Phosphite-Mediated Suppression of Anthocyanin Accumulation Regulated by Mitochondrial ATP Synthesis and Sugars in Arabidopsis

Phosphite-Mediated Suppression of Anthocyanin Accumulation Regulated by Mitochondrial ATP... Abstract Despite the essential role of phosphate (Pi) in plant growth and development, how plants sense and signal the change of Pi supply to adjust its uptake and utilization is not yet well understood. Pi itself has been proposed to be a signaling molecule that regulates Pi starvation responses (PSRs) because phosphite (Phi), a non-metabolized Pi analog, suppresses several PSRs. In this study, we identified a phosphite-insensitive1 (phi1) mutant which retained anthocyanin, a visible PSR, in Phi-containing but Pi-deficient medium. phi1 mutants were impaired in the gene encoding an FAd subunit of mitochondrial F1Fo-ATP synthase and showed a reduced mitochondrial ATP level in roots, growth hypersensitivity to oligomycin and an increased mitochondrial membrane potential, suggesting that this gene has a crucial role in mitochondrial ATP synthesis. phi1 mutants accumulated a high level of sugars in shoots, which may account for the increased accumulation of anthocyanin and starch in Phi-containing conditions. Gene expression analysis showed that a subset of genes involved in carbohydrate metabolism in phi1 was misregulated in response to Phi. The majority of genes were repressed by Pi starvation and, unlike wild-type plants, their repression in phi1 was not affected by the addition of Phi. Our findings show that defective mitochondrial ATP synthesis results in sugar accumulation, leading to alteration of Phi-mediated suppression of PSRs. This study reinforces the role of sugars, and also reveals a cross-talk among ATP, sugars and Pi/Phi molecules in mediating PSRs. Introduction Phosphorus (P) is an essential element for cell growth and development. It serves as a component of phospholipids and nucleic acids, and is also an energy source and participates in signaling transduction (Marschner 2012). Plant roots predominately absorb P in the form of inorganic phosphate (Pi, HPO42– or H2PO4–); however, the availability of Pi in the soil is often limited because Pi easily forms aggregates with cations or is converted into its organic form (Raghothama 1999). To cope with the low availability of Pi, plants have developed a series of adaptive reactions, collectively known as Pi starvation responses (PSRs), to enhance Pi uptake from the soil as well as to readjust Pi distribution and usage within plants. Some of the well-studied PSRs include modification of root system architecture, enhanced expression of Pi transporters, secretion of phosphatases and RNases, remodeling of membrane lipids, accumulation of anthocyanin and increased starch formation (Lopez-Arredondo et al. 2014). The transcriptional regulation underlying these PSRs, in which a large number of genes, ranging from 500 to 2,000 depending on the experimental conditions, were up-regulated upon Pi starvation, has been a focus of several studies (Misson et al. 2005, Bustos et al. 2010, Secco et al. 2013, Liu et al. 2016). A MYB-related transcription factor family in Arabidopsis, PHOSPHATE STARVATION RESPONSE1 (PHR1) and PHR1-LIKE1 (PHL1), are involved in activating the expression of a considerable number of Pi starvation-induced (PSi) genes (collectively named PHR1 regulons), which generally harbor a PHR1 binding sequence (P1BS) motif in their promoter (Rubio et al. 2001, Bustos et al. 2010). Nonetheless, 30% and 50% of PSi genes in shoots and roots, respectively, are not significantly reduced in phr1 phl1 mutants (Bustos et al. 2010), suggesting the existence of PHR1-independent regulation. On the other hand, the transcriptional regulation of Pi starvation-repressed (PSr) genes is virtually unexplored. In comparison with our knowledge about the downstream PSRs, how Pi starvation signaling is perceived and transduced is less understood. Several compounds, such as sugars, hormones and microRNA, have been proposed to be potential signaling molecules (Chiou and Lin 2011, Hammond and White 2011, Lin et al. 2014), and several findings support a role for sugar in regulating PSRs. For example, induction of PSi genes in Arabidopsis and white lupin roots was reduced by blocking shoot to root translocation of sugars, through either the impairment of phloem-specific SUCROSE TRANSPORTER2 (SUC2) or stem girdling experiments (Krapp and Stitt 1995, Lloyd and Zakhleniuk 2004, Liu et al. 2005, Lei et al. 2011). In contrast, overexpression of sucrose transporters resulted in accumulation of sugars in both shoots and roots, coinciding with the activation of PSi genes (Lei et al. 2011, Dasgupta et al. 2014). Moreover, exogenous application of sugars augmented the expression of several PSi genes under Pi deficiency (Liu et al. 2005). Transcriptomic analysis of Arabidopsis leaf also revealed the interaction between sugar and Pi in regulating gene expression, in which a cluster of PSi genes specifically involved in alleviating Pi starvation was further induced by sucrose supply (Müller et al. 2007). Therefore, sucrose was considered to be a global regulator of PSRs although the underlying mechanism is unknown. Dasgupta et al. (2014) proposed that the effect of sugars on the PSRs may result from the sequestration of Pi by excess sugars or the disruption of carbon (C) to P balance. In addition, the Pi molecule itself has been proposed to be a signal for the regulation of PSRs. This is supported by the finding that a non-metabolized analog of Pi, phosphite (Phi, H2PO3– or HPO32−), suppressed a number of PSRs, such as a high root to shoot ratio, root hair formation, anthocyanin accumulation, acid phosphatase activity, membrane lipid remodeling and expression of many PSi genes (Ticconi et al. 2001, Varadarajan et al. 2002, Kobayashi et al. 2006, Berkowitz et al. 2013, Jost et al. 2015). Because of the structural similarity between Pi and Phi and the results from competitive inhibition of Phi on Pi transport, Phi is believed to be taken up through the PHOSPHATE TRANSPORTER 1 (PHT1) family proteins (Danova-Alt et al. 2008, Jost et al. 2015). Once inside the root, Arabidopsis PHT1;8 and PHT1;9 Pi transporters were shown to be involved in root to shoot Phi translocation (Jost et al. 2015). How Pi or Phi regulates the expression of PSR genes has remained unclear until recently. Puga et al. (2014) found that SYG1/Pho81/XPR1 (SPX) domain-containing proteins, SPX1 or SPX2, bind to PHR1 in the presence of Pi or Phi, and prevent PHR1 from activating its target genes in Arabidopsis. Conversely, PHR1 is liberated from SPX1 binding when Pi is limited, thus enabling activation of PHR1-dependent PSi genes. Similar findings were also reported for the interaction of OsSPX1 or OsSPX2 and OsPHR2 in rice (Wang et al. 2014). These results support the notion that Pi is a signaling molecule and imply the potential of SPX proteins as Pi sensors. As Pi sources for fertilizer usage are finite, attention has been paid to breeding crops with improved P use efficiency. To achieve this aim, understanding how plants sense P availability in the rhizosphere and how the signal is transmitted to initiate adaptive responses to increase Pi uptake/usage by plants is essential. In this study, we employed Phi in a genetic screen to try to identify the regulatory components involved in the PSRs mediated by Pi molecules. We isolated a phosphite-insensitive1 (phi1) mutant which retained anthocyanin accumulation in shoots even in the presence of Phi. Further genetic and metabolite analyses revealed that the phi1 mutant is impaired in the gene encoding the FAd subunit of mitochondrial F1-Fo ATP synthase, and accumulates a high level of sugar which may account for the retention of anthocyanin under Phi treatment. In agreement with the abnormal carbohydrate metabolism, gene expression analysis revealed Phi-dependent misexpression of several starch- and sugar-related genes in phi1. Our study highlights a cross-talk among ATP, sugars and Pi/Phi molecules to co-ordinate the PSRs. Results Identification of a Phi-insensitive mutant To identify the components involved in the regulation of PSR, we established a system to select Arabidopsis mutants that responded differently to Phi as compared with wild-type (WT) plants. We first grew Arabidopsis WT seedlings in half-strength Hoagland’s medium containing different combinations of Phi and Pi concentrations to evaluate the effects on the suppression of anthocyanin accumulation by Phi. Because anthocyanin accumulation is an easily visible PSR, we employed it as a screening phenotype. To compare the anthocyanin accumulation between Pi-sufficient (500 μM Pi, +Pi) and deficient conditions (no supply of Pi, –Pi), a medium containing 500 μM Phi but without supply of Pi (+Phi) was chosen as a screening condition because it was effective in suppressing anthocyanin accumulation and caused less harm to growth. Using this condition, we screened a pool (∼60,000 lines) of activation tagging mutants of Arabidopsis (Weigel et al. 2000) and identified a mutant, phosphite insensitive1 (phi1), that retained a substantial amount of anthocyanin in the shoot compared with WT seedlings (Fig. 1A). Although Phi suppressed the growth of both WT and phi1 plants, phi1 mutants showed reduced biomass and growth of primary and lateral roots compared with the WT under all growth conditions (Fig. 1A, Supplementary Fig. S1A–D). Fig. 1 View largeDownload slide A phosphite-insensitive mutant (phi1) retains anthocyanin under +Phi conditions. The phenotype (A) and RNA level of three anthocyanin biosynthesis genes, CHS, ANS and DFR (C) of 12-day-old WT and phi1 plants grown under Pi-sufficient (+Pi), Pi-deficient (–Pi) or Phi-containing (+Phi) conditions for 6 d. Enlargement of the abaxial side of leaves under +Phi is shown. The anthocyanin content in 16-day-old WT and phi1 shoots which were treated for 10 d after 6 d germination (B). Letters represent significant differences among genotypes and growth conditions, P < 0.05 (ANOVA). Error bars represent SEs from three biological replicates. Scale bars = 1 cm. Fig. 1 View largeDownload slide A phosphite-insensitive mutant (phi1) retains anthocyanin under +Phi conditions. The phenotype (A) and RNA level of three anthocyanin biosynthesis genes, CHS, ANS and DFR (C) of 12-day-old WT and phi1 plants grown under Pi-sufficient (+Pi), Pi-deficient (–Pi) or Phi-containing (+Phi) conditions for 6 d. Enlargement of the abaxial side of leaves under +Phi is shown. The anthocyanin content in 16-day-old WT and phi1 shoots which were treated for 10 d after 6 d germination (B). Letters represent significant differences among genotypes and growth conditions, P < 0.05 (ANOVA). Error bars represent SEs from three biological replicates. Scale bars = 1 cm. When anthocyanin was measured in shoots (normalized by fresh weight), both the WT and phi1 had a very low level under +Pi, and it increased remarkably under –Pi, although phi1 had a slightly higher level than the WT (Fig. 1B). In response to a 10 d Phi treatment, the WT showed a significant reduction of anthocyanin, whereas phi1 retained a substantial amount, approximately six times that in the WT (Fig. 1B). To take into account the possible effect due to reduced shoot biomass in phi1, we normalized anthocyanin based on the number of shoots and found that phi1 still retained higher anthocyanin than the WT under +Phi (Supplementary Fig. S1E). phi1 true leaves had a higher cell density than those of the WT due to reduced cell size, but the cell number per leaf (cotyledon or true leaf) of phi1 was comparable with that of the WT (Supplementary Fig. S2). Anthocyanin was also accumulated in the cotyledons of phi1, in which cell density and number are similar to those of the WT. Therefore, the increased anthocyanin accumulation in phi1 shoots under +Phi is not caused by impaired growth. Consistent with the accumulation of anthocyanin, the expression of three selected anthocyanin biosynthetic genes, namely CHALCONE SYNTHASE (CHS), ANTHOCYANIDIN SYNTHASE (ANS) and DIHYDROFLAVONOL 4-REDUCTASE (DFR), in phi1 was also higher than that in the WT, especially under +Phi (Fig. 1C). Therefore, we concluded that Phi-mediated suppression of anthocyanin accumulation is impaired in phi1 mutants. phi1 mutants are defective in a subunit of mitochondrial ATP synthase To determine the gene responsible for the phi1 phenotype, we carried out thermal asymmetric interlaced (TAIL)-PCR (Singer and Burke 2003) and identified a T-DNA insertion at the 3'-untranslated region (UTR) of AT2G21870 encoding a plant-specific FAd subunit of mitochondrial F1-Fo ATP synthase in phi1 mutants (Fig. 2A). This T-DNA insertion resulted in significantly reduced expression of AT2G21870 mRNA in both shoots and roots under all conditions tested, as measured by quantitative reverse transcription–PCR (qRT–PCR) (Fig. 2B) and RNA gel blot (Supplementary Fig. S3) analyses. Because the phi1 mutant was generated by activation tagging, we also examined the expression of two flanking genes (AT2G21860 and AT2G21880) and found no significant differences from the WT under nearly all growth conditions (Supplementary Fig. S4), suggesting that these two genes do not contribute to the phi1 phenotype. Because of the absence of activation of neighboring genes, we suspect that the enhancer could have been lost in the T-DNA. Fig. 2 View largeDownload slide phi1 mutants are defective in MGP1. A T-DNA was inserted in the 3'-UTR of MGP1 (AT2G21870) in phi1 mutants (A), which resulted in a reduction in the expression of MGP1 in all growth conditions (+Pi, –Pi and +Phi) (B). The arrows in (A) indicate the positions of primers used for qRT–PCR. A negative correlation between the expression level of MGP1 and anthocyanin accumulation in plants treated under +Phi for 6 d (C). The data were obtained from the WT, phi1, RNAi and genomic complementation (GC) whole seedlings. The RNAi (RNAi#26) and GC (GC#11) lines used for the subsequent experiments are indicated. r represents Pearson’s correlation coefficient. Letters represent significant differences among genotypes and growth conditions, P < 0.05 (ANOVA), and error bars represent SEs of three biological replicates. Fig. 2 View largeDownload slide phi1 mutants are defective in MGP1. A T-DNA was inserted in the 3'-UTR of MGP1 (AT2G21870) in phi1 mutants (A), which resulted in a reduction in the expression of MGP1 in all growth conditions (+Pi, –Pi and +Phi) (B). The arrows in (A) indicate the positions of primers used for qRT–PCR. A negative correlation between the expression level of MGP1 and anthocyanin accumulation in plants treated under +Phi for 6 d (C). The data were obtained from the WT, phi1, RNAi and genomic complementation (GC) whole seedlings. The RNAi (RNAi#26) and GC (GC#11) lines used for the subsequent experiments are indicated. r represents Pearson’s correlation coefficient. Letters represent significant differences among genotypes and growth conditions, P < 0.05 (ANOVA), and error bars represent SEs of three biological replicates. AT2G21870, previously named MGP1 (MALE GAMETOPHYTE DEFECTIVE1), is essential for pollen development (Li et al. 2010). Loss of function of MGP1 led to death of pollen grains, and no homozygous mutant plants could be produced in the progeny of the heterozygous mgp1–/+ mutant. Anthocyanin accumulation in phi1 mutants under Phi treatment can be restored to the WT level by introducing the genomic sequence of MGP1 (Fig. 2C), suggesting that MGP1 is responsible for the phi1 phenotype. Because no additional mutants are available, we generated MGP1-RNAi (RNA interference) lines for further confirmation. As shown in Fig. 2C, the RNAi lines with reduced MGP1 expression accumulated a higher amount of anthocyanin than WT plants under +Phi. The expression of MGP1 was negatively correlated with anthocyanin accumulation under Phi treatment (Pearson’s correlation coefficient r = –0.83) (Fig. 2C), suggesting that MGP1 is involved in Phi-mediated suppression of anthocyanin accumulation. To segregate potential additional T-DNA insertion loci, we backcrossed the phi1 mutant to WT plants. All F1 progeny showed the WT phenotype of no anthocyanin accumulation under +Phi treatment, suggesting the recessive nature of the phi1 mutation. Mutant lines which retained anthocyanin under +Phi and had low expression of MGP1 from the F2 population (Supplementary Fig. S5) were analyzed in subsequent experiments, together with one RNAi (RNAi#26) line and one genomic complementation (GC) line, GC#11. The phi1 mutant is impaired in mitochondrial ATP synthesis Because the phi1 mutant is impaired in the gene encoding a subunit of mitochondrial ATP synthase, it is likely that ATP metabolism is compromised. We thus measured the ATP level. In the roots, phi1 had a reduced amount of ATP under +Pi, approximately 60% of that of the WT (Fig. 3A). The ATP level was greatly decreased after 6 d treatment with –Pi or +Phi, but no differences were observed between the WT and phi1 (Fig. 3A). Surprisingly, the shoot ATP level remained relatively unchanged in all growth conditions and phi1 showed no significant differences from the WT regardless of treatments (Fig. 3B). When grown in medium containing 5 µM oligomycin which specifically inhibits the activity of mitochondrial ATP synthase (Pagliarani et al. 2013), phi1 mutants were more sensitive to oligomycin than WT plants, supporting the role of MGP1 in mitochondrial ATP metabolism (Fig. 3C). Because mitochondrial ATP synthesis is coupled with the proton motive force, impairment of ATP synthesis results in accumulation of protons at the intermembrane space, leading to a higher mitochondrial membrane potential. We thus examined whether the mitochondrial membrane potential is altered in phi1 mutants. Tetramethylrhodamine ethyl ester (TMRE) is a fluorescent dye used to measure the membrane potential of mitochondria within a cell (Ehrenberg et al. 1988). By applying TMRE to the roots grown under +Pi, we found that the phi1 root hairs displayed a stronger signal of fluorescence than those of the WT (Fig. 3D), which indicates an increased mitochondrial membrane potential. As a control, addition of oligomycin markedly increased the signal in both WT and phi1 plants (Fig. 3D). It is conceivable that the increased mitochondrial membrane potential in phi1 mutants is a result of diminished proton pumping caused by the impairment of mitochondrial ATP synthesis. Fig. 3 View largeDownload slide phi1 mutants are impaired in ATP synthesis and sensitive to oligomycin. The relative ATP level in 12-day-old WT and phi1 roots (A) and shoots (B) after 6 d +Pi, –Pi or +Phi treatment. Letters represent significant differences compared with the WT, P < 0.05 (ANOVA). Growth phenotypes of the WT and phi1 after oligomycin treatment (C). Examination of mitochondrial membrane potential by TMRE fluorescent dye in the root hairs of the WT and phi1 with (+) or without (−) oligomycin treatment (D). The enhanced signal of red dots represents mitochondria with increased membrane potential. The microscopic fluorescent intensity was adjusted to a gain (master) value of 750 and 660 for control and oligomycin treatment, respectively. The plants in (C) and (D) were grown under +Pi conditions. Fig. 3 View largeDownload slide phi1 mutants are impaired in ATP synthesis and sensitive to oligomycin. The relative ATP level in 12-day-old WT and phi1 roots (A) and shoots (B) after 6 d +Pi, –Pi or +Phi treatment. Letters represent significant differences compared with the WT, P < 0.05 (ANOVA). Growth phenotypes of the WT and phi1 after oligomycin treatment (C). Examination of mitochondrial membrane potential by TMRE fluorescent dye in the root hairs of the WT and phi1 with (+) or without (−) oligomycin treatment (D). The enhanced signal of red dots represents mitochondria with increased membrane potential. The microscopic fluorescent intensity was adjusted to a gain (master) value of 750 and 660 for control and oligomycin treatment, respectively. The plants in (C) and (D) were grown under +Pi conditions. Accumulation of Pi and Phi in phi1 mutants We next examined whether Pi or Phi content is altered in phi1 mutants. Compared with WT and GC plants, there were no striking differences in shoot Pi content except a small reduction in the phi1 mutant grown under +Pi conditions and in the RNAi line grown under –Pi and +Phi conditions (Fig. 4A). Nevertheless, both phi1 and RNAi plants showed a higher root Pi content than the WT and GC plants under both +Pi and +Phi conditions (Fig. 4B). Fig. 4 View largeDownload slide Pi and Phi content in phi1 mutants. The Pi content in the shoot (A) and root (B) of the WT, phi1, RNAi and genomic complementation (GC) plants grown under +Pi, –Pi and +Phi conditions. The Phi content in the shoot and root of plants grown under +Phi conditions (C). Letters represent significant differences among genotypes and growth conditions, P < 0.05 (ANOVA). Error bars are SEs from 5–9 biological replicates. Fig. 4 View largeDownload slide Pi and Phi content in phi1 mutants. The Pi content in the shoot (A) and root (B) of the WT, phi1, RNAi and genomic complementation (GC) plants grown under +Pi, –Pi and +Phi conditions. The Phi content in the shoot and root of plants grown under +Phi conditions (C). Letters represent significant differences among genotypes and growth conditions, P < 0.05 (ANOVA). Error bars are SEs from 5–9 biological replicates. When Phi was measured, while all the root samples had a comparable level, the shoot of phi1 but not of the RNAi line had a reduced level, about 60% of that of the WT (Fig. 4C). This argues against the possibility of anthocyanin accumulation in Phi-treated phi1 and RNAi lines as a consequence of reduced Phi accumulation. To examine further if this was the case, pho1 mutants impaired in root to shoot translocation of Pi (Poirier et al. 1991), and probably also Phi, were inspected. Under +Phi conditions, pho1 mutants showed diminished anthocyanin similar to WT plants (Supplementary Fig. S6A, B) but accumulated a level of shoot Phi comparable with phi1 (Supplementary Fig. S6C). This suggests that low Phi accumulation in phi1 shoots is unlikely to be the reason for anthocyanin accumulation, and additional factors are involved. High sugar in phi1 mutants antagonizes Phi-mediated suppression of anthocyanin and starch accumulation Since phi1 showed impaired mitochondrial ATP synthesis, the cellular metabolic processes are probably affected. It has been documented that sugars positively regulate anthocyanin accumulation (Solfanelli et al. 2006), and both sugars and anthocyanin are increased upon Pi starvation (Ciereszko and Barbachowska 2000, Pant et al. 2015); we therefore quantified the amount of sugar. As expected, the levels of sucrose, fructose and glucose in WT and phi1 shoots were increased under –Pi (Fig. 5A). However, it is surprising that their level under +Phi remained high or even higher than in –Pi in all genotypes. Of note, phi1 shoots accumulated about 1.5-fold the WT level of sucrose and glucose under +Pi and +Phi, and at least 2-fold the WT level of fructose under all growth conditions (Fig. 5A). Similar to phi1 mutants, the RNAi line also had a higher level of sugars (except for glucose) than the WT under +Phi conditions. The accumulation of sugars was restored to the WT level in the GC lines (Fig. 5A), indicating that the reduced expression of MGP1 leads to overaccumulation of sugar in shoots. Fig. 5 View largeDownload slide Effect of sugar on the Phi-mediated suppression of anthocyanin and starch accumulation. Relative amount of sucrose, fructose and glucose in the shoot of the WT, phi1, RNAi and genomic complementation (GC) plants (A). Anthocyanin content in WT plants treated with various combinations of sucrose and Pi or Phi as indicated (B). Anthocyanin accumulation in the shoot of WT, suc2-5 (C) and hps1 (D) grown under +Pi, –Pi and +Phi conditions. Staining of starch by Lugol’s iodine is shown in dark blue (E). Double and single asterisks represent significant difference with P < 0.01 and P < 0.05 (Student’s t-test), respectively, in the comparison with the WT. Letters represent significant differences among genotypes and growth conditions, P < 0.05 (ANOVA). Error bars represent SEs. Scale bars = 1 cm. Fig. 5 View largeDownload slide Effect of sugar on the Phi-mediated suppression of anthocyanin and starch accumulation. Relative amount of sucrose, fructose and glucose in the shoot of the WT, phi1, RNAi and genomic complementation (GC) plants (A). Anthocyanin content in WT plants treated with various combinations of sucrose and Pi or Phi as indicated (B). Anthocyanin accumulation in the shoot of WT, suc2-5 (C) and hps1 (D) grown under +Pi, –Pi and +Phi conditions. Staining of starch by Lugol’s iodine is shown in dark blue (E). Double and single asterisks represent significant difference with P < 0.01 and P < 0.05 (Student’s t-test), respectively, in the comparison with the WT. Letters represent significant differences among genotypes and growth conditions, P < 0.05 (ANOVA). Error bars represent SEs. Scale bars = 1 cm. We speculated that overaccumulation of sugar in phi1 mutants may over-ride Phi-mediated suppression of anthocyanin accumulation. We increased the sucrose supplement in the +Phi media from 1% to 5% or 10% and found a 2- to 6-fold increase of anthocyanin in WT plants, which backed up our supposition (Fig. 5B). Furthermore, two mutants that accumulate a high level of endogenous sucrose in shoots, suc2 and hypersensitive to phosphate starvation1 (hps1) (Lloyd and Zakhleniuk 2004, Lei et al. 2011), also accumulated a higher level of anthocyanin than the WT plants when grown in the +Phi medium (Fig. 5C, D). These results indicate that sugars antagonize Phi-mediated suppression of anthocyanin accumulation. Therefore, the enhanced anthocyanin accumulation under Phi treatment in phi1 and RNAi plants is probably attributed to the increased level of sugars. In addition to sugars, Pi starvation enhances starch accumulation because the activity of a key starch biosynthesis enzyme, ADP-glucose pyrophosphorylase (AGPase), is allosterically inhibited by Pi, while Pi depletion and a high level of sugars up-regulate its activity (Geigenberger 2011). We therefore examined the effect of Phi on Pi starvation-induced starch accumulation. As revealed by Lugol’s iodine staining, Phi treatment suppressed Pi starvation-induced starch accumulation in WT and GC lines, but not in phi1 and RNAi lines (Fig. 5E), indicating that MGP1 is involved in Phi-mediated suppression of starch accumulation. A subset of carbohydrate metabolism genes is misexpressed in phi1 in response to Phi Because phi1 mutants accumulated excessive sugar and starch, we next examined the expression of a subset of PSR genes involved in sugar and starch metabolism. These genes were selected according to a preliminary microarray analysis which showed prominent differences in expression between the WT and phi1 under +Phi. qRT–PCR analyses were carried out to examine the gene expression in WT, GC, RNAi and phi1 seedlings grown under +Pi, –Pi and +Phi conditions. As illustrated by the heat map, up-regulation of several genes (e.g. CHS, ANS and DFR) involved in anthocyanin biosynthesis under –Pi was retained under +Phi in phi1 and RNAi lines, in contrast to being suppressed in WT and GC plants (Fig. 6A, Group I; see original data in Supplementary Table S1). A similar situation was also observed for a starch biosynthetic gene encoding the large subunit of ADP-glucose pyrophosphorylase (APL3), PHOSPHOGLUCAN WATER DIKINASE (PWD) involved in starch degradation and GLYCOSYL HYDROLASE 9B8 (GH9B8) (Fig. 6A, Group I), reflecting the disturbance of starch metabolism in phi1 and RNAi plants under +Phi. The increased expression of PWD is likely to be the result of a feedback regulation to avoid overaccumulation of starch because it was up-regulated in all Pi-starved samples. The expression of two selected sugar-related PSi genes, 2,3-bisphosphoglycerate-independent PHOSPHOGLYCERATE MUTASE 2 (iPGAM2) and SUCROSE SYNTHASE 3 (SUS3) was significantly enhanced in phi1 and RNAi plants under +Phi (Fig. 6A, Group II). In contrast to the other PSi genes suppressed by Phi, iPGAM2 and SUS3 transcripts are further up-regulated in response to Phi in all samples, correlated with the enhanced sugar accumulation (Fig. 5A). The activity of SUS3 and APL3 in starch and sucrose metabolism is indicated in Supplementary Fig. S7A. Fig. 6 View largeDownload slide Differential expression of PSR genes in phi1 mutants responsive to Phi. The heat maps show the relative expression of a subset of PSR genes involved in anthocyanin biosynthesis (highlighted with purple), and starch (brown) or sugar (green) metabolism (A), and several PHR1-regulated PSi genes (B) in the WT, phi1, RNAi and genomic complementation (GC) plants grown under +Pi, –Pi and +Phi conditions. The expression level is presented as the Z-score. In (A), the genes are classified into the groups illustrated by the line graphs at the side. The original data of qRT–PCR are given in Supplementary Table S1. Fig. 6 View largeDownload slide Differential expression of PSR genes in phi1 mutants responsive to Phi. The heat maps show the relative expression of a subset of PSR genes involved in anthocyanin biosynthesis (highlighted with purple), and starch (brown) or sugar (green) metabolism (A), and several PHR1-regulated PSi genes (B) in the WT, phi1, RNAi and genomic complementation (GC) plants grown under +Pi, –Pi and +Phi conditions. The expression level is presented as the Z-score. In (A), the genes are classified into the groups illustrated by the line graphs at the side. The original data of qRT–PCR are given in Supplementary Table S1. We also investigated the expression of several sugar-related PSr genes. The repression of these PSr genes by Phi was relieved in WT plants but remained in phi1 and RNAi lines (Fig. 6A, Group III). The weak response to Phi-mediated derepression of these PSr genes in phi1 mutants could be explained by an increased level of sugars, as these sugar-related genes are induced under low carbon sources, such as DARK INDUCIBLE 10/RAFFINOSE SYNTHASE 6 (DIN10/RS6) (Fujiki et al. 2001), AKINBETA1 (Li et al. 2009), MYO-INOSITOL OXYGENASE 2 (MIOX2) (Alford et al. 2012) and SUGAR TRANSPORTER 1 (STP1) (Cordoba et al. 2015). According to the metabolic pathway in the Kyoto Encyclopedia of Genes and Genomes (KEGG) (http://www.genome.jp/kegg-bin/show_organism?org=ath), DIN10/RS6 and SEED IMBIBITION 2/RAFFINOSE SYNTHASE 2 (AtSIP2/RS2) are assigned to galactose metabolism (Supplementary Fig. S7B), while FRUCTOSE-BISPHOSPHATE ALDOLASE 5 (FBA5) and FRUCTOSE-1,6-BISPHOSPHATASE (FBP) are involved in biosynthesis of fructose 1,6-biphosphate and fructose 6-phosphate, respectively, in the glycolysis/gluconeogenesis pathway (Supplementary Fig. S7C). These results suggest that impaired mitochondrial ATP synthase counteracts Phi-mediated derepression of PSr genes involved in carbohydrate metabolism. To examine whether PHR1-mediated regulation of PSi genes was perturbed in phi1 mutants, we analyzed the expression of several typical PHR1-regulated PSi genes, such as INDUCED BY PHOSPHATE STARVATION1 (IPS1), At4, SPX1, SPX2, SPX3, PHT1;4, PHT1;8, MONOGALACTOSYL-DIACYLGLYCEROL SYNTHASE 2 (MGD2) and SULFOQUINOVOSYL-DIACYLGLYCEROL 1 (SQD1). We found that the expression of these genes was suppressed by Phi to a similar extent in all the lines (Fig. 6B), suggesting that impaired mitochondrial ATP synthase activity does not have a major impact on PHR1 regulons. Discussion To investigate the Pi-mediated regulatory pathway of PSRs, we employed Phi in a genetic screen and identified an insensitive mutant (phi1) which retained anthocyanin in the presence of Phi. phi1 mutants are impaired in the MGP1 gene involved in mitochondrial ATP synthesis. Further analyses suggested that the excessive accumulation of sugars in phi1 mutants may account for the retention of anthocyanin. This phenotype was not caused by sucrose supplementation in the growth medium because phi1 still accumulated a high level of anthocyanin when grown on sucrose-free medium (Supplementary Fig. S5C). In addition to the anthocyanin biosynthetic genes, several carbohydrate metabolism-related PSR genes are misregulated in phi1 mutants, particularly under +Phi conditions. This study not only highlights the importance of sugars in the Pi regulatory pathway but also delineates the role of MGP1 in mitochondrial ATP synthesis. It should be noted that anthocyanin is also induced by other stresses. Improvement in screening using a reporter driven by the promoter of a Phi-suppressed PSi gene is needed in the future. Role of MGP1 in mitochondrial ATP synthesis The MGP1 gene encodes a plant-specific FAd subunit of mitochondrial F1-Fo ATP synthase which was initially identified in spinach and soybean (Hamasur and Glaser 1992, Smith et al. 1994). The FAd subunit is thought to be a third component of eukaryotic ATP synthase, an associated component likely to be involved in linking the F1 and Fo complexes (Nagley 1988, Smith et al. 1994). Based on a search in the protein database of the National Center for Biotechnology Information (NCBI) (https://blast.ncbi.nlm.nih.gov/Blast.cgi) and EnsemblPlants (http://plants.ensembl.org/index.html), we found that homologs of this subunit are present extensively in angiosperms and can been identified in a gymnosperm, Picea sitchensis (Sitka spruce), and three ancient plants, Physcomitrella patens (moss), Selaginella moellendorffii (spikemoss) and Marchantia polymorpha (common liverwort). No homologs were identified in the animal kingdom. We postulate that MGP1 emerged during land-plant evolution. Phylogenetic analysis of 41 MGP1 homologs in angiosperms showed that the homologs in monocotyledons and eudicotyledons were categorized distinctly (Supplementary Fig. S8), suggesting that they diverged before the speciation of monocotyledons and eudicotyledons. Li et al. (2010) proposed that MGP1 acts as an inhibitor of ATP hydrolysis during the dehydration stage of pollen development to restore disrupted mitochondrial membrane potential. Loss of function of this gene resulted in a pollen defect and produced no homozygous seeds (Li et al. 2010), pointing to an essential role for MGP1 in plant survival and the success of land-plant evolution. The phi1 mutation we identified in the present study is caused by reduced expression of MGP1, and homozygous seeds are available for characterization. Several lines of evidence from our characterization of phi1 mutants suggest that MGP1 has a crucial role in the synthesis of mitochondrial ATP. First, phi1 mutants were hypersensitive to oligomycin (Fig. 3C), a specific inhibitor of mitochondrial ATP synthase. Secondly, the mitochondrial membrane potential in phi1 mutants was elevated very probably due to impaired mitochondrial ATP synthesis (Fig. 3D). Thirdly, the ATP content in phi1 roots was reduced under +Pi even though it remained unchanged in shoots (Fig. 3A, B). The heterogeneity in the abundance of mitochondrial proteins between photosynthetic and non-photosynthetic tissues (Lee et al. 2008) may partly explain our observation about the discrepancy between roots and shoots. Moreover, because chloroplasts can contribute approximately three times more ATP synthesis than mitochondria under light conditions (Cheung et al. 2014), it is possible that the ATP generated from chloroplasts may compensate the defect of ATP synthesis in the mitochondria of phi1 shoots. However, we did not detect the differences between WT and phi1 shoots after 1 d dark treatment even though the ATP level dropped to 50% of that of the light controls (Supplementary Fig. S9). Although the basis for this is currently unknown, we speculate that reduction of mitochondrial respiration during darkness (Lee et al. 2010) may diminish the difference between the WT and phi1. Our observations also suggest a complex interaction between chloroplasts and mitochondria in ATP metabolism. Impaired mitochondrial ATP synthesis caused abnormal sugar accumulation As Pi is an essential substrate for ATP synthesis, it is intuitive to expect a reduction of ATP under Pi starvation. Indeed, several early studies reported the reduction of ATP in Pi-starved plants (Duff et al. 1989, Rao and Terry 1995, Mikulska et al. 1998, Morcuende et al. 2007). However, a recent analysis of primary metabolites showed an increasing level of ATP in 3 d and 6 d Pi-starved Arabidopsis and barley leaves, respectively (Alexova et al. 2017, Kuo et al., 2018). This discrepancy could be explained by differences in plant species, tissue types or treatment conditions. Nevertheless, conservation of Pi and ATP consumption is an important adaptive response to Pi starvation, which is achieved by several metabolic bypass enzymes involved in glycolysis, the tricarboxylic acid (TCA) cycle and the mitochondrial electron transfer chain (Plaxton and Tran 2011). A number of studies have shown that defects in mitochondrial components involved in the respiratory pathway affected photosynthetic activity and carbon assimilation (Carrari et al. 2003, Nunes-Nesi et al. 2005). Likewise, our study on phi1 mutants showed that reduced mitochondrial ATP synthesis could perturb carbon metabolism, leading to increased sugar accumulation. Further quantitative analysis of metabolites in phi1 mutants, especially those in the TCA cycle, may elucidate its high sugar phenotype and offer additional insights into the interaction between respiratory flux and carbon assimilation. Is Pi a signal? The direct evidence for Pi being a signal molecule comes from the prevalent suppression of PSRs by Phi (Ticconi et al. 2001, Varadarajan et al. 2002, Jost et al. 2015). The dependence of the interaction between SPX and PHR proteins on Pi or Phi (Puga et al. 2014, Wang et al. 2014) not only explains how Pi inhibits the transcriptional activity of PHR but also provides further support for the role of Pi as a signal molecule to regulate the expression of PSR genes negatively. However, a recent structural–functional analysis revealed that inositol polyphosphates (InsPs) or inositol pyrophosphates (PP-InsPs) act as better ligands than Pi for the interaction between rice OsSPX4 and OsPHR2 (Wild et al. 2016), which argues against Pi being a signal in regulating PHR1 activity. Although InsPs or PP-InsPs had a much higher affinity than Pi in in vitro binding analysis (Wild et al. 2016), it has to be considered that the in vivo cellular concentration of InsPs or PP-InsPs is much lower than that of Pi. Several PHR1 regulons are suppressed in Pi-starved plants upon Phi addition (Jost et al. 2015), supporting the role of Pi in suppression of PHR1 activity. However, we could not rule out the possibility that addition of Phi facilitates the release of Pi from other cellular fractions, which is used to generate InsPs, or InsPs are restricted to certain cellular compartments during Pi starvation and released upon Pi or Phi replenishment. In spite of technical challenges, simultaneous quantification of the cellular concentration of Pi, InsP and PP-InsP and their change in response to different external Pi supply should provide further insights (kuo et al., 2018). As InsP or PP-InsP biosynthesis requires ATP, and phi1 had a reduced ATP level, it would be interesting to examine the amount of InsP or PP-InsP in phi1. However, Phi-mediated suppression of the PHR1 regulon is not altered by phi1 mutation (Fig. 6B), suggesting that misregulation of the PSR genes in phi1 mutants is PHR1 independent. How the levels of Pi, InsP or PP-InsP, and ATP are co-ordinated to exert PSRs will be a focus of future study. Cross-talk between sugar metabolism and PSRs The importance of sugars in regulating the PSR has been well documented. Results from alteration of the endogenous sugar level through controlling the phloem loading of sucrose or exogenous application of sugars showed that sugars could activate the expression of a number of PSi genes, or, in other words, sugars antagonize the effect of Pi on the suppression of PSi genes (Liu et al. 2005, Müller et al. 2007, Dasgupta et al. 2014). In accordance with the positive effect of sugar on PSRs, Pi starvation leads to an elevated level of sugars, starch and anthocyanin, but reduced abundance of hexose phosphates, and their amounts are restored upon Pi replenishment (Pant et al. 2015). The enhancement of starch and anthocyanin under Pi starvation could be a consequence of the build up of sugar. Our study showed that phi1 mutants overaccumulate sugar concomitantly with the development of starch and anthocyanin under +Phi conditions, indicating that sugars antagonize the sensitivity of phi1 mutants to Phi-mediated suppression of anthocyanin and starch accumulation. Unlike WT plants, the expression of several sugar-related PSr genes in phi1 remained suppressed even in the presence of Phi (Fig. 6A;Supplementary Table S1), suggesting alteration of carbon metabolism in Phi-treated phi1 plants. Several of these PSr genes, such as DIN10/RS6, AKINBETA1 and MIOX2, have been reported to be suppressed by a high level of sugars, which supports the notion of the effect of sugars on these genes (Fujiki et al. 2001, Li et al. 2009, Alford et al. 2012). These results underline the close interaction between carbon metabolism and regulation of PSRs. Based on the observation of an antagonistic relationship between sugar and Pi/Phi in regulating PSRs, we hypothesize that the extent of PSRs may be modulated by the ratio of sugar to Pi/Phi. Similarly, Dasgupta et al. (2014) also speculated that plants may perceive a disruption of C to P homeostasis to trigger the PSRs. In this scenario, a high sugar to Pi ratio (e.g. Pi deficiency) leads to activation of PSRs, whereas a low sugar to Pi ratio (e.g. Pi sufficiency) results in suppression of PSRs. In the presence of Phi, the decrease of the sugar to Pi/Phi ratio would favor the attenuation of PSRs. Retention of PSRs in phi1 mutants under +Phi conditions could be explained by an elevated ratio of sugar to Pi/Phi due to overaccumulation of sugar (Figs 4C, 5A). Nevertheless, additional regulatory factors, such as hormones (Rubio et al. 2009) or cell cycle activity (Lai et al. 2007), must be involved because the change in the sugar to Pi ratio cannot explain all the PSRs. Future investigation to test this hypothesis will be needed. Materials and Methods Plant materials and growth conditions Activation tagging mutants of Arabidopsis thaliana (stock CS31100) were purchased from the Arabidopsis Biological Resource Center. The accession Columbia-7 (Col-7) was used as a WT control for the phi1 mutant. In the case of pho1, suc2-5 (SALK_087046) and hps1 mutants, Col-0 was used as the WT. Seeds were surface sterilized and germinated on agar medium of half-strength modified Hoagland’s nutrient solution containing 250 μM KH2PO4, 1% sucrose and 0.8% Bactoagar, with pH adjusted to 5.7 (Aung et al. 2006). hps1 mutants were germinated in medium without sucrose supplementation. After germination, 6- to 8-day-old seedlings were transferred to the media with different P sources for 6 d unless specified, designated as +Pi, –Pi and +Phi. The +Pi and +Phi media contained 500 μM KH2PO4 and 500 μM phosphorous acid (H2PO3, Sigma), respectively, and no P source was supplied in the –Pi medium. For oligomycin treatment, 6-day-old seedlings were transferred to the +Pi, –Pi and +Phi media containing 5 μM oligomycin and grown for an additional 9 d. For exogenous sucrose treatment, sucrose was supplied to +Phi medium at three concentrations, 1% (∼30 mM), 5% (∼150 mM) or 10% (∼300 mM). Plants were grown at 22°C under a cycle of 16 h light (white fluorescent light at 100–150 μE m–2 s–1) and 8 h dark. Quantification of Pi, Phi and anthocyanin Pi content was determined as described (Chiou et al. 2006). Phi was measured by nuclear magnetic resonance (NMR). A sample mixture (400 μl) comprising 300 μl of homogenate (in 1% acetic acid), 80 μl of 10 mM EDTA and 20 μl of deuterium water was dispensed into NMR tubes. The NMR instrument was operated using Bruker Avance 500 AV and analyzed by TopSpin software (Bruker) at the High Field Nuclear Magnetic Resonance Center (HFNMRC) in Academia Sinica. Quantification was made based on a standard curve plotted by a range of known concentrations of H2PO3. Anthocyanin content was measured using a modified protocol (Lange et al. 1971). Shoot tissues were weighed and homogenized in 1 ml of extraction buffer (propanol:HCl:H2O in a ratio of 18:1:81). The sample tubes were immersed in boiled water for 1.5 min and then cooled immediately on ice for at least 3 min before a 10 min centrifugation. Supernatants were collected and measured at 535 and 650 nm. Anthocyanin content was calculated based on the formula Abs(535 nm – 2×650 nm)/FW. Generation of RNAi and complementation lines A 336 bp DNA fragment located within the coding region of MGP1 (AT2G21870) and an approximately 5.8 kb genomic DNA fragment of MGP1 including a 2.1 kb promoter and 1.2 kb downstream of the 3'-UTR were cloned into pCR8/GW/TOPO entry vectors (Invitrogen) for generating RNAi and complementation lines, respectively. The DNA fragments were subcloned into the Gateway destination vectors pB7GWIWG2(I) (Karimi et al. 2002) or pMDC99 (Curtis and Grossniklaus 2003) via LR Clonase enzyme mix (Invitrogen). The constructs were transformed into Agrobacterium tumefaciens GV3101 followed by the transformation into WT or phi1 plants using the floral dip method (Clough and Bent 1998). The putative transgenic RNAi and complementary plants were selected in media containing 100 μg ml–1 BASTA and hygromycin, respectively. Measurement of sugar and ATP For the measurement of sugar, metabolites were extracted from 100 mg of tissues in 1 ml of 80% methanol (containing 12 μg ml–1 ribitol) by sonication at 4°C for 30 min. Ribitol was used as an external control for normalization. After centrifugation, the supernatant was vacuum-dried and subjected to gas chromatography–mass spectrometry (GC-MS) analyses. The amount of the three main sugar species (sucrose, glucose and fructose) was quantified in relation to that of the WT grown under +Pi conditions. ATP was measured according to the protocol of Cho et al. (2016). Tissues were homogenized in 2.3% trichloroacetic acid containing ribitol (20 μg ml–1) as an external control. After centrifugation, the supernatant was neutralized to pH 6.5–7 with KOH and subjected to analysis using a Triple Quadrupole (TSQ) mass spectrometer. Observation of mesophyll cell size Cotyledons and true leaves of 12-day-old seedlings grown under +Phi for 6 d were fixed in 70% ethanol and cleared using chloral hydrate solution (200 g of chloral hydrate, 20 g of glycerol and 50 ml of dH2O) as described (Tsuge et al. 1996). The samples were observed under an upright microscope (Zeiss Imager, Z1) using differential interference contrast. Mito-tracker analysis Nine-day-old seedlings were submerged in Hoagland’s medium supplied with 20 nM TMRE fluorescent dye (Molecular Probes) for 30 min in the dark. The staining in root hairs was observed using an inverted confocal microscope (Zeiss LSM 510 Meta NLO DuoScan) with an objective C-Apochromat ×40/1.2 W, and excitation at 561 nm and emission 610 nm. Starch staining assay Lugol’s solution was prepared by mixing potassium iodide and iodine in a ratio of 2:1. Shoots of 12-day-old seedlings were immersed in ethanol to remove Chl, followed by the addition of Lugol’s solution. Analysis by qRT–PCR Twelve-day-old seedlings were ground and homogenized in liquid nitrogen, and RNA was isolated using RNAzol RT reagent (Molecular Research Center). cDNA was synthesized using oligo(dT) and Moloney murine leukemia virus reverse transcriptase (Invitrogen) from RNA. The quality of cDNA was inspected by the integrity of ACTIN2 (ACT2) RNA. The gene expression level was determined using Power SYBR Green PCR Master Mix kit in the Real Time PCR system (185-5200 CFX Connect, Bio-Rad), following the manufacturer’s instructions. The expression level was calculated by normalization to that of a housekeeping gene, UBIQUITIN 10 (UBQ10). The relative expression was converted into a Z-score and is presented as a heat map. The original data are shown in Supplementary Table S1. The sequences of primers used in this study are listed in Supplementary Table S2. Statistical analysis Statistical analyses were carried out using ANOVA of CoStat version 6.4 (CoHort Software) or the simple Student t-test of Microsoft Excel. Supplementary Data Supplementary data are available at PCP online. Funding This work was supported by the Ministry of Science and Technology of the Republic of China [grant No. NSC94-2311-B-001-057] and Academia Sinica, Taiwan. Acknowledgments We thank the ABRC Metabolomics Core Laboratory for the measurement of sugars and ATP, HFNMRC at Academia Sinica for quantification of Phi, and the Confocal Microscopic Core Facility at Academia Sinica for microscopy images. We are grateful to Kyaw Aung for helping with the TAIL-PCR cloning and Der-Fen Suen (ABRC, Academia Sinica, Taiwan) for helping with the setup of Mito-Tracker analysis. hps1 mutants were kindly provided by Dong Liu (Tsinghua University, China). We also appreciate Der-Fen Suen, Teng-Kuei Huang and Hui-Fen Kuo for constructive discussion and comments on the manuscript. Disclosures The authors have no conflicts of interest to declare. References Alexova R. , Nelson C.J. , Millar A.H. ( 2017 ) Temporal development of the barley leaf metabolic response to Pi limitation . Plant Cell Environ . 40 : 645 – 657 . Google Scholar CrossRef Search ADS PubMed Alford S.R. , Rangarajan P. , Williams P. , Gillaspy G.E. ( 2012 ) myo-Inositol oxygenase is required for responses to low energy conditions in Arabidopsis thaliana . Front. Plant Sci. 3 : 69 . Google Scholar CrossRef Search ADS PubMed Aung K. , Lin S.I. , Wu C.C. , Huang Y.T. , Su C.L. , Chiou T.J. ( 2006 ) pho2, a phosphate overaccumulator, is caused by a nonsense mutation in a microRNA399 target gene . Plant Physiol . 141 : 1000 – 1011 . Google Scholar CrossRef Search ADS PubMed Berkowitz O. , Jost R. , Kollehn D.O. , Fenske R. , Finnegan P.M. , O’Brien P.A. , et al. . ( 2013 ) Acclimation responses of Arabidopsis thaliana to sustained phosphite treatments . J. Exp. Bot . 64 : 1731 – 1743 . Google Scholar CrossRef Search ADS PubMed Bustos R. , Castrillo G. , Linhares F. , Puga M.I. , Rubio V. , Perez-Perez J. , et al. . ( 2010 ) A central regulatory system largely controls transcriptional activation and repression responses to phosphate starvation in Arabidopsis . PLoS Genet. 6 : e1001102 . Google Scholar CrossRef Search ADS PubMed Carrari F. , Nunes-Nesi A. , Gibon Y. , Lytovchenko A. , Loureiro M.E. , Fernie A.R. ( 2003 ) Reduced expression of aconitase results in an enhanced rate of photosynthesis and marked shifts in carbon partitioning in illuminated leaves of wild species tomato . Plant Physiol . 133 : 1322 – 1335 . Google Scholar CrossRef Search ADS PubMed Cheung C.Y.M. , Poolman M.G. , Fell D.A. , Ratcliffe R.G. , Sweetlove L.J. ( 2014 ) A diel flux balance model captures interactions between light and dark metabolism during day–night cycles in C3 and Crassulacean acid metabolism leaves . Plant Physiol . 165 : 917 . Google Scholar CrossRef Search ADS PubMed Chiou T.J. , Aung K. , Lin S.I. , Wu C.C. , Chiang S.F. , Su C.L. ( 2006 ) Regulation of phosphate homeostasis by microRNA in Arabidopsis . Plant Cell 18 : 412 – 421 . Google Scholar CrossRef Search ADS PubMed Chiou T.J. , Lin S.I. ( 2011 ) Signaling network in sensing phosphate availability in plants . Annu. Rev. Plant Biol. 62 : 185 – 206 . Google Scholar CrossRef Search ADS PubMed Cho H.Y. , Wen T.N. , Wang Y.T. , Shih M.C. ( 2016 ) Quantitative phosphoproteomics of protein kinase SnRK1 regulated protein phosphorylation in Arabidopsis under submergence . J. Exp. Bot. 67 : 2745 – 2760 . Google Scholar CrossRef Search ADS PubMed Ciereszko I. , Barbachowska A. ( 2000 ) Sucrose metabolism in leaves and roots of bean (Phaseolus vulgaris L.) during phosphate deficiency . J. Plant Physiol . 156 : 640 – 644 . Google Scholar CrossRef Search ADS Clough S.J. , Bent A.F. ( 1998 ) Floral dip: a simplified method for Agrobacterium-mediated transformation of Arabidopsis thaliana . Plant J . 16 : 735 – 743 . Google Scholar CrossRef Search ADS PubMed Cordoba E. , Aceves-Zamudio D.L. , Hernandez-Bernal A.F. , Ramos-Vega M. , Leon P. ( 2015 ) Sugar regulation of SUGAR TRANSPORTER PROTEIN 1 (STP1) expression in Arabidopsis thaliana . J. Exp. Bot . 66 : 147 – 159 . Google Scholar CrossRef Search ADS PubMed Curtis M.D. , Grossniklaus U. ( 2003 ) A gateway cloning vector set for high-throughput functional analysis of genes in planta . Plant Physiol . 133 : 462 – 469 . Google Scholar CrossRef Search ADS PubMed Danova-Alt R. , Dijkema C. , De Waard P. , Kock M. ( 2008 ) Transport and compartmentation of phosphite in higher plant cells—kinetic and P nuclear magnetic resonance studies . Plant Cell Environ . 31 : 1510 – 1521 . Google Scholar CrossRef Search ADS PubMed Dasgupta K. , Khadilkar A.S. , Sulpice R. , Pant B. , Scheible W.R. , Fisahn J. , et al. . ( 2014 ) Expression of sucrose transporter cDNAs specifically in companion cells enhances phloem loading and long-distance transport of sucrose but leads to an inhibition of growth and the perception of a phosphate limitation . Plant Physiol . 165 : 715 – 731 . Google Scholar CrossRef Search ADS PubMed Duff S.M. , Moorhead G.B. , Lefebvre D.D. , Plaxton W.C. ( 1989 ) Phosphate starvation inducible ‘bypasses’ of adenylate and phosphate dependent glycolytic enzymes in Brassica nigra suspension cells . Plant Physiol . 90 : 1275 – 1278 . Google Scholar CrossRef Search ADS PubMed Ehrenberg B. , Montana V. , Wei M.D. , Wuskell J.P. , Loew L.M. ( 1988 ) Membrane potential can be determined in individual cells from the nernstian distribution of cationic dyes . Biophys. J . 53 : 785 – 794 . Google Scholar CrossRef Search ADS PubMed Fujiki Y. , Yoshikawa Y. , Sato T. , Inada N. , Ito M. , Nishida I. , et al. . ( 2001 ) Dark-inducible genes from Arabidopsis thaliana are associated with leaf senescence and repressed by sugars . Physiol. Plant. 111 : 345 – 352 . Google Scholar CrossRef Search ADS PubMed Geigenberger P. ( 2011 ) Regulation of starch biosynthesis in response to a fluctuating environment . Plant Physiol . 155 : 1566 – 1577 . Google Scholar CrossRef Search ADS PubMed Hamasur B. , Glaser E. ( 1992 ) Plant mitochondrial F0F1 ATP synthase. Identification of the individual subunits and properties of the purified spinach leaf mitochondrial ATP synthase . Eur. J. Biochem. 205 : 409 – 416 . Google Scholar CrossRef Search ADS PubMed Hammond J.P. , White P.J. ( 2011 ) Sugar signaling in root responses to low phosphorus availability . Plant Physiol . 156 : 1033 – 1040 . Google Scholar CrossRef Search ADS PubMed Jost R. , Pharmawati M. , Lapis-Gaza H.R. , Rossig C. , Berkowitz O. , Lambers H. , et al. . ( 2015 ) Differentiating phosphate-dependent and phosphate-independent systemic phosphate-starvation response networks in Arabidopsis thaliana through the application of phosphite . J. Exp. Bot . 66 : 2501 – 2514 . Google Scholar CrossRef Search ADS PubMed Karimi M. , Inze D. , Depicker A. ( 2002 ) GATEWAY vectors for Agrobacterium-mediated plant transformation . Trends Plant Sci . 7 : 193 – 195 . Google Scholar CrossRef Search ADS PubMed Kobayashi K. , Masuda T. , Takamiya K. , Ohta H. ( 2006 ) Membrane lipid alteration during phosphate starvation is regulated by phosphate signaling and auxin/cytokinin cross-talk . Plant J . 47 : 238 – 248 . Google Scholar CrossRef Search ADS PubMed Krapp A. , Stitt M. ( 1995 ) An evaluation of direct and indirect mechanisms for the ‘sink-regulation’ of photosynthesis in spinach: changes in gas-exchange, carbohydrates, metabolites, enzyme activities and steady-state transcript levels after cold-girdling source leaves . Planta 195 : 313 – 323 . Google Scholar CrossRef Search ADS Kuo H.F. , Hsu Y.Y. , Lin W.C. , Chen K.Y. , Munnik T. , Brearley C. , et al. . ( 2018 ) Arabidopsis inositol phosphate kinases, IPK1 and ITPK1, constitute a metabolic pathway in maintaining phosphate homeostasis. bioRxiv 270355, https://doi.org/10.1101/270355. Lai F. , Thacker J. , Li Y. , Doerner P. ( 2007 ) Cell division activity determines the magnitude of phosphate starvation responses in Arabidopsis . Plant J . 50 : 545 – 556 . Google Scholar CrossRef Search ADS PubMed Lange H. , Shropshire W. , Mohr H. ( 1971 ) An analysis of phytochrome-mediated anthocyanin synthesis . Plant Physiol . 47 : 649 – 655 . Google Scholar CrossRef Search ADS PubMed Lee C.P. , Eubel H. , Millar A.H. ( 2010 ) Diurnal changes in mitochondrial function reveal daily optimization of light and dark respiratory metabolism in Arabidopsis . Mol. Cell. Proteomics 9 : 2125 – 2139 . Google Scholar CrossRef Search ADS PubMed Lee C.P. , Eubel H. , O’Toole N. , Millar A.H. ( 2008 ) Heterogeneity of the mitochondrial proteome for photosynthetic and non-photosynthetic Arabidopsis metabolism . Mol. Cell. Proteomics 7 : 1297 – 1316 . Google Scholar CrossRef Search ADS PubMed Lei M. , Liu Y. , Zhang B. , Zhao Y. , Wang X. , Zhou Y. , et al. . ( 2011 ) Genetic and genomic evidence that sucrose is a global regulator of plant responses to phosphate starvation in Arabidopsis . Plant Physiol . 156 : 1116 – 1130 . Google Scholar CrossRef Search ADS PubMed Li L. , Foster C.M. , Gan Q. , Nettleton D. , James M.G. , Myers A.M. , et al. . ( 2009 ) Identification of the novel protein QQS as a component of the starch metabolic network in Arabidopsis leaves . Plant J . 58 : 485 – 498 . Google Scholar CrossRef Search ADS PubMed Li W.Q. , Zhang X.Q. , Xia C. , Deng Y. , Ye D. ( 2010 ) MALE GAMETOPHYTE DEFECTIVE 1, encoding the FAd subunit of mitochondrial F1F0-ATP synthase, is essential for pollen formation in Arabidopsis thaliana . Plant Cell Physiol . 51 : 923 – 935 . Google Scholar CrossRef Search ADS PubMed Lin W.Y. , Huang T.K. , Leong S.J. , Chiou T.J. ( 2014 ) Long-distance call from phosphate: systemic regulation of phosphate starvation responses . J. Exp. Bot . 65 : 1817 – 1827 . Google Scholar CrossRef Search ADS PubMed Liu J. , Samac D.A. , Bucciarelli B. , Allan D.L. , Vance C.P. ( 2005 ) Signaling of phosphorus deficiency-induced gene expression in white lupin requires sugar and phloem transport . Plant J . 41 : 257 – 268 . Google Scholar CrossRef Search ADS PubMed Liu T.Y. , Huang T.K. , Yang S.Y. , Hong Y.T. , Huang S.M. , Wang F.N. , et al. . ( 2016 ) Identification of plant vacuolar transporters mediating phosphate storage . Nat. Commun . 7 : 11095 . Google Scholar CrossRef Search ADS PubMed Lloyd J.C. , Zakhleniuk O.V. ( 2004 ) Responses of primary and secondary metabolism to sugar accumulation revealed by microarray expression analysis of the Arabidopsis mutant, pho3 . J. Exp. Bot . 55 : 1221 – 1230 . Google Scholar CrossRef Search ADS PubMed Lopez-Arredondo D.L. , Leyva-Gonzalez M.A. , Gonzalez-Morales S.I. , Lopez-Bucio J. , Herrera-Estrella L. ( 2014 ) Phosphate nutrition: improving low-phosphate tolerance in crops . Annu. Rev. Plant Biol. 65 : 95 – 123 . Google Scholar CrossRef Search ADS PubMed Marschner H. ( 2012 ) Marschner’s Mineral Nutrition of Higher Plants . Academic Press , London . Mikulska M. , Bomsel J.-L. , Rychter A.M. ( 1998 ) The influence of phosphate deficiency on photosynthesis, respiration and adenine nucleotide pool in bean leaves . Photosynthetica 35 : 79 – 88 . Google Scholar CrossRef Search ADS Misson J. , Raghothama K.G. , Jain A. , Jouhet J. , Block M.A. , Bligny R. , et al. . ( 2005 ) A genome-wide transcriptional analysis using Arabidopsis thaliana Affymetrix gene chips determined plant responses to phosphate deprivation . Proc. Natl. Acad. Sci. USA 102 : 11934 – 11939 . Google Scholar CrossRef Search ADS Morcuende R. , Bari R. , Gibon Y. , Zheng W. , Pant B.D. , Blasing O. , et al. . ( 2007 ) Genome-wide reprogramming of metabolism and regulatory networks of Arabidopsis in response to phosphorus . Plant Cell Environ. 30 : 85 – 112 . Google Scholar CrossRef Search ADS PubMed Müller R. , Morant M. , Jarmer H. , Nilsson L. , Nielsen T.H. ( 2007 ) Genome-wide analysis of the Arabidopsis leaf transcriptome reveals interaction of phosphate and sugar metabolism . Plant Physiol . 143 : 156 – 171 . Google Scholar CrossRef Search ADS PubMed Nagley P. ( 1988 ) Eukaryote membrane genetics: the Fo sector of mitochondrial ATP synthase . Trends Genet. 4 : 46 – 51 . Google Scholar CrossRef Search ADS PubMed Nunes-Nesi A. , Carrari F. , Lytovchenko A. , Smith A.M.O. , Loureiro M.E. , Ratcliffe R.G. , et al. . ( 2005 ) Enhanced photosynthetic performance and growth as a consequence of decreasing mitochondrial malate dehydrogenase activity in transgenic tomato plants . Plant Physiol . 137 : 611 – 622 . Google Scholar CrossRef Search ADS PubMed Pagliarani A. , Nesci S. , Ventrella V. ( 2013 ) Modifiers of the oligomycin sensitivity of the mitochondrial F1F0-ATPase . Mitochondrion 13 : 312 – 319 . Google Scholar CrossRef Search ADS PubMed Pant B.D. , Pant P. , Erban A. , Huhman D. , Kopka J. , Scheible W.R. ( 2015 ) Identification of primary and secondary metabolites with phosphorus status-dependent abundance in Arabidopsis, and of the transcription factor PHR1 as a major regulator of metabolic changes during phosphorus limitation . Plant Cell Environ. 38 : 172 – 187 . Google Scholar CrossRef Search ADS PubMed Plaxton W.C. , Tran H.T. ( 2011 ) Metabolic adaptations of phosphate-starved plants . Plant Physiol. 156 : 1006 – 1015 . Google Scholar CrossRef Search ADS PubMed Poirier Y. , Thoma S. , Somerville C. , Schiefelbein J. ( 1991 ) Mutant of Arabidopsis deficient in xylem loading of phosphate . Plant Physiol . 97 : 1087 – 1093 . Google Scholar CrossRef Search ADS PubMed Puga M.I. , Mateos I. , Charukesi R. , Wang Z. , Franco-Zorrilla J.M. , de Lorenzo L. , et al. . ( 2014 ) SPX1 is a phosphate-dependent inhibitor of Phosphate Starvation Response 1 in Arabidopsis . Proc Natl. Acad. Sci. USA 111 : 14947 – 14952 . Google Scholar CrossRef Search ADS Raghothama K.G. ( 1999 ) Phosphate acquisition . Annu. Rev. Plant Physiol. Plant Mol. Biol. 50 : 665 – 693 . Google Scholar CrossRef Search ADS PubMed Rao I.M. , Terry N. ( 1995 ) Leaf phosphate status, photosynthesis, and carbon partitioning in sugar beet (IV. Changes with time following increased supply of phosphate to low-phosphate plants) . Plant Physiol. 107 : 1313 – 1321 . Google Scholar CrossRef Search ADS PubMed Rubio V. , Bustos R. , Irigoyen M.L. , Cardona-Lopez X. , Rojas-Triana M. , Paz-Ares J. ( 2009 ) Plant hormones and nutrient signaling . Plant Mol. Biol. 69 : 361 – 373 . Google Scholar CrossRef Search ADS PubMed Rubio V. , Linhares F. , Solano R. , Martin A.C. , Iglesias J. , Leyva A. , et al. . ( 2001 ) A conserved MYB transcription factor involved in phosphate starvation signaling both in vascular plants and in unicellular algae . Genes Dev . 15 : 2122 – 2133 . Google Scholar CrossRef Search ADS PubMed Secco D. , Jabnoune M. , Walker H. , Shou H. , Wu P. , Poirier Y. , et al. . ( 2013 ) Spatio-temporal transcript profiling of rice roots and shoots in response to phosphate starvation and recovery . Plant Cell 25 : 4285 – 4304 . Google Scholar CrossRef Search ADS PubMed Singer T. , Burke E. ( 2003 ) High-throughput TAIL-PCR as a tool to identify DNA flanking insertions. In Plant Functional Genomics . Edited by Grotewold E. pp. 241 – 272 . Humana Press , Totowa, NJ . Google Scholar CrossRef Search ADS Smith M.K. , Day D.A. , Whelan J. ( 1994 ) Isolation of a novel soybean gene encoding a mitochondrial ATP synthase subunit . Arch. Biochem. Biophys . 313 : 235 – 240 . Google Scholar CrossRef Search ADS PubMed Solfanelli C. , Poggi A. , Loreti E. , Alpi A. , Perata P. ( 2006 ) Sucrose-specific induction of the anthocyanin biosynthetic pathway in Arabidopsis . Plant Physiol . 140 : 637 – 646 . Google Scholar CrossRef Search ADS PubMed Ticconi C.A. , Delatorre C.A. , Abel S. ( 2001 ) Attenuation of phosphate starvation responses by phosphite in Arabidopsis . Plant Physiol . 127 : 963 – 972 . Google Scholar CrossRef Search ADS PubMed Tsuge T. , Tsukaya H. , Uchimiya H. ( 1996 ) Two independent and polarized processes of cell elongation regulate leaf blade expansion in Arabidopsis thaliana (L.) Heynh . Development 122 : 1589 – 1600 . Google Scholar PubMed Varadarajan D.K. , Karthikeyan A.S. , Matilda P.D. , Raghothama K.G. ( 2002 ) Phosphite, an analog of phosphate, suppresses the coordinated expression of genes under phosphate starvation . Plant Physiol . 129 : 1232 – 1240 . Google Scholar CrossRef Search ADS PubMed Wang Z. , Ruan W. , Shi J. , Zhang L. , Xiang D. , Yang C. , et al. . ( 2014 ) Rice SPX1 and SPX2 inhibit phosphate starvation responses through interacting with PHR2 in a phosphate-dependent manner . Proc. Natl. Acad. Sci. USA 111 : 14953 – 14958 . Google Scholar CrossRef Search ADS Weigel D. , Ahn J.H. , Blazquez M.A. , Borevitz J.O. , Christensen S.K. , Fankhauser C. , et al. . ( 2000 ) Activation tagging in Arabidopsis . Plant Physiol. 122 : 1003 – 1013 . Google Scholar CrossRef Search ADS PubMed Wild R. , Gerasimaite R. , Jung J.Y. , Truffault V. , Pavlovic I. , Schmidt A. , et al. . ( 2016 ) Control of eukaryotic phosphate homeostasis by inositol polyphosphate sensor domains . Science 352 : 986 – 990 . Google Scholar CrossRef Search ADS PubMed Abbreviations Abbreviations GC genomic complementarion InsP inositol polyphosphate MGP1 MALE GAMETOPHYTE DEFECTIVE1 NMR nuclear magnetic resonance Phi phosphite PHR1 PHOSPHATE STARVATION RESPONSE1 PP-InsP inositol pyrophosphate PSi phosphate starvation induced PSR phosphate starvation response PSr phosphate starvation repressed qRT–PCR quantitative reverse transcription–PCR RNAi RNA interference TMRE tetramethylrhodamine ethyl ester UTR untranslated region WT wild type © The Author(s) 2018. Published by Oxford University Press on behalf of Japanese Society of Plant Physiologists. All rights reserved. For permissions, please email: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Plant and Cell Physiology Oxford University Press

Phosphite-Mediated Suppression of Anthocyanin Accumulation Regulated by Mitochondrial ATP Synthesis and Sugars in Arabidopsis

Loading next page...
 
/lp/ou_press/phosphite-mediated-suppression-of-anthocyanin-accumulation-regulated-5ybgJsRkfl
Publisher
Oxford University Press
Copyright
© The Author(s) 2018. Published by Oxford University Press on behalf of Japanese Society of Plant Physiologists. All rights reserved. For permissions, please email: journals.permissions@oup.com
ISSN
0032-0781
eISSN
1471-9053
D.O.I.
10.1093/pcp/pcy051
Publisher site
See Article on Publisher Site

Abstract

Abstract Despite the essential role of phosphate (Pi) in plant growth and development, how plants sense and signal the change of Pi supply to adjust its uptake and utilization is not yet well understood. Pi itself has been proposed to be a signaling molecule that regulates Pi starvation responses (PSRs) because phosphite (Phi), a non-metabolized Pi analog, suppresses several PSRs. In this study, we identified a phosphite-insensitive1 (phi1) mutant which retained anthocyanin, a visible PSR, in Phi-containing but Pi-deficient medium. phi1 mutants were impaired in the gene encoding an FAd subunit of mitochondrial F1Fo-ATP synthase and showed a reduced mitochondrial ATP level in roots, growth hypersensitivity to oligomycin and an increased mitochondrial membrane potential, suggesting that this gene has a crucial role in mitochondrial ATP synthesis. phi1 mutants accumulated a high level of sugars in shoots, which may account for the increased accumulation of anthocyanin and starch in Phi-containing conditions. Gene expression analysis showed that a subset of genes involved in carbohydrate metabolism in phi1 was misregulated in response to Phi. The majority of genes were repressed by Pi starvation and, unlike wild-type plants, their repression in phi1 was not affected by the addition of Phi. Our findings show that defective mitochondrial ATP synthesis results in sugar accumulation, leading to alteration of Phi-mediated suppression of PSRs. This study reinforces the role of sugars, and also reveals a cross-talk among ATP, sugars and Pi/Phi molecules in mediating PSRs. Introduction Phosphorus (P) is an essential element for cell growth and development. It serves as a component of phospholipids and nucleic acids, and is also an energy source and participates in signaling transduction (Marschner 2012). Plant roots predominately absorb P in the form of inorganic phosphate (Pi, HPO42– or H2PO4–); however, the availability of Pi in the soil is often limited because Pi easily forms aggregates with cations or is converted into its organic form (Raghothama 1999). To cope with the low availability of Pi, plants have developed a series of adaptive reactions, collectively known as Pi starvation responses (PSRs), to enhance Pi uptake from the soil as well as to readjust Pi distribution and usage within plants. Some of the well-studied PSRs include modification of root system architecture, enhanced expression of Pi transporters, secretion of phosphatases and RNases, remodeling of membrane lipids, accumulation of anthocyanin and increased starch formation (Lopez-Arredondo et al. 2014). The transcriptional regulation underlying these PSRs, in which a large number of genes, ranging from 500 to 2,000 depending on the experimental conditions, were up-regulated upon Pi starvation, has been a focus of several studies (Misson et al. 2005, Bustos et al. 2010, Secco et al. 2013, Liu et al. 2016). A MYB-related transcription factor family in Arabidopsis, PHOSPHATE STARVATION RESPONSE1 (PHR1) and PHR1-LIKE1 (PHL1), are involved in activating the expression of a considerable number of Pi starvation-induced (PSi) genes (collectively named PHR1 regulons), which generally harbor a PHR1 binding sequence (P1BS) motif in their promoter (Rubio et al. 2001, Bustos et al. 2010). Nonetheless, 30% and 50% of PSi genes in shoots and roots, respectively, are not significantly reduced in phr1 phl1 mutants (Bustos et al. 2010), suggesting the existence of PHR1-independent regulation. On the other hand, the transcriptional regulation of Pi starvation-repressed (PSr) genes is virtually unexplored. In comparison with our knowledge about the downstream PSRs, how Pi starvation signaling is perceived and transduced is less understood. Several compounds, such as sugars, hormones and microRNA, have been proposed to be potential signaling molecules (Chiou and Lin 2011, Hammond and White 2011, Lin et al. 2014), and several findings support a role for sugar in regulating PSRs. For example, induction of PSi genes in Arabidopsis and white lupin roots was reduced by blocking shoot to root translocation of sugars, through either the impairment of phloem-specific SUCROSE TRANSPORTER2 (SUC2) or stem girdling experiments (Krapp and Stitt 1995, Lloyd and Zakhleniuk 2004, Liu et al. 2005, Lei et al. 2011). In contrast, overexpression of sucrose transporters resulted in accumulation of sugars in both shoots and roots, coinciding with the activation of PSi genes (Lei et al. 2011, Dasgupta et al. 2014). Moreover, exogenous application of sugars augmented the expression of several PSi genes under Pi deficiency (Liu et al. 2005). Transcriptomic analysis of Arabidopsis leaf also revealed the interaction between sugar and Pi in regulating gene expression, in which a cluster of PSi genes specifically involved in alleviating Pi starvation was further induced by sucrose supply (Müller et al. 2007). Therefore, sucrose was considered to be a global regulator of PSRs although the underlying mechanism is unknown. Dasgupta et al. (2014) proposed that the effect of sugars on the PSRs may result from the sequestration of Pi by excess sugars or the disruption of carbon (C) to P balance. In addition, the Pi molecule itself has been proposed to be a signal for the regulation of PSRs. This is supported by the finding that a non-metabolized analog of Pi, phosphite (Phi, H2PO3– or HPO32−), suppressed a number of PSRs, such as a high root to shoot ratio, root hair formation, anthocyanin accumulation, acid phosphatase activity, membrane lipid remodeling and expression of many PSi genes (Ticconi et al. 2001, Varadarajan et al. 2002, Kobayashi et al. 2006, Berkowitz et al. 2013, Jost et al. 2015). Because of the structural similarity between Pi and Phi and the results from competitive inhibition of Phi on Pi transport, Phi is believed to be taken up through the PHOSPHATE TRANSPORTER 1 (PHT1) family proteins (Danova-Alt et al. 2008, Jost et al. 2015). Once inside the root, Arabidopsis PHT1;8 and PHT1;9 Pi transporters were shown to be involved in root to shoot Phi translocation (Jost et al. 2015). How Pi or Phi regulates the expression of PSR genes has remained unclear until recently. Puga et al. (2014) found that SYG1/Pho81/XPR1 (SPX) domain-containing proteins, SPX1 or SPX2, bind to PHR1 in the presence of Pi or Phi, and prevent PHR1 from activating its target genes in Arabidopsis. Conversely, PHR1 is liberated from SPX1 binding when Pi is limited, thus enabling activation of PHR1-dependent PSi genes. Similar findings were also reported for the interaction of OsSPX1 or OsSPX2 and OsPHR2 in rice (Wang et al. 2014). These results support the notion that Pi is a signaling molecule and imply the potential of SPX proteins as Pi sensors. As Pi sources for fertilizer usage are finite, attention has been paid to breeding crops with improved P use efficiency. To achieve this aim, understanding how plants sense P availability in the rhizosphere and how the signal is transmitted to initiate adaptive responses to increase Pi uptake/usage by plants is essential. In this study, we employed Phi in a genetic screen to try to identify the regulatory components involved in the PSRs mediated by Pi molecules. We isolated a phosphite-insensitive1 (phi1) mutant which retained anthocyanin accumulation in shoots even in the presence of Phi. Further genetic and metabolite analyses revealed that the phi1 mutant is impaired in the gene encoding the FAd subunit of mitochondrial F1-Fo ATP synthase, and accumulates a high level of sugar which may account for the retention of anthocyanin under Phi treatment. In agreement with the abnormal carbohydrate metabolism, gene expression analysis revealed Phi-dependent misexpression of several starch- and sugar-related genes in phi1. Our study highlights a cross-talk among ATP, sugars and Pi/Phi molecules to co-ordinate the PSRs. Results Identification of a Phi-insensitive mutant To identify the components involved in the regulation of PSR, we established a system to select Arabidopsis mutants that responded differently to Phi as compared with wild-type (WT) plants. We first grew Arabidopsis WT seedlings in half-strength Hoagland’s medium containing different combinations of Phi and Pi concentrations to evaluate the effects on the suppression of anthocyanin accumulation by Phi. Because anthocyanin accumulation is an easily visible PSR, we employed it as a screening phenotype. To compare the anthocyanin accumulation between Pi-sufficient (500 μM Pi, +Pi) and deficient conditions (no supply of Pi, –Pi), a medium containing 500 μM Phi but without supply of Pi (+Phi) was chosen as a screening condition because it was effective in suppressing anthocyanin accumulation and caused less harm to growth. Using this condition, we screened a pool (∼60,000 lines) of activation tagging mutants of Arabidopsis (Weigel et al. 2000) and identified a mutant, phosphite insensitive1 (phi1), that retained a substantial amount of anthocyanin in the shoot compared with WT seedlings (Fig. 1A). Although Phi suppressed the growth of both WT and phi1 plants, phi1 mutants showed reduced biomass and growth of primary and lateral roots compared with the WT under all growth conditions (Fig. 1A, Supplementary Fig. S1A–D). Fig. 1 View largeDownload slide A phosphite-insensitive mutant (phi1) retains anthocyanin under +Phi conditions. The phenotype (A) and RNA level of three anthocyanin biosynthesis genes, CHS, ANS and DFR (C) of 12-day-old WT and phi1 plants grown under Pi-sufficient (+Pi), Pi-deficient (–Pi) or Phi-containing (+Phi) conditions for 6 d. Enlargement of the abaxial side of leaves under +Phi is shown. The anthocyanin content in 16-day-old WT and phi1 shoots which were treated for 10 d after 6 d germination (B). Letters represent significant differences among genotypes and growth conditions, P < 0.05 (ANOVA). Error bars represent SEs from three biological replicates. Scale bars = 1 cm. Fig. 1 View largeDownload slide A phosphite-insensitive mutant (phi1) retains anthocyanin under +Phi conditions. The phenotype (A) and RNA level of three anthocyanin biosynthesis genes, CHS, ANS and DFR (C) of 12-day-old WT and phi1 plants grown under Pi-sufficient (+Pi), Pi-deficient (–Pi) or Phi-containing (+Phi) conditions for 6 d. Enlargement of the abaxial side of leaves under +Phi is shown. The anthocyanin content in 16-day-old WT and phi1 shoots which were treated for 10 d after 6 d germination (B). Letters represent significant differences among genotypes and growth conditions, P < 0.05 (ANOVA). Error bars represent SEs from three biological replicates. Scale bars = 1 cm. When anthocyanin was measured in shoots (normalized by fresh weight), both the WT and phi1 had a very low level under +Pi, and it increased remarkably under –Pi, although phi1 had a slightly higher level than the WT (Fig. 1B). In response to a 10 d Phi treatment, the WT showed a significant reduction of anthocyanin, whereas phi1 retained a substantial amount, approximately six times that in the WT (Fig. 1B). To take into account the possible effect due to reduced shoot biomass in phi1, we normalized anthocyanin based on the number of shoots and found that phi1 still retained higher anthocyanin than the WT under +Phi (Supplementary Fig. S1E). phi1 true leaves had a higher cell density than those of the WT due to reduced cell size, but the cell number per leaf (cotyledon or true leaf) of phi1 was comparable with that of the WT (Supplementary Fig. S2). Anthocyanin was also accumulated in the cotyledons of phi1, in which cell density and number are similar to those of the WT. Therefore, the increased anthocyanin accumulation in phi1 shoots under +Phi is not caused by impaired growth. Consistent with the accumulation of anthocyanin, the expression of three selected anthocyanin biosynthetic genes, namely CHALCONE SYNTHASE (CHS), ANTHOCYANIDIN SYNTHASE (ANS) and DIHYDROFLAVONOL 4-REDUCTASE (DFR), in phi1 was also higher than that in the WT, especially under +Phi (Fig. 1C). Therefore, we concluded that Phi-mediated suppression of anthocyanin accumulation is impaired in phi1 mutants. phi1 mutants are defective in a subunit of mitochondrial ATP synthase To determine the gene responsible for the phi1 phenotype, we carried out thermal asymmetric interlaced (TAIL)-PCR (Singer and Burke 2003) and identified a T-DNA insertion at the 3'-untranslated region (UTR) of AT2G21870 encoding a plant-specific FAd subunit of mitochondrial F1-Fo ATP synthase in phi1 mutants (Fig. 2A). This T-DNA insertion resulted in significantly reduced expression of AT2G21870 mRNA in both shoots and roots under all conditions tested, as measured by quantitative reverse transcription–PCR (qRT–PCR) (Fig. 2B) and RNA gel blot (Supplementary Fig. S3) analyses. Because the phi1 mutant was generated by activation tagging, we also examined the expression of two flanking genes (AT2G21860 and AT2G21880) and found no significant differences from the WT under nearly all growth conditions (Supplementary Fig. S4), suggesting that these two genes do not contribute to the phi1 phenotype. Because of the absence of activation of neighboring genes, we suspect that the enhancer could have been lost in the T-DNA. Fig. 2 View largeDownload slide phi1 mutants are defective in MGP1. A T-DNA was inserted in the 3'-UTR of MGP1 (AT2G21870) in phi1 mutants (A), which resulted in a reduction in the expression of MGP1 in all growth conditions (+Pi, –Pi and +Phi) (B). The arrows in (A) indicate the positions of primers used for qRT–PCR. A negative correlation between the expression level of MGP1 and anthocyanin accumulation in plants treated under +Phi for 6 d (C). The data were obtained from the WT, phi1, RNAi and genomic complementation (GC) whole seedlings. The RNAi (RNAi#26) and GC (GC#11) lines used for the subsequent experiments are indicated. r represents Pearson’s correlation coefficient. Letters represent significant differences among genotypes and growth conditions, P < 0.05 (ANOVA), and error bars represent SEs of three biological replicates. Fig. 2 View largeDownload slide phi1 mutants are defective in MGP1. A T-DNA was inserted in the 3'-UTR of MGP1 (AT2G21870) in phi1 mutants (A), which resulted in a reduction in the expression of MGP1 in all growth conditions (+Pi, –Pi and +Phi) (B). The arrows in (A) indicate the positions of primers used for qRT–PCR. A negative correlation between the expression level of MGP1 and anthocyanin accumulation in plants treated under +Phi for 6 d (C). The data were obtained from the WT, phi1, RNAi and genomic complementation (GC) whole seedlings. The RNAi (RNAi#26) and GC (GC#11) lines used for the subsequent experiments are indicated. r represents Pearson’s correlation coefficient. Letters represent significant differences among genotypes and growth conditions, P < 0.05 (ANOVA), and error bars represent SEs of three biological replicates. AT2G21870, previously named MGP1 (MALE GAMETOPHYTE DEFECTIVE1), is essential for pollen development (Li et al. 2010). Loss of function of MGP1 led to death of pollen grains, and no homozygous mutant plants could be produced in the progeny of the heterozygous mgp1–/+ mutant. Anthocyanin accumulation in phi1 mutants under Phi treatment can be restored to the WT level by introducing the genomic sequence of MGP1 (Fig. 2C), suggesting that MGP1 is responsible for the phi1 phenotype. Because no additional mutants are available, we generated MGP1-RNAi (RNA interference) lines for further confirmation. As shown in Fig. 2C, the RNAi lines with reduced MGP1 expression accumulated a higher amount of anthocyanin than WT plants under +Phi. The expression of MGP1 was negatively correlated with anthocyanin accumulation under Phi treatment (Pearson’s correlation coefficient r = –0.83) (Fig. 2C), suggesting that MGP1 is involved in Phi-mediated suppression of anthocyanin accumulation. To segregate potential additional T-DNA insertion loci, we backcrossed the phi1 mutant to WT plants. All F1 progeny showed the WT phenotype of no anthocyanin accumulation under +Phi treatment, suggesting the recessive nature of the phi1 mutation. Mutant lines which retained anthocyanin under +Phi and had low expression of MGP1 from the F2 population (Supplementary Fig. S5) were analyzed in subsequent experiments, together with one RNAi (RNAi#26) line and one genomic complementation (GC) line, GC#11. The phi1 mutant is impaired in mitochondrial ATP synthesis Because the phi1 mutant is impaired in the gene encoding a subunit of mitochondrial ATP synthase, it is likely that ATP metabolism is compromised. We thus measured the ATP level. In the roots, phi1 had a reduced amount of ATP under +Pi, approximately 60% of that of the WT (Fig. 3A). The ATP level was greatly decreased after 6 d treatment with –Pi or +Phi, but no differences were observed between the WT and phi1 (Fig. 3A). Surprisingly, the shoot ATP level remained relatively unchanged in all growth conditions and phi1 showed no significant differences from the WT regardless of treatments (Fig. 3B). When grown in medium containing 5 µM oligomycin which specifically inhibits the activity of mitochondrial ATP synthase (Pagliarani et al. 2013), phi1 mutants were more sensitive to oligomycin than WT plants, supporting the role of MGP1 in mitochondrial ATP metabolism (Fig. 3C). Because mitochondrial ATP synthesis is coupled with the proton motive force, impairment of ATP synthesis results in accumulation of protons at the intermembrane space, leading to a higher mitochondrial membrane potential. We thus examined whether the mitochondrial membrane potential is altered in phi1 mutants. Tetramethylrhodamine ethyl ester (TMRE) is a fluorescent dye used to measure the membrane potential of mitochondria within a cell (Ehrenberg et al. 1988). By applying TMRE to the roots grown under +Pi, we found that the phi1 root hairs displayed a stronger signal of fluorescence than those of the WT (Fig. 3D), which indicates an increased mitochondrial membrane potential. As a control, addition of oligomycin markedly increased the signal in both WT and phi1 plants (Fig. 3D). It is conceivable that the increased mitochondrial membrane potential in phi1 mutants is a result of diminished proton pumping caused by the impairment of mitochondrial ATP synthesis. Fig. 3 View largeDownload slide phi1 mutants are impaired in ATP synthesis and sensitive to oligomycin. The relative ATP level in 12-day-old WT and phi1 roots (A) and shoots (B) after 6 d +Pi, –Pi or +Phi treatment. Letters represent significant differences compared with the WT, P < 0.05 (ANOVA). Growth phenotypes of the WT and phi1 after oligomycin treatment (C). Examination of mitochondrial membrane potential by TMRE fluorescent dye in the root hairs of the WT and phi1 with (+) or without (−) oligomycin treatment (D). The enhanced signal of red dots represents mitochondria with increased membrane potential. The microscopic fluorescent intensity was adjusted to a gain (master) value of 750 and 660 for control and oligomycin treatment, respectively. The plants in (C) and (D) were grown under +Pi conditions. Fig. 3 View largeDownload slide phi1 mutants are impaired in ATP synthesis and sensitive to oligomycin. The relative ATP level in 12-day-old WT and phi1 roots (A) and shoots (B) after 6 d +Pi, –Pi or +Phi treatment. Letters represent significant differences compared with the WT, P < 0.05 (ANOVA). Growth phenotypes of the WT and phi1 after oligomycin treatment (C). Examination of mitochondrial membrane potential by TMRE fluorescent dye in the root hairs of the WT and phi1 with (+) or without (−) oligomycin treatment (D). The enhanced signal of red dots represents mitochondria with increased membrane potential. The microscopic fluorescent intensity was adjusted to a gain (master) value of 750 and 660 for control and oligomycin treatment, respectively. The plants in (C) and (D) were grown under +Pi conditions. Accumulation of Pi and Phi in phi1 mutants We next examined whether Pi or Phi content is altered in phi1 mutants. Compared with WT and GC plants, there were no striking differences in shoot Pi content except a small reduction in the phi1 mutant grown under +Pi conditions and in the RNAi line grown under –Pi and +Phi conditions (Fig. 4A). Nevertheless, both phi1 and RNAi plants showed a higher root Pi content than the WT and GC plants under both +Pi and +Phi conditions (Fig. 4B). Fig. 4 View largeDownload slide Pi and Phi content in phi1 mutants. The Pi content in the shoot (A) and root (B) of the WT, phi1, RNAi and genomic complementation (GC) plants grown under +Pi, –Pi and +Phi conditions. The Phi content in the shoot and root of plants grown under +Phi conditions (C). Letters represent significant differences among genotypes and growth conditions, P < 0.05 (ANOVA). Error bars are SEs from 5–9 biological replicates. Fig. 4 View largeDownload slide Pi and Phi content in phi1 mutants. The Pi content in the shoot (A) and root (B) of the WT, phi1, RNAi and genomic complementation (GC) plants grown under +Pi, –Pi and +Phi conditions. The Phi content in the shoot and root of plants grown under +Phi conditions (C). Letters represent significant differences among genotypes and growth conditions, P < 0.05 (ANOVA). Error bars are SEs from 5–9 biological replicates. When Phi was measured, while all the root samples had a comparable level, the shoot of phi1 but not of the RNAi line had a reduced level, about 60% of that of the WT (Fig. 4C). This argues against the possibility of anthocyanin accumulation in Phi-treated phi1 and RNAi lines as a consequence of reduced Phi accumulation. To examine further if this was the case, pho1 mutants impaired in root to shoot translocation of Pi (Poirier et al. 1991), and probably also Phi, were inspected. Under +Phi conditions, pho1 mutants showed diminished anthocyanin similar to WT plants (Supplementary Fig. S6A, B) but accumulated a level of shoot Phi comparable with phi1 (Supplementary Fig. S6C). This suggests that low Phi accumulation in phi1 shoots is unlikely to be the reason for anthocyanin accumulation, and additional factors are involved. High sugar in phi1 mutants antagonizes Phi-mediated suppression of anthocyanin and starch accumulation Since phi1 showed impaired mitochondrial ATP synthesis, the cellular metabolic processes are probably affected. It has been documented that sugars positively regulate anthocyanin accumulation (Solfanelli et al. 2006), and both sugars and anthocyanin are increased upon Pi starvation (Ciereszko and Barbachowska 2000, Pant et al. 2015); we therefore quantified the amount of sugar. As expected, the levels of sucrose, fructose and glucose in WT and phi1 shoots were increased under –Pi (Fig. 5A). However, it is surprising that their level under +Phi remained high or even higher than in –Pi in all genotypes. Of note, phi1 shoots accumulated about 1.5-fold the WT level of sucrose and glucose under +Pi and +Phi, and at least 2-fold the WT level of fructose under all growth conditions (Fig. 5A). Similar to phi1 mutants, the RNAi line also had a higher level of sugars (except for glucose) than the WT under +Phi conditions. The accumulation of sugars was restored to the WT level in the GC lines (Fig. 5A), indicating that the reduced expression of MGP1 leads to overaccumulation of sugar in shoots. Fig. 5 View largeDownload slide Effect of sugar on the Phi-mediated suppression of anthocyanin and starch accumulation. Relative amount of sucrose, fructose and glucose in the shoot of the WT, phi1, RNAi and genomic complementation (GC) plants (A). Anthocyanin content in WT plants treated with various combinations of sucrose and Pi or Phi as indicated (B). Anthocyanin accumulation in the shoot of WT, suc2-5 (C) and hps1 (D) grown under +Pi, –Pi and +Phi conditions. Staining of starch by Lugol’s iodine is shown in dark blue (E). Double and single asterisks represent significant difference with P < 0.01 and P < 0.05 (Student’s t-test), respectively, in the comparison with the WT. Letters represent significant differences among genotypes and growth conditions, P < 0.05 (ANOVA). Error bars represent SEs. Scale bars = 1 cm. Fig. 5 View largeDownload slide Effect of sugar on the Phi-mediated suppression of anthocyanin and starch accumulation. Relative amount of sucrose, fructose and glucose in the shoot of the WT, phi1, RNAi and genomic complementation (GC) plants (A). Anthocyanin content in WT plants treated with various combinations of sucrose and Pi or Phi as indicated (B). Anthocyanin accumulation in the shoot of WT, suc2-5 (C) and hps1 (D) grown under +Pi, –Pi and +Phi conditions. Staining of starch by Lugol’s iodine is shown in dark blue (E). Double and single asterisks represent significant difference with P < 0.01 and P < 0.05 (Student’s t-test), respectively, in the comparison with the WT. Letters represent significant differences among genotypes and growth conditions, P < 0.05 (ANOVA). Error bars represent SEs. Scale bars = 1 cm. We speculated that overaccumulation of sugar in phi1 mutants may over-ride Phi-mediated suppression of anthocyanin accumulation. We increased the sucrose supplement in the +Phi media from 1% to 5% or 10% and found a 2- to 6-fold increase of anthocyanin in WT plants, which backed up our supposition (Fig. 5B). Furthermore, two mutants that accumulate a high level of endogenous sucrose in shoots, suc2 and hypersensitive to phosphate starvation1 (hps1) (Lloyd and Zakhleniuk 2004, Lei et al. 2011), also accumulated a higher level of anthocyanin than the WT plants when grown in the +Phi medium (Fig. 5C, D). These results indicate that sugars antagonize Phi-mediated suppression of anthocyanin accumulation. Therefore, the enhanced anthocyanin accumulation under Phi treatment in phi1 and RNAi plants is probably attributed to the increased level of sugars. In addition to sugars, Pi starvation enhances starch accumulation because the activity of a key starch biosynthesis enzyme, ADP-glucose pyrophosphorylase (AGPase), is allosterically inhibited by Pi, while Pi depletion and a high level of sugars up-regulate its activity (Geigenberger 2011). We therefore examined the effect of Phi on Pi starvation-induced starch accumulation. As revealed by Lugol’s iodine staining, Phi treatment suppressed Pi starvation-induced starch accumulation in WT and GC lines, but not in phi1 and RNAi lines (Fig. 5E), indicating that MGP1 is involved in Phi-mediated suppression of starch accumulation. A subset of carbohydrate metabolism genes is misexpressed in phi1 in response to Phi Because phi1 mutants accumulated excessive sugar and starch, we next examined the expression of a subset of PSR genes involved in sugar and starch metabolism. These genes were selected according to a preliminary microarray analysis which showed prominent differences in expression between the WT and phi1 under +Phi. qRT–PCR analyses were carried out to examine the gene expression in WT, GC, RNAi and phi1 seedlings grown under +Pi, –Pi and +Phi conditions. As illustrated by the heat map, up-regulation of several genes (e.g. CHS, ANS and DFR) involved in anthocyanin biosynthesis under –Pi was retained under +Phi in phi1 and RNAi lines, in contrast to being suppressed in WT and GC plants (Fig. 6A, Group I; see original data in Supplementary Table S1). A similar situation was also observed for a starch biosynthetic gene encoding the large subunit of ADP-glucose pyrophosphorylase (APL3), PHOSPHOGLUCAN WATER DIKINASE (PWD) involved in starch degradation and GLYCOSYL HYDROLASE 9B8 (GH9B8) (Fig. 6A, Group I), reflecting the disturbance of starch metabolism in phi1 and RNAi plants under +Phi. The increased expression of PWD is likely to be the result of a feedback regulation to avoid overaccumulation of starch because it was up-regulated in all Pi-starved samples. The expression of two selected sugar-related PSi genes, 2,3-bisphosphoglycerate-independent PHOSPHOGLYCERATE MUTASE 2 (iPGAM2) and SUCROSE SYNTHASE 3 (SUS3) was significantly enhanced in phi1 and RNAi plants under +Phi (Fig. 6A, Group II). In contrast to the other PSi genes suppressed by Phi, iPGAM2 and SUS3 transcripts are further up-regulated in response to Phi in all samples, correlated with the enhanced sugar accumulation (Fig. 5A). The activity of SUS3 and APL3 in starch and sucrose metabolism is indicated in Supplementary Fig. S7A. Fig. 6 View largeDownload slide Differential expression of PSR genes in phi1 mutants responsive to Phi. The heat maps show the relative expression of a subset of PSR genes involved in anthocyanin biosynthesis (highlighted with purple), and starch (brown) or sugar (green) metabolism (A), and several PHR1-regulated PSi genes (B) in the WT, phi1, RNAi and genomic complementation (GC) plants grown under +Pi, –Pi and +Phi conditions. The expression level is presented as the Z-score. In (A), the genes are classified into the groups illustrated by the line graphs at the side. The original data of qRT–PCR are given in Supplementary Table S1. Fig. 6 View largeDownload slide Differential expression of PSR genes in phi1 mutants responsive to Phi. The heat maps show the relative expression of a subset of PSR genes involved in anthocyanin biosynthesis (highlighted with purple), and starch (brown) or sugar (green) metabolism (A), and several PHR1-regulated PSi genes (B) in the WT, phi1, RNAi and genomic complementation (GC) plants grown under +Pi, –Pi and +Phi conditions. The expression level is presented as the Z-score. In (A), the genes are classified into the groups illustrated by the line graphs at the side. The original data of qRT–PCR are given in Supplementary Table S1. We also investigated the expression of several sugar-related PSr genes. The repression of these PSr genes by Phi was relieved in WT plants but remained in phi1 and RNAi lines (Fig. 6A, Group III). The weak response to Phi-mediated derepression of these PSr genes in phi1 mutants could be explained by an increased level of sugars, as these sugar-related genes are induced under low carbon sources, such as DARK INDUCIBLE 10/RAFFINOSE SYNTHASE 6 (DIN10/RS6) (Fujiki et al. 2001), AKINBETA1 (Li et al. 2009), MYO-INOSITOL OXYGENASE 2 (MIOX2) (Alford et al. 2012) and SUGAR TRANSPORTER 1 (STP1) (Cordoba et al. 2015). According to the metabolic pathway in the Kyoto Encyclopedia of Genes and Genomes (KEGG) (http://www.genome.jp/kegg-bin/show_organism?org=ath), DIN10/RS6 and SEED IMBIBITION 2/RAFFINOSE SYNTHASE 2 (AtSIP2/RS2) are assigned to galactose metabolism (Supplementary Fig. S7B), while FRUCTOSE-BISPHOSPHATE ALDOLASE 5 (FBA5) and FRUCTOSE-1,6-BISPHOSPHATASE (FBP) are involved in biosynthesis of fructose 1,6-biphosphate and fructose 6-phosphate, respectively, in the glycolysis/gluconeogenesis pathway (Supplementary Fig. S7C). These results suggest that impaired mitochondrial ATP synthase counteracts Phi-mediated derepression of PSr genes involved in carbohydrate metabolism. To examine whether PHR1-mediated regulation of PSi genes was perturbed in phi1 mutants, we analyzed the expression of several typical PHR1-regulated PSi genes, such as INDUCED BY PHOSPHATE STARVATION1 (IPS1), At4, SPX1, SPX2, SPX3, PHT1;4, PHT1;8, MONOGALACTOSYL-DIACYLGLYCEROL SYNTHASE 2 (MGD2) and SULFOQUINOVOSYL-DIACYLGLYCEROL 1 (SQD1). We found that the expression of these genes was suppressed by Phi to a similar extent in all the lines (Fig. 6B), suggesting that impaired mitochondrial ATP synthase activity does not have a major impact on PHR1 regulons. Discussion To investigate the Pi-mediated regulatory pathway of PSRs, we employed Phi in a genetic screen and identified an insensitive mutant (phi1) which retained anthocyanin in the presence of Phi. phi1 mutants are impaired in the MGP1 gene involved in mitochondrial ATP synthesis. Further analyses suggested that the excessive accumulation of sugars in phi1 mutants may account for the retention of anthocyanin. This phenotype was not caused by sucrose supplementation in the growth medium because phi1 still accumulated a high level of anthocyanin when grown on sucrose-free medium (Supplementary Fig. S5C). In addition to the anthocyanin biosynthetic genes, several carbohydrate metabolism-related PSR genes are misregulated in phi1 mutants, particularly under +Phi conditions. This study not only highlights the importance of sugars in the Pi regulatory pathway but also delineates the role of MGP1 in mitochondrial ATP synthesis. It should be noted that anthocyanin is also induced by other stresses. Improvement in screening using a reporter driven by the promoter of a Phi-suppressed PSi gene is needed in the future. Role of MGP1 in mitochondrial ATP synthesis The MGP1 gene encodes a plant-specific FAd subunit of mitochondrial F1-Fo ATP synthase which was initially identified in spinach and soybean (Hamasur and Glaser 1992, Smith et al. 1994). The FAd subunit is thought to be a third component of eukaryotic ATP synthase, an associated component likely to be involved in linking the F1 and Fo complexes (Nagley 1988, Smith et al. 1994). Based on a search in the protein database of the National Center for Biotechnology Information (NCBI) (https://blast.ncbi.nlm.nih.gov/Blast.cgi) and EnsemblPlants (http://plants.ensembl.org/index.html), we found that homologs of this subunit are present extensively in angiosperms and can been identified in a gymnosperm, Picea sitchensis (Sitka spruce), and three ancient plants, Physcomitrella patens (moss), Selaginella moellendorffii (spikemoss) and Marchantia polymorpha (common liverwort). No homologs were identified in the animal kingdom. We postulate that MGP1 emerged during land-plant evolution. Phylogenetic analysis of 41 MGP1 homologs in angiosperms showed that the homologs in monocotyledons and eudicotyledons were categorized distinctly (Supplementary Fig. S8), suggesting that they diverged before the speciation of monocotyledons and eudicotyledons. Li et al. (2010) proposed that MGP1 acts as an inhibitor of ATP hydrolysis during the dehydration stage of pollen development to restore disrupted mitochondrial membrane potential. Loss of function of this gene resulted in a pollen defect and produced no homozygous seeds (Li et al. 2010), pointing to an essential role for MGP1 in plant survival and the success of land-plant evolution. The phi1 mutation we identified in the present study is caused by reduced expression of MGP1, and homozygous seeds are available for characterization. Several lines of evidence from our characterization of phi1 mutants suggest that MGP1 has a crucial role in the synthesis of mitochondrial ATP. First, phi1 mutants were hypersensitive to oligomycin (Fig. 3C), a specific inhibitor of mitochondrial ATP synthase. Secondly, the mitochondrial membrane potential in phi1 mutants was elevated very probably due to impaired mitochondrial ATP synthesis (Fig. 3D). Thirdly, the ATP content in phi1 roots was reduced under +Pi even though it remained unchanged in shoots (Fig. 3A, B). The heterogeneity in the abundance of mitochondrial proteins between photosynthetic and non-photosynthetic tissues (Lee et al. 2008) may partly explain our observation about the discrepancy between roots and shoots. Moreover, because chloroplasts can contribute approximately three times more ATP synthesis than mitochondria under light conditions (Cheung et al. 2014), it is possible that the ATP generated from chloroplasts may compensate the defect of ATP synthesis in the mitochondria of phi1 shoots. However, we did not detect the differences between WT and phi1 shoots after 1 d dark treatment even though the ATP level dropped to 50% of that of the light controls (Supplementary Fig. S9). Although the basis for this is currently unknown, we speculate that reduction of mitochondrial respiration during darkness (Lee et al. 2010) may diminish the difference between the WT and phi1. Our observations also suggest a complex interaction between chloroplasts and mitochondria in ATP metabolism. Impaired mitochondrial ATP synthesis caused abnormal sugar accumulation As Pi is an essential substrate for ATP synthesis, it is intuitive to expect a reduction of ATP under Pi starvation. Indeed, several early studies reported the reduction of ATP in Pi-starved plants (Duff et al. 1989, Rao and Terry 1995, Mikulska et al. 1998, Morcuende et al. 2007). However, a recent analysis of primary metabolites showed an increasing level of ATP in 3 d and 6 d Pi-starved Arabidopsis and barley leaves, respectively (Alexova et al. 2017, Kuo et al., 2018). This discrepancy could be explained by differences in plant species, tissue types or treatment conditions. Nevertheless, conservation of Pi and ATP consumption is an important adaptive response to Pi starvation, which is achieved by several metabolic bypass enzymes involved in glycolysis, the tricarboxylic acid (TCA) cycle and the mitochondrial electron transfer chain (Plaxton and Tran 2011). A number of studies have shown that defects in mitochondrial components involved in the respiratory pathway affected photosynthetic activity and carbon assimilation (Carrari et al. 2003, Nunes-Nesi et al. 2005). Likewise, our study on phi1 mutants showed that reduced mitochondrial ATP synthesis could perturb carbon metabolism, leading to increased sugar accumulation. Further quantitative analysis of metabolites in phi1 mutants, especially those in the TCA cycle, may elucidate its high sugar phenotype and offer additional insights into the interaction between respiratory flux and carbon assimilation. Is Pi a signal? The direct evidence for Pi being a signal molecule comes from the prevalent suppression of PSRs by Phi (Ticconi et al. 2001, Varadarajan et al. 2002, Jost et al. 2015). The dependence of the interaction between SPX and PHR proteins on Pi or Phi (Puga et al. 2014, Wang et al. 2014) not only explains how Pi inhibits the transcriptional activity of PHR but also provides further support for the role of Pi as a signal molecule to regulate the expression of PSR genes negatively. However, a recent structural–functional analysis revealed that inositol polyphosphates (InsPs) or inositol pyrophosphates (PP-InsPs) act as better ligands than Pi for the interaction between rice OsSPX4 and OsPHR2 (Wild et al. 2016), which argues against Pi being a signal in regulating PHR1 activity. Although InsPs or PP-InsPs had a much higher affinity than Pi in in vitro binding analysis (Wild et al. 2016), it has to be considered that the in vivo cellular concentration of InsPs or PP-InsPs is much lower than that of Pi. Several PHR1 regulons are suppressed in Pi-starved plants upon Phi addition (Jost et al. 2015), supporting the role of Pi in suppression of PHR1 activity. However, we could not rule out the possibility that addition of Phi facilitates the release of Pi from other cellular fractions, which is used to generate InsPs, or InsPs are restricted to certain cellular compartments during Pi starvation and released upon Pi or Phi replenishment. In spite of technical challenges, simultaneous quantification of the cellular concentration of Pi, InsP and PP-InsP and their change in response to different external Pi supply should provide further insights (kuo et al., 2018). As InsP or PP-InsP biosynthesis requires ATP, and phi1 had a reduced ATP level, it would be interesting to examine the amount of InsP or PP-InsP in phi1. However, Phi-mediated suppression of the PHR1 regulon is not altered by phi1 mutation (Fig. 6B), suggesting that misregulation of the PSR genes in phi1 mutants is PHR1 independent. How the levels of Pi, InsP or PP-InsP, and ATP are co-ordinated to exert PSRs will be a focus of future study. Cross-talk between sugar metabolism and PSRs The importance of sugars in regulating the PSR has been well documented. Results from alteration of the endogenous sugar level through controlling the phloem loading of sucrose or exogenous application of sugars showed that sugars could activate the expression of a number of PSi genes, or, in other words, sugars antagonize the effect of Pi on the suppression of PSi genes (Liu et al. 2005, Müller et al. 2007, Dasgupta et al. 2014). In accordance with the positive effect of sugar on PSRs, Pi starvation leads to an elevated level of sugars, starch and anthocyanin, but reduced abundance of hexose phosphates, and their amounts are restored upon Pi replenishment (Pant et al. 2015). The enhancement of starch and anthocyanin under Pi starvation could be a consequence of the build up of sugar. Our study showed that phi1 mutants overaccumulate sugar concomitantly with the development of starch and anthocyanin under +Phi conditions, indicating that sugars antagonize the sensitivity of phi1 mutants to Phi-mediated suppression of anthocyanin and starch accumulation. Unlike WT plants, the expression of several sugar-related PSr genes in phi1 remained suppressed even in the presence of Phi (Fig. 6A;Supplementary Table S1), suggesting alteration of carbon metabolism in Phi-treated phi1 plants. Several of these PSr genes, such as DIN10/RS6, AKINBETA1 and MIOX2, have been reported to be suppressed by a high level of sugars, which supports the notion of the effect of sugars on these genes (Fujiki et al. 2001, Li et al. 2009, Alford et al. 2012). These results underline the close interaction between carbon metabolism and regulation of PSRs. Based on the observation of an antagonistic relationship between sugar and Pi/Phi in regulating PSRs, we hypothesize that the extent of PSRs may be modulated by the ratio of sugar to Pi/Phi. Similarly, Dasgupta et al. (2014) also speculated that plants may perceive a disruption of C to P homeostasis to trigger the PSRs. In this scenario, a high sugar to Pi ratio (e.g. Pi deficiency) leads to activation of PSRs, whereas a low sugar to Pi ratio (e.g. Pi sufficiency) results in suppression of PSRs. In the presence of Phi, the decrease of the sugar to Pi/Phi ratio would favor the attenuation of PSRs. Retention of PSRs in phi1 mutants under +Phi conditions could be explained by an elevated ratio of sugar to Pi/Phi due to overaccumulation of sugar (Figs 4C, 5A). Nevertheless, additional regulatory factors, such as hormones (Rubio et al. 2009) or cell cycle activity (Lai et al. 2007), must be involved because the change in the sugar to Pi ratio cannot explain all the PSRs. Future investigation to test this hypothesis will be needed. Materials and Methods Plant materials and growth conditions Activation tagging mutants of Arabidopsis thaliana (stock CS31100) were purchased from the Arabidopsis Biological Resource Center. The accession Columbia-7 (Col-7) was used as a WT control for the phi1 mutant. In the case of pho1, suc2-5 (SALK_087046) and hps1 mutants, Col-0 was used as the WT. Seeds were surface sterilized and germinated on agar medium of half-strength modified Hoagland’s nutrient solution containing 250 μM KH2PO4, 1% sucrose and 0.8% Bactoagar, with pH adjusted to 5.7 (Aung et al. 2006). hps1 mutants were germinated in medium without sucrose supplementation. After germination, 6- to 8-day-old seedlings were transferred to the media with different P sources for 6 d unless specified, designated as +Pi, –Pi and +Phi. The +Pi and +Phi media contained 500 μM KH2PO4 and 500 μM phosphorous acid (H2PO3, Sigma), respectively, and no P source was supplied in the –Pi medium. For oligomycin treatment, 6-day-old seedlings were transferred to the +Pi, –Pi and +Phi media containing 5 μM oligomycin and grown for an additional 9 d. For exogenous sucrose treatment, sucrose was supplied to +Phi medium at three concentrations, 1% (∼30 mM), 5% (∼150 mM) or 10% (∼300 mM). Plants were grown at 22°C under a cycle of 16 h light (white fluorescent light at 100–150 μE m–2 s–1) and 8 h dark. Quantification of Pi, Phi and anthocyanin Pi content was determined as described (Chiou et al. 2006). Phi was measured by nuclear magnetic resonance (NMR). A sample mixture (400 μl) comprising 300 μl of homogenate (in 1% acetic acid), 80 μl of 10 mM EDTA and 20 μl of deuterium water was dispensed into NMR tubes. The NMR instrument was operated using Bruker Avance 500 AV and analyzed by TopSpin software (Bruker) at the High Field Nuclear Magnetic Resonance Center (HFNMRC) in Academia Sinica. Quantification was made based on a standard curve plotted by a range of known concentrations of H2PO3. Anthocyanin content was measured using a modified protocol (Lange et al. 1971). Shoot tissues were weighed and homogenized in 1 ml of extraction buffer (propanol:HCl:H2O in a ratio of 18:1:81). The sample tubes were immersed in boiled water for 1.5 min and then cooled immediately on ice for at least 3 min before a 10 min centrifugation. Supernatants were collected and measured at 535 and 650 nm. Anthocyanin content was calculated based on the formula Abs(535 nm – 2×650 nm)/FW. Generation of RNAi and complementation lines A 336 bp DNA fragment located within the coding region of MGP1 (AT2G21870) and an approximately 5.8 kb genomic DNA fragment of MGP1 including a 2.1 kb promoter and 1.2 kb downstream of the 3'-UTR were cloned into pCR8/GW/TOPO entry vectors (Invitrogen) for generating RNAi and complementation lines, respectively. The DNA fragments were subcloned into the Gateway destination vectors pB7GWIWG2(I) (Karimi et al. 2002) or pMDC99 (Curtis and Grossniklaus 2003) via LR Clonase enzyme mix (Invitrogen). The constructs were transformed into Agrobacterium tumefaciens GV3101 followed by the transformation into WT or phi1 plants using the floral dip method (Clough and Bent 1998). The putative transgenic RNAi and complementary plants were selected in media containing 100 μg ml–1 BASTA and hygromycin, respectively. Measurement of sugar and ATP For the measurement of sugar, metabolites were extracted from 100 mg of tissues in 1 ml of 80% methanol (containing 12 μg ml–1 ribitol) by sonication at 4°C for 30 min. Ribitol was used as an external control for normalization. After centrifugation, the supernatant was vacuum-dried and subjected to gas chromatography–mass spectrometry (GC-MS) analyses. The amount of the three main sugar species (sucrose, glucose and fructose) was quantified in relation to that of the WT grown under +Pi conditions. ATP was measured according to the protocol of Cho et al. (2016). Tissues were homogenized in 2.3% trichloroacetic acid containing ribitol (20 μg ml–1) as an external control. After centrifugation, the supernatant was neutralized to pH 6.5–7 with KOH and subjected to analysis using a Triple Quadrupole (TSQ) mass spectrometer. Observation of mesophyll cell size Cotyledons and true leaves of 12-day-old seedlings grown under +Phi for 6 d were fixed in 70% ethanol and cleared using chloral hydrate solution (200 g of chloral hydrate, 20 g of glycerol and 50 ml of dH2O) as described (Tsuge et al. 1996). The samples were observed under an upright microscope (Zeiss Imager, Z1) using differential interference contrast. Mito-tracker analysis Nine-day-old seedlings were submerged in Hoagland’s medium supplied with 20 nM TMRE fluorescent dye (Molecular Probes) for 30 min in the dark. The staining in root hairs was observed using an inverted confocal microscope (Zeiss LSM 510 Meta NLO DuoScan) with an objective C-Apochromat ×40/1.2 W, and excitation at 561 nm and emission 610 nm. Starch staining assay Lugol’s solution was prepared by mixing potassium iodide and iodine in a ratio of 2:1. Shoots of 12-day-old seedlings were immersed in ethanol to remove Chl, followed by the addition of Lugol’s solution. Analysis by qRT–PCR Twelve-day-old seedlings were ground and homogenized in liquid nitrogen, and RNA was isolated using RNAzol RT reagent (Molecular Research Center). cDNA was synthesized using oligo(dT) and Moloney murine leukemia virus reverse transcriptase (Invitrogen) from RNA. The quality of cDNA was inspected by the integrity of ACTIN2 (ACT2) RNA. The gene expression level was determined using Power SYBR Green PCR Master Mix kit in the Real Time PCR system (185-5200 CFX Connect, Bio-Rad), following the manufacturer’s instructions. The expression level was calculated by normalization to that of a housekeeping gene, UBIQUITIN 10 (UBQ10). The relative expression was converted into a Z-score and is presented as a heat map. The original data are shown in Supplementary Table S1. The sequences of primers used in this study are listed in Supplementary Table S2. Statistical analysis Statistical analyses were carried out using ANOVA of CoStat version 6.4 (CoHort Software) or the simple Student t-test of Microsoft Excel. Supplementary Data Supplementary data are available at PCP online. Funding This work was supported by the Ministry of Science and Technology of the Republic of China [grant No. NSC94-2311-B-001-057] and Academia Sinica, Taiwan. Acknowledgments We thank the ABRC Metabolomics Core Laboratory for the measurement of sugars and ATP, HFNMRC at Academia Sinica for quantification of Phi, and the Confocal Microscopic Core Facility at Academia Sinica for microscopy images. We are grateful to Kyaw Aung for helping with the TAIL-PCR cloning and Der-Fen Suen (ABRC, Academia Sinica, Taiwan) for helping with the setup of Mito-Tracker analysis. hps1 mutants were kindly provided by Dong Liu (Tsinghua University, China). We also appreciate Der-Fen Suen, Teng-Kuei Huang and Hui-Fen Kuo for constructive discussion and comments on the manuscript. Disclosures The authors have no conflicts of interest to declare. References Alexova R. , Nelson C.J. , Millar A.H. ( 2017 ) Temporal development of the barley leaf metabolic response to Pi limitation . Plant Cell Environ . 40 : 645 – 657 . Google Scholar CrossRef Search ADS PubMed Alford S.R. , Rangarajan P. , Williams P. , Gillaspy G.E. ( 2012 ) myo-Inositol oxygenase is required for responses to low energy conditions in Arabidopsis thaliana . Front. Plant Sci. 3 : 69 . Google Scholar CrossRef Search ADS PubMed Aung K. , Lin S.I. , Wu C.C. , Huang Y.T. , Su C.L. , Chiou T.J. ( 2006 ) pho2, a phosphate overaccumulator, is caused by a nonsense mutation in a microRNA399 target gene . Plant Physiol . 141 : 1000 – 1011 . Google Scholar CrossRef Search ADS PubMed Berkowitz O. , Jost R. , Kollehn D.O. , Fenske R. , Finnegan P.M. , O’Brien P.A. , et al. . ( 2013 ) Acclimation responses of Arabidopsis thaliana to sustained phosphite treatments . J. Exp. Bot . 64 : 1731 – 1743 . Google Scholar CrossRef Search ADS PubMed Bustos R. , Castrillo G. , Linhares F. , Puga M.I. , Rubio V. , Perez-Perez J. , et al. . ( 2010 ) A central regulatory system largely controls transcriptional activation and repression responses to phosphate starvation in Arabidopsis . PLoS Genet. 6 : e1001102 . Google Scholar CrossRef Search ADS PubMed Carrari F. , Nunes-Nesi A. , Gibon Y. , Lytovchenko A. , Loureiro M.E. , Fernie A.R. ( 2003 ) Reduced expression of aconitase results in an enhanced rate of photosynthesis and marked shifts in carbon partitioning in illuminated leaves of wild species tomato . Plant Physiol . 133 : 1322 – 1335 . Google Scholar CrossRef Search ADS PubMed Cheung C.Y.M. , Poolman M.G. , Fell D.A. , Ratcliffe R.G. , Sweetlove L.J. ( 2014 ) A diel flux balance model captures interactions between light and dark metabolism during day–night cycles in C3 and Crassulacean acid metabolism leaves . Plant Physiol . 165 : 917 . Google Scholar CrossRef Search ADS PubMed Chiou T.J. , Aung K. , Lin S.I. , Wu C.C. , Chiang S.F. , Su C.L. ( 2006 ) Regulation of phosphate homeostasis by microRNA in Arabidopsis . Plant Cell 18 : 412 – 421 . Google Scholar CrossRef Search ADS PubMed Chiou T.J. , Lin S.I. ( 2011 ) Signaling network in sensing phosphate availability in plants . Annu. Rev. Plant Biol. 62 : 185 – 206 . Google Scholar CrossRef Search ADS PubMed Cho H.Y. , Wen T.N. , Wang Y.T. , Shih M.C. ( 2016 ) Quantitative phosphoproteomics of protein kinase SnRK1 regulated protein phosphorylation in Arabidopsis under submergence . J. Exp. Bot. 67 : 2745 – 2760 . Google Scholar CrossRef Search ADS PubMed Ciereszko I. , Barbachowska A. ( 2000 ) Sucrose metabolism in leaves and roots of bean (Phaseolus vulgaris L.) during phosphate deficiency . J. Plant Physiol . 156 : 640 – 644 . Google Scholar CrossRef Search ADS Clough S.J. , Bent A.F. ( 1998 ) Floral dip: a simplified method for Agrobacterium-mediated transformation of Arabidopsis thaliana . Plant J . 16 : 735 – 743 . Google Scholar CrossRef Search ADS PubMed Cordoba E. , Aceves-Zamudio D.L. , Hernandez-Bernal A.F. , Ramos-Vega M. , Leon P. ( 2015 ) Sugar regulation of SUGAR TRANSPORTER PROTEIN 1 (STP1) expression in Arabidopsis thaliana . J. Exp. Bot . 66 : 147 – 159 . Google Scholar CrossRef Search ADS PubMed Curtis M.D. , Grossniklaus U. ( 2003 ) A gateway cloning vector set for high-throughput functional analysis of genes in planta . Plant Physiol . 133 : 462 – 469 . Google Scholar CrossRef Search ADS PubMed Danova-Alt R. , Dijkema C. , De Waard P. , Kock M. ( 2008 ) Transport and compartmentation of phosphite in higher plant cells—kinetic and P nuclear magnetic resonance studies . Plant Cell Environ . 31 : 1510 – 1521 . Google Scholar CrossRef Search ADS PubMed Dasgupta K. , Khadilkar A.S. , Sulpice R. , Pant B. , Scheible W.R. , Fisahn J. , et al. . ( 2014 ) Expression of sucrose transporter cDNAs specifically in companion cells enhances phloem loading and long-distance transport of sucrose but leads to an inhibition of growth and the perception of a phosphate limitation . Plant Physiol . 165 : 715 – 731 . Google Scholar CrossRef Search ADS PubMed Duff S.M. , Moorhead G.B. , Lefebvre D.D. , Plaxton W.C. ( 1989 ) Phosphate starvation inducible ‘bypasses’ of adenylate and phosphate dependent glycolytic enzymes in Brassica nigra suspension cells . Plant Physiol . 90 : 1275 – 1278 . Google Scholar CrossRef Search ADS PubMed Ehrenberg B. , Montana V. , Wei M.D. , Wuskell J.P. , Loew L.M. ( 1988 ) Membrane potential can be determined in individual cells from the nernstian distribution of cationic dyes . Biophys. J . 53 : 785 – 794 . Google Scholar CrossRef Search ADS PubMed Fujiki Y. , Yoshikawa Y. , Sato T. , Inada N. , Ito M. , Nishida I. , et al. . ( 2001 ) Dark-inducible genes from Arabidopsis thaliana are associated with leaf senescence and repressed by sugars . Physiol. Plant. 111 : 345 – 352 . Google Scholar CrossRef Search ADS PubMed Geigenberger P. ( 2011 ) Regulation of starch biosynthesis in response to a fluctuating environment . Plant Physiol . 155 : 1566 – 1577 . Google Scholar CrossRef Search ADS PubMed Hamasur B. , Glaser E. ( 1992 ) Plant mitochondrial F0F1 ATP synthase. Identification of the individual subunits and properties of the purified spinach leaf mitochondrial ATP synthase . Eur. J. Biochem. 205 : 409 – 416 . Google Scholar CrossRef Search ADS PubMed Hammond J.P. , White P.J. ( 2011 ) Sugar signaling in root responses to low phosphorus availability . Plant Physiol . 156 : 1033 – 1040 . Google Scholar CrossRef Search ADS PubMed Jost R. , Pharmawati M. , Lapis-Gaza H.R. , Rossig C. , Berkowitz O. , Lambers H. , et al. . ( 2015 ) Differentiating phosphate-dependent and phosphate-independent systemic phosphate-starvation response networks in Arabidopsis thaliana through the application of phosphite . J. Exp. Bot . 66 : 2501 – 2514 . Google Scholar CrossRef Search ADS PubMed Karimi M. , Inze D. , Depicker A. ( 2002 ) GATEWAY vectors for Agrobacterium-mediated plant transformation . Trends Plant Sci . 7 : 193 – 195 . Google Scholar CrossRef Search ADS PubMed Kobayashi K. , Masuda T. , Takamiya K. , Ohta H. ( 2006 ) Membrane lipid alteration during phosphate starvation is regulated by phosphate signaling and auxin/cytokinin cross-talk . Plant J . 47 : 238 – 248 . Google Scholar CrossRef Search ADS PubMed Krapp A. , Stitt M. ( 1995 ) An evaluation of direct and indirect mechanisms for the ‘sink-regulation’ of photosynthesis in spinach: changes in gas-exchange, carbohydrates, metabolites, enzyme activities and steady-state transcript levels after cold-girdling source leaves . Planta 195 : 313 – 323 . Google Scholar CrossRef Search ADS Kuo H.F. , Hsu Y.Y. , Lin W.C. , Chen K.Y. , Munnik T. , Brearley C. , et al. . ( 2018 ) Arabidopsis inositol phosphate kinases, IPK1 and ITPK1, constitute a metabolic pathway in maintaining phosphate homeostasis. bioRxiv 270355, https://doi.org/10.1101/270355. Lai F. , Thacker J. , Li Y. , Doerner P. ( 2007 ) Cell division activity determines the magnitude of phosphate starvation responses in Arabidopsis . Plant J . 50 : 545 – 556 . Google Scholar CrossRef Search ADS PubMed Lange H. , Shropshire W. , Mohr H. ( 1971 ) An analysis of phytochrome-mediated anthocyanin synthesis . Plant Physiol . 47 : 649 – 655 . Google Scholar CrossRef Search ADS PubMed Lee C.P. , Eubel H. , Millar A.H. ( 2010 ) Diurnal changes in mitochondrial function reveal daily optimization of light and dark respiratory metabolism in Arabidopsis . Mol. Cell. Proteomics 9 : 2125 – 2139 . Google Scholar CrossRef Search ADS PubMed Lee C.P. , Eubel H. , O’Toole N. , Millar A.H. ( 2008 ) Heterogeneity of the mitochondrial proteome for photosynthetic and non-photosynthetic Arabidopsis metabolism . Mol. Cell. Proteomics 7 : 1297 – 1316 . Google Scholar CrossRef Search ADS PubMed Lei M. , Liu Y. , Zhang B. , Zhao Y. , Wang X. , Zhou Y. , et al. . ( 2011 ) Genetic and genomic evidence that sucrose is a global regulator of plant responses to phosphate starvation in Arabidopsis . Plant Physiol . 156 : 1116 – 1130 . Google Scholar CrossRef Search ADS PubMed Li L. , Foster C.M. , Gan Q. , Nettleton D. , James M.G. , Myers A.M. , et al. . ( 2009 ) Identification of the novel protein QQS as a component of the starch metabolic network in Arabidopsis leaves . Plant J . 58 : 485 – 498 . Google Scholar CrossRef Search ADS PubMed Li W.Q. , Zhang X.Q. , Xia C. , Deng Y. , Ye D. ( 2010 ) MALE GAMETOPHYTE DEFECTIVE 1, encoding the FAd subunit of mitochondrial F1F0-ATP synthase, is essential for pollen formation in Arabidopsis thaliana . Plant Cell Physiol . 51 : 923 – 935 . Google Scholar CrossRef Search ADS PubMed Lin W.Y. , Huang T.K. , Leong S.J. , Chiou T.J. ( 2014 ) Long-distance call from phosphate: systemic regulation of phosphate starvation responses . J. Exp. Bot . 65 : 1817 – 1827 . Google Scholar CrossRef Search ADS PubMed Liu J. , Samac D.A. , Bucciarelli B. , Allan D.L. , Vance C.P. ( 2005 ) Signaling of phosphorus deficiency-induced gene expression in white lupin requires sugar and phloem transport . Plant J . 41 : 257 – 268 . Google Scholar CrossRef Search ADS PubMed Liu T.Y. , Huang T.K. , Yang S.Y. , Hong Y.T. , Huang S.M. , Wang F.N. , et al. . ( 2016 ) Identification of plant vacuolar transporters mediating phosphate storage . Nat. Commun . 7 : 11095 . Google Scholar CrossRef Search ADS PubMed Lloyd J.C. , Zakhleniuk O.V. ( 2004 ) Responses of primary and secondary metabolism to sugar accumulation revealed by microarray expression analysis of the Arabidopsis mutant, pho3 . J. Exp. Bot . 55 : 1221 – 1230 . Google Scholar CrossRef Search ADS PubMed Lopez-Arredondo D.L. , Leyva-Gonzalez M.A. , Gonzalez-Morales S.I. , Lopez-Bucio J. , Herrera-Estrella L. ( 2014 ) Phosphate nutrition: improving low-phosphate tolerance in crops . Annu. Rev. Plant Biol. 65 : 95 – 123 . Google Scholar CrossRef Search ADS PubMed Marschner H. ( 2012 ) Marschner’s Mineral Nutrition of Higher Plants . Academic Press , London . Mikulska M. , Bomsel J.-L. , Rychter A.M. ( 1998 ) The influence of phosphate deficiency on photosynthesis, respiration and adenine nucleotide pool in bean leaves . Photosynthetica 35 : 79 – 88 . Google Scholar CrossRef Search ADS Misson J. , Raghothama K.G. , Jain A. , Jouhet J. , Block M.A. , Bligny R. , et al. . ( 2005 ) A genome-wide transcriptional analysis using Arabidopsis thaliana Affymetrix gene chips determined plant responses to phosphate deprivation . Proc. Natl. Acad. Sci. USA 102 : 11934 – 11939 . Google Scholar CrossRef Search ADS Morcuende R. , Bari R. , Gibon Y. , Zheng W. , Pant B.D. , Blasing O. , et al. . ( 2007 ) Genome-wide reprogramming of metabolism and regulatory networks of Arabidopsis in response to phosphorus . Plant Cell Environ. 30 : 85 – 112 . Google Scholar CrossRef Search ADS PubMed Müller R. , Morant M. , Jarmer H. , Nilsson L. , Nielsen T.H. ( 2007 ) Genome-wide analysis of the Arabidopsis leaf transcriptome reveals interaction of phosphate and sugar metabolism . Plant Physiol . 143 : 156 – 171 . Google Scholar CrossRef Search ADS PubMed Nagley P. ( 1988 ) Eukaryote membrane genetics: the Fo sector of mitochondrial ATP synthase . Trends Genet. 4 : 46 – 51 . Google Scholar CrossRef Search ADS PubMed Nunes-Nesi A. , Carrari F. , Lytovchenko A. , Smith A.M.O. , Loureiro M.E. , Ratcliffe R.G. , et al. . ( 2005 ) Enhanced photosynthetic performance and growth as a consequence of decreasing mitochondrial malate dehydrogenase activity in transgenic tomato plants . Plant Physiol . 137 : 611 – 622 . Google Scholar CrossRef Search ADS PubMed Pagliarani A. , Nesci S. , Ventrella V. ( 2013 ) Modifiers of the oligomycin sensitivity of the mitochondrial F1F0-ATPase . Mitochondrion 13 : 312 – 319 . Google Scholar CrossRef Search ADS PubMed Pant B.D. , Pant P. , Erban A. , Huhman D. , Kopka J. , Scheible W.R. ( 2015 ) Identification of primary and secondary metabolites with phosphorus status-dependent abundance in Arabidopsis, and of the transcription factor PHR1 as a major regulator of metabolic changes during phosphorus limitation . Plant Cell Environ. 38 : 172 – 187 . Google Scholar CrossRef Search ADS PubMed Plaxton W.C. , Tran H.T. ( 2011 ) Metabolic adaptations of phosphate-starved plants . Plant Physiol. 156 : 1006 – 1015 . Google Scholar CrossRef Search ADS PubMed Poirier Y. , Thoma S. , Somerville C. , Schiefelbein J. ( 1991 ) Mutant of Arabidopsis deficient in xylem loading of phosphate . Plant Physiol . 97 : 1087 – 1093 . Google Scholar CrossRef Search ADS PubMed Puga M.I. , Mateos I. , Charukesi R. , Wang Z. , Franco-Zorrilla J.M. , de Lorenzo L. , et al. . ( 2014 ) SPX1 is a phosphate-dependent inhibitor of Phosphate Starvation Response 1 in Arabidopsis . Proc Natl. Acad. Sci. USA 111 : 14947 – 14952 . Google Scholar CrossRef Search ADS Raghothama K.G. ( 1999 ) Phosphate acquisition . Annu. Rev. Plant Physiol. Plant Mol. Biol. 50 : 665 – 693 . Google Scholar CrossRef Search ADS PubMed Rao I.M. , Terry N. ( 1995 ) Leaf phosphate status, photosynthesis, and carbon partitioning in sugar beet (IV. Changes with time following increased supply of phosphate to low-phosphate plants) . Plant Physiol. 107 : 1313 – 1321 . Google Scholar CrossRef Search ADS PubMed Rubio V. , Bustos R. , Irigoyen M.L. , Cardona-Lopez X. , Rojas-Triana M. , Paz-Ares J. ( 2009 ) Plant hormones and nutrient signaling . Plant Mol. Biol. 69 : 361 – 373 . Google Scholar CrossRef Search ADS PubMed Rubio V. , Linhares F. , Solano R. , Martin A.C. , Iglesias J. , Leyva A. , et al. . ( 2001 ) A conserved MYB transcription factor involved in phosphate starvation signaling both in vascular plants and in unicellular algae . Genes Dev . 15 : 2122 – 2133 . Google Scholar CrossRef Search ADS PubMed Secco D. , Jabnoune M. , Walker H. , Shou H. , Wu P. , Poirier Y. , et al. . ( 2013 ) Spatio-temporal transcript profiling of rice roots and shoots in response to phosphate starvation and recovery . Plant Cell 25 : 4285 – 4304 . Google Scholar CrossRef Search ADS PubMed Singer T. , Burke E. ( 2003 ) High-throughput TAIL-PCR as a tool to identify DNA flanking insertions. In Plant Functional Genomics . Edited by Grotewold E. pp. 241 – 272 . Humana Press , Totowa, NJ . Google Scholar CrossRef Search ADS Smith M.K. , Day D.A. , Whelan J. ( 1994 ) Isolation of a novel soybean gene encoding a mitochondrial ATP synthase subunit . Arch. Biochem. Biophys . 313 : 235 – 240 . Google Scholar CrossRef Search ADS PubMed Solfanelli C. , Poggi A. , Loreti E. , Alpi A. , Perata P. ( 2006 ) Sucrose-specific induction of the anthocyanin biosynthetic pathway in Arabidopsis . Plant Physiol . 140 : 637 – 646 . Google Scholar CrossRef Search ADS PubMed Ticconi C.A. , Delatorre C.A. , Abel S. ( 2001 ) Attenuation of phosphate starvation responses by phosphite in Arabidopsis . Plant Physiol . 127 : 963 – 972 . Google Scholar CrossRef Search ADS PubMed Tsuge T. , Tsukaya H. , Uchimiya H. ( 1996 ) Two independent and polarized processes of cell elongation regulate leaf blade expansion in Arabidopsis thaliana (L.) Heynh . Development 122 : 1589 – 1600 . Google Scholar PubMed Varadarajan D.K. , Karthikeyan A.S. , Matilda P.D. , Raghothama K.G. ( 2002 ) Phosphite, an analog of phosphate, suppresses the coordinated expression of genes under phosphate starvation . Plant Physiol . 129 : 1232 – 1240 . Google Scholar CrossRef Search ADS PubMed Wang Z. , Ruan W. , Shi J. , Zhang L. , Xiang D. , Yang C. , et al. . ( 2014 ) Rice SPX1 and SPX2 inhibit phosphate starvation responses through interacting with PHR2 in a phosphate-dependent manner . Proc. Natl. Acad. Sci. USA 111 : 14953 – 14958 . Google Scholar CrossRef Search ADS Weigel D. , Ahn J.H. , Blazquez M.A. , Borevitz J.O. , Christensen S.K. , Fankhauser C. , et al. . ( 2000 ) Activation tagging in Arabidopsis . Plant Physiol. 122 : 1003 – 1013 . Google Scholar CrossRef Search ADS PubMed Wild R. , Gerasimaite R. , Jung J.Y. , Truffault V. , Pavlovic I. , Schmidt A. , et al. . ( 2016 ) Control of eukaryotic phosphate homeostasis by inositol polyphosphate sensor domains . Science 352 : 986 – 990 . Google Scholar CrossRef Search ADS PubMed Abbreviations Abbreviations GC genomic complementarion InsP inositol polyphosphate MGP1 MALE GAMETOPHYTE DEFECTIVE1 NMR nuclear magnetic resonance Phi phosphite PHR1 PHOSPHATE STARVATION RESPONSE1 PP-InsP inositol pyrophosphate PSi phosphate starvation induced PSR phosphate starvation response PSr phosphate starvation repressed qRT–PCR quantitative reverse transcription–PCR RNAi RNA interference TMRE tetramethylrhodamine ethyl ester UTR untranslated region WT wild type © The Author(s) 2018. Published by Oxford University Press on behalf of Japanese Society of Plant Physiologists. All rights reserved. For permissions, please email: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

Journal

Plant and Cell PhysiologyOxford University Press

Published: Mar 5, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off