tree Physiology Online. PCR products were purified using a Gel Purification Kit (Omega Bio-tek). Purified products were cloned into the pMD18-T simple T/A clone vector (TaKaRa Bio) for sequencing using GenScript USA Inc. The promoters were verified by carrying an overlapping sequence with the coding sequence of LbPT genes.
Complementation analysis of PHT1 genes in yeast
Complementation analysis was conducted as previously described (Ai et al. 2009). The yeast (Saccharomyces cerevisiae) Pi uptake-defective mutant MB192 (MATa pho3-1 ade2 leu2-3,112 his3-532 trp1-289 ura3-1,2 can Δpho84::HIS3) (Bun-Ya et al. 1991), which was provided by the National BioResource Project (NBRP) of the Ministry of Education, Culture, Sports, Science and Technology (MEXT), Japan, and the expression vector in MB192, pGADT7 (kindly provided by Prof. Zhiying Zhang, College of Animal Science & Technology, Northwest A&F University) were used for these experiments. The coding sequence of the LbPT genes was cut and subcloned into the XbaI or BamHI to XhoI sites of the yeast shuttle vector pGADT7. Yeast strains (YpGAD-LbPT1, YpGAD-LbPT2, YpGAD-LbPT3, YpGAD-LbPT4, YpGAD-LbPT5, YpGAD-LbPT7, wild type and YpGADT7) were cultured in YNB liquid medium, then harvested at the exponential phase and re-suspended in Pi-free medium. The yeast strains were then incubated at 30 °C for 4 h before plating the cells onto YNB agar plates containing three different Pi concentrations (10 mM, 100 μM or 50 μM), followed by incubation at 30 °C for 3 days. In addition, YNB liquid media containing two different Pi concentrations (10 or 100 μM) were also inoculated with these yeast strains and incubated at 30 °C for 24 h. Yeast growth was measured using bromocresol purple as a pH indicator: a color change from purple to yellow would indicate the acidification of the liquid medium.
The yeast pho84 mutant (MB192), which lacks the high-affinity phosphate transporter PHO84, has severely impaired phosphate uptake. Because MB192 cells cannot accumulate phosphate, they produce a repressible acid phosphatase (rAPase), even during growth on high-phosphate medium containing levels of phosphate sufficient to repress rAPase production in wild-type cells (Harrison and van Buuren 1995). In order to detect the production of acid phosphatase in yeast cells, yeast strains YpGAD-LbPT1, YpGAD-LbPT2, YpGAD-LbPT3, YpGAD-LbPT4, YpGAD-LbPT5, YpGAD-LbPT7, wild type and YpGADT7 were plated onto YNB agar plates containing 10 mM Pi and incubated at 30 °C for 3 days. rAPase activity was detected as previously described (Toh-E and Oshima 1974).
Subcellular localization of PHT1 genes in yeast and tobacco
The coding sequences of LbPT1, LbPT2, LbPT3, LbPT4, LbPT5 and LbPT7 were translationally fused with GFP gene, and the fusion sequences were inserted into the yeast expression vector pGADT7 and the plant expression vector modified pCambia0380 with a 35S promoter and NOS terminator (kindly provided by Prof. Weixing Shan, College of Plant Protection, Northwest A&F University). For expression in yeast, the generated constructs and free GFP gene were transformed into MB192. Photographs were taken using a Nikon A1R (Nikon, Japan) confocal microscope. The excitation laser wavelength was 488 nm and emission spectra were collected at 500–540 nm.
For transient expression in tobacco leaves, all the generated constructs, free GFP gene and PM-rk constructs (Nelson et al. 2007) were individually transformed into Agrobacterium tumefaciens Gv3101. The tobacco leaves were prepared, transformed and photographed using previously described methods (Pan et al. 2016). For transient expression in tobacco roots, the Agrobacterium culture carrying LbPT3, free GFP gene and PM-rk constructs were harvested and suspended in the identical infiltration media used for leaves, and adjusted to a final concentration with an OD600 of 0.1. The 4- to 6-week-old tobacco plant roots (AM roots and NM roots) were submerged in the infiltration media for 20 min; photographs were taken 2 days post-infiltration, similar to the method used for the transient expression in tobacco leaves.
Transcriptional responses of PHT1 genes to Pi and water stress
In order to investigate the transcriptional responses of PHT1 genes to Pi and water stress, we set up two independent experiments.
Experiment 1: transcriptional responses of PHT1 genes to different Pi levels
Seedlings were transplanted into plastic pots (height, 16.5 cm; diameter, 13.0 cm; volume, 400 ml) filled with sterilized sand, and cultivated in a climate incubator with 12 h of full light (6000 lx) per day at 28 °C and 60% relative humidity.
For 40 days post-inoculation (dpi), each seedling was fertilized with 20 ml of half-strength Hoagland's solution every 10 days, and watered with 5 ml of distilled water every 2 days. On the 41st day, each pot was immersed in a beaker filled with distilled water: the water was changed five times, once every 2 h, to remove the residual nutrients in the substrates. The pots were then placed back in the incubator for another 5 days. The AM- and NM-treatment groups were then divided into two subgroups (10 pots in each subgroup). Every 5 days the seedlings in one subgroup received 20 ml of modified Hoagland's solution containing 0.02 mM NaH2PO4 and the seedlings in the other subgroup received 20 ml of modified Hoagland's solution containing 1 mM NaH2PO4. Twenty days after the seedlings received the Pi treatment, six seedlings were randomly selected: all the roots and leaves were sampled and stored at −80 °C until required for subsequent experiments. Some root parts were fixed in FAA fixative (Feder and O'Brien 1968) to measure the AM colonization.
Experiment 2: transcriptional responses of PHT1 genes to different water stresses
For Experiment 2, seedlings were transplanted into plastic cylinders (height, 16.0 cm; diameter, 12.0 cm), which were inserted into the top of plastic pots (height, 17.0 cm; diameter, 13.5 cm) (to make a total volume of 1600 ml) to increase the depth of the pot to facilitate root elongation (for details see Figure S1 available as Supplementary Data at Tree Physiology Online). The growth substrate was a sand–soil mixture (sand:loamy soil = 1:1, in volume; the pH of soil was 8.2; the available K, P, NO3-N and NH4-N content of the soil was 159.1, 11.2, 36.9, and 5.4 mg kg–1, respectively, and the organic matter content of the soil was 14.6 g kg–1). The seedlings were cultivated in a glass house with natural light. During the growth period, the average temperature was 25 °C and the relative humidity was 35%.
Seedlings were fertilized with 200 ml of half-strength Hoagland's solution every 10 days. The water content in each pot was monitored every day using TDR100 (Spectrum Technologies Inc., Plainfield, IL, USA), and was adjusted to 50% of field capacity (percentage of volume water content: 45%) by adding corresponding water if required. At 60 dpi, the supplement of Hoagland's solution was stopped, and the seedlings in each treatment group were divided into three subgroups (14 pots in each subgroup). The water content of the three subgroups in each treatment (AM-treatment or NM-treatment) was adjusted to 50%, 30% or 15% of field capacity and then maintained at corresponding level using TDR100 monitor daily. Six seedlings were randomly selected and sampled from each subgroup at 90 dpi. Leaves, stems and roots of each plant were sampled independently and stored at –80 °C. Some root parts were fixed in FAA fixative to examine AM colonization.
AM colonization rate measurement
The roots fixed in FAA fixative (from Experiments 1 and 2) were washed and cut into pieces (2 cm in length), and stained with trypan blue (Phillips and Hayman 1970). The AM colonization rate was quantified using the magnified intersections method (McGonigle et al. 1990, Sun and Tang 2012).
Determination of the total P concentration of the plant
The frozen samples were milled (Experiments 1 and 2), oven-dried at 80 °C for 24 h, and then digested with concentrated H2SO4 and H2O2. The P concentration was determined using the molybdate-blue colorimetric method (Thomas et al. 1967).
Anthocyanin analysis
The frozen leaf samples (Experiment 2) were milled and the anthocyanins were extracted and measured using the method previously described by Bieza and Lois (2001).
RNA extraction, cDNA synthesis, PCR and quantitative real-time PCR (qPCR)
Total RNA was extracted separately from roots and leaves using the E.Z.N.A™ plant RNA kit (Omega Bio-tek). The integrity of the RNA was evaluated by performing agarose gel electrophoresis, and was quantified using spectrophotometry with a NanoDrop 2000 (Thermo Scientific, Pittsburgh, PA, USA). Any genomic DNA contaminations were removed using DNase I treatment (TaKaRa Bio). PCR was performed using the primers of the L. barbarum actin gene (HQ415754) (see Table S3 available as Supplementary Data at Tree Physiology Online) to ensure the removal of any DNA contamination after DNase I treatment according to a previously described method (Labouyrie et al. 1999). cDNA was synthesized using the PrimerScript® 1st Strand cDNA Synthesis Kit (TaKaRa Bio), and was used as a template in the PCRs. PCR was performed in a C1000 thermocycler (Bio-Rad, Hercules, CA, USA) following the manufacturer's instructions, using gene-specific primers (see Table S3 available as Supplementary Data at Tree Physiology Online). The 20 μl reaction volume contained 0.5 μl of each gene-specific primer (10 μM), 1.0 μl of cDNA equivalent to 50 ng of total RNA, 8 μl of RNase free H2O and 10 μl of 2× EcoTaq PCR SuperMix (TransGen Biotech, Beijing, China). An initial 5 min step at 94 °C was followed by 35 cycles of denaturation at 94 °C for 15 s, annealing at 56 °C for 30 s, extension at 72 °C for 30 s, with a final extension for 10 min at 72 °C (different primers had a different annealing temperature, time and extension time).
For qPCR, a fragment of the L. barbarum actin gene was used as an internal control for the relative quantification assay (primer pair: Actin-f and Actin-r, see Table S3 available as Supplementary Data at Tree Physiology Online). The SYBR Green-based qPCR was carried out to quantify the transcript levels of cloned genes relative to those of the actin gene using gene-specific primers (see Table S3 available as Supplementary Data at Tree Physiology Online). The 20 μl reaction volume contained 0.4 μl of each gene-specific primer (10 μM), 2.0 μl of cDNA equivalent to 100 ng of total RNA, 7.2 μl of RNase free H2O and 10 μl of SYBR Green PCR master mix (Roche Diagnostics, Basel, Switzerland). All qPCR experiments were performed using a CF96X real-time PCR system (Bio-Rad, Hercules, CA, USA), and the thermal cycling consisted of 95 °C for 10 min, followed by 40 cycles of 95 °C for 15 s, 55 °C for 15 s, 72 °C for 25 s, followed by heating at 60– 95 o C, increasing at a rate of 0.5 °C per 10 s. The melting curve indicated the specificity of the primer pairs. CT values obtained from qPCR were compared using the 2–ΔΔCT method (Livak and Schmittgen 2001). Three biological replicates with three technical repetitions were conducted for all reactions.
Data analysis
All the data were analyzed by performing an ANOVA (SPSS 20.0, IBM, Armonk, NY, USA), and the means were compared using Tukey's HSD test (P < 0.05) (SPSS 20.0, IBM, Armonk, NY, USA). All the column charts were constructed using the OriginPro 8.5.0 (OriginLab Corporation, Northampton, MA, USA) and Adobe Photoshop CS4 (Adobe Systems Inc., San Jose, CA, USA).
Results
Cloning and computational analysis of PHT1 genes in L. barbarum
Putative cis-regulatory elements in the promoter regions of L. barbarum PHT1 genes (a) and putative MYCS in the promoter regions of LbPT3–LbPT5 (b). PIBS, GNATATNC; mycorrhiza transcription factor binding sequence (MYCS), TTTCTTGTTCT, or with one nucleotide variation; PHO, CACGTG; W-box, TTGACY; AAAGAT and CTCTT are OSE1ROOTNODULE and NODCON2GM motifs, respectively.
We constructed an unrooted phylogenetic tree to analyze the evolutionary relationships of the PHT1 proteins found in L. barbarum and other plant species. All PiT proteins cloned in this study were orthologs of members of the PHT1 family found in species belonging to the Solanaceae (see Figure S4 available as Supplementary Data at Tree Physiology Online). Four PHT1 proteins belonged to group I (LbPT1, LbPT2, LbPT3 and LbPT7), two PHT1 proteins belonged to group III (LbPT4 and LbPT5), and L. barbarum possessed two phylogenetically distant AM-associated PHT1 proteins, LbPT3 and LbPT4/5 (see Figure S4 available as Supplementary Data at Tree Physiology Online).
Functional analysis of cloned PHT1 genes in yeast
Functional characterization of L. barbarum PiTs in the yeast mutant strain MB192. (a) Yeast MB192 cells harboring either the empty expression vector pGADT7 (control) or the full-length cDNAs of LbPT1, LbPT2, LbPT3, LbPT4, LbPT5 or LbPT7 constructs and the wild type were grown in SD medium at pH 5.8 to an OD600 = 1. Equal volumes of 10-fold serial dilutions were plated onto plates containing 10 mM Pi, 100 μM Pi or 50 μM Pi and then incubated at 30 °C for 3 days. (b) An acid phosphatase activity test was performed using yeast strain MB192 cells harboring either the empty expression vector pGADT7 (control) or the full-length cDNAs of LbPT1, LbPT2, LbPT3, LbPT4, LbPT5 or LbPT7 constructs and the wild type. Phosphatase activity was detected by staining, following the method of Toh-E and Oshima (1974). (c) The yeast strain cells harboring either the empty expression vector pGADT7 (control) or the full-length cDNAs of LbPT1, LbPT3 or LbPT7 and wild type were grown in solution that had been stained for acidification. The pH indicator bromocresol purple did not change from blue to yellow until the yeast cells had achieved significant growth in culture. The media contained either 10 μM or 100 μM Pi. (d) Growth curves of the wild type, MB192 and MB192 transformed with YpGAD-LbPT1, YpGAD-LbPT3 and YpGAD-LbPT7 generated from a 45-h culture in 25 mM MES buffer and 100 μM Pi.
To validate whether the function of these PiTs was same as the function of PHO84 in yeast, the rAPase activity of MB192 expressing these PiTs was tested. MB192 transformants expressing LbPT3 and LbPT7 showed wild-type rAPase activity and remained pale, indicating that the expression of LbPT3 and LbPT7 complemented the pho84 mutant phenotype (Figure 2b). Although the MB192 transformants expressing LbPT1 could grow on low Pi medium, it still showed the pho84 phenotype and stained dark red (Figure 2a and b).
Subcellular localization of cloned PHT1 genes
We examined the subcellular localization of LbPT1, LbPT2, LbPT3, LbPT4, LbPT5 and LbPT7 in yeast and tobacco. LbPT1-GFP, LbPT3-GFP and LbPT7-GFP were located in the yeast cell membrane. LbPT7-GFP was also localized in the intracellular membranes. LbPT2-GFP was centralized at several points in the yeast cell membrane, while LbPT4-GFP and LbPT5-GFP were not located in the yeast cell membrane (see Figure S5 available as Supplementary Data at Tree Physiology Online).
Subcellular localization of LbPT1-GFP, LbPT2-GFP, LbPT3-GFP, LbPT4-GFP, LbPT5-GFP and LbPT7-GFP in tobacco leaves or roots. The LbPT-GFP constructs or free GFP construct were transiently expressed in tobacco leaves or roots and the fluorescence was observed under a confocal microscope. (a) Free GFP acted as the control, LbPT1:GFP, LbPT2:GFP, LbPT4:GFP, LbPT5:GFP and LbPT7:GFP fusion proteins localized to the plasma membrane (PM) in tobacco leaves, while LbPT3:GFP did not. The PM marker is in red and GFP is in green. (b) Free GFP acted as the control, the LbPT3:GFP fusion protein localized to the PM in tobacco roots. Bars represent 25 μm in length.
Arbuscular mycorrhizal formation in L. barbarum
As found in our previous investigations in the field, R. irregularis formed AM with L. barbarum, and typical mycorrhizal structures (arbuscules, vesicles, inter-radical spores) were observed in the roots (see Figure S6 available as Supplementary Data at Tree Physiology Online). The colonization type belonged to the Arum-type. Plants inoculated with sterile inoculum did not form AM. Arbuscular mycorrhizal colonization rates were significantly suppressed by high Pi supply but were not affected by water stress (Table 1).
AM colonization rates under Pi and water stresses.
| Experiment | Subgroup | Infection rate of AM fungi | Infection rate of arbuscule |
|---|---|---|---|
| 1 | AM LP | 0.58 ± 0.23a | 0.32 ± 0.20a |
| AM HP | 0.12 ± 0.07b | 0.01 ± 0.00b | |
| 2 | AM50 | 0.34 ± 0.05 | 0.19 ± 0.04 |
| AM30 | 0.47 ± 0.06 | 0.33 ± 0.06 | |
| AM15 | 0.39 ± 0.11 | 0.21 ± 0.07 |
| Experiment | Subgroup | Infection rate of AM fungi | Infection rate of arbuscule |
|---|---|---|---|
| 1 | AM LP | 0.58 ± 0.23a | 0.32 ± 0.20a |
| AM HP | 0.12 ± 0.07b | 0.01 ± 0.00b | |
| 2 | AM50 | 0.34 ± 0.05 | 0.19 ± 0.04 |
| AM30 | 0.47 ± 0.06 | 0.33 ± 0.06 | |
| AM15 | 0.39 ± 0.11 | 0.21 ± 0.07 |
AM LP and AM HP stand for AM fungi inoculation at low and high Pi levels, respectively; AM50, AM30, AM15 stand for 50%, 30%, 15% of field capacity level with AM fungi, respectively. Data are presented as the mean ± SD. Different letters indicate that the means are significantly different between treatments. Analysis of variance followed by Tukey's HSD test when the whole test was significantly different (P < 0.05); n = 6.
AM colonization rates under Pi and water stresses.
| Experiment | Subgroup | Infection rate of AM fungi | Infection rate of arbuscule |
|---|---|---|---|
| 1 | AM LP | 0.58 ± 0.23a | 0.32 ± 0.20a |
| AM HP | 0.12 ± 0.07b | 0.01 ± 0.00b | |
| 2 | AM50 | 0.34 ± 0.05 | 0.19 ± 0.04 |
| AM30 | 0.47 ± 0.06 | 0.33 ± 0.06 | |
| AM15 | 0.39 ± 0.11 | 0.21 ± 0.07 |
| Experiment | Subgroup | Infection rate of AM fungi | Infection rate of arbuscule |
|---|---|---|---|
| 1 | AM LP | 0.58 ± 0.23a | 0.32 ± 0.20a |
| AM HP | 0.12 ± 0.07b | 0.01 ± 0.00b | |
| 2 | AM50 | 0.34 ± 0.05 | 0.19 ± 0.04 |
| AM30 | 0.47 ± 0.06 | 0.33 ± 0.06 | |
| AM15 | 0.39 ± 0.11 | 0.21 ± 0.07 |
AM LP and AM HP stand for AM fungi inoculation at low and high Pi levels, respectively; AM50, AM30, AM15 stand for 50%, 30%, 15% of field capacity level with AM fungi, respectively. Data are presented as the mean ± SD. Different letters indicate that the means are significantly different between treatments. Analysis of variance followed by Tukey's HSD test when the whole test was significantly different (P < 0.05); n = 6.
Responses of cloned PHT1 genes to Pi availability and AM inoculation (Experiment 1)
We performed semi-quantitative RT-PCR and compared the expression of PHT1 genes in the roots and leaves of seedlings that received a low supply of Pi. As indicated in Figure S7 available as Supplementary Data at Tree Physiology Online, LbPT1, LbPT2 and LbPT7 were expressed in both roots and leaves, and the AM colonization appeared to hinder slightly the expression of these three genes in roots; LbPT3, LbPT4 and LbPT5 were highly induced in mycorrhizal roots, and LbPT5 was also detected in the roots of seedlings that received the NM-treatment although the transcript level was quite low; and LbPT3, LbPT4 and LbPT5 were not detected in leaves.
LbPT1, LbPT2, LbPT3, LbPT4, LbPT5 and LbPT7 gene expression in seedlings that received the AMF and Pi treatments. AMF, inoculated with AM fungi; NM, mock inoculated with inactive AMF fungi; A+P, AMF treatment and high Pi treatment; A–P, AMF treatment and low Pi treatment; N+P, NM treatment and high Pi treatment; N–P, NM treatment and low Pi treatment. Each column represents the mean ± SE. Significant effect of two-way ANOVA: *P < 0.05, **P < 0.01, ***P < 0.001, NS, no significant effect. Different letters indicate that the means are significantly different between treatments. Analysis of variance followed by Tukey's HSD test when the whole test was significantly different (P < 0.05).
Plant growth, plant P status, anthocyanin content and the expression patterns of cloned PHT1 genes under water stress (Experiment 2)
Arbuscular mycorrhiza formation significantly promoted the growth of roots and stems of seedlings subjected to the three different water conditions (Table 2). Seedlings grown under severe water stress (15% of field capacity) produced the least leaf biomass in both the AM-treatment and the NM-treatment groups. The root biomass produced by seedlings in both the AM-treatment and NM-treatment groups increased initially but then decreased under severe water stress conditions. Arbuscular mycorrhizall colonization significantly improved the P content of roots. Although less significant, the P content of the roots of seedlings in both the AM-treatment and NM-treatment initially showed an increasing trend but then decreased under severe water stress conditions. In addition, the P content of the leaves of seedlings that received the AM treatment declined under severe water stress conditions.
The effects of AM colonization and water stress on biomass and P content of L. barbarum.
| Treatment | Dry weight (g) | P content (mg) | ||||
|---|---|---|---|---|---|---|
| Root | Stem | Leaf | Root | Stem | Leaf | |
| AM50 | 0.364 ± 0.057ab | 0.870 ± 0.091a | 0.299 ± 0.014ab | 1.360 ± 0.076ab | 1.387 ± 0.353a | 1.383 ± 0.178a |
| AM30 | 0.437 ± 0.014a | 0.952 ± 0.030a | 0.308 ± 0.022a | 1.583 ± 0.154a | 1.763 ± 0.245a | 1.133 ± 0.200ab |
| AM 15 | 0.346 ± 0.005b | 0.566 ± 0.046b | 0.171 ± 0.003c | 1.230 ± 0.079b | 1.350 ± 0.010a | 1.010 ± 0.101b |
| NM50 | 0.169 ± 0.010c | 0.359 ± 0.056c | 0.238 ± 0.02b | 0.307 ± 0.106d | 0.367 ± 0.045b | 0.373 ± 0.035c |
| NM30 | 0.324 ± 0.044b | 0.349 ± 0.069c | 0.251 ± 0.047ab | 0.570 ± 0.076c | 0.300 ± 0.082b | 0.513 ± 0.091c |
| NM15 | 0.175 ± 0.012c | 0.272 ± 0.043c | 0.139 ± 0.007c | 0.407 ± 0.050cd | 0.273 ± 0.038b | 0.400 ± 0.020c |
| Mycorrhiza | ** | ** | ** | *** | *** | *** |
| Water | ** | ** | ** | ** | NS | NS |
| Mycorrhiza × Water | NS | ** | NS | NS | NS | * |
| Treatment | Dry weight (g) | P content (mg) | ||||
|---|---|---|---|---|---|---|
| Root | Stem | Leaf | Root | Stem | Leaf | |
| AM50 | 0.364 ± 0.057ab | 0.870 ± 0.091a | 0.299 ± 0.014ab | 1.360 ± 0.076ab | 1.387 ± 0.353a | 1.383 ± 0.178a |
| AM30 | 0.437 ± 0.014a | 0.952 ± 0.030a | 0.308 ± 0.022a | 1.583 ± 0.154a | 1.763 ± 0.245a | 1.133 ± 0.200ab |
| AM 15 | 0.346 ± 0.005b | 0.566 ± 0.046b | 0.171 ± 0.003c | 1.230 ± 0.079b | 1.350 ± 0.010a | 1.010 ± 0.101b |
| NM50 | 0.169 ± 0.010c | 0.359 ± 0.056c | 0.238 ± 0.02b | 0.307 ± 0.106d | 0.367 ± 0.045b | 0.373 ± 0.035c |
| NM30 | 0.324 ± 0.044b | 0.349 ± 0.069c | 0.251 ± 0.047ab | 0.570 ± 0.076c | 0.300 ± 0.082b | 0.513 ± 0.091c |
| NM15 | 0.175 ± 0.012c | 0.272 ± 0.043c | 0.139 ± 0.007c | 0.407 ± 0.050cd | 0.273 ± 0.038b | 0.400 ± 0.020c |
| Mycorrhiza | ** | ** | ** | *** | *** | *** |
| Water | ** | ** | ** | ** | NS | NS |
| Mycorrhiza × Water | NS | ** | NS | NS | NS | * |
AM50, AM30, AM15 and NM50, NM30, NM15 stand for 50%, 30%, 15% of field capacity level with/without AM fungi, respectively. Data are presented as the mean ± SD. Significant effect of two-way ANOVA: *P < 0.05, **P < 0.01, ***P < 0.001, NS, no significant effect. Different letters indicate that the means are significantly different between treatments. Analysis of variance followed by Tukey's HSD test when the whole test was significantly different (P < 0.05); n = 6.
The effects of AM colonization and water stress on biomass and P content of L. barbarum.
| Treatment | Dry weight (g) | P content (mg) | ||||
|---|---|---|---|---|---|---|
| Root | Stem | Leaf | Root | Stem | Leaf | |
| AM50 | 0.364 ± 0.057ab | 0.870 ± 0.091a | 0.299 ± 0.014ab | 1.360 ± 0.076ab | 1.387 ± 0.353a | 1.383 ± 0.178a |
| AM30 | 0.437 ± 0.014a | 0.952 ± 0.030a | 0.308 ± 0.022a | 1.583 ± 0.154a | 1.763 ± 0.245a | 1.133 ± 0.200ab |
| AM 15 | 0.346 ± 0.005b | 0.566 ± 0.046b | 0.171 ± 0.003c | 1.230 ± 0.079b | 1.350 ± 0.010a | 1.010 ± 0.101b |
| NM50 | 0.169 ± 0.010c | 0.359 ± 0.056c | 0.238 ± 0.02b | 0.307 ± 0.106d | 0.367 ± 0.045b | 0.373 ± 0.035c |
| NM30 | 0.324 ± 0.044b | 0.349 ± 0.069c | 0.251 ± 0.047ab | 0.570 ± 0.076c | 0.300 ± 0.082b | 0.513 ± 0.091c |
| NM15 | 0.175 ± 0.012c | 0.272 ± 0.043c | 0.139 ± 0.007c | 0.407 ± 0.050cd | 0.273 ± 0.038b | 0.400 ± 0.020c |
| Mycorrhiza | ** | ** | ** | *** | *** | *** |
| Water | ** | ** | ** | ** | NS | NS |
| Mycorrhiza × Water | NS | ** | NS | NS | NS | * |
| Treatment | Dry weight (g) | P content (mg) | ||||
|---|---|---|---|---|---|---|
| Root | Stem | Leaf | Root | Stem | Leaf | |
| AM50 | 0.364 ± 0.057ab | 0.870 ± 0.091a | 0.299 ± 0.014ab | 1.360 ± 0.076ab | 1.387 ± 0.353a | 1.383 ± 0.178a |
| AM30 | 0.437 ± 0.014a | 0.952 ± 0.030a | 0.308 ± 0.022a | 1.583 ± 0.154a | 1.763 ± 0.245a | 1.133 ± 0.200ab |
| AM 15 | 0.346 ± 0.005b | 0.566 ± 0.046b | 0.171 ± 0.003c | 1.230 ± 0.079b | 1.350 ± 0.010a | 1.010 ± 0.101b |
| NM50 | 0.169 ± 0.010c | 0.359 ± 0.056c | 0.238 ± 0.02b | 0.307 ± 0.106d | 0.367 ± 0.045b | 0.373 ± 0.035c |
| NM30 | 0.324 ± 0.044b | 0.349 ± 0.069c | 0.251 ± 0.047ab | 0.570 ± 0.076c | 0.300 ± 0.082b | 0.513 ± 0.091c |
| NM15 | 0.175 ± 0.012c | 0.272 ± 0.043c | 0.139 ± 0.007c | 0.407 ± 0.050cd | 0.273 ± 0.038b | 0.400 ± 0.020c |
| Mycorrhiza | ** | ** | ** | *** | *** | *** |
| Water | ** | ** | ** | ** | NS | NS |
| Mycorrhiza × Water | NS | ** | NS | NS | NS | * |
AM50, AM30, AM15 and NM50, NM30, NM15 stand for 50%, 30%, 15% of field capacity level with/without AM fungi, respectively. Data are presented as the mean ± SD. Significant effect of two-way ANOVA: *P < 0.05, **P < 0.01, ***P < 0.001, NS, no significant effect. Different letters indicate that the means are significantly different between treatments. Analysis of variance followed by Tukey's HSD test when the whole test was significantly different (P < 0.05); n = 6.
The relative anthocyanin content of leaves under AM colonization (AMF) and water stress. Water50, Water30 and Water15 indicate that the water content of the soil was maintained at 50%, 30% and 15% of field capacity level, respectively. Each column represents the mean ± SE. Significant effect of two-way ANOVA: *P < 0.05, **P < 0.01, ***P < 0.001. Different letters indicate that the means are significantly different between treatments. Analysis of variance followed by Tukey's HSD test when the whole test was significantly different (P < 0.05).
LbPT1, LbPT2 and LbPT7 gene expression in leaves at different growth stages. New, juvenile leaves (leaves at the expansion stage); mature, mature leaves (leaves fully expanded); old, senescent leaves (leaves turning yellow, starting with the apex). Each column represents the mean ± SE. Different letters indicate that the means are significantly different between treatments. Analysis of variance followed by Tukey's HSD test when the whole test was significantly different (P < 0.05).
LbPT1, LbPT2, LbPT3, LbPT4, LbPT5 and LbPT7 gene expression under AM colonization (AMF) and water stress. AM50, AM30, AM15 and NM50, NM30, NM15 indicate that the water content of the soil was maintained at 50%, 30% and 15% of field capacity level with/without AM fungi, respectively. Each column represents the mean ± SE. Significant effect of two-way ANOVA: *P < 0.05, **P < 0.01, ***P < 0.001, NS, no significant effect. Different letters indicate that the means are significantly different between treatments. Analysis of variance followed by Tukey's HSD test when the whole test was significantly different (P < 0.05).
(a) A conceptual model showing the functions of LbPTs in roots and shoots. (b) A schematic model for growth performance of L. barbarum and relative expression levels of LbPT1, LbPT2, LbPT3, LbPT4, LbPT5 and LbPT7 in leaves and roots under different soil water content conditions and the presence/absence of AM fungi. The relative expression levels of LbPTs were based on the qPCR results in this study. The results were analyzed using the 2−ΔΔCT method and visualized as a heatmap, which was generated using the package ‘gplots’ in R and Adobe Photoshop CS4 (Adobe Systems Inc., San Jose, CA, USA). The PiTs that complement MB192 are in green font, and the others are in red font.
Discussion
To date, little is known about the PHT1 family members in ligneous plant species belonging to the Solanaceae and their transcriptional response to AM and environmental stress. In the present study, six L. barbarum PHT1 genes were cloned, their functional analysis and subcellular localization were studied using a yeast pho84 mutant (MB192) and tobacco leaves or roots, and their transcriptional responses to Pi deficiency, AM and water stress were also investigated.
Evolutionary relationship of L. barbarum PHT1 genes and members of the PHT1 family in other species
Based on bioinformatics analysis, we confirmed that all the LbPTs had the characteristics of PiT proteins, which are members of the PHT1 family (Rausch and Bucher 2002), part of the major facilitator superfamily (MFS) (Reddy et al. 2012). Phylogenetic analysis revealed that these six LbPTs clustered with previously known Solanaceae PiTs (Kai et al. 2002, Nagy et al. 2005, Chen et al. 2007, Chen et al. 2014). Like other members of the Solanaceae PHT1 family, the six L. barbarum PiTs were assigned to dicot-specific cluster I (LbPT1, LbPT2, LbPT3 and LbPT7) and mycorrhiza-inducible cluster III (LbPT4 and LbPT5), with very high levels of sequence identity in each group. The known regulatory motif P1BS (GNATATNC), an element associated with Pi starvation signaling (Rubio et al. 2001), was present in five of the promoter regions, but was not present in the promoter region of LbPT1. This finding was similar to that of many PHT1 genes in other species (Chiou and Lin 2011). Furthermore, a PHO binding site (CACGTG), an element involved in Pi starvation signaling in the PHO pathway (Oshima 1997), was present only in the LbPT1 promoter region. The LbPT1 promoter region was similar to the OsPT1 of 13 PHT1 genes in rice, except for the absence of P1BS in the promoter region of OsPT1, and the presence of a PHO binding site (Sun et al. 2012). Furthermore, W-box (TTGACY), a WRKY75 binding motif involved in Pi acquisition (Devaiah et al. 2007), was present in the promoter region of LbPT1, LbPT4 and LbPT7. These findings suggested that all six LbPTs were involved in the P signaling network, regulated by upstream transcription factors. In addition, LbPT3 and LbPT4/LbPT5, together with their orthologs from other Solanaceae species, formed independent Solanaceae clades to the exclusion of other dicot homologs, possessing two phylogenetically distant AM-associated PHT1 proteins. Moreover, the mycorrhiza transcription factor binding sequence (MYCS, CTTC-motif: TTTCTTGTTCT; Karandashov et al. 2004) was present in the promoter region of LbPT4, or with one nucleotide variation in the promoter region of LbPT3 and LbPT5. The MYCS in the promoter region of LbPT3 and LbPT5 was identical to that of NtPT3 and SmPT5, respectively (Chen et al. 2011). The presence of the MYCS and P1BS motif suggested that LbPT3–LbPT5 would be induced by AM when the Pi supply was low, like their orthologs from tomato, tobacco, pepper and eggplant (Chen et al. 2011). These characteristics suggested that the L. barbarum PiTs have a similar evolutionary history to that of other Solanaceae species (Chen et al. 2014).
Functional conservation and divergence of PiT proteins in L. barbarum
Since PHT1 genes were first discovered in Arabidopsis thaliana (Muchhal et al. 1996), many PHT1 genes have been identified in various species (Nussaume et al. 2011). Multiple experiments have suggested that PHT1 members not only play a crucial role in Pi uptake, but also participate in Pi translocation and mobilization within/across particular tissues or cells during plant growth (Kai et al. 2002, Sun et al. 2012, Fan et al. 2013, Chen et al. 2014, Li et al. 2015). In the present study, all six PHT1 genes were detected in AM roots; the transcripts of three of these genes (LbPT1, LbPT2 and LbPT7) were also detected in leaves; only very low levels of LbPT2 were detected in leaves. LbPT3, LbPT4 and LbPT5 were highly induced in mycorrhizal roots, but no visible expression was detected in roots of the NM control seedlings by semi-quantitative RT-PCR. This was similar to the expression patterns of orthologs of these genes (PT3, PT4 and PT5) in other Solanaceae species (i.e., potato, tomato, eggplant and tobacco) (Chen et al. 2011). The expression patterns of LbPT3, LbPT4 and LbPT5 were consistent with conjectures based on phylogenetic and promoter region analyses. These analyses indicated that L. barbarum had three mycorrhiza-induced PiT proteins, similar to other solanaceous species (Chen et al. 2011). Functional analysis of LbPT3, LbPT4 and LbPT5 in yeast indicated that only the MB192 cell expressing LbPT3 grew well on a low Pi medium (Figure 3a), which showed the same phenotype as the wild type in the rAPase analysis (Figure 3b). The complement feature of LbPT3, which was the same as that reported for StPT3 of potato, can complement the yeast mutant PAM2, which lacks the high-affinity phosphate transporter (Rausch et al. 2001). LbPT4/5 was similar to StPT4/5 and LePT4/5, which cannot complement PAM2 (Nagy et al. 2005). Meanwhile, the subcellular localization in yeast indicated that LbPT3, a PM-localized protein, was different from LbPT4 and LbPT5. This incorrect localization of LbPT4 and LbPT5 in yeast might result in the loss of the Pi transport function. Like the majority of PiT proteins that have been reported in other plants, LbPT4 and LbPT5 were located in the PM in tobacco leaves whereas LbPT3 did not locate in the PM of tobacco leaves. Furthermore, it was interesting that LbPT3 was located in the PM in tobacco roots. Although LbPT3, LbPT4 and LbPT5 were induced in AM roots, LbPT3 was different from LbPT4 and LbPT5 in terms of its function and localization. In conclusion, functional divergence of LbPT3 and LbPT4/5 was the same as that reported for the solanaceous model plants tomato and potato, PT3 showed high-affinity PiT activity in the yeast pho84 mutant, PT4/5 might be functionally redundant in mycorrhizal Pi uptake (Nagy et al. 2005).
The transcripts of the other PHT1 genes, LbPT1, LbPT2 and LbPT7, were detected in the roots and leaves. Functional analysis of LbPT1, LbPT2 and LbPT7 in yeast indicated that the MB192 cell expressing LbPT1 and LbPT7 grew well on a low Pi medium (Figure 3a), and only the MB192 cell expressing LbPT7 showed the same phenotype as the wild type in the rApase analysis (Figure 3b). These findings suggested that LbPT1 and LbPT7 play a role in Pi uptake like PHO84 in yeast; however, LbPT1 might not have the function of PHO84 in terms of regulating PHO5 expression (Mouillon and Persson 2005). In addition, LbPT2 and LbPT7, similar to StPT2 in potato (Leggewie et al. 1997), were more responsive than LbPT1 to phosphate deprivation. These findings combined with the feature of their promoter region, P1BS was present in LbPT2 and LbPT7 but not in LbPT1 while the PHO binding motif was only present in LbPT1, suggested that LbPT2 and LbPT7 were regulated by a phosphate starvation response pathway (Rubio et al. 2001). LbPT1 might be regulated by a PHO-regulated Pi pathway, like OsPT1, a constitutively expressed PiT protein modulating Pi uptake and translocation in rice (Sun et al. 2012). Furthermore, the investigation of the transcript levels of LbPT1, LbPT2 and LbPT7 at the three different leaf developmental stages revealed that the LbPT7 transcript level was significantly higher in senescent leaves, the LbPT2 transcript level was lowest in mature leaves. The expression patterns of LbPT7 were consistent with that of PhPT1, which is expressed mainly in senescent corolla in petunia (Chapin and Jones 2009), Pht1;5, which is expressed mainly in older leaves at the onset of senescence in Arabidopsis (Mudge et al. 2002, Nagarajan et al. 2011), and OsPT8, which is expressed mainly in the source organs in rice (Jia et al. 2011, Li et al. 2015). These findings suggested that LbPT7 has a function in remobilizing Pi between source and sink organs, similar to the function of PhPT1, Pht1;5 and OsPT8. Furthermore, LbPT2, which has a relatively weak transcript level in leaves, the transcript level of which was analogous to that of LePT2, is expressed in green and ripe tomato fruit (Chen et al. 2014). In contrast to LbPT7, LbPT2 cannot complement MB192, although it was responsive to Pi starvation and its expression in old leaves was noticeable. These characteristics of LbPT2 were similar to those of OsPT2, which is responsible for root–shoot Pi transport in rice (Ai et al. 2009). Meanwhile, the cellular localization analysis in yeast indicated that LbPT1 and LbPT7 were located in the yeast cell membrane while LbPT2 was centralized at several points in the yeast cell membrane. Furthermore, the cellular localization analysis in tobacco leaves indicated that LbPT1, LbPT2 and LbPT7 were PM-localized proteins. In conclusion, our results suggest that LbPT1 modulates Pi uptake and translocation; LbPT2 plays a role in modulating Pi translocation; and LbPT7 modulates Pi uptake and remobilization (Figure 8a).
Diverse role of PiT proteins in L. barbarum to cope with water stress with/without AM fungi
Plants have evolved several strategies for drought avoidance/tolerance, such as investing more nutrients and carbohydrates in the roots, shedding of older leaves, induction of senescence, assimilating remobilization/diversion of nutrients from vegetative to reproductive growth and ceasing growth (Borchert and Rivera 2001, Chaves et al. 2003, Munné-Bosch and Alegre 2004, Jaramillo et al. 2013). During leaf senescence caused by water stress, the anthocyanin content increases, and the ordered degradation of the cellular content occurs, resulting in the remobilization of nutrients (Lim et al. 2007). As plants try to cope with water stress, investing more nutrients and carbohydrates in the roots and remobilizing nutrients from the senescent leaves, the essential macronutrient P has to be taken into consideration. With regards to P metabolic processes, PiT proteins play a crucial role not only in Pi uptake, but also in translocation and remobilization (Rausch and Bucher 2002, Nussaume et al. 2011, Fan et al. 2013, Chen et al. 2014).
In the absence of AM fungi, seedlings that experienced severe water stress accelerated the accumulation of anthocyanin and significantly suppressed leaf growth, indicating that water stress exacerbated leaf senescence. Similar results have been reported for A. thaliana (Sperdouli and Moustakas 2012). The transcript levels of LbPT1 and LbPT7 were higher in the senescent leaves and, consequently, the relative expression of LbPT1 and LbPT7 in leaves was enhanced under water stress, particularly the expression of LbPT7. These findings suggested that Pi was being transported from the senescent leaves to the rest of the plant (i.e., the youngest leaves or roots) under water stress conditions (Munné-Bosch and Alegre 2004). Accordingly, we speculated that LbPT1 and LbPT7 played a role in translocating Pi from the senescent leaves to other developing tissues, which may be specific for the perennial lifestyle of trees (Loth-Pereda et al. 2011). As the water content of the soil decreases, the Pi availability also decreases (Sardans and Peñuelas 2004), and under severe water stress conditions plant growth ceases (Chaves et al. 2003). In addition, the higher P content and greater root and leaf dry weights might correlate with the up-regulation of LbPT1, LbPT2 and LbPT7 under moderate water stress. The biomass and P content of L. barbarum roots increased initially and then decreased as the water stress became more severe, which suggest that the plant invested more carbohydrate and nutrients in the roots under moderate water stress, but reduced the investment under severe water stress. This may explain the transcriptional patterns of LbPT1, LbPT2 and LbPT7 in roots under water stress, which were upregulated initially as nutrients from the aerial parts were invested in the roots and for Pi uptake from the soil, and then expression decreased sharply as plant metabolism almost ceased as a result of severe water stress.
The P content and the biomass (except for leaves) of plants that received the AM treatment were enhanced. Water stress did not affect the anthocyanin content significantly, indicating that seedlings that formed an AM association were protected against water stress, which may be because Pi uptake was improved as a result of the AM association, as well as other non-nutrient benefits (Bolan 1991, Ruiz-Lozano et al. 2012). The Pi uptake by AM fungi was due to the expression of mycorrhiza-induced PHT1 genes, LbPT3, LbPT4 and LbPT5, whose transcript levels were unaffected under water stress. This result indicated that the mycorrhizal Pi uptake pathway was efficient, regardless of water stress intensity. The improved nutritional status may help mycorrhizal plants to acquire water from a larger area of soil and to delay leaf senescence under water stress. In addition, the transcript levels of LbPT1, LbPT2 and LbPT7 were suppressed in seedlings that received the AM-treatment. This finding was similar to the expression patterns of three orthologous transporters (LePT1, LePT2 and LePT7) in tomato roots (Chen et al. 2014): the expression of LePT1, LePT2 and LePT7 was lower in AM roots than in non-AM roots. It is interesting that LbPT1 was upregulated by water stress in AM roots, which suggests that LbPT1 may play a role in Pi translocation from roots to shoots; water stress exacerbated Pi translocation from the roots to the shoots in AM roots. These findings suggested that the supply of Pi by AM fungi was sufficient to satisfy the demand for Pi in the roots and replaced the direct Pi uptake of the plant by non-AM-induced PiT proteins in the roots (Smith et al. 2003, 2011).
To summarize, once water stress has occurred, the plant has several strategies that can be used to cope with water deficiency and the fundamental Pi requirement, such as greater investment of nutrients in the roots, induction of leaf senescence, remobilization of Pi from senescent organs to the rest of the plant, the acceleration of Pi uptake in roots and Pi translocation from shoots to roots (Schachtman et al. 1998). Our findings suggest that seedlings that have formed a symbiotic relationship with AM fungi can obtain enough Pi to meet the plant's requirements and, therefore, can economize on energy for Pi uptake and translocation/remobilization. This suggests that mycorrhizal plants rely more on the mycorrhizal Pi uptake pathway and reduce the translocation/remobilization of Pi by PiT proteins. Furthermore, an adequate nutrient supply enables the host plant to explore the soil for water more effectively.
Conclusions
Here, we characterized six PHT1 genes of L. barbarum, which had a similar evolutionary history to that of other annual herbaceous plant species belonging to the Solanaceae family. The PiT proteins encoded by these PHT1 genes play distinct roles in Pi acquisition and translocation in leaves and roots at different development stages and under different growth conditions. Three of these PiT proteins (LbPT3, LbPT4 and LbPT5) are involved in the mycorrhizal Pi pathway. The results of this study suggested that the expression patterns of the non-mycorrhiza-induced PiT proteins (LbPT1, LbPT2 and LbPT7) were induced by Pi deficiency and affected by water stress, and that the mycorrhiza-induced PiT proteins (LbPT3, LbPT4 and LbPT5) were inhibited by Pi supply but not regulated by water stress. The non-mycorrhiza-induced PiT proteins are likely to play a role in Pi translocation and remobilization along with leaf senescence and the investment of nutrients in roots to cope with water stress. Mycorrhiza-induced PiT proteins play a role in Pi uptake with AM fungi, which is unaffected by water stress. Future studies should investigate the expression patterns of mycorrhiza-induced PHT1 genes under other abiotic stresses, such as salt or heavy metal stress.
Supplementary Data
Supplementary Data for this article are available at Tree Physiology Online.
Acknowledgments
We thank Prof. Zhiying Zhao for providing the yeast expression vector pGADT7, Prof. Weixing Shan for providing the transient expression vector modified pCambia0380, and Dr Xianan Xie for advice about the yeast complementation analysis.
Author contribution statement
W.H. and M.T. conceived and designed the research study. W.H., H.Z. and X.Z. conducted the experiments. H.C. analyzed the data. W.H. and H.Z. wrote the manuscript. All authors read and approved the manuscript.
Conflict of interest
The authors declare that they have no conflict of interest.
Funding
This research was supported by the National Natural Science Foundation of China (41671268, 31270639), the Program for Changjiang Scholars and Innovative Research Team in University of China (IRT1035), the Shaanxi Science and Technology Innovation Project plan (2016KTCL02-07) and the Northwest A&F University doctoral research start-up fund (Z109021503).
