Phosphate Transporter BnaPT37 Regulates Phosphate Homeostasis in Brassica napus by Changing Its Translocation and Distribution In Vivo

Inorganic phosphate (Pi) is actively taken up by Pi transporters (PTs) from the soil and transported into the plant. Here, we functionally characterized the Brassica napus gene BnaPT37, which belongs to the PHT1 family. BnaPT37 is a plasma membrane-localized protein containing 534 amino acids. Expression of BnaPT37 increased significantly under Pi deficiency in various tissues, especially in fully expanded leaves. Expression of the β-glucuronidase reporter gene driven by the BnaPT37 promoter showed that BnaPT37 is expressed in the root, stem, calyx, and leaf under Pi deficiency. BnaPT37 can complement a yeast mutant strain defective in five Pi transporters and can restore the growth of the Arabidopsis atpt1/2 double mutant under Pi deprivation. Overexpression of BnaPT37 in rapeseed significantly increased Pi translocation from root to shoot. Moreover, the movement of Pi from fully expanded leaves to new leaves and roots was enhanced in the transgenic lines compared to the wild type. However, the overexpression of BnaPT37 inhibited the flowering time, plant height, and Pi accumulation in seeds. In conclusion, BnaPT37 functions as a plasma membrane-localized Pi transporter and might be involved in Pi translocation from root to shoot and Pi distribution from source to sink in B. napus.


Introduction
Phosphorus (P) is one of the essential macronutrients for plant growth and development.P is an important component of phosphate (Pi), nucleoprotein, phospholipid, and phytate and is involved in various metabolic activities in plants, such as photosynthesis, carbohydrate metabolism, fat metabolism, and so on [1].Moreover, P can improve the drought resistance and disease resistance of plants [2][3][4].In soil, there is both organic and inorganic P; plants, however, mainly absorb inorganic P [5].Although the total P content is very high in soil, about 80% of P exists in the organic form, and Pi ions are easily fixed by the soil, which makes it difficult for plants to use [6].Therefore, the phytoavailability of P in soil is often very low.P fertilizer can provide sufficient nutrients for crops, but the input of a large amount of P fertilizer may result in environmental pollution, such as eutrophication in water systems, and damage the environmental ecosystem [7].In addition, the massive use of P fertilizer accelerates the depletion of Pi rock, a non-renewable resource [8,9].
When the soil is short of P, that is, the content of available P in the soil solution is less than 10 µM [10], the plant may produce a series of morphological adaptive changes to cope, such as changing root architecture, increasing root exudation, and inducing the expression of P starvation response genes [9,11].During P deficiency, plants can induce the expression of Pi transporters (PHTs) to uptake P from the rhizosphere soil.The high-affinity P transporters can absorb P at very low external P concentrations [12,13].The PHT1 family belongs to the ninth subfamily of the major facilitator superfamily (MFS).It is strongly induced by low P stress, involved in P absorption by plant roots and transport in plants, and is extremely important for plant growth [14][15][16].The PHT1 family genes contain 12 transmembrane domains (TMs) and generally have a hydrophilic ring between the sixth and seventh transmembrane domains [17,18].The PHT1 family mainly transports Pi from soil to root cells through Pi/H + co-transport, and there are significant differences in the characteristics of Pi transport among its members [19][20][21][22][23].
PHT1 family genes play important roles in Pi uptake and transport in plants.The Arabidopsis PHT1 family has nine members, and eight of them are mainly expressed in the root [19].Under high and low P conditions, the Pi uptake efficiency of the pht1;1 and pht1;4 mutants were significantly lower than that of the wild type, and the dual mutants exhibited additive effects, indicating that AtPHT1;1 and AtPHT1;4 play a key role in Pi uptake [14,24,25].AtPHT1;5 showed a higher transcript level in leaves than that in roots under P-deficient conditions, mediating P translocation between source and sink organs, and its mutation affected P transport to the shoot [26], while AtPHT1;8 and AtPHT1;9 were involved in Pi translocation from root to shoot [15].In addition, AtPHT1;8 and AtPHT1;9 might participate in Pi translocation together with AtPHT1;3 and AtPHT1;4 [19].In rice (Oryza sativa), there are 13 members of the PHT1 family [13].OsPHT1;1, expressed constitutively in the root and the shoot, was involved in Pi uptake and transport under P-sufficient conditions [27].OsPHT1;2 is a low-affinity P transporter expressed in the main root of the central column of lateral root, mediating Pi transfer into root vasculature [20].Os-PHT1;3 was strongly induced by an extremely low P stress condition, and the P absorption efficiency of the ospht1;3 mutant significantly decreased compared to the wild type [28].The Pi uptake ability of OsPHT1;8 RNAi lines was significantly reduced compared to the wild type, indicating its role in mediating Pi uptake in rice [29,30].Overexpression of OsPHT1;9 and OsPHT1;10 significantly increased the Pi uptake rate of rice, while the inorganic P concentration of the double mutant significantly decreased [16].OsPHT1;7 is highly expressed in anthers and plays an important role in P transport, redistribution, and anther development [31].The mutation of OsPHT1;7 significantly reduces P redistribution into new leaves, leaving more P stuck in old leaves, leading to a significant decrease in yield and P concentration in the anthers [31].OsPHT1;4 is strongly expressed in roots and embryos and plays an important role in embryonic development [32].Both OsPHT1;11 and OsPHT1;13 are specifically induced by mycorrhizal fungi, and are involved in Pi uptake by the arbuscular mycorrhizal symbiotic pathway [33,34].Similar to Arabidopsis, there were collaborations between Pi transport proteins in rice; for example, the function of OsPHT1;2 was probably enhanced via other OsPHT family members [35].A total of 13 PHT1 family members have been identified in maize, and ZmPHT1;7 plays a central role in Pi uptake as well as in P redistribution from old to young leaves [36].
Brassica napus is an important oil crop worldwide, but it is sensitive to P stress and needs massive amounts of P for high seed yield and oil content [37].In addition, B. napus has suffered from a constant P deficiency worldwide [38].Therefore, it is crucial to breed B. napus cultivars with increased P-use efficiency and improved seed production.The PHT1 family genes play an important role in Pi uptake and translocation in plants [12,13].In our previous research, we found that there are 49 members of the PHT1 family in B. napus with different expression patterns both in the root and shoot, indicating that the rapeseed PHT1 family genes may play different roles in P homeostasis in B. napus [39].The reported BnPHT1;4 gene localizes in the plasma membrane, is induced by P deficiency in the root, and is expressed in the early cotyledon [40].Overexpression of BnPHT1;4 significantly reduces P accumulation in cotyledons while increasing inorganic P content and promoting seed germination, indicating that BnPHT1;4 is involved in plant P uptake and early seed germination [41].In this study, BnaPT37, a member of the PHT1 family that was induced by P deficiency in root and shoot, was isolated and characterized in B. napus.To investigate the function of BnaPT37 in P homeostasis, we developed transgenic lines with overexpression of BnaPT37 in B. napus.Our results revealed that BnaPT37 is involved in Pi uptake and the distribution of Pi to new leaves.

Sequence Alignment and Phylogenetic Analysis of BnaPT37 Gene
There are 49 members of the putative phosphate transporter PHT1 family in B. napus named BnaPT1-BnaPT49 [39].Sequence analysis showed that the coding sequence (CDS) of the BnaPT37 gene consists of 1605 bp.The protein sequence of BnaPT37 contains 534 amino acids, which share about 95% of their identity with AtPT4 in Arabidopsis (Figure 1A,B).The protein structure of BnaPT37 is similar to that of AtPT4, both of which contain 12 transmembrane domains and a conserved PHT1 signature (GGDYPLSATIMSE) (Figures 1A,C and S1).

The Expression Pattern of BnaPT37 in Response to Different P Supplies
Most of the reported PHT1 family members are regulated by the external P environment, mainly by P deficiency [13].A time-course treatment was performed to investigate the responses of BnaPT37 expression to a Pi shortage signal.In roots, BnaPT37 was significantly induced by Pi starvation after 6 h of treatment (Figure 2A).The induction rate grew steadily throughout the course of the treatment, peaking at Day 9, and a 1-day resupply of Pi reduced BnaPT37 expression to levels equivalent to those observed before P stress treatment (Figure 2A).In shoots, the induction of BnaPT37 expression was observed after 12 h of P deprivation treatment.It reached its maximum after 9 days of P stress treatment and decreased sharply to those at the beginning of P stress treatment after a 1-day resupply of Pi (Figure 2B).In order to further clarify the expression pattern of the BnaPT37 gene in different tissues, we divided B. napus seedlings into seven tissues, including the root, hypocotyl, basal node, petiole, cotyledon, fully expanded leaf, and new leaf, to determine the expression level of the BnaPT37 transcript.Consistent with the expression pattern at different time points, BnaPT37 was significantly up-regulated by P deficiency in all tissues, but the highest expression was observed in the fully expanded leaf under P deficiency, followed by the petiole, root, and cotyledon (Figure 2C).The expression pattern of the BnaPT37 gene in the maturing stage was also analyzed.The result showed that BnaPT37 was highly induced by low-P treatment in carpopodium but inhibited in fully expanded leaves.No significant difference in BnaPT37 expression was observed in other tissues under both P treatments (Figure 2D).

BnaPT37 Is P-Starvation-Induced and Plasma Membrane-Localized Protein
We transiently inserted BnaPT37 fused with GFP into Arabidopsis leaf protoplasts under the control of the 35S promoter to examine the subcellular location.The results showed that the GFP signal of BnaPT37-GFP was identified as a fine ring at the cell periphery (Figure 3A), indicating that BnaPT37 is a plasma membrane-localized protein.To investigate the tissue specificity and Pi responsiveness of BnaPT37, we generated transgenic B. napus lines carrying a 2 kb promoter region of BnaPT37 fused with the βglucuronidase (GUS) reporter gene.The results showed that GUS driven by the BnaPT37 promoter was highly expressed in various cells of the root, stem, and leaf under P-deficient conditions (Figure 3B).In addition, the expression of BnaPT37 was also detected in the calyx under P-deficient conditions at the flowering stage but not in the young pod (Figure 3B).

BnaPT37 Mediates Pi Transport in Yeast and in Arabidopsis
To acquire biochemical support for the function of BnaPT37, we conducted complementation tests using the yeast mutant strain EY917, which is defective in five Pi transporters (PHO84, PHO89, PHO87, PHO90, and PHO91) [42,43].At the same time, it can activate the Gal-inducible promoter to express the yeast Pi transporter gene PHO84 in the presence of galactose, so that the yeast mutant EY917 can absorb Pi in the medium and resume growth (Figure 4A).In addition, PHO84, a reported yeast Pi transporter gene, was used as the positive control.On the defective growth medium with glucose as the carbon source and different Pi concentrations, the transformed cells with either BnaPT37 or PHO84 grew very well on the medium with different Pi concentrations, but the yeast transformed with an empty vector was completely inhibited (Figure 4A).Compared to the transformed yeast cells with BnaPT37, the cells with PHO84 could grow better.Then, the growth curves of yeast at different time points were further verified.At 60 µM Pi, the growth state of yeast expressing BnaPT37 is slightly lower than that of yeast expressing PHO84 but still much higher than that of yeast expressing an empty vector (Figure 4B).These results indicate that BnaPT37 can partially restore the growth of yeast mutant EY917.To further confirm that BnaPT37 can transport Pi, BnaPT37 was transformed into the Arabidopsis atpt1/2 double mutant that lacks the main Pi transporters AtPHT1;1 and AtPHT1;4.Three independent complementary lines were selected for phenotypic analysis (Figure S2A).Under 15 µM P treatment, the root and shoot fresh weights of the complementary materials were significantly higher than those of the atpt1/2 double mutant and wild type, while the shoot fresh weights of the complementary materials were slightly higher than those of the atpt1/2 double mutant under 625 µM P conditions (Figure S2B-D).The inorganic P concentrations in the shoot and root of the complementary materials were significantly higher than those of the atpt1/2 double mutant, even higher than those of the wild-type plants under both high-and low-P treatments (Figure S2E,F).These results indicate that BnaPT37 has Pi uptake activity and can recover the P-deficient phenotype of the Arabidopsis atpt1/2 double mutant.These results suggest that BnaPT37 is a phosphate transporter.

Overexpression of BnaPT37 Retards Rapeseed Growth at Seedling Stage
To determine the functions of BnaPT37 in plant growth and Pi transportation, three independent BnaPT37 homozygous overexpression lines were created.Reverse-transcription PCR (RT-PCR) was used to confirm the expression levels of BnaPT37 in the three overexpression lines (Figure S3).After treatment with low P conditions, the overexpression lines showed more severe chlorosis and necrosis on the cotyledon of rapeseed compared to the wild-type plants.The same phenotype was observed on both the cotyledon and leaf1, which is the first true leaf of rapeseed under high P supplies (Figure 5A,E).We then detected the SPAD value and the contents of chlorophyll a, chlorophyll b, and carotenoid in the cotyledon and leaf1 of both transgenic and wild-type plants under high and low P stress environments.The results showed that they were lower in the cotyledon of the BnaPT37 overexpression lines than in the wild type under two P levels (Figure S4).In addition, the root and shoot dry weight of the transgenic plants was much lower than that of the wild-type plants under high-P conditions, but no significant differences were detected under P-deficient conditions (Figure 5B,F).

Pi Translocation and Distribution Changes in the BnaPT37 Transgenic Plants
Higher total P concentrations were observed only in the shoots of the three overexpression lines under high-P conditions compared to wild-type plants (Figure 5C,G).
Furthermore, the inorganic P concentrations in the cotyledon and leaf1 of the overexpression lines were significantly higher than those in wild-type plants under normal-P conditions, whereas higher inorganic P concentrations were detected only in the cotyledons of the overexpression lines than in the wild-type plants under low-P treatments (Figure 5D,H).However, overexpression of BnaPT37 had little effect on the P accumulation in the whole plant, and only the root P accumulations of the overexpression lines were slightly reduced under high-P conditions (Figure S5).We then further calculated the root-to-shoot translocation of P and the P concentration in the xylem sap.Under high-P treatment, the Pi translocation from root to shoot of the overexpression lines increased significantly compared to the wild-type plants, and similar results were observed for the P concentrations in the xylem sap under both P treatments (Figure 6).These results indicate that overexpression of BnaPT37 may enhance rapeseed Pi translocation from root to shoot.The Pi concentrations in the cotyledon and leaf1 of the overexpression lines increased significantly compared to those in the wild type (Figure 5), suggesting that overexpression of BnaPT37 may affect P distribution in plants.Thus, we analyzed the P contents and distributions in different tissues of the overexpression lines and the wild type.The P content increased in the cotyledon of the overexpression lines compared to the wild-type plants but decreased in leaf 4, the youngest leaf, under low-P conditions (Figure 7A).Under high-P conditions, higher P contents were observed in the cotyledon, leaf1 and leaf2 of the overexpression lines compared to those of the wild type (Figure 7B).Then, we calculated the P distribution ratio among different tissues of the plants.It showed that P distribution in the cotyledon of the overexpression lines was significantly higher than in the wild type under low-P treatment but lower in the newest leaf 4 (Figure 7C).Under high P levels, the distribution of P was high in cotyledon leaf1 and leaf2 but lower in leaf4 of the BnaPT37 overexpression lines compared to that of the wild type (Figure 7D).These results suggest that BnaPT37 may be involved in P allocation in B. napus.
To further check whether BnaPT37 functions in P redistribution, we cultivated the transgenic and wild-type plants under high-P environments for 15 d and then transferred them to P-free treatments for 6 d.The P concentrations and contents in different tissues were analyzed before and after P-free treatments.We found that ∆P contents were higher in the newest leaf 3, root, and stem but lower in fully expanded leaf2 and leaf1 of the BnaPT37 overexpression lines compared to the wild type (Figure 8), indicating that overexpression of BnaPT37 accelerated P reallocation from source to sink organs in B. napus.

Overexpression of BnaPT37 Affects Plant Height and P Distribution at the Flowering Stage
BnaPT37 is expressed highly in rapeseed shoots (Figure 2), and overexpression of BnaPT37 affects P allocation in rapeseed seedlings (Figures 6-8).Thus, we speculated that BnaPT37 may also affect rapeseed growth at the reproductive stage.Then, we planted the transgenic lines and wild type using a pod culture system in real soil.The flowering time of the overexpression lines was delayed compared to the wild type under both sufficient and deficient P supplies (Figure 9A,B), while the plant height of the overexpression lines was lower than that of the wild-type plants (Figure 9C).However, the dry weight of the stem, lower leaves, and upper leaves of the overexpression lines was slightly higher than that of the wild-type plants (Figure 9D,E).In addition, overexpression of BnaPT37 had little effect on P concentration in various rapeseed tissues (Figure S6A,D).Nevertheless, the P distribution was significantly inhibited in flower buds but enhanced in the upper leaves of B. napus by overexpression of BnaPT37 under sufficient P supplies (Figure S6F).

BnaPT37 Is Involved in the Translocation of P into Grains at the Ripening Stage
Seed yield and agronomic performance were investigated at the ripening stage.The plant height, first branch height, branch number, seed number per pod, and thousandseed weight were not affected by the overexpression of BnaPT37 under both P levels (Figure S7A-G).However, pod numbers in the main inflorescence and lateral branches were significantly reduced in the transgenic lines compared to the wild type under low-P treatment, as was the straw weight (Figure S7H-K).In addition, overexpression of BnaPT37 significantly reduced both seed yield and pod number per plant under low-P treatment, while no difference was observed under normal-P treatment (Figure 10A,B).We further determined the P concentrations and contents in different organs of the BnaPT37 overexpression lines and wild-type plants.Under low-P conditions, P concentrations in different organs of the overexpression lines were higher than in the wild type (Figure 10C), but they were slightly lower under sufficient-P conditions (Figure 10D).Then, we calculated total P contents in different organs and found that overexpression of BnaPT37 significantly reduced P accumulation in seeds under both sufficient and deficient conditions (Figure 10E,F).These results indicate that BnaPT37 is involved in the transportation of P to seeds in B. napus.

Discussion
The Pi transporter of the PHT1 family has been reported to contain 12 transmembrane domains in many plant species, including a hydrophilic ring between the sixth and seventh transmembrane domains [17,18,40].The BnaPT37 gene identified here was found to be homologous with AtPT4 through phylogenetic tree analysis and sequence comparison, and the amino acid similarity reached more than 98% (Figure 1).Importantly, AtPT4 has a typical molecular structural feature (GGTYPLSATIMSE) of the PHT1 family [18,44].The homologous gene BnaPT37 in B. napus was similar, indicating that it may also play an important role in regulating Pi transport (Figure 1).The examination of the threedimensional protein structure revealed that BnaPT37's structure was similar to that of Arabidopsis AtPT4 (Figure S1), which was consistent with the structural characteristics of PHT1 family genes [17].Research shows that the PHT1 family encodes a Pi transporter located on the plasma membrane [32,40,45].The same result was observed here (Figure 3A), indicating that BnaPT37 is a plasma membrane-localized Pi transporter.PHT1 family members may express in different tissues with varied intensity in response to P deficiency [19][20][21][22][23].The PHT1;4 gene in rapeseed was expressed in early cotyledons, and its overexpression significantly reduced the accumulation of P in B. napus cotyledons [41].AtPHT1;5 is highly expressed in the P-deficient shoot, and the mutation of AtPHT1;5 affects P allocation to the shoot, indicating that AtPHT1;5 is involved in P homeostasis in Arabidopsis [26].The abundance of BnaPT37 was induced by P deficiency in all tissues at the rapeseed seedling stage, but the highest expression levels were observed in leaves, followed by roots (Figure 2).The tissue-specific localization analysis showed that the protein level of BnaPT37 was enhanced in the root, cotyledon, and leaf under P deficiency (Figure 3B).Arabidopsis AtPHT1;4 is induced by P deficiency in Arabidopsis root and shoot, but the expression level in the root is higher than in the shoot.The mutation of AtPHT1;4 significantly reduces the Pi absorption rate, indicating that it mainly participates in Pi uptake [14].However, little is known about the function of AtPT4 in Arabidopsis.Here, we speculated based on the expression pattern of the AtPT4 homologous gene, BnaPT37, that it may be involved in Pi translocation and distribution but not in Pi uptake (Figures 2 and 3).
PHT1 family genes mainly transport Pi from soil to root cells through Pi/H + cotransport [12,13,19].The high-affinity PHT1 Pi transporters can absorb Pi under extremelylow-external-Pi concentration conditions, and their Km values are generally in the micromole range [5,21,28].PHO84 is the first high-affinity P transporter identified in yeast [46].The yeast mutant strain with its functional defect is often used to study the transport activity of Pi transporters.We found that BnaPT37 can restore the growth defects of yeast mutant stain to a certain extent, even when 0.1 mM of Pi is supplied externally (Figure 4).Moreover, we introduced BnaPT37 to the Arabidopsis double-mutant atpt1/2 and found that the concentrations of inorganic P in the complementary lines were significantly increased, and the complementary lines grew better compared to the double-mutant and the wild type (Figure S2), indicating that BnaPT37 can transport Pi [29,36,40].
PHT1 family members are mainly up-regulated in the root under P-deficient conditions, but some of them are also expressed in leaf and other tissues, such as OsPHT1;4, BnPHT1;4, and GmPHT1;5, which results in functional differentiation [32,41,47].BnaPT37 has the highest homology with Arabidopsis AtPT4 (Figure 1), and it is induced by P deficiency in both root and shoot, especially in the leaf (Figure 2).It is suggested that BnaPT37 may not be involved in the Pi uptake, but functions in the Pi translocation or distribution.Previous research showed that the P concentration in the old leaves of the ospht1;7 mutant was significantly higher than in the wild type at both low and normal P levels.The P allocation to the new leaves of the mutant was significantly reduced compared to the wild type, which affected the development of anthers and ultimately led to a decline in yield [31].In addition, Arabidopsis AtPHT1;5 overexpression lines senesced earlier than in the wild type, with chlorosis in old leaves, a significant decrease in P concentration in rosette leaves, and an increase in inflorescence stems and siliques, indicating that AtPHT1;5 promoted the redistribution of Pi from source to sink [26].Here, the overexpression of BnaPT37 led to the accumulation of Pi in plants under high P supplies as well as high inorganic P concentrations (Figure 5).It is noteworthy that under both P supplies, BnaPT37 overexpression plants showed Pi toxic symptoms (Figure 5A,E).Similar results were observed by overexpressing OsPHT1;3 and OsPHT1;8 [28,29].Moreover, the overexpression of BnaPT37 significantly reduced P distribution to new leaves compared to that of the wild type (Figure 7).This is different from ZmPHT1;7 and OsPHT1;8.Overexpression of ZmPHT1;7 led to more P accumulation in new leaves, while overexpression of OsPHT1;8 significantly increased the inorganic P concentration in new leaves at low P levels [29,36].Here, we found that overexpression of BnaPT37 significantly increased the inorganic P concentration in the xylem sap as well as the ratio of root to shoot translocation (Figure 6).Similar results were observed by overexpressing OsPHT1;4 in rice [32].In addition, overexpression of BnaPT37 facilitated the movement of P from a fully expanded leaf to a new leaf and root (Figure 8).This is quite different from the Arabidopsis homologous gene previously reported.Furthermore, overexpression of BnaPT37 decreased flowering time, plant height, and P accumulation in seeds under both P treatments (Figures 9 and 10).However, seed yield was not affected under normal P supply (Figure 10).
Hydroponic culture experiment showed that overexpression of BnaPT37 significantly increased the transport of Pi from root to shoot, but more Pi was accumulated in cotyledon and old leaves, and the distribution of P to new leaves was significantly reduced (Figures 6 and 7).This is different from Arabidopsis AtPHT1;5 overexpression lines, which showed a significant decrease in P concentration in rosette leaves, and P in siliques increased significantly [26].Here, overexpression of BnaPT37 led to a decrease in yield and P accumulation in seeds under low P conditions, and significantly reduced P accumulation in seeds under normal-P treatment (Figure 10).We speculated that the overexpression of BnaPT37 in the rapeseed ripening stage might retain more P in the fallen leaves, leading to a decrease in Pi transport to the seeds.Taken together, our data imply that BnaPT37 is a plasma membrane-localized Pi transporter that might be involved in Pi translocation from root to shoot and Pi distribution from source organs to sink organs in B. napus.

Plant Materials and Growth Conditions
The B. napus cultivar "Westar 10" and the gene overexpression lines were used in this study.The plants were grown in hydroponic conditions in an illuminated culture room.The light intensity was 300-320 µmol m −2 s −1 , the light period was 16 h light/8 h dark, and the temperature was 22 • C. Seeds were surface-sterilized for 12 min using 1% NaClO and washed five times with pure water.Then, the surface-sterilized seeds were immersed for 24 h in deionized water prior to germination on moistened gauze.After 5 d of growth in a solution containing 0.5 mM CaCl 2 , the uniformly sized seedlings were transferred to Hoagland's nutrient solution [48].Two P levels including high P (1 mM) and low P (10 µM) were set.The nutrient solution was refreshed every 3 d.The Arabidopsis seedlings were grown on a 1/2 MS solid medium [49] and treated with 625 µM Pi and 15 µM Pi conditions, respectively.The Arabidopsis seedlings and transgenic materials of B. napus were cultured for about 15 days and 19 days, respectively.Then, at least 3 biological replicates were sampled.
For the field trial, the B. napus cultivar "Zhongshuang 11" was grown in the field under normal (CK, 90 kg ha −1 P 2 O 5 )-and low-P (LP, 15 kg ha −1 P 2 O 5 ) conditions.Except for P fertilizer, the application rates of other fertilizers were as follows: nitrogen (N) 180 kg N ha −1 , K 2 O 120 kg ha −1 and borax 15 kg ha −1 .Before sowing, K (potassium chloride) and P (ordinary superphosphate) were applied.N as urea was divided into 120 kg before planting, 30 kg during the seedling stage, and 30 kg at the bolting stage.Different tissues after sampling were used for RNA isolation.Three biological replicates with two plants each were sampled at harvest.For the pot culture, two P treatments were normal (CK, 150 mg kg −1 P 2 O 5 ) and low P (LP, 20 mg kg −1 P 2 O 5 ).Both the BnaPT37 overexpression lines and the wild-type plants were grown in a pot containing 7 kg of soil.The basic properties of the soil were as follows: pH 6.36 (soil:water ratio of 1:2.5), organic matter 5.92 g kg −1 , alkaline hydrolysis N 9.46 mg kg −1 , available P 5.98 mg kg −1 , available potassium 113.38 mg kg −1 .Four replicates for each treatment were applied.Each pot was soaked in 2000 mL of water containing 3.01 g of KNO 3 , 1.75 g of MgSO 4 •7H 2 O, 4.64 g of (NH 4 ) 2 SO 4 and 7 mL of the Arnon storage solution (1000×), 7 mL of FeSO 4 •EDTA and the Hogland storage solution (200×).P concentration was measured at rapeseed flowering and ripening stages, and agronomic traits and seed yield were analyzed at the ripening stage.

Gene Structure and Sequence Analysis
According to the gene ID of BnaPT37 (BnaC04g46050D) from the "Darmor-bzh" genome database (http://www.genoscope.cns.fr/brassicanapus/,accessed on 15 May 2017) [50], the reference genome sequences, CDS, and amino acid sequences were obtained.The amino acid sequence of the B. napus BnaPT37 gene was compared to that of the homologous gene in A. thaliana on the Clustal Omega website (https://www.ebi.ac.uk/Tools/msa/ clustalo/, accessed on 20 May 2017).Transmembrane domains and conservative domains were predicted using HMMTOP (http://www.enzim.hu/hmmtop/html/submit.html, accessed on 15 July 2022).The transmembrane domain model was displayed using TOPO2 (http://www.sacs.ucsf.edu/TOPO2/,accessed on 16 July 2022) software.The Expasy website (https://prosite.expasy.org/prosite.html,accessed on 16 July 2022) was used to predict the significant domains.The significant domains were predicted by Expasy (https://prosite.expasy.org/prosite.html,accessed on 16 July 2022).MEGA version 7.0 software was used for phylogenetic relationship analysis with 1000 bootstrap replicates.The CDS and amino acid sequences of the BnaPT37 gene and the Arabidopsis homologous gene were compared by DNAMAN version 5.2.9 software.The SWISS-MODEL online website (https://swissmodel.expasy.org/interactive,accessed on 23 September 2022) was used to predict the three-dimensional structure of proteins and then displayed through ChimeraX-1.1 software.

RNA Extraction and Real-Time Quantitative PCR Analysis
Total RNA extraction of plant samples was performed using the Eastep ® Super Total RNA Extraction Kit (Qiagen, Promega, Shanghai, China).First-strand cDNA was synthesized using the ReverTra Ace qPCR RT Master Mix with the gDNA Remover Kit (Toyobo, Osaka, Japan).Real-time RT-PCR was performed using the ABI7300 Real-time Detection System (Applied Biosystems, Foster City, CA, USA) with the SYBR Premix Ex TaqTM II (Toyobo, Shanghai, China).The AtUBC9 and BnaEF1-a genes were used as reference genes.The relative expression levels of the target genes were calculated by the 2 −∆∆CT method [51].All the primers used in this study are listed in Table S1.

Vector Construction and Transformation in Arabidopsis and B. napus
For the overexpression of BnaPT37 in Arabidopsis and B. napus, the 1605 bp CDS was cloned into the PBI121s vector driven by the 35S promoter.The purified PCR product was digested with XbaI and SmaI and cloned into the PBI121s vector.For the tissue localization vector of BnaPT37, a 2000 bp DNA fragment upstream of the translation start codon was amplified from the genomic DNA.The purified PCR product was digested with XmaI and cloned into the DX2181b vector.All constructs were transformed into Agrobacterium agrobacterium strain GV3101 by the electroporation method and then transformed into the Arabidopsis atpt1/2 double mutant [52] and the B. napus cultivar "Westar 10" [53].The Arabidopsis seeds were selected on a 1/2 MS medium containing 50 mg of L −1 kanamycin.And the B. napus transgenic plants were identified by PCR amplification of target fragments.The homozygous lines were used for phenotype analysis.The primers used in this study are listed in Table S1.

Subcellular Localization Analysis
For transient expression analysis in Arabidopsis protoplasts, the open reading frame sequence of BnaPT37 was fused in frame into the position between the 35S promoter and the green fluorescent protein (GFP) sequence in the PM999 plasmid.Because the PM999 vector has a short fragment, the PM999 vector was linearized by PCR amplification with the PM999-GFP-F/R primer, and the target fragment was connected to the vector through In-Fusion cloning.The Arabidopsis mesophyll protoplasts were extracted and the BnaPT37-GFP vector was transformed into the protoplasts using a PEG-mediated method.The transformation of Arabidopsis mesophyll protoplasts was performed using the method described previously [54].After incubation at room temperature in the dark for 12-16 h, the GFP signal was observed using a laser confocal microscope (TCS SP8; Leica, Wetzlar, Germany).The primers used in this study are listed in Table S1.

Tissue Localization Analysis
In order to clarify the tissue localization of the BnaPT37 gene, we amplified the 2 kb promoter region of the BnaPT37 gene and cloned it into the DX2181b vector to construct the pBnaPT37:GUS vector.The pBnaPT37:GUS transgenic B. napus plants, together with the wild-type plants, were cultivated in 1 mM P conditions for 7 days, and then transferred to 1 mM or 0 mM P conditions for additional 7 days.The GUS activity was examined according to the methods described previously [55].

Pi Transport Activity Assay in Yeast
To generate vectors for the yeast complementation test, the coding sequences of B. napus BnaPT37 and Arabidopsis PHO84 were amplified from "Westar 10" and Col-0, respectively, and cloned into the PRS426-ADH1 vector linearized with BamHI and NotI using the In-Fusion HD Cloning kit (PT5162-1; Takara Bio, Kusatsu, Japan).The above constructs and empty vector were transformed independently into the yeast mutant strain EY917, which is defective in the Pi uptake [42].The vector plasmid was transformed into yeast by the PEG/lithium acetate (LiAc) method.The transformants were screened in the growth medium lacking uracil and tryptophan (−Ura−Trp).The functional complementation assay of BnaPT37 in the yeast mutant strain was performed according to the methods described previously [28].The OD 600 values were measured at 15 h, 20 h, 30 h, 35 h, and 40 h, and were used to draw the growth curve.The primers used in this study are listed in Table S1.

Measurement of Inorganic P and Total P Concentrations in Plants
The shoot, root, and other tissues of the wild-type and transgenic seedlings were sampled separately.The Pi concentrations were measured using the malachite green methods described previously [56].About 25-50 mg of fresh tissue samples was mixed with 25 µL of 5 M H 2 SO 4 and 1.5 mL of distilled water.After centrifugation at 12,000× g at 4 • C, the supernatant was collected and diluted to an appropriate concentration.The diluted supernatant was mixed with a malachite green reaction solution (19.4 mM H 3 BO 3 , 27.64 mM (NH 4 ) 6 MO 7 O 24 •4H 2 O, 2.38 M H 2 SO 4 , 627.5 µM malachite green solid reagent, and 0.1% polyvinyl alcohol) at a 3:1 ratio.After the reaction for 30 min, 200 µL of the reaction mixture were taken to measure the absorption values at 650 nm using ELIASA (Spark; TECAN, Männedorf, Switzerland).For the determination of Pi concentrations in samples, a standard curve was developed using varying concentrations of KH 2 PO 4 .
The total P concentrations were measured using the H 2 SO 4 -H 2 O 2 methods described previously [57].About 50 mg of dried plant samples were predigested overnight in glass tubes with 2 mL of H 2 SO 4 .Then, the tubes were heated to 120 • C with four to five drops of 30% H 2 O 2 every 30 min until the solution became colorless.The total P content was measured by molybdenum blue colorimetry at 700 nm using ELISA (Spark; TECAN) after the reaction at 30 • C for 30 min.For the determination of total P content in samples, a standard curve was developed using different concentrations of KH 2 PO 4 .

Xylem Sap Collection
Five-day-old hydroponically grown B. napus seedlings were exposed to 1 mM Pi and 10 µM Pi for 14 days; then, xylem sap samples were collected.Briefly, stems were cut with a razor blade 2-3 cm above the basal stems to collect xylem sap.The xylem sap from the first 10 min was discarded to avoid contamination.About 200-300 µL were collected.Each replicate consisted of 8-9 plants, and 5 replicates collected.After a certain dilution of the collected xylem sap, the diluent was mixed with malachite green reaction solution at a 3:1 ratio.After the reaction for 30 min, 200 µL of the reaction mixture was taken to measure the absorption values at 650 nm using ELIASA (Spark; TECAN).

Statistical Analysis of Data
Statistics were performed by Duncan's test or Student's t test.Data significantly different from those in the corresponding controls are indicated as * p < 0.05, ** p < 0.01, *** p < 0.001.

Figure 1 .
Figure 1.Characterization of BnaPT37 and phylogenetic relationship with other phosphate (Pi) transporters in Arabidopsis.(A) Alignment of amino acid sequences between BnaPT37 and AtPT4.Amino acid sequences are aligned by Clustal Omega.The predicted 12 transmembrane domains of BnaPT37 were underlined and numbered as I-XII.The prediction of transmembrane domains was carried out using HMMTOP.The red box is the conserved PHT1 signature (GGDYPLSATIMSE).(B) The phylogenetic relationship of BnaPT37 with other Pi transporters in Arabidopsis.The minimum evolution tree was constructed using MEGA7 with 1000 bootstrap replicates.(C) The transmembrane model was drawn using TOPO2 website The significant domains are predicted by Expasy.Enlarged symbols indicate sites of significant structure-function importance: brown, N-glycosylation; green, protein kinase C phosphorylation; cyan, casein kinase II phosphorylation; magenta, tyrosine kinase phosphorylation; purple, amidation; and blue, N-myristoylation.* indicates conservative amino acid sequence.

Figure 1 .
Figure 1.Characterization of BnaPT37 and phylogenetic relationship with other phosphate (Pi) transporters in Arabidopsis.(A) Alignment of amino acid sequences between BnaPT37 and AtPT4.Amino acid sequences are aligned by Clustal Omega.The predicted 12 transmembrane domains of BnaPT37 were underlined and numbered as I-XII.The prediction of transmembrane domains was carried out using HMMTOP.The red box is the conserved PHT1 signature (GGDYPLSATIMSE).(B) The phylogenetic relationship of BnaPT37 with other Pi transporters in Arabidopsis.The minimum evolution tree was constructed using MEGA7 with 1000 bootstrap replicates.(C) The transmembrane model was drawn using TOPO2 website The significant domains are predicted by Expasy.Enlarged symbols indicate sites of significant structure-function importance: brown, N-glycosylation; green, protein kinase C phosphorylation; cyan, casein kinase II phosphorylation; magenta, tyrosine kinase phosphorylation; purple, amidation; and blue, N-myristoylation.* indicates conservative amino acid sequence.

Figure 2 .Figure 2 .
Figure 2. Expression pattern of BnaPT37 in response to phosphate (Pi) starvation in Brassica napus.(A,B) Gene expression patterns in the root (A) and shoot (B) in a time-course treatment.Twelveday-old seedlings were treated with +P (1 mM Pi) or −P (0 mM Pi) for 9 days, then resupplied with 1 mM Pi for 3 days.The root and shoot of the plants were sampled at 0 h, 6 h, 12 h, 24 h, 3 days, 5 days, 7 days and 9 days after the treatment, and 1 day (R1d) and 3 days (R3d) after the resupply of Pi. (C) Gene expression in different organs at seedling stage.Two-week-old seedlings were transferred into +P (1 mM Pi) or −P (0 mM Pi) for 7 days.The root, hypocotyl, basal node, petiole, cotyledon, fully expanded leaf (FEL), and new leaf were sampled.(D) Gene expression in different organs at the reproductive stage.The mature leaf (ML), cauline leaves (CL), inflorescence stems (IS), bud (Bu), flower (Fl), Husk (Hu), Seeds (Se), and carpopodium (Ca) were sampled separately at the ripen stage after CK (90 kg ha −1 P2O5) and LP (15 kg ha −1 P2O5) treatments in a field.Gene expression levels were determined by RT-qPCR.BnaEF1-a (Accession number DQ312264) was used as an internal control.Values represent means ± SD of biological replicates (n = 3).Data significantly different from those in the corresponding controls are indicated as * p < 0.05, ** p < 0.01, and *** p < 0.001 by Student's t test.

Plants 2023 , 10 Figure 3 .
Figure 3. Subcellular and tissue-specific localization of BnaPT37 protein.(A) Subcellular localization of BnaPT37.Constructs for C-terminal fusion of BnaPT37 with green fluorescent protein (GFP) were transformed into Arabidopsis protoplasts.The green signals indicate GFP, and the red signals indicate chloroplast auto-fluorescence (AF).Scale bars = 8 µm.BF, bright field; (B) pBnaPT37:GUS transgenic rapeseed plants were cultured at +P (1 mM Pi) and −P (0 µM Pi) for seven days, and stained at root (a and b), stem (c and d), node (e and j), leaf (f and k), cotyledon (g and l), flower (h and m) and silique (i and n) under +P (a,c,e-i) and −P (b,d,j-n).Shoot bar = 2 mm.Root bar = 0.5 mm.

Plants 2023 , 10 Figure 4 .
Figure 4. Evaluation of the phosphate (Pi) transport activity of BnaPT37 in a yeast mutant.(A) Complementation of yeast mutant EY917 (Δpho84Δpho87Δpho89Δpho90Δpho91) defective in Pi uptake by BnaPT37.Yeast cells harboring either the PHO84 cNDA construct (positive control) or an empty expression vector (negative control) and the BnaPT37 cDNA construct were grown in a synthetic dropout (−Ura/−Trp) liquid medium containing a 2% (w/v) galactose to OD600 = 1.Equal volumes of 10-fold serial gradient dilutions were applied to YNB without the Pi (YNB-P; pH 5.5) medium with different Pi concentrations and a 2% (w/v) glucose, and incubated at 28 °C for 4 d.(B) Growth curve of EY917 transformed with BnaPT37, PHO84, and an empty expression vector under a 60 µM Pi medium.The yeast solution was washed with sterile water and adjusted to 1 under OD600.Then, 40 µL aliquots were added to 2 mL of a synthetic dropout (−Ura/−Trp) liquid medium containing a 60 µM Pi and a 2% (w/v) glucose.The culture was incubated at 28 °C, and the OD600 was measured by sampling at different time points.

Figure 4 .
Figure 4. Evaluation of the phosphate (Pi) transport activity of BnaPT37 in a yeast mutant.(A) Complementation of yeast mutant EY917 (∆pho84∆pho87∆pho89∆pho90∆pho91) defective in Pi uptake by BnaPT37.Yeast cells harboring either the PHO84 cNDA construct (positive control) or an empty expression vector (negative control) and the BnaPT37 cDNA construct were grown in a synthetic dropout (−Ura/−Trp) liquid medium containing a 2% (w/v) galactose to OD600 = 1.Equal volumes of 10-fold serial gradient dilutions were applied to YNB without the Pi (YNB-P; pH 5.5) medium with different Pi concentrations and a 2% (w/v) glucose, and incubated at 28 • C for 4 d.(B) Growth curve of EY917 transformed with BnaPT37, PHO84, and an empty expression vector under a 60 µM Pi medium.The yeast solution was washed with sterile water and adjusted to 1 under OD600.Then, 40 µL aliquots were added to 2 mL of a synthetic dropout (−Ura/−Trp) liquid medium containing a 60 µM Pi and a 2% (w/v) glucose.The culture was incubated at 28 • C, and the OD600 was measured by sampling at different time points.

10 Figure 5 .
Figure 5. Overexpression of BnaPT37 affects rapeseed growth and phosphorus (P) concentration under two contrasting P conditions.The 5−day−old seedlings of wild-type and BnaPT37 overexpression plants were transferred to 10 µM and 1 mM P conditions for 14 d.Then, the phenotypes (A,E), dry weight (B,F), total P concentration (C,G), and Pi concentration (D,H) were investigated under 10 µM (A−D) and 1 mM (E−H) P conditions.Error bars indicate standard deviation (n = 4).Data significantly different from those in the corresponding controls are indicated as * p < 0.05, ** p < 0.01, and *** p < 0.001 by Student's t test.

Figure 5 .
Figure 5. Overexpression of BnaPT37 affects rapeseed growth and phosphorus (P) concentration under two contrasting P conditions.The 5−day−old seedlings of wild-type and BnaPT37 overexpression plants were transferred to 10 µM and 1 mM P conditions for 14 d.Then, the phenotypes (A,E), dry weight (B,F), total P concentration (C,G), and Pi concentration (D,H) were investigated under 10 µM (A−D) and 1 mM (E−H) P conditions.Error bars indicate standard deviation (n = 4).Data significantly different from those in the corresponding controls are indicated as * p < 0.05, ** p < 0.01, and *** p < 0.001 by Student's t test.

Plants 2023 , 10 Figure 6 .Figure 6 .
Figure 6.Overexpression of BnaPT37 increased the root-to-shoot translocation of phosphorus (P) in Brassica napus.The 5−day−old seedlings were transplanted into 1 mM and 10 µM P for 14 d.(A) The root−to−shoot translocation of P (n = 4); (B) Pi concentration in the xylem sap of Westar10 and BnaPT37 overexpression lines under two P treatments (n = 5).Error bars indicate the standard deviation.Data significantly different from those in the corresponding controls are indicated as * p < 0.05, ** p < 0.01, and *** p < 0.001 by Student's t test.

Figure 7 .Figure 7 .
Figure 7. Overexpression of BnaPT37 affects the allocation of phosphorus (P) to new leaves in Brassica napus.(A,B), The total P content in different tissues of Westar10 and BnaPT37 overexpression plants under low (A) − and high (B) −P conditions.The 5−day−old seedlings were transferred to 10 µM and 1 mM P until the fourth leaf appeared (about 16 d).The root, stem, cotyledon, and four leaves (leaf1 to leaf4) were sampled separately for the determination of P content; (C,D), P distribution of of Westar10 and BnaPT37 overexpression plants under low (C) − and high (D) −P conditions.Leaf1 indicates the oldest leaf, and leaf4 indicates the youngest leaf.Error bars indicate the standard deviation (n = 4).Data significantly different from those in the corresponding controls are indicated as * p < 0.05, ** p < 0.01, and *** p < 0.001 by Student's t test.

Figure 8 . 1 )Figure 8 .
Figure 8. Redistribution of phosphorus (P) in the BnaPT37 overexpression lines and the wild−type (Westar10) plants.The 5−day−old Westar 10 and BnaPT37 overexpression seedlings were transferred to a nutrient solution with 1 mM P until leaf2 appeared.Then, all the plants were transferred to a nutrient solution without P for 6 d.Different tissues, including the root, stem, cotyledon, leaf1, leaf2, and leaf 3, were harvested separately before and after P starvation treatment.The difference in P content (∆P) of each part was calculated.Error bars indicate the standard deviation (n = 4).Data significantly different from those in the corresponding controls are indicated as * p < 0.05, ** p < 0.01, and *** p < 0.001 by the Student's t test.

Plants 2023 , 10 Figure 9 .
Figure 9. Phenotyping, plant height, and dry weight of Westar10 and BnaPT37 overexpression lines at the flowering stage.(A,B) The growth performance.(C) Plant height.(D) Dry weight under low P (LP) supply.(E) Dry weight under normal P (CK) supply.The Westar10 and BnaPT37 overexpression plants were grown under CK (P2O5, 150 mg kg −1 ) and LP (P2O5, 20 mg kg −1 ) conditions until flowering in a pot trial.The stem, lower leaves, upper leaves, and flower bud were sampled separately.Scale bars = 10 cm.Each treatment contained four biological replicates.Error bars indicate the standard deviation.Data significantly different from those in the corresponding controls are indicated as * p < 0.05, ** p < 0.01, and *** p < 0.001 by Student's t test.

Figure 9 .
Figure 9. Phenotyping, plant height, and dry weight of Westar10 and BnaPT37 overexpression lines at the flowering stage.(A,B) The growth performance.(C) Plant height.(D) Dry weight under low P (LP) supply.(E) Dry weight under normal P (CK) supply.The Westar10 and BnaPT37 overexpression plants were grown under CK (P 2 O 5 , 150 mg kg −1 ) and LP (P 2 O 5 , 20 mg kg −1 ) conditions until flowering in a pot trial.The stem, lower leaves, upper leaves, and flower bud were sampled separately.Scale bars = 10 cm.Each treatment contained four biological replicates.Error bars indicate the standard deviation.Data significantly different from those in the corresponding controls are indicated as * p < 0.05, ** p < 0.01, and *** p < 0.001 by Student's t test.

Figure 10 .− 1 )Figure 10 .
Figure 10.Overexpression of BnaPT37 reduced phosphorus (P) transport to the seeds under both normal (CK) and low (LP) P supplies at the ripening stage.(A) Pod numbers per plant; (B) seed yield per plant; (C) P concentration under low−P conditions; (D) P concentration under sufficient−P conditions; (E) P content under low-P conditions; (F) P content under sufficient−P conditions.The Westar10 and BnaPT37 overexpression plants were grown under CK (P2O5, 150 mg kg −1 ) and LP (P2O5, 20 mg kg −1 ) conditions until the seed ripened in a pot trial.Different tissues of Westar10 and BnaPT37 transgenic plants, including root pod hull, straw, and seeds, were harvested separately for the determination of P concentrations and contents.Each treatment contains four biological replicates.Error bars indicate the standard deviation.Data significantly different from those in the corresponding controls are indicated as * p < 0.05, ** p < 0.01, and *** p < 0.001 by Student's t test.