Five Post-Translational Modification Residues of CmPT2 Play Key Roles in Yeast and Rice

Chrysanthemum (Chrysanthemum morifolium Ramat.) is one of the largest cut flowers in the world. Phosphate transporter Pht1 family member CmPht1;2 protein (CmPT2) plays an important role in response to low-phosphate (LP) stress in chrysanthemum. Post-translational modification (PTM) can modulate the function of proteins in multiple ways. Here, we used yeast and rice systems to study the role of putative PTM in CmPT2 by determining the effect of mutation of key amino acid residues of putative glycosylation, phosphorylation, and myristoylation sites. We chose nine amino acid residues in the putative PTM sites and mutated them to alanine (A) (Cmphts). CmPT2 recovered the growth of yeast strain MB192 under LP conditions. However, G84A, G222A, T239A, Y242A, and N422A mutants could not grow normally under LP conditions. Analysis of phosphorus absorption kinetics showed that the Km of CmPT2 was 65.7 μM. Among the nine Cmphts, the expression of five with larger Km (124.4–397.5 μM) than CmPT2 was further evaluated in rice. Overexpression of CmPT2-OE increased plant height, effective panicle numbers, branch numbers, and yield compared with that of wild type ‘Wuyunjing No. 7’ (W7). Overexpression of Cmphts-OE led to decreased plant height and effective panicle numbers compared with that of the CmPT2-OE strain. The Pi content in roots of CmPT2-OE was higher than that of the W7 under both high (normal) phosphate (HP) and LP conditions. However, the Pi content in the leaves and roots was significantly lower in the N422A-OE strain than in the CmPT2-OE strain under both HP and LP conditions. Under LP conditions, the phosphorus starvation response (PSR) genes in CmPT2-OE were inhibited at the transcription level. The expression patterns of phosphorus-related genes in T239A, Y242A, and N422A-OE under LP conditions were different from those of CmPT2-OE. In conclusion, these five post-translational modification residues of CmPT2 play key roles in modulating the function of CmPT2. This work boosters our understanding of the function of phosphate transporters and provides genetic resources for improving the efficiency of phosphorus utilization in crop plants.


Introduction
Phosphorus (P) is one of the three essential mineral nutrients required for plants [1]. It is involved in the vast majority of plant metabolic pathways, including energy metabolism, nucleic acid synthesis, signal transduction, and phosphorylation [2]. Soluble phosphates (HPO 4 2− or H 2 PO 4 − ) are the main forms of phosphorus taken up and utilized by plants. The immobilization of most phosphate from fertilizer in the soil and the organic forms

Yeast Complementation
The verification of complementary growth indicated that yeast strain MB192 cells heterologously expressing CmPT2 (Yp112-CmPT2) and WT grew well under both 20 and 60 μM Pi (LP) conditions. G84A, G222A, T239A, Y242A, and N422A of the Cmphts yeast transformant did not grow well under LP conditions and the transformation of empty carrier p112A1NE, whereas others had no obvious change with Yp112-CmPT2. However, under high phosphorus (HP, 100 μM Pi and YPDA) conditions, all yeast transformants grew normally ( Figure 2). According to the OD600 value curves of the WT, Yp112, Yp112-CmPT2, and Yp112-Cmphts at different culture times, the growth rate of Yp112-CmPT2 was significantly higher than that of Yp112. However, it did not reach the growth rate of the wild type, indicating that CmPT2 can partially restore the growth of mutant MB192. Cultures with point mutations G84A, G222A, T239A, Y242A, and N422A did not grow as well as Yp112-CmPT2 at different culture times under LP conditions (60 μM) ( Figure 3). Yp112-CmPT2, WT, and most point mutation yeast strains grew best at a pH of 4.0. However, T239A mutants were little affected by changes in pH. N422A mutants grew best at pH 7.0. G449A and G456A transformants grew best at pH 6.0. The S510A transformant grew best at a pH of 5.0 ( Figure 4). The dynamic curves of the point mutation transformants G84A, G222A, T239A, Y242A, and N422A all deviated from normal Michaelis-Menten kinetics. Radioisotope 32 P-labeled phosphate determination of Km of Yp112-CmPT2 was 65.7 µM, with a maximum absorption rate of 170 pmol Pi (mg yeast cells min −1 ). In the mutants, the kinetic curves of G84A, G222A, T239A, Y242A, and N422A did not conform to the Michaelis-Menten equation. The Km values ranged from 124.4 to 397.5 µM, much larger than the Km value of Yp112-CmPT2 ( Figure 5).

Yeast Complementation
The verification of complementary growth indicated that yeast strain MB192 cells heterologously expressing CmPT2 (Yp112-CmPT2) and WT grew well under both 20 and 60 μM Pi (LP) conditions. G84A, G222A, T239A, Y242A, and N422A of the Cmphts yeast transformant did not grow well under LP conditions and the transformation of empty carrier p112A1NE, whereas others had no obvious change with Yp112-CmPT2. However, under high phosphorus (HP, 100 μM Pi and YPDA) conditions, all yeast transformants grew normally ( Figure 2). According to the OD600 value curves of the WT, Yp112, Yp112-CmPT2, and Yp112-Cmphts at different culture times, the growth rate of Yp112-CmPT2 was significantly higher than that of Yp112. However, it did not reach the growth rate of the wild type, indicating that CmPT2 can partially restore the growth of mutant MB192. Cultures with point mutations G84A, G222A, T239A, Y242A, and N422A did not grow as well as Yp112-CmPT2 at different culture times under LP conditions (60 μM) ( Figure 3). Yp112-CmPT2, WT, and most point mutation yeast strains grew best at a pH of 4.0. However, T239A mutants were little affected by changes in pH. N422A mutants grew best at pH 7.0. G449A and G456A transformants grew best at pH 6.0. The S510A transformant grew best at a pH of 5.0 ( Figure 4). The dynamic curves of the point mutation transformants G84A, G222A, T239A, Y242A, and N422A all deviated from normal Michaelis-Menten kinetics. Radioisotope 32 P-labeled phosphate determination of Km of Yp112-CmPT2 was 65.7 µM, with a maximum absorption rate of 170 pmol Pi (mg yeast cells min −1 ). In the mutants, the kinetic curves of G84A, G222A, T239A, Y242A, and N422A did not conform to the Michaelis-Menten equation. The Km values ranged from 124.4 to 397.5 µM, much larger than the Km value of Yp112-CmPT2 ( Figure 5).

PCR and RT-PCR Identification of the Transgene Rice
We transformed the CmPT2 and its five point mutants that did not complement the growth of MB192 into wild type W7. DNA was extracted from the T1 and T2 generations of hygromycin-resistant plants using sodium dodecyl sulfate (SDS), and the transgenic plants were identified with primers CmPT2-1100-F and GFP-R with a target fragment size of approximately 590 bp. The results are presented in Supplementary Table S3. Among the 17 CmPT2-OE lines, OE-8-4, OE-11-1, OE-11-5, and OE-11-11 were all homozygous, while the others were heterozygous. Three G84A-10 strains were homozygous. G222A-4-5 was homozygous, while the others were heterozygous. Nine lines of T239A-3 (3 strains), Y242A-4 (3 strains), and N422A-7 (3 strains), were heterozygous.
After PCR identification, RNA was extracted from the roots of three positive plants per T1 generation line. The transcript level of CmPT2-OE-8 was defined as 1. The expression levels of CmPT2-OE-9, -10, and -19 were 4.3, 4.96, and 3.1 times than those of CmPT2-OE-8, respectively. The expression levels of CmPT2-OE-1 and -11 were 2.48 and 2.25 times those of CmPT2-OE-8, respectively. The expression levels of CmPT2-OE-2 and -29 were 7% and 9% of CmPT2-OE-8, respectively (Supplemental Figure S1A). Expression of the post-translational modification-mutated Cmphts was tested in various lines of transgenic rice G84A, G222A, T239A, Y242A, and N422A. The RT-PCR results showed that among the 10 lines of G84A, the expression levels of CmPT2 G84A in G84A-10 and G84A-5 were 4.42 and 1.5 times those of CmPT2 in CmPT2-OE-8. In nine lines of G222A, the expression levels of CmPT2 G222A were lower than those of CmPT2 in CmPT2-OE-8. The expression levels of G222A-4 and -8 were 62% and 54% of CmPT2-OE-8, respectively. Among the 10 T239A lines, the highest relative expression of CmPT2 T239A was in T239A-3, approximately 1.16 times CmPT2 expression in CmPT2-OE-8. The expression level of CmPT2 T239A in T239A-10 was 94% of that of CmPT2 in CmPT2-OE-8. Among the nine lines of Y242A, the highest expression levels of CmPT2 Y242A were observed in Y242-A-4, approximately 1.23-fold higher than that of CmPT2 in CmPT2-OE-8. Among the nine lines of N422A, the highest CmPT2 N422A expression was observed in N422A-7, approximately 2.17 times CmPT2 expression in CmPT2-OE-8 (Supplemental Figure S1B-F).

P/Pi Content of T2 Generation in the CmPT2/Cmphts Overexpression Plants
The effective phosphorus content was determined from rice leaves and roots after treatment under different P i conditions. Under both HP and LP conditions, the phosphate content in leaves was not significantly different among CmPT2-OE, Cmphts-OE, and W7 ( Figure 8A,C). However, under LP-stress conditions, CmPT2-OE strains had a higher phosphate content than wild-type roots ( Figure 8B). The P i content in the roots of CmPT2-OE-4-1 and CmPT2-OE-9-1 was 2.33 and 1.725 times that of W7 under LP stress, respectively ( Figure 8C). Notably, the P i content in the roots of N422A-7-1 was lower than that of CmPT2-OE-8-1 under LP conditions, although the P i content of the other mutants was not significantly different from that of CmPT2-OE-8-1 ( Figure 8D).

P/Pi Content of T2 Generation in the CmPT2/Cmphts Overexpression Plants
The effective phosphorus content was determined from rice leaves and roots after treatment under different Pi conditions. Under both HP and LP conditions, the phosphate content in leaves was not significantly different among CmPT2-OE, Cmphts-OE, and W7 (Figure 8 A,C). However, under LP-stress conditions, CmPT2-OE strains had a higher phosphate content than wild-type roots ( Figure 8B). The Pi content in the roots of CmPT2-OE-4-1 and CmPT2-OE-9-1 was 2.33 and 1.725 times that of W7 under LP stress, respectively ( Figure 8C). Notably, the Pi content in the roots of N422A-7-1 was lower than that of CmPT2-OE-8-1 under LP conditions, although the Pi content of the other mutants was not significantly different from that of CmPT2-OE-8-1 ( Figure 8D).

Expression Analysis of CmPT2/Cmphts-OE in T2 Generation Materials
The expression levels of phosphorus-related genes in transgenic and wild-type plants showed that the expression levels of OsSPX2, OsIPS1, OsPT2, OsPT3, OsPT6, OsPT8, and OsPT10 were not significantly different between the W7 and CmPT2-OE plants under normal Pi conditions. Under LP conditions, the expression of these phosphorus starvation response (PSR) genes in transgenic plants was significantly lower than that in wild-type plants. This suggests that transgenic plants are less sensitive to LP stress than wild-type plants. The degree of PSR gene expression varied among different transgenic lines and was related to CmPT2 expression levels. The PSR genes were expressed at low levels in CmPT2-OE-9-1, which had a high CmPT2 expression level. However, they were highly expressed in CmPT2-OE-1-7 and CmPT2-OE-11-5, which showed low CmPT2 expression levels in these lines (Figure 9).

Expression Analysis of CmPT2/Cmphts-OE in T2 Generation Materials
The expression levels of phosphorus-related genes in transgenic and wild-type plants showed that the expression levels of OsSPX2, OsIPS1, OsPT2, OsPT3, OsPT6, OsPT8, and OsPT10 were not significantly different between the W7 and CmPT2-OE plants under normal P i conditions. Under LP conditions, the expression of these phosphorus starvation response (PSR) genes in transgenic plants was significantly lower than that in wild-type plants. This suggests that transgenic plants are less sensitive to LP stress than wild-type plants. The degree of PSR gene expression varied among different transgenic lines and was related to CmPT2 expression levels. The PSR genes were expressed at low levels in CmPT2-OE-9-1, which had a high CmPT2 expression level. However, they were highly expressed in CmPT2-OE-1-7 and CmPT2-OE-11-5, which showed low CmPT2 expression levels in these lines (Figure 9).

CmPT2 Could Complete the Growth of Yeast Mutant MB192
Plants need to absorb phosphate at the root/soil interface through specific phosphate transporter proteins and complete the transmembrane transport of phosphorus between cells [6]. The heterologous transformation of genes is an important means of identifying gene function. There have been many reports on the functional identification of the Pht

CmPT2 Could Complete the Growth of Yeast Mutant MB192
Plants need to absorb phosphate at the root/soil interface through specific phosphate transporter proteins and complete the transmembrane transport of phosphorus between cells [6]. The heterologous transformation of genes is an important means of identifying gene function. There have been many reports on the functional identification of the Pht gene from different plant sources with functionally assisted validation by complementation of yeast phosphate transport-related mutants [14]. Analysis of the enzymatic kinetic activity of phosphate transport proteins can be performed using a 32 P radioisotopic assay, which was used to verify the function of CmPT1 (AGK29560) [15]. The sequence identity between chrysanthemum gene CmPT2 and CmPT1 is 90.8% [13]. In this study, CmPT2 was able to complement the ability of the yeast phosphate transport deletion mutant MB192 to grow in LP environments, with a K m of 65.7 µM using a 32 P radioisotope. Yp112-CmPT2 grew best at pH 4.0, indicating that CmPT2 belongs to the hydrogen proton co-transport high-affinity phosphate transporter protein and plays an important role in the phosphate absorption process in yeast. These results indicated that CmPT2 is a high-affinity phosphate transporter protein in chrysanthemums.

Five of the Nine Mutants of CmPT2 Could Not Complete the Growth of MB192
Furthermore, nine sites of protein post-translational modification (glycosylation, phosphorylation, myristoylation of CmPT2 protein) were analyzed. When serine 514 Ser was mutated to aspartate (Asp), the plasma membrane localization changed to the endoplasmic reticulum (ER) [16]. However, in the subcellular localization results of transient expression in the onion epidermis, the extent of plasma membrane localization did not change after mutation in the post-translational modified amino acid sites (Supplementary Figure S1). Studies in Arabidopsis have found that a mutation in Y in the AtPht1;1 protein disrupts the formation of its homodimer and enhances P i uptake [17]. Here, complementation experiments with yeast mutant MB192 showed that G84A, G222A, T239A, Y242A, and N422A yeast transformants could not grow normally under LP conditions. The optimum pH of the five mutant yeast transformants was also changed. The greater the K m value, the lower the affinity of the enzyme for the substrate and the lower the enzyme activity. The K m values of these five mutants were much greater than that of Yp112-CmPT2, and we speculated that the affinity of the transporter for P i was decreased. These data indicate that the loss of function or alteration in yeast was influenced by mutations in CmPT2 post-modified amino acid sites. We speculated that changes in the kinetic curve may be due to morphological changes in the protein underpinning its function.

CmPT2 May Improve Rice Yield by Increasing Phosphate Absorption
Heterologous transformation in gene function analysis has been reported in plants.
Overexpression of the Arabidopsis high-affinity phosphate transporter gene PHT1 in tobacco culture cells can promote cell growth under LP conditions [18]. On expressing the barley high-affinity phosphate transporter gene Pht1;1 in rice, the phosphate absorption rate significantly increased [19]. Overexpression of the tobacco high-affinity phosphate transporter gene NtPT1 promoted phosphate uptake and accumulation in transgenic rice [20]. The Brassica napus phosphate transporter gene BnPht1;4 can promote phosphate resorption and improve the root architecture in Arabidopsis [21]. In the present study, statistics on the mature traits of T2 generation CmPT2-OE transgenic materials and wild-type plants showed that the growth status of CmPT2-OE materials was improved under field management conditions. The high P i content in the roots of CmPT2-OE lines compared to that of W7 indicated that the overexpression of CmPT2 may promote the absorption of phosphorus in rice and consequently improve yield in rice. Considering the RT-PCR results, we surmise that moderate expression levels of CmPT2 may promote rice growth (CmPT2-OE-1-7). CmPT2-OE-9-1 lines had decreased plant height, yellow leaf tip, and reduced tiller number, whereas CmPT2 was highly expressed and P i content was the high-est in CmPT2-OE-9-1. When the expression level of CmPT2 was high, a certain degree of the phosphorus poisoning phenotype appeared. These results were similar to those of OsPht1; 8-overexpressing rice, which demonstrated excessive phosphorus accumulation and symptoms of phosphorus toxicity [7]. However, CmPT2 was functionally redundant in a phosphate-adequate (300 µM) environment. Compared to W7, CmPT2-OE was less sensitive to LP stress and may be more adaptable to LP conditions.

Four of the Five Mutants Affected the Function of CmPT2 in Rice
To further confirm the effects of these five sites on CmPT2, we transformed them into the rice. The height of the T2-generation of G84A and N422A transgenic plants was lower than the CmPT2-OE-8-1-expressing plant, indicating that they affected the increase in growth by CmPT2 in rice. The primary branch numbers of G84, T239A, and Y242A transgenic plants decreased, suggesting that they dropped from development by CmPT2 in rice. PTMs play an important role in response to nutrient stress. Surprisingly, O-GlcNAcylation has extensive crosstalk with phosphorylation, where it serves as a nutrient/stress sensor to modulate signaling, transcription, and cytoskeletal functions [22]. The responses to P i vs. phosphate (H 2 PO 3 − ) vary in P i -starved Arabidopsis suspension cells, and the differences between them are mainly at the protein phosphorylation level [23]. Here, the effective phosphate content in the roots and leaves was lower in the N422A-7-1 strain than in the CmPT2-OE-8-1 strain, suggesting that the glycosylation of N422 of CmPT2 may be important for the phosphate transport function of CmPT2. N422 is a glycosylation site located at TM2 and TM10. These results indicated that the transmembrane domains TM2 and TM10 play critical roles in the phosphate transport process of CmPT2 proteins. This may be related to the binding of H + to P i [9]. Myristoylation promotes protein association with the cell membrane and regulates membrane protein expression. However, this targeting mechanism during plant signal processing remains unknown [24]. In addition, T239A and Y242A, which serve as phosphorylation sites located on the intermediate hydrophilic ring, also play an important role in the phosphate transport process of CmPT2. Phosphorylation regulates the reception and transduction of cellular signals. It is a highly efficient signal for post-translational modifications of proteins in cells and acts in plants in response to abiotic and biotic stress conditions [25]. Therefore, four PTMs, G84, T239, Y242, and N422, might be the key amino acid sites in the function of CmPT2 in rice. However, G222A might have had little effect here. This suggests that these post-translational modification sites may be closely related to signal transduction at the cell membrane, perhaps via the recognition of H + by P i binding.

CmPT2 May Regulate the Expression of the PSR Genes Directly or Indirectly
PHR1 is considered a central regulator of the plant response to phosphorus deficiency stress, an intermediate vector from phosphorus signaling to downstream high-affinity phosphate transporters. It is involved in the expression of most downstream phosphorusdeficiency-responsive genes, including SPX (SPX domain protein) and IPS (induced by phosphate starvation) [26]. The SPX domain can indicate the phosphate state in fungal, plant, and human cells. In eukaryotes, inorganic phosphorus uptake, transport, storage, and signal transduction are inseparable from SPX domain-containing proteins [27]. In rice, a negative feedback regulatory pathway of phosphorus signaling and balance exists between SPX2 and PHR2 [28]. Rice IPS1, a member of the TPSll/MT4 family, is involved in the regulation of the inorganic phosphate signaling pathway and homeostasis [29]. RT-PCR results showed that SPX2 and IPS1 were induced by phosphorus deficiency at the transcriptional level in wild-type rice. SPX2 and IPS1 were significantly inhibited, and the inhibition level correlated with the expression level of CmPT2 in CmPT2-OE lines. These results suggest that CmPT2 reduced the sensitivity of transgenic plants to LP stress. However, the expression patterns of P-related genes in the T239A-3-1, Y242A-4-1, and N422A-7-1 overexpression lines were different from those in the CmPT2-OE-8-1 strain.
The rice Pht1 family has 13 genes (OsPT1-OsPT13). OsPT2 is a low-affinity phosphate transporter. Overexpression of OsPT2 caused a P-poisoning phenotype and a large increase in P content. Inhibition of OsPT2 expression results in a low rate of shoot P uptake and a low total phosphorus content [30]. OsPHT1;3 regulates the uptake and remobilization of P i under extreme LP conditions [31]. OsPT6, a high-affinity phosphate transporter, can complement the yeast phosphate uptake mutant. Inhibition of OsPT6 expression also reduces the rate of shoot P uptake and affects the ability to transport P to the shoots [32]. OsPT8, similar to OsPT6, fully complemented the yeast pho84 mutants. Overexpression of OsPT8 causes a phosphotoxic phenotype in rice [7]. OsPT9 and OsPT10 are specifically induced by P starvation in the root epidermis, root hair, and lateral roots. Downregulation of OsPT9 or OsPT10 did not significantly reduce P uptake under low-or high-P conditions [33]. The function of OsPT9 or OsPT10 may be redundant for phosphate uptake. In this study, the expression of Pht1 family genes was tested in wild-type and CmPT2-OE rice treated with different phosphate concentrations. Under LP conditions, the expression of phosphate transporter protein genes, OsPT2, OsPT3, OsPT6, OsPT8, and OsPT10, was strongly induced. The expression of OsPT2, OsPT3, OsPT6, and OsPT8 was significantly lower in the CmPT2-OE lines than that in W7. The degree of PSR gene expression varied among transgenic lines and correlated with CmPT2 expression levels. This shows that chrysanthemum phosphate transporter CmPT2 might replace the phosphate transporter in rice. However, the expression level of OsPT10 was not significantly different between CmPT2-OE and W7 plants, confirming the redundancy of OsPT10. However, under LP conditions, OsPT3 and OsPT6 expression levels were significantly higher in the T239A-3-1, Y242A-4-1, and N422A-7-1 lines than in the CmPT2-OE-8-1 strain. OsPT8 expression was significantly increased, whereas OsPT10 expression was significantly decreased in the Y242A-4-1 OE line. Thus, mutations in these three sites (two phosphorylation sites and one glycosylation site) may be important for the regulation of P-related genes by CmPT2 directly or indirectly.

Site-Directed Mutagenesis
Based on the restriction enzyme site in the yeast expression vector (p112A1NE), fusion high-fidelity PCR was performed to construct the yeast expression vector p112A1NE-CmPT2 with 112-EcoR-F and 112-Not-R. After analysis of CmPT2 protein transmembrane domain and protein post-translational modification (glycosylation, phosphorylation, myristoylation) sites [9,13]. PCR-based site-directed mutagenesis was applied to the p112A1NE-CmPT2 construct using fusion high-fidelity DNA polymerase based on primers designed according to the protocol of the Quickchange ® Site-Directed Mutagenesis Kit (Stratagene, La Jolla, CA, USA) [34]. All mutated primer sequences are listed in Supplemental Table  S1. Primers were synthesized by Generay Biotech Co., Ltd. (Shanghai, China). The nine mutants were named G75A, G84A, G222A, T239A, Y242A, N422A, T425A, G456A, and S510A. All plasmid constructs were sequenced to ensure that no unexpected mutations or cloning errors occurred.

Construction of a Green Fluorescent Protein (GFP) Fusion Vector and Intracellular Localization Analysis
Site-directed mutagenesis was performed using the plasmid pENTR TM 1A-CmPT2 constructed in our previous studies [13]. All mutated constructs were recombined into the C-terminus GFP fusion vector pMDC43 by the LR reaction (as described in Gateway ® Technology with Clonase ® II) [35]. Plasmid DNA was bombarded into onion (Allium cepa) epidermal cells using a helium-driven gene gun (PDS-1000; Bio-Rad, Hercules; CA, USA) according to the manufacturer's instructions [36]. The onion epidermal cells were incubated on Murashige and Skoog (MS) solid medium plates in the dark for 16-20 h. GFP expression was monitored using confocal laser scanning microscopy at 488 nm (Zeiss, Oberkochen, Germany) [37].

Functional Complementation Assay of CmPT2 and Cmphts in Yeast
The yeast expression vector p112A1NE carrying the p112A1NE empty vector and CmPT2 and Cmphts open reading frame was transformed into MB192 (named Yp112, Yp112-CmPT2, and Yp112-Cmphts, respectively). The wild-type yeast strains, MB192, Yp112, Yp112-CmPT2, and Yp112-Cmphts were grown to the logarithmic phase and then spotted onto yeast nitrogen base medium plates containing different P i concentrations (20, 60, and 100 mM) evenly [34]. The cells in the logarithmic phase were also transferred into the YNB medium containing 60 µM P i for 40 h. The optical density of the yeast culture was measured every 8 h. To substantiate the pH dependence of P i uptake, different extracellular pH values from 4 to 8 were used at a fixed amount of 80 mM KH 2 PO 4 . For 32 P uptake experiments in yeast, approximately 5 mg of fresh yeast cell sample was used following a previously described method [38].

Regeneration of CmPT2 and Cmphts-Overexpressing Rice
The rice overexpression vectors pCAMBIA1300-CmPT2 and pCAMBIA1300-Cmphts were constructed using p112A1NE-CmPT2 and p112A1NE-Cmphts with the recombinant primers CmPT2-CE-F and CmPT2-CE-R, respectively. Primers were synthesized by Beijing Qingke Xinye Biotechnology Co., Ltd. (Beijing, China). The transgenic expression of CmPT2/Cmphts was carried out at Jian Gu Biotechnology Co., Ltd., Hefei, Anhui, China. Transgenic hygromycin-resistant plants overexpressing CmPT2 or Cmphts were identified by PCR and RT-PCR at the DNA and RNA levels and named CmPT2 or Cmphts-OE, respectively. Primer sequences are shown in Supplemental Table S1. The relative expression of CmPT2 was calculated using the 2 −∆∆Ct method [39].

Treatments of Plant Materials
All experimental materials were planted in the experimental base of the China Rice Research Institute in Fuyang District, Hangzhou in summer (sown on 24 May, 2020 and transplanted on 20 June, 2020) and southern propagation base of the Hainan Lingshui Rice Research Institute in winter (sowed on 14 December, 2020 and transplanted on 14 January, 2021), with conventional field management. Fertilization occurred at the tillering and heading stages. The amount of N fertilizer was 165 kg/hm 2 , with m (base fertilizer):m (tillering fertilizer):m (panicle fertilizer) = 5:2:3. The amount of K fertilizer was 165 kg/hm 2 , with m (tiller fertilizer):m (ear fertilizer) = 7:3. The amount of P fertilizer was 90 kg/hm 2 as base fertilizer. The phenotypes were counted under these conditions after ripening.
The rice materials were treated with nutrient solutions (Supplementary Table S2) of different phosphate concentrations. After seven days of hydroponics in a 96-cell rice hydroponic box, the seedlings were transferred to a plastic incubator filled with 20 L of different phosphate concentrations (300 µM P i , 10 µM P i ) for 10 d [7] and subjected to RT-PCR and P i content determination. Each rice seedling was fixed on a colonization plate with a colonization sponge. The nutrient solution was changed every three days during these periods. The expression of PSR genes and characteristics of T2 generations, W7, CmPht1,2-OE, G84A, G222A, T239A, Y242A, and N422A, were investigated. The soluble phosphate content was determined using the molybdenum blue method with three biological replicates [40].

Expression Analysis of Pi Absorption-Related Genes in CmPT2 and Cmphts-Overexpressing Rice
The template used for quantitative RT-PCR was the cDNA reverse recorded in the above step. The enzyme used was ChamQ Universal SYBR qPCR Master Mix (Novzan Biotechnology Co., Ltd., Nanjing, China). The relative expression of target genes was calculated using the 2 −∆∆Ct method [39]. Primers were synthesized by Qingke Xinye Biotechnology Co., Ltd. (Beijing, China) and are shown in Supplemental Table S1.

Statistical Analysis
All the significance discriminate analyses were carried out using the SPSS software (IBM SPSS Statistics Version 20), and the phenotypes of the OE lines and WT were compared by Student's t-test at a 5% probability level.

Conclusions
This study explored the function of the chrysanthemum phosphate transporter CmPT2 in phosphate absorption in yeast and rice. CmPT2 belongs to the hydrogen proton cotransport high-affinity phosphate transporter. It plays an important role in the growth and phosphate uptake of yeast and rice. We also defined key post-translational modification amino acids affecting the function of CmPT2. The four PTM residues, G84, T239, Y242, and N422, might be the key amino acid sites for the function of CmPT2.
Supplementary Materials: The following are available online at https://www.mdpi.com/article/10 .3390/ijms24032025/s1. Author Contributions: G.X., J.J., F.C. and S.C. designed the study. J.T. and Y.T. performed the transgenic expression of CmPT2 in rice and its relative phenotype. C.L. generated the PTM point mutants and yeast. All authors carried out field experiments. All authors have read and agreed to the published version of the manuscript.