Regulation of PPi Levels Through the Vacuolar Membrane H+-Pyrophosphatase
Inorganic pyrophosphate (PPi) is a high-energy compound, although the free energy change of its hydrolysis is approximately 60 % that of ATP. PPi is generated as a by-product of macromolecule biosyntheses in plants, especially in proliferating cells. In living cells, the accumulation of PPi causes the suppression of these metabolic processes and the formation of insoluble Ca–PPi complexes. To avoid these negative effects, the vacuolar H+-pyrophosphatase (H+-PPase) hydrolyzes PPi and pumps H+ across the vacuolar membrane to maintain their acidic state. Importantly, recent studies on fugu5, the H+-PPase loss-of-function mutants, have clearly demonstrated that their phenotype is rescued by the expression of the yeast cytosolic PPase IPP1, which hydrolyzes cytosolic PPi but has no effect on vacuolar acidification, thus strongly suggesting that the role of the H+-PPase lies in the consumption of the inhibitory PPi rather than vacuolar acidification. In this chapter we describe the chemical properties and metabolic role of PPi, in addition to the physiological functions of H+-PPase and soluble PPase revealed by using several mutant lines.
KeywordsArabidopsis H+-PPase PPi homeostasis Proton pump Metabolism Germination Gluconeogenesis Sucrose IPP1 Vacuolar pH Storage lipid mobilization Leaf development Cotyledon Cell proliferation Cell expansion Compensation fugu5 mutants
1 Introduction: PPi in Plant Cells
Living organisms have evolved metabolic networks that enable efficient interplay between energy-producing and energy-consuming reactions for survival and propagation of their offspring. In cells, whether simple or complex, the manner by which energy-rich molecules are made and consumed is basically conserved. In eukaryotes, ATP can be produced by a number of distinct cellular processes that generate energy, most of which take place in the mitochondria and the chloroplasts.
Inorganic pyrophosphate (PPi; diphosphate) is known as a biological high-energy compound because its standard free energy of hydrolysis is comparable to that of ATP. However, the importance of PPi has been obscured by ATP. In living cells, PPi is principally produced by hydrolysis of ATP. For example, biosyntheses of macromolecules, such as DNA, RNA, and proteins, generate PPi as a by-product of these reactions that use ATP or other nucleotide triphosphates such as GTP. Hydrolysis of PPi by cytosolic soluble pyrophosphatases (sPPases) and membrane-bound H+-translocating pyrophosphatases (H+-PPases) produces inorganic phosphate, which is reused for generation of ATP, other nucleotides, and phosphate compounds. Therefore, the relationship of ATP and PPi is that of light and shadow.
The scientific literature on the biological role of PPi published from the 1940s to the end of the year 1999 has been well reviewed and critically evaluated (Heinonen 2001). About 195 known biochemical reactions that produce PPi are catalogued into different categories, most of them being fundamentally important for cell life (Maeshima 2000; Heinonen 2001).
PPi has been proposed to provide a pyrophosphate bond as an energy donor instead of ATP during the origin of life on Earth (Baltcheffsky and Baltcheffsky 1992). Indeed, H+-PPase utilizes the pyrophosphate bond of PPi instead of ATP to perform active transport of protons across the membrane. This PPi-dependent enzymatic activity was first reported for the photosynthetic bacterium Rhodospirillum rubrum in 1967 (Baltcheffsky 1967), and for plant vacuolar membrane fractions in the middle of the 1980s (Chanson et al. 1985; Rea and Poole 1986). Enzymes were then identified in mung bean (Vigna radiata) (Maeshima and Yoshida 1989), R. rubrum (Nore et al. 1991), and other plants (for review, see Maeshima 2000; Gaxiola et al. 2007). H+-PPase has attracted biochemists and plant scientists at viewpoints of novel energy transducing proton pump and as a regulator of PPi level in plant cells.
The concentration of PPi directly affects the chemical equilibrium of macromolecule biosyntheses and other PPi-generating metabolic processes such as fatty acid β-oxidation in cells. Indeed, a high concentration of PPi stops the DNA polymerization reaction. In cells, the PPi level is maintained as a balance between the generation and hydrolysis of PPi. Several biosynthetic reactions, which are catalyzed by, for example, aminoacyl-tRNA synthetase, RNA polymerase, and fatty acyl-CoA synthetase, generate PPi, and some enzymes, including H+-PPase, consume PPi (Heinonen 2001). How is the PPi level maintained by these enzymes? What is the physiological importance of PPi homeostasis in plant cells? PPi is notable as an energy donor comparable to ATP, a regulator of metabolism, and an effector of cotyledon morphogenesis as described later.
Here, we focus on PPi in plants and describe the physicochemical and biochemical properties of PPi, the metabolic processes of generation and hydrolysis of PPi, and the physiological relevance of the maintenance of PPi levels in cells.
2 Physicochemistry of PPi
2.1 Biochemistry of PPi Hydrolysis and Synthesis
PPi can be chemically prepared by heating orthophosphoric acid, thus producing pyrophosphate (pyro from the Greek, meaning “fire”). In living cells, PPi is generated mainly from hydrolysis of ATP and other nucleotide triphosphates, which are used as phosphate donors in the biosyntheses of DNA, RNA, polypeptides, glycogen, and cellulose. Displacement reactions of ATP with R-O (R, chemical moiety or residue) can be categorized into three types: (1) phosphoryl transfer (produces R-O-Pi and ADP), (2) pyrophosphoryl transfer (R-O-PPi and AMP), and (3) adenylyl transfer (R-O-Pi-O-ribose-adenine and PPi). Macromolecule polymerization processes include the above reaction (3) and produce PPi in cells.
The accumulation of crystals of calcium pyrophosphate (Ca2PPi) is known as a rheumatologic disorder in human connective tissues. Indeed, Ca2PPi is practically insoluble in water. This may be one of the reasons why PPi cannot become the main energy currency in living cells instead of ATP. In plant cells, Ca2PPi complexes are negligible under normal physiological conditions, because the concentrations of Ca2+ and PPi in the cytosol are kept at around 0.1 mM and 0.1−0.3 mM, respectively. In plant cells, some PPi may exist in an Mg2PPi complex.
2.2 Quantification of PPi in Cells and Tissues
A few research groups have focused on the PPi concentration, which has been calculated to be at 0.1–0.3 mM in the plant cell cytosol. High concentrations of PPi increase the risk of formation of Ca–PPi complexes, which are insoluble even in cells as described above. PPi has been reported to exist predominantly in the cytosol in plants as mentioned above, and if PPi were to accumulate in vacuoles, it might easily form insoluble Ca2PPi complexes.
In growing hypocotyls of mung bean, the PPi concentration is at the range of 64–84 nmol/g fr wt (Maeshima 1990). The cytosolic PPi level is around 0.2 mM with the assumption that the cytosol occupies about 30 % of the total cell volume, including the cell wall space. In other plants and other tissues, PPi levels are comparatively lower than those of mung bean hypocotyls (for review, see Heinonen 2001). It is expected that the level is higher in growing cells that actively synthesize macromolecules. In etiolated young seedlings of Arabidopsis, the PPi level has been reported to be 51 nmol/g fr wt (Ferjani et al. 2011). If the cytosol occupies 30 % of the seedling cells, the cytosolic PPi concentration is calculated to be 0.17 mM. What effects do cytosolic PPi concentrations have on PPi-utilizing enzymes? H+-PPase can express its full activity only at more than 0.1 mM as demonstrated by patch clamp analysis of mung bean H+-PPase expressed in yeast (Nakanishi et al. 2003). The mung bean enzyme has a Km for the substrate Mg2PPi, an actual form of the substrate, of 4.8 μM (Nakanishi et al. 2003). This value means 18 μM PPi with Mg2+ at 1 mM.
Several biochemical methods have been used to determine the PPi concentration in tissue extracts. Researchers have used three approaches to ensure accurate PPi quantification: (1) stop the biosynthesis and hydrolysis of PPi immediately during or after tissue homogenization, (2) prepare the PPi fraction with high recovery, and (3) measure the PPi concentration with sensitive methods. In most cases, tissues are frozen in liquid nitrogen and homogenized in chilled 80 % ethanol (Ferjani et al. 2011) or 0.45 N perchloric acid (Takeshige and Tazawa 1989; Maeshima 1990) to prevent enzyme reactions. After centrifugation or gel filtration of the tissue extract, the soluble fractions are used for PPi measurements, such as in enzyme-based methods, liquid chromatography, or gas chromatography. Also, H+-PPase can be used for PPi quantification theoretically. A critical point of this method is to remove Ca2+ and Pi from the tissue extracts, because these ions partially inhibit H+-PPase. In addition to recent advances in PPi quantification, we need new methods to determine or monitor PPi levels in real time in living cells to understand the physiological relevance of PPi homeostasis.
3 PPi in Metabolism
3.1 PPi Producing Processes
In 1941, Cori found that PPi accumulated in rat liver extracts, which represents the first report about PPi formation in a biological system (see Cori et al. 1951). Later, in 1948, Kornberg described the first biological reaction that led to PPi production, which he named pyrophosphorolysis with reference to the well-known phosphorolysis (Kornberg 1948). In 1957, he proposed that pyrophosphorylases mostly act in the direction of PPi production favoring the formation of several stable biochemical compounds (Kornberg 1957). He stated later in 1962 that coupling of hydrolysis of PPi by inorganic PPase makes these reactions practically irreversible, a hypothesis that is now widely accepted (Kornberg 1962). Then, a larger number of pyrophosphorylases were reported in the early 1960s that were subdivided into different groups based on the type of reactions they catalyze (Imsande and Handler 1961). It is now generally accepted that most PPi comes from hydrolysis of nucleotide triphosphates (NTPs) in vivo.
In plant cells, PPi is a by-product of biosynthetic processes characteristic of actively growing cells, such as macromolecule biosyntheses and ß-oxidation of fatty acids (Maeshima 2000; Heinonen 2001). Nucleic acid syntheses (polymerization of DNA and RNA), their subsequent modifications such as polyadenylation and capping of mRNA, and pyrimidine and purine nucleotides syntheses de novo are all sources of PPi (Heinonen 2001). The biosyntheses of amino acids, such as histidine and tryptophan, aminoacyl-tRNA (for protein synthesis), and their modifications also produce PPi (Heinonen 2001). Of course, besides all the reactions presented so far, the hydrolysis of NTPs (as stated above), and the synthesis of cyclic nucleotides (cNMP) by the action of adenylate cyclase, liberates PPi in a stoichiometric manner. The activation of fatty acids (FAs, below) by a long chain acyl-CoA synthetase (LACS) is usually coupled to the hydrolysis of ATP to AMP and PPi, and generates acyl-CoAs (Fulda et al. 2004). In plants, syntheses of starch and sucrose (Suc, below) go through an ADPG pyrophosphorylase (AGPase, below) reaction, where one PPi is formed for each glucose unit incorporated (Stitt et al. 1985).
In C4 plants, a very high rate of PPi production is expected to occur. Oxaloacetate, produced in the reaction between PEP and CO2, catalyzed by PEP carboxylase, is then transferred to the bundle sheath cells, where CO2 is released to be used in carbon assimilation (Edwards and Huber 1981). PEP regeneration in the mesophyll chloroplasts by pyruvate orthophosphate dikinase (PPDK) releases one PPi for each CO2 assimilated. However, this high PPi production in the mesophyll chloroplasts is counterbalanced by their high PPase activity, which is at least one order of magnitude lower in C3 plants, and is likely to be an adaptation to C4 photosynthesis (Hatch and Slack 1970).
Gluconeogenesis, on the other hand, is also a “hot spot” for PPi production in plant cytosols in which PPi is generated by the reaction of PPi-dependent phosphofructokinase (PFP, below). The reaction catalyzed by UDPG pyrophosphorylase (UGPase, below) with UTP and Glc-1-P produces UDPG and PPi. Finally, PPi is produced in plastids during the conversion of the Glc-1-P pool to produce ADPG, which is the substrate for starch synthesis, by the action of AGPase.
3.2 PPi-Utilizing Reactions
Plants have four enzymes, PFP, UGPase, PPDK, and H+-PPase, that can use PPi as a phosphoryl donor. PFP is a cytosolic enzyme, widely distributed in the plant kingdom, that catalyzes a readily reversible reaction between fructose-6-phosphate (Fru-6-P) and fructose-1,6-bisphosphate (Fru-1,6-bisP; Stitt et al. 1982; Carnal and Black 1983; Kruger et al. 1983; Kubota and Ashihara 1990). Assays in vitro using PFP purified from potato (Solanum tuberosum) tubers have demonstrated that Pi inhibits the reaction in the direction of Fru-6-P phosphorylation (glycolysis) and that PPi is inhibitory to the opposite reaction (gluconeogenesis; Stitt 1989).
UGPase acts in both directions producing PPi in the synthesis of and consuming it in the mobilization of Suc. In Suc synthesis it acts together with Suc phosphate synthase and Suc phosphatase. On the other hand, during mobilization of Suc, the first enzyme acting is Suc synthase, which in spite of its name does not make Suc in vivo, but rather breaks it down. It is obvious that the mobilization of Suc by Suc synthase and UGPase saves biochemical energy compared to the alternative invertase–hexokinase pathway (Taiz and Zeiger 2010). PPDK is a well-known enzyme of the C4 photosynthetic pathway where it catalyzes the ATP- and Pi-dependent formation of PEP, the primary CO2 acceptor molecule, from pyruvate (Chastain et al. 2002). Finally, the H+-PPase of the tonoplast and other endomembranes transports protons in vivo at the expense of PPi. These H+-PPases are the main focus of this review; therefore, they will be introduced in detail later on.
3.3 Deficiency and Excess of PPi
In all animals and many microbes, the large amount of cytosolic PPi that is generated as a by-product of anabolism in actively proliferating cells is thought to be tightly regulated through immediate hydrolysis by abundant inorganic PPases in a highly exergonic reaction (Kornberg 1962; Josse and Wong 1971; Maeshima 2000; Heinonen 2001). In plants, the plastids contain very high PPase activity (Gross and ap Rees 1986; Weiner et al. 1987) and very low PPi levels (Weiner et al. 1987). In contrast, in the plant cytosol, PPi is not wasted because there is little or no PPase activity (allowing a significant pool of PPi to accumulate), and PPi-dependent enzymes exist that can use PPi instead of ATP to maintain several cellular activities (Weiner et al. 1987).
Treatments of plant tissues in vitro using inhibitors, or the overexpression of some key enzymes that consume PPi, have in extreme cases resulted in a several-fold change of PPi levels. For example, the incubation of spinach leaf discs with imidodiphosphate (IDP), a potential inhibitor of the vacuolar H+-PPase, but not of PFP, raises PPi levels up to fivefold (Neuhaus and Stitt 1991). This led to the conclusion that H+-PPase plays an important role in the removal of PPi from the cytosol.
In contrast, the overexpression of PPase from Escherichia coli in potato and tobacco plants leads to a twofold increase in total PPase activity, which is comparable with that normally found in the chloroplast (Sonnewald 1992). PPi contents were later measured in the same transgenic potato tubers and estimated to be half that of the wild type (Geigenberger et al. 1998). In conclusion, introducing a cytosolic PPase into transgenic plants results in an alteration in photoassimilate partitioning that seems to be shifted towards soluble sugar accumulation (Sonnewald 1992). Finally, transgenic sugarcane clones (Saccharum spp. hybrids) with varying degrees of reduced PFP activity display no visible phenotypical changes, but significant changes are evident in their metabolite profiles. In fact, decreased PFP leads to a reduction of PPi levels in older internodes, consistent with PFP catalyzing a net gluconeogenic (PPi-generating) flux in aged internodes (van der Merwe et al. 2010). In summary, all these studies point to a robust interaction between PPi levels and cellular metabolism. Yet, despite all these efforts, direct evidence of the role of H+-PPase in vivo is still missing, leaving our understanding about this important issue fragmentary and speculative.
4 PPi Hydrolysis by H+-PPase
H+-PPase is a key enzyme that regulates the PPi balance in the cytosol. This has been clearly demonstrated by loss-of-function mutants of H+-PPase as described in the next section (Ferjani et al. 2011). Here, we overview the H+-PPases in plants. In addition to its role as a scavenger of PPi, H+-PPase has a role in proton pumping across the vacuolar membrane of plants as the fourth proton pump. Plants have the mitochondrial and chloroplast F-type ATPase, the plasma membrane-type H+-ATPase (P-type ATPase), and the vacuolar-type H+-ATPase (V-type ATPase) (Gaxiola et al. 2007; Martinoia et al. 2007). The F-type ATPase functions as an ATP synthetase in mitochondria or chloroplasts, and the P-type ATPase functions in plasma membranes to acidify the extracellular space and maintain cytosolic pH. Indeed, the pH optimum of plasma membrane ATPases is relatively acidic at pH 6.5–7.0 (Faraday and Spanswick 1992). This means that ATPases efflux H+ from the cytosol across the plasma membrane when the cytosolic pH decreases. The third H+-ATPase, V-ATPase, acidifies the vacuoles and maintains the cytosolic pH together with H+-PPase. The V-ATPase is also located in the Golgi apparatus, although its main localization is the vacuolar membrane.
The fourth H+ pump, H+-PPase, has several characteristic features: (1) the enzyme consists of a single polypeptide of approximately 80 kDa; (2) its substrate is the unique, simple, and high-energy compound PPi; and (3) H+-PPase is found in a limited number of organisms, such as plants and several photosynthetic bacteria, and is not found in yeast or animal cells (Maeshima 2000). Point three above might be tightly linked to the specific physiological properties of plants, especially the PPi balance in cells, as described in this chapter. For point one above, the membrane topology of mung bean H+-PPase has been investigated (Mimura et al. 2004; Nakanishi et al. 2001), and recently a clear crystal structure of the enzyme has been solved (Lin et al. 2012). The tertiary structure of plant H+-PPase shows a high similarity with Na+-pumping PPase (Na+-PPase) of a hyperthermophilic bacteria Thermotoga maritime, which is a sodium pump.
Most plants have two types of H+-PPases, I and II, which differ in their primary sequence and K+ dependence of enzyme function (Drozdowicz et al. 2000). The type I H+-PPase (At1g15690) requires K+ at more than 30 mM for maximal activity and functions in vacuolar membranes (Maeshima and Yoshida 1989). The type II enzyme (At1g78920 and At1g16780), which does not require K+, is localized in the Golgi apparatus and related vesicles (Segami et al. 2010). In Arabidopsis, there is no difference in the molecular activities of the type I and II enzymes, and the protein amount of the type II is less than 0.2 % of the type I (vacuolar H+-PPase). Thus, only the vacuolar H+-PPase should be considered in understanding PPi homeostasis in plant tissues.
4.2 Soluble PPases
It is impossible to understand the physiological regulation of PPi levels in plant cells without considering soluble PPases (sPPases) as well as H+-PPases. For a long time, it was thought that the activities of sPPases are negligible in plant cells (Weiner et al. 1987). In addition to vacuolar H+-PPase, however, cDNAs of sPPases (EC 184.108.40.206) have been cloned from green algae and land plants (Kieber and Signer 1991; Rojas-Beltrán et al. 1999; Gómez-García et al. 2006). Arabidopsis has six sPPase genes: AtPPa1 (At1g01050), AtPPa2 (At2g18230), AtPPa3 (At2g46860), AtPPa4 (At3g53620), AtPPa5 (At4g01480), and AtPPa6 (At5g09650). These homologues are divided into two groups: prokaryotic (AtPPa1 to AtPPa5) and eukaryotic types (AtPPa6) (Rojas-Beltrán et al. 1999; Gómez-García et al. 2006). Both types share common properties with the vacuolar H+-PPase, such as the requirement of Mg2+ for activity. The two types, however, exhibit differences in molecular size and tertiary structure. For example, the yeast enzyme (eukaryotic type) is larger than the E. coli enzyme (prokaryotic type) and functions as a homodimer (Salminen et al. 2002), while the E. coli enzyme forms a homohexamer (Avaeva et al 1999).
AtPPa6 has a cleavable transit peptide and localizes to the chloroplast stroma (Schulze et al. 2004). For the Chlamydomonas reinhardtii enzyme (Cr-sPPase I, eukaryotic type), its Km for PPi is 10.5 μM and Kcat 315 s−1 (Gómez-García et al. 2006). The Km is very small when compared with H+-PPases and other sPPases. The AtPPa6 gene is induced by Glc, Fru, and Suc in Arabidopsis. Therefore, these plastid-type (eukaryotic type) sPPases are thought to enhance the reaction of AGPase, which generates PPi and stimulates starch biosynthesis (Schulze et al. 2004; Gómez-García et al. 2006).
For prokaryotic types, AtPPa1 and AtPPa4 expressed in E. coli have been used for enzymatic characterization. These two isoforms have similar molecular masses of approximately 24 kDa and have similar kinetic properties: AtPPa1 (Km for PPi, 114 μM; kcat, 9.28 s−1) and AtPPa4 (Km, 101 μM; kcat, 9.20 s−1) (Sancha et al. 2007). AtPPa1 is expressed in most tissues (Sancha et al. 2007), consistent with information from open databases such as Genevestigator (http://www.genevestigator.ethz.ch/). As for the phenotypic properties when AtPPa1 is overexpressed in seeds, a decrease in the amount of stored lipids and an increase in the levels of free sugar and starch have been reported (Meyer et al. 2012). In mutants of AtPPa1 and AtPPa4 knocked-down by RNAi techniques, the stored lipid content is increased.
sPPase is involved in the self-incompatibility of Papaver rhoeas (poppy). In the process of self-incompatibility, a sPPase (Pr-p26.1s) in the cytosol of pollens is inactivated by phosphorylation in response to calcium signaling (de Graaf et al. 2006). As the result of an increase in intracellular PPi levels, macromolecule biosynthesis is inhibited and pollen tube growth is suppressed. This prevents self-fertilization. In Arabidopsis, which is a self-compatible plant, the vacuolar H+-PPase exists in pollen at relatively high levels, but loss-of-function mutant plants show normal self-fertilization (Segami, unpublished data). Therefore, sPPase, which may not be inactivated by phosphorylation, hydrolyzes PPi in the cytosol of H+-PPase knockout plants (AVP1/VHP1;1) even during fertilization. Furthermore, recent reports suggest that sPPases are involved in the recycling of phosphate. A member (AtPPsPase1) of the haloacid dehydrogenase superfamily possesses pyrophosphatase activity and is induced under phosphate-deficient conditions in Arabidopsis (May et al. 2011).
The activity of sPPases in the cytosol was thought to be very low and negligible for a long time. At present, we need detailed information on the total and individual activities of sPPases and on the physiological changes of their activities for understanding their physiological roles and contributions when compared with H+-PPases.
5 PPi Balance in Seedling Development
In this section, we will briefly introduce the isolation of H+-PPase loss-of-function fugu5 mutants of A. thaliana, and the analyses that led to the discovery of previously unrecognized important roles of this enzyme in plant growth and development.
5.1 Compensation and Plant Development
Animal and plant forms display a spectacular diversity that is familiar to everyone; however, how their organ sizes are predetermined has been a long-standing issue in biological research (Conlon and Raff 1999; Tsukaya 2008). Which factors, genetic pathways, and developmental mechanisms promote the progression of organ growth and then restrict it when appropriate organ size is reached? Is size a function of proliferative growth of single cells and their final size, or is it rather under developmental programs acting organ-wide? In order to deepen our knowledge of coordinated organ-size control in multicellular organisms, answers to these questions and other closely related issues are eagerly awaited.
Intrinsic regulatory mechanisms inscribed in the plant’s genome orchestrate developmental programs that when properly executed determine the appropriate size of an organ, though this is significantly affected by various environmental cues, such as light, temperature, water, etc. (Tsukaya 2005). Owing to their determinate fate, leaves have the potential to grow for only a defined period of time; hence, their final size could be interpreted as a simple function of the number and size of component cells. Importantly, an organ-wide control system has been suggested by the so-called “compensation” phenomenon, in which a decrease in cell number, caused by a mutation that compromises cell cycling, triggers excessive cell enlargement (Tsukaya 2002, 2006, 2008; Beemster et al. 2003; Horiguchi et al.2006a; Ferjani et al. 2007, 2008, 2010; Micol 2009; Kawade et al. 2010; Horiguchi and Tsukaya 2011). Based on the above observations, non-cell-autonomous regulation should be assumed as fundamentally important for proper organogenesis. As such, compensation-exhibiting mutants represent excellent models in which the coordinated regulation between organ-size determinants (i.e., cell number and size) is abolished. Cloning the mutated genes and analyzing their functions and contribution to developing organs is a promising approach for solving the long-standing size regulation enigma.
Five decades have passed since the earliest report about compensation (Haber 1962). Since then, this phenomenon has been reported to occur in transgenics in which the cell cycle is inhibited (Hemerly et al. 1995; De Veylder et al. 2001), and in mutants defective in genes that positively regulate cell cycling (Mizukami and Fischer 2000; Kim and Kende 2004; Horiguchi et al. 2005; Ferjani et al. 2007, 2011). Hence, compensation appears to be a universal phenomenon that is not restricted to the model organism Arabidopsis thaliana (Horiguchi and Tsukaya 2011 and Table 1 therein), but also occurs in a wide range of plant species such as Nicotiana tabacum, Oryza sativa, and Antirrhinum majus (Hemerly et al. 1995; Barrôco et al. 2006; Delgado-Benarroch et al. 2009, respectively).
To uncover compensation mechanisms, large-scale screenings have been carried out that have identified several Arabidopsis mutants with leaf size and/or shape defects (Horiguchi et al. 2006b). Among them, five mutants that are collectively called fugu1-fugu5 exhibit typical compensation (Ferjani et al. 2007). For the last decade or so, analyses of the developmental dynamics in a large number of mutants and transgenics that exhibit compensation have been fruitful and have shed light on several key features of the compensation phenomenon (summarized in Horiguchi and Tsukaya 2011).
5.2 Discovery of fugu5 Mutants
AVP1 has a dual molecular function: consumption of PPi and pumping of protons into the vacuoles. Given the broad belief that AVP1 is important as a proton pump, discrepancy among the avp1-1 and fugu5 mutant phenotypes has again raised a fundamental question that awaits a clear answer: What is the biological role of H+-PPase? Which one of the two functions of AVP1 is critical for plant development?
5.3 H+-PPase, a Master Regulator of Cytosolic PPi Homeostasis
In plant cells, the vacuolar H+-ATPase together with the H+-PPase is assumed to be responsible for vacuole acidification, and to cooperatively establish a transmembrane proton gradient as the driving force for the transport of solutes (Maeshima 2001; Gaxiola et al. 2007). Although a major role of H+-PPase as a proton pump has been reported (Li et al. 2005), the importance of the enzymatic hydrolysis of PPi has generated less interest. Therefore, to gain new insights into this fundamental issue, it has been necessary to evaluate the contribution of these two functions separately.
To address this, Ferjani and colleagues have used an elegant approach that consists of checking the effect of specific removal of cytosolic PPi alone on fugu5 mutant phenotypes. For this, they have used the cytosolic PPase IPP1 of Saccharomyces cerevisiae that can only hydrolyze PPi, but does not perform proton pumping (Lundin et al. 1991). In doing so, transgenics in the fugu5 background expressing IPP1 protein (AVP1pro::IPP1, below) should recover their ability to hydrolyze cytosolic PPi, but still lack the vacuolar H+-PPase functioning as a pump. In their construct, the AVP1 promoter was used to express IPP1 in a natural leaf-developmental context, avoiding any ambiguous side effects due to ectopic expression. Amazingly, all of the fugu5 mutant phenotypes recognized so far have been perfectly rescued by complementation with IPP1 (Fig. 2; see also Ferjani et al. 2011 for more details). Moreover, the vacuolar pH in the AVP1pro::IPP1 transgenic lines remains equal to that of fugu5, demonstrating that the slight upshift of vacuolar pH has no direct correlation with the observed fugu5 phenotypes (Ferjani et al. 2011). Taken together, the above empirical evidence indicates that in early seedling development, characterized by an active metabolism, the important role of the H+-PPase lies in the consumption of the inhibitory PPi rather than vacuolar acidification (Ferjani et al. 2011; Bertoni 2011).
Additionally, fugu5 displays a short hypocotyl phenotype in absolute darkness that is rescued by Suc supply or in AVP1pro::IPP1 transgenics where PPi has been specifically removed. The Suc-dependent growth feature of fugu5 phenocopies that of mutants defective in seed storage lipid mobilization (Graham 2008). Whereas triacylglycerol (TAG) lipolysis (Fig. 2b) and FA catabolism via ß-oxidation are unaffected, fugu5 seedlings accumulate a much higher amount of PPi and less Suc than wild-type plants (Ferjani et al. 2011). This data confirms that gluconeogenesis is partially inhibited by the high cytosolic PPi levels in fugu5. Given the broad range of biochemical reactions inhibited by PPi, including gluconeogenesis, this data shows that H+-PPase maintains a normal leaf cell number and size through PPi hydrolysis (Ferjani et al. 2011). The phenotypes of fugu5 are specifically rescued upon Suc supply at the onset of seed imbibition, highlighting the heterotrophic nature of growth during the earliest stages of plant development. The fact that such phenotypes are no longer observed in leaves formed at later stages indicates a transition to autotrophic growth in which the leaves are exclusively nourished from photosynthesized carbohydrates.
Very recently, Meyer and colleagues reported a dominant low-seed-oil mutant (lo15571) of Arabidopsis generated by enhancer tagging in which the conversion of photoassimilates to oil is reduced (Meyer et al. 2012). Immunoblot analysis revealed increased levels of AtPPa1 (At1g01050) protein in developing siliques of lo15571. Interestingly, AtPPa1 encodes a cytosolic sPPase and is one of five closely related genes that share predicted cytosolic localization (Meyer et al. 2012, see also Sect. 4.2). In their scenario, they emphasized that the rate of cytosolic glycolysis of Suc mobilization, as the major route providing precursors for seed oil biosynthesis, is strongly influenced by the expression of endogenous sPPases (i.e., limited PPi pools), although no data on PPi contents have been provided.
Altogether, these novel findings disagree with the decades-old belief that vacuolar acidification through H+-PPase is crucial for plant growth. Rather, the overall metabolic context that produces PPi during germination provides a more elaborate view into the role of the vacuolar H+-PPase. In addition, as described in previous sections, the Arabidopsis genome contains two other genes, AtVHP2;1 and AtVHP2;2, encoding for type II enzymes that are exclusively localized in the Golgi apparatus (Mitsuda et al. 2001; Segami et al. 2010, see also Sect. 4.1). The physiological contribution of the type II H+-PPases in vacuolar acidification and cytosolic PPi hydrolysis is fairly negligible, consistent with the absence of detectable PPi hydrolyzing activity in the total membrane fraction of fugu5 (Hruz et al. 2008; Winter et al. 2007; Segami et al. 2010; Ferjani et al. 2011, 2012). For these reasons, we should consider AVP1 as the master PPase regulating cytosolic PPi levels. Indeed, this important conclusion represents a milestone for future studies related to PPi homeostasis in plants.
6 Conclusions and Future Prospects
The central role of plant H+-PPases in PPi homeostasis has been uncovered, but this discovery does not mark the end of the road, which has been circuitous. In fact, many critical questions remain unanswered. First, does our actual understanding of a major role for vacuolar H+-PPase in PPi homeostasis totally rule out its role as a proton pump? In vha-a2 vha-a3, a mutant lacking tonoplast-specific V-ATPase activity, the vacuolar pH is elevated (pH 6.4) but remains significantly more acidic than the cytosol (pH 7.4; Krebs et al. 2010). Also, a lack of the vacuolar H+-PPase does not significantly affect vacuolar pH (Ferjani et al. 2011). In agreement with these findings, overexpression of the vacuolar H+-PPase in the vha-a2 vha-a3 mutant background did not restore their phenotype, and the triple mutants vha-a2 vha-a3 fugu5-1 are viable (Kriegel et al. ICAR 2012 abstract book). Based on this, a contribution of the endosomal V-ATPase to vacuolar pH has been strongly suggested. Besides, we believe that comparative studies taking advantage of fugu5 mutants and AVP1pro::IPP1 transgenics together with vha-a2 vha-a3 should help resolve this intriguing issue. Also, provided that cold and anoxia are two kinds of environmental stresses that can specifically induce H+-PPase expression, while V-ATPase is severely inhibited, the potential role of H+-PPase as a proton pump might be masked as most phenotypes described here were deduced from studies conducted under standard culture conditions.
Second, while huge strides have been made, the target sites of PPi inhibition starting from TAGs seed reserves and ending up with Suc syntheses de novo have yet to be determined. For this, quantitative high-throughput metabolomics should be very useful. Last, the crystal structure of a V. radiata H+-PPase (VrH+-PPase) complexed with IDP, a non-hydrolyzable PPi analogue, at 2.35 Å resolution has been recently reported (Lin et al. 2012). Given that the amino acid residues involved in PPi binding, hydrolysis, and proton translocation are highly conserved among various plant species, the point mutations in fugu5-1 and fugu5-2 and the relatively short deletion in fugu5-3 (Fig. 1) should be useful for comparative analyses of the structure–function relationship between the H+-PPases in a wide range of other plant species.
This work was supported by Grants-in-Aid from the Japan Society for the Promotion of Science (Grant 16-04179 to A.F.), Grant-in-Aid for Young Scientists (B) (21770036 and 24770039 to A.F.), Scientific Research (23248017 and 24114706, to M.M.), and the Steel Foundation for Environmental Protection Technology (to M.M.). The contribution to the fugu5-related research project of past and present members of Tsukaya laboratory (The University of Tokyo), Horiguchi laboratory (Rikkyo University), Ferjani laboratory (Tokyo Gakugei University), and Maeshima laboratory (Nagoya University) is gratefully acknowledged.
- Barrôco RM, Peres A, Droual AM, De Veylder L, le Nguyen SL, De Wolf J, Mironov V, Peerbolte R, Beemster GT, Inzé D, Broekaert WF, Frankard V (2006) The cyclin-dependent kinase inhibitor Orysa; KRP1 plays an important role in seed development of rice. Plant Physiol 142:1053–1064PubMedCrossRefGoogle Scholar
- Chastain CJ, Fries JP, Vogel JA, Randklev CL, Vossen AP, Dittmer SK, Watkins EE, Fiedler LJ, Wacker SA, Meinhover KC, Sarath G, Chollet R (2002) Pyruvate, orthophosphate dikinase in leaves and chloroplasts of C3 plants undergoes light-/dark-induced reversible phosphorylation. Plant Physiol 128:1368–1378PubMedCrossRefGoogle Scholar
- Edwards GE, Huber SG (1981) The C4 pathway. In: Hatch MD, Boarman NK (eds) The biochemistry of plants. A comprehensive treatise, vol 8. Academic, New York, pp 533–574Google Scholar
- Ferjani A, Yano S, Horiguchi G, Tsukaya H (2008) Control of leaf morphogenesis by long- and short-distance signaling: differentiation of leaves into sun or shade types and compensated cell enlargement. In: Bögre L, Beemster GTS (eds) Plant growth signaling, vol 10, Plant cell monograph series. Springer, Berlin, pp 47–62CrossRefGoogle Scholar
- Geigenberger P, Hajirezaei M, Geiger M, Deiting U, Sonnewald U, Stitt M (1998) Overexpression of pyrophosphatase leads to increased sucrose degradation and starch synthesis, increased activities of enzymes for sucrose-starch interconversions, and increased levels of nucleotides in growing potato tubers. Planta 205:428–437PubMedCrossRefGoogle Scholar
- Imsande J, Handler P (1961) Pyrophosphorylases. In: Boyer PD, Lardy H, Myrbäck K (eds) The enzymes, 2nd edn. Academic, New York, pp 281–304Google Scholar
- Josse J, Wong SCK (1971) Inorganic pyrophosphatase of Escherichia coli. In: Boyer PD (ed) The enzymes, 3rd edn. Academic, New York, pp 499–527Google Scholar
- Kornberg A (1957) Pyrophosphorylases and phosphorylases in biosynthetic reactions. Adv Enzymol 18:191–240Google Scholar
- Kornberg A (1962) On the metabolic significance of phosphorolytic and pyrophosphorolytic reactions. In: Kasha H, Pullman B (eds) Horizons in biochemistry. Academic, New York, pp 251–264Google Scholar
- Kriegel A, Krebs M, Schumacher K (2012) Vacuolar pH – who is in charge? The 23rd international conference on arabidopsis research (ICAR2012) abstract book, pp 195Google Scholar
- Taiz L, Zeiger E (2010) Plant physiology, 5th edn. Sinauer Associates, Sunderland, MAGoogle Scholar