Metabolism and Long-distance Translocation of Cytokinins

Authors


*Corresponding author
Tel: +81 45 503 9576; Fax: +81 45 503 9609; E-mail: sakaki@riken.jp

Abstract

During plant development, distantly-located organs must communicate in order to adapt morphological and physiological features in response to environmental inputs. Among the recognized signaling molecules, a class of phytohormones known as the cytokinins functions as both local and long-distance regulatory signals for the coordination of plant development. This cytokinin-dependent communication system consists of orchestrated regulation of the metabolism, translocation, and signal transduction of this phytohormone class. Here, to gain insight into this elaborate signaling system, we summarize current models of biosynthesis, trans-membrane transport, and long-distance translocation of cytokinins in higher plants.

inline imageHitoshi Sakakibara
(Corresponding author)

Introduction

Higher plants are composed of multiple organ systems that are functionally differentiated, such as photosynthetic and non-photosynthetic organs and vegetative and reproductive organs. Plant organs interact with each other to optimize both metabolic and developmental processes to allow the organism to accommodate to environmental inputs. For these mutual interactions, local and long-distance communication among cells and organs are essential. Messenger molecules, such as phytohormones, mRNA, small RNAs and proteins, are involved in this communication system and are transported throughout the plant by the vascular system (Ruiz-Medrano et al. 2001; Mouchel and Leyser 2007; Liu et al. 2009). Cytokinins, a class of phytohormones, are one of these long-distance messengers transported through the plant vascular system.

Cytokinins are defined as substances that induce cytokinesis in the presence of auxin. To date, a variety of natural cytokinin species, including trans-zeatin (tZ), N6-(Δ2-isopentenyl)adenine (iP), cis-zeatin (cZ), and their conjugates have been identified; the active cytokinin species are the free-base type (Mok and Mok 2001). In addition to their action as inducers of cytokinesis, cytokinins are also involved in regulating various biological processes: senescence (Gan and Amasino 1995; Kim et al. 2006), apical dominance (Sachs and Thimann 1967; Tanaka et al. 2006; Shimizu-Sato et al. 2009), root proliferation (Werner et al. 2001, 2003), phyllotaxis (Giulini et al. 2004), and reproductive competence (Ashikari et al. 2005). To regulate such plant developmental processes, cytokinin activity must be finely controlled.

Cytokinin activity in an organ is regulated at diverse steps, including de novo synthesis, activation, conjugation, and degradation. Spatial distribution of cytokinin signaling systems (i.e. receptors and response regulators) specifies the domain in which a cytokinin response can occur. In addition, local and long-distance transport systems are involved in regulating cytokinin action. In the present review, we summarize the role of cytokinins as a signaling messenger with focus on the biosynthesis, transmembrane transport, and long-distance translocation. For an update on intracellular cytokinin signal transduction, readers are directed to other excellent reviews (Hwang et al. 2002; Heyl and Schmülling 2003; Kakimoto 2003; Mizuno 2004; Argueso et al. 2009).

Cytokinin Metabolism in Higher Plants

In the past decade, our understanding of cytokinin biosynthesis has greatly progressed due in large part to the identification of key pathway genes encoding adenosine phosphate-isopentenyltransferase (IPT; Kakimoto et al. 2001; Takei et al. 2001a; Sakamoto et al. 2006), tRNA-isopentenyltransferase (tRNA-IPT; Miyawaki et al. 2004, 2006; Sakamoto et al. 2006), cytokinin trans-hydroxylase, CYP735A (Takei et al. 2004b) and the cytokinin nucleoside 5′-monophosphate phosphoribohydrolase, LONELY GUY (LOG; Kurakawa et al. 2007). Based on these findings, a basic scheme for the cytokinin biosynthesis pathway was proposed (Sakakibara 2006; Hirose et al. 2008; Kamada-Nobusada and Sakakibara 2009) and is illustrated in Figure 1.

Figure 1.

Simplified model for cytokinin biosynthesis and degradation pathways.
Blue arrows indicate reactions for enzymes with known genes, and grey arrows indicate pathway genes that remain to be identified. In this scheme, only biosynthesis and degradation steps are illustrated; further details can be found in Sakakibara (2006). CKX, cytokinin oxidase/dehydrogenase; cZ, cis-zeatin; DMAPP, dimethylallyl diphosphate; iP, N6-(Δ2-isopentenyl)adenine; IPT, adenosine phosphate-isopentenyltransferase; LOG, LONELY GUY; tRNA-IPT, tRNA-isopentenyltransferase; tZ, trans-zeatin.

In Arabidopsis, the initial step of iP and tZ biosynthesis is catalyzed by IPT using dimethylallyl diphosphate (DMAPP) and adenosine 5′-diphosphate (ADP), or adenosine 5′-triphosphate (ATP), to generate iP-ribotides. These iP-ribotides are then hydroxylated to tZ-ribotides by CYP735A1 or CYP735A2 (Takei et al. 2004b). On the other hand, biosynthesis of cZ is initiated by tRNA-IPTs that catalyze the prenylation of tRNA using DMAPP; however, the enzyme for cis-hydroxylation has yet to be identified in plants. Conversion of iP-, tZ-, and cZ-riboside 5′-monophosphate to their active forms occurs by two pathways: the LOG and two-step pathways. In the former, cytokinin riboside 5′-monophosphates are directly converted to free-base cytokinins by LOG (Kurakawa et al. 2007). In the latter pathway, the ribotides are dephosphorylated to the ribosides and subsequently converted to free-base cytokinins (Chen and Kristopeit 1981a, 1981b), but the corresponding genes have not yet been identified.

Inactivation of cytokinins is carried out by degradation or conjugation. Degradation is catalyzed by cytokinin oxidase/dehydrogenase (CKX; Galuszka et al 2001; Schmülling et al. 2003; Figure 1). Glucose-conjugation to cytokinins occurs at the N3, N7 and N9 positions of the purine ring, or in the hydroxyl group of the prenyl side chain. Recent studies have demonstrated that the degradation step plays an important role in regulating cytokinin activity (Werner et al. 2001, 2003; Ashikari et al. 2005).

Traditionally it was thought that cytokinins were synthesized in the root and transported to the shoots through the xylem (Letham and Palni 1983; Beveridge et al. 1997). However, recent studies on the spatial distribution of cytokinin metabolism have demonstrated that cytokinins are produced not only in roots, but also in various sites within the aerial parts of the plant. In Arabidopsis, the IPT genes are expressed in numerous organs including roots, leaves, stems, flowers, and siliques (Miyawaki et al. 2004; Takei et al. 2004a), whereas the CYP735A genes are expressed predominantly in roots (Takei et al. 2004b). Recent studies on the LOG family genes in rice and Arabidopsis suggest that activation of cytokinin occurs in nearly all parts of the plant (Kurakawa et al. 2007; Kuroha et al. 2009). Superimposition of the expression patterns for the IPT and CYP735A genes, in Arabidopsis, reveals the differential distribution of de novo synthesis pathways for iP and tZ. For instance, AtIPT3 is expressed in phloem tissue in rosette leaves, whereas expression of CYP735As in rosette leaves is scarcely detectable. Alternatively, both IPTs and CYP735As are expressed in roots. Such differential distribution of these cytokinin biosynthesis genes might be important to produce the various cytokinin species in underground and aboveground organs.

Cytokinin Transport across the Plasma Membrane

As cytokinins are a mobile class of phytohormones, it is likely that higher plants have import and export systems to mobilize the cytokinin across the plasma membrane (Cedzich et al. 2008; Hirose et al. 2008). Characterization of cytokinin transport in Arabidopsis cell cultures suggested the presence of proton-coupled multiphasic cytokinin transport systems (Cedzich et al. 2008). To date, the purine permease (PUP) family and the equilibrative nucleoside transporter (ENT) family have been proposed as candidates for cytokinin transporters. Among Arabidopsis PUP family proteins (Gillissen et al. 2000), the ability of AtPUP1 and AtPUP2 to transport tZ and iP was shown using a yeast system (Bürkle et al. 2003); however, genetic studies on plant PUPs using loss-of-function or gain-of-function mutants have not been reported.

For plant ENT proteins, competitive uptake studies in yeast cells suggested that Arabidopsis ENT3, ENT6, ENT7 and rice ENT2 can transport iP-riboside (iPR) and tZ-riboside (tZR) (Hirose et al. 2005, 2008). Genetic screening for suppressor mutants of cytokinin overproduction (Sun et al. 2003) resulted in the identification of T-DNA insertion lines in which AtENT8 expression was downregulated (Sun et al. 2005). Although these results suggest that plant ENT proteins are involved in the transport of cytokinin ribosides, clear and definitive evidence has not been provided. Figure 2 summarizes the manner in which these putative cytokinin membrane transporters function. Clearly, further genetic studies on PUPs, ENTs, and other transporters are needed to fully understand such putative cytokinin transport systems establishing concentration gradients within specific tissues/organs.

Figure 2.

A possible model of cytokinin transport across the plasma membrane.
Arabidopsis purine permease 1 (AtPUP1) and AtPUP2 can transport free-base cytokinins, such as iP and tZ, in a proton-coupled manner (Gillissen et al. 2000; Bürkle et al. 2003). Some equilibrative nucleoside transporter (ENT) family proteins of Arabidopsis and rice can transport cytokinin ribosides such as iPR and tZR (Hirose et al. 2005, 2008; Sun et al. 2005). ENT family proteins, which have been identified also in mammals, fungi, and bacteria, facilitate diffusion of nucleosides along a concentration gradient (Hyde et al. 2001). iP, N6-(Δ2-isopentenyl)adenine; iPR, N6-(Δ2-isopentenyl)adenine riboside; tZ, trans-zeatin; tZR, trans-zeatin riboside.

Long-Distance Transport of Cytokinins

In higher plants, long-distance translocation of cytokinins is mediated by the xylem, an acropetal transport system that occurs by transpiration flow, and the phloem translocation system that delivers photosynthate throughout the body of the plant. Systemic translocation of cytokinins was implied by early tracer experiments. Although radioactive cytokinins applied to leaves are strongly retained at the treated site, a small proportion of the labeled cytokinins are translocated to other plant parts (Vonk and Davelaar 1981; Badenoch-Jones et al. 1984; Abo-hamed et al. 1984; Letham 1994). In xylem sap, the major form of cytokinin is tZR (Beveridge et al. 1997; Takei et al. 2001b; Hirose et al. 2008), and in phloem sap, the major forms are iP-type cytokinins, such as iPR and iP-ribotides (Corbesier et al. 2003; Hirose et al. 2008). Thus, it is conceivable that plants might use tZR as an acropetal messenger and iP-type cytokinins as systemic or basipetal messengers (Figure 3).

Figure 3.

A model for long-distance cytokinin transport through the plant vascular system.
In the xylem (pale red column) and phloem (pale blue column), tZR and iP-type cytokinins are major transported species, respectively (Corbesier et al. 2003; Hirose et al. 2008). Supplying nitrate to Arabidopsis roots induces expression of AtIPT3, a cytokinin biosynthesis gene (Takei et al. 2004a) (indicated by circled red letters), which subsequently upregulates the translocation of cytokinins (tZR) through the xylem (red broken arrow). Cytokinin trans-hydroxylases (CYP735A1 and CYP735A2 in Arabidopsis) are involved in the synthesis of tZR (Takei et al. 2004b). Xylem cytokinins (tZR) are translocated acropetally (red arrow) by the transpiration stream (cyan arrow). Nitrate ions are also transported via xylem (black arrow) and assimilated into amino acids in the leaves. Phloem cytokinins (iP) are translocated systemically or basipetally (blue arrows). Cytokinin biosynthesis and response (purple arrows) occur at numerous sites throughout the plant. CK, cytokinin; iP-CK, N6-(Δ2-isopentenyl)adenine-type cytokinins; IPT, adenosine phosphate-isopentenyltransferase; N, nitrogen; tZ, trans-zeatin; tZR, trans-zeatin riboside.

This hypothesis is supported by a recent grafting experiment using an atipt1;3;5;7 mutant, in which the content of both iP-type and tZ-type cytokinins decreased in comparison with wild-type plants (Miyawaki et al. 2006). Wild-type root-stocks recovered the tZ-type cytokinins in the mutant shoot-scions but not the iP-type cytokinins (Matsumoto-Kitano et al. 2008). Wild-type shoot-scions recovered the iP-type cytokinins in the mutant root-stocks to normal levels, whereas the tZ-type cytokinins were only partially recovered (Matsumoto-Kitano et al. 2008). Reciprocal grafting experiments also restored visible mutant phenotypes, such as defects in the thickening growth of roots and inflorescence stems (Matsumoto-Kitano et al. 2008). Summary of this grafting study is illustrated in Figure 4A and C.

Figure 4.

Schematic illustrations of grafting studies used to explore the long-distance translocation of cytokinins.
(A) An Arabidopsis mutant, atipt1;3;5;7, in which the content of both iP-type (iP-CK) and tZ-type (tZ-CK) cytokinins is low, were reciprocally grafted with wild type plants (WT). WT root-stocks recovered the tZ-CK in the mutant shoot-scions but not the iP-CK. WT shoot-scions recovered the iP-CK in mutant root-stocks to normal levels, whereas the tZ-CK level was only partially recovered. Small, middle, and large ellipses around iP-CK or tZ-CK schematically indicate each level of these cytokinins, representing decreased, partially recovered and normal levels, respectively. Based on the work of Matsumoto-Kitano et al. (2008).
(B) Increased-branching mutants of pea (rms4; Beveridge et al. 1996, 1997) and Arabidopsis (max2; Stirnberg et al. 2002; Foo et al. 2007), in which xylem cytokinin content is decreased, were reciprocally grafted with WT plants. Although the WT shoot-scion could restore the xylem cytokinin content, the WT root-stock could not. Cytokinin levels in xylem sap from root-stock are indicated with thick (for the normal level) or narrow (for a low level) red arrows.
(C) Speculative models to explain the results of two grafting experiments described in (A) (left side) and (B) (right side). The graft experiments in (A) indicate that iP-CK can move basipetally but not acropetally (blue arrow) and that tZ-CK can move acropetally (thick orange arrow) rather than basipetally (narrow orange arrow). This supports the hypothesis that plants might use tZR as an acropetal messenger and iP-CK as basipetal messenger (Figure 3). The partial recovery of tZ-CK in atipt1;3;5;7 root-stocks grafted to WT scions is possibly caused by biosynthesis of tZ-CK from recovered iP-CK, or basipetal translocation of tZ-CK. The latter result (B) is consistent with the hypothesis that some endogenous and basipetal signals (broken green arrow) regulate xylem cytokinin translocation (red arrow).

Cytokinin translocation via the xylem is controlled both by environmental and endogenous signals. The tZR content and flow rate of the xylem sap are significantly increased by nitrate supplement in barley (Samuelson et al. 1992) and maize (Takei et al. 2001b), implying that tZR acts as a messenger for nitrate signaling. Xylem cytokinins upregulated by nitrate supplement induced the accumulation of cytokinin-responsive gene transcripts in leaves (Sakakibara et al. 1998; Takei et al. 2001b). In Arabidopsis roots, the accumulation of AtIPT3 transcripts is induced by nitrate, followed by that of tZ-ribotides and tZR (Takei et al. 2002, 2004a) (Figure 3). Furthermore, in an AtIPT3-deficient mutant, the nitrate-dependent accumulation of cytokinins was markedly reduced or diminished (Takei et al. 2004a), indicating that AtIPT3 is a key gene for the nitrate-dependent de novo biosynthesis of cytokinins. Promoter-reporter analyses with transgenic Arabidopsis plants showed that the AtIPT3 promoter is active in phloem companion cells rather than xylem tissues (Takei et al. 2004a). Thus, there may well be a cytokinin translocation system operating between the phloem and xylem tissues.

Studies on the increased-branching mutants of pea (rms4; Beveridge et al. 1996) and Arabidopsis (max2; Stirnberg et al. 2002; Foo et al. 2007), in which the cytokinin content of the xylem sap is dramatically reduced; imply a novel control mechanism for cytokinin delivery via the xylem transpiration stream. Reciprocal grafting experiments carried out between rms4 and wild type plants showed that the reduction in xylem cytokinin concentration and increase in branching occurs in a scion-dependent manner (Beveridge et al. 1996). A scion-dependent reduction in xylem cytokinin content was also observed by reciprocal grafting experiments between max2 and wild type plants (Foo et al. 2007). These results imply that some endogenous and basipetal signals are involved in controlling long-distance cytokinin movement. Summary of these grafting studies are illustrated in Figure 4B and C. Currently, in addition to cytokinin and auxin, strigolactone is proposed as a novel phytohormone involved in branching (Gomez-Roldan et al. 2008; Umehara et al. 2008), but the identity of the signals regulating xylem cytokinin levels remains to be elucidated.

Future Perspectives

Over the past decade, identification and characterization of cytokinin-related genes has greatly advanced our understanding of cytokinin metabolism and signal transduction; however, to fully elucidate the global signaling system for cytokinins, the following issues need to be resolved. First, tZ-type and iP-type cytokinins are differentially distributed in xylem and phloem tissues, implying that they transfer different biological messages. At present, the physiological meaning of side chain structural variation remains to be solved. An important approach to answering this question is to use loss-of-function mutants of CYP735As, in which tZ-type cytokinin is expected to decrease or disappear. In addition, further characterization of ligand specificity and functional differentiation of cytokinin receptors will provide a basis to address this question. Second, it is now known that nitrate acts as an environmental variable to control long-distance cytokinin transport. However, additional factors are likely to be involved in regulating this transport system. For instance, light conditions also affect cytokinin delivery to the shoots by changing the transpiration rate (Boonman et al. 2007). Elucidation of mutual interactions between cytokinin delivery and these factors will be necessary in order to full understand the physiological role of long-distance cytokinin translocation for plant growth and development.

Agricultural Implications

An increase of agricultural food production worldwide over the past four decades has been associated with a remarkable increase in the use of fertilizers (Hirel et al. 2007). Concomitantly, this fertilizer usage also caused environmental concerns such as the eutrophication of freshwater (London 2005) and marine ecosystems (Beman et al. 2005). Since the world population is growing, it will be essential over the foreseeable future that increases in food production be achieved without further negative impacts on the global environment. To meet this important requirement, plant scientists will need to advance our understanding of approaches that can be used to develop crop plants with enhanced nutrient use efficiency. Plant productivity can be regulated by two factors: the morphological (e.g. grain number) and metabolic (e.g. uptake and efficient use of nutrients) attributes. Cytokinins are closely relevant to both of these characteristics.

In morphological aspects, a quantitative trait loci analysis of rice grain number revealed that cytokinin activity is an important factor to define grain number (Ashikari et al. 2005). The loss-of-function mutations in the LOG gene caused severe reduction in panicle size, abnormal branching patterns, and a decrease in the number of flowers and stamens (Kurakawa et al. 2007). In metabolic aspects, cytokinins are involved in the regulation of various genes encoding transporters for nitrate, ammonium, sulfate, phosphate, and iron (Sakakibara et al. 2006; Séguéla et al. 2008). Since such transporters mediate primary uptake and proper allocation of essential nutrients, they play an important role in efficient nutrient acquisition and usage.

Cytokinins also increase sink strength as initially demonstrated by movement of radioactive metabolites, such as carbohydrates and amino acid, from other plant parts to cytokinin-treated sites (Mothes et al. 1961; Mothes and Engelbrecht 1963; Kuiper 1993; Werner et al. 2008). In the rice grain, zeatin content per grain undergoes an increase after heading and the highest levels are present during the period of maximal grain weight (Oritani and Yoshida 1976). Similarly in barley, it was reported that cytokinins participate in the regulation of grain size, possibly by influencing both accumulation and the duration of the filling period (Mechael and Seiler-Kelbitsch 1972). These findings indicate that cytokinins are likely involved in grain filling, a process that involves the active mobilization of nutrients from a vascular system to the endosperm. Thus, advancing our understanding of the regulatory network of cytokinin signaling, metabolism, translocation, and action may well provide a key to opening the door for further improvement of agricultural outputs, including grain yield.

(Co-Editor: William J. Lucas)

Acknowledgements

We thank Dr Norihito Nakamichi, RIKEN Plant Science Center, Japan, for his helpful comments and critical reading of the manuscript.

Ancillary