HAN Jiao, YU Guo-hong, WANG Li, LI Wei, HE Rui, WANG Bing, HUANG Sheng-cai, CHENG Xianguo

1 College of Life Science, Shanxi Normaluniversity, Linfen 041000, P.R.China

2 Institute of Agricultural Resources and Regional Planning, Chinese Academy of Agricultural Sciences, Beijing 100081, P.R.China

3 College of Land and Environment, Shenyang Agriculturaluniversity, Shenyang 110866, P.R.China

Abstract Phosphate transporters play an important role in promoting the uptake and transport of phosphate in plants. In this study,the McPht gene from the Mesembryanthemum crystallinum, a mitochondrial phosphate transporter, was isolated and constructed onto a constitutive expression vector carrying 35S::GFP, and the recombinant constructs were transferred into Oryza sativa japonica L. cv. Kitaake to investigate the regulatory role of the McPht gene under phosphorus deficiency. The McPht gene encodes a protein of 357 amino acids with six transmembrane domains and is located to the mitochondria, and the mRNA transcripts of the McPht gene are highly accumulated in the shoots of M. crystallinum in response to phosphorus deficiency. However, more mRNA transcripts of the McPht gene were accumulated in the roots of the transgenic rice under phosphorus deficiency. Measurements showed that the transgenic rice demonstrated an enhanced promotion in the root development, the root activities, and phosphate uptake under phosphorus deficiency. Transcriptome sequencing showed that the transgenic rice exhibited totalof 198 differentially expressed genes. Of these, totalof 154 differentially expressed genes were up-regulated and total 44 genes were down-regulated comparing to the wild type in response to phosphorus deficiency. The selective six genes of the up-regulated differentially expressed genes showed an enhanced increase in mRNA transcripts in response to phosphorus deficiency, however, the transcripts of the mitochondrial carrier protein transporter in rice, a homologous gene of the McPht, in both the transgenic line and the wild type had no obvious differences. Functional enrichment analyses revealed that the most of the up-regulated genes are involved in the cytoplasmic membrane-bounded vesicle, and most of the down-regulated genes are involved in the mitochondrion and cytoplasmic membrane-bounded vesicle. The differentially expressed genes were highly enriched in plant secondary metabolisms and plant-pathogen interaction. These results indicated that the overexpression of the McPht gene might participate in the physiologicaladaptive modulation of the transgenic rice to phosphorus deficiency by up- or down-regulating the differentially expressed genes.

Keywords: McPht gene, phosphorus deficiency, phosphate transporter, transcriptome sequencing, transgenic rice

1. Introduction

Phosphorus (Pi) nutrition is essential for plant growth and development, and widely involved in the physiologicalbiochemical metabolic processes (Richardson 2009). As an important component of nucleic acids and phospholipids,phosphate plays important regulatory roles in the photosynthesis, energy metabolism, signal transduction,gene expressions, and enzymatic reactions (Yang and Finnegan 2010). The phosphate uptake by plants depends on the amounts of available phosphate, which is usually present in the forms of HPO42-and H2PO4

-, and phosphate availability is relatively limited in soils although the phosphate contents exceed 40 mmol L-1in plants (Bollons and Barraclough 1997), thus indicating that the phosphate uptake by plant is an inverse process accompanying concentration gradient energy dissipation. Therefore, the establishment of an adaptive mechanism for improving phosphate uptake efficiency (PUE) is necessary by altering root morphology and accumulation of anthocyanins under phosphorus deficiency (Raghothama 1999).

The uptake and transport of phosphate in plants are mainly participated by the high-affinity phosphate transporter and the low-affinity phosphate transporter (Christine et al.2002). Currently, the phosphate transporters are divided into the Pht1, Pht2, Pht3, and Pht4 subfamilies (Lin et al. 2009).The Pht1 subfamily is composed of a cluster of high affinity phosphate transporters, which contain 12 transmembrane regions with the N- and C-terminal ends mounting inside of cell (Raghothama 1999; Bucher et al. 2001; Christine et al. 2002). Many Pht1 subfamily members have been functionally profiled in tomato (Xu et al. 2007), Arabidopsis(Nagarajan et al. 2011), and rice (Ai et al. 2009). Total 13 of phosphate transporters in rice were confirmed to be the Pht1 subfamily (Christine et al. 2002). Both the Pht1;1 and Pht1;4 are highly expressed in the roots of Arabidopsis under phosphorus deficiency (Shin et al. 2004). The transgenic rice overexpressing OsPht1;4 revealed higher Pi concentrations in the roots and shoots under different phosphate levels compared with the wild type (Zhang et al.2015). Similar to the Pht1, the Pht2 subfamily members also posses 12 transmembrane regions with a central hydrophilic loop in the middle at the 8th and 9th regions (Daram et al.1999), but most of the Pht2 subfamily members have been identified to function as low-affinity phosphate transporters such as AtPht2;1 (Daram et al. 1999), MtPht2;1 (Zhao et al.2003), and TaPHT2;1 (Guo et al. 2013). Study showed that the PHT2;1 affected the phosphate accumulation and gene expression in the leaves under phosphorus deficiency (Versaw and Harrison 2002). The expression of the MtPHT 2;1 gene in leaves was confirmed to be closely related to the supply levels of phosphate and located to the chloroplast envelope (Zhao et al. 2003). Unlike the Pht1 and Pht2, the Pht3 subfamily members belong to the high-affinity phosphate transporters which are composed of six transmembrane domains, and have been functionally characterized in soybean, maize, rice, and Arabidopsis(Takabatake et al. 1999). Stappen and Kramer (1994)reported that the Pht3 subfamily functions in promoting the phosphate exchange through the co-transport of Pi/H+and the reverse transport of Pi/OH-in cells. The mitochondrial phosphate transporter (MPT) gene in Arabidopsis has been identified. More accumulation of the AtMPT mRNA led to higher ATP contents in the transgenic Arabidopsis, and a large number of genes is related to gibberellin synthesis and regulated by changing available energy under salt stress (Zhu et al. 2012). Like the Pht3, the Pht4 subfamily members are also composed of 6 transmembrane regions, but most of them were functionally identified in Arabidopsis. The PHT4 gene was mainly expressed in the leaves and roots of Arabidopsis thaliana, and confirmed to be located to the plastid envelope or the Golgi apparatus (Guo et al. 2008).

Rice is one of the most important food crops, and has been used as model for gene functionalanalyses. In this study, the McPht gene, a mitochondrial phosphate transporter, was isolated from Mesembryanthemum crystallinum, which has an ecologicaladaptation to high saline soils or infertile soils or desert soils (Adams et al.1998), thus showing that M. crystallinum reveals stronger adaptive mechanism to phosphorus deficiency and phosphate transporters might play crucial regulatory roles in the uptake of limited available phosphate in poor soils.To date, no detailed study on mitochondrial phosphate transporter from M. crystallinum was reported (Winter and Holtum 2007). Therefore, we constructed a plant transformation vector carrying the McPht gene and 35S promoter, and transferred this construct into rice cv. Kitaake(Oryza sativa japonica L.), and three T2generations of the transgenic rice lines with stable hereditary were selected.Both the transgenic lines and the wild type were cultured in hydroponics to investigate the changes of molecular responses, the physiological metabolism, and transcriptome profiles in the transgenic rice under phosphorus deficiency,and thus evaluating possible regulatory role of the McPht gene in promoting the phosphate uptake. Our data showed that the overexpressions of the McPht gene might confer an adaptive modulation in enhancing the uptake and transport of phosphate in the transgenic rice under phosphorus deficiency.

2. Materials and methods

2.1. Plant materials and stress treatments

Rice seeds from T2generation of the transgenic rice and the wild type were germinated on the glass culture dishes by fresh water in dark at 30°C for 4 d, and transplanted into a net plastic plates mounting in the containers, and continuously cultured in a growth chamber at 28°C with 70%humidity for 1 wk (Zheng et al. 2013). Uniform seedlings were selected and transferred to the box containing 10 loffresh Hoagland’s nutrient solution for further culture in a greenhouse at 28°C with 70-80% relative humidity and light 12 h/dark 12 h for 1 wk. The rice seedlings of 1-wk-old were separately subjected to the nutrient solution of 4 μmol L-1P2O5(phosphorus deficiency) or the nutrient solution of 12.5 mmol L-1P2O5(sufficient phosphorus), and continuously cultured for 4 wk, and plant samples were collected for RNA extraction and physiological measurements. The seeds of M. crystallinum L. were sowed in the vermicular and peat soil mixture of 1:1, and cultured at 23°C with relative humidity of 70% and 14 h light/10 h dark for 4 wk in a growth chamber.M. crystallinum seedlings of the 4-wk-old were subjected to phosphorus deficiency for 2 wk at the same culture condition.All treatments were designed by three biological replicates in this study.

2.2. Plant transformation

The McPht cDNA fragments were obtained by PCR using the specific primer (Table 1), and inserted into the BamHI and PstI sites of a pCAMBIA1305 expression vector.The resulting construct of McPht::pCAMBIA1305 was transformed into Agrobacterium tumefaciens GV3101, and the transformants were used to infect callus of Kitaake rice(O. sativa japonica L.). The infected callus was cultured under sterile condition, and the generating seedlings were transplanted in the mixture of vermicular and peat soils (1:2),and all calluses were continually cultured for regeneration and further selected on the MS medium containing hygromycin by standard protocols (Matsumoto et al. 2009),and three transgenic rice lines of T2generation with stable hereditary were selected for subsequent experiments.

2.3. RNA extraction and quantitative RT-PCR

Total RNA of plants were extracted with the EasyPure Plant RNA Kit (TransGen Biotech, China), and RNA concentrations were determined by ND2000 spectrophotometer (Thermo Scientific Co., USA). Total 1 μg of RNA was reverse-transcripted into the cDNA using TransScript One-Step gDNA Removaland cDNA Synthesis SuperMix (TransGen Biotech, China) according to the manufacturer’s manual. Quantitative PCR procedures were performed in a 20-μlof reaction solution containing 10 μlof 2×TransStart®Top Green qPCR Super Mix, 0.4 μlof sense and antisense primers (Table 1), 1 μlof cDNA template and 8.2 μlof RNA-free water using the TransStart Top Green qPCR SuperMix (TransGen Biotech, China). Quantitative PCR procedures were performed with the IQ5 Multicolor Real-Time PCR Detection System (Bio-Rad, Richmond, CA)under the following circles: 94°C for 30 s, 45 cycles of 94°C for 5 s, 55°C for 15 s, and 72°C for 10 s according to the manufacturer’s instructions. The relative expression levels of targets genes in the transgenic rice were calculated by the 2-∆∆CTformula (Livak and Schmittgen 2001) using OsActin as an internal control. The relative expression levels of the McPht gene in M. crystallinum L. were calculated by the 2-∆∆CTformula using McActin as an internal control. All nucleotides sequences of the specific primers for qRT-PCR analyses were shown in Table 1.

Table 1 Gene-specific primers used in this study

2.4. Subcellular localization analysis

The McPht cDNA fragments were amplified by RT-PCR using a pair of specific primers with PstI/NcoI sites (Table 1),and inserted into the 16318hGFP expression vector under the controlof the 35S promoter with PstI/NcoI-specific sites(Wang et al. 2013). Both the McPht::GFP fusion construct and the 16318hGFP empty vector were transformed into the leave protoplasts of the 3-wk-old Arabidopsis by a PEG-mediated method (Locatelli et al. 2003). The transformants were placed at room temperature over 18 h, and the images of the transgenic protoplasts were photographed by a LSM700 Confocal Laser Microscope (Carl Zeiss AG,Oberkochen, Germany).

2.5. Roots recording and physiological measurements

Roots scanning was performed by a scanner of EPSONLA,UK, and the data were analyzed by WINRHIZO Pro software of 2004c version (WINRHIZO, Regent Co., Canada). Roots fresh weights were weighed under natural dehydration. Root activities were measured at 485 nm by triphenyltetrazolium chloride (TTC) method using a spectrophotometer (HACH DR/4000U, USA) (Lindström and Nyström 1987). Total 0.2 g of dry matter were weighted, and digested in the solution containing 5 mlof concentrated H2SO4and 2 mlof H2O2,and then the generating colorless digestion solutions were cooled and diluted to the volume of 100 mL by distilled water for determination of total phosphate in plants.

2.6. cDNA library construction and transcriptome sequencing

Fresh plant leaves were sampled and stored in liquid nitrogen, and approximately 1 μg of RNA was extracted for constructing the cDNA library, which was generated using NEBNext®Ultra™ RNA Library Prep Kit for Illumina®(NEB,USA) following the manufacturer’s recommendations. The mRNA preparation was performed by poly-Toligo-attached magnetic beads. Fragmentation was carried out using divalent cations under elevated temperature in NEB Next First Strand Synthesis Reaction Buffer (5×). the first strand cDNA was synthesized using random hexamer primer and M-MuLV Reverse Transcriptase (RNase H-). The second strand cDNA synthesis was subsequently performed using DNA Polymerase I and RNase H. Remaining overhangs were converted into blunt ends via exonuclease and polymerase. Alladaptors were ligated after adenylation of DNA fragments at 3´-end. The library fragments were purified with AMPure XP System (Beckman Coulter, Beverly,USA). Agilent 2100 Bioanaylzer (Agilent Technologeis Co., USA) and ABI StepOnePlus Real-Time PCR System(Applied Biosystems ABI Co., USA) are used for qualifying the sample library and transcriptome sequencing.

2.7. Sequencing datafiltering

Sequencing data were gained by Illumina HiSeq TM2000 platform with PE125 method. The original images were transformed into the raw reads by base recognition, and the raw reads werefiltered to obtain clean reads, which further were assembled using Software Trinity Program(Grabherr et al. 2011). All contigs were generated by the assembly of the information between overlap sequences,and then partially data were assembled, and the unigenes were produced using non-redundant protein sequences (Nr),clusters of orthologous groups of proteins (COG), euKaryotic ortholog groups (KOG), gene ontology (GO), and Kyoto encyclopedia of genes and genomes (KEGG).

2.8. Differentially expressed gene, GO enrichment,and KEGG analyses

The reads of gene number were counted by the HTSeq software. The gene expression levels were calculated by the fragment per kb per million fragments (FPKM) method(Trapnell et al. 2010). Both DESeq and DESeq2 were used to detect the differentially expressed genes between the wild rice and the transgenic rice. Differences in genetic screening conditions were indicated by the difference in multiples ≥2 and q-value (or FDR)≤0.01. GOseq was used for GO enrichment analysis of differentially expressed genes, and GO analysis with p-value≤0.05 were assigned as differentially expressed genes (Young et al. 2010). Pathway enrichment analysis of DEGs was performed by the KEGG(the major public pathway-related database).

3. Results

3.1. The McPht gene is a member of Pht3 subfamily and located to the mitochondria

McPht cDNA encodes a protein of 357 amino acids with the molecular weight of 37.77 kDa and isoelectric point of 9.30 (accession no: AGC00815.1). The McPht phosphate transporter is composed of six conserved domains, in which the 1st, 2nd, 5th, and 6th transmembrane domains are separately located outside the membrane (Fig. 1-A).PHYRE2 analyses showed that the McPht phosphate transporter has three tandem domains with different hydrophilicity regions and six alpha helices (Fig. 1-B).Multiple alignment showed that the McPht phosphate transporter exhibits more than 80% identity with the Pht3 members from 15 dicotyledonous plants (Fig. 1-C), and is a typical homolog of mitochondrial phosphate transporters(Fig. 2), indicating that the McPht gene is a member of the Pht3 subfamily.

To localize the McPht protein, the 16318hGFP vector and the McPht::GFP fusion construct under the controlof the 35S promoter were separately transferred into Arabidopsis protoplasts by PEG-mediated method. As shown in Fig. 3,the transformant carrying the McPht::GFP construct seems to exhibit strong fluorescence signals on the mitochondria,but the transformant carrying the 16318hGFP vector revealed entire green fluorescence signals throughout Arabidopsis protoplasts, suggesting that the McPht protein was very likely localized to the mitochondria.

Fig. 1 Structural profiles of McPht protein and phylogenetic analysis of McPht gene. A, transmembrane profiles of McPht phosphate transport protein. B, structural characteristics of the McPht protein. C, phylogenetic tree analysis.

3.2. Relative expression profiles of the McPht gene under phosphorus deficiency

To profile the relative expressions of the McPht gene in both M. crystallinum and the transgenic rice under phosphorus deficiency, the quantitative PCR procedures were performed.The data showed that phosphorus deficiency affected the expression profiles of the McPht gene in roots and shoots of plants. As shown in Fig. 4-A, phosphorus deficiency induced the expression of the McPht gene in the roots of the transgenic lines OE-1, OE-5, and OE-7, and the transcripts of the McPht gene increased by 8.96, 17.21, and 1.32 folds comparing to sufficient phosphate supply, respectively.However, the relative expression levels of the McPht gene in the shoots of the transgenic lines seemed to be remarkably lower than those in the roots under phosphorus deficiency.Especially, the OE-1 line showed an enhanced decrease of 0.79 folds in the mRNA accumulations of the McPht gene in the shoots, and both the OE-5 and OE-7 lines slightly increased the mRNA accumulations of the McPht gene by 1.02 and 0.97 folds relative to those under sufficient phosphorus supply, respectively (Fig. 4-B). Similarly, the relative expression of the McPht gene in the shoots of M. crystallinum were significantly higher than those in the roots under phosphorus deficiency (Fig. 4-C). These data suggested that mRNA transcripts of the McPht gene in the shoots are higher than those in the roots under phosphorus deficiency.

Fig. 2 Homology profiles between the McPht and homologous Pht3 proteins. The pink colors indicate 75% amino acid similarity and blue colors indicate 100% amino acid similarity.

Fig. 3 Subcellular localization of the McPht. A, the expressed McPht::GFP fusion protein signals in the Arabidopsis mesophyll protoplasts. B, the control expressing GFP (green fluorescent protein, 16318hGFP) signals in the Arabidopsis mesophyll protoplasts.

Fig. 4 The relative expression levels of the McPht gene under phosphorus deficiency. A, relative expressions in the roots. B,relative expressions in the shoots. C, the relative expressions of the McPht gene in the roots and shoots of Mesembryanthemum crystallinum. OE-1, OE-5, and OE-7, transgenic lines. All data represent mean±SD of three replicates.

3.3. Overexpression of the McPht gene referred to an accommodation of the transgenic rice to phosphorus deficiency

To functionally evaluate the McPht gene in the transgenic rice in response to phosphorus deficiency, both the transgenic rice and the wild type were continuously cultured by the pots containing 10 lof Hoagland’s nutrient solutions for 4 wk, and then rice seedlings were subjected to sufficient phosphorus supply of 12.5 mmol L-1P2O5(Fig. 5-A) and phosphorus deficiency of 4 μmol L-1P2O5(Fig. 5-B), and further cultured for 2 wk. Photo images showed that three transgenic rice lines revealed better growth than the wild type under sufficient phosphorus supply (Fig. 5-C) or phosphorus deficiency (Fig. 5-D). Roots scanning showed that the total root lengths of the transgenic lines were significantly higher than those of the wild type under phosphorus deficiency(Fig. 6-A), and the lateral root number of the transgenic lines OE-1 and OE-5 were higher than those of the wild type (Fig. 6-B). Conversely, the total roots length and lateral roots number of the transgenic lines were lower than those of the wild type under sufficient phosphorus supply(Fig. 6-B), and total roots projected area of the transgenic rice significantly higher than those of the wild type under phosphorus deficiency (Fig. 6-C). While the roots biomass of the transgenic rice seems to be corresponding to the changes of roots architecture (Fig. 6-D). These data indicated that the transgenic lines carrying the McPht gene might induce the changes of roots morphology in response to phosphorus deficiency.

Fig. 5 Phenotypic images of the transgenic lines and the wild type. A and C, phenotypes of the wild type and the transgenic rice OE-1, OE-5, and OE-7 under supply of sufficient phosphorus. B and D, phenotypic images of the wild type and the transgenic rice OE-1, OE-5, and OE-7 under phosphorus deficiency.

Fig. 6 Morphologic parameters and activities of the roots and total phosphate contents. A, the total roots length. B, the number of lateral roots. C, total roots projected area. D, roots weights. E, the roots activities. F, total phosphate contents per plant. All data represent mean±SD of three replicates.

Physiological measurements showed that the roots activities of the transgenic rice were higher than those of the wild type under phosphorus deficiency. Comparing to wildtype, the root activities of the transgenic lines OE-1, OE-5,and OE-7 were increased by 0.53, 0.47, and 1.27 folds under phosphorus deficiency, respectively (Fig. 6-E). However,the wild type demonstrated higher roots activities than the transgenic rice under sufficient phosphorus supply. Both sufficient phosphorus supply and phosphorus deficiency led to an enhanced increase in the contents of total phosphate in the transgenic lines compared with the wild type (Fig. 6-F).Under phosphorus deficiency, the transgenic rice lines,OE-1, OE-5, and OE-7, separately showed an increase of 1.24, 1.19, and 1.15 mg g-1in phosphate uptake per plant(Fig. 6-F).

3.4. Relative expressions of differentially expressed gene

Transcriptome sequencing showed that total 21 578 genes in the wild type and 21 935 genes in the transgenic lines were detected, total 21 119 genes were co-expressed,total 459 genes were differentially expressed in the wild type and total 816 genes were differentially expressed in the transgenic lines in response to phosphorus deficiency,respectively (Fig. 7-A). Statisticalanalyses showed that total 198 genes were differentially expressed in the transgenic lines, and 154 genes of these differentially expressed genes were up-regulated and 44 genes were downregulated comparing to the wild type (Fig. 7-B). To profile the expressions of the McPht gene’s homologous gene in the transgenic rice line under phosphorus deficiency, we performed qRT-PCR to determine the mRNA accumulation of the Os04g0448800 gene, a mitochondrial carrier protein gene (accession no. AP014960.1). Data showed that both the transgenic line and the wild type have no obvious differences in mRNA accumulations of the Os04g0448800 gene in the shoots (Fig. 7-C). Additionally, we selected six genes of the differentially expressed genes, which were up-regulated in the transgenic rice (Table 2), to determine the expression profiles of these genes in response to phosphorus deficiency. qRT-PCR showed that the transgenic rice exhibits an enhanced increase in mRNA accumulations of these differentially expressed genes comparing to the wild type, and the relative expression levels of Os02g0602300, Os07g0650600, Os02g0106100,Os03g0348200, Os10g0190800, and Os08g0157500 genes increased by 3.38, 1.52, 0.55, 1.37, 1.09, and 0.23 folds in response to phosphorus deficiency, respectively(Fig. 7-C). This result seems to be identical with the RNAseq analyses of transcriptome (Table 2). As shown in Table 2,these differentially expressed genes are mainly involved in carbohydrate metabolic process, oxidoreductase activity,flavone 3’-O-methyltransferase 1, magnesium ion binding,isoaspartyl peptidase/L-asparaginase 2, alpha-humulene synthase, threonine-protein kinase, transmembrane receptor protein serine/threonine kinase signaling pathway,and dephosphorylation 5´-nucleotidase activity.

Fig. 7 Expression profiles and distributions of endogenous genes in the transgenic rice. A, gene expression of Venn diagrams.Each circle represents the total number of detected genes, and the overlapped portions of the circles indicate the detected genes in both samples. OE, line 1 (OE-1); WT, wild type. B, the distribution of differentially expressed genes (DEGs). Each point represents a gene, and red points represent the genes of up-regulation, blue points represent the genes of down-regulation, and green points represent the co-expressed genes. FDR, false discovery rate. Diff, differential expression. C, relative expression levels of endogenous genes in rice under phosphate stress. The WT was used as the control. All data represent mean±SD of three replicates.

3.5. GO functional enrichment analysis on differentially expressed genes

Functional enrichment analyses of differentially expressed genes were performed by the Gene Ontology Database and the Japan Genome Annotation Database. The results showed that total 198 DEGs between the transgenic lines and the wild type were functionally annotated, and total 30 Go Terms of DEGs were involved in the biological functions.In detail, these DEGs were separately enriched by 17 Go Terms involved in molecular function (MF), 10 Go Terms involved in biological process (BP), and three Go Terms involved in cellular component (CC) (Fig. 8-A). The upregulated genes participating in the molecular function mainly are gathered in the most enriched terms like peptide receptor activity, transmembrane receptor protein serine/threonine kinase activity ubiquitin ligase binding protein,ubiquitin protein ligase binding, and transmembrane receptor protein serine/threonine kinase activity. The down-regulated genes were enriched by calcium ion binding and O-methyltransferase activity (Fig. 8-B). The genes involved in the biological process were mainly enriched by transmembrane receptor protein serine/threonine kinase signaling pathway, hormone-mediated signaling pathway,and protein autophosphorylation. Most of the up-regulated genes were represented by the hormone-mediated signaling pathway, protein autophosphorylation, and transmembrane receptor protein serine/hreonine kinase signaling pathway. In the cellular component, the cytoplasmic membrane-bounded vesicle demonstrated the highest enrichments of DEGs,and followed by plasmodesma and plasma membrane.Most of the up-regulated genes were represented by the cytoplasmic membrane-bounded vesicle, and most of the down-regulated genes were represented by the clusters of the mitochondrion and cytoplasmic membrane-bounded vesicle.

3.6. Pathway enrichment analyses of differentially expressed genes

To further understand the enrichment profiles of differentially expressed genes involved in metabolic pathways, the differentially expressed genes were compared against the KEGG database by pathway enrichment analysis.The enrichments of 30 pathway terms were selected to statistically determine the DEGs number of up- or downregulation in each pathway term (Fig. 9). The largest cluster of up-regulated genes was revealed by plant-pathogen interaction, and the number of the down-regulated genes was lower than the number of the up-regulated genes.Data showed that these metabolic pathways are mainly involved in the plant secondary metabolic pathways such as phenylpropanoid synthesis, phenylalanine metabolism,and the secondary metabolite synthesis.

4. Discussion

4.1. McPht gene is induced by Pi deficiency and participates in the growth regulation of the roots

Homology alignment showed that the McPht gene from M. crystallinum demonstrates a typical structure of the Pht3 subfamily members containing six transmembrane domains, and is a homolog of the mitochondrial phosphate transporters, which have been confirmed to be located in the mitochondria of plant cells (Stappen and Kramer 1994). Studies have shown that phosphate transporters play a positive role in plant growth under environmental stress (Ai et al. 2009; Li et al. 2015). Our data showed anenhanced increase in the expression of the McPht gene under phosphate deficiency, indicating that phosphate deficiency seems to induce the expression of the McPht gene. Wen et al. (2014) reported that phosphorus deficiency increased the expression levelof the OsPT6 gene, and the transgenic vegetable soybean overexpressing the OsPT6 gene had higher root weights than the wild type under lowphosphate stress. Previous reports showed that phosphate transporters were preferentially expressed in the roots under phosphorus deficiency (Daram et al. 1998; Liu et al. 1998).

Table 2 A partial descriptions of the differentially expressed genes (DEGs)

Fig. 8 Histogram of the most enriched gene ontology (GO) terms and differentially expressed genes (DEGs) number. A, the DEGs enrichment GO terms. Different colors represent biological process (BP), cellular component (CC), and molecular function (MF).＊ means significant enrichment GO term. B, statistics of up-regulation or down-regulation DEGs of enriched GO term.

Fig. 9 Statistics of the up- or down-regulation of differentially expressed genes (DEGs) enriched pathway term.

Study showed that both the low affinity phosphate system and the high affinity phosphate system are two essential pathways for the uptake and transport of phosphate in plants, and thus leading to an increase in biomass (Furihata et al. 1992). The high-affinity systems are induced by low-phosphorus conditions (Furihata et al. 1992), while the low-affinity system seems to be constitutive in plants(Raghothama 1999). Plaxton and Carswell (1999) reported that multiple plasmalemma phosphorus transporters differentially exhibit expression under varying phosphorus nutritional regimes. Interestingly, the high-affinity transporter mRNA transcripts in the roots were increased under phosphate deficiency because of an enhanced increase in the root capacities for P uptake (Duncan and Carrow 1999;Shenoy and Kalagudi 2005). These studies suggested that the high-affinity transporters play a critical role in the phosphate acquisition under phosphorus deficiency. In our study, the McPht gene is a typical member of the highaffinity phosphate transport systems, and might be induced at low external phosphate concentration, thereby increasing the relative expression levels of the McPht gene in the transgenic rice comparing to sufficient phosphorus supply.Our data showed that phosphorus deficiency induced the changes of roots architecture and remarkably promoted the roots growth of the transgenic rice, and increased the roots activities and the ratio of the roots to shoots of the transgenic rice although no detailed data were shown.Therefore, we think that the differential expression of the McPht gene seem to be caused by the changes in exudation levels of organic acids from the roots under different levels of phosphorus supply.

Zhou et al. (2008) found that the transgenic rice overexpressing the OsPHR2 gene demonstrated an increase in the root length under phosphorus deficiency compared with the wild type. Our data showed that the transgenic rice overexpressing the McPht gene increased the root length and lateral root number in response to phosphorus deficiency than the wild type, and similar result also has been reported (Ren et al. 2014). This study indicated that the overexpression of the McPht gene in the transgenic rice increased the root activities, thus probably promoting the root growth and the root respiration under phosphorus deficiency. Wang et al. (2014) confirmed that the transgenic rice overexpressing the OsPHT1;9 and OsPHT1;10 genes significantly increased the phosphate uptake than the wild type in responses to phosphorus deficiency or excess phosphorus.

4.2. The McPht gene regulates the differentially expressed genes in the secondary metabolites

Specific changes in different metabolic pathways and signal transduction pathways were associated with regulatory profiles of the differentially expressed genes under phosphorus deficiency (Marioni et al. 2008; Mortazavi et al.2008). In our study, the differentially expressed genes are mainly involved in plasmodesma, plasma membrane,cytoplasmic membrane-bounded vesicle, oxidoreductase,transmembrane receptor kinase, the secondary metabolites,sugar metabolism, and the other biological processes.Study showed that differentially expressed genes in maize are mainly involved in sugar synthesis, protein synthesis, amino acid degradation, and the secondary metabolic pathways under phosphorus deficiency (Sun et al. 2016). The up-regulation of genes associated with metabolic pathways contributed to the plant growth, and the differentially expressed genes were related to amino acid metabolism, which supplies a source of C in barley roots under phosphorus deficiency (Carlos et al. 2008; Huang et al. 2008), suggesting that these differentially expressed genes might involve in the metabolism processes of C and N to maintain phosphorus balance in plant cells under phosphorus deficiency. Our results showed that phosphorus deficiency triggered a series of changes in plant secondary metabolic pathways and plant-pathogen interaction. Sun et al. (2016) reported that phosphorus deficiency affected various metabolic pathways, the biosynthesis of the secondary metabolites, and plant-pathogen interactions in maize, and also increased accumulation of anthocyanin in the plants (Zhang et al. 2014).

5. Conclusion

The McPht protein is located to the mitochondria, and belongs to a member of the Pht3 subfamily. Phosphorus deficiency induced the expressions of the McPht gene in the M. crystallinum and the transgenic rice, and the overexpression of the McPht gene in the transgenic rice significantly promoted the phosphate uptake and the root development, thus leading to an enhanced increase in the root activities under phosphorus deficiency. Both the up- and down-regulated genes in the transgenic rice are mainly involved in the regulations of molecular function,biological process, and cell component. The enrichments of differentially expressed genes were associated with the up-regulation by participating in the processes of phenylpropanoid synthesis, phenylalanine metabolism, and the secondary metabolite synthesis. This study suggested that the overexpression of the McPht gene in the transgenic rice might participate in an acclimation modulation in the specific up-regulation of differentially expressed genes,and thus promoting the phosphate uptake and physiological metabolism.

Acknowledgements

This work was supported by the National Key Project for Cultivation of New Varieties of Genetically Modified Organisms, Ministry of Agriculture, China (2016ZX08002-005), and the National Basic Research Program of China(2015CB150800). Special thanks to the research team of Prof. Wan Jianmin (Chinese Academy of Agricultural Sciences) for their support on rice transformation.