Purine salvage in plants

Phytochemistry 147 (2018) 89e124

Contents lists available at ScienceDirect

Phytochemistry journal homepage: www.elsevier.com/locate/phytochem

Review

Purine salvage in plants Hiroshi Ashihara a, *, Claudio Stasolla b, Tatsuhito Fujimura c, Alan Crozier d a

Department of Biology, Ochanomizu University, Bunkyo-ku, Tokyo, 112-8610, Japan Department of Plant Science, University of Manitoba, Winnipeg, R3T 2N2, Canada c Faculty of Life and Environmental Sciences, University of Tsukuba, Tsukuba, 305-8572, Japan d Department of Nutrition, University of California, Davis, CA, 95616-5270, USA b

a r t i c l e i n f o

a b s t r a c t

Article history: Received 26 June 2017 Received in revised form 10 December 2017 Accepted 14 December 2017

Purine bases and nucleosides are produced by turnover of nucleotides and nucleic acids as well as from some cellular metabolic pathways. Adenosine released from the S-adenosyl-L-methionine cycle is linked to many methyltransferase reactions, such as the biosynthesis of caffeine and glycine betaine. Adenine is produced by the methionine cycles, which is related to other biosynthesis pathways, such those for the production of ethylene, nicotianamine and polyamines. These purine compounds are recycled for nucleotide biosynthesis by so-called “salvage pathways”. However, the salvage pathways are not merely supplementary routes for nucleotide biosynthesis, but have essential functions in many plant processes. In plants, the major salvage enzymes are adenine phosphoribosyltransferase (EC 2.4.2.7) and adenosine kinase (EC 2.7.1.20). AMP produced by these enzymes is converted to ATP and utilised as an energy source as well as for nucleic acid synthesis. Hypoxanthine, guanine, inosine and guanosine are salvaged to IMP and GMP by hypoxanthine/guanine phosphoribosyltransferase (EC 2.4.2.8) and inosine/guanosine kinase (EC 2.7.1.73). In contrast to de novo purine nucleotide biosynthesis, synthesis by the salvage pathways is extremely favourable, energetically, for cells. In addition, operation of the salvage pathway reduces the intracellular levels of purine bases and nucleosides which inhibit other metabolic reactions. The purine salvage enzymes also catalyse the respective formation of cytokinin ribotides, from cytokinin bases, and cytokinin ribosides. Since cytokinin bases are the active form of cytokinin hormones, these enzymes act to maintain homeostasis of cellular cytokinin bioactivity. This article summarises current knowledge of purine salvage pathways and their possible function in plants and purine salvage activities associated with various physiological phenomena are reviewed. © 2017 Elsevier Ltd. All rights reserved.

This article dedicated to our friend and colleague at the University of Calgary, Professor Trevor A. Thorpe. Keywords: Review Purine metabolism Purine nucleotides Salvage enzymes Phosphoribosyltransferases Purine nucleoside kinases Nucleosidases Nucleotidases Caffeine biosynthesis Cytokinin interconversion

1. Introduction Purine nucleotides, such as ATP and GTP, are important metabolites as the energy source and building blocks of nucleic acids (Moffatt and Ashihara, 2002; Stasolla et al., 2003; Zrenner and Ashihara, 2011; Zrenner et al., 2006). In plants, purine nucleotides also act as substrates for the biosynthesis of nicotinamide adenine dinucleotides (NAD and NADP) (Ashihara et al., 2015; Zrenner and Ashihara, 2011), purine alkaloids (Ashihara et al., 2008b, 2017) and cytokinins (Ashihara et al., 2013b; Hirose et al., 2008). In most organisms, with the exception of parasites, purine nucleotides are synthesized de novo (Berens et al., 1995). In addition, purine nucleotides are synthesized by salvage pathways, recycling preformed

* Corresponding author. E-mail address: [email protected] (H. Ashihara). https://doi.org/10.1016/j.phytochem.2017.12.008 0031-9422/© 2017 Elsevier Ltd. All rights reserved.

purine bases and nucleosides for nucleotide synthesis (Moffatt and Ashihara, 2002). The salvage pathways are not supplementary routes leading to the biosynthesis of purine nucleotides. Although net formation of purine nucleotides is performed by the de novo pathway, rapid turnover of nucleic acids, especially RNA, is required for nucleotide production by the salvage pathways. Reduced salvage activity inhibits the normal growth of plants and other organisms (see Section 8.1.2). In mammals, genetic deficiency of a purine salvage enzyme, hypoxanthine/guanine phosphoribosyltransferase (EC 2.4.2.8), causes the Lesch-Nyhan syndrome (Nyhan, 1997). In plants, adenine phosphoribosyltransferase (EC 2.4.2.7) appears to be a more important enzyme in purine salvage than hypoxanthine/ guanine phosphoribosyltransferase. In addition to producing adenine nucleotides for energy metabolism and nucleic acid biosynthesis (Stasolla et al., 2003; Zrenner and Ashihara, 2011; Zrenner et al., 2006), adenine phosphoribosyltransferase is

90

H. Ashihara et al. / Phytochemistry 147 (2018) 89e124

involved in the removal of adenine and adenosine, which act as inhibitors of several reactions in cells (Moffatt and Ashihara, 2002), as well as being utilised for caffeine biosynthesis (Ashihara and Crozier, 2001), and activation and inactivation of cytokinin functions (Auer, 2002; Mok and Mok, 2001). Furthermore, deficiency of this enzyme causes male-sterility (Gaillard et al., 1998). Thus, purine salvage is not only the energy-saving route of nucleotide biosynthesis, but is an essential pathway in plants. Since purine salvage is a topic of interest in medical science, including parasitology (Freitas-Mesquita and Meyer-Fernandes, 2017) and neurochemistry (Balestri et al., 2007), many reviews have been published. However, there is currently no review on purine salvage in plants. This is the first comprehensive article on this topic in plants which covers purine salvage literature published up to September, 2017. This review will first describe the metabolic fate of purine bases and nucleosides (Section 2.1) and enzyme activity profiles in plant cells and tissues (Section 2.2), and then summarize the purine salvage enzymes (Section 3). After a brief summary of pathways for the supply of substrates for the salvage enzymes (Section 4), enzymes involved in substrate production for purine salvage enzymes (Section 5) and cellular concentration of substrates and products of purine salvage enzymes are summarised (Section 6). The transport mechanism of purine bases and nucleosides is then described (Section 7). The second section of this review covers physiological aspects of purine salvage in plants (Section 8). This includes information on changes in purine salvage activity accompanied by physiological events (Section 8.1), the involvement of purine salvage enzymes in other aspects of metabolism, such as caffeine biosynthesis (Section 8.2), interconversion of cytokinins (Section 8.3) and various phenomena including male sterility (Section 8.4), pathogen responses (Section 8.5) and gravitropism (Section 8.6). Finally, recent omics studies which may reveal new functions of purine salvage in plants are discussed (Section 8.7). This article aims at describing the current knowledge of purine salvage studies at the biochemical, molecular biological and physiological levels, together with the historical survey of this topic. There are some reviews on general plant nucleotide metabolism including biosynthesis and degradation pathways of purine nucleotides, (Moffatt and Ashihara, 2002; Ross, 1981; Stasolla et al., 2003; Wagner and Backer, 1992; Wasternack, 1982; Zrenner and Ashihara, 2011; Zrenner et al., 2006). For related pathways, recent reviews are available on caffeine biosynthesis (Ashihara et al., 2013a, b; 2017), cytokinin metabolism (Hluska et al., 2016; Sakakibara, 2006, 2010), ethylene biosynthesis (Lin et al., 2009) and S-adenosyl-L-methionine (SAM) metabolism (Roje, 2006). Except for recombinant enzymes derived from special genes, enzyme names recommended by the IUBMB are used throughout the article. 2. Metabolism of purine bases and nucleosides in plants 2.1. Metabolic fate of purine bases and nucleosides Purine bases and nucleosides are metabolised by three different routes (i) salvage pathways, (ii) catabolic (degradation) pathways, and (iii) in some plant species, purine alkaloid biosynthesis. The pathways that have been established in plants are summarised in Fig. 1 (Ashihara and Crozier, 1999; Moffatt and Ashihara, 2002; Stasolla et al., 2003; Zrenner and Ashihara, 2011). The metabolic fate of purine bases and nucleosides has been investigated using radiolabelled purine precursors. 14C-Labelled purine bases and nucleosides are taken up rapidly and immediately metabolised in plant cells and tissues. Purine salvage activity is usually estimated

from a summation of the radioactivity incorporated into nucleotides and nucleic acids. A portion of purines is catabolised by the conventional purine catabolic pathway and radioactivity from 14Clabelled purine bases and nucleosides is incorporated into the purine catabolites, allantoin and allantoic acid as well as CO2. In some plant species such as Camellia sinensis (tea) and Coffea arabica (coffee), purine bases and nucleosides are utilised for the biosynthesis of purine alkaloids, (Ashihara and Crozier, 1999; Ashihara et al., 2008a, b, 2017). Profiles of the metabolic fate of a series of purine bases and nucleosides have been comprehensively investigated in A. thaliana cells (Yin et al., 2014), potato tubers (Katahira and Ashihara, 2006a), leaves and roots of tea seedlings (Deng and Ashihara, 2010) and cacao leaves (Koyama et al., 2003). The metabolic fate of [8-14C]adenine, [8-14C]hypoxanthine, [8-14C]guanine and [8-14C]xanthine in the exponential growing phase of A. thaliana cells (Yin et al., 2014), disks of young developing potato tubers (Katahira and Ashihara, 2006a), young leaves of tea (Deng and Ashihara, 2010) and cacao (Koyama et al., 2003) are shown in Fig. 2AeD. Of the four purine bases investigated, adenine is the favoured precursor for purine salvage being converted to AMP (step 3 in Fig. 1) which is converted to ATP by subsequent phosphorylation reaction steps and then incorporated into RNA by RNA polymerase. The second most efficient salvaged purine base precursor is guanine. Guanine is converted to GMP (step 7) and then used for GTP and RNA synthesis. The conversion of AMP to GMP occurs in plants, but conversion of GMP to AMP is limited. In several species radioactivity from [8-14C]adenine is distributed in adenine residues (~70%), and the remainder (~30%) is incorporated in guanine residues, whereas radioactivity from [8-14C]guanine is incorporated principally into guanine residues (>95%) (Deng and Ashihara, 2010; Katahira and Ashihara, 2006a; Yin et al., 2014). This suggests that some AMP produced by adenine salvage is converted to IMP by AMP deaminase (EC 3.5.4.6) and then converted to GMP via XMP (steps 11e13 in Fig. 1). Conversion of GMP to AMP is negligible, because GMP reductase (EC 1.7.1.7, step 14 in Fig. 1) is absent in plants (Stasolla et al., 2003). Small amounts of radioactivity from [8-14C]hypoxanthine is recovered in the guanine residues of RNA. This implies that IMP derived from hypoxanthine is utilised preferentially in GMP synthesis, and conversion to AMP is restricted. In potato tubers, adenine and guanine are salvaged predominantly for nucleotide synthesis, but <2% of these purine bases is catabolised (Katahira and Ashihara, 2006a) (Fig. 2B). Since deaminases for adenine and guanine are absent in plants (Stasolla et al., 2003; Katahira and Ashihara, 2006a), adenine appears to be converted to AMP (step 3), which is then deaminated to IMP (step 11) and converted to xanthine via inosine and hypoxanthine (steps 10, 8 and 17a). Xanthine is catabolised via the conventional purine catabolic pathway (step 17e22). In the case of guanine, it is first converted to GMP (step 7) and then catabolised via a GMP / guanosine / xanthosine / xanthine pathway (steps 10, 15 and 9). The salvage activity of hypoxanthine in plants is usually lower than that of adenine and guanine, although relatively high hypoxanthine salvage is observed in species such as cacao (Fig. 2D). Since hypoxanthine is a substrate for both purine salvage (hypoxanthine/guanine phosphoribosyltransferase, step 7) and purine catabolism enzymes (xanthine dehydrogenase, step 17), the relative activity of purine salvage is reduced in plant materials where the purine degradation is pronounced. In purine alkaloid-forming species, xanthosine is a key substrate for the biosynthesis of theobromine and caffeine (steps 23e25). In contrast to other purine bases, xanthine is degraded extensively (steps 17e22) and only limited amounts are utilised for purine alkaloid synthesis in tea and cacao leaves (Fig. 2C and D). The metabolic fate of [8-14C]adenosine, [8-14C]inosine, [8-14C]

H. Ashihara et al. / Phytochemistry 147 (2018) 89e124 91

Fig. 1. Metabolic pathways of purine compounds in plants inferred from the fate of exogenously supplied [8-14C]purine nucleosides and nucleobases. The boxed structure shows the purine numbering system. Reactions and enzymes involved in purine salvage, degradation and alkaloid synthesis are shown with red, blue and green arrows and numbers, respectively. Numbers indicate the reaction steps catalysed by the enzymes (EC numbers) shown below. 1, Adenosine kinase (EC 2.7.1.20); 2, adenosine nucleosidase (3.2.2.7); 3, adenine phosphoribosyltransferase (2.4.2.7); 4, deoxyadenosine kinase (2.7.1.76); 5, deoxyguanosine kinase (2.7.1.113); 6, inosine/guanosine kinase (2.7.1.73); 7, hypoxanthine/guanine phosphoribosyltransferase (2.4.2.8); 8, inosine/guanosine nucleosidase (3.2.2.2); 9, purine (xanthosine) nucleosidase (3.2.2.1); 10, 50 -nucleotidase (3.1.3.5) or various phosphatases (3.1.3.-); 11, AMP deaminase (3.5.4.6); 12, IMP dehydrogenase (1.1.1.205); 13, GMP synthase (6.3.5.2); 14, GMP reductase (1.7.1.7) (this enzyme has not yet been discovered in any plants); 15, guanosine deaminase (3.5.4.15); 16, guanine deaminase (3.5.4.3); 17a,b, xanthine dehydrogenase (1.1.1.204); 18, uricase (1.7.3.3); 19, allantoinase (3.5.2.5); 20, allantoicase (3.5.3.4); 21, ureidoglycolate lyase (4.3.2.3); 22, urease (3.5.1.5); 23, 7-methylxanthosine synthase (2.1.1.158); 24, N-methylnucleosidase (3.2.2.25); 25a¡d, caffeine synthase (2.1.1.160); 25a,c, theobromine synthase (2.1.1.159). Adapted from Deng and Ashihara (2010). (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)

92

H. Ashihara et al. / Phytochemistry 147 (2018) 89e124

Fig. 2. The metabolic fate of exogenously supplied adenine, hypoxanthine, guanine and xanthine in A) suspension cultured cells of Arabidopsis thaliana cells (Yin et al., 2014), B) growing potato tubers (Katahira and Ashihara, 2006a), C) tea leaves (Deng and Ashihara, 2010) and D) cacao leaves (Koyama et al., 2003). Incubation time was 4 h for A. thaliana cells and potato tubers, 10 h for tea leaves 18 h for cacao leaves. The rate of incorporation into the salvage products, catabolites and purine alkaloids are shown as percentage of total radioactivity taken up by the samples. A, adenine; H, hypoxanthine, G, guanine, X, xanthine.

guanosine and [8-14C]xanthosine in cultured cells of A. thaliana (Yin et al., 2014), slices of potato tubers (Katahira and Ashihara, 2006a), young leaves of tea (Deng and Ashihara, 2010) and cacao (Koyama et al., 2003) are shown in Fig. 3AeD. The fate of [8-14C]deoxyadenosine and [8-14C]deoxyguanosine in potato tubers and tea leaves are shown in Fig. 3B and C. Except for cacao leaves, the order of the salvage activity is adenosine > guanosine > inosine (Fig. 3AeC). In cacao leaves, similar levels of inosine and guanosine salvage are found (Fig. 3D). No xanthosine salvage activity has been detected in any of the

plant materials shown in Fig. 3, including cultured cells of A. thaliana. However, conflicting results, in the form of active xanthosine salvage activity have been reported in intact A. thaliana seedlings by Riegler et al. (2011) who found a significant amount of radioactivity (14% of the total radioactivity) from [8-14C]xanthosine was incorporated into RNA. The reason for this discrepancy is unclear, but it is probably due to differences in experimental procedure. It is possible that the RNA fraction by Riegler et al. (2011) was contaminated with acid-soluble metabolites due to inadequate washing (Ashihara, 2012). In addition, it is noteworthy that

H. Ashihara et al. / Phytochemistry 147 (2018) 89e124

93

Fig. 3. The metabolic fate of exogenously supplied adenosine, inosine, guanosine, xanthosine, deoxyadenosine and deoxyguanosine in (A) suspension cultured cells of Arabidopsis thaliana cells (Yin et al., 2014), (B) growing potato tubers (Katahira and Ashihara, 2006a), (C) tea leaves (Deng and Ashihara, 2010) and (D) cacao leaves (Koyama et al., 2003). Incubation time was 4 h for A. thaliana cells and potato tubers, 10 h for tea leaves 18 h for cacao leaves. The rate of incorporation into the salvage products, catabolites and purine alkaloids are shown as percentage of total radioactivity taken up by the samples. AR, adenosine; IR, inosine; GR, guanosine; XR, xanthosine; AdR, deoxyadenosine and GdR, deoxyguanosine.

no in vitro xanthosine kinase activity has been reported in any plant extracts (Table 1). Adenosine is converted to AMP by adenosine kinase (EC 2.7.1.20, step 1 in Fig. 1) and inosine and guanosine are salvaged by inosine/ guanosine kinase (EC 2.7.1.73, step 6). When [8-14C]adenosine was used for the feeding experiments, approximately 70% of the radioactivity in RNA was recovered as adenine residues, and 30% as guanine residues, but the radioactivity from [8-14C]guanosine and [8-14C]inosine was recovered entirely as guanine residues (Deng and Ashihara, 2010; Katahira and Ashihara, 2006a; Yin et al.,

2014). These patterns are very similar to the results of both [8-14C]adenine and [8-14C]guanine discussed above. Since there are no reports of adenosine deaminase in most plants (Stasolla et al., 2003) except for very low activity in roots and foliage of alfalfa plants (Edwards, 1996), catabolism appears to be initiated mainly from AMP as outlined above. Inosine is converted to hypoxanthine by inosine/guanosine nucleosidase (EC 3.2.2.2, step 8) and guanosine is converted to xanthosine by guanosine deaminase (EC 3.5.4.15, step 15). These metabolites are converted to xanthine (step 17 and 9, respectively) and enter the conventional

94

H. Ashihara et al. / Phytochemistry 147 (2018) 89e124

Table 1 Profile of activity of enzymes involved in purine salvage in growing potato tubers (Katahira and Ashihara, 2006a) and young tea leaves (Deng and Ashihara, 2010). Enzyme activity is expressed as pkat (mg protein)1. Enzyme (EC number) (Abbreviation)

Reaction (Step number in Fig. 1) (*14C-Labelled substrates and products)

Potato tubers

Tea leaves

Adenine phosphoribosyltransferase (2.4.2.7) (APRT) Hypoxanthine/guanine phosphoribosyltransferase (2.4.2.8) (HGPRT)

Adenine* þ PRPP / AMP* þ PPi (3) Hypoxanthine* þ PRPP / IMP* þ PPi (7) Guanine* þ PRPP / GMP* þ PPi (7) Xanthine* þ PRPP / XMP* þ PPi Adenine* þ Ribose-1-phosphate / Adenosine* þ Pi Hypoxanthine* þ Ribose-1-phosphate / Inosine* þ Pi Guanine* þ Ribose-1-phosphate / Guanosine* þ Pi Xanthine* þ Ribose-1-phosphate / Xanthosine* þ Pi Adenosine* þ ATP / AMP* þ ADP (1) Deoxyadenosine* þ ATP / dAMP* þ ADP (4) Inosine* þ ATP / IMP* þ ADP (6) Guanosine* þ ATP / GMP* þ ADP (6) Xanthosine* þ ATP / XMP* þ ADP Deoxyguanosine* þ ATP / dGMP* þ ADP (5) Adenosine* þ AMP / AMP* þ Adenosine (1) Inosine* þ AMP / IMP* þ Adenosine (6) Guanosine* þ AMP / GMP* þ Adenosine (7) Xanthosine* þ AMP / XMP* þ Adenosine Xanthosine* þ IMP / XMP* þ Inosine Deoxyadenosine* þ AMP / dAMP* þ Adenosine (4) Deoxyguanosine* þ AMP / dGMP* þ Adenosine (5) Adenosine* / Adenine* þ Ribose (2) Deoxyadenosine* / Adenine* þ Deoxyribose (2) Inosine* / Hypoxanthine* þ Ribose (8) Guanosine* / Guanine* þ Ribose (8) Deoxyguanosine* / Guanine* þ Deoxyribose (8) Xanthosine* / Xanthine* þ Ribose (9)

101.4 6.4 4.8 nd nd nd nd nd 76.3 89.8 16.8 15.5 nd 53.8 39.7 19.8 20.9 nd nd 89.8 55.0 9.5 13.2 1.1 0.1 <0.1 8.7

96.9 0.79 1.26 0.22 nd nd nd nd 34.7 e 0.37 0.13 nd e e e

Xanthine phosphoribosyltransferase (2.4.2.22) (XPRT) Purine-nucleoside phosphorylase (2.4.2.1) (PNP)

Adenosine kinase (2.7.1.20) (AK) Deoxyadenosine kinase (2.7.1.76) Inosine/guanosine kinase (2.7.1.73) (IGK)

Deoxyguanosine kinase (2.7.1.113) Nucleoside phosphotransferase (2.7.1.77) (NPT)

Adenosine nucleosidase (3.2.2.7) (ARN) Inosine/guanosine nucleosidase (3.2.2.2)

Purine nucleosidase (3.2.2.1)

e e e e 84.9 e 9.21 8.53 e 29.1

nd, not detected. , not determined.

purine catabolism pathway (steps 17e22). In tea and cacao leaves, adenosine, inosine and guanosine are utilised for purine alkaloid biosynthesis via xanthosine (steps 23e25 in Fig. 1). Deoxyadenosine and deoxyguanosine are readily salvaged to deoxyAMP and deoxyGMP which are further metabolised to deoxynucleoside triphosphates and utilised for nucleic acid synthesis. In potato tubers, almost all radioactivity from these deoxynucleosides is found in the salvage products. In tea leaves, a fraction of the deoxynucleosides may be converted to adenine and guanine by adenosine nucleosidase (step 2) and inosine/guanosine nucleosidase (step 8) respectively, and then utilised for the synthesis of purine alkaloids (Fig. 3C). Changes in the metabolic fate of 14Clabelled purines have been linked to physiological events, including seed germination, cell growth, embryogenesis and organogenesis (see Section 8.1). 2.2. Profile of activity of purine salvage enzymes In most organisms including plants, purine nucleotides are synthesized de novo from glycine, glutamine, aspartate, 5phosphoribosyl-1-pyrophosphate (PRPP), 10-formyl tetrahydrofolate and CO2 (Henderson and Paterson, 1973) (Fig. 4). In contrast to de novo synthesis, which consists of energy consuming multistep reactions, purine nucleotides can be synthesized by more simple savage pathways when purine-skeleton containing compounds are available. In salvage pathways, only one mole of PRPP or ATP is utilised for the synthesis of one mole of these nucleotides from a purine base or a purine nucleoside. In plant cells, purine bases and nucleosides originate from the intercellular breakdown of nucleic acids and nucleotides and other reactions which release purine bases and nucleosides. In addition, they are transported from other tissues, such as storage organs and senescent leaves. It is also feasible that they may be taken up by the roots from the surrounding soil where the purines may be breakdown products of

fallen leaves, dead insects and bacteria. There are many reports on the salvage of purine bases, but profiles of activities of the enzymes involved have only been investigated in potato tubers (Katahira and Ashihara, 2006a) and young tea leaves (Deng and Ashihara, 2010) (Table 1). Two distinct phosphoribosyltransferases, adenine phosphoribosyltransferase (EC 2.4.2.7) and hypoxanthine/guanine phosphoribosyltransferase (EC 2.4.2.8) contribute to purine base salvage in plants. In mammalian tissues, hypoxanthine/guanine phosphoribosyltransferase has a central role in the regulation of purine nucleotide biosynthesis (Adams and Harkness, 1976). However, in plant tissues, activity of this enzyme is much lower (~6%) than that of adenine phosphoribosyltransferase. Theoretically, purine bases can also be converted to their respective purine nucleosides using ribose-1-phosphate by purine nucleoside phosphorylase (EC 2.4.2.1) which occurs in bacteria and animals (Bzowska et al., 2000) but not plant tissues (Table 1). Purine nucleosides, namely, adenosine, inosine, guanosine, deoxyadenosine and deoxyguanosine are converted to their respective nucleoside monophosphates. There are four distinct nucleoside kinases, adenosine kinase (EC 2.7.1.20), inosine/guanosine kinase (EC 2.7.1.73), deoxyadenosine kinase (EC 2.7.1.76) and deoxyguanosine kinase (EC 2.7.1.113) which catalyse the conversion of purine nucleoside to 50 -nucleoside monophosphates in plant tissues (Table 1). Nucleoside phosphotransferase (EC 2.7.1.77) can also participate in purine nucleoside salvage (Table 1). Since activity of adenosine nucleosidase (EC 3.2.2.7), inosine/ guanosine nucleosidase (EC 3.2.2.2), and purine nucleosidase (EC 3.2.2.1) occur in potato tubers and tea leaves (Table 1), simple hydrolysis of purine nucleosides to purine bases can occur in plants. Purine base formation from purine nucleosides catalysed by phosphorylase, or by the reverse reaction of phosphoribosyltransferases, may not participate in the hydrolysis in plants (Stasolla et al., 2003). Although salvage pathways of xanthine and

H. Ashihara et al. / Phytochemistry 147 (2018) 89e124

95

Fig. 4. The pathway of de novo purine nucleotide biosynthesis in plants. Numbers indicate the reaction steps catalysed by the enzymes (EC numbers) shown below. 1, PRPP amidotransferase (2.4.2.14); 2, GAR synthetase (6.3.4.13); 3, GAR formyl transferase (2.1.2.2); 4, FGAM synthetase (6.3.5.3); 5, AIR synthetase (6.3.3.1); 6, AIR carboxylase (4.1.1.21); 7, SAICAR synthetase (6.3.2.6); 8, adenylosuccinate lyase (4.3.2.2); 9, AICAR formyl transferase (2.1.2.3); 10, IMP cyclohydrolase (3.5.4.10); 11, SAMP synthetase (6.3.4.4); 12, adenylosuccinate lyase (4.3.2.2); 13, IMP dehydrogenase (1.1.1.205); 14, GMP synthetase (6.3.5.2). Metabolites: PRA, 5-phosphoribosyl amine; GAR, glycineamide ribonucleotide; FGAR, formylglycineamide ribonucleotide; FGRAM, formylglycine amidine ribonucleotide; AIR, 5-aminoimidazole ribonucleotide; CAIR, 5-aminoimidazole 4-carboxylate ribonucleotide; SCAIR, 5-aminoimidazole-4-N-succinocarboxyamide ribonucleotide; AICAR, 5-aminoimidazole-4-carboxyamide ribonucleotide; FAICAR, 5-formamidoimidazole-4-carboxyamide ribonucleotide; SAMP, adenylosuccinate.

96

H. Ashihara et al. / Phytochemistry 147 (2018) 89e124

xanthosine are not functional in many plant tissues (Figs. 2 and 3), a low phosphoribosyltransferase activity which catalyses the formation of XMP from xanthine occurs in tea leaf extracts (Table 1). However it is unclear if this activity is due to a side reaction of hypoxanthine/guanine phosphoribosyltransferase or a distinct xanthine phosphoribosyltransferase (see Section 3.2). 3. Enzymes of purine salvage 3.1. Purine phosphoribosyltransferases 3.1.1. Adenine phosphoribosyltransferase Adenine phosphoribosyltransferase (AMP: diphosphate phospho-D-ribosyltransferase EC 2.4.2.7) catalyses the formation of AMP from adenine and 5-phosphoribosyl-1-pyrophosphate (PRPP). This enzyme was first isolated from yeast (Kornberg et al., 1955), and subsequently it has been found in bacteria, animals and plants (Henderson and Paterson, 1973; Stasolla et al., 2003). Biochemical properties of the native and recombinant purine phosphoribosyltransferase from plants are listed in Table 2. Studies on adenine phosphoribosyltransferase with various plants suggest that the enzyme catalyses the conversion of adenine to AMP with PRPP, and the Km values of adenine are usually low (1e10 mM). In contrast, the Km values for PRPP can fluctuate between 1 and 300 mM. The enzyme also catalyses the conversion of cytokinin bases, isopentenyladenine, benzyladenine and transzeatin, to their respective cytokinin ribotides. Unlike the low Km values for adenine, the values for cytokinins are extremely high (50e300 mM). The endogenous concentrations of cytokinins are low (see Section 6.3) and probably never reach the levels equivalent to the Km for cytokinins. Hence it is proposed that although

cytokinins could be metabolised by adenine phosphoribosyltransferase, the physiological role of adenine phosphoribosyltransferase in cytokinin metabolism is negligible (Lee and Moffatt, 1993). Adenine phosphoribosyltransferase from A. thaliana purified to apparent homogeneity was obtained by Lee and Moffatt (1993). The enzyme is a homodimer of ~54 kDa. Cloning and characterization of A. thaliana adenine phosphoribosyltransferase reported by Moffatt et al. (1992). An intact cDNA of 729 nucleotides was obtained which predicts a protein of 27 KDa. The deduced amino acid sequence was compared with those of other adenine phosphoribosyltransferases (APT) and shown to be most similar to the E. coli protein. There are five sequences annotated as coding for APT or APT-like enzymes in the A. thaliana genome. These have been designated APT15 (Moffatt and Ashihara, 2002; Zhang et al., 2013). APT13 have been cloned, over-expressed in E. coli and compared using kinetic analyses by Allen et al. (2002). At a cytosolic pH, APT13 bind adenine efficiently based on their Km values (0.8e2.6 mM), although APT1 metabolizes adenine at a rate 31- and 35-fold faster than APT2 and APT3, respectively. Each isoform also uses the cytokinin bases, isopentenyl adenine, benzyladenine and trans-zeatin as substrates. Based on their Km values, APT2 and APT3 (15e440 mM) have much higher affinities than APT1 (1800e2500 mM) for all three cytokinins; conversely the Vmax values for APT2 and APT3 with these cytokinin substrates showed the opposite trend, being 4- to 19-fold lower than those of APT1. The parameter Vmax/Km, which is an estimate of catalytic potential of an enzyme, predicts that the three APT isoforms are very similar in their utilization of these cytokinin substrates in vitro. Since none of the predicted amino acid sequences for the APTs of A. thaliana appear to contain transit signalling peptides (see Moffatt and Ashihara, 2002), APTs are

Table 2 Properties of the native and recombinant purine phosphoribosyltransferase from plants. (A) Adenine phosphoribosyltransferase (APRT) Enzyme source Jerusalem artichokea Strawberrya Catharansus roseusa Brassica junceaa Wheat germa Tomato rootsa Tomato leavesa Hevea brasiliensisa Prunus persica Arabidopsis thalianaa Arabidopsis thalianab

Isozyme e e e e e e e e e e APT1 APT2 APT3 APT4 APT5

Purification crude crude 260-fold 4000-fold 200-fold e e 1500-fold 600-fold 3800-fold e e e e e

Molecular mass (kDa) e e e 54 23 27 47 54 (28 x 2) 27 54 (27 x 2) 28 28 28 28 28

Optimum pH

Km values (mM) A PRPP tZ 64 130 9.6 15 e 2.0 1.0 250 25 290 55 e e e e

e e e e e e e e 300 e 1800 330 76 e e

5.56.5 7.5 7.68.0 9.2 7.5 ~8 ~9 7.0 8.5 8.8 (37 C) 8.8 7.47.5 7.47.5 e e

5.5 10 7.0 3.8 74 4.0 2.0 3.8 0.7 4.5 1.0 2.6 0.8 e e

Optimum pH

Km values (mM) HX G X

iP

BA

e e e e 130 e e e e

e e e e 154 50 125

c

c

2500 110 440 e e

2400 15 370 e e

c

Localization

Ref No.

e e e e e e e Cytosol e e Cytosol, Chloroplast e Cytosol Cytosol Cytosol

1 2 3 4 5 6 7 8 9 10, 11

(B) Hypoxanthine/guanine phosphoribosyltransferase (HGPRT) Enzyme source Jerusalem artichokea Arabidopsis thalianab

Isozyme e e

Purification crude e

Molecular mass (kDa) e e

7.58.5 e

6.1 176

5.5 29

e e

Localization PRPP 8.8 

 

12 13

Abbreviations: Purine bases: A, adenine; G, guanine; HX, hypoxanthine; X, xanthine. Cytokinin bases: BA, benzyladenine; iP, isopentenyladenine; tZ, trans-zeatin. Phosphoribosyl-donor: PRPP, 5-phosphoribosyl-1-pyrophosphate. References: (1) Le Floc'h and Lafleuriel (1978); (2) Robert and Petel (2000); (3) Hirose and Ashihara (1983a); (4) Moffatt and Somerville (1990); (5) Chen et al. (1982); (6) Burch and Stuchbury (1986); (7) Gallois et al. (1996); (8) Lecomte and Le Floc'h (1999); (9) Lee and Moffatt (1993); (10) Allen et al. (2002); (11) Zybailov et al. (2008): (12) Le Floc'h and Lafleuriel (1981a); (13) Liu et al. (2007). a Native enzyme. b Recombinant enzyme. e, not determined. c Activity was found but the Km value is not available.

H. Ashihara et al. / Phytochemistry 147 (2018) 89e124

assumed to be located in the cytosol. However, current findings suggest that an isoform of APT1 has transit peptide for chloroplasts (Zybailov et al., 2008). Traditional biochemical analysis indicated that in addition to cytosol, adenine phosphoribosyltransferase is also distributed in plastids and mitochondria in other plants (Ashihara and Ukaji, 1985; Koshiishi et al., 2001; Le Floc'h and Lafleuriel, 1983). Despite the fact that the presence of adenine phosphoribosyltransferase in mitochondria has not yet been confirmed at a molecular level, purified intact mitochondria from C. roseus are capable of synthesizing adenine nucleotides from 14Clabelled adenine (Ukaji and Ashihara, 1986). There are many reports on physiological events associated with changes in the activity of adenine phosphoribosyltransferase (see Table 6 and Section 8.1). Functional analysis using adenine phosphoribosyltransferase-deficient mutants is also discussed in Section 8.1.1. The roles of adenine phosphoribosyltransferase in cytokinin interconversion and male sterility are covered in Sections 8.3 and 8.4, respectively. Omics studies on adenine phosphoribosyltransferase are listed in Section 8.7.

3.1.2. Hypoxanthine/guanine phosphoribosyltransferase Hypoxanthine-guanine phosphoribosyltransferase (IMP (GMP): diphosphate phospho-D-ribosyltransferase, EC 2.4.2.8) catalyses the formation of IMP (or GMP) from inosine (or guanine) and PRPP. Although activity of hypoxanthine/guanine phosphoribosyltransferase is generally high in mammalian tissues (Adams and Harkness, 1976), it is usually much lower than that of adenine phosphoribosyltransferase in plants. For example, the activity of hypoxanthine/guanine phosphoribosyltransferase is 20-fold lower than that of adenine phosphoribosyltransferase in potato tubers (Katahira and Ashihara, 2006a) and 100 times lower in tea leaves (Deng and Ashihara, 2010) (Table 1). Le Floc'h and Lafleuriel (1981a) reported partially purified hypoxanthine/guanine phosphoribosyltransferase from Jerusalem artichoke shoots (Table 2). Relatively low Km values (5e10 mM) for hypoxanthine, guanine and PRPP were observed. The rate of IMP formation from hypoxanthine was greater than GMP formation from guanine. The enzyme activity was inhibited by the products, IMP and GMP. Due to the instability of the enzyme, further purification and characterization was not achieved. Molecular and functional analysis of hypoxanthine/guanine phosphoribosyltransferase of A. thaliana were carried out by Liu et al. (2007). The kinetic analysis of the recombinant hypoxanthine/guanine phosphoribosyltransferase revealed that the enzyme catalyses the conversion of guanine and hypoxanthine to their respective nucleoside monophosphates, but xanthine is not as a substrate. The Km value of this recombinant enzyme for hypoxanthine (176 mM) was 30 times higher than that of the native enzyme extracted from Jerusalem artichoke shoots (6 mM). The relatively low affinity of the A. thaliana recombinant enzyme for hypoxanthine is different from other sources including humans (Keough et al., 1999), yeast (Ali and Sloan, 1982) as well as the native plant enzyme (Le Floc'h and Lafleuriel, 1981a) which display similar Km values towards guanine and hypoxanthine. The reason why the recombinant A. thaliana hypoxanthine/guanine phosphoribosyltransferase possesses a substantially higher affinity to guanine than to hypoxanthine has not yet been resolved. In most bacteria and nearly all eukaryotes, hypoxanthine/guanine phosphoribosyltransferase catalyses the salvage of hypoxanthine and guanine and, in a few cases, xanthine. Several bacteria possess two distinct enzymes for the salvage of 6-oxopurines (Craig and Eakin, 2000). In bacteria the primary substrate for hypoxanthine/guanine phosphoribosyltransferase is hypoxanthine, and guanine is

97

utilised with reduced efficiency (Lee et al., 1998). The other enzyme, xanthine phosphoribosyltransferase (EC 2.4.2.22), has a preference for catalysing the salvage of guanine and xanthine (Krenitsky et al., 1970). Recombinant hypoxanthine/guanine phosphoribosyltransferase (HGPT) from A. thaliana lacks activity towards xanthine (Liu et al., 2007). Deng and Ashihara (2010) reported the activity of xanthine phosphoribosyltransferase in tea leaves was 0.22 pkat/mg protein. This was the first report of a plant enzyme which catalyses the conversion of xanthine to xanthosine 50 -monophosphate (XMP), although the activity was 440 and ~5 times lower, respectively than that of adenine phosphoribosyltransferase and hypoxanthine/guanine phosphoribosyltransferase, (Table 1). It is unclear whether this xanthine phosphoribosyltransferase activity is due to a distinct enzyme or if it is side reaction of hypoxanthine/guanine phosphoribosyltransferase. In situ tracer experiments using [14C] xanthine indicate that xanthine is not salvaged to nucleotides and nucleic acids in several plant species including A. thaliana (Zrenner and Ashihara, 2011; Deng and Ashihara, 2010; Katahira and Ashihara, 2006a; Koyama et al., 2003; Yin et al., 2014). Functional analysis of hypoxanthine/guanine phosphoribosyltransferase genes (HGPT) which is related to seed germination are discussed in Section 8.1.1. 3.1.3. Structures of phosphoribosyltransferase (PRT) family proteins Molecular studies indicate that bacterial and animal adenine phosphoribosyltransferase (APT) and hypoxanthine/guanine phosphoribosyltransferase (HGPT) proteins belong to the phosphoribosyltransferase (PRT) family (Craig and Eakin, 2000; Liu et al., 2007; Sinha and Smith, 2001). Structurally, the PRT proteins share a highly conserved PRPP binding site, which represents the core fold of these proteins, consisting of three structural loops: the PPi- PRPP- and flexible loops. The PPi loop is involved in binding to the PPi group of PRPP, whereas the PRPP loop binds to the ribose-5phosphate group of PRPP. The flexible loop folds over the bound PRPP and contributes to the binding and release of PPi. These PRT proteins have a highly conserved amino acid sequence motif composed of 13 residues, nine of which are directly involved in forming the PRPP loop. In contrast to the PRPP loop, the amino acid composition of the PPi loop is more variable, although its conformation is generally conserved among different APT and HGPT members. Except for the well conserved Ser-Tyr dipeptide, the length and amino acid composition of the flexible loop are also variable among different PRT proteins. Liu et al. (2007) reported that the basic molecular features of plant HGPTs are similar to those of previously characterized hypoxanthine/guanine phosphoribosyltransferase enzymes. 3.2. Purine nucleoside kinases 3.2.1. Adenosine kinase Adenosine kinase (ATP: adenosine 50 -phosphotransferase, EC 2.7.1.20) catalyses the phosphorylation of adenosine to AMP using ATP as a phosphate donor. This enzyme was first isolated from yeast and animal tissues (Caputto, 1951; Kornberg and Pricer, 1951). Adenosine kinase is widely distributed in plant cells and tissues (Stasolla et al., 2003) and there are reports of both purified native and recombinant enzymes (Table 3A). Guranowski (1979b) purified adenosine kinase from seeds of yellow lupin (Lupinus luteus) to an almost homogeneous state. The enzyme is a single polypeptide chain with a molecular mass of 38 kDa. The Km value for adenosine is low (1.5 mM), but for ATP is higher (300 mM). This trend is also found with adenosine kinase from other plant sources (Table 3A). ATP is the main phosphate

98

H. Ashihara et al. / Phytochemistry 147 (2018) 89e124

Table 3 Properties of the native and recombinant purine nucleoside kinases from plants. (A) Adenosine kinase (AK) Enzyme source Wheat germ Yellow lupin seedsa Peach flower budsa Tobacco BY-2 cellsa Arabidopsis thalianab Tobacco BY-2 cellsb

Physcomitrella (moss)b

Isoform     ADK1 ADK2 ADK1T ADK1S ADK2T ADK2S 

Purification

Molecular mass (kDa)

Optimum pH

Km values (mM) AR ATP tZR

20-fold 7300-fold         

 38 45 40 38 38 38 38 38 38 28

6.87.2 7.07.5 9.5  8.0 9.5 7.5 7.0 8.5 8.8 (37 C) 

8.7 1.5 1.8  0.5 0.3 4.0 1.9 7.3 3.9 

 300 200  350 370     

      2.4 2.5 3.2 2.2 

Purification

Molecular mass (kDa)

Optimum pH

Km values (mM) IR GR XR

iPR

DHZR

31    3.2 4.8 0.3 6.1 8.7 1.7

      3.9 3.9 2.6 3.8 

c

Localization

Ref No.

    cytosol cytosol     

1 2 3 4 5 5 6 6 6 6 7

(B) Inosine/guanosine kinase Enzyme source Jerusalem artichokea

Isoform 

crude



7.6

70

Molecular mass (kDa)

Optimum pH

Km values (mM) IR AR GR

14



Localization ATP 1450

mitochondria

8

(C) Nucleoside phosphotransferase (NPT) Enzyme source Carrot rootsa Yellow lupin cotyledonsa Barley seedlingsa

Isoform e e e

Purification 940-fold 120-fold e

45 72 50 x 2

5.0 8.0 e

e 400 e

Localization AdR

AMP

c

c

c

c

400 540

e 2130

e 540

400 e

e e e

9 10 11

Abbreviations: Purine nucleosides: AR, adenosine; AdR, deoxyadenosine; GR, guanosine; IR, inosine; XR, xanthosine. Cytokinin ribosides: DHZR, dihydrozeatin riboside; iPR, isopentenyladenosine; tZR, t-zeatin riboside. References: (1) Chen and Eckert (1977); (2) Guranowski (1979b); (3) Faye and Le Floc'h (1997); (4) Laukens et al. (2003); (5) Moffatt et al. (2000); (6) Kwade et al. (2005); (7)
s et al. (1989); (9) Brunngraber and Chargaff (1967); (10) Guranowski (1979a); (11) Prasher et al. (1982). von Schwartzenberg et al. (1998); (8) Combe a Native enzyme. b Recombinant enzyme. c Activity was found but the Km value is not available. e, not determined.

donor. Other 50 -nucleoside triphosphates, inosine-50 -triphosphate, dATP, GTP, and xanthosine-50 -triphosphate can substitute for ATP but are less effective. From the view point of phosphorylation of cytokinins (see also Section 8.3), Chen and Eckert (1977) partially purified adenosine kinase from wheat germ which rapidly phosphorylated adenosine and the cytokinin riboside, isopentenyladenine riboside. The Km for isopentenyladenine riboside (31 mM) was ~3.5 times larger than that for adenosine (8.7 mM). The Vmax for isopentenyladenine riboside (140 pkat/mg protein) was ~6 times lower than for adenosine (770 pkat/mg protein). These findings indicate that adenosine is a better substrate for the wheat germ adenosine kinase than the cytokinin. Subsequently, adenosine kinase was purified from the flower buds of peach (Faye and Le Floc'h, 1997) and tobacco cells (Laukens et al., 2003). The kinetic properties of these enzymes are summarised in Table 3A. Cloning and characterization of adenosine kinase was first performed in the moss Physcomitrella patens (von Schwartzenberg et al., 1998). A gene (adk) was cloned from a cDNA library by functional complementation of an E. coli purine auxotrophic strain. The length of the entire cDNA clone was 1175 bp with an open reading frame coding for a protein with a predicted molecular mass of 37.3 kDa. Southern analysis indicated the presence of a single adk gene within the Physcomitrella genome. The deduced amino acid sequence had a 52% identity with the human adk. The phosphorylation of adenosine and isopentenyladenine riboside was shown by in vitro enzyme assays using crude extracts from E. coli mutants expressing the adk cDNA clone and from Physcomitrella chloronemal tissue. Subsequently, the cDNAs and genes encoding two isoforms of adenosine kinase were isolated from A. thaliana

(Moffatt et al., 2000). The adk1- and adk2-coding sequences were very similar, respectively sharing 92% and 89% amino acid and nucleotide identity. Each cDNA was overexpressed in E. coli, and the catalytic activity of each isoform was determined. Both adenosine kinases had similar catalytic properties (Table 3A). The Km values for adenosine and for isopentenyladenine riboside were 0.3e0.5 mM and 3e5 mM, respectively. While Vmax values of respective substrates were 45e112 pkat and 1e11 pkat/mg protein. This suggests that adenosine is the much preferred substrate for both adenosine kinase isoforms. The Kms of ADK1 and ADK2 for isopentenyladenine riboside are essentially the same but the Vmax of ADK2 with isopentenyladenine riboside is 10-fold higher than that of ADK1. The physiological concentrations of endogenous cytokinins in A. thaliana are ~103-fold lower than these apparent Km values (see Section 6.1) and the Vmax/Km ratios for isopentenyladenine riboside are much lower than those for adenosine. From these in vitro kinetic results, it is clear that cytokinins are not the primary in vivo substrates of either ADK1 or ADK2. However, their relatively low Km values for isopentenyladenine riboside (<5 mM) suggest that ADK isozymes contribute to some extent to the metabolism of cytokinin ribosides. Kwade et al. (2005) identified four adenosine kinase isoforms, designated 1S, 2S, 1T, and 2T, in tobacco BY2 cells. In contrast to adenosine kinase from other plant sources, all four tobacco adenosine kinase isoforms displayed a high affinity for adenosine and three distinct types of cytokinin ribosides: isopentenyladenine riboside; trans-zeatin riboside; and dihydrozeatin riboside (Table 3A). The Vmax/Km values suggest that in vitro AK2S exhibits an overall higher efficiency in the metabolism of cytokinin

H. Ashihara et al. / Phytochemistry 147 (2018) 89e124

ribosides than the other three isoforms. Schoor et al. (2011) investigated the functional significance of adenosine kinase activity in A. thaliana development and cytokinin metabolism using ADK-deficient lines created by transgene silencing (see Section 8.3.2). Adenosine kinase activity associated with various physiological events are presented in Table 6 and discussed in Section 8.1. Role of adenosine kinase in the homeostasis of cytokinins, plant pathogen responses and root gravitropism are noted in Sections 8.3, 8.5 and 8.6, respectively. 3.2.2. Inosine/guanosine kinase Inosine/guanosine kinase (ATP: inosine/guanosine 50 -phosphotransferase, EC 2.7.1.73) catalyses the phosphorylation of inosine to IMP and that of guanosine to GMP using ATP as a phosphate donor. This enzyme activity was first found in cell-free extracts of Ehrlich ascites cells (Pierre and LePage, 1968). In contrast to adenosine kinase, distribution of inosine/guanosine kinase is limited. Unusually, a few prokaryotes, including E. coli and Salmonella typhimurium, contain inosine/guanosine kinase rather than adenosine kinase (Nygaard, 1983). This enzyme has been found in potato tubers (Katahira and Ashihara, 2006a) and tea leaves (Deng and Ashihara, 2010) (Table 1). The partial purification of inosine/guanosine kinase from mitochondria of Jerusalem artichoke was re
s et al. (1989) (Table 3B). The enzyme seems to be ported by Combe located in intermembrane space of mitochondria. The Km values for guanosine (14 mM) are lower than for inosine (70 mM). Although the inosine/guanosine kinase gene has been cloned in E. coli (Harlow et al., 1995; Mori et al., 1995), there are no reports of the plant enzyme being cloned. Changes in the activity of inosine/guanosine kinase have been found during cell growth, embryogenesis and organogenesis in plants (see Table 6 and Sections 8.1.2 and 8.1.3). 3.2.3. Structures of ribokinase (RBSK) family proteins Molecular studies with bacteria and animals indicate that adenosine kinase and ribokinase are two important enzymes belonging to the RBSK (also known as RK) family of proteins that share a number of unique primary and tertiary structural elements (Park and Gupta, 2008). Ribokinase catalyses the phosphorylation of ribose to ribose-5-phosphate, using ATP as a phosphate donor. Ribokinase encoding genes (RBSKs) have been found in both prokaryotes and eukaryotes, and sequence comparisons have shown that they belong to the PfkB (phosphofructokinase B) family of carbohydrate kinases. Members of the RBSK family proteins are identified by the presence of two highly conserved sequence motifs. The first motif is located in a glycinerich area in proximity of the N-terminal region of these enzymes. This motif includes two consecutive glycine residues that form part of the hinge between the lid domain and the aba domain. The second motif, which is found in the C-terminal region, is involved in ATP binding and the formation of an anion hole. This motif also contains a conserved aspartic acid residue which serves as a catalytic base. Although the overall sequence identity between the members of the RBSK family is less than 30%, their crystal structures are very similar. Despite large differences in their primary and tertiary structures, enzymes in the RBSK families show similar substrate specificity and carry out identical functions. The human RBSK gene has a much higher degree of sequence identity to the RBSK sequences from protists and bacteria than to the homologues from A. thaliana and Saccharomyces cerevisiae (Park et al., 2007). It is, thus, possible that RBSKs from plants and fungi species originated independently from those of animal and protist species. Recently, Riggs et al. (2016) identified RBSK in A. thaliana

99

and characterized ribokinase activity in the protein. However, they did not mention its adenosine kinase activity. The authors proposed that this enzyme participates in the salvage of ribose released from nucleosides. 3.3. Nucleoside phosphotransferase Nucleoside phosphotransferase (nucleotide: nucleoside 50 phosphotransferase, EC 2.7.1.77) catalyses the conversion of nucleosides to nucleoside monophosphate using 50 -nucleoside monophosphate, for example, inosine þ AMP / inosine-50 monophosphate (IMP) þ adenosine. This is a non-specific nucleoside phosphotransferase. In potato tubers, the enzyme converts adenosine, inosine, guanosine, deoxyadenosine and deoxyguanosine to AMP, IMP, GMP, dAMP and dGMP, respectively (Katahira and Ashihara, 2006a). Potentially xanthosine may be converted to XMP but it has not been tested as a substrate (Table 1). Hence, in addition to nucleoside kinases, nucleoside phosphotransferase acts as purine nucleoside salvage for purine nucleotide synthesis. This enzyme is widely distributed (Brawerman and Chargaff, 1955). Highly purified native nucleoside phosphotransferase has been obtained from carrot roots (Brunngraber and Chargaff, 1967, 1970), cotyledons of yellow lupin seedlings (Guranowski, 1979a) and barley seedlings (Prasher et al., 1982) (Table 3C). The enzyme from yellow lupin seedling cotyledons (Guranowski, 1979a) is similar to the enzyme purified from a bacterium, Erwinia herbicola (Chao, 1976). With the yellow lupin nucleoside phosphotransferase, purine and pyrimidine nucleosides are good phosphate acceptors and 50 -nucleotides are more effective phosphate donors than 30 -AMP. The high Km values indicate that nucleoside phosphotransferase binds purine nucleosides but with much lower affinity than purine nucleoside kinases (see Section 3.2). Changes of nucleoside phosphotransferase activity in relation to seed germination, growth of cells and somatic embryos are shown in Table 6 and noted in Sections 8.1.18.1.3. Correlations between nucleoside phosphotransferase activity and glutathione redox state are described in Section 8.1.11. 3.4. Purine-nucleoside phosphorylase Purine bases may also be salvaged via purine nucleosides by the reverse reaction of purine-nucleoside phosphorylase (purinenucleoside: phosphate ribosyltransferase, EC 2.4.2.1) using ribose1-phosphate. Purine-nucleoside phosphorylase catalyses both phosphorylytic cleavage of purine nucleosides and ribosylation of purine bases. In bacteria and animals, purine-nucleoside phosphorylase is an important enzyme in purine metabolism. However, an absence of purine-nucleoside phosphorylase activity has been often reported when purine bases were used as substrates. For example, the enzyme activity could not be detected in extracts of tobacco leaf protoplasts (Barankiewicz and Paszkowski, 1980), Acer pseudoplatanus cells (Doree, 1973), cotyledons of germinating lupin seeds (Guranowski and Barankiewicz, 1979) and developing potato tubers (Katahira and Ashihara, 2006a). Likewise, the enzyme does not occur in Euglena gracilis (Guranowski and Wasternack, 1982). The presence of purine nucleoside phosphorylase activity has been claimed by cytokinin researchers (Table 4C). Chen and Petschow (1978) reported the partial purification from wheat germ of purine-nucleoside phosphorylase, which catalyses the ribosylation of isopentenyladenine, kinetin and adenine to form the corresponding nucleosides. Recently Bromley et al. (2014) reported on a cytokinin riboside phosphorylase (StCKP1) from potato. This

100

H. Ashihara et al. / Phytochemistry 147 (2018) 89e124

Table 4 Properties of the native and recombinant purine nucleosidase and purine nucleoside phosphorylase from plants. (A) Adenosine nucleosidase (ARN) Enzyme source Brussels sproutsa Spinach beet leavesa Barley leavesa Tea leavesa

Isoform e e I, II, III

e Wheat germa Yellow lupin seedsa e Coffee leavesa e a Jerusalem artichoke tubers e a e Potato

Molecular mass (kDa) Optimum pH Km values (mM) AR AdR IR

GR

iPAR

17-fold 40-fold 100-fold e

e e 33 x 2 68

e e e e

e e e e

e e e e

e e e e

1 2 3 4

46-fold 146-fold 5800-fold crude crude

59 72 (33 x 2) 72 (35 x 2) e e

e e e e

e e e e

2.4 e e e e

e e e e Cell wall

5 6 7 8 9

Purification

Molecular mass (kDa) Optimum pH Km values (mM) IR GR XR

Localization

crude 65-fold 1700-fold

e 62 80

e e e

Purification

3.54.5 4.5 4.75.4 4.0 (I, III) 4.5 (II) 4.7 7.5 6.0 57 e

2400 e 11 e 0.82.3 120 c

c

1.4 4.8 6.3 17

e e e e e

c

c

Localization Ref No.

c

(B) Inosine/guanosine nucleosidase Enzyme source

Isoform

Jerusalem artichoke tubersa e Yellow lupin seedsa e Yellow lupin seedsa e

57 8 4.75.5

2.5 65 2.7

8.5 e 2.7

e e e

8 10 11

(C) Nucleoside hydrolase Enzyme source Arabidopsis thalianab Maizeb Physcomitrella patensb

Recombinant protein Major substrate URH1 ZmNRH3 PpNRH1

Uridine Inosine Xanthosine

Molecular mass (kDa) Optimum pH Km values (mM) AR IR GR e 76 81

e 79 79

700 39 113

Localization XR

1400 e e 201 104 396 78 61 116

iPAR e 196 e

e e e

12 13 13

(D) Purine nucleoside phosphorylase (PNP) Enzyme source

Recombinant protein Major substrate

Potatob

StCKP1

Molecular mass (kDa) Optimum pH Km values (mM) AR iPR tZR DHZR Pi

Cytokinin ribosides Cytokinin 40 bases

e e

128 A 3.3

13 7 7 iP tZ DHZ 0.05 0.35 0.56

c

Localization e

14

Ribose-1-P c

Abbreviations: Purine nucleosides and bases: A, adenine; AR, adenosine; AdR, deoxyadenosine; GR, guanosine; IR, inosine; XR, xanthosine. Cytokinin ribosides and bases: DHR, dihydrozeatin; DHZR, dihydrozeatin riboside; iP, isopentenyladenine; iPR, isopentenyladenosine; tZ, trans-zeatin; tZR, trans-zeatin riboside. References: (1) Mazelis and Creveling (1963); (2) Poulton and Butt (1976); (3) Guranowski and Schneider (1977); (4) Imagawa et al. (1979); (5) Chen and Kristopeit (1981); (6) Abusamhadneh et al. (2000); (7) Campos et al. (2005); (8) Le Floc'h and Lafleuriel (1981b); (9) Riewe et al. (2008b); (10) Guranowski (1982); (11) Szuwart et al. (2006); (12)
et al. (2013), (14) Bromley et al. (2014). Jung et al. (2009); (13) Kope cna a Native enzyme. b Recombinant enzyme. c Activity is found but no Km values are shown. e, not determined.

recombinant enzyme catalyses both ribosylation and phosphorolytic cleavage reactions. StCKP1 has higher affinity for cytokinin bases and nucleosides than for adenine and adenosine. The Km value for isopentenyladenine (0.05 mM) and isopentenyladenine riboside (10 mM) is lower than those for adenine (4 mM) and adenosine (130 mM). These results suggest that purine-nucleoside phosphorylase may to be involved in cytokinin metabolism, but its role in relation to adenine, guanine and hypoxanthine salvage is very limited, because as yet no in vitro enzyme activity has been detected in plant extracts (Table 1). Information on the function of purine-nucleoside phosphorylase in endodormancy of potato tubers is presented in Section 8.3.3.

formed as a by-product of a number of biosynthesis pathways, including those of ethylene and polyamines (Waduwara-Jayabahu et al., 2012) (see Section 4.4).

4. Pathways for supply of purine bases and nucleosides

4.2. Degradation of nucleoside triphosphates

Intracellular purine bases and nucleosides are mainly produced by degradation of nucleic acids (4.1) and nucleotides (4.2). In addition, adenosine is a product of the S-adenosyl-L-methionine (SAM) cycle (Moffatt and Weretilnyk, 2001) (4.3). Adenine is produced by the degradation of 5-methylthioadenosine which is

Conversion of nucleoside triphosphate to diphosphate is coupled with many chemical reactions including biosynthetic and various kinase reactions. Formation of AMP from ATP is also coupled with diphosphotransferase reactions such as 5phosphoribosyl-1-pyrophosphate (PRPP) synthetase (EC 2.7.6.1)

4.1. Degradation of nucleic acids Degradation of various RNAs and DNAs occurs in the turnover of nucleic acids and in senescent plant tissues. Various nucleases participate in this degradation (Green, 1994; Wilson, 1975, 1982) and 30 - or 50 -nucleoside monophosphate or deoxynucleosides monophosphates are produced and enter nucleotide pools. Purine nucleotides produced by nucleases are further catabolised to purine nucleosides and purine bases (see Voet and Voet, 2010; Buchanan et al., 2015).

H. Ashihara et al. / Phytochemistry 147 (2018) 89e124

101

Fig. 5. Purine nucleoside producing reactions in plants. (A) 50 -nucleosidase (EC 3.1.3.5) or some phosphatases (3.1.3.), (B) 30 -nucleosidase (3.1.3.6) or phosphatases (3.1.3.), (C) Sadenosyl-L-homocysteine (SAH) hydrolase and (D) S-adenosyl-L-methionine (SAM) cycle consist of (1) SAM synthetase (2.5.1.6), (2) SAM-dependent methyltransferases (2.1.1.), (3) SAH hydrolase (3.3.1.1) and (4) methionine synthase (2.1.1.13).

102

H. Ashihara et al. / Phytochemistry 147 (2018) 89e124

and thiamine diphosphokinase (EC 2.7.6.2). In addition, apyrases which hydrolyse both the g- and b-phosphates of nucleoside triphosphates and nucleoside diphosphates contribute to the degradation of nucleoside triphosphates. 4.3. Production of purine nucleosides Adenosine, guanosine, deoxyadenosine and deoxyguanosine are formed from respective 50 - or 30 -purine nucleotides. Inosine and

xanthosine are produced, respectively, from 50 -IMP and xanthosine-50 -monophosphate (50 -XMP). For the example, adenosine formation pathways are illustrated in Fig. 5. Purine nucleosides are formed by the reactions catalysed by 50 -nucleotidase (EC 3.1.3.5) and 30 -nucleotidase (EC 3.1.3.6), hydrolytic enzymes which catalyse the hydrolysis of nucleotide into nucleoside and inorganic phosphate (Pi) (step 1 and 2 in Fig. 5A and B). Acid phosphatase (EC 3.1.3.2) and alkaline phosphatase (EC 3.1.3.1), with broad substrate specificity, may also participate in these reactions. Purine

Fig. 6. Purine base producing reactions in plants. (A) (1) adenosine nucleosidase (EC 3.2.2.7), and (2) inosine-guanosine nucleosidase (3.2.2.2), (B) 50 -methylthioadenosine nucleosidase (3.2.2.16) and (C) S-adenosyl-L-homocysteine nucleosidase (3.2.2.9).

H. Ashihara et al. / Phytochemistry 147 (2018) 89e124

nucleosides are also produced from cyclic nucleotides, namely cAMP and cGMP, after conversion to 50 -AMP and 50 -GMP by cyclic nucleotide phosphodiesterase (EC 3.1.4.17). Adenosine is also produced from S-adenosyl-L-homocysteine (SAH) (step 3 in Fig. 5C) released from the SAM cycle (Fig. 5D). There are many SAM-dependent methyltransferase-catalysed reactions (Struck et al., 2012). The methyl group of SAM is transferred to the methyl acceptors (Fig. 5D) and SAH is formed which is then hydrolysed to homocysteine and adenosine (step 3 in Fig. 5C) by Sadenosyl-L-homocysteine hydrolase (EC 3.3.1.1). 4.4. Production of purine bases In plants purine bases are produced from purine nucleosides mainly by adenosine nucleosidase (EC 3.2.2.7), and inosine/guanosine nucleosidase (EC 3.2.2.2) (Fig. 6A). In addition, adenine is produced from 50 -methylthioadenosine by 50 -methylthioadenosine nucleosidase (EC 3.2.2.16) (Fig. 6B) and from SAH by S-adenosyl-Lhomocysteine nucleosidase (EC 3.2.2.9) (Fig. 6C). 50 -Methylthioadenosine is produced in the biosynthesis of ethylene, polyamines and nicotianamine. Ethylene, a volatile plant hormone, is synthesized from SAM in a two-step reaction. The first step, catalysed by 1-aminocyclopropane-1-carboxylic acid synthase (EC 4.4.1.14) releases 50 -methylthioadenosine. In turn, 50 -methylthioadenosine is hydrolysed to 50 -methylthioribose and adenine, which is reused in the methionine salvage cycle, the so-called Yang cycle (Yang and Hoffman, 1984). In the biosynthesis of polyamines, including spermidine, spermine, and thermospermine, S-adenosylmethioninamine (decarboxylated SAM), which is produced by SAM decarboxylase, is utilised. In this process, 50 -methylthioadenosine is released. For example, spermidine synthase (EC 2.5.1.16) catalyses the reaction, putrescine þ S-adenosylmethioninamine / spermidine þ 50 -methylthioadenosine (Takahashi and Kakehi, 2010). In nicotianamine synthesis, three moles of SAM are needed and two moles of 50 -methylthioadenosine are released as a by-product in the reaction catalysed by nicotianamine synthase (EC 2.5.1.43) (Itai et al., 2000). In addition to the above reactions, limited amounts of purine bases and nucleosides are produced by many degradation reactions of purine containing metabolites. For example, very small amounts of adenine are produced by the degradation of cytokinins by cytokinin oxidase/dehydrogenase (EC 1.5.99.12) (Trifunovi
c et al., 2015). Production of adenosine and adenine by these reactions is related to various physiological events (see Sections 8.1.6, 8.1.7 and 8.2). 4.5. Production of 5-phosphoribosyl-1-pyrophosphate 5-Phosphoribosyl-1-pyrophosphate (PRPP) is a phosphoribosyl donor for AMP and GMP synthesis in reactions catalysed by adenine phosphoribosyltransferase and hypoxanthine/guanine phosphoribosyltransferase. PRPP is synthesized from ribose-5-phosphate and ATP by PRPP synthetase (EC 2.7.6.1). In plants, ribose-5phosphate is provided by the photosynthetic reductive pentose phosphate pathway (Calvin-Benson-Bassham cycle) and/or the oxidative pentose phosphate pathway (Ashihara, 2016).

103

5.1. Apyrase Apyrases (nucleoside triphosphate phosphohydrolase [nucleoside monophosphate-forming], EC 3.6.1.5) are distinct from other phosphatases with respect to their specific activity, nucleotide substrate specificity, divalent cation requirement, and sensitivity to inhibitors. The enzymes can be divided into two main classes: ectoapyrases and endo-apyrases. In plant tissues, apyrases are cytosolic, nuclear or membrane-bound proteins (Kettlun et al., 1982). Many reports, mainly on potato apyrases, have been published. For example, Kettlun et al. (2005) reported the substrate specificity of apyrases from two potato tuber clonal varieties by measuring their kinetic parameters, Km and kcat, on deoxyribonucleotides and fluorescent analogues of ATP and ADP. Both apyrases showed a broad specificity towards dATP, dGTP, dTTP, dCTP and thio-dATP. Cloning of the gene encoding a soluble potato apyrase was reported by Handa and Guidotti (1996). Localization of apyrase in apoplasts, the space outside the plasma membrane, has been demonstrated in soybean (Day et al., 2000) and potato (Riewe et al., 2008a). On the basis of these results, a model was proposed that links apyrase to growth via regulation of extracellular ATP. Riewe et al. (2008b) demonstrated that incubation of potato tuber slices with ATP led to the formation of ADP, AMP, adenosine, adenine and ribose, indicating operation of apyrase, 50 -nucleotidase and nucleosidase. 5.2. Nucleotidases and phosphatases 30 - and 50 -Nucleotide monophosphates are hydrolysed to their respective nucleosides by several phosphohydrolase enzymes including acid phosphatase (EC 3.1.3.2), 50 -nucleotidase (EC 3.1.3.5), and 30 -nucleotidase (EC 3.1.3.6). Acid phosphatases are enzymes which catalyse the breakdown of a wide variety of phosphate esters and usually exhibit a pH optima below 6.0 (Ullah and Gibson, 1988). 50 -Nucleotidase catalyses the phosphorylytic cleavage of 50 -nucleotides and this group of enzymes is widely distributed in animal tissues, plants and bacteria (Sharma et al., 1986). Soluble and membrane-bound 50 -nucleotidase have been found in plants (Polya and Ashton, 1973; Sharma et al., 1986). The soluble wheat enzyme has Km values for 50 -AMP and 30 -AMP of 1.4 and 6.1 mM, respectively (Polya and Ashton, 1973). Membrane-bound 50 -nucleotidase has been detected in the plasma membrane (Sharma et al., 1986) and Golgi apparatus of peanut cotyledons (Mittal et al., 1988). The plasma membrane enzyme is specific for 50 -AMP. Other nucleotides, namely GMP, UMP, CMP, ADP, GDP, ATP, GTP and UTP, were not hydrolysed (Sharma et al., 1986). In contrast to animal enzymes, acidic pH optimum has been described for both membrane-bound and soluble forms of plant 50 -nucleotidases. These enzymes also hydrolyse 30 -nucleotides (Zimmermann, 1992). In contrast to 50 -nucleotidase, there are few reports of plant 30 nucleotidases (30 -ribonucleotide phosphohydrolase, EC 3.1.3.6). Potato 30 -nucleotidase dephosphorylated ribonucleoside 30 -phosphates and hydrolysed RNA, DNA and synthetic polyribonucleotides. It has been suggested that a single enzyme is responsible for both types of hydrolysis (Nomura et al., 1971; Suno et al., 1973). 50 -Ribonucleotides are competitive inhibitors of 30 nucleotides. For further details, readers are referred to an excellent review on plant 50 -nucleotidases by Zimmermann (1992).

5. Enzymes which contribute to the production of the substrates of purine salvage

5.3. Nucleosidases

There are many RNases, DNases, non-specific acid- and alkalinephosphatases in plants. However, such enzymes are not a topic of this review which focuses on enzymes which are closely related to the formation of purine nucleosides and nucleobases.

Nucleosidases (nucleoside hydrolases) are a class of enzymes that hydrolyse the N-glycosidic bond of selected nucleosides between the base and sugar. They have been isolated from a number of sources including bacteria, parasitic protozoans, plants, marine

104

H. Ashihara et al. / Phytochemistry 147 (2018) 89e124

invertebrates and baker's yeast, but not mammals (Arivett et al., 2014). Following traditional enzyme purification and characterization studies, adenosine nucleosidase (adenosine ribohydrolase, EC 3.2.2.7) and inosine/guanosine nucleosidase (inosine ribohydrolase, EC 3.2.2.2) have been reported (Table 4A and B). In contrast, A. thaliana recombinant nucleosidases, obtained using the homologs of the well-characterized nucleosidase of Crithidia fasciculate, a species of parasitic excavates, were used to generate recombinant proteins, which in turn had their enzymatic capacity assessed (see Section 5.3.3.) Recombinant enzymes expressed in E. coli and yeast have multiple substrate specificities and are called uridine ribohydrolase (URH) or nucleoside ribohydrolase (NRH) (Table 4C). These recombinant enzymes have as yet not been detected as native proteins isolated from plants. 5.3.1. Adenosine nucleosidase Soluble adenosine nucleosidase has been purified from Brussels sprouts (Mazelis and Creveling, 1963), wheat germ (Chen and Kristopeit, 1981), yellow lupin seeds (Abusamhadneh et al., 2000), Jerusalem artichoke tubers (Le Floc'h and Lafleuriel, 1981b), leaves of spinach beet (Poulton and Butt, 1976), barley (Guranowski and Schneider, 1977), tea (Imagawa et al., 1979) and coffee (Campos et al., 2005). Cell wall bound adenosine nucleosidase which may be involved in the salvage of extracellular ATP has been also detected potato tubers (Riewe et al., 2008b). Properties of enzymes are summarised in Table 4. Most enzymes are specific for adenosine and deoxyadenosine, and do not hydrolyse other purine and pyrimidine nucleosides. Wheat germ adenosine nucleosidase can hydrolyse the cytokinin, isopentenyladenine riboside (Chen and Kristopeit, 1981) (see 8.3). Adenosine nucleosidase activity fluctuates during the course of many physiological events including seed formation and germination, growth of cells and embryos, organogenesis, and in response to stresses including salt, wounding and phosphate deficiency (see Section 8.1). This enzyme may also be involved in cytokinin interconversions (Section 8.3). 5.3.2. Inosine/guanosine nucleosidase Inosine/guanosine nucleosidase (inosine/guanosine ribohydrolase, EC 3.2.2.2) activity has been found in E. coli (Koch and Lamont, 1956), fish (Tarr, 1955) and plants (Deng and Ashihara, 2010; Katahira and Ashihara, 2006a; Stasolla et al., 2003). This enzyme, with a slightly different substrate specificity, has been detected in Jerusalem tubers (Le Floc'h and Lafleuriel, 1981b) and yellow lupin seeds (Guranowski, 1982; Szuwart et al., 2006) (Table 4B). Guranowski and his co-workers reported two distinct enzymes which hydrolyse inosine and guanosine. Guranowski (1982) partially purified inosine nucleosidase from yellow lupin seeds that catalysed the hydrolysis of adenosine and guanosine but at lower rates than inosine. Subsequently, the Guranowski group reported on a highly purified guanosine/inosine-preferring nucleosidase from yellow lupin seeds (Szuwart et al., 2006). 5.3.3. Other nucleosidases An enzyme which possesses broader substrate specificity has been purified to homogeneity from the soluble, plant cell fraction of N2-fixing nodules of the tropical legume cowpea (Atkins et al., 1989). Among the purine nucleosides, the Vmax values are in the ratio 28:7:1:0.4 for xanthosine, inosine, adenosine and guanosine, respectively. The enzyme can, therefore, be classified as a purine nucleosidase (nucleoside hydrolase; EC 3.2.2.1). In addition to purine nucleosides, the isolated enzyme hydrolysed pyrimidine nucleosides, namely, uridine, thymidine and cytidine. Some studies on plant nucleosidases have been carried out

using recombinant enzymes expressed in E. coli or yeast (Jung et al.,
et al., 2013). Jung et al. (2009) describe the 2009, 2011; Kopecna cloning of a nucleosidase from A. thaliana, uridine ribohydrolase 1 (URH1) which was characterized by complementation of a yeast mutant. Furthermore, URH1 was synthesized as a recombinant protein in E. coli. The purified recombinant protein exhibited hydrolase activity for both pyrimidine and purine nucleosides, namely, uridine, inosine, adenosine and isopentenyladenine riboside; the Km values of the E. coli expressed enzyme for the corresponding substrates were 800, 1400, 700 and 400 mM, respectively. A much higher Km value of 2100 mM was found for uridine in the recombinant enzyme expressed in yeast. These values are ~1000 times higher than that of most native adenosine nucleosidase and inosine/guanosine nucleosidase (Table 4). In a subsequent paper, it was reported that xanthosine, as well as uridine, were better substrates for NSH1 (formerly designation URH1) than inosine and adenosine (Jung et al., 2011). The apparent Km value for NSH1 for xanthosine cleavage (1700 mM) is high, and the catalytic activity with xanthosine is 6-fold lower than with uridine. However, extremely low affinity of substrates and broad substrate specificity observed with this recombinant enzyme may reflect differences in the post-translational modification of the protein expressed in E. coli or yeast and in plants. Characterization of recombinant nucleosidases in two model plants, Physcomitrella patens (PpNRH) and maize (ZmNRH) has
et al., 2013). The substrate specificity of been reported (Kope cna these enzymes are broad, as they can hydrolyse xanthosine, inosine, adenosine, guanosine, uridine and cytokinin-ribosides with different efficiencies. Two isoforms of nucleosidases were found; one isoform preferentially used inosine and xanthosine and the other hydrolyses uridine and xanthosine. Both enzymes catalysed the hydrolysis of isopentenyladenine riboside. The Km values of typical recombinant enzymes are shown in Table 4. The values (100e400 mM) are much higher than those observed with native nucleosidases (1e10 mM) in plants. As shown in Section 4.3, adenine is produced by hydrolysis of Smethyl-50 -thioadenosine catalysed by 50 -methylthioadenosine nucleosidase (EC 3.2.2.16) (Fig. 6B) and hydrolysis of SAH catalysed by S-adenosyl-L-homocysteine nucleosidase (EC 3.2.2.9) (Fig. 6C). 50 -Methylthioadenosine nucleosidase which catalyses hydrolytic cleavage of 50 -methylthioadenosine with the formation of adenine and 50 -methylthioribose, has been purified to homogeneity from yellow lupin seeds (Guranowski et al., 1981). The nucleosidase exhibits highest specificity towards the natural substrate with a Km of 0.4 mM for 5-methylthioadenosine. Two similar nucleosidases for hydrolysing 5-methylthioadenosine occur in A. thaliana (AtMTAN1 and AtMTAN2), but only AtMTAN2 shows broad substrate specificity for hydrolysis of both 5methylthioadenosine and SAH (Park et al., 2009). The molecular structure of AtMTAN1 and AtMTAN2 has been investigated by Siu et al. (2008, 2011). 5.4. Phosphoribosylpyrophosphate synthetase PRPP synthetase (ribose-phosphate diphosphokinase, EC 2.7.6.1) catalyses the reaction, ribose 5-phosphate þ ATP / PRPP þ AMP. Since a nucleotide consists of a phosphoribosyl moiety and a base, PRPP synthetase is recognised as an enzyme involved in nucleotide biosynthesis (see Buchanan et al., 2015; Voet and Voet, 2010). Characterization of native PRPP synthetase from black gram seedlings (Ashihara and Komamine, 1974), spinach leaves (Ashihara, 1977a, b) and rubber tree latex (Gallois et al., 1997) has been reported. Krath et al. (1999) clonied and sequenced four PRPP synthases from A. thaliana encoding cDNAs which were then expressed in E. coli. The four cDNAs were designated PRS14 and

H. Ashihara et al. / Phytochemistry 147 (2018) 89e124

their respective gene products PRPP synthetase isozymes PRS14. PRS1 and PRS2 require Pi for activity and therefore resemble the PRPP synthetase obtained from E. coli, Salmonella typhimurium, Bacillus subtilis and mammals (class I). In contrast, PRS3 and PRS4 are novel plant-specific Pi-independent enzymes (class II). This is based on both their low sequence homology and apparent phylogenetic divergence from the bacterial and mammalian PRPP synthetases. The class II enzyme represents the predominant PRPP synthetase activity in plants and is located mainly in cytosol. In contrast, class I isoenzymes are located in chloroplasts (Krath and Hove-Jensen, 1999, 2001). A number of nucleoside monophosphates and nucleoside diphosphates inhibit the activity of this enzyme. Furthermore the enzyme is influenced by the cellular adenylate energy charge and is tightly controlled by cellular nucleotide levels. A comprehensive review on plant PRPP synthetase has been published elsewhere (Ashihara, 2016). 6. Cellular concentration of substrates and products of purine salvage Together with pyrimidines, purines are major chemical constituents of RNA and DNA. They are also present as free nucleotides, nucleosides and nucleobases. In plant, bacterial and animal cells the levels of free nucleosides and bases are usually much lower than those of nucleotides (Stasolla et al., 2003). Current techniques for assaying nucleotide-related compounds in biological fluids include capillary electrophoresis, reversedphase or ion HPLC and hydrophilic interaction liquid chromatography (HILIC) combined with absorbance or MS detection. Currently, ultra-high-performance liquid chromatography coupled with triple-quadrupole tandem MS is the preferred method for the simultaneous determination of metabolites of nucleotides (Mateos-

105

Vivas et al., 2015). However, a routine HPLC analysis using absorbance detection is useful to profile well-known nucleotides, such as ATP, ADP and AMP. Due to the presence of hydrolysing enzymes for nucleotides and nucleosides, such as non-specific acid phosphatases and nucleosidases, extraction of nucleotide-related compounds from plant material is extremely challenging. These hydrolytic enzymes are not denatured during extraction and remain active in cold methanol-extracts. These problems, which were highlighted long ago by Bieleski (1964), are often not taken into account by current day investigator. 6.1. Purine nucleotides Purine nucleotides are products of de novo and salvage purine biosynthesis. Some examples of the concentrations of purine and pyrimidine nucleotides in cultured plant cells and leaves are presented in Table 5. The results were obtained from cultured cells of A. thaliana, Nicotiana tabacum (tobacco), Lycopersicon esculentum (tomato), and Catharanthus roseus (Madagascar periwinkle) at the exponential stage of growth. For comparison, nucleotide levels of cultured C. roseus cells grown in a phosphate (Pi)-deficient medium are also shown. Table 5 also lists the levels in young leaf tissues of A. thaliana, N. tabacum, L. esculentum and Camellia sinensis (tea) (Ashihara and Crozier, 1999; Riondet et al., 2005). The profiles of nucleotide pools in all these plant materials are similar. The pools of adenine nucleotides and uridine nucleotides (¼uracil nucleotides) are always larger than the guanine nucleotide pool. The cytidine nucleotide pool is the smallest of these pools. The nucleoside triphosphate levels are usually much higher than those of nucleosidemono- and diphosphates. The intracellular levels of nucleotides in plant cells are influenced by several environmental factors. For example, Pi starvation

Table 5 Concentration of purine and pyrimidine nucleotides in plant cells and tissues. The values shown are obtained from the cultured cells at the exponential growth phases and young leaves. For comparison, the values from phosphate (Pi)-deficient Catharanthus roseus cells are also shown. Nucleotide level (nmol/g f.w.)

ATP ADP AMP GTP GDP GMP IMP XMP UTP UDP UMP CTP CDP CMP Relative ratec SAN (%) SGN (%) SUN (%) SCN (%) Energy status ECd ATP/ADP a b c d

Arabidopsis thalianaa

Nicotiana tabacuma

Lycopersicon esculentuma

Camellia sinensisb

Catharansus roseusb

Cells

Leaves

Cells

Leaves

Cells

Leaves

Leaves

Cells þPi

Pi

145 19 17 26 e tr e e 18 tr 178 9 tr e

40 77 59 tr e 9 8 e 20 18 81 tr e tr

124 8 9 23 e tr e e 49 e 187 4 e e

48 23 33 tr e e 20 e 7 23 74 12 9 tr

190 38 e 24 e 10 e e 27 59 174 e e e

50 30 55 tr e e 20 e 6 16 50 e e tr

177 35 14 26 tr tr e 29 88 46 55 tr tr 32

100 18 16 26 3 3 e e 55 29 33 7 4 6

22 16 12 8 3 2 e e 10 22 24 2 2 3

44 6 48 2

56 3 38 0

35 6 59 1

42 0 42 8

44 7 50 0

60 0 0 32

48 6 40 7

45 11 39 6

40 10 44 6

0.85 7.6

0.45 0.5

0.91 14.9

0.57 2.0

0.92 5.1

0.48 1.7

0.86 3.3

0.81 5.6

0.60 1.4

Data are taken from Riondet et al. (2005). Ashihara and Crozier (1999). Relative rates are shown as % of total nucleotides. SAN, SGN, SUN and SCN mean total adenine, guanine, uracil and cytosine nucleotides, respectively. EC means “adenylate energy charge” proposed by Atkinson (1977). tr, trace; , not detectable.

106

H. Ashihara et al. / Phytochemistry 147 (2018) 89e124

Table 6 Developmental and physiological events associated with purine salvage enzyes. Events

Section in text

Plant materials

Seed germination

8.1.1

Wheat embryo Yellow lupin cotyledons Yellow lupin cotyledons Catharanthus roseus cells Catharanthus roseus cells Carrot cells White spruce cells White spruce embryos White spruce embryos White spruce embryos White spruce explants Peach bud Straw berry Catharanthus roseus cells Catharanthus roseus cells Avicennia marina leaves Barley root Potato tubers Arabidopsis thaliana Tobacco cells Brassica napus Brassica napus Brassica napus Camellia sinensis Arabidopsis thaliana Arabidopsis thaliana Potato tubers Arabidopsis thaliana Arabidopsis thaliana Rice Tobacco Arabidopsis thaliana

Seed formation and germination Cell growth

Somatic embryo drying Somatic embryo development Somatic embryo germination Somatic organogenesis Dormancy break

8.1.1 8.1.2

Phosphate starvation Phosphate starvation recovery Salt stress Iron-deficiency Wound stress Oxidative stress Programmed cell death Effect of glutathione Effect of brassinolide Embryogenic capacity Caffeine biosynthesis Cytokinin inactivation Cytokinin interconversion Endodormancy Exogenous cytokinin transport Male sterility

8.1.3 8.1.3 8.1.3 8.1.3 8.1.4 8.1.4 8.1.5 8.1.5 8.1.6 8.1.7 8.1.6 8.1.9 8.1.10 8.1.11 8.1.12 8.1.13 8.2 8.3.1 8.3.2 8.3.3 8.3.4 8.4

Plant pathogen response Root gravitropism

8.5 8.6

Purine base salvage

Purine nucleoside salvage

Reference

APRT, HGPRT,

AK AK, NPT ARN

Price and Murray (1969) Guranowski and Barankiewicz (1979) Guranowski and Pawelkiewicz (1978) Shimazaki et al. (1982) Hirose and Ashihara (1984a) Ashihara and Nygaard (1989) Ashihara et al. (2000) Stasolla et al. (2001b) Ashihara et al. (2001) Stasolla et al. (2001a) Stasolla et al. (2006) Faye and Le Floc'h (1999)
tel (2000) Robert and Pe Shimano and Ashihara (2006) Yin et al. (2007) Suzuki et al. (2003) Itai et al. (2000) Katahira and Ashihara (2006b) Sukrong et al. (2012) Stasolla et al. (2005) Belmonte et al. (2003) Belmonte et al. (2011) Elhiti et al. (2011) Koshiishi et al. (2001) Zhang et al. (2013) Schoor et al. (2011) Bromley et al. (2011) Bishop et al. (2015) Moffatt and Somerville (1988) Zhou et al. (2006) Wang et al. (2003) Young et al. (2006)

APRT APRT, HGPRT APRT APRT APRT APRT APRT APRT APRT APRT APRT APRT APRT APRT APRT APRT APRT APRT APRT APRT APRT

AK, AK AK, AK, AK, AK, AK, AK AK AK, AK, AK,

ARN IGK, IGK, IGK, IGK, IGK,

NPT, ARN NPT, ARN NPT NPT, ARN ARN

ARN ARN ARN

AK, ARN AK AK, AK, AK, AK,

NPT, ARN IK IK ARN

AK CNP APRT APRT APRT AK AK

Abbreviations of enzymes: AK, adenosine kinase; APRT, adenine phosphoribosyltransferase; ARN, adenosine nucleosidase; HGPRT, hypoxanthine/guanine phosphoribosyltransferase; IGK, inosine/guanosine kinase; NPT, nucleoside phosphotransferase, CNP, cytokinin nucleoside phosphorylase.

of cultured plant cells results in a marked decrease in the levels of nucleotides, especially nucleoside triphosphates (Table 6) (see also Section 8.1.5). The proportion of adenylates, i.e. the adenylate energy charge ([ATP] þ ½[ADP])/(ATP þ ADP þ AMP) originally proposed by Atkinson (1977), in actively growing plant cells is maintained at an almost constant 0.8e0.9. The 0.86 energy charge in young tea leaves is within this range, but leaves of A. thaliana, tobacco and tomato are lower at 0.5e0.6 because of the higher concentrations of ADP and AMP. While the energy charge fluctuates during diverse growth conditions, some changes might be ascribed to the presence of hydrolytic enzymes in the extract. The ATP/ADP ratio is also used as a major parameter of interest in investigations of metabolic aspects of bioenergetics. The ratios are usually high in actively growing cells and tissues. 6.2. Purine bases and nucleosides Purine bases and nucleosides are substrates of purine salvage enzymes. It is well-known that bacteria and animal cells do not contain significant cellular pools of nucleobases. This could be caused by the high activities of salvage enzymes resulting in a rapid conversion of purine bases into their respective nucleotides (Henderson and Paterson, 1973; Wagner and Backer, 1992). In many plant cells, adenine and guanine pools are also markedly smaller than those of adenine and guanine nucleotides (Ashihara et al., 1990; Brown, 1963; Grzelczak and Buchowicz, 1975). For example, the adenine and guanine content of wheat grain are 4 and 5 nmol/g d.w., respectively (Grzelczak and Buchowicz, 1975). Shimano and Ashihara (2006) reported that only trace amounts of adenine and guanine are found in growing

Catharanthus roseus cells, but these purine bases accumulate after long Pi starvation when adenine nucleotide contents are reduced (see Section 8.1.5). A much higher content of adenine, up to 300 nmol/g d.w., has been reported in the youngest upper leaves of barley (Sawert et al., 1988) and in soybean seeds (Yokozawa and Oura, 1986). Cellular pool sizes of purine nucleoside are usually smaller than those of purine nucleotides. However, compared with purine bases, sizeable amounts of purine nucleosides have been observed in plants (Ashihara et al., 1990; Grzelczak and Buchowicz, 1975; Sawert et al., 1988; Shimano and Ashihara, 2006). For example, the adenosine, guanosine and xanthosine content of mature pea seeds are 152, 67 and 91 nmol/g d.w., respectively (Brown, 1963). Accumulation of nucleosides occurs in some cereal leaves. In young wheat leaves, the pool size of adenosine (300 nmol/g d.w.) is similar to that of ATP (280 nmol/g d.w.) (Sawert et al., 1988). Nucleosides may be stored in vacuoles of plant cells, as excess cytosolic nucleosides can have metabolic consequences (Oikawa et al., 2011). 6.3. Cytokinin bases, nucleosides and nucleotides Cytokinin bases and nucleosides are substrates of purine salvage enzymes (see Section 3). Compared to purine nucleotides (Table 5), the concentration of cytokinins is very low. The cytokinin bases, nucleosides and nucleotides in A. thaliana seedlings
k et al. (2008) shown below is as an example. reported by Nova The concentration of isopentenyladenine cytokinins, isopentenyladenine, isopentenyladenine riboside and isopentenyladenine ribotide are 0.4, 2.4 and 5.7 pmol/g f.w.,

H. Ashihara et al. / Phytochemistry 147 (2018) 89e124

respectively. While those of trans-zeatin cytokinins, trans-zeatin, trans-zeatin riboside and trans-zeatin ribotide are 0.6, 2.8 and 1.6 pmol/g f.w., respectively. Levels of dihydrozeatin and N6-benzyladenine cytokinins are in the low femtomolar range (<20 fmol). It is noteworthy that these cytokinin concentrations are much lower than the Km values of most adenine phosphoribosyltransferase and adenosine kinase enzymes (see Section 3.1.1 and 3.2.1). The possible function of cytokinin conversion by purine salvage enzymes is discussed in Section 8.1. 6.4. 5-Phosphoribosyl-1-pyrophosphate PRPP is a substrate of adenine phosphoribosyltransferase and hypoxanthine/guanine phosphoribosyltransferase (see Section 3.1). The pool size of PRPP is low compared with most purine nucleotides (Table 5) although the turnover of PRPP is rapid. For example, PRPP levels varied from 0.41 to 2.2 nmol/g f.w. during the growth of cultured C. roseus cells, but the rates of PRPP production in planta fluctuated from 33 to 125 nmol/h/g f.w. (Hirose and Ashihara, 1983c). 7. Transport of purine bases and nucleosides 7.1. Uptake of purine bases and nucleosides by plant cells and tissues Plant cells and tissues take up exogenously supplied purine bases and nucleosides. Fig. 7 shows the relative uptake velocities of a series of purine bases and nucleosides in four different plant materials. The uptake was estimated by summation of

Fig. 7. The uptake of 14C-precursors by suspension-cultured cells of Arabidopsis thaliana (Yin et al., 2014), Catharanthus roseus (Hirose and Ashihara, 1983b), discs of growing potato tubers (Katahira and Ashihara, 2006a), and segments of tea leaves (Deng and Ashihara, 2010). Samples (100e200 mg f.w.) were incubated with 10 mM 14 C-precursors in 2 ml potassium phosphate buffer or Murashige-Skoog culture medium (pH 5.7). Incubation time was 1e10 h as shown in the figure. Rates of uptake are linear during the incubation periods. The values are shown as percentage of the rate of adenine uptake (control). * Data are not available.

107

radioactivity incorporated into soluble and insoluble materials of cells and tissues and released 14CO2, and relative values are expressed as percentage of the uptake rate of adenine. The purine compounds are taken up at differing rates during the incubation period. In suspension cultured cells of A. thaliana (Yin et al., 2014) and Catharanthus roseus (Hirose and Ashihara, 1983b), uptake rates of hypoxanthine, guanine, adenosine, inosine and guanosine are similar to that of adenine, but in the disks of potato tubers (Katahira and Ashihara, 2006a) and tea leaves (Deng and Ashihara, 2010), a significant reduction in the uptake of hypoxanthine, guanine, inosine and guanosine is observed. The uptake rate of xanthine and xanthosine is lower than that of adenine in all plant materials examined. An extremely rapid uptake of deoxyadenosine is found in potato tubers but not tea leaves. Arguably, these results suggest that different transport systems exist for different purine compounds in different plant species. 7.2. Transporters of purine bases and nucleosides There are different families of transporters for purine bases and nucleosides (Girke et al., 2014). One of the purine base transport proteins is the purine permease transporter (PUP) mediating intra- and intercellular transport of purine bases including cytokinins. A family of high-affinity transporters for adenine, cytosine, and purine derivatives has been detected in A. thaliana (Gillissen et al., 2000). Complementation of a yeast mutant deficient in adenine uptake (fcy2) with a cDNA expression library of A. thaliana enabled the identification of a gene encoding purine permease1 (AtPUP1) which belongs to a large gene family (AtPUP1 to AtPUP15) encoding a new class of small, integral membrane proteins. AtPUP1 transports adenine and cytokinins with high affinity. Uptake is energy dependent and occurs against a concentration gradient. Competition studies show that some purine derivatives including hypoxanthine, zeatin and caffeine are potent inhibitors of adenine and cytosine uptake. Inhibition by cytokinins is competitive indicating that cytokinins are transported by this system. Desimone et al. (2002) characterized a transporter, ureide permease (UPS) in A. thaliana for allantoin and other oxoderivatives of heterocyclic nitrogen compounds such as xanthine and uric acid. The transporter is a member of a family of plasma membrane-localized plant nucleobase transporters. Xanthine, allantoin and uracil, but not adenine, act as substrates for the transporter. The Km values for xanthine of AtUSP1 and AtUSP2 are 7 and 24 mM, respectively (Schmidt et al., 2006). The nucleobase-ascorbate transporters (NATs), also known as nucleobase: cation-symporter 2 (NCS2) proteins, represent the largest and most conserved class of nucleobase transporters in plants (De Koning and Diallinas, 2000). Identification and expression analysis of twelve members of the nucleobaseeascorbate transporter (NAT) gene family in A. thaliana was carried out by Maurino et al. (2006). Niopek-Witz et al. (2014) characterized two NAT proteins, NAT3 and NAT12 from A. thaliana after their heterologous expression in Escherichia coli UraA knockout mutants. Both proteins were shown to transport adenine, guanine and uracil with high affinities. The apparent Km values for adenine and guanine were 10.1 mM and 4.9 mM for NAT3 and 1.7 mM and 2.4 mM for NAT12. Competition studies with the three substrates suggest hypoxanthine as a further substrate of both transporters. Since the transport of nucleobases is inhibited markedly by low concentrations of a proton uncoupler, NAT3 and NAT12 appear to act as protonenucleobase symporters. Other similar transporters, nucleobase symporter 1 (NCS1) (Witz et al., 2014) and AZGA-like protein-like transporter (Mansfield et al.,

108

H. Ashihara et al. / Phytochemistry 147 (2018) 89e124

2009) have been found to contribute to adenine and guanine transport. In contrast to nucleobase transport, nucleoside transport in plants is mediated by equilibrative nucleoside transporters (ENTs). Biochemical characterization of some ENTs from A. thaliana (AtENT) (Li et al., 2003; Wormit et al., 2004) and rice (OsENT) (Hirose et al., 2005) have been reported. Substrate specificity of each AtENT is slightly different, and, in addition to purine nucleosides (adenosine and guanosine), the ENTs transport pyrimidine nucleosides (cytidine, uridine and thymidine). Km values of AtENTs range from 3.0 to 15.5 mM for adenosine and 7.3e18.0 mM for guanosine. AtENT6 transports cytokinins. The Km values for isopentenyladenine riboside and trans-zeatin riboside are 17 and 630 mM, respectively (Hirose et al., 2008). OsENT2 also transports adenosine, isopentenyladenine riboside and trans-zeatin riboside. The Km values for these substrates are 3, 32 and 660 mM, respectively (Hirose et al., 2005). The high Km value for transzeatin riboside indicates that it is difficult to transport this compound in planta assuming OsENT2 are the only transporters capable of transfer of t-zeatin. Information on the substrate specificity of all transporters is fragmentary and the Km values mentioned above are provided from the recombinant proteins expressed in yeast or E. coli. Thus, the data may be not necessarily reflect the properties of the native transporters in plants. Characterization of transporters isolated from plant membranes is needed to elucidate an understanding of the transport system operating in planta. 8. Physiological aspects of purine salvage in plants Purine nucleotide synthesis by the salvage pathways is energetically favourable for cells, since the salvage of one mole of purine base or purine nucleoside requires only one mole of PRPP or ATP for nucleoside monophosphate synthesis. Purine bases and nucleosides released as nucleotide and nucleic acid degradation products are recycled by these pathways. If purine bases or nucleosides are available in plant cells, purine salvage reactions are preferentially utilised over de novo purine biosynthesis. De novo biosynthesis is inhibited by AMP which is synthesized by adenine salvage (Hirose and Ashihara, 1984b). Operation of the salvage pathway also reduces the intracellular levels of purine bases and nucleosides which are inhibitory to other metabolic reactions. For example, in the SAM cycle, adenosine is produced by the SAH hydrolysis (Fig. 6D). SAH hydrolase catalyses a reversible reaction that is favoured to move in the direction of SAH hydrolysis by removal of adenosine. Removal of adenosine by adenosine kinase or adenosine nucleosidase is necessary to accelerate the cycle (Moffatt and Weretilnyk, 2001; Suzuki et al., 2003). Specific examples are shown in Section 8.1.6. The purine bases have very limited solubility in water. Nucleosides and nucleotides have greater solubility (Dawson et al., 1986), due, respectively, to the presence of polar sugars, or sugars and charged phosphate groups. Therefore, purine base salvage may contribute to the solubilisation of purine compounds. In addition, functional analysis of genes encoding purine salvage enzymes revealed additional unique functions: for example, adenine phosphoribosyltransferase deficient mutants are male-sterile (Section 8.4) (Moffatt and Somerville, 1988) and adenine phosphoribosyltransferase reduces the excess accumulation of the active form of cytokinins (Section 8.3.1) thereby affecting cytokinin activity. Here, we introduce research which shows that purine salvage activity is accompanied by a series of physiological events and explain possible metabolic reasons for this. Physiological events associated with purine salvage enzymes are listed in Table 6.

8.1. Changes in purine salvage activity accompanied by physiological events 8.1.1. Seed germination and purine salvage Changes in purine salvage activity during germination of seeds have been investigated in some Fabaceae species, including castor bean (Ricinus communis) and wheat (Triticum aestivum). One of the earliest detectable metabolic processes occurring in imbibed seeds is the rapid increase of ATP. In some instances the increase in ATP is accounted for by a decrease in AMP yet in many cases there is insufficient AMP to account for all of the ATP that accumulates during early stages of germination. It has, however, been shown that the net increase in adenine nucleotides is due mainly to the salvage of adenosine and adenine catalysed by adenosine kinase and adenine phosphoribosyltransferase (Anderson, 1977a, b). In a Fabaceae plant, black gram (Phaseolus mungo), the relative rate of purine salvage and its degradation in the cotyledons (a storage organ) and embryonic axes were determined by 14C-tracer experiments during germination (Ashihara, 1983). In the cotyledons, the salvage of adenine, adenosine and guanine, assessed through the incorporation of 14C in the nucleotides and nucleic acids. is higher 12e48 h after imbibition and the initiation of germination, than at 0e12 h and 48e96 h. In contrast, appreciable ureide (allantoin and allantoic acid) formation from purines is found after 48 h of germination. In the embryonic axes, a higher incorporation of 14C-labelled purines into salvage products is observed during the early stage of germination, whereas significant incorporation into ureides (allantoin and allantoic acid) is detected in the later stage. The pathways of ureide synthesis from purines are shown in Fig. 1. 14C-Labelled adenine taken up by the cotyledons during imbibition is translocated to the embryonic axes where the radioactivity is associated with ureides. The results suggest that purine derivatives stored in the cotyledons of black gram seeds are utilised for nucleotides and nucleic acids especially during the early stage of germination, and that they are degraded to allantoin and allantoic acid by the conventional purine catabolic pathway in the later stages and reutilized as a nitrogen source for seedling growth. The level of PRPP is extremely low in dry seeds, but after imbibition increase rapidly in cotyledons (Ross and Murray, 1971; Ashihara and Kameyama, 1989). Production of purine nucleotides via the salvage pathway during the early stages of germination of Fabaceae seeds occurs mainly at the expense of adenine, adenosine, guanine, and guanosine which are stored in the cotyledons (Brown, 1963). During germination of yellow lupin seed and black gram salvage of purines is catalysed predominantly by adenine phosphoribosyltransferase and adenosine kinase (Guranowski and Barankiewicz, 1979; Nobusawa and Ashihara, 1983). The alternative route of adenine salvage, purine nucleoside phosphorylase and adenosine kinase, is unlikely to operate in seeds because they seemingly lack purine-nucleoside phosphorylase activity (Guranowski and Barankiewicz, 1979). During the early phases of germination, there are indications that adenosine salvage is catalysed by adenosine kinase (step 1 in Fig. 1). The activity of this enzyme increased in the cotyledons during germination of legume seeds (Guranowski and Barankiewicz, 1979; Nobusawa and Ashihara, 1983). The adenosine salvage by adenosine nucleosidase and adenine phosphoribosyltransferase (steps 2 and 3 in Fig. 1) becomes operative only during the later stages of germination, because adenosine nucleosidase activity is absent in dry seeds (Guranowski and Pawelkiewicz, 1978). The contribution of nucleoside phosphotransferase seems to be important only in the cotyledons during the later period of germination (Guranowski and Barankiewicz, 1979). This suggests that purine salvage enzymes

H. Ashihara et al. / Phytochemistry 147 (2018) 89e124

109

adenine phosphoribosyltransferase and adenosine kinase may contribute to rapid ATP synthesis and the subsequent nucleic acid synthesis. Alternative enzymes, adenosine nucleosidase and nucleoside phosphotransferase, may be related to the salvage of purines which is derived from catabolites of nucleic acids and nucleotides in senescing cotyledons. The pattern of purine salvage in wheat embryos is slightly different from that occurring in Fabaceae seeds. The activities of adenine phosphoribosyltransferase and hypoxanthine/guanine phosphoribosyltransferase increase during imbibition, while adenosine kinase and PRPP synthetase activities are high in extracts from dry embryos (Price and Murray, 1969). Castor bean is a Euphorbiaceous plant, in which the endosperm is a major storage organ. In germinating castor bean seedlings, purine nucleosides and bases are transported from the endosperm to cotyledons (Kombrink and Beevers, 1983). Therefore, in growing parts of seedlings, purine salvage occurs using these transported precursors. In the later stage of germination, degradation of nucleic acids and nucleotides takes place accompanied by senescence of endosperm and development of growing tissues. Ureides may be transported to newly growing parts of the seedlings and used as nitrogen sources. The total content of nucleic acids and nucleotides declines rapidly between day 4 and day 8 of seedling development. The function of adenosine transported into developing cotyledons from endosperm in castor bean was estimated from the effect of exogenously supplied adenosine in cotyledons of 5-day-old seedlings after removal of the endosperm (Florchinger et al., 2006). Most of 14C-labelled adenosine administered in detached cotyledons was incorporated into nucleotides, RNA and DNA. Import and conversion of adenosine improved the energy content of cotyledons as revealed by a substantially increased ATP/ADP ratio. This effect was accompanied by increases in respiratory activity and a stimulation of glycolytic flux by activation of phosphofructokinase. Adenosine also stimulates the biosynthesis of starch. Thus, adenosine imported from mobilizing endosperm into developing castorbean cotyledons fulfils an important function as it promotes anabolic reactions in this rapidly developing tissue. A relationship between purine salvage activity and germination has been suggested from a genetic perspective. It has been reported that the APT-deficient mutants of A. thaliana germinate and grow more slowly than the wild-type (Moffatt and Somerville, 1988). Furthermore, functional analysis of hypoxanthine/guanine phosphoribosyltransferase (HGPT) revealed that HGPT expression is dynamically regulated in growth during seed germination (Liu et al., 2007). The germination rates of wild-type and knockout mutant (athgpt1-1) seeds of A. thaliana are similar during the early stages (up to 48 h after sowing) but the rate is ~20% reduced in the mutant after 60 h. While the germination rates (20e40%) of the overexpression mutants (Ho3-8 and Ho4-4) are significantly higher than the rate of wild-type (~10%) 36 h after imbibition. The contrasting germination rates exhibited by wild-type and the AtHGPT knockout and overexpression mutants are probably caused by the differences in AtHGPT enzyme activity levels among the three genotypes. Although no direct measurement of enzyme activity was performed this is supported indirectly by the differential sensitivities of the three genotypes to the cytotoxic compound 8azaguanine which can be efficiently salvaged through HGPT enzyme activity. These mutant studies support the theory that salvage reactions catalysed by adenine phosphoribosyltransferase and hypoxanthine/guanine phosphoribosyltransferase contribute to seed germination as proposed in early physiological studies.

carrots. The growth of C. roseus cells is divided into (i) the lag, (ii) cell division, (iii) cell expansion, and (iv) stationary phases. Adenylate levels in C. roseus cells changed markedly during growth. A rapid increase in ATP content is observed in the lag phase. The activity of adenine phosphoribosyltransferase also increases markedly during the lag phase of cell growth. The adenine nucleotide content decreases concomitantly with cell division and then increases again as cells expanded (Shimazaki et al., 1982). The activities of adenine phosphoribosyltransferase, hypoxanthine/guanine phosphoribosyltransferase and adenosine kinase increase during the lag phase of the cells and then rapidly decrease. The activities increase again in the stationary phase. The patterns of change in the activities of several degradation enzymes including DNase, RNase, acid phosphatase, 30 - and 50 -nucleotidases, and purine nucleosidase decrease soon after stationary phase cells are transferred to a fresh medium and reached a minimum during the cell division and expansion phases. The activities then increase during the stationary phase (Hirose and Ashihara, 1984a). The purine salvage activity was investigated with the de novo biosynthesis of purine nucleotides in cultured carrot cells (Ashihara and Nygaard, 1989). The highest rates of incorporation of 14C-labelled adenosine into nucleotides occurred when the stationary cells are transferred to the fresh culture medium. This suggests that purine salvage capacity is retained at the late growth stages, while de novo synthesis is dominant in the cell division and expansion phases. This pattern is also reflected by increased levels of PRPP amidotransferase (EC 2.4.2.14) involved in de novo purine synthesis which occurs only at the cell division phase (Ashihara and Nygaard, 1989). In addition to purine salvage, a similar activation of pyrimidine salvage has been also found in C. roseus cells (Kanamori-Fukuda et al., 1981). The effect of various purine bases and nucleosides on the growth of mutant Datura innoxia cells which lack de novo purine biosynthetic activity has been examined (Ashihara et al., 1991; King et al., 1980; Mitsui and Ashihara, 1988). Originally designated as an adenine-requiring auxotroph (King et al., 1980), the mutant cell line (Ad1) is useful for investigating the importance of individual purine nucleosides and bases for growth. The relative growth rates with these purines are as follows; adenosine (100) > inosine (89) > adenine (85) > 5-aminoimidazole-4-carboxamide riboside (AICAR) (80). In contrast, no growth is detected with hypoxanthine, guanosine, guanine and 5-aminoimidazole-4-carboxamide (AICA) (Ashihara et al., 1991). 14C-Feeding experiments suggest that the Ad1 cells can readily salvage adenosine, adenine, inosine, guanosine and guanine for the biosynthesis of nucleotides and nucleic acids. Conversion of adenine nucleotides to guanine nucleotides occurs, probably via the reaction catalysed by AMP deaminase (step 11 in Fig. 1). However, the reverse conversion, guanine nucleotides to adenine nucleotides, occurs at an extremely low rate, because of a virtual lack of GMP reductase activity (step 14 in Fig. 1). Consequently, growth of the Ad1 cells is not supported by guanine and guanosine salvage. Hypoxanthine is catabolised predominantly by the conventional purine catabolism pathways (Fig. 1). AICAR appears to be readily phosphorylated by adenosine kinase (Page, 1989), and the resultant, AICA-ribotide, an intermediate of the later steps of de novo purine biosynthesis (after step 9 in Fig. 4), seems to be utilised for the synthesis of both adenine and guanine nucleotides (steps 9e14 in Fig. 4). In contrast, AICA is not converted to AICA-ribotide by purine phosphoribosyltransferases. These results indicate that adenosine and adenine salvage occupy a central position in purine salvage for the growth of plants.

8.1.2. Cell growth and purine salvage The relationship between purine salvage and plant cell growth has been studied in cultured cells of Catharanthus roseus and

8.1.3. Purine salvage during embryogenesis and organogenesis Due to the inability to effectively dissect zygotic embryos from the maternal tissue, especially during the early stages of

110

H. Ashihara et al. / Phytochemistry 147 (2018) 89e124

Fig. 8. The developmental pathway of white spruce somatic embryos with culture treatments and sampling points. In the maintenance medium the embryogenic tissue consists of early filamentous embryos. Upon subsequent transfers of the tissue into liquid and solid media for development, the embryos increase in size. A welldeveloped shoot and root pole become visible after 21 days on solid medium. During the following days in culture, the embryos develop further, and a ring of cotyledons emerges from the shoot apical region. Fully developed embryos, observed after 36 days on solid medium, are partially dried for 10 days before being transferred on germination medium. Formation of a functional root and shoot can be observed between day 4 and 6. 2,4-dichlorophenoxyacetic acid (2,4-D); benzyladenine (BA); abscisic acid (ABA); partial drying treatment (PDT). Numbers in bold are days in culture. Adapted from Stasolla and Thorpe (2004).

embryogenesis, biochemical studies on purine metabolism have been conducted using embryos derived from somatic cells. This technique was developed using white spruce (Picea glauca) as an embryogenic model system (Stasolla and Thorpe, 2004). In this system four distinct embryogenic phases are identified: (i) maintenance of embryogenic tissue, (ii) development of somatic embryos, (iii) maturation of somatic embryos and (iv) germination of somatic embryos (Fig. 8). In the maintenance medium, the embryogenic tissue consists of early filamentous embryos composed of elongated suspensor cells subtending highly cytoplasmic cells of the embryo proper. Upon subsequent transfers of the tissue to liquid (7 days) and solid (36 days) development medium, the embryos increase in size. A well-developed shoot and root pole become visible after 21 days on solid medium. During subsequent days in culture, the embryos develop further, and a ring of cotyledons emerges from the shoot apical region. Fully developed embryos, observed after 36 days on solid medium, are desiccated by a partial drying treatment (PDT) for 10 days before being transferred onto germination medium. Formation of a functional root and shoot can be observed between day 4 and 6 (Stasolla and Thorpe, 2004). Overall, during somatic embryogenesis adenine phosphoribosyltransferase and adenosine kinase are the major purine

salvage enzymes, while there is only low activity of inosine/guanosine kinase and nucleoside phosphotransferase (Table 7). Purine salvage during the maintenance stage of embryogenic tissues was investigated using 14C-purine feeding experiments and by measuring the activity of the salvage enzymes at day 1, 4, and 7 (Fig. 8) (Ashihara et al., 2000; Belmonte et al., 2003). In situ activity of adenine and adenosine salvage in the cells of the lag (day1) and logarithmic (day 4) phases is higher than that of stationary phase cells (day 7). Inosine salvage activity is relatively low during maintenance and no significant differences in activity are found among the different growth stages. High activities of adenosine kinase and adenine phosphoribosyltransferase are detected, while the activities of inosine/guanosine kinase and nucleoside phosphotransferase are low. No measureable adenosine nucleosidase activity is found. Based on these results it is concluded that adenine and adenosine salvage in white spruce cells is mainly catalysed by adenine phosphoribosyltransferase and adenosine kinase. The contribution of adenine, adenosine and inosine salvage to purine nucleotide and nucleic acid biosynthesis during the development of the white spruce somatic embryos was examined by Ashihara et al. (2001). The salvage of adenine and adenosine is elevated during the initial stages of embryo growth, which is characterized by rapid cell proliferation, but it declines with further embryo development demarked by tissue patterning. It is concluded that adenosine salvage played a major role during the development of white spruce embryos. Studies on the metabolic fate of 14C-labelled adenosine suggested that turnover of adenine nucleotides is rapid, and some are utilised for nucleic acid synthesis. In contrast, most of 14C-inosine taken up by the embryos is not salvaged for nucleotide synthesis but rather catabolised by the conventional purine catabolic pathway via ureides (steps 8, 17e22 in Fig. 1). This trend is observed at all stages of embryo development. The imposition of a partial drying treatment on fully developed white spruce somatic embryos is a necessary maturation step required for the successful germination and conversion into plantlets (Kong and Yeung, 1995). Purine salvage activity was investigated during the partial drying treatment (Stasolla et al., 2001b). Adenine and adenosine are extensively salvaged by adenine phosphoribosyltransferase and adenosine kinase, respectively. The activity of adenine phosphoribosyltransferase increases during the PDT. Based on these results it is suggested that the imposition of the partial drying treatment might be required to increase the activity of adenine phosphoribosyltransferase, which could contribute to the enlargement of the ATP pool required to sustain growth at the onset of germination. Purine salvage during germination of somatic embryos of white

Table 7 Activity of enzymes involved in purine salvage during maintenance, development, maturation and germination of white spruce somatic embryos. Enzyme activity is expressed as pkat/mg protein. Developmental stages are shown in Fig. 8. Enzymes

Maintenance

Development Liquid medium

Solid medium

Maturation

Germination

Duration of culture (days)

1

4

7

4

7

7

14

21

28

36

5

10

2

4

6

Adenine phosphoribosyltransferase (APRT) Adenosine kinase (AK) Inosine/guanosine kinase (IGK) (inosine)a Nucleoside phosphotransferase (NPT) (adenosine)b Nucleoside phosphotransferase (NPT) (inosine)c Adenosine nucleosidase (ARN) 5-Phosphoribosyl-1-pyrophosphate synthetase (PRPPS)

245 233 12 16 9 nd e

269 1130 16 13 3 nd 75

289 1680 13 11 4 nd 30

432 968 21 12 12 nd 91

487 566 14 17 9 nd 63

476 1705 22 15 16 nd 57

732 978 28 14 15 nd 105

629 866 18 14 15 nd 97

614 845 29 19 19 nd 107

153 156 4 2 2 nd 16

172 114 1 4 2 nd 31

200 148 nd 3 2 nd 32

210 153 2 3 2 nd 22

591 232 3 3 3 5 40

593 212 2 2 2 5 27

The data are taken from Stasolla and Thorpe (2004). nd, not detected. , data is not available. a IGK activity was measured with inosine. NPT activity with adenosineb and inosinec are shown.

H. Ashihara et al. / Phytochemistry 147 (2018) 89e124

spruce was also examined (Stasolla et al., 2001a). Similar to seed germination (Ashihara, 1983), the metabolism of purine bases and nucleoside is extremely low during the early stages of somatic embryo germination. This is related to the fact that during the initial hours of germination the tissue is not fully imbibed and, therefore, the whole machinery involved in purine salvage is unlikely to be fully functional. The second stage of purine metabolism is the salvage phase, in which both adenine and adenosine are actively salvaged for nucleotide and nucleic acid synthesis. Adenine salvage activity estimated by 14C-feeding experiments increased ~3.5 times from germination days 2e6. Similar, but less pronounced changes are observed for adenosine. Stimulation of adenine salvage activity during embryo germination appears to be caused by the increased activity of adenine phosphoribosyltransferase. The adenosine kinase-mediated route (step 3 in Fig. 1) is responsible for the salvage of adenosine during maintenance of embryogenic tissue, development, and maturation of somatic embryos. In contrast, adenosine salvage via the adenosine nucleosidase and adenine phosphoribosyltransferase route (steps 2 and 3 in Fig. 1) appears to be operative after day 4 in the germination medium, since adenosine nucleosidase activity is observed only during the later stages of somatic embryo germination as is found in seed germination (Section 8.1.1.). Collectively, it appears that extensive utilization of adenine and adenosine for the synthesis of ATP and other salvage products appears to be critical for the successful germination of white spruce somatic embryos. This concept was reinforced by Stasolla and Yeung (1999) who showed that applications of ascorbic acid increased the germination frequency of the somatic embryos, and enhanced the intake of adenine and adenosine throughout the germination period. Furthermore, the higher enzymatic activities of adenosine kinase and adenine phosphoribosyltransferase in the ascorbic acid-treated embryos are responsible for a larger proportion of adenine and adenosine being salvaged in the treated embryos compared to their control counterparts. Thus, a positive correlation appears to exist between the ability to salvage adenine and adenosine and successful embryo conversion (Stasolla et al., 2001a). Purine salvage was also investigated during white spruce organogenesis using similar experimental methods to those outlined above (Stasolla et al., 2006). White spruce epicotyl explants cultured on shoot-forming (SF) medium utilised adenine and adenosine for nucleotide and nucleic acid synthesis more effectively than tissue cultured on non-shoot forming (NSF) medium. High levels of salvage products are detected in the shoot-forming tissue after 10 days in culture, when shoot formation is initiated along the epicotyl axis of the explants. Such a differential utilization of purine precursors is ascribed principally to the higher specific activity of adenine phosphoribosyltransferase and adenosine kinase in the shoot-forming tissue. This indicates that higher adenine and adenosine salvage for nucleotide and nucleic acid synthesis in the shoot-forming tissue represents a physiological switch during the initiation of organogenesis. Metabolism of adenine and inosine in developing microsporederived embryos of canola (Brassica napus) was investigated by Ashihara et al. (2008a). Due to the difficulty of isolating and manipulating canola zygotic embryos, investigations on biochemical aspects of embryo development have focused on microsporederived embryogeny, the process whereby the pollen developmental program of immature microspores is re-routed towards the formation of haploid embryos. The metabolic fate of 14C-labelled purines were compared in embryos at different developmental stages: globular stage (day 10), early cotyledonary stage (day 20), late cotyledonary stage (day 25), and fully developed stage (day 35). The feeding experiments with [8-14C]adenine and [8-14C]inosine indicate that purine salvage activity is extremely high in the

111

globular stage and decreases gradually until the late cotyledonary stage, before increasing again in the fully developed embryos. Overall it appears that purine salvage is required in support of active growth during the initial phases of embryogenesis and at the end of the maturation period, in preparation for post-embryonic growth. 8.1.4. Dormancy break and purine salvage Temperate perennial plants are dormant during winter. This resting state, which typically is initiated during late autumn and early winter, is broken by low temperatures. Plants recover their growth potential, facilitating the initiation of vegetative growth the following spring. This change appears to be caused by an increase in nucleoside triphosphates. French researchers estimated the dormancy break by the “nucleotide test”, namely, the ability of isolated buds to increase the nucleoside triphosphate pool subsequent to an exogenous supply of adenosine (Gendraud, 1975). This ex vivo adenosine salvage ability test estimates the intensity of the inhibition of growth of the isolated bud. The nucleotides test was applied to the buds of several species including ash (Barnola et al., 1986), peach (Balandier et al., 1993; Faye and Le Floc'h, 1999; Le
tel, Floc'h and Faye, 1995) and strawberry plants (Robert and Pe 2000). Peach seedlings have adenosine kinase, adenosine nucleosidase and adenine phosphoribosyltransferase activity, hence, adenosine salvage may be performed by the direct (adenosine kinase) (step 1 in Fig. 1) and the indirect (adenosine nucleosidase and adenine phosphoribosyltransferase) (steps 2 and 3 in Fig. 1) pathways. In the case of the “nucleotide test”, with isolated vegetative buds of the trees, the indirect route appeared to be the salvage pathway of exogenously supplied adenosine in situ, because the radioactivity from [U-14C]adenosine was found predominantly in adenine moiety, and not in ribose of adenine nucleotides. However, there is evidence that the activity of adenosine kinase is closely related to the dormancy-break. Variations in activity of adenosine kinase in peach buds during four autumn-winter periods showed high increases in specific activity in late December/early January in flower buds, but in late February/early March in vegetative buds. These rises in enzyme activity coincided with the actual start of growth. After dormancy release, the enzyme activity increased as soon as buds are subjected to favourable temperature conditions, whereas their rise appeared much later in conditions of cold deprivation. In peach, adenosine kinase can be considered as a marker enzyme of the initiation of flower and vegetative bud growth (Faye and Le Floc'h, 1999). Using apical buds of strawberry, time-course activity of adenosine kinase and adenine phosphoribosyltransferase in relation to the dormancy break and the nucleotide synthesis capability was examined by Robert and Petel (2001). The results showed the changing pattern of the two purine salvage enzymes is different. An increase in adenine phosphoribosyltransferase activity is found during winter re-acquisition of nucleotide synthesis capability while an increase in adenosine kinase activity occurs during spring growth of the strawberry plants. These results suggest that adenine phosphoribosyltransferase activity is a good marker of the break of strawberry dormancy, while adenosine kinase is a marker of subsequent spring growth. The involvement of cytokinin nucleoside phosphorylase in endodormancy of potato tubers is discussed in Section 8.3.3. 8.1.5. Phosphate and purine salvage There are some reports on phosphate and purine salvage in suspension-cultured plant cells. Among the many inorganic nutrients, inorganic phosphate (Pi) has been shown to participate in many metabolic processes affecting plant cell growth and

112

H. Ashihara et al. / Phytochemistry 147 (2018) 89e124

development. Rapid uptake of Pi by the stationary cultured cells is generally achieved immediately after inoculation into fresh medium (Wylegalla et al., 1985; Ashihara and Ukaji, 1986). In C. roseus cells, the absorption of Pi increases for at least 6 h after Pi-addition and more than 85% of the absorbed Pi is present as free Pi and the rest is converted into organic phosphates within the cells. The ATP level in the cells is increased by Pi and a maximum level is observed when the initial medium Pi concentration is 10e20 mM. During the initial 3 h of incubation with 10 mM Pi, the ATP level is 4e5 times that of the Pi-starved control cells. Since only a slight decrease in ADP and AMP level is observed with the Pi-treatment, the marked increase in ATP level with Pi seems to depend upon the net synthesis of adenine nucleotides (Ashihara and Ukaji, 1986). When the stationary-phase cells of C. roseus cells are transferred to fresh complete and Pideficient medium and incubated for 24 h, the level of ATP and GTP in the Pi-sufficient cells are respectively 4.6 and 3.4 times higher than in Pi-deficient cells. Simultaneously, the adenine and adenosine salvage activity estimated by the 14C-feeding experiments is 9.7 and 3.0 times higher in the previously Pi-treated cells. Therefore, adenine and adenosine salvage are the most likely candidate for this increased purine nucleotide content (Ashihara et al., 1988) and this increase is caused by an increase in activity of adenine phosphoribosyltransferase and adenosine kinase (Hirose and Ashihara, 1984a). In addition to the intracellular level of ATP, the level of PRPP, a substrate of adenine phosphoribosyltransferase, also increases substantially during the 24 h after the addition of Pi. Turnover of PRPP (see 6.3) is also elevated by addition of Pi (Ukaji and Ashihara, 1987). The effect of long-term Pi starvation of up to 3 weeks on the levels of purine nucleotides and related compounds was examined using suspension-cultured C. roseus cells. Levels of adenine and guanine nucleotides, especially ATP and GTP, were markedly reduced during Pi-starvation. There was an increase in the activity of RNase, DNase, 50 - and 30 -nucleotidases and acid phosphatase, which may participate in the hydrolysis of nucleic acids and nucleotides. Accumulation of adenosine, adenine, guanosine and guanine was observed during the long-term Pi starvation. Longterm Pi starvation markedly depressed the flux of transport of exogenously supplied 14C-labelled adenosine and adenine, but these labelled compounds which were taken up by the cells were readily converted to adenine nucleotides even in Pi-starved cells, in which RNA synthesis from these precursors is significantly reduced. The activities of adenosine kinase, adenine phosphoribosyltransferase and adenosine nucleosidase are maintained at a high level in long-term Pi starved cells (Shimano and Ashihara, 2006). Involvement of rapid nucleotide synthesis in recovery from Pi starvation of C. roseus cells has been demonstrated by Yin et al. (2007). Growth of suspension-cultured C. roseus cells ceased during Pi starvation, but recommenced upon addition of Pi even after long-term starvation. The feeding experiments with 33Pi show the most heavily labelled organic compounds are nucleotides, followed by sugar phosphates in the Pi-starved cells. The RNA, protein, and free nucleotide content all decrease gradually during Pi starvation, however, these compounds, especially nucleotides, increase markedly in the 24 h after addition of Pi. These responses are found in short- and long-term Pi-starved cells, although the total amounts of these compounds are lower in the long-term Pi-deficient cells. Of the nucleotides, a large increase is observed in nucleoside triphosphates rather than nucleoside mono- and diphosphates. The transcript level of phosphate transporter and the activities of acid phosphatase, 50 - and 30 nucleotidase, and adenosine nucleosidase are all reduced by the addition of Pi. In contrast, the activity of adenine

phosphoribosyltransferase is markedly increased in the Pi-fed cells. Little or no increase is observed in adenosine kinase. Several studies have suggested that expression of many genes depends on the Pi level in plants. For example, the phosphate transporter gene is strongly expressed in Pi-deficient plants (Dong et al., 1999; Kai et al., 2002), probably an adaptation to obtain more Pi from outside the cells. The synthesis of many mRNAs still occurs in Pi-deficient cells. However, levels of nucleotide triphosphates, which are substrates for RNA synthesis, are decreased markedly in the cells, and net synthesis of RNA must, therefore, be greatly reduced by Pi-deficiency. This implies that studies on the gene expression and investigation at the metabolite level are both important in determining the Pi-dependent mechanism of control of cell growth. Based on the 33P feeding study, it is speculated that an initial event in recovery from Pi starvation involves the synthesis of nucleotides, especially nucleoside triphosphates. In the Pi-starved cells, AMP and ADP dominate the adenine nucleotides, and the adenylate energy charge is much lower than in growing cells (Shimano and Ashihara, 2006). When Pi is supplied to the Pideficient culture ATP is formed from ADP and Pi, mainly by respiration (stage 1). Turnover, i.e. utilization and regeneration, of ATP is accelerated, and phosphate groups at the g-position of ATP molecules are transferred to various nucleoside mono- and diphosphates and free sugars, so that many nucleoside triphosphates and sugar phosphates are formed (stage 2). The net increase in nucleotides is initiated by salvage reactions utilizing the nucleosides and nucleobases which accumulate in the Pistarved cells (Shimano and Ashihara, 2006). ATP and PRPP are used for nucleoside and nucleobase salvage reactions by kinases and phosphoribosyltransferases, respectively. Once the total nucleotide level has increased, it triggers the net synthesis of RNA and protein (stage 3). It has been reported that free amino acids accumulate in Pi-starved cells where the protein synthesis is limited (Ukaji and Ashihara, 1987), so that no amino acid synthesis may be required for protein synthesis during the early phase of recovery. DNA duplication then takes place and cell division begins (stage 4). A greater increase in nucleotide level and a net increase in RNA and protein content and in cell numbers are detected in the 24 h Pi-fed cells. A possible sequence of metabolic activation beginning from Pi starvation is shown in Fig. 9. The activities of hydrolysing enzymes, APase, and nucleotidases are significantly reduced in the Pi-fed cells in which the availability of free Pi increases. These findings are supported by the fact that expression of the genes of APase and some hydrolysing enzymes is rapidly reduced in Pi-fed A. thaliana seedlings (Muller et al., 2004). It is reported previously that activities of adenosine kinase, adenosine nucleosidase and adenine phosphoribosyltransferase are maintained at a high level even in long-term Pistarved cells, and serve for the limited turnover of nucleotides in these cells (Shimano and Ashihara, 2006). These enzymes may also have an important function in the resumption of cell growth via the increase in net synthesis of nucleotides during the early stage of recovery from Pi starvation. Of the three enzymes measured, adenine phosphoribosyltransferase and adenosine kinase activities increased or are maintained by the addition of Pi, whereas adenosine nucleosidase is reduced in the Pi-fed cells. This differing response suggests that adenosine nucleosidase is not involved in purine salvage but takes part in the degradation process. The present results strongly suggest that rapid reformation of nucleotides by salvage reactions are key events in recovery from Pi starvation. 8.1.6. Salt stress and adenosine salvage Some plant species accumulate glycine betaine in leaves as a compatible osmolate. Choline, a direct precursor of glycine

H. Ashihara et al. / Phytochemistry 147 (2018) 89e124

113

Fig. 9. Possible metabolic events during recovery from Pi starvation in suspension-cultured Catharanthus roseus cells. When Pi is supplied to the Pi-deficient culture, ATP is formed from ADP and Pi mainly by respiration (stage 1). Turnover of ATP is accelerated, and phosphate groups at the g-position of ATP molecules are transferred to various nucleoside mono- and diphosphates and free sugars, so that many nucleoside triphosphates and sugar phosphates are formed (stage 2). Once the total nucleotide level has increased, it triggers the net synthesis of RNA and protein (stage 3). DNA duplication then takes place and cell division begins (stage 4). Adapted from Yin et al. (2007).

betaine, is produced from ethanolamine via three SAM-dependent N-methyltransferase reactions. SAH produced in these reactions is a potent inhibitor of SAM-dependent methyltransferases (Fig. 6D). SAH is hydrolysed to homocysteine and adenosine by SAH hydrolase. In glycine betaine-forming spinach and sugar beet leaves, SAH hydrolase and adenosine kinase activities increase in leaves subjected to 100e300 mM of NaCl and these changes reflect increased levels of SAH hydrolase and ADK protein and transcripts. In contrast, tobacco and canola, which do not accumulate glycine betaine, did not show comparable changes in these enzymes (Weretilnyk et al., 2001). Since SAH hydrolase activity is inhibited by its products, adenosine kinase is not a stressresponsive enzyme per se, but plays a pivotal role in sustaining trans-methylation reactions in general by serving as a coarse metabolic control to reduce the cellular concentration of free adenosine and SAH (Moffatt and Weretilnyk, 2001). Similar results have been observed in Avicennia marina, a halophytic mangrove shrub that grows in tidal swamps (Suzuki et al., 2003). It accumulates considerable quantities of glycine betaine in its leaves and the levels are increased by salt stress (Ashihara et al., 1997). The feeding experiments revealed that the incorporation of the radioactivity from [1,2-14C]ethanolamine into glycine betaine, and that from [methyl-14C]SAM into choline, phosphocholine and glycine betaine are stimulated by salt stress. In vitro activity of adenosine kinase and adenosine nucleosidase increases after the leaf disks are incubated with 250 and 500 mM NaCl. In A. marina, adenosine nucleosidase also participates to remove adenosine from the cycle (Suzuki et al., 2003). In addition to the ability to remove adenosine from the SAM cycle (see Fig. 11), halophytes, such as mangrove plants, generally possess the active purine salvage mechanism which may require an adequate supply of ATP for removal of salts from the cytoplasm of the cells. The salvage activity of adenosine has been compared in seven different mangrove species and in poplar (Ashihara et al., 2003). In four mangrove species, Rhizophora stylosa, Bruguiera gymnorrhiza, Kandelia candel and Sonneratia alba, the adenosine salvage activity is higher than in poplar and ATP synthesis from adenosine is stimulated by 250 mM NaCl. An adequate supply of ATP appears to be essential for the normal growth of these mangrove trees. In contrast, the adenosine salvage activity of leaves

Fig. 10. A role of adenine phosphoribosyltransferase (APRT) in biosynthesis of mugineic acid family phytosiderophores (MAs) in barley roots. The synthesis of one mol of nicotianamine requires three mol of SAM and produces three molecules of 50 methylthioadenosine (MTA) (step 2). MTA enters the methionine cycle for recycling carbons of the ribosyl group. In this process, three molecules of adenine are released from the methionine cycle for the synthesis of one molecule of MAs. The methionine cycle shown here is the same as the Yang cycle for ethylene biosynthesis except for the step 2. Instead of nicotianamine synthase, 1-aminocyclopane-1-carboxylic acid synthase participates in the Yang cycle (Yang and Hoffman, 1984). Numbers indicate the reaction steps catalysed by the enzymes (EC numbers) shown below. (1) SAM synthetase (EC 2.5.1.6); (2) nicotianamine synthase (2.5.1.43); (3) 50 -methylthioadenosine nucleosidase (3.2.2.16); (4) S-methyl-5-thioribose kinase (EC 2.7.1.100); (5) IDI 1 (iron-deficient induced cDNA1), Recent studies suggest that conversion of MTR-1-P to KMTB (step 5) is performed via acireductone. The reaction appears to be catalysed sequentially by an isomerase, a dehydrataseeenolaseephosphatase and an acireductone dioxygenase (Rzewuski and Sauter, 2008); (6) transaminase; (7) adenine phosphoribosyltransferase (2.7.2.4). Metabolites: KMTB, a-keto-g-methylthiobutyric acid; MAs, mugineic acid family phytosiderophores; Met, methionine; MTA, 50 methylthioadenosine; MTR, 50 -methylthioribose; MTR-1-P, 50 -methylthioribose-1phosphate; PPi, inorganic pyrophosphate; PRPP, 5-phosphoribosyl-1-pyrophosphate. Adapted from Itai et al. (2000).

114

H. Ashihara et al. / Phytochemistry 147 (2018) 89e124

of the mangrove shrubs, Avicennia marina and Lumnitzera racemosa, is low. The difference in adenosine salvage ability observed between large mangrove trees and small mangrove shrubs is not easily explained, but it might be due to the different requirements of energy for growth. 8.1.7. Fe-deficient stress and adenine phosphoribosyltransferase In Fe-deficient Graminaceous plants, production of the mugineric acid family of phytosiderophores (MAs), which solubilize inorganic Fe (III) compounds by chelation is increased. Itai et al. (2000) isolated a protein from barley roots which was induced by Fe-deficiency treatment. Based on the amino acid sequence of this protein, a cDNA (HvAPT1) encoding an adenine phosphoribosyltransferase was cloned from a cDNA library prepared from Fedeficient barley roots. Southern analysis suggested that there are at least two genes encoding adenine phosphoribosyltransferase in barley. Fe deficiency increased HvAPT1 expression in barley roots and resupplying Fe to the Fe-deficient plants rapidly negated the increase in HvAPT1 mRNA. A 4e9 fold increase in adenine phosphoribosyltransferase activity by Fe deficiency was also observed in roots of other MAs-producing graminaceous plants, such as rye, maize and rice. Nicotianamine, a precursors of MAs is synthesized from SAM catalysed by nicotianamine synthase (step 2 in Fig. 10). The reaction is 3 SAM / 3 S-methyl-50 -thioadenosine þ nicotianamine. 50 -Methylthioadenosine is converted to 5methylthioribose by 50 -methylthioadenosine nucleosidase (step 3 in Fig. 10) and adenine is released by this reaction. 50 -Methylthioribose is metabolised by the Yang methionine cycle for ethylene biosynthesis (Fig. 10) (Yang and Hoffman, 1984). Adenine released from the nucleosidase reaction is salvaged by adenine phosphoribosyltransferase and utilised for ATP synthesis. This may be required for maintaining an ample supply of ATP for SAM generation. 8.1.8. Wound stress and adenosine salvage Role of adenosine salvage in wound-induced adenylate biosynthesis has been reported (Katahira and Ashihara, 2006b). Activation of plant storage tissue after excision and incubation is known as the ‘wounding’ or ‘ageing’ phenomenon (Kahl, 1974). “Wound respiration” of potato tubers has been investigated since the late 19th century (Muller-Thungau, 1882). Many studies have been performed on the effect of wounding on cyanide-sensitive or cyanide-resistant respiration and mitochondrial membrane proteins (Hiser and McIntosh, 1990), but there are only a few studies on the synthesis of ATP during the wounding process. Levels of ATP and other nucleotides increased 2e7 times in “wounded” potato tuber slices maintained on moist paper for 24 h. The relative expression intensity of genes encoding adenosine kinase and adenine phosphoribosyltransferase in tuber slices is greater than that of genes of the de novo pathway, encoding glycineamide ribonucleotide formyltransferase (GART) (step 3 in Fig. 4) and 5aminoimidazole ribonucleotide synthetase (AIRS) (step 5 in Fig. 4). In vitro activities of adenosine kinase and adenine phosphoribosyltransferase doubled following wounding. Adenosine nucleosidase activity is negligible in freshly prepared slices, but its activity increases substantially as the slices age. In situ adenosine salvage activity, estimated from the incorporation of radioactivity from exogenously supplied [8-14C]adenosine into nucleotides and RNA, increased more than five times in the wounded slices (Katahira and Ashihara, 2006b). These results strongly suggest that greater expression of the genes encoding enzymes of adenosine salvage during wounding is closely related to the increased supply of adenine nucleotides in the wounded slices. From an energetics viewpoint, the wounding process of potato tuber slices is similar to the dormancy break of intact potato tubers.

Accelerated energy generation is required for these processes. In order to maintain high levels of ATP in potato tuber slices after wounding, the salvage pathways of adenine and adenosine appear to be very effective. The results suggest that the enhanced adenine and adenosine salvage activity in the wounded tuber slices is directly attributable to the pronounced expression of ADK and APT, and the subsequent increase in adenosine kinase and adenine phosphoribosyltransferase activities. The substantial increase in adenosine and adenine salvage activities in wounded tubers may participate in the salvage of adenine and/or adenosine, which are produced by rapid turnover of ATP (Katahira and Ashihara, 2006b). These salvage pathways may also contribute to the recycling of nucleosides derived from RNA. A 4-fold increase in activity of RNases has been observed 24 h after wounding induced by the slicing of potato tubers (Sacher et al., 1972). 8.1.9. Oxidative stress and adenine phosphoribosyltransferase Sukrong et al. (2012) reported that adenine phosphoribosyltransferase activity was negatively associated with oxidative stress caused by a reduction of cellular adenine levels. A genetic screen aimed at isolating oxidative stress-tolerant lines of A. thaliana identified oxt1, a line that exhibits improved tolerance to oxidative stress and elevated temperature without any apparent deleterious growth effects under non-stress conditions. Oxt1 harbours a mutation that arises from the altered expression of a gene encoding APT1, indicating a link between purine metabolism, whole-plant growth responses, and stress acclimation. The oxt1 mutation results in decreased APT1 expression that leads to reduced adenine phosphoribosyltransferase activity. Correspondingly, oxt1 plants possess elevated pools of adenine. Decreased adenine phosphoribosyltransferase activity directly correlates with stress resistance in transgenic lines that ectopically express APT1. The metabolic alteration in oxt1 plants also changes the expression of several antioxidant defence genes and the response of these genes to oxidative challenge. In addition, it has been reported that manipulation of adenine levels can induce stress tolerance in wild-type plants. These results show that alterations in cellular adenine levels can trigger stress tolerance and improve growth, leading to increases in plant biomass. However, high concentrations of adenine beyond a threshold level leads to deleterious growth inhibition. Therefore, control of adenine phosphoribosyltransferase activity is important to maintain the proper cellular adenine homeostasis. 8.1.10. Programmed cell death and purine salvage Alterations in the pattern of purine nucleotide synthesis and degradation were investigated during programmed cell death of tobacco BY-2 cells, induced by a simultaneous increase in the endogenous levels of nitric oxide and hydrogen peroxide (Stasolla et al., 2005). During the early phases of programmed cell death, increases in the salvage activity of adenine and adenosine are observed, which are related to the high activity of adenine phosphoribosyltransferase and adenosine kinase. During the later stages, a large fraction of purine nucleotide is also produced through the de novo pathway, suggesting a tight regulation between salvage and de novo synthesis. These changes are strictly associated with programmed cell death, as they did not occur if, nitric oxide or hydrogen peroxide is increased, or when actinomycin, which inhibits the death program, is added to the medium in the presence of nitric oxide and hydrogen peroxide. These changes in purine nucleotide synthesis represent an early metabolic switch which may be needed to ensure the proper execution of all the high-energy demand processes characteristic of the death programme. Independent experiments revealed that programmed cell death is associated with a reduction in adenine

H. Ashihara et al. / Phytochemistry 147 (2018) 89e124

nucleotide translocase activity leading to the inhibition of ATP synthesis. Depletion of cellular energy requirements might be a contributing factor triggering programmed cell death (Zhivotovsky et al., 2009). 8.1.11. Effect of glutathione redox state on purine salvage Relationship of glutathione redox state and purine metabolism in plants was investigated by Belmonte et al. (2003). Changes in the glutathione redox state affect plant growth and more specifically cell proliferation. Inclusion of reduced glutathione in the maintenance medium of spruce somatic embryogenesis promotes the proliferation of the embryogenic tissue. The balance between the reduced (GSH) and oxidized (GSSG) forms of glutathione and ultimately the GSH:GSSG ratio could be correlated with the growth of the embryogenic tissue over time. Treatment of embryogenic tissue with 0.2 mM GSH gave higher uptake of both 14 C-adenine and 14C-adenosine, especially during the rapid growth phase of the culture period, where a large proportion of these substrates is converted into nucleic acids and ATP. Inclusion of GSH sharply increased adenosine kinase activity 1 day after the treatment. The activity of adenine phosphoribosyltransferase in GSH-treated tissues was higher than that in GSSG-treated tissue throughout the course of the experiment. GSH appeared to promote nucleoside phosphotransferase activity. Thus, there is a strong correlation between the glutathione redox state and adenine and adenosine salvage during proliferation of white spruce embryogenic tissue. 8.1.12. Effect of brassinolide and purine salvage Belmonte et al. (2011) reported that brassinolide-induced improved development of Brassica napus microspore-derived embryos is associated with increased activities of purine salvage enzymes. Cellular brassinolide levels regulate the development of the microspore-derived embryos. Purine metabolism was compared in developing embryos treated with brassinolide and with brassinazole, an inhibitor of brassinolide biosynthesis. The utilization of adenine and adenosine for the synthesis of nucleotides and nucleic acids increased significantly in brassinolide-treated embryos. These metabolic changes were ascribed to enhanced adenine phosphoribosyltransferase and adenosine kinase activities, which are induced by brassinolide applications. In embryos treated with brassinazole purine salvage activity was reduced and this was associated to structural abnormalities and poor embryonic performance. 8.1.13. Embryogenic capacity and purine salvage activity Distinct fluctuations in purine salvage accompany the enhanced embryogenic capacity of Brassica napus cells over-expressing SHOOTMERISTEMLESS (STM) (Elhiti et al., 2011). The overexpression of STM enhances the number of microspore-derived embryos produced in culture and their ability to regenerate viable plants. This is in contrast to the down-regulation of this gene repressing the embryogenic process (Elhiti et al., 2010). Synthesis of purine nucleotides was measured in developing microsporederived embryos generated from B. napus lines ectopically expressing or down-regulating STM. Purine metabolism was estimated by using 14C-labelled adenine, adenosine and inosine. The improvement in embryo number and quality, induced by the ectopic expression of STM, was linked to the increased purine salvage activity during the early phases of embryogenesis and the enlargement of the adenylate pool required for the active growth of the embryos. This is due to an increase in transcriptional and enzymatic activity of adenine phosphoribosyltransferase and adenosine kinase. The highly operative salvage pathway induced by the ectopic expression of STM is associated with a reduced

115

catabolism of nucleotides, suggesting the presence of an antagonist mechanism controlling the rate of salvage and degradation pathways. These events are precluded by the down-regulation of STM which represses the formation of the embryos and their postembryonic performance.

8.2. Purine salvage and caffeine biosynthesis Purine alkaloids, such as caffeine and theobromine, are derived from purine nucleotides. In radiolabelled feeding experiments, purine bases and nucleosides are often used as precursors of caffeine biosynthesis. These compounds are salvaged to AMP, GMP and IMP and then converted to xanthosine which enters the four step caffeine biosynthesis pathway (see Fig. 1) (Ashihara et al., 2008a, 2011, 2017). In young tea leaves, caffeine biosynthesis begins with IMP formed directly by the de novo pathway (steps 1-10 in Fig. 4) (Ito and Ashihara, 1999). In addition, it also starts with adenine and guanine nucleotides which are pre-formed by the de novo (see Fig. 4) or by salvage reactions (see Fig. 1). An alternative route of caffeine biosynthesis utilises adenosine released from the SAM cycle (Koshiishi et al., 2001). The caffeine biosynthetic pathway contains three methylation steps that utilise SAM as the methyl donor (Fig. 11). In the process SAM is converted to SAH which in turn is hydrolysed to homocysteine and adenosine. Significant amounts of radioactivity from [methyl-14C]methionine and [methyl-14C]SAM are incorporated into theobromine and caffeine in young tea leaf segments, and very high SAH hydrolase activity is found in cell-free extracts from young tea leaves. Substantial amounts of radioactivity from [adenosyl-14C]SAH are also recovered as theobromine and caffeine in tea leaf segments, indicating that adenosine derived from SAH is utilised for the synthesis of the purine ring of caffeine. From the profiles of activity of related enzymes in tea leaf extracts, it was proposed that the major route

Fig. 11. A role of adenosine salvage in the S-adenosyl-L-methionine (SAM) route of caffeine biosynthesis in tea and coffee leaves. The synthesis of one mole of caffeine requires three moles of SAM and three moles of adenosine are released from the SAM cycle. Adenosine is salvaged to AMP by a single reaction catalysed by adenosine kinase (AK) and/or two steps reactions by adenosine nucleosidase (ARN) and adenine phosphoribosyltransferase (APRT). AMP is utilised for the caffeine biosynthesis. Numbers indicate the reaction steps catalysed by the enzymes (EC numbers) shown below. (1) SAM synthetase (2.5.1.6), (2) SAM-dependent N-methyltransferases (2.1.1.), (3) SAH hydrolase (3.3.1.1), (4) methionine synthase (2.1.1.13), (5) adenosine nucleosidase (3.2.2.7), (6) adenine phosphoribosyltransferase (2.4.2.7), (7) adenosine kinase (2.7.1.20), (8) AMP deaminase (3.5.4.6); (9) IMP dehydrogenase (1.1.1.205), (10) 50 nucleotidase (3.1.3.5), (11) 7-methylxanthosine synthase (2.1.1.158), (12) N-methylnucleosidase (3.2.2.25), (13) theobromine synthase (2.1.1.159) or dual-functional caffeine synthase (2.1.1.160), and (14) caffeine synthase (2.1.1.160). Metabolites: Sadenosyl-L-homocysteine (SAH); S-adenosyl-L-methionine (SAM). Adapted from Ashihara et al. (2017).

116

H. Ashihara et al. / Phytochemistry 147 (2018) 89e124

from SAM to caffeine is a SAM / SAH / adenosine / adenine / AMP / IMP / XMP / xanthosine / 7-methylxanthosine / 7methylxanthine / theobromine / caffeine pathway (steps 2, 3, 5, 6 and 8e14 in Fig. 11). In addition, direct adenosine kinase-

catalysed formation of AMP (step 7 in Fig. 11) from adenosine may also participate. In the SAM route of caffeine biosynthesis, not only the methyl groups but also the purine ring of caffeine is derived from SAM and adenosine salvage.

Fig. 12. The cytokinin interconversion pathway. Numbers indicate the reaction steps catalysed by the enzymes shown below. EC numbers for cytokinin specific enzymes have not yet been allocated. (1) cytokinin base (adenine) phosphoribosyltransferase; (2) cytokinin riboside (adenosine) kinase; (3) cytokinin ribotide phosphoribohydrolase (LONELY GUY, LOG); (4) 50 -nucleotidase; (5) cytokinin riboside (purine nucleoside) phosphorylase (StCKP1) (6) cytokinin riboside (adenosine) nucleosidase. Cytokinin ribotide phosphoribohydrolase (LOG) is distinct from AMP phosphoribohydrolase (Kuroha et al., 2009). Cytokinin riboside phosphorylase (StCKP1) has a stronger affinity for iPR than for adenosine (Bromley et al., 2014). Adapted from Bishop et al. (2015).

H. Ashihara et al. / Phytochemistry 147 (2018) 89e124

8.3. Cytokinin and purine salvage Cytokinins, such as isopentenyladenine, benzyladenine and trans-zeatin are adenine-derived plant hormones. Major sites of cytokinin biosynthesis are roots, cambium, and other actively dividing tissues. Cytokinins play vital roles in the regulation of multiple physiological processes in plants including cell proliferation and differentiation, nutrient allocation, leaf senescence, and responses to external biotic and abiotic signals (Crozier et al., 2000; Schoor et al., 2011; Zhang et al., 2013; Bishop et al., 2015). Therefore, cytokinin biosynthesis, interconversions (activation and inactivation) and degradation are involved in the regulation of cytokinin homeostasis during plant development (Åstot et al., 2000; Bishop et al., 2015; Kwade et al., 2005; Romanov et al., 2006; Kurakawa et al., 2007; Schoor et al., 2011). Cytokinins are synthesized as ribonucleotides and three forms of cytokinins, namely cytokinin nucleotides, nucleosides and free bases occur in plant cells (Sakakibara, 2006, 2010) (see Section 6.3). Cytokinins are effective long-distance signals molecules transported within the plant in xylem and phloem. The xylemtransported form of cytokinins are predominantly ribosides, but the cytokinin receptors show a preference for the free base, therefore, conversion of nucleoside forms to bases is necessary. In vegetative tissues of A. thaliana, the major route for the production of free base forms of cytokinins from ribonucleotides is mediated by cytokinin ribotide phosphoribohydrolase (named LONELY GUY proteins) (Sakakibara, 2006, 2010). The reaction of this enzyme is very similar to that catalysed by AMP phosphoribohydrolase (EC 3.2.2.4), however, AMP cannot be hydrolysed by this enzyme (Kuroha et al., 2009). In addition to LONELY GUY, degradation enzymes of purine nucleotides and nucleosides also contribute to the production of the free base form of cytokinins. Chen and Kristopeit (1981) reported that two isoforms of 50 -nucleotidases from wheat germ have high affinity for isopentenyladenine riboside 50 -monophosphate. Interconversion of cytokinins are shown in Fig. 12. Two purine salvage enzymes, adenine phosphoribosyltransferase and adenosine kinase, have been shown to contribute to the control of cytokinin activity as discussed below. 8.3.1. Role of adenine phosphoribosyltransferase in cytokinin inactivation Zhang et al. (2013) reported that APT1 was the causal gene of the high-dose cytokinin-resistant mutants of A. thaliana. APT1 is functionally predominant among the five members of the adenine phosphoribosyl transferase family. Loss of APT1 activity in plants leads to excess accumulation of cytokinin bases, and evokes a myriad of cytokinin-regulated responses, such as delayed leaf senescence, anthocyanin accumulation, and downstream gene expression. These results indicated that APT1 is a key metabolic enzyme participating in the cytokinin inactivation by phosphoribosylation. Like adenine, exogenously applied cytokinin bases are immediately salvaged to cytokinin ribotides by adenine phosphoribosyltransferase in planta, therefore, cytokinin bases transported to plant cells appear to be temporarily stored as cytokinin nucleotides, an inactive form, after which they are changed to the active form by hydrolysing cytokinin activation enzymes shown above. The nucleotide formation by adenine phosphoribosyltransferase plays a key role in metabolising exogenous cytokinins (Bishop et al., 2015). 8.3.2. Role of adenosine kinase in homeostasis of cytokinin in planta The contribution of adenosine kinase to cytokinin interconversions was reported by Schoor et al. (2011). Use was made of lines of A. thaliana silenced in adk expression to

117

understand the contribution of this enzyme activity to in vivo cytokinin metabolism. Both small interfering RNA- and artificial microRNA-mediated silencing of adk led to impaired root growth, small, crinkled rosette leaves, and reduced apical dominance. ADK-deficient roots and leaves exhibited irregular cell division. Root tips had an uneven arrangements of root cap cells, reduced meristem size, and enlarged cells in the elongation zone; rosette leaves exhibited decreased cell size but increased cell abundance. In vivo feeding of 3H-isopentenyladenine riboside and 3H-transzeatin riboside in ADK-deficient leaves showed a reduced incorporation of radioactivity into the isopentenyladenine ribotide and trans-zeatin ribotide, respectively. There were higher levels of cytokinin ribosides in adk-silencing plants. From these metabolic and phenotypic analyses of ADK deficient lines, it was concluded that ADK contributes to cytokinin homeostasis in planta (Schoor et al., 2011). 8.3.3. Purine nucleoside phosphorylase and endodormancy of potato tubers Endodormancy of potato tubers is an internally imposed state of zero growth and minimal metabolism. Duration of endodormancy is shortened by chill treatment: a period of exposure to temperatures near to but not below freezing, mimicking the early stages of a cold season. Termination of endodormancy in response to the chill treatment allows buds of temperate perennials to resume growth whenever the cold season ends. It has been shown that the exit from endodormancy in potato tubers is linked to an increase in cytokinins and a higher cell sensitivity to cytokinins. These factors are amplified by chilling (Turnbull and Hanke, 1985a, b). Bromley et al. (2014) reported a nucleoside phosphorylase superfamily protein, isolated from potato shoot apices, which interconverts cytokinin ribosides and free bases. The protein, isolated by affinity chromatography from tuberizing potato stolons, exhibited cytokinin-binding activity. The predicted polypeptide includes a cleavable signal peptide and motifs for purine nucleoside phosphorylase activity. Isopentenyladenine, trans-zeatin, and dihydrozeatin are converted into ribosides in the presence of ribose-1-phosphate. In the reversed reaction the conversion of cytokinin ribosides to cytokinin bases in the presence of Pi, StCKP1 had no detectable purine nucleosidase activity. Evidence was presented that StCKP1 is active in tubers as a negative regulator of cytokinins, prolonging dormancy by a chill-reversible mechanism. These results suggest that conversion of cytokinin riboside to cytokinin bases are involved in cytokinin activation. Although StCKP1 was discovered as a cytokinin-binding nucleoside phosphorylase in extracts of tuberizing potato stolon tips (Bromley et al., 2014), purine nucleoside phosphorylase activity which catalyses the reaction, purine nucleoside þ Pi / purine bases þ ribose-1phosphate, is not detectable in most plants including growing potato tubers (see Table 1). In contrast purine nucleosidase, which catalyses the reaction purine nucleoside þ H2O / purine base þ ribose, participates to the hydrolysis of purine nucleosides, such as adenosine and guanosine (Katahira and Ashihara, 2006a; Moffatt and Ashihara, 2002) (Table 1). These results suggest that the cytokinin phosphorylase may be a specific enzyme for cytokinin interconversion. 8.4. Purine salvage and male sterility Three mutants of A. thaliana deficient in adenine phosphoribosyltransferase activity (apt13) were isolated by selecting for germination of seeds on a medium containing 0.1 mM 2,6diaminopurine by Moffatt and Somerville (1988). In each of the mutants, 2,6-diaminopurine resistance is due to a recessive nuclear

118

H. Ashihara et al. / Phytochemistry 147 (2018) 89e124

mutation at a locus designated apt. The mutants grow more slowly than wild-type plants, and are male sterile due to the abortion of pollen after the meiotic divisions of the pollen mother cells. The basis of the male sterile phenotype of apt1-deficient mutants has been investigated by light and electron microscopy (Regan and Moffatt, 1990; Zhang et al., 2002). Histochemical analyses indicate that the first defects in pollen development in apt1-3 mutants occur when the microspores are released from the tetrad, just after meiosis. At this stage mutant microspores have atypical cell walls that stain abnormally with conventional histochemical stains. Following further development, vacuole formation is delayed and the microspores begin deteriorating with very few undergoing mitotic divisions. Vital stains indicate that mutant microspores are less metabolically active following meiosis. An A. thaliana mutant lacking the major APT isoform, APT1, is male sterile due to defects that become apparent soon after meiosis. Electron microscopebased studies were conducted to define if the effects of APT1 deficiency on pollen development were attributable to adenine or cytokinin metabolism (Zhang et al., 2002). Changes were observed in mutant anthers in both tapetal and pollen mother cells prior to meiosis with additional defects being found at later stages in both compartments. Major changes included altered lipid accumulation in the tapetal cells, changes in pollen cell wall development, and a loss of synchrony in the development of the tapetum and microspores. The microspores of the mutant initiated meiosis earlier than their wild-type counterparts. Cytokinins may contribute to the signal for the onset of meiosis, suggesting that the apt1-3 mutants may experience a transient increase in cytokinins due to reduced APT1 activity. The later defects in pollen development are probably explained by a deficiency in adenine nucleotides. These results suggest that APT1 deficiency causes a decrease in energy metabolism, due to the lack of adenine salvage into adenine nucleotides, which ultimately leads to pollen abortion. The early onset of meiosis in the mutant may be associated with altered cytokinin metabolism. Involvement of the adenine phosphoribosyl transferase gene (OsAPT2) in the thermo-sensitive genic male sterility (TGMS) in rice plants has been suggested by Zhou et al. (2006). The rice OsAPT2 encodes a putative adenine phosphoribosyl transferase and the levels of its transcripts detected in the young panicles of the TGMS mutant line ‘Annong S-1’ decreased at 29  C. This temperature is required for the induction of ‘Annong S-1’ fertility conversion. Since the panicle is likely the thermo-sensitive organ at the early stages of pollen fertility alternation, the observed heat-induced change in the OsAPT2 expression pattern in young panicles may mediate, at least in part, thermo-sensitive genic male sterility. An antisense strategy was used to suppress the expression of the OsAPT2 homolog in A. thaliana, and the resultant homozygous, transgenic plants had a reduced AMP content, displayed lower pollen germination rates and exhibited abnormalities in leaf phenotypes and flowering timing. These data suggest that OsAPT2 may be involved in TGMS in the rice. 8.5. Plant pathogen response and adenosine kinase Geminiviruses are single-stranded DNA viruses that infect a wide range of plant species. Wang et al. (2003) reported that adenosine kinase is inactivated by geminivirus AL2 and L2 proteins which are encoded by the Begomovirus and Curtovirus genera, respectively. The AL2 and L2 proteins inactivate adenosine kinase in vitro and after co-expression in E. coli and yeast. The adenosine kinase activity is reduced in transgenic plants expressing the viral proteins and in geminivirus-infected plant tissues. In contrast, adenosine kinase activity is increased after inoculation of plants with a variety of RNA viruses or a geminivirus lacking a functional

L2 gene. Consistent with its ability to interact with multiple cellular kinases, AL2 is present in both the nucleus and the cytoplasm of infected plant cells. These data indicate that adenosine kinase is targeted by viral pathogens, and that adenosine kinase may be a part of host defence responses. AL2 and L2 also interact with and inactivate SNF1 kinase (sucrose non-fermenting 1 protein kinase, a yeast homologue of mammalian AMP-activated protein kinase), a global regulator of metabolism. These observations suggest that metabolic alterations mediated by SNF1 are an important component of innate antiviral defences and that the inactivation of adenosine kinase and SNF1 by the geminivirus proteins represents a dual strategy to counter this defence. Infection of tobacco plants by geminiviruses leads to the expression of viral proteins (AL2 and L2) that inactivate both SNF kinase and adenosine kinase in order to disable this plant defence system. 8.6. Root gravitropism and cap morphogenesis and adenosine kinase Young et al. (2006) reported that adenosine kinase is a modulator of root cap morphogenesis and gravitropism in A. thaliana. Upon gravistimulation, ADK protein levels and activity increased in the root tip. Mutation in ADK1, one of two ADK genes, resulted in cap morphogenesis defects, along with alterations in root sensitivity to gravistimulation and slower kinetics of root gravitropic curvature. The kinetics of defect were partially rescued by adding spermine to the growth medium, whereas the defects in cap morphogenesis and gravitropic sensitivity were not altered. The root morphogenesis and gravitropism defects of adk1-1 are accompanied by altered expression of the auxin efflux facilitator (PIN3) in the cap and decreased expression of the auxin-responsive reporter (DR5-GUS). Furthermore, PIN3 failed to re-localise to the bottom membrane of statocytes and thus adk1-1 roots did not develop a lateral auxin gradient across the cap that is necessary for the curvature response. The adk1-1 does not affect gravity-induced cytoplasmic alkalinisation of the root statocytes, suggesting either that ADK1 acts downstream of the cytoplasmic alkalinisation and upstream of PIN3 relocalization in a linear pathway, or that the pH and PIN3-relocalization responses to gravistimulation occupy distinct branches of the pathway. These data are consistent with a role for adenosine kinase on the activation of the SAM cycle, which provides the SAM for the production of polyamines in the control of root gravitropism and cap morphogenesis. 8.7. Omics studies related to the purine salvage in plants 8.7.1. Transcriptome studies Studies on the global levels of transcripts aid our understanding on how plants adjust their metabolic responses during development and following stress conditions. Advances in this area have benefited from the development of high throughput technologies, such as RNA-Seq (RNA sequencing). Changes in transcript levels of purine salvage enzymes, for example, adenine phosphoribosyl transferase (APT) and adenosine kinase (ADK) genes, accompanied by growth and development and different environmental conditions can be found in the databases of transcriptome, for example, Arabidopsis Information Resource (TAIR, http://www.arabidopsis.org/) and in the Rice Expression Profile Database (RiceXPro, http://ricexpro.dna.affrc.go.jp/). It has been speculated that transcription of ADK increases in association with SAM-dependent methylation activities, because ADK transcript abundance increased in association with changes in SAHH (S-adenosyl-L-homocysteine hydrolase) gene transcript levels, presumably due to changes in the flux of the SAM cycle (Schoor and Moffatt, 2004) (see 8.1.6).

H. Ashihara et al. / Phytochemistry 147 (2018) 89e124

8.7.2. Proteome studies Since mRNAs are not always translated into proteins, the proteome information can assist in understanding the regulation of purine salvage at the molecular level. Up-regulation of APT1 (adenine phosphoribosyltransferase 1) in response to stress was
found in several studies (Chen et al., 2016; Deng et al., 2015; Kosova et al., 2013). A proteomics study by Chen et al. (2016) describing protein profiles of roots of Glycine max in response to Mn toxicity indicated 31 proteins related to stress. Among these proteins APT1 was upregulated, an observation suggesting the involvement of adenine salvage in energy supply during the modification of the cell wall which is highly affected by the stress. There are also several proteomics reports on adenosine kinase protein (ADK). For example, in the differential proteomic analysis of anthers between cytoplasmic male sterile and maintainer lines in Capsicum annuum, adenosine kinase 1 (ADK1)-like protein was down-regulated in the sterile anthers along with more than 20 other proteins. These included ATP synthase D chain, formate dehydrogenase, a-mannosidase, chloroplast manganese stabilizing protein-II, glutathione S-transferase, putative caffeoyl-CoA 3-Omethyltransferase, and glutamine synthetase (GS). The results support the link between adenosine kinase and male sterility discussed in Section 8.4. Up-regulated ADK proteins have been documented during different stress conditions, but its function was not clearly evident (Bertolde et al., 2014; Chen et al., 2014, 2015; Salavati et al., 2012). 8.7.3. Metabolome studies In general, metabolomics studies can be identified as (i) comparison of pool sizes of metabolites and (ii) metabolic flux analysis using isotopically labelled substrates. Feeding experiments have been often performed in purine salvage studies using radioactive substrates (see Section 2.1), but utilization of stable isotopes and analysis by NMR- and GC-MS- and LC-MS-based metabolomics has proved more informative in elucidating substrate utilization in response to specific physiological conditions (Liu and Locasale, 2017; Zaimenko et al., 2017). Although this type of metabolomics studies has not yet performed in nucleotide-related fields in plant cells, this approach might be useful in revealing the role of purine salvage in plants. The special techniques and pitfalls of watersoluble primary metabolites including nucleotides are reviewed by Lu et al. (2017). 9. Concluding remarks Purine salvage is catalysed principally by adenine phosphoribosyltransferase and adenosine kinase in plants. This recycling pathway is not only energetically advantageous for adenine nucleotide formation but also to the removal of purine bases and nucleosides which inhibit the flow of metabolism such as the SAM cycle. In addition, other purine bases, hypoxanthine and guanine, and purine nucleoside, inosine and guanosine, are salvaged by hypoxanthine/guanine phosphoribosyltransferase and inosine/ guanosine kinase, respectively. Deoxyadenosine and deoxyguanosine are also salvaged by deoxyadenosine kinase and deoxyguanosine kinase. In some cases, nucleoside phosphotransferase and adenosine nucleosidase contribute to purine salvage. Purine salvage is not a supplementary pathway for the de novo purine biosynthesis. Rapid turnover of nucleotides needs the salvage reactions, so that purine salvage is required for normal cellular growth and homeostasis. The salvage activity is closely associated with physiological events in plants. Onset of seed germination, cell division and growth, dormancy break and recovery from several stresses require purine nucleotides produced by these purine salvage reactions. Purine salvage enzymes also

119

contribute the conversion of cytokinin bases and nucleosides to nucleotides. Cytokinin ribotides or ribosides are major forms of endogenous cytokinins and are biologically inactive. The reverse reaction, formation of bioactive cytokinin bases appears to be a key activation step. Molecular studies using mutants which suppress the expression of adenine phosphoribosyltransferase and adenosine kinase suggest that these purine salvage enzymes participate in cytokinin-mediated processes through activation-deactivation mechanisms. In this review, we summarised the present knowledge of the purine salvage in plants. Many potential roles of purine salvage proposed at the physiological level and discussed in this review are likely to be confirmed at the molecular level. Furthermore, omics research may reveal new functions of purine salvage in plants. Acknowledgements The authors would like to thank Professor Takao Yokota, Teikyo University, Utsunomiya, Japan for his advice on chemical structures illustrated in this paper. Our own purine research cited in this paper were supported by the Grant-in-Aid for Scientific Research from the Japanese Ministry of Education, Science and Culture, and from the Japan Society for the Promotion of Science. Parts of studies are also supported by research grants from the Sapporo Bioscience Foundation, the All Japan Coffee Association, the Society for Research on Umami Taste, the Kumagai Foundation for Science and Technology, the Moritani Scholarship Foundation, the Asahi Group Foundation, the Fujiwara Natural History Foundation, the Salt Science Research Foundation and the Scandinavia-Japan Sasakawa Foundation for H.A. The travel funds from the Royal Society, UK and from the British Council are also acknowledged by H.A. and A.C. References Abusamhadneh, E., McDonald, N.E., Kline, P.C., 2000. Isolation and characterization of adenosine nucleosidase from yellow lupin (Lupinus luteus). Plant Sci. 153, 25e32. Adams, A., Harkness, R.A., 1976. Developmental changes in purine phosphoribosyltransferases in human and rat tissues. Biochem. J. 160, 565e576. Ali, L.Z., Sloan, D.L., 1982. Studies of the kinetic mechanism of hypoxanthineguanine phosphoribosyltransferase from yeast. J. Biol. Chem. 257, 1149e1155. Allen, M., Qin, W., Moreau, F., Moffatt, B., 2002. Adenine phosphoribosyltransferase isoforms of Arabidopsis and their potential contributions to adenine and cytokinin metabolism. Physiol. Plant. 115, 56e68. Anderson, J.D., 1977a. Adenylate metabolism of embryonic axes from deteriorated soybean seeds. Plant Physiol. 59, 610e614. Anderson, J.D., 1977b. Responses of adenine nucleotides in germinating soybean embryonic axes to exogenously applied adenine and adenosine. Plant Physiol. 60, 689e692. Arivett, B., Farone, M., Masiragani, R., Burden, A., Judge, S., Osinloye, A., Minici, C., Degano, M., Robinson, M., Kline, P., 2014. Characterization of inosine-uridine nucleoside hydrolase (RihC) from Escherichia coli. Biochim. Biophys. Acta 1844, 656e662. Ashihara, H., 1977a. Characterization of phosphoribosylpyrophosphate synthetase from spinach leaves. Z. Pflanzenphysiol. 83, 379e392. Ashihara, H., 1977b. Regulation of the activity of spinach phosphoribosylpyrophosphate synthetase by “energy charge” and endproducts. Z. Pflanzenphysiol. 85, 383e392. Ashihara, H., 1983. Changes in activities of purine salvage and ureide synthesis during germination of black gram (Phaseolus mungo) seeds. Z. Pflanzenphysiol. 113, 47e60. Ashihara, H., 2012. Xanthosine metabolism in plants: metabolic fate of exogenously supplied 14C-labelled xanthosine and xanthine in intact mungbean seedlings. Phytochem. Lett. 5, 100e103. Ashihara, H., 2016. Biosynthesis of 5-phosphoribosyl-1-pyrophosphate in plants: a review. Eur. Chem. Bull. 5, 314e323. Ashihara, H., Adachi, K., Otawa, M., Yasumoto, E., Fukushima, Y., Kato, M., Sano, H., Sasamoto, H., Baba, S., 1997. Compatible solutes and inorganic ions in the mangove plant Avicennia marina and their effects on the activities of enzymes. Z. Naturforsch. 52c, 433e440. Ashihara, H., Crozier, A., 1999. Biosynthesis and metabolism of caffeine and related purine alkaloids in plants. Adv. Bot. Res. 30, 117e205. Ashihara, H., Crozier, A., 2001. Caffeine: a well known but little mentioned

120

H. Ashihara et al. / Phytochemistry 147 (2018) 89e124

compound in plant science. Trends Plant Sci. 6, 407e413. Ashihara, H., Kato, M., Crozier, A., 2011. Distribution, biosynthesis and catabolism of methylxanthines in plants. In: Fredholm, B.B. (Ed.), Methylxanthines (Handbook of Experimental Pharmacology), vol. 200. Springer, pp. 11e31. Ashihara, H., Kameyama, Y., 1989. Biosynthesis of pyrimidines during germination of black gram seeds: changes in the level of participant enzymes. Curr. Sci. 58, 889e891. Ashihara, H., Komamine, A., 1974. Regulatory properties of a plant phosphoribosylpyrophospate synthetase. Plant Sci. Lett. 2, 119e123. Ashihara, H., Li, X.N., Ukaji, T., 1988. Effect of inorganic phosphate on the biosynthesis of purine and pyrimidine nucleotides in suspension-cultured cells of Catharanthus roseus. Ann. Bot. 61, 225e232. Ashihara, H., Ludwig, I., Katahira, R., Yokota, T., Fujimura, T., Crozier, A., 2015. Trigonelline and related nicotinic acid metabolites: occurrence, biosynthesis, taxonomic considerations, and their roles in planta and in human health. Phytochem. Rev. 14, 765e798. Ashihara, H., Luit, B., Belmonte, M., Stasolla, C., 2008a. Metabolism of nicotinamide, adenine and inosine in developing microspore-derived canola (Brassica napus) embryos. Plant Physiol. Biochem. 46, 752e759. Ashihara, H., Mitsui, K., Yabuki, N., Nygaard, P., 1991. Adenosine metabolism and growth of adenosine-requiring mutant cells of Datura innoxia. Int. J. Purine Pyrimidine Res. 2, 129e135. Ashihara, H., Mizuno, K., Yokota, T., Crozier, A., 2017. Xanthine alkaloids: occurrence, biosynthesis, and function in plants. In: Kinghorn, A.D., Falk, H., Gibbons, S., Kobayashi, J. i (Eds.), Progress in the Chemistry of Organic Natural Products 105. Springer, pp. 1e88. Ashihara, H., Nygaard, P., 1989. Purine nucleotide synthesis and RNA synthesis in suspension cultured cells of carrots. Physiol. Plant. 75, 31e36. Ashihara, H., Sano, H., Crozier, A., 2008b. Caffeine and related purine alkaloids: biosynthesis, catabolism, function and genetic engineering. Phytochemistry 69, 841e856. Ashihara, H., Stasolla, C., Loukanina, N., Thorpe, T.A., 2000. Purine and pyrimidine metabolism in cultured white spruce (Picea glauca) cells: metabolic fate of 14Clabeled precursors and activity of key enzymes. Physiol. Plant 108, 25e33. Ashihara, H., Stasolla, C., Loukanina, N., Thorpe, T.A., 2001. Purine metabolism during white spruce somatic embryo development: salvage of adenine, adenosine, and inosine. Plant Sci. 160, 647e657. Ashihara, H., Ukaji, T., 1985. Presence of adenine phosphoribosyltransferase and adenosine kinase in chloroplasts of spinach leaves. Int. J. Biochem. 17, 1275e1277. Ashihara, H., Ukaji, T., 1986. Inorganic phosphate absorption and its effect on adenosine 5'-triphosphate level in suspension cultured cells of Catharanthus roseus. J. Plant Physiol. 124, 77e85. Ashihara, H., Wakahara, S., Suzuki, M., Kato, A., Sasamoto, H., Baba, S., 2003. Comparison of adenosine metabolism in leaves of several mangrove plants and a poplar species. Plant Physiol. Biochem. 41, 133e139. Ashihara, H., Yabuki, N., Mitsui, K., 1990. A high-performance liquid chromatography method for separation of purine bases, nucleosides and ureides: application to studies on purine catabolism in higher plants. J. Biochem. Biophys. Methods 21, 59e63. Ashihara, H., Yokota, T., Crozier, A., 2013a. Biosynthesis and catabolism of purine alkaloids. Adv. Bot. Res 68, 111e138. Ashihara, H., Yokota, T., Crozier, A., 2013b. Purine alkaloids, cytokinins, and purine
rillon, J.-M. (Eds.), Natural like neurotoxin alkaloids. In: Ramawat, K.G., Me Products: Phytochemistry, Botany and Metabolism of Alkaloids, Phenolics and Terpenes. Springer, Berlin, pp. 953e975. € m, A., Wang, Q., Kunkel, T., Moritz, T., Chua, N.-H., Åstot, C., Dolezal, K., Nordstro Sandberg, G., 2000. An alternative cytokinin biosynthesis pathway. Proc. Natl. Acad. Sci. U. S. A. 97, 14778e14783. Atkins, C.A., Storer, P.J., Shelp, B.J., 1989. Purification and properties of purine nucleosidase from N2-fixing nodules of cowpea (Vigna unguiculata). J. Plant Physiol. 134, 447e452. Atkinson, D.E., 1977. Cellular Energy Metabolism and its Regulation. Academic Press, New York, pp. 1e293. Auer, C.A., 2002. Discoveries and dilemmas concerning cytokinin metabolism. J. Plant Growth Regul. 21, 24e31. Balandier, P., Gendraud, M., Rageau, R., Bonhomme, M., Richard, J.P., Parisot, E., 1993. Bud break delay on single node cuttings and bud capacity for nucleotide accumulation as parameters for endo- and paradormancy in peach trees in a tropical climate. Sci. Hort. 55, 249e261. Balestri, F., Barsotti, C., Lutzemberger, L., Camici, M., Ipata, P.L., 2007. Key role of uridine kinase and uridine phosphorylase in the homeostatic regulation of purine and pyrimidine salvage in brain. Neurochem. Int. 51, 517e523. Barankiewicz, J., Paszkowski, J., 1980. Purine metabolism in mesophyll protoplasts of tobacco (Nicotiana tabacum) leaves. Biochem. J. 186, 343e350. Barnola, P., Lavarenne, S., Gendraud, M., Jallut, N., 1986. Dormance des bourgeons ^ne (Fraxinus excelsior L.):e
valuation du pool des nucle
osides triapicaux du fre
ventail des tempe
ratures actives sur le de
bourrement des phosphates et e
riode de dormance. Ann. For. Sci. 43, 339e350. bourgeons en pe Belmonte, M., Elhiti, M., Ashihara, H., Stasolla, C., 2011. Brassinolide-improved development of Brassica napus microspore-derived embryos is associated with increased activities of purine and pyrimidine salvage pathways. Planta 233, 95e107. Belmonte, M., Stasolla, C., Loukanina, N., Yeung, E.C., Thorpe, T.A., 2003. Glutathione modulation of purine metabolism in cultured white spruce embryogenic tissue.

Plant Sci. 165, 1377e1385. Berens, R.L., Krug, E.C., Marr, J.J., 1995. Purine and pyrimidine metabolism. In: Marr, J.J., Müller, M. (Eds.), Biochemistry and Molecular Biology of Parasites. Academic Press, San Diego, pp. 89e117. Bertolde, F.Z., Almeida, A.A.F., Pirovani, C.P., 2014. Analysis of gene expression and proteomic profiles of clonal genotypes from Theobroma cacao subjected to soil flooding. PLoS One 9 (10), e108705. Bieleski, R.L., 1964. The problem of halting enzyme action when extracting plant tissues. Anal. Biochem. 9, 431e442. Bishop, G., Sakakibara, H., Seo, M., Yamaguchi, S., 2015. Biosynthesis of hormones. In: Buchanan, B.B., Gruissem, W., Jones, R.L. (Eds.), Biochemistry and Molecular Biology of Plants, second ed. John Wiley and Sons, Chichester, UK, pp. 769e833. Brawerman, G., Chargaff, E., 1955. On the distribution and biological significance of the nucleoside phosphotransferases. Biochim. Biophys. Acta 16, 524e532. Bromley, Jennifer R., Warnes, Barbara J., Newell, Christine A., Thomson, Jamie C.P., James, Celia M., Turnbull, Colin G.N., Hanke, David E., 2014. A purine nucleoside phosphorylase in Solanum tuberosum L. (potato) with specificity for cytokinins contributes to the duration of tuber endodormancy. Biochem. J. 458, 225e237. Brown, E., 1963. Purine and pyrimidine derivatives in mature pea seeds. Biochem. J. 88, 498e504. Brunngraber, E.F., Chargaff, E., 1967. Purification and properties of a nucleoside phosphotransferase from carrot. J. Biol. Chem. 242, 4834e4840. Brunngraber, E.F., Chargaff, E., 1970. Nucleoside phosphotransferase from carrot: kinetic studies and exploration of active sites. J. Biol. Chem. 245, 4825e4831. Burch, L.R., Stuchbury, T., 1986. Metabolism of purine nucleotides in the tomato plant. Phytochemistry 25, 2445e2449. Buchanan, B.B., Gruissen, W., Jones, R.L., 2015. Biochemistry and Molecular Biology of Plants. American Society of Plant Biologists, Rockville. Bzowska, A., Kulikowska, E., Shugar, D., 2000. Purine nucleoside phosphorylases: properties, functions, and clinical aspects. Pharmacol. Ther. 88, 349e425. Campos, A., Rijo-Johansen, M.J., Carneiro, M.F., Fevereiro, P., 2005. Purification and characterisation of adenosine nucleosidase from Coffea arabica young leaves. Phytochemistry 66, 147e151. Caputto, R., 1951. The enzymatic synthesis of adenylic acid; adenosine kinase. J. Biol. Chem. 189, 801e814. Chao, H.M., 1976. Nucleoside phosphotransferase from Erwinia herbicola, a new membrane-bound enzyme. J. Biol. Chem. 251, 2330e2333. Chen, C.-M., Eckert, R.L., 1977. Phosphorylation of cytokinin by adenosine kinase from wheat germ. Plant Physiol. 59, 443e447. Chen, C.-M., Melitz, D.K., Clough, F.W., 1982. Metabolism of cytokinin: phosphoribosylation of cytokinin bases by adenine phosphoribosyltransferase from wheat germ. Arch. Biochem. Biophys. 214, 634e641. Chen, C.M., Kristopeit, S.M., 1981. Metabolism of cytokinin: deribosylation of cytokinin ribonucleoside by adenosine nucleosidase from wheat germ cells. Plant Physiol. 68, 1020e1023. Chen, C.M., Petschow, B., 1978. Metabolism of cytokinin. Ribosylation of cytokinin bases by adenosine phosphorylase from wheat germ. Plant Physiol. 62, 871e874. Chen, C., Song, Y., Zhuang, K., Li, L., Xia, Y., Shen, Z., 2015. Proteomic analysis of copper-binding proteins in excess copper-stressed roots of two rice (Oryza sativa L.) varieties with different Cu tolerances. PLoS One 10 (4), e0125367. Chen, Y., Chen, X., Wang, H., Bao, Y., Zhang, W., 2014. Examination of the leaf proteome during flooding stress and the induction of programmed cell death in maize. Proteome Sci. 12, 33. Chen, Z., Yan, W., Sun, L., Tian, J., Liao, H., 2016. Proteomic analysis reveals growth inhibition of soybean roots by manganese toxicity is associated with alteration of cell wall structure and lignification. J. Proteom. 143, 151e160.
s, A., Lafleuriel, J., Le Floc'h, F., 1989. The inosine-guanosine kinase activity of Combe mitochondria in tubers of Jersalem artichoke. Plant Physiol. Biochem. 27, 729e736. Craig, S.P., Eakin, A.E., 2000. Purine phosphoribosyltransferases. J. Biol. Chem. 275, 20231e20234. Crozier, A., Kamita, Y., Bishop, G., Yokota, T., 2000. Biosynthesis of hormones and elicitor molecules. In: Buchanan, B.B., Gruissen, W., Jones, R.L. (Eds.), Biochemistry and Molecular Biology of Plants. American Society of Plant Physiologists, Rockville, pp. 850829. Dawson, R.M.C., Elliott, D.C., Elliott, W.H., Jones, K.M., 1986. Data for Biochemical Research, third ed. Clarendon Press, Oxford. Day, R.B., McAlvin, C.B., Loh, J.T., Denny, R.L., Wood, T.C., Young, N.D., Stacey, G., 2000. Differential expression of two soybean apyrases, one of which is an early nodulin. Mol. Plant-Microbe Interact. 13, 1053e1070. De Koning, H., Diallinas, G., 2000. Nucleobase transporters (review). Mol. Membr. Biol. 17, 75e94. Deng, W.-W., Ashihara, H., 2010. Profiles of purine metabolism in leaves and roots of Camellia sinensis seedlings. Plant Cell Physiol. 51, 2105e2118. Deng, W.-W., Sasamoto, H., Ashihara, H., 2015. Effect of caffeine on the expression pattern of water-soluble proteins in rice (Oryza sativa) seedlings. Nat. Prod. Commun. 10, 733e736. Desimone, M., Catoni, E., Ludewig, U., Hilpert, M., Schneider, A., Kunze, R., Tegeder, M., Frommer, W.B., Schumacher, K., 2002. A novel superfamily of transporters for allantoin and other oxo derivatives of nitrogen heterocyclic compounds in Arabidopsis. Plant Cell 14, 847e856. Dong, B., Ryan, P.R., Rengel, Z., Delhaize, E., 1999. Phosphate uptake in Arabidopsis

H. Ashihara et al. / Phytochemistry 147 (2018) 89e124 thaliana:dependence of uptake on the expression of transporter genes and internal phosphate concentrations. Plant Cell Environ. 22, 1455e1461. Doree, M., 1973. Metabolism of exogenous adenine by Acer Pseudoplatanus cells. Phytochemistry 12, 2101e2108. Edwards, R., 1996. S-Adenosyl-L-methionine metabolism in alfalfa cell cultures following treatment with fungal elicitors. Phytochemistry 43, 1163e1169. Elhiti, M., Ashihara, H., Stasolla, C., 2011. Distinct fluctuations in nucleotide metabolism accompany the enhanced in vitro embryogenic capacity of Brassica cells over-expressing SHOOTMERISTEMLESS. Planta 234, 1251e1265. Elhiti, M., Tahir, M., Gulden, R.H., Khamiss, K., Stasolla, C., 2010. Modulation of embryo-forming capacity in culture through the expression of Brassica genes involved in the regulation of the shoot apical meristem. J. Exp. Bot. 61, 4069e4085. Faye, F., Le Floc'h, F., 1997. Adenosine kinase of peach tree flower buds: purification and properties. Plant Physiol. Biochem. 35, 15e22. Faye, F., Le Floc'h, F., 1999. Changes in the activity of peach bud adenosine kinase from the dormancy period to bud break. J. Plant Physiol. 154, 471e476. Florchinger, M., Zimmermann, M., Traub, M., Neuhaus, H., Mohlmann, T., 2006. Adenosine stimulates anabolic metabolism in developing castor bean (Ricinus communis L.) cotyledons. Planta 223, 340e348. Freitas-Mesquita, A.L., Meyer-Fernandes, J.R., 2017. 30 -nucleotidase/nuclease in protozoan parasites: molecular and biochemical properties and physiological roles. Exp. Parasitol. 179, 1e6. Gaillard, C., Moffatt, B.A., Blacker, M., Laloue, M., 1998. Male sterility associated with APRT deficiency in Arabidopsis thaliana results from a mutation in the gene APT1. Mol. Gen. Genet. 257, 348e353. Gallois, R., Prevot, J.C., Clement, A., Jacob, J.L., 1996. Purification and characterization of an adenine phosphoribosyl-transferase from rubber tree latex physiological implications. Plant Physiol. Biochem. 34, 527e537. Gallois, R., Prevot, J.C., Clement, A., Jacob, J.L., 1997. Purification and characterization of phosphoribosylpyrophosphate synthetase from rubber tree latex. Plant Physiol. 115, 847e852.  l'e
tude du me
tabolisme des nucle
osides diet Gendraud, M., 1975. Contribution a
s in vitro en rapport avec triphosphates de tubercules de topinambour cultive
s morphoge
ne
tiques. Plant Sci. Lett. 4, 53e59. leurs potentialite
, B., Kühn, C., Rentsch, D., Brandl, B., Frommer, W.B., Gillissen, B., Bürkle, L., Andre 2000. A new family of high-affinity transporters for adenine, cytosine, and purine derivatives in Arabidopsis. Plant Cell 12, 291e300. € hlmann, T., 2014. Nucleobase and Girke, C., Daumann, M., Niopek-Witz, S., Mo nucleoside transport and integration into plant metabolism. Front. Plant Sci. 5, 443. Green, P.J., 1994. The ribonucleases of higher plants. Annu. Rev. Plant Physiol. Plant Mol. Biol. 45, 421e445. Grzelczak, Z., Buchowicz, J., 1975. Purine and pyrimidine bases and nucleosides of germinating Triticum aestivum seeds. Phytochemistry 14, 329e331. Guranowski, A., 1979a. Nucleoside phosphotransferase from yellow lupin seedling cotyledons. Biochim. Biophys. Acta 569, 13e22. Guranowski, A., 1979b. Plant adenosine kinase: purification and some properties of the enzyme from Lupinus luteus seeds. Arch. Biochem. Biophys. 196, 220e226. Guranowski, A., 1982. Purine catabolism in plants. Purification and some properties of inosine nucleosidase from yellow lupin (Lupinus leteus L.) seeds. Plant Physiol. 70, 344e349. Guranowski, A., Barankiewicz, J., 1979. Purine salvage in cotyledons of germinating lupin seeds. FEBS Lett. 104, 95e98. Guranowski, A., Pawelkiewicz, J., 1978. Adenosylhomocysteinase and adenosine nucleosidase activities in Lupinus luteus cotyledons during seed formation and germination. Planta 139, 245e247. Guranowski, A., Schneider, Z., 1977. Purification and characterization of adenosine nucleosidase from barley leaves. Biochim. Biophys. Acta 482, 145e158. Guranowski, A.B., Chiang, P.K., Cantoni, G.L., 1981. 50 -Methylthioadenosine nucleosidase. Purification and characterization of the enzyme from Lupinus luteus seeds. Eur. J. Biochem. 114, 293e299. Guranowski, A., Wasternack, C., 1982. Adenine and adenosine metabolizing enzymes in cell-free extracts from Euglena gracilis. Comp. Biochem. Physiol. B 71, 483e488. Handa, M., Guidotti, G., 1996. Purification and cloning of a soluble ATPdiphosphohydrolase (apyrase) from potato tubers (Solanum tuberosum). Biochem. Biophys. Res. Commun. 218, 916e923. Harlow, K.W., Nygaard, P., Hove-Jensen, B., 1995. Cloning and characterization of the gsk gene encoding guanosine kinase of Escherichia coli. J. Bacteriol. 177, 2236e2240. Henderson, J.F., Paterson, A.R.P., 1973. Nucleotide Metabolism-An Introduction. Academic Press, New York. Hirose, F., Ashihara, H., 1983a. Adenine phosphoribosyltransferase of Catharanthus roseus cells: purification, properties and regulation. Z. Pflanzenphysiol. 110, 135e145. Hirose, F., Ashihara, H., 1983b. Comparison of purine metabolism in suspension cultured cells of different growth phases and stem tissue of Catharanthus roseus. Z. Naturforsch. 38C, 375e381. Hirose, F., Ashihara, H., 1983c. Content and availability of 5-phosphoribosyl-1pyrophosphate in cultured cells of Catharanthus roseus. Z. Pflanzenphysiol. 110, 183e190. Hirose, F., Ashihara, H., 1984a. Changes in activity of enzymes involved in purine “salvage” and nucleic acid degradation during growth of Catharanthus roseus cells in suspension culture. Physiol. Plant 60, 532e538.

121

Hirose, F., Ashihara, H., 1984b. Fine control of purine nucleotide biosynthesis in intact cells of Catharanthus roseus. J. Plant Physiol. 116, 417e423. Hirose, N., Makita, N., Yamaya, T., Sakakibara, H., 2005. Functional characterization and expression analysis of a gene, OsENT2, encoding an equilibrative nucleoside transporter in rice suggest a function in cytokinin transport. Plant Physiol. 138, 196e206. Hirose, N., Takei, K., Kuroha, T., Kamada-Nobusada, T., Hayashi, H., Sakakibara, H., 2008. Regulation of cytokinin biosynthesis, compartmentalization and translocation. J. Exp. Bot. 59, 75e83. Hiser, C., McIntosh, L., 1990. Alternative oxidase of potato is an integral membrane protein synthesized de novo during aging of tuber slices. Plant Physiol. 93, 312e318.
, D., Fre
bortov

k, D., Kope Hluska, T., Dobrev, P.I., Tarkowska a, J., Zalaba cný, D., Plíhal, O., Kok
as, F., Briozzo, P., Zatloukal, M., Motyka, V., Galuszka, P., 2016. Cytokinin metabolism in maize: novel evidence of cytokinin abundance, interconversions and formation of a new trans-zeatin metabolic product with a weak anticytokinin activity. Plant Sci. 247, 127e137. Imagawa, H., Yamano, H., Inoue, K., Takino, Y., 1979. Purification and properties of adenosine nucleosidases from tea leaves. Agric. Biol. Chem. 43, 2337e2343. Itai, R., Nakanishi, H., Nishizawa, N.K., Yoshimura, E., Mori, S., Suzuki, K., Yamaguchi, H., 2000. Induced activity of adenine phosphoribosyltransferase (APRT) in iron-deficient barley roots: a possible role for phytosiderophore production. J. Exp. Bot. 51, 1179e1188. Ito, E., Ashihara, H., 1999. Contribution of purine nucleotide biosynthesis de novo to the formation of caffeine in young tea (Camellia sinensis) leaves. J. Plant Physiol. 154, 145e151. €rchinger, M., Kunz, H.-H., Traub, M., Wartenberg, R., Jeblick, W., Jung, B., Flo €hlmann, T., 2009. Uridine-ribohydrolase is a key regulator in Neuhaus, H.E., Mo the uridine degradation pathway of Arabidopsis. Plant Cell 21, 876e891. € hlmann, T., 2011. Arabidopsis nucleoside hydrolases Jung, B., Hoffmann, C., Mo involved in intracellular and extracellular degradation of purines. Plant J. 65, 703e711. Kahl, G., 1974. Metabolism of plant storage tissue slices. Bot. Rev. 40, 263e314. Kai, M., Takazumi, K., Adachi, H., Wasaki, J., Shinano, T., Osaki, M., 2002. Cloning and characterization of four phosphate transporter cDNAs in tobacco. Plant Sci. 163, 837e846. Kanamori-Fukuda, I., Ashihara, H., Komamine, A., 1981. Pyrimidine nucleotide biosynthesis in Vinca rosea cells: changes in the activity of de novo and salvage pathways during growth in a suspension culture. J. Exp. Bot. 32, 69e78. Katahira, R., Ashihara, H., 2006a. Profiles of purine biosynthesis, salvage and degradation in disks of potato (Solanum tuberosum L.) tubers. Planta 225, 115e126. Katahira, R., Ashihara, H., 2006b. Role of adenosine salvage in wound-induced adenylate biosynthesis in potato tuber slices. Plant Physiol. Biochem. 44, 551e555. Keough, D., Ng, A.-L., Winzor, D., Emmerson, B., de Jersey, J., 1999. Purification and characterization of Plasmodium falciparum hypoxanthineeguanineexanthine phosphoribosyltransferase and comparison with the human enzyme. Mol. Biochem. Parasitol. 98, 29e41. Kettlun, A.M., Espinosa, V., García, L., Valenzuela, M.A., 2005. Potato tuber isoapyrases: substrate specificity, affinity labeling, and proteolytic susceptibility. Phytochemistry 66, 975e982. Kettlun, A.M., Uribe, L., Calvo, V., Silva, S., Rivera, J., Mancilla, M., Antonieta, M., Valenzuela, Traverso-Cori, A., 1982. Properties of two apyrases from Solanum tuberosum. Phytochemistry 21, 551e558. King, J., Horsch, R.B., Savage, A.D., 1980. Partial characterization of two stable auxotrophic cell strains of Datura innoxia Mill. Planta 149, 480e484. Koch, A.L., Lamont, W.t.t.a.o.W.A., 1956. Some enzymes of nucleoside metabolism of Escherichia coli. J. Biol. Chem. 223, 535e549. Kombrink, E., Beevers, H., 1983. Transport of purine and pyrimidine bases and nucleosides from endosperm to cotyledons in germinating castor bean seedlings. Plant Physiol. 73, 370e376. Kong, L., Yeung, E.C., 1995. Effects of silver nitrate and polyethylene glycol on white spruce (Picea glauca) somatic embryo development: enhancing cotyledonary embryo formation and endogenous ABA content. Physiol. Plant 93, 298e304.
, M., Blaschke, H., Kope Kope cna cný, D., Vigouroux, A., Kon citíkov
a, R., Nov
ak, O.,
ra, S., von Schwartzenberg, K., 2013. Structure and Kotland, O., Strnad, M., More function of nucleoside hydrolases from Physcomitrella patens and maize catalyzing the hydrolysis of purine, pyrimidine, and cytokinin ribosides. Plant Physiol. 163, 1568e1583. Kornberg, A., Lieberman, I., Simms, E.S., 1955. Enzymatic synthesis of purine nucleotides. J. Biol. Chem. 215, 417e427. Kornberg, A., Pricer, W.E., 1951. Enzymatic phosphorylation of adenosine and 2,6diaminopurine riboside. J. Biol. Chem. 193, 481e496. Koshiishi, C., Kato, A., Yama, S., Crozier, A., Ashihara, H., 2001. A new caffeine biosynthetic pathway in tea leaves: utilisation of adenosine released from the Sadenosyl-L-methionine cycle. FEBS Lett. 499, 50e54.
mva
s, P., Planchon, S., Renaut, J., Vankova
, R., Pra
sil, I.T., 2013. ProKosov
a, K., Víta teome analysis of cold response in spring and winter wheat (Triticum aestivum) crowns reveals similarities in stress adaptation and differences in regulatory processes between the growth habits. J. Proteome Res. 12, 4830e4845. Koyama, Y., Tomoda, Y., Kato, M., Ashihara, H., 2003. Metabolism of purine bases, nucleosides and alkaloids in theobromine-forming Theobroma cacao leaves. Plant Physiol. Biochem. 41, 977e984. Krath, B.N., Eriksen, T.A., Poulsen, T.S., Hove-Jensen, B., 1999. Cloning and

122

H. Ashihara et al. / Phytochemistry 147 (2018) 89e124

sequencing of cDNAs specifying a novel class of phosphoribosyl diphosphate synthase in Arabidopsis thaliana. Biochim. Biophys. Acta 1430, 403e408. Krath, B.N., Hove-Jensen, B., 1999. Organellar and cytosolic localization of four phosphoribosyl diphosphate synthase isozymes in spinach. Plant Physiol. 119, 497e505. Krath, B.N., Hove-Jensen, B., 2001. Class II recombinant phosphoribosyl diphosphate synthase from spinach: phosphate independence and diphosphoryl donor specificity. J. Biol. Chem. 276, 17851e17856. Krenitsky, T.A., Neil, S.M., Miller, R.L., 1970. Guanine and xanthine phosphoribosyltransfer activities of Lactobacillus casei and Escherichia coli. J. Biol. Chem. 245, 2605e2611. Kurakawa, T., Ueda, N., Maekawa, M., Kobayashi, K., Kojima, M., Nagato, Y., Sakakibara, H., Kyozuka, J., 2007. Direct control of shoot meristem activity by a cytokinin-activating enzyme. Nature 445, 652e655. Kuroha, T., Tokunaga, H., Kojima, M., Ueda, N., Ishida, T., Nagawa, S., Fukuda, H., Sugimoto, K., Sakakibara, H., 2009. Functional analyses of LONELY GUY cytokinin-activating enzymes reveal the importance of the direct activation pathway in arabidopsis. Plant Cell 21, 3152e3169.

, D., Van Onckelen, H., Roef, L., Kwade, Z., Swiaa ˛ tek, A., Azmi, A., Goossens, A., Inze 2005. Identification of four adenosine kinase isoforms in tobacco BY-2 cells and their putative role in the cell cycle-regulated cytokinin metabolism. J. Biol. Chem. 280, 17512e17519. Laukens, K., Lenobel, R., Strnad, M., Van Onckelen, H., Witters, E., 2003. Cytokinin affinity purification and identification of a tobacco BY-2 adenosine kinase. FEBS Lett. 533, 63e66. Le Floc'h, F., Faye, F., 1995. Metabolic fate of adenosine and purine metabolism enzymes in young plants of peach tree. J. Plant Physiol. 145, 398e404. Le Floc'h, F., Lafleuriel, J., 1978. L'adeninephosphoribosyltransferase des pousses de topinambour Helianthus tuberosus. Phytochemistry 17, 643e646. Le Floc'h, F., Lafleuriel, J., 1981a. L'hypoxanthine-guanine phosphoribosyltransferase des pousses de topinambour Helianthus tuberosus. Z. Pflanzenphysiol. 104, 331e340. Le Floc'h, F., Lafleuriel, J., 1981b. The purine nucleosidases of Jerusalem artichoke shoots. Phytochemistry 20, 2127e2129. Le Floc'h, F., Lafleuriel, J., 1983. The role of mitochondria in the recycling of adenine into purine nucleotides in the Jerusalem artichoke (Helianthus tuberosus L.). Z. Pflanzenphysiol. 113, 61e71. Lecomte, I., Le Floc'h, F., 1999. Adenine phosphoribosyltransferase of peach tree leaves: purification and properties. J. Plant Physiol. 154, 289e295. Lee, C.C., Craig, S.P., Eakin, A.E., 1998. A single amino acid substitution in the human and a bacterial hypoxanthine phosphoribosyltransferase modulates specificity for the binding of guanine. Biochemistry 37, 3491e3498. Lee, D., Moffatt, B.A., 1993. Purification and characterization of adenine phosphoribosyl transferase from Arabidopsis thaliana. Physiol. Plant 87, 483e492. Li, G., Liu, K., Baldwin, S.A., Wang, D., 2003. Equilibrative nucleoside transporters of Arabidopsis thaliana. cDNA cloning, expression pattern, and analysis of transport activities. J. Biol. Chem. 278, 35732e35742. Lin, Z., Zhong, S., Grierson, D., 2009. Recent advances in ethylene research. J. Exp. Bot. 60, 3311e3336. Liu, X., Locasale, J.W., 2017. Metabolomics: a primer. Trend. Biochem. Sci. 42, 274e284. Liu, X., Qian, W., Liu, X., Qin, H., Wang, D., 2007. Molecular and functional analysis of hypoxanthine-guanine phosphoribosyltransferase from Arabidopsis thaliana. New Phytol. 175, 448e461. Lu, W., Su, X., Klein, M.S., Lewis, I.A., Fiehn, O., Rabinowitz, J.D., 2017. Metabolite measurement: pitfalls to avoid and practices to follow. Annu. Rev. Biochem. 86, 277e304. Mansfield, T.A., Schultes, N.P., Mourad, G.S., 2009. AtAzg1 and AtAzg2 comprise a novel family of purine transporters in Arabidopsis. FEBS Lett. 583, 481e486.
mez, D., Carabias-Martínez, R., Mateos-Vivas, M., Rodríguez-Gonzalo, E., García-Go 2015. Hydrophilic interaction chromatography coupled to tandem mass spectrometry in the presence of hydrophilic ion-pairing reagents for the separation of nucleosides and nucleotide mono-, di- and triphosphates. J. Chromatogr. A 1414, 129e137. Maurino, V.G., Grube, E., Zielinski, J., Schild, A., Fischer, K., Flügge, U.-I., 2006. Identification and expression analysis of twelve members of the nucleobaseeascorbate transporter (NAT) gene family in Arabidopsis thaliana. Plant Cell Physiol. 47, 1381e1393. Mazelis, M., Creveling, R.K., 1963. An adenosine hydrolase from Brussels sprouts. J. Biol. Chem. 238, 3358e3361. Mitsui, K., Ashihara, H., 1988. Nucleotide pools and purine metabolism in tissue cultures of wild-type Dature innoxia and adenine-requiring auxotroph. Plant Cell Physiol. 29, 1171e1183. Mittal, R., Das, J., Sharma, C.B., 1988. Purification and characterization of 50 -nucleotidase from the Golgi apparatus of peanut cotyledons. Plant Sci. 55, 93e101. Moffatt, B., Somerville, C., 1988. Positive selection for male-sterile mutants of Arabidopsis lacking adenine phosphoribosyl transferase activity. Plant Physiol. 86, 1150e1154. Moffatt, B.A., Ashihara, H., 2002. Purine and pyrimidine nucleotide synthesis and metabolism. The Arabidopsis Book. Am. Soc. Plant Biol. 1, 1e20 e0018. Moffatt, B.A., McWhinnie, E.A., Burkhart, W.E., Pasternak, J.J., Rothstein, S.J., 1992. A complete cDNA for adenine phosphoribosyltransferase from Arabidopsis thaliana. Plant Mol. Biol. 18, 653e662. Moffatt, B.A., Somerville, C.R., 1990. Purification of adenine phosphoribosyltransferase from Brassica juncea. Arch. Biochem. Biophys. 283, 484e490.

Moffatt, B.A., Wang, L., Allen, M.S., Stevens, Y.Y., Qin, W., Snider, J., von Schwartzenberg, K., 2000. Adenosine kinase of Arabidopsis. Kinetic properties and gene expression. Plant Physiol. 124, 1775e1785. Moffatt, B.A., Weretilnyk, E.A., 2001. Sustaining S-adenosyl-L-methionine-dependent methyltransferase activity in plant cells. Physiol. Plant. 113, 435e442. Mok, D.W.S., Mok, M.C., 2001. Cytokinin metabolism and action. Annu. Rev. Plant Physiol. Plant Mol. Biol. 52, 89e118. Mori, H., Iida, A., Teshiba, S., Fujio, T., 1995. Cloning of a guanosine-inosine kinase gene of Escherichia coli and characterization of the purified gene product. J. Bacteriol. 177, 4921e4926. Muller-Thungau, H., 1882. Uber Zuckranhaufung in pflanzentheilen in forge niederer temperature. Landw. Jahrb. 5, 751e828. Muller, R., Nilsson, L., Krintel, C., Hamborg Nielsen, T., 2004. Gene expression during recovery from phosphate starvation in roots and shoots of Arabidopsis thaliana. Physiol. Plant. 122, 233e243. €hlmann, T., 2014. Biochemical characNiopek-Witz, S., Deppe, J., Lemieux, M.J., Mo terization and structureefunction relationship of two plant NCS2 proteins, the nucleobase transporters NAT3 and NAT12 from Arabidopsis thaliana. Biochim. Biophys. Acta 1838, 3025e3035. Nobusawa, E., Ashihara, H., 1983. Purine metabolism in cotyledons and embryonic axes of black gram (Phaseolus mungo L.) seedlings. Int. J. Biochem. 15, 1059e1065. Nomura, A., Suno, M., Mizuno, Y., 1971. Studies on 30 -nucleotidase-nuclease from potato tubers: I. Purification and some properties of the enzyme. J. Biochem. 70, 993e1001.
k, O., Hauserov
Nova a, E., Amakorov
a, P., Dole zal, K., Strnad, M., 2008. Cytokinin profiling in plant tissues using ultra-performance liquid chromatographyeelectrospray tandem mass spectrometry. Phytochemistry 69, 2214e2224. Nygaard, P., 1983. Utilization of preformed purine bases and nucleosides. In: Munch-Petersen, A. (Ed.), Metabolism of Nucleotides, Nucleosides and Nucleobases in Microorganisms. Academic Press, London, pp. 27e93. Nyhan, W.L., 1997. The recognition of Lesch-Nyhan syndrome as an inborn error of purine metabolism. J. Inherit. Metab. Dis. 20, 171e178. Oikawa, A., Matsuda, F., Kikuyama, M., Mimura, T., Saito, K., 2011. Metabolomics of a single vacuole reveals metabolic dynamism in an alga Chara australis. Plant Physiol. 157, 544e551. Page, T., 1989. Purine nucleotide production in normal and HPRT- cells. Int. J. Biochem. 21, 1377e1381. Park, E.Y., Choi, W.S., Oh, S.-I., Kim, K.-N., Shin, J.S., Song, H.K., 2009. Biochemical and structural characterization of 50 -methylthioadenosine nucleosidases from Arabidopsis thaliana. Biochem. Biophys. Res. Commun. 381, 619e624. Park, J., Gupta, R.S., 2008. Adenosine kinase and ribokinase e the RK family of proteins. Cell. Mol. Life Sci. 65, 2875e2896. Park, J., van Koeverden, P., Singh, B., Gupta, R.S., 2007. Identification and characterization of human ribokinase and comparison of its properties with E. coli ribokinase and human adenosine kinase. FEBS Lett. 581, 3211e3216. Pierre, K.J., LePage, G.A., 1968. Formation of inosine-50 -monophosphate by a kinase in cell-free extracts of Ehrlich ascites cells in vitro. Proc. Soc. Exp. Biol. Med. 127, 432e440. Polya, G.M., Ashton, A.R., 1973. Inhibition of wheat seedling 50 (30 )-ribonucleotide phosphohydrolase by adenosine 30 ,50 -cyclic monophosphate. Plant Sci. Lett. 1, 349e357. Poulton, J.E., Butt, V.S., 1976. Partial purification and properties of adenosine nucleosidase from leaves of spinach beet (Beta vulgaris L.). Planta 131, 179e185. Prasher, D.C., Carr, M.C., Ives, D.H., Tsai, T.C., Frey, P.A., 1982. Nucleoside phosphotransferase from barley. Characterization and evidence for ping pong kinetics involving phosphoryl enzyme. J. Biol. Chem. 257, 4931e4939. Price, C.E., Murray, A.W., 1969. Purine metabolism in germinating wheat embryos. Biochem. J. 115, 129e133. Regan, S.M., Moffatt, B.A., 1990. Cytochemical analysis of pollen development in wild-type Arabidopsis and a male-sterile mutant. Plant Cell 2, 877e889. Riegler, H., Geserick, C., Zrenner, R., 2011. Arabidopsis thaliana nucleosidase mutants provide new insights into nucleoside degradation. New Phytol. 191, 349e359. Riggs, J.W., Rockwell, N.C., Cavales, P.C., Callis, J., 2016. Identification of the plant ribokinase and discovery of a role for Arabidopsis ribokinase in nucleoside metabolism. J. Biol. Chem. 291, 22572e22582. Riewe, D., Grosman, L., Fernie, A.R., Wucke, C., Geigenberger, P., 2008a. The potatospecific apyrase is apoplastically localized and has influence on gene expression, growth, and development. Plant Physiol. 147, 1092e1109. Riewe, D., Grosman, L., Fernie, A.R., Zauber, H., Wucke, C., Geigenberger, P., 2008b. A cell wall-bound adenosine nucleosidase is involved in the salvage of extracellular ATP in Solanum tuberosum. Plant Cell Physiol. 49, 1572e1579. Riondet, C., Morel, S., Alcaraz, G., 2005. Determination of total ribonucleotide pool in plant materials by high-pH anion-exchange high-performance liquid chromatography following extraction with potassium hydroxide. J. Chromatogr. A 1077, 120e127.
tel, G., 2000. Seasonal changes in adenine phosphoribosyltransferase Robert, F., Pe and adenosine kinase activities as markers of the dormancy and the growth of Fragaria x ananassa. Plant Physiol. Biochem. 38, 395e402. Robert, F., Petel, G., 2001. Nucleotide synthesis capability as a marker of the petiole elongation of strawberry plants (Fragaria X ananassa Duch.). J. Horticul. Sci. Biotechnol. 75, 690e696. Roje, S., 2006. S-Adenosyl-L-methionine: beyond the universal methyl group donor. Phytochemistry 67, 1686e1698.

H. Ashihara et al. / Phytochemistry 147 (2018) 89e124 Romanov, G.A., Lomin, S.N., Schmülling, T., 2006. Biochemical characteristics and ligand-binding properties of Arabidopsis cytokinin receptor AHK3 compared to CRE1/AHK4 as revealed by a direct binding assay. J. Exp. Bot. 57, 4051e4058. Ross, C., Murray, M.G., 1971. Development of pyrimidine-metabolizing enzymes in cotyledons of germinating peas. Plant Physiol. 48, 626e630. Ross, C.W., 1981. Biosynthesis of nucleotides. In: Stump, P.K., Conn, E.E. (Eds.), Biochemistry of Plants, vol. 6. Academic Press, New York, pp. 169e205. Rzewuski, G., Sauter, M., 2008. Ethylene biosynthesis and signaling in rice. Plant Sci. 175, 32e42. Sacher, J.A., Towers, G.H.N., Davies, D.D., 1972. Effect of light and ageing on enzymes, particularly phenylalanine ammonia lyase, in discs of storage tissue. Phytochemistry 11, 2383e2391. Sakakibara, H., 2006. Cytokinins: activity, biosynthesis, and translocation. Annu. Rev. Plant Biol. 57, 431e449. Sakakibara, H., 2010. Cytokinin biosynthesis and metabolism. In: Davies, P.J. (Ed.), Plant Hormones: Biosynthesis, Signal Transduction, Action! Springer, Dordrecht, pp. 95e114. Salavati, A., Khatoon, A., Nanjo, Y., Komatsu, S., 2012. Analysis of proteomic changes in roots of soybean seedlings during recovery after flooding. J. Proteom. 75, 878e893. Sawert, A., Backer, A.I., Wagner, K.G., 1988. Age-dependent decrease of nucleoside pools in cereal leaves. Plant Cell Physiol. 29, 61e65. Schmidt, A., Baumann, N., Schwarzkopf, A., Frommer, W., Desimone, M., 2006. Comparative studies on ureide permeases in Arabidopsis thaliana and analysis of two alternative splice variants of AtUPS5. Planta 224, 1329e1340. Schoor, S., Moffatt, B.A., 2004. Applying high throughput techniques in the study of adenosine kinase in plant metabolism and development. Front. Biosci. 9, 1771e1781. Schoor, S., Farrow, S., Blaschke, H., Lee, S., Perry, G., von Schwartzenberg, K., Emery, N., Moffatt, B., 2011. Adenosine kinase contributes to cytokinin interconversion in Arabidopsis. Plant Physiol. 157, 659e672. Sharma, C.B., Mittal, R., Tanner, W., 1986. Purification and properties of a glycoprotein adenosine 50 -monophosphatase from the plasma membrane fraction of Arachis hypogaea cotyledons. Biochim. Biophys. Acta 884, 567e577. Shimano, F., Ashihara, H., 2006. Effect of long-term phosphate starvation on the levels and metabolism of purine nucleotides in suspension-cultured Catharanthus roseus cells. Phytochemistry 67, 132e141. Shimazaki, A., Hirose, F., Ashihara, H., 1982. Changes in adenine nucleotide levels and adenine salvage during the growth of Vinca rosea cells in suspension culture. Z. Pflanzenphysiol. 106, 191e198. Sinha, S.C., Smith, J.L., 2001. The PRT protein family. Curr. Opin. Struct. Biol. 11, 733e739. Siu, K.K.W., Lee, J.E., Sufrin, J.R., Moffatt, B.A., McMillan, M., Cornell, K.A., Isom, C., Howell, P.L., 2008. Molecular determinants of substrate specificity in plant 50 methylthioadenosine nucleosidases. J. Mol. Biol. 378, 112e128. Siu, K.K.W., Asmus, K., Zhang, A.N., Horvatin, C., Li, S., Liu, T., Moffatt, B., Woods, V.L., Howell, P.L., 2011. Mechanism of substrate specificity in 50 -methylthioadenosine/ S-adenosylhomocysteine nucleosidases. J. Struct. Biol. 173, 86e98. Stasolla, C., Katahira, R., Thorpe, T.A., Ashihara, H., 2003. Purine and pyrimidine nucleotide metabolism in higher plants. J. Plant Physiol. 160, 1271e1295. Stasolla, C., Loukanina, N., Ashihara, H., Yeung, E.C., Thorpe, T.A., 2001a. Ascorbic acid changes the pattern of purine metabolism during germination of white spruce somatic embryos. Tree Physiol. 21, 359e367. Stasolla, C., Loukanina, N., Ashihara, H., Yeung, E.C., Thorpe, T.A., 2001b. Purine and pyrimidine metabolism during the partial drying treatment of white spruce (Picea glauca) somatic embryos. Physiol. Plant 111, 93e101. Stasolla, C., Loukanina, N., Yeung, E.C., Thorpe, T.A., 2005. Progression of programmed cell death in tobacco BY-2 cells is delineated by specific changes in de novo and salvage synthesis of purine nucleotides. Physiol. Plant 123, 254e261. Stasolla, C., Loukanina, N., Ashihara, H., Yeung, E.C., Thorpe, T.A., 2006. Changes of purine and pyrimidine nucleotide biosynthesis during shoot initiation from epicotyl explants of white spruce (Picea glauca). Plant Sci. 171, 345e354. Stasolla, C., Thorpe, T.A., 2004. Purine and pyrimidine nucleotide synthesis and degradation during in vitro morphogenesis of white spruce (Picea glauca). Front. Biosci. 9, 1506e1519. Stasolla, C., Yeung, E.C., 1999. Ascorbic acid improves the conversion of white spruce somatic embryos. In Vitro Cell Dev. Biol. Plant 35, 316e319. Struck, A.-W., Thompson, M.L., Wong, L.S., Micklefield, J., 2012. S-Adenosyl-methionine-dependent methyltransferases: highly versatile enzymes in biocatalysis, biosynthesis and other biotechnological applications. Chem Bio Chem 13, 2642e2655. Sukrong, S., Yun, K.Y., Stadler, P., Kumar, C., Facciuolo, T., Moffatt, B.A., Falcone, D.L., 2012. Improved growth and stress tolerance in the arabidopsis oxt1 mutant triggered by altered adenine metabolism. Mol. Plant 5, 1310e1332. Suno, M., Nomura, A., Mizuno, Y., 1973. Studies on 3'-nucleotidase-nuclease from potato tubers II. Further studies on substrate specificity and mode of action. J. Biochem. 73, 1291e1297. Suzuki, M., Yasumoto, E., Baba, S., Ashihara, H., 2003. Effect of salt stress on the metabolism of ethanolamine and choline in leaves of the betaine-producing mangrove species Avicennia marina. Phytochemistry 64, 941e948. Szuwart, M., Starzynska, E., Pietrowska-Borek, M., Guranowski, A., 2006. Calciumstimulated guanosine-inosine nucleosidase from yellow lupin (Lupinus luteus). Phytochemistry 67, 1476e1485. Takahashi, T., Kakehi, J.-I., 2010. Polyamines: ubiquitous polycations with unique roles in growth and stress responses. Ann. Bot. 105, 1e6.

123

Tarr, H.L.A., 1955. Fish muscle riboside hydrolases. Biochem. J. 59, 386e391. Trifunovi
c, M., Motyka, V., Cingel, A., Suboti
c, A., Jevremovi
c, S., Petri
c, M., Holík, J.,  2015. Changes in cytokinin content and Malbeck, J., Dobrev, P.I., Dragi
cevi
c, I.C., altered cytokinin homeostasis in AtCKX1 and AtCKX2-overexpressing centaury (Centaurium erythraea Rafn.) plants grown in vitro. Plant Cell Tissue Organ Cult. 120, 767e777. Turnbull, C.G.N., Hanke, D.E., 1985a. The control of bud dormancy in potato tubers. Planta 165, 359e365. Turnbull, C.G.N., Hanke, D.E., 1985b. The control of bud dormancy in potato tubers. Measurement of the seasonal pattern of changing concentrations of zeatincytokinins. Planta 165, 366e376. Ukaji, T., Ashihara, H., 1986. Purine salvage in mitochondria of cultured Catharanthus roseus cells. J. Plant Physiol. 125, 191e197. Ukaji, T., Ashihara, H., 1987. Effect of inorganic phosphate on synthesis of 5phosphoribosyl-1-pyrophosphate in cultured plant cells. Int. J. Biochem. 19, 1127e1131. Ullah, A.H.J., Gibson, D.M., 1988. Purification and characterization of acid phosphatase from cotyledons of germinating soybean seeds. Arch. Biochem. Biophys. 260, 514e520. Voet, D., Voet, J.G., 2010. Biochemistry, 4th Edition. Wiley, Hoboken, NJ, USA. von Schwartzenberg, K., Kruse, S., Reski, R., Moffatt, B., Laloue, M., 1998. Cloning and characterization of an adenosine kinase from Physcomitrella involved in cytokinin metabolism. Plant J. 13, 249e257. Waduwara-Jayabahu, I., Oppermann, Y., Wirtz, M., Hull, Z.T., Schoor, S., Plotnikov, A.N., Hell, R., Sauter, M., Moffatt, B.A., 2012. Recycling of methylthioadenosine is essential for normal vascular development and reproduction in Arabidopsis. Plant Physiol. 158, 1728e1744. Wagner, K.G., Backer, A.I., 1992. Dynamics of nucleotides in plants studied on a cellular basis. In: Jeon, K.W., Friedlander, M. (Eds.), International Review of Cytology, vol. 134. Academic Press, San Diego, pp. 1e84. Wang, H., Hao, L., Shung, C.-Y., Sunter, G., Bisaro, D.M., 2003. Adenosine kinase is inactivated by geminivirus AL2 and L2 proteins. Plant Cell 15, 3020e3032. Wasternack, C., 1982. Metabolism of pyrimidines and purines. In: Pirson, A., Zimmermann, M.H. (Eds.), Encyclopedia of Plant Physiology, New Ser, 14B. Springer, Berlin, pp. 263e301. Weretilnyk, E.A., Alexander, K.J., Drebenstedt, M., Snider, J.D., Summers, P.S., Moffatt, B.A., 2001. Maintaining methylation activities during salt stress. The involvement of adenosine kinase. Plant Physiol. 125, 856e865. Wilson, C.M., 1975. Plant nucleases. Annu. Rev. Plant Physiol. 26, 187e208. Wilson, C.M., 1982. Plant nucleases: biochemistry and development of multiple molecular forms. Isozymes 6, 33e54. Witz, S., Panwar, P., Schober, M., Deppe, J., Pasha, F.A., Lemieux, M.J., Moḧlmann, T., 2014. Structure-function relationship of a plant NCS1 member - homology modeling and mutagenesis identified residues critical for substrate specificity of PLUTO, a nucleobase transporter from arabidopsis. PLoS One 9. € € Wormit, A., Traub, M., FlOrchinger, M., Neuhaus, H.E., MOhlmann, T., 2004. Characterization of three novel members of the Arabidopsis thaliana equilibrative nucleoside transporter (ENT) family. Biochem. J. 383, 19e26. Wylegalla, C., Meyer, R., Wagner, K.G., 1985. Nucleotide pools in suspensioncultured cells of Datura innoxia II. Correlation with nutrient uptake and macromolecular synthesis. Planta 166, 446e451. Yang, S.F., Hoffman, N.E., 1984. Ethylene biosynthesis and its regulation in higher plants. Annu. Rev. Plant Physiol. Plant Mol. Biol. 35, 155e189. Yin, Y., Katahira, R., Ashihara, H., 2014. Metabolism of purine nucleosides and bases in suspension-cultured Arabidopsis thaliana cells. Eur. Chem. Bull. 3, 925e934. Yin, Y., Shimano, F., Ashihara, H., 2007. Involvement of rapid nucleotide synthesis in recovery from phosphate starvation of Catharanthus roseus cells. J. Exp. Bot. 58, 1025e1033. Yokozawa, T., Oura, H., 1986. Contents of free adenine in soybeans from several countries. Agric. Biol. Chem. 50, 1317e1319. Young, L.-S., Harrison, B.R., Narayana Murthy, U.M., Moffatt, B.A., Gilroy, S., Masson, P.H., 2006. Adenosine kinase modulates root gravitropism and cap morphogenesis in Arabidopsis. Plant Physiol. 142, 564e573. Zaimenko, I., Lisec, J., Stein, U., Brenner, W., 2017. Approaches and techniques to characterize cancer metabolism in vitro and in vivo. Biochim. Biophys. Acta 1868, 412e419. Zhang, C., Guinel, F.C., Moffatt, B.A., 2002. A comparative ultrastructural study of pollen development in Arabidopsis thaliana ecotype Columbia and male-sterile mutant apt1-3. Protoplasma 219, 59e71. Zhang, X., Chen, Y., Lin, X., Hong, X., Zhu, Y., Li, W., He, W., An, F., Guo, H., 2013. Adenine phosphoribosyl transferase 1 is a key enzyme catalyzing cytokinin conversion from nucleobases to nucleotides in arabidopsis. Mol. Plant 6, 1661e1672. Zhivotovsky, B., Galluzzi, L., Kepp, O., Kroemer, G., 2009. Adenine nucleotide translocase: a component of the phylogenetically conserved cell death machinery. Cell Death Differ. 16, 1419e1425. Zhou, C.-J., Li, J., Zou, J.-C., Liang, F.-S., Ye, C.-J., Jin, D.-M., Weng, M.-L., Wang, B., 2006. Cloning and characterization of a second form of the rice adenine phosphoribosyl transferase gene (OsAPT2) and its association with TGMS. Plant Mol. Biol. 60, 365e376. Zimmermann, H., 1992. 50 -Nucleotidase: molecular structure and functional aspects. Biochem. J. 285, 345e365. Zrenner, R., Ashihara, H., 2011. Nucleotide metabolism. In: Ashihara, H., Crozier, A., Komamine, A. (Eds.), Plant Metabolism and Biotechnology. Wiley, Chichester, UK, pp. 135e162.

124

H. Ashihara et al. / Phytochemistry 147 (2018) 89e124

Zrenner, R., Stitt, M., Sonnewald, U., Boldt, R., 2006. Pyrimidine and purine biosynthesis and degradation in plants. Annu. Rev. Plant Biol. 57, 805e836. Zybailov, B., Rutschow, H., Friso, G., Rudella, A., Emanuelsson, O., Sun, Q., van Wijk, K.J., 2008. Sorting signals, N-terminal modifications and abundance of the chloroplast proteome. PLoS One 3, e1994.

Hiroshi Ashihara is an Emeritus Professor at Ochanomizu University in Tokyo. In 1973 he was appointed as a Research Associate at Ochanomizu University and he continued to work there as an Associate Professor and then a Full Professor for nearly 40 years. He obtained his Ph.D from the University of Tokyo in 1975 and carried out postdoctoral research in the Department of Biochemistry at the University of Sheffield in the UK from 1977 to 1979. He began research on plant carbohydrate metabolism and then expanded to nucleotide metabolism including the biosynthesis of purine and pyridine alkaloids. He has published over 200 papers, more than 20 reviews and has co-edited three books on plant metabolism. He has been a visiting professor at the University of Copenhagen, Denmark and a visiting scientist at the University of Calgary, Canada.

Claudio Stasolla is a Professor of Plant Physiology at the University of Manitoba, Canada. He obtained his Ph.D from the University of Calgary, Canada, in 2001, and did postdoctoral research in the Forest Biotechnology Group at the University of North Carolina, USA. He has published over 100 papers on research related to plant morphogenesis and development using experimental in-vitro systems, such as somatic embryogenesis and organogenesis of conifers and angiosperms. Over the years he has contributed to elucidating changes in purine and pyrimidine metabolism during the formation of embryos and organs. His current research focuses on cell fate acquisition and retention during conditions of stress.

Tatsuhito Fujimura is an Emeritus Professor at the University of Tsukuba in Japan. He obtained his Ph.D in Plant Physiology from the University of Tokyo in 1979. He was a research leader of the Plant Biotechnology Group at the Mitsui Toatsu Chemical Company working on genetic manipulation of rice. He has published over 100 papers on research related to genetics, breeding, physiology, molecular biology and biotechnology of rice and other crop plants.

Alan Crozier obtained a Ph.D from the University of London and carried out postdoctoral research at the University of Calgary. He is currently a Senior Researcher in the Department of Nutrition at the University of California, Davis and a consultant for Mars Inc. He is also an Honorary Senior Research Fellow in the School of Medicine, Dentistry and Nursing at the University of Glasgow where he was previously Professor of Plant Biochemistry and Human Nutrition. He has published over 300 papers and edited nine books. His research interests include purine alkaloids, and dietary flavonoids and related phenolic compounds in fruits, vegetables and beverages and their bioavailability following ingestion. He was elected an Eminent Scientist of RIKEN, the Institute of Physical and Chemical Research in Japan, in 1999 for his achievements in the field of plant hormones and secondary metabolites. In 2013 he was awarded the Mars Prize for his research on “Flavonoid Metabolism/Chemistry” at the Sixth International Conference on Polyphenols and Health, in Buenos Aires, Argentina. He was a Thomson Reuters Highly Cited Researcher in 2014, 2015 and 2016.