INTRODUCTION
In land plants, phosphorus (P) is one of the essential macro-nutrients required to maintain their growth and reproduce seeds for the next generation. P is required in the plant cells as structural constituent of DNA, RNA, phospholipids, and energy coins (ATP, ADP, and AMP) (Hawkesford et al., 2012). In addition, P plays an important role in regulating enzymatic reactions and signaling processes through the protein phosphorylation/dephosphorylation mechanism (Hawkesford et al., 2012). Owing to its role in various physiological functions, the requirement of P in land plants is high followed by the requirements of nitrogen, potassium, calcium, and magnesium among the 17 essential mineral nutrients, and generally, the P content corresponds to approximately 0.2 % of the dry matter of plants (Kirkby, 2012). Land plants can absorb P only in its inorganic phosphate form (Pi) present in the soil through their roots or mycorrhizae (Bieleski, 1973). Land plants possess two kinds of Pi transporters, namely low-affinity and high-affinity transporters (Furihata et al., 1992; Muchhal et al., 1996; Kai et al., 1997; Leggewie et al., 1997; Liu et al., 1998), and their Km values are estimated to be 50-100 µM and 2.5-12.3 µMin planta, respectively, based on the radioactive Pi uptake experiments (Nussaume et al., 2011). These transporter proteins commonly harbor 12 trans-membrane-spanning regions, with a large hydrophilic charged part dividing the protein molecules into two distinct domains containing of six transmembrane regions in its structure. These transporter proteins transport one Pi with two to four protons (H+) into the root against the electrochemical and concentration gradients across the root surface (Ullrich-Eberius et al., 1981, 1984; Sakano, 1990; Mimura, 1999; Raghothama, 1999). In this process, the plasma membrane H+-ATPase contributes to generating the H+ electrochemical gradient and maintaining the cytoplasmic pH for Pi/H+ symport (Ullrich-Eberius et al., 1981, 1984; Mimura, 1999; Raghothama, 1999). The Pi acquisition strategy in land plants is sophisticated and well-regulated from the transcriptional- to the post-translational level (Secco et al., 2012; Gu et al., 2016). Under conditions of low Pi availability, the expression of Pi transporter genes is activated by Pi-starvation responsive transcription factors such as PHR protein-harboring MYB-domain and WRKY-proteins (Rubio et al., 2001; Zhou et al., 2008; Gu et al., 2016). In addition, microRNA399 and microRNA827 support Pi uptake and accumulation in plants at the post-transcriptional step (Fujii et al., 2005; Aung et al.,2006; Bari et al., 2006; Chiou et al., 2006; Franco-Zorrilla et al., 2007; Lin et al., 2010; Secco et al., 2012; Gu et al., 2016). Simultaneously, a negative feedback system to suppress the excess Pi uptake is switched on, such as the increase in class 1 SPX domain-containing proteins, which suppress the expression of Pi-starvation responsive genes, and the non-protein cording geneIPS1, which acts as a target-mimicry of microRNA399 under Pi starvation conditions (Franso-Zorrilla et al., 2007; Wang et al., 2009; Liu et al., 2010; Secco et al., 2012; Puga et al., 2014; Wang et al., 2014). In contrast, under conditions of adequate Pi availability, the Pi transporters are actively degraded by an E2 ubiquitin conjugase-related protein, PHO2, thus leading to the down-regulation of Pi absorption (Delhaize & Randall, 1995; Dong et al., 1998; Aung et al., 2006; Bari et al., 2006). After Pi absorption from the rhizosphere, the Pi homeostasis functions in plant cells (Biddulph et al., 1958; Lee et al., 1990; Mimura, 1995, 1999). When Pi is sufficiently supplied to the cells, Pi is stored in vacuoles to maintain its concentration in the cytosol and organelles such as chloroplasts and mitochondria (Mimura et al., 1990, 1992, 1996; Pratt et al., 2009). On the other hand, under Pi deficiency, Pi is exported from the vacuoles to the cytosol and is preferentially distributed into various organelles for maintaining cellular physiological functions (Mimura et al., 1990, 1992, 1996; Pratt et al., 2009).
As mentioned above, although Pi is essential for plant growth, excess Pi application to plants leads to chlorosis and necrosis in the leaves and finally withering in whole plants. These phenomena have been recognized as P toxicity (Rossiter, 1951; Bhatti & Loneragan, 1970; Clarkson & Scattergood, 1982). To our knowledge, P toxicity has been first reported in 1917 by John W. Shive (Shive, 1918). He examined the effect of different levels of P application on soybean growth by the soil and water culture method, concluding that excess Pi application causes specific injury in soybean leaves (Shive, 1918). Such P toxicity symptoms have been observed in the leaves of various land plant such as rice, wheat, barley and Arabidopsis (Bhatti & Loneragan, 1970; Aung et al., 2006; Chiou et al., 2006; Wang et al., 2009; Liu et al., 2010). In general, P toxicity occurs when the Pi content exceeds approximately 1% of the dry matter of leaves (Bhatti & Loneragan, 1970). P toxicity has been assumed to be caused by zinc (Zn)-deficiency in the leaves because the Zn content decreases depending on the dosage of Pi application in plants (Singh et al., 1988; Zhu et al., 2001; Zhang et al., 2002; Hawkesford et al., 2012). Several hypothetical mechanisms have been proposed to explain P-induced Zn-deficiency in plants. Some of the major mechanisms include: 1. the dilution effect of Zn on the tissue growth stimulated by P application; 2. inhibition of Zn absorption by root under excess P application through mineral interaction in soil (Loneragan et al., 1979); 3. suppression of Zn translocation from root to shoot (Singh et al., 1988); 4. inhibition of Zn acquisition depending on mycorrhizae under excess P application (Ova et al., 2015). In contrast, several studies have shown that P toxicity is observed in plant leaves despite normal Zn accumulation in the leaves (Loneragan et al., 1979; Cakmak & Marschner, 1987; Ova et al., 2015). Therefore, P toxicity cannot be explained only by the suppression of Zn acquisition from soil to leaves. Interestingly, Delhaize and Randall (1995) observed that light intensity directly modulates the P toxicity symptoms in a Pi-accumulating Arabidopsis mutant (pho2 ), and limiting the illumination alleviates P toxicity. This observation implies the involvement of photosynthesis in the occurrence of P toxicity, but the detailed molecular mechanisms of P toxicity have not yet been addressed in land plants.
In this study, we investigated the detailed mechanisms of P toxicity that cause withering in land plants, especially in the leaves. Firstly, we examined the effects of Pi application on photosynthesis, and subsequently, found that high Pi application simultaneously limits photosynthesis and decreases in the scavenging activity of reactive oxygen species (ROS), as a result of the lower availability of metals caused by phytic acid synthesis in leaves. Thereafter we discussed the detailed whole phenomenon of P toxicity in land plants.