A Phosphate Transporter from the Root Endophytic Fungus Piriformospora indica Plays a Role in Phosphate Transport to the Host Plant*

Because pure cultures and a stable transformation system are not available for arbuscular mycorrhizal fungi, the role of their phosphate transporters for the symbiotic interaction with the plant up till now could not be studied. Here we report the cloning and the functional analysis of a gene encoding a phosphate transporter (PiPT) from the root endophytic fungus Piriformospora indica, which can be grown axenically. The PiPT polypeptide belongs to the major facilitator superfamily. Homology modeling reveals that PiPT exhibits twelve transmembrane helices divided into two halves connected by a large hydrophilic loop in the middle. The function of the protein encoded by PiPT was confirmed by complementation of a yeast phosphate transporter mutant. The kinetic analysis of PiPT (Km 25 μm) reveals that it belongs to the high affinity phosphate transporter family (Pht1). Expression of PiPT was localized to the external hyphae of P. indica colonized with maize plant root, which suggests that external hyphae are the initial site of phosphate uptake from the soil. To understand the physiological role of PiPT, knockdown transformants of the gene were prepared using electroporation and RNA interference. Knockdown transformants transported a significantly lower amount of phosphate to the host plant than wild-type P. indica. Higher amounts of phosphate were found in plants colonized with wild-type P. indica than that of non-colonized and plants colonized with knockdown PiPT P. indica. These observations suggest that PiPT is actively involved in the phosphate transportation and, in turn, P. indica helps improve the nutritional status of the host plant.

Phosphorous is one of the most essential mineral nutrients for plant growth and development and constitutes up to 0.5% of the dry weight of plant cell. It plays diverse regulatory, structural, and energy transfer roles and consequently is required in significant quantities (1,2). In the soil phosphorous presents mainly in the form of sparingly soluble complexes that are not directly accessible to plants. Thus, it is the nutrient that limits crop production throughout the world (3). In arbuscular mycorrhizal associations, plants acquire phosphate from the extensive network of fine extra radical hyphae of fungus, which extend beyond root depletion zones to mine new regions of the soil (4). Plants possess two distinct modes of phosphate uptake, direct uptake by its own transporters and indirect uptake through mycorrhizal associations. High affinity phosphate transporters have been identified and characterized in several plant and fungal species, including Arabidopsis, Medicago truncatula, Lycopersicon esculentum, Solenum tuberosum, Saccharomyces cerevisiae, and Neurospora crassa (5)(6)(7)(8)(9)(10)(11)(12). However, in the case of arbuscular mycorrhizal fungal species, including Glomus versiforme, G. intraradices, and G.mosseae, the role of phosphate transporters could not be verified due to the lack of a stable transformation system (4,13,14). Involvement of P. indica was reported in high salt tolerance, disease resistance, and growth-promoting activities leading to enhancement of host plant yield (15)(16)(17). However, the role of this root-endophytic fungus in plant nutrition has not yet been demonstrated.
In present study, a high affinity phosphate transporter has been isolated, identified, and functionally characterized from root endophyte fungus P. indica. Recently, a transformation system based on the polyethylene glycol method has been established for P. indica (18). Using electroporation for transformation and an RNAi 4 approach as powerful tools for gene silencing in fungi (19), we demonstrate that PiPT is essential for phosphate transport to the host plant. We suggest that exploitation of P. indica and its PiPT not only complements cropgrowing strategies but may also serve as a model system to study molecular mechanism and indirect uptake of phosphate by plants.

EXPERIMENTAL PROCEDURES
Plant, Fungi, Bacteria, and Yeast Strains-Zea mays (var. pro33) plant and fungus P. indica (15) were used throughout the study, and Escherichia coli XL1-Blue was used for cloning purposes (20). Yeast phosphate transporter mutant MB192 (MATa: pho3-7 pho84::HIS3 ade2 leu2-3,7 12 his3-532 trpl-encodes phosphate transporter, we established a cDNA library of P. indica, which was grown in low phosphate Aspergillus minimal medium (AMM) for 1 week (23). Total RNA was isolated by TRIzol reagent (Invitrogen) and pooled followed by poly(A) ϩ RNA extraction from 2 mg of RNA by using oligo(dT) cellulose beads (Amersham Biosciences). cDNA synthesis was performed using a cDNA synthesis kit (Stratagene). After size fractionation, cDNA Ͼ800 bp long were selected to create a library in the -ZAP vector and recombinant independent clones (5 ϫ 10 7 ) were obtained with lengths ranging from 0.5 to 3 kb.
To isolate full-length PiPT, a homologous probe was made using a partial PiPT gene, which was amplified using phosphate transporter-degenerate primers (forward, 5Ј-atgggtrtyggiathggiggigaytaycc and reverse, 5Ј-gtcgtrttiggiccraarttygraaraaa), where y represents a pyrimidine, r represents a purine, and i represents inosine (14). In brief, high molecular weight genomic DNA was extracted from ϳ500 mg of P. indica culture (24). PCR reactions were carried out in a final volume of 50 l containing 10 mM Tris-HCl (pH 8.3); 50 mM KCl; 1.5 mM MgCl 2 ; 0.01% (w/v) gelatin; 200 M of dNTPs; 3 M of each primer; 3 units of DNA Taq polymerase, and 60 -100 ng of genomic DNA (as template). The PCR program used was as follows: 94°C for 3 min (1 cycle), 92°C for 45 s, 55°C for 45 s, 72°C for 1 min, 15 s (30 cycles), and 72°C for 5 min (1 cycle).
The PCR product was cloned in p-drive vector (Qiagen) and sequenced by using M13 primers.
After obtaining homologous probe (928 bp), cDNA library was screened to get full-length PiPT cDNA by using Hybond-N nylon membranes (Amersham Biosciences) harboring recombinant phages, which were hybridized overnight at 60°C in hybridization buffer containing 7% SDS, 0.5 M phosphate buffer (pH 7.5), 1 mM EDTA (pH 8), and 1% bovine serum albumin. Membranes were washed twice at 25°C, once at 60°C in 0.5 ϫ SSC, and 0.2% SDS, before autoradiography. After three rounds of screening, the plaque-purified phages were converted to pBluescript-II SK derivatives by in vivo excision according to the manufacturer's instructions and sequenced.
Northern  PO 4 ). Fungus was harvested at 1, 5, 10, and 15 days, and total RNA was isolated. 15 g of total RNA was electrophoretically separated on 1.2% denaturing formaldehyde agarose gels and blotted onto nylon membrane. For Southern blot, genomic DNA (25 g) was digested with restriction enzymes and separated on 0.8% agarose gel, denatured, and transferred to a nylon membrane. For both analyses, hybridization and washing conditions were kept the same as mentioned in the previous text. An homologous probe used before to get full-length PiPT gene was also used for Northern blots.
For the expression studies of PiPT, two-step RT-PCR was performed with 5 g of total RNA, which was isolated from colonized (with P. indica) and non-colonized maize roots. The first strand of cDNA was synthesized with a Superscript cDNA synthesis kit (Clontech), and this was subsequently used as a template for PCR with gene-specific primers (forward, 5Ј-gctcgctgatctcgacca and reverse, 5Ј-atccggagatggtaacaatc). The cycling conditions were as follows: denaturation was done at 94°C for 5 min for one cycle; however, for 35 cycles denaturation was done for 40 s. Annealing was done at 59°C for 40 s, and extension was done at 72°C for 90 s. Translational elongation factor gene of P. indica (PiTef) was used as a reference gene as described before (17).
Computational Analyses and Homology Modeling-The functional sites and their pattern were determined using the PROSITE databank (available on-line). For identification purposes, the blastX algorithm (www.ncbi.nlm.nih.gov) was used. For modeling PiPT transmembrane (TM) domains, TM helix segments were predicted with various computer programs; e.g. SOSUI, HMMTOP, TMHMM, and TMpred (26 -29). A consensus of twelve TM helices emerged, and the boundary of each predicted helix was identified based on empirical rules derived from known membrane protein structures (30). Two models, PiPT1 and PiPT2, were constructed based upon homology with known crystal structures of glycerol-3-phosphate transporter (GlPT) and lactose permease (LacY) of E. coli (31,32). Sequence alignments were done with ClustalW and BLOSUM62 with a gap penalty of 10 for insertion and 5 for extension (33,34). All predicted TM domains of PiPT were used as secondary structural constraints in the input file for the modeling program. The final PiPT model having 1-522 amino acids was build using the MODELLER program based on GlPT (35). Statistical validation for bond lengths, bond angles, and Ramachandran analysis was obtained with PROCHECK (36). Adjustments were made whenever residues in the PiPT model had stereochemical violations.
Yeast Complementation-For this purpose, yeast high affinity phosphate transporter mutant strain MB192 was used (11).
PiPT cDNA was cloned into the EcoRI-BamHI site of the yeast expression vector p112A1NE by using sequence-specific primers (forward, 5Ј-cgaattcctcgtcgcgacaccttcatc and reverse, 5Ј-aagatctgtggctacattaatgtatctcattc). MB192 was transformed with recombinant p112A1NE vector (9) having PiPT inserted by the LiCl-PEG method (37,38). Procedures and medium used for growth and selection of transformants were similar as described elsewhere (11). For repressible acid phosphatase assay, cells were grown on YNB medium/high phosphate (11 mM) containing 3% glycerol (in place of glucose). Repressible acid phosphatase was measured after 36 h by using intact yeast cells as an enzyme source and p-nitrophenyl phosphate as substrate (39). MB192 cells transformed with empty vector were used as a control. Uptake of Radioactive Orthophosphate-Transformed MB192 cells (recombinant p112A1NE vector having the PiPT insert) were grown to the logarithmic phase (A 650 0.8 -1.2) on modified YNB medium containing 20 M KH 2 PO 4 . Before uptake assay, transformed MB192 cells were preincubated with 10 mm glucose for 5 min. Cells were washed in phosphatefree medium and re-suspended in the same medium (A 650 ϭ 1.0). Uptake was performed as described (40). In brief, cells (A 650 ϭ 1.0) were incubated in a solution containing 20 M phosphate together with 100 nCi of phosphorus-32 (Amersham Biosciences). The incorporation reaction was stopped by adding 4 ml of ice-cold water and centrifuging (15,800 ϫ g, 10 min), and supernatant was discarded. The pellet was washed twice with 4 ml of ice-cold water, and radioactivity incorporated by the cells was determined using a liquid scintillation counter (Beckman Instruments). The uptake assay was performed at room temperature (25°C). The experiment was repeated three times independently, and each time three replicates were taken. MB192 cells transformed with empty vector were used as a control. The amount of phosphate transported by control (background) was deducted from the amount of phosphate trans-  (72). The tree is drawn to scale, with branch lengths in the same units as those of the evolutionary distances used to infer the phylogenetic tree. The evolutionary distances were computed using the Maximum Composite Likelihood method (73) and are in the units of the number of base substitutions per site. All positions containing gaps and missing data were eliminated from the dataset (Complete Deletion option). Phylogenetic analyses were done by using MEGA4 (74).
Phosphate Transporter from P. indica ported by complimented MB192 with PiPT. GraphPad Prism 4 was used to plot nonlinear regression for phosphate uptake rate.

Localization of PiPT Expression and Relative
Quantitative RT-PCR Analysis-To determine the PiPT expression in external hyphae and in internal hyphae of the P. indica colonized maize plant root, relative quantitative RT-PCR was performed. For colonization purpose, radicals of maize seedlings were first submerged into macerated P. indica and incubated for two more days on water agar plates and finally transferred to MN containing 10 M phosphate. At day 5 external hyphae projecting out from the surface of the colonized root were collected by forceps. Approximately 2 mg of hyphae was collected per sample. In case of internal hyphae sample collection, first external hyphae were removed using forceps and or brushed off with a paint brush as described before (4). Small pieces (5-10 mm) of colonized root were collected. Colonization was also confirmed from these collected root pieces as described previously (17). RNA was isolated from these two samples, and cDNA was synthesized with oligo(dT) and Superscript II and then subjected to real-time PCR with specific primer pairs and SYBR Green I using a ABI 7500 Real-Time PCR System (Applied Biosystems) according to manufacturer's instructions. The forward and reverse primer sets for the PiPT and PiTef gene were RT PiPT FOR 5Ј-ctcgctcatcccagccttt, RT PiPT REV 5Ј-ccacagtcgaagagcttgca, PiTefFOR 5Ј-tcgtcgctgtcaacaagatg, and PiTefREV 5Ј-gagggctcgagcatgttgt, respectively. The reaction mixture was heated at 95°C for 10 min and then subjected to 40 PCR cycles of 95°C for 15 s, 61°C for 1 min, and 72°C for 20 s and the resulting fluorescence was monitored. For the constitutive PiTef gene, the mixture was heated at 95°C for 10 min, and then subjected to 40 PCR cycles of 95°C for 15 s, 57°C for 1 min, and 72°C for 20 s. The heat dissociation curves confirmed that a single PCR product was amplified for each gene. The melting temperatures were 81.3°C and 85.3°C for the PCR products of the PiPT and PiTef gene, respectively. The level of target mRNA, relative to the mean of reference housekeeping genes, was calculated by the comparative C t method as described by the manufacturer.
RNAi Cassette Formation for Knockdown of PiPT-A 350-bp unique fragment of PiPT gene was selected using the BLAST tool and further analyzed for RNA 2°structures. This unique fragment was amplified using specific primers pSD-1GFor 5Ј-ccggaattcggctcagcgaaatcagaagccatg and pSD-1GRev5Ј-ccggaattcaggtcaagacgcggttgccgtccc and cloned into a pGEM-T cloning vector. The 350-bp insert was digested out from pGEM-T with EcoRI and was subcloned into pSD-1G vector at unique EcoRI site. (For vector information please see supplemental Fig. 1). Positive clones were confirmed with PCR, EcoRI digestion, and sequencing. This construct was named as pSPiPTD-1G.
Transformation of pSPiPTD-1G into P. indica-For this purpose minimum inhibitory concentration of Geneticin (G418) was first determined by growing P. indica at 30°C at 200 rpm for 7 days. This culture of P. indica was macerated and inoculated in fresh media for 24 h. The culture was centrifuged at 1000 ϫ g for 5 min, and the mycelia pellet was resuspended in 1 ml of AMM. 100 l of this culture was mixed with 10 ml of top agar containing 100, 300, 500, and 700 g/ml G418 in separate 15-ml tubes. This mixture was poured on AMM agar plates and was incubated at 30°C for 2 weeks. We have found that P. indica was unable to grow at Ն700 g/ml G418. Here in every P. indica electroporation-mediated transformation experiment, 1000 g/ml G418 was used. pSPiPTD-1G was introduced into the P. indica by using electroporation. For the first time we have developed a transformation protocol using electroporation as follows: 7-day-old macerated mycelia were grown in AMM for 24 h at 30°C. To this ␤-glucuronidase (1 mg/ml) (Sigma: Helix pomatia) was added (41). After 2 h of growth, mycelia were centrifuged at 1000 ϫ g for 5 min, washed, and suspended in 1 mM HEPES buffer (pH 7.0) and 50 mM mannitol. From this 100 l was taken in a tube and mixed with 2-7 g of pSPiPTD-1G and subjected to electroporation using the Bio-Rad GenePulser apparatus at the field strength to 12.5 KV/cm, at 25-microfarad capacitance and 5-ms pulse length. After electroporation, 1 ml of AMM was added, and incubation was resumed at 30°C for 24 h. Transformants were selected and named as "KD-PiPT transformants." PiPT Transcript Abundance and siRNA Analysis in KD-PiPT P. indica-To determine the abundance of PiPT transcripts, relative quantitative RT-PCR was done. For this purpose, total RNA was extracted from wild type and two colonies of KD-PiPT P. indica using TRIzol reagent. For RNA preparation and PCR analysis, experimental conditions were kept the same as mentioned previously in localization of PiPT experiment.
After checking the PiPT transcripts abundance analysis, siRNA analysis was performed to show whether KD construct leads to siRNA accumulation or not. For this purpose, small RNAs were extracted and probed with some modifications (42). In brief, total RNA was extracted from KD-PiPT P. indica and wild-type P. indica by using TRIzol reagent. The pellet was Phosphate Transporter from P. indica AUGUST 20, 2010 • VOLUME 285 • NUMBER 34 JOURNAL OF BIOLOGICAL CHEMISTRY 26535 dissolved in di-ethyl pyro-corbonate (DEPC) water, heated to 65°C for 5 min, and then placed on ice. Polyethylene glycol (molecular weight of 8000, Sigma) was added to a final concentration of 5% and NaCl to a final concentration of 0.5 m. After 30-min incubation on ice, the RNA was centrifuged at 10,000 ϫ g for 10 min. Three volumes of ethanol were added to the supernatant, and the RNA was precipitated at Ϫ20°C for at least 2 h. The low molecular weight RNAs were pelleted by centrifugation for 10 min at 10,000 ϫ g. The pellet was dissolved in diethyl pyro-corbonate (DEPC) water and heated to 65°C for 5 min and a one-third volume of 4ϫ loading solution was added (2 ϫ TBE (1 ϫ TBE is 0.09 M Tris borate, pH 8.0, and 0.002 M EDTA), 40% sucrose, and 0.1% bromphenol blue) before loading on 15% urea PAGE in 1ϫ TBE. The RNA samples were electrophoresed at 2.5 V/cm and then blotted to a Hybond N ϩ membrane (Amersham Biosciences) and UV cross-linked. The membrane was prehybridized in 50% formamide, 7% SDS, 50 mM NaHPO 4 /NaH 2 PO 4 , pH 7.0, 0.3 M NaCl, 5ϫ Denhardt's solution (1ϫ Denhardt's solution is 0.02% Ficoll, 0.02% poly-vinyl pyrrolidone, and 0.02% bovine serum albumin), and 100 mg/ml sheared, denatured salmon sperm DNA at 37°C for at least 3 h. Probe was prepared by end labeling of the PiPT primer (5Ј-gctcgctgatctgacca) using [␥-32 P]ATP and polynucleotide kinase as per the instructions manual (Molecular Labeling and Detection, Fermentas) and was added to pre-hybridization solution. The hybridization was performed at 37°C overnight, and the membrane was subsequently washed at 37°C in 2 ϫ SSC (1 ϫ SSC is 0.15 M NaCl and 0.015 M sodium citrate (C 6 H 5 Na 3 O 7 .2H 2 O)) and 0.2% SDS for 15 min twice. Final washing was given only with 2 ϫ SSC at room temperature for 10 min and autoradiography was done.
Role of PiPT in Phosphate Transport from P. indica to Host Plant-To prove the role of PiPT in phosphate transport, a bi-compartment assay was performed using an in vitro culture system (43) as shown in Fig. 9. In the bi-compartment experiment, to make a physical barrier between both compartments, a 6-cm Petri dish (compartment 2) was placed inside a 15-cm Petri dish (compartment 1). Murashige-Skoog and AMM was  (75) are shown as helices over the corresponding amino acid sequences indicated by numerals (TM1-TM12). Prediction of functional motifs in PiPT polypeptide was performed with PROSITE data base (available on-line) (52). *, signature tag of the MFS transporter. (Consensus symbols: "!" is any sequence of IV, "$" is any sequence of LM, "%" is any sequence of FY, and "#" is any sequence of NDQEBZ.) Green boxed sequences are potential phosphorylation sites for cAMP-and cGMP-dependent protein kinase.
supplied to compartment 1, while compartment 2 was supplied with MN media. Surface-sterilized maize seeds were placed in compartment 1. The leafy shoots protruded through a groove cut in the lid of each dish and were fixed in one position by wrapping a sterile non-absorbent cotton wool around the portion of the subtending rhizome as it passed through the groove. Three sets were prepared for the experiment: (a) maize plants colonized with wild-type P. indica (WT), (b) maize plants colonized with KD-PiPT P. indica (KD), and (c) maize plants grown alone without P. indica. In all the cases 10 M phosphate concentration was used in compartment 1 as well as in compartment 2. For sets "a" and "b" to establish colonization between maize roots (compartment 1) and P. indica (compartment 2) a connective bridge was made by placing a 4-to 5-cm long agar strip so that P. indica can cross into the compartment 1 (Fig. 9). In the case of set "c" a connecting bridge was also made to check any transfer of radioactive phosphate from compartment 2 to 1 due to diffusion, and this set was used as a control.
As the colonization develops extra-radical hyphae proliferate in the medium surrounding the roots in compartment 1 where they ramify and later sporulate. After colonization was established, the MN media in compartment 2 of all three sets was replaced with fresh MN media containing 10 M phosphate and 1 M of 32 P (specific activity, 200 mCi/mmol). Radioactivity was determined in all the three sets by autoradiography, and the amount of 32 P incorporated was measured by liquid scintillation analyzer (Packard). Experiment was conducted three times independently.
Effect of PiPT on Phosphate Nutrition-To know whether phosphate has a effect on plant growth, we have determined the total phosphate content in 1) maize plant colonized with wild-type P. indica, 2) only maize plant, and 3) maize plant colonized with KD-PiPT P. indica. In this case, for colonization, radicals of maize seedlings (4 days old) were first submerged into macerated wild type as well as KD-PiPT P. indica and were incubated for 2 more days on water agar plates and finally transferred to MN containing 10 M phosphate and grown for 10 more days. Control plants were grown under similar conditions using autoclaved macerated fungal mycelium. Inorganic phosphate was determined using acid extraction of fresh plant material by a method described by Irving and Bouma (44). Shoots were collected, and biomass was measured in terms of fresh weight from all the three sets of same age. To these three set of samples, 40 l of 5 M H 2 SO 4 was added per 20 mg of each sample. The sam-   AUGUST 20, 2010 • VOLUME 285 • NUMBER 34

JOURNAL OF BIOLOGICAL CHEMISTRY 26537
ples were ground in a hand-held device, and to this 3 ml of distilled water was added. The resulting solution was filtered using Whatman No. 4 filter paper and a subsample, ranging from 20 l to 1.5 ml, (depending on the phosphate concentration) was analyzed for phosphate. The subsample was made up to 1.5 ml with water and to this 0.5 ml of malachite green reagent was added (45); the total contents were then mixed vigorously. After at least 30 min, the A 650 of the solution was measured. Standards in the range of 125 nm to 50 M of phosphate as KH 2 PO 4 were used. To understand the role of P. indica in phosphate nutrition improvements, the growth-promoting performance of P. indica was analyzed at low (10 M) and at sufficient or high phosphate concentration (1 mM). We have selected high phosphate concentration as the same has been used in the Hoagland solution (21). For this purpose, four sets were prepared 1) maize plants grown in low phosphate and treated with autoclaved macerated fungal mycelium (served as a control for low phosphate condition), 2) maize plants colonized with wild-type P. indica and grown at low phosphate condition, 3) maize plants grown at high phosphate condition and treated with macerated fungal mycelium (served as a control for high phosphate condition), and 4) maize plants colonized with wild-type P. indica and grown at high phosphate condition. As mentioned above, all four experimental sets were grown in acid-washed sand fertilized with modified 0.5ϫ Hoagland solution (21) containing respective phosphate concentrations. After 4 weeks, plants were harvested, and biomass was measured in terms of fresh weight.

Isolation and Organization of PiPT-Full-length PiPT
(GenBank TM accession no. DQ899728) cDNA clone, encoding the P. indica phosphate transporter, was isolated from a low phosphate-supplied P. indica cDNA library. PiPT shares 35% sequence identity with GvPT and PHO84 phosphate transporters, from G. verisforme and S. cerevisiae, respectively. Low sequence identity (30%) was observed with plant phosphate transporters ( Table 1). The PiPT is 1815 bp in length, and the open reading frame is flanked by a 90-bp untranslated sequence, at the 5Ј-end and 156 bp of untranslated sequence, including the poly(A ϩ ), at the 3Ј-end. An unrooted phylogenetic tree demonstrates the close relationship between the PiPT protein and members of the plant Pht1 family and high affinity phosphate transporters of fungi (Fig. 1). Southern blot analysis suggests the presence of a single PiPT gene in the genome of P. indica (Fig. 2).
Homology Modeling-PiPT encodes a polypeptide of 522 amino acids having a relative molecular mass of 57.7 kDa. Based on the hydropathy plots, the encoded polypeptide is predicted to be an integral membrane protein containing twelve hydrophobic membrane-spanning domains (TM) having a large hydrophilic loop between TM6 and TM7, resulting in a 6 ϩ 6 configuration. There are five potential protein kinase C-medi-  (Fig. 3).
Blast searches with the transporter classification database (TCDB, University of California at San Diego) indicate that PiPT belongs to the MFS family. The PiPT2 model has substantially more gaps in the helical regions; therefore, further analysis was done only with PiPT1 model. Structure-based sequence alignment for the TMDs of GlPT and PiPT1 was generated. We observed low homology in the 1-6 TMDs between GlPT and PiPT1. The alignment scheme was generated, and 1-6 TMDs sequence segments were then mapped to the reference alignment. We were able to build 522 residues of PiPT1 model based on the GlPT coordinates. Ramachandran analysis suggested that PiPT1 has no residues in the disallowed region and has an excellent score (0.4) for the overall G factor (bond lengths and bond angles). The schematic representation of the three-dimensional structure of the PiPT1 homology model indicates the pattern of characteristic of the MFS fold. The predicted PiPT shape is trapezoidal with dimensions of ϳ46 ϫ 27 Å from the bottom and ϳ33 ϫ 26 Å from the top, and its height is ϳ42 Å (Fig. 4).
PiPT Expression Is Dependent on the Phosphate Availability and Colonization-Increased expression of PiPT was detected at all the time intervals when 50 M or less phosphate was supplied (Fig. 5a). This increased express ion of PiPT under low phosphate supplied conditions reveals the high affinity nature of PiPT. RT-PCR data showed that PiPT is expressed during the interaction with the maize plants colonized with P. indica at all the time points, i.e. 5 days onward, whereas no expression was observed in roots of non-colonized plants (Fig. 5b). Furthermore, the increased expression of PiPT was observed with a gradual increase of colonization with time (Fig. 5c).
PiPT Encodes a High Affinity Phosphate Transporter-To determine whether PiPT encoded a functional phosphate transporter, the cDNA was ligated into a yeast expression vector p112A1NE and transformed into a yeast strain MB192 carrying the pho84 mutation. PCR-based confirmation was done for the presence of either PiPT cDNA or PHO84 cDNA in the yeast mutants used for complementation and uptake kinetics. PCR was carried out with plasmids as template (isolated from respective yeast cells) and primers that were specific to either PHO84 or PiPT. A 1.7-kb DNA fragment was obtained for PiPT, whereas a 0.8-kb DNA fragment was obtained for PHO84. No amplification was observed when native p112A1NE was used as a template. These positive yeast cells were further used for the repressible acid phosphatase assay and uptake kinetics of PiPT. The yeast pho84 mutant lacks the high affinity phosphate transporter and consequently has severely impaired phosphate uptake. Because they cannot accumulate phosphate, pho84 mutant cells produce a repressible acid phosphatase (EC 3.1.3.2) even during growth on high phosphate media containing levels of phosphate sufficient to repress repressible acid phosphatase production in wild-type cells. MB192 transformants expressing either PiPT or PHO84 displayed wild-type repressible acid phosphatase activity indicating that expression of PiPT or PHO84 complements the pho84 phenotype. MB192 transformants carrying the expression vector (without transporter) displayed the mutant pho84 FIGURE 7. Localization of PiPT expression in external (EH) and internal hyphae (IH) from maize plant root colonized with P. indica. For the determination of relative expression of PiPT level, cDNA was synthesized from RNA isolated from EH and IH and subjected to real-time PCR using specific primers and SYBR green I. The comparative C t method was applied to analyze the data. For experimental samples, targeted (PiPT) quantity was determined and divided by the target quantity of the calibrator (Tef). Thus, the calibrator becomes the 1ϫ sample, and all other quantities are expressed as an n-fold difference relative to the Tef. The values obtained for PiPT expression for EH and IH were 1-and 0.055-fold, respectively, relative to Tef. The means Ϯ S.D. of three independent determinations are presented. Asterisks indicate significant differences from EH at p Ͻ 0.01. Wild-type control (WTC); transformed colony 2 (TC2) and TC5, with an RNAi construct. The P. indica colonies were first grown in AMM, at day 4 AMM was replaced with fungal minimal media with a 10 m supplement of phosphate, and RNA was extracted from P. indica colonies at 3 days after transfer of media. Expression levels of PiPT were determined by real-time PCR as described in the legend of Fig. 7. The values obtained for PiPT expression for WTC, TC2, and TC5 were 1-, 0.4-, and 0.39-fold, respectively, relative to Tef. The means Ϯ S.D. of three independent determinations are presented. Asterisks indicate significant differences from WTC at p Ͻ 0.01. b, Northern blot analysis of siRNAs of the PiPT in P. indica transformants. Blot was exposed overnight to show detectable siRNA accumulation in the KD-PiPT P. indica (TC2 and TC5), C, wild-type P. indica. DNA oligonucleotides (16 and 22 nucleotides (nt)) were used as molecular size markers for siRNA analysis. Equal loading of total RNA was estimated by ethidium bromide staining of rRNAs (predominant RNA). AUGUST 20, 2010 • VOLUME 285 • NUMBER 34 phenotype with high repressible acid phosphatase activity (Fig. 6a).

Phosphate Transporter from P. indica
Phosphate uptake by MB192 expressing PiPT was further confirmed by measuring uptake of 32 P from solution. MB192 expressing PiPT accumulated 32 P at a rate significantly (p Ͻ 0.001) higher than the background rate shown by control MB192 cells (Fig. 6b) (Fig. 6c).
Localization of PiPT Expression-To determine the specific localization of PiPT expression, transcript abundance was measured in internal hyphae and external hyphae that ramify out of the colonized root into the media using relative quantitative RT-PCR. We have found that PiPT transcripts were 18-fold higher in the external hyphae as compared with internal hyphae, and this difference was found to be significant (p Ͻ 0.01) (Fig. 7).
Characterization of KD-PiPT P. indica-We have obtained transformation efficiencies between 5 and 20 transformants/ microgram of DNA (supplemental Fig. S2). The KD for PiPT was analyzed for transcript abundance and siRNA. Real-time RT-PCR analysis revealed a lower abundance (60%) of PiPT in KD-PiPT P. indica as compared with wild-type P. indica, and this difference was found to be significant (p Ͻ 0.05) (Fig. 8a). Accumulation of siRNA was observed in the Northern blot in the case of KD-PiPT P. indica, and no siRNA was detected in the case of wild-type P. indica (Fig. 8b).
Analysis of Fungus-to-Plant Transfer of Phosphate-Transportation of phosphate was observed using the bi-compartment system (Fig. 9). In the first set (set a) autoradiography revealed extensive labeling of maize plants by uptake of radiolabeled 32 P from wild-type P. indica (Fig. 10A, panel a). The 32 P was transferred to maize plants through the fungal mycelium and across the hyphal bridge between both compartments. Very little radioactive counts were observed in the agar media of the second compartment confirming that the amount of 32 P present in the maize plants was exclusively transferred by P. indica and not because of leaching by the fungus in the second compartment. In set b, very less radioactivity was detected in maize plants colonized with KD-PiPT transformants of P. indica, confirming the direct role of PiPT in phosphate transport to maize plants (Fig. 10A, panel b). In the case of set c no radioactivity was observed (Fig. 10A, panel c), and hence, the movement of 32 P from one chamber to another was not due to diffusion but through fungus only. We observed that 349 pmol of phosphate was transported by wild-type P. indica to the host plant as compared with 36 pmol in the case of KD-PiPT P. indica, and this difference was found to be significant (p Ͻ 0.05) (Fig. 10B). We have also observed that the colonization of both wild-type and KD transformants of P. indica into maize plants was found similar in both cases, i.e. 70% at 20 days after inoculation (Fig. 10A, panels a-c).
Role of PiPT in Phosphate Nutritional Improvements of Host Plant-We found that inoculation of P. indica significantly increased the average above ground biomass of maize plants by 2.4-and 2.8-fold than that of the non-colonized plants as well as from the KD-PiPT P. indica colonized plants, respectively (Fig.  11a). Further we found 2.5-and 3.75-fold higher phosphate content in plants colonized with wild-type P. indica as compared with non-colonized plants and KD-PiPT P. indica-colonized plants, respectively, and this difference was found to be statistically significant (p Ͻ 0.05) (Fig. 11b). In a separate analysis, to know the performance of P. indica in growth promotion activity at low and high phosphate concentrations (as mentioned under "Experimental Procedures"), we have found that the growth (in terms of fresh weight) of maize plants colonized with P. indica was 1.2-fold higher when grown at high phosphate condition (1 mM), whereas it was 2-fold higher in the case of maize plants colonized with P. indica grown at low phosphate condition (10 M) in comparison to their respective controls (Fig. 11c).

DISCUSSION
In present study, a high affinity phosphate transporter has been isolated, identified, and functionally characterized in P. indica. The deduced amino acid sequences of the PiPT share significant similarities (including homology and topology) with those of known high affinity phosphate transporters reported so far from higher plants and fungi. Based on our phylogenetic dendrogram data we have observed that PiPT falls into the group of high affinity phosphate transporters of fungi far distinct to plant phosphate transporters.
Hydrophobicity analysis and sequence alignment suggests that PiPT contains 12 membrane-spanning domains with a large hydrophilic loop separating them into two groups of six. This is a characteristic of phosphate transporter, and many other transporter proteins belong to the MFS family and thus support our data (46 -50). It was suggested that MFS proteins are typically 400 -600 amino acids in length having similar transmembrane topology and signature sequences in two cytoplasmic loops. Additionally, position and spacing of these membrane-spanning regions of PiPT are very similar to those reported in other phosphate transporters from higher plants and fungi (51). Based on secondary structure analyses using the PROSITE database of protein domains, families, and functional sites (52,53), we have observed that both the N and C termini and hydrophilic domains of the polypeptides are located toward the cytoplasmic side of the plasma membrane. The central channel of the PiPT is formed by TM1, TM2, TM7, and TM11. Unlike other fungal and plant phosphate transporters, cAMP-and cGMP-dependent protein kinase phosphorylation sites have been observed in PiPT polypeptide. Consequently, we speculate that the cAMP-and cGMP-dependent downstream signaling pathway is involved in phosphate transporter regulation. Although these data are not supported by the experimental evidence, further investigation is needed.
So far, there is no report available on crystallization nor a well defined structure of inorganic phosphate transporters that can be used for modeling of PiPT; therefore, we used GlPT as a reference template for the modeling of PiPT (31). The PiPT model contains 522 amino acid residues, with an unmodeled loop at the N and C termini and polypeptides of 82 amino acids between TM6 and TM7. Because PiPT is only a modeled structure, hence, a realistic evaluation of its quality should be investigated experimentally.
The expression of PiPT does not depend on the culture time, and a steady-state high transcript level of PiPT was observed FIGURE 10. In A: transport of phosphorus to maize plants by P. indica carried out in the bi-compartment Petri dish culture system. Radioactivity incorporated in plants was demonstrated by autoradiography. Radioactivity count intensities are shown in false color code (vertical bar, low to high). Panels: i, whole maize plant before autoradiography; ii, false-color autoradiograph of the maize plant obtained after 3 h of exposure of the maize plant; and iii, microscopic view of a sample of plant root before autoradiography. a, maize plants were colonized with wild-type P. indica (WT); b, maize plants were colonized with KD-P. indica (KD); c, maize plants were grown alone without P. indica (C). In B: amount of 32 P transferred to the maize plant components by P. indica. Radioactivity was measured three times independently (number of transformants used was n ϭ 2). The mean Ϯ S.D. of three independent measurements are shown. The bar labeled with the asterisk represents significance as compared with the wild-type P. indica (p Ͻ 0.05). upon micromolar phosphate supply (Յ50 M). Based on structural similarities and a physiological response to a phosphate deficiency, PiPT is clearly part of the high affinity transporter system. Similar findings have been observed by Chung et al. (54) in algae, Tetraselmis chui, for a high affinity phosphate transporter gene (TcPHO), which in turn supports our data.
Further, the expression studies showed that the expression of PiPT is regulated by the amount of phosphate present in the outside only, and it seems that there is no role for the intracellular phosphate pool in the regulation of PiPT. The PiPT expression level increases as colonization proceeds as shown in Fig. 5 (b and c). We have also observed similar colonization of P. indica in maize plant (70% at 20 days after inoculation) at low as well as at high phosphate concentration; therefore, we suggest that the colonization is independent of phosphate availability.
The functionality and kinetics of PiPT were investigated by their ability to complement the yeast high affinity phosphate transporter pho84 mutant. These data suggest that PiPT is a functional stretch that has potential to complement yeast pho84 mutant. Further, kinetic data of PiPT follow Michaelis-Menten kinetics with an apparent K m of 25.27 Ϯ 3.66 M. These findings along with high level expression of PiPT in micromolar phosphate supply suggest that PiPT belongs to the high affinity phosphate transporters. Similar results were also reported previously with GvPT and PHO84, and for these the K m has been estimated at 18 and 8.2 M, respectively (4,11). To localize the PiPT expression in roots of maize plants colonized with P. indica, we have collected external hyphae projecting out from the colonized root as well as internal hyphae (within the root). Quantitative RT-PCR analysis shown that PiPT transcripts were 18-fold higher in the external hyphae as compared with internal hyphae, indicating that external hyphae is the main site of PiPT expression. Similar findings were also observed by Harrison and Buuren (4) in the external hyphae projecting out from the surface of the mycorrhizal roots of Allium porrum colonized with G.versiformae and in turn support our data.
We have successfully used electroporation-mediated transformation system in P. indica and obtained good transformation efficiency, and similar findings have been reported by Zuccaro et al. (18) that support our data. However in their study they used protoplast preparation for transformation. To determine the physiological function of PiPT in vivo, we prepared KD transformants of PiPT and found that the transcript level of PiPT was specifically and effectively reduced. We have also observed the accumulation of siRNA in the case of KD-PiPT transformants. The formation of siRNA indicates that in FIGURE 11. Effect of PiPT on phosphate nutrition and plant growth. Shown are biomass (a) and total phosphate (b) content. Error bars denote the ϮS.D. of the mean from plants of three replicate plates. *, significant difference from the maize plant colonized with wild-type P. indica (taken as a control). c, growth-promoting performance of P. indica at low (10 m) and high (1 mm) phosphate concentrations. The mean Ϯ S.D. of three independent measurements are shown. The bar labeled with the asterisk represents significant (p Ͻ 0.05) as compared with their respective control.