Plant Cell Physiol. 46(5): 716–728 (2005) doi:10.1093/pcp/pci074, available online at www.pcp.oupjournals.org JSPP © 2005 Tissue Localization of in Maize: Possible Involvement of Quinone Species Generated from Plant Phenolics by Other Enzymatic Systems in the Catalytic Reaction

Petr Galuszka 1, Jitka Frébortová 2, Lenka Luhová 3, Kristin D. Bilyeu 4, James T. English 5 and Ivo Frébort 1, 6 1 Division of Molecular Biology, Department of Biochemistry, Palacký University, Šlechtitelu 11, 783 71 Olomouc, Czech Republic 2 Laboratory of Growth Regulators, Palacký University/Institute of Experimental Botany of the Academy of Science, Šlechtitelu 11, 783 71 Olomouc, Czech Republic 3 Department of Biochemistry, Palacký University, Šlechtitelu 11, 783 71 Olomouc, Czech Republic 4 United States Department of Agriculture / Agricultural Research Service, University of Missouri, Colombia, MO 65211, USA 5 Department of Plant Microbiology and Pathology, University of Missouri, Columbia, MO 65211, USA

; The degradation of cytokinins in plants is controlled Introduction by the cytokinin dehydrogenase (EC 1.5.99.12). Cytokinin dehydrogenase from maize showed the ability to Cytokinins are plant hormones that contribute to the regu- use oxidation products of guaiacol, 4-methylcatechol, aceto- lation of numerous developmental processes including apical syringone and several other compounds as electron accep- dominance, flower and fruit development, leaf senescence and tors. These results led us to explore the cability for indirect seed germination (reviewed in Haberer and Kieber 2002). production of suitable electron acceptors by different qui- Diverse cytokinins have been characterized in a variety of plant none-generating . The results reported here re- species (Auer 1997). Differences among cytokinins are related vealed that the electron acceptors may be generated in vivo predominantly to structural features, including the isoprenoid from plant phenolics by other enzymatic systems such as side chain that is attached to an adenine moiety (Mok and Mok peroxidase and tyrosinase/laccase/catechol oxidase. Histo- 2001). The small size of active cytokinins is thought to con- chemical localization of cytokinin dehydrogenase by activ- tribute to their common occurrence in the intercellular or ity staining and immunochemistry using optical and apoplastic region of plant tissues (Corbesier et al. 2003). Here, confocal microscopy showed that cytokinin dehydrogenase they bind to cell surface receptors and initiate a signal trans- is most abundant in the aleurone layer of maize kernels and duction cascade that leads to activation of genes involved in in phloem cells of the seedling shoots. Cytokinin de- tissue development (Yamada et al. 2001, Hutchison and Kieber hydrogenase was confirmed to be present in the apoplast of 2002, Schmülling 2002). cells. Co-staining of activity for laccase, an enzyme Apoplastic cytokinins can also be transported into an adja- poised to function on the cell wall in the apoplast, in those cent cell by interaction with purine transporters (Bürkle et al. tissues suggests a possible cooperation of the enzymes in 2003); however, the fate of imported active cytokinins is uncer- cytokinin degradation. Additionally, the presence of pre- tain because no cytoplasmic receptors for the hormone are cursors for electron acceptors of cytokinin dehydrogenase was detected in phloem exudates collected from maize seed- known. It is possible that active transport into the cell is a lings, suggestive of an enzymatic capacity to control cytoki- mechanism to position cytokinins for further metabolism while nin flux through the vasculature. A putative metabolic preventing them from activating the signal transduction cas- connection between cytokinin degradation and conversion cade poised at the cell membrane. Cytoplasmic cytokinins can of plant phenolics by oxidases was proposed. be inactivated by glucosylation or metabolized by non-secreted cytokinin dehydrogenases (CKXs; historically categorized as Keywords: Cytokinin — Cytokinin dehydrogenase — Laccase cytokinin oxidases) (Mok and Mok 2001, Bilyeu et al. 2001). — Maize — Plant phenolics. Apoplastic concentrations of cytokinins reflect the balance of accumulation via local synthesis and transport, vs. cell uptake Abbreviations: BSA, bovine serum albumin; CKX, cytokinin or inactivation. Cytoplasmic concentrations of cytokinins dehydrogenase (EC 1.5.99.12); DCPIP, 2,6-dichlorophenol indo- depend on the rate of transport into the cell from the apoplast phenol; DMSO, dimethylsulfoxide; EST, expressed sequence tag; iP, vs. the rates of reversible and irreversible inactivation. Lack of 6 6 N -(2-isopentenyl)adenine; iPR, N -(2-isopentenyl)adenosine; PBS, clarity about the relative affinity of cytokinins for transporters, phosphate-buffered saline; Q , 2,3-dimethoxy-5-methyl-1,4-benzoqui- 0 receptors and metabolizing enzymes present in a tissue or cel- none; RT–PCR, reverse transcription–polymerase chain reaction; ZmCKX1, the CKX enzyme from maize, encoded by the gene Zmckx1 lular environment has hindered understanding of the mecha- (GenBank AF044603). nism of cytokinin signaling (Heyl and Schmülling 2003).

6 Corresponding author: E-mail, [email protected]; Fax, +420-585634933.

716 Localization of cytokinin dehydrogenase in maize 717

Over the past decade, many genes have been identified included members of the metabolic resource pool in tissues that affect cytokinin synthesis, transport, metabolism and their where the is expressed. function in growth regulation. In Arabidopsis, for instance, it The uniqueness or redundancy of CKX gene family mem- appears that a gene family of 20 or fewer members controls bers can be deciphered only after mechanistic details of their these processes (Heyl and Schmülling 2003, Kakimoto 2003, individual function have been characterized in native tissue and Popelková et al. 2004). The role of individual members of cellular environments. As a first step, we describe in this report these families is beginning to be uncovered, although compre- details of the spatial distribution of ZmCKX1 in various organ hensive understanding of genetic redundancy is lacking. tissues. We also describe ZmCKX1 activity as affected by Despite advances in gene discovery, mechanistic details of molecular components of the local metabolic resource pool that cytokinin metabolism and function are also lacking. We have can serve as electron acceptors directly or after oxidative modi- investigated one metabolic component, that of irreversible fication. cytokinin inactivation. This is an important process involved in maintenance of optimal hormone concentration. CKXs are the Results only known enzymes capable of irreversible degradation of cytokinins. They inactivate the hormone by removal of the iso- Localization of active cytokinin dehydrogenase in maize prenoid side chain of zeatin, N6-(2-isopentenyl)adenine (iP) Comparisons of CKX activity among plant tissues are and their ribosides to produce adenine or adenosine and the commonly made on the basis of relative activity in tissue corresponding aldehydes (Galuszka et al. 2000). Inactivation of homogenates. Recent studies have shown that a variety of cyto- cytokinins by CKX may be coordinated with specific plant plasmic and cell wall-associated constituents can serve as elec- developmental processes. In maize, for instance, zeatin riboside tron acceptors for ZmCKX1, but with variable efficiency levels increase transiently in developing kernels, but they (Galuszka et al. 2001, Frébortová et al. 2004). The distribution decline during kernel maturation in conjunction with up-regu- of potential electron acceptors may vary among plant tissues lation of CKX (Dietrich et al. 1995). and cell types. In tissue preparations, cellular structure is CKX was first isolated from kernels of maize (Zea mays) destroyed, and diverse CKXs are blended along with meta- and designated as ZmCKX1 (Houba-Herin et al. 1999, Morris bolic constituents that may normally be compartmentalized. et al. 1999). The respective gene Zmckx1 was subsequently Consequently, estimation and comparison of CKX activity in cloned. Since then, other members of the CKX gene family partially purified, homogenized tissues can be confounded by have been identified in developing maize kernels (Bilyeu et al. differences in CKX concentration and proportional influences 2003). It soon became clear that other plants contain small of members of the mixed electron acceptor pool. families of CKX genes (Bilyeu et al. 2001, Werner et al. 2001, To standardize comparisons of CKX activity, we devel- Bilyeu et al. 2003, Galuszka et al. 2004, Popelková et al. 2004). oped an activity assay based on our previous characterization In Arabidopsis and rice, complete genome sequence informa- of the enzymatic reaction with saturating levels of the artificial tion has enabled CKX genes to be categorized based on the electron acceptor 2,6-dichlorophenol indophenol (DCPIP) presence or absence of an N-terminal signal peptide-coding (Frébort et al. 2002). Providing saturating amounts of a suita- sequence that presumably specifies secretion of the enzymes to ble CKX electron acceptor eliminates the dependence of the apoplast (Schmülling et al. 2003). enzyme activity on possibly limited amounts of naturally Recent studies by Brugière et al. (2003) have established occurring electron acceptors in crude enzyme preparations. that Zmckx1 is expressed in the vascular bundles of kernels, However, the activity assay is not capable of discriminating seedling roots and coleoptiles. They also showed that Zmckx1 between separate CKX family members that can utilize DCPIP. expression is inducible by natural cytokinins. Zmckx1 is one of At saturating DCPIP concentrations (Frébort et al. 2002), at least five CKX genes known to occur in maize (Massonneau CKX activity was detectable in all tissues of developing ker- et al. 2004). Zmckx1 is an important member of this gene fam- nels (Table 1). CKX was active using either iP or N6-(2-isopen- ily because it is the predominant CKX family member present tenyl)adenosine (iPR) as , and at pHs of 6 and 8. Of in developing maize kernels. the kernel tissues examined, minimal CKX activity occurred in Expression studies have been important for establishing the pericarp, regardless of assay substrate or pH. Under most the locations of CKX transcription; however, details of protein conditions, maximum activity was detected in tissues of the distribution within metabolically active tissues are as yet un- aleurone layer distal to the embryo. Activities in this region defined. In addition, to understand the efficiency of CKX were, at times, >100-fold greater than in pericarp tissue. In con- in a particular cell type or tissue, it is important to estab- trast to reproductive tissues, CKX activity was very low in lish the mechanistic details of protein function. For instance, seedling roots for either substrate or pH. Frébortová et al. (2004) recently confirmed earlier reports The location of CKX activity was confirmed visually by (Galuszka et al. 2001, Laskey et al. 2003) that ZmCKX1 func- in situ assays in which a purple precipitate formed as the result tioned preferentially with electron acceptors other than . of electron transfer through a redox indicator dye combination The molecules that served as electron acceptors for ZmCKX1 that reflected dehydrogenase activity (Šebela et al. 2001). Dye 718 Localization of cytokinin dehydrogenase in maize

Table 1 Cytokinin dehydrogenase activity in reproductive and vegetative tissues of maize Maize tissue CKX activity per g FW (pkat g–1) e pH 8.0 pH 6.0 IP iPR iP iPR Kernel tissues Whole kernel a ,c 1,330 ± 75 762 ± 14 1,070 ± 48 909 ± 32 Pericarp a, d 175 ± 19 73 ± 6 95 ± 52 120 ± 12 Embryo a, d 920 ± 83 227 ± 52 424 ± 81 500 ± 37 Aleurone layer (proximal to embryo) a, d 8,450 ± 918 2,360 ± 457 3,370 ± 1,870 5,710 ± 381 Aleurone layer (distal to embryo) a, d 17,500 ± 2,870 8,120 ± 914 10,700 ± 2240 18,100 ± 3,050 Starchy endosperm a, d 1,380 ± 87 490 ± 72 466 ± 51 852 ± 109 Milky endosperm a, d 827 ± 81 306 ± 47 391 ± 95 794 ± 45 Other reproductive tissues Maternal tissue (pedicel) a, d 899 ± 112 509 ± 154 901 ± 85 348 ± 29 Maternal leaf a,d 191 ± 85 540 ± 237 364 ± 48 394 ± 158 Cob leaf a, c 66.1 ± 3.1 17.3 ± 0.6 17.9 ± 0.7 20.0 ± 1.1 Cob stalk a, c 21.3 ± 1.5 8.1 ± 0.3 6.7 ± 0.3 8.3 ± 0.3 Whole silk a, c 25.2 ± 1.8 6.6 ± 0.2 12.4 ± 0.7 8.7 ± 0.2 Vegetative tissues Seedling root apex (5-day-old) b ,c 54.9 ± 2.7 21.1 ± 0.8 43.5 ± 0.9 52.3 ± 2.1 Seedling root (10-day-old ) b,c 2.2 ± 0.1 1.3 ± 0.1 6.7 ± 0.2 5.7 ± 0.2 Seedling shoot apex (5-day-old) b, c 24.8 ± 0.9 11.1 ± 0.3 13.1 ± 0.4 12.3 ± 0.4 Seedling shoot (10-day-old ) b, c 10.1 ± 0.2 4.1 ± 0.1 6.0 ± 0.3 4.8 ± 0.2 a Tissues were collected from field-grown maize 3 weeks after pollination. b Vegetative tissues included growth chamber-grown seedlings. c Tissues were homogenized with liquid nitrogen in order to disrupt the cells (total activity measured). d Tissues from a single kernel were dissected and used for measuring activity without cell disruption (presumably apoplastic enzyme measured). e Activity was measured by the 4-aminophenol method (Frébort et al. 2002) with 0.15 mM of iP or iPR as substrates. Values show the means and SDs of six replicate activity measurements. precipitation in control tissues that were not treated with CKX was detected in the vasculature associated with maternal tissue substrate was considered to be background staining. The induc- (pedicel). tion of reaction precipitates is more appropriate for tissue stain- In developing seedlings, CKX activity staining was ing than reactions with DCPIP because that electron acceptor detected in phloem cells of the stem (Fig. 3), but roots exhib- bleaches with enzyme activity. ited only a minimal amount of diffuse staining (not shown). In both longitudinal and transverse sections of developing maize kernels (Fig. 1), intense staining, reflecting high levels Immunohistochemical detection of ZmCKX1 in maize kernels of CKX activity, was detected in the aleurone layer, embryo Immunohistochemical staining provided a means of and base of the silk (Fig. 2). In addition, redox indicator dye directly visualizing the location of ZmCKX1 within tissues

Fig. 1 Schematic anatomy of the maize kernel: (A) longitudinal and (B) transverse dissection. Localization of cytokinin dehydrogenase in maize 719

Fig. 2 Histochemical detection of cytokinin dehydrogenase activity in maize kernel. (A) Longitudinal and (B) transverse section of maize ker- nel (3 weeks after pollination) through the central part, (C) transverse section through maternal tissue, (D) aleurone layer, (E) close-up of the aleu- rone layer, (F) close-up of the embryo. Left and right panels show activity staining of the sections with iP substrate and control experiments without iP, respectively. The activity is expressed as purple precipitate of nitroblue tetrazolium formazan. en, endosperm; em, embryo; pc, peri- carp; al, aleurone layer; sb, silk base; vb, vascular bundles of maternal tissue; pe, pedicel. known to have enzymatic activity. For characterization of acceptors in plants (Galuszka et al. 2001, Frébortová et al. enzyme distribution, we used a polyclonal antibody raised 2004). A variety of phenolic compounds are present in plant against the ZmCKX1 peptide fragment, CF400 (Morris et al. tissues that could enter the quinone pool via modification by 1999, Bilyeu et al. 2003). In agreement with activity assessments, gold labeling of ZmCKX1 protein with CF400-directed antibody was obvious throughout the aleurone layer, embryo cells and pedicel of ker- nels (Fig. 4). Staining was also evident in kernel pericarp and seedling stem phloem cells that showed lower levels of total CKX activity (Table 1). Weak representation of the enzyme in phloem cells was consistent with low detectable CKX activity in vascular and other non-reproductive tissues. Further evaluation of ZmCKX1 distribution using confo- cal microscopy showed that cross-reacting antibody signals demarked the outline of the cells in aleurone layer, embryo and other kernel tissues (Fig. 5). The distribution of ZmCKX1 in all tissues was consistent with apoplastic expression of the enzyme as predicted by the presence of an N-terminal signal peptide (Morris et al. 1999). Fig. 3 Histochemical detection of cytokinin dehydrogenase activity Identification of ZmCKX1 electron acceptors produced by other in maize seedlings (2 weeks after germination). (A) Stem cross-sec- tion and (B) close-up of phloem cells. Left and right panels show enzymatic systems activity staining of the sections with iP substrate and control experi- Our previous experiments suggested that quinones are ments without iP, respectively. The activity is expressed as purple pre- likely to be important, naturally occurring ZmCKX1 electron cipitate of nitroblue tetrazolium formazan. xy, xylem; ph, phloem. 720 Localization of cytokinin dehydrogenase in maize

Fig. 4 Gold labeling of ZmCKX1 with specific antibody. (A–E) Maize kernel (3 weeks after pollination): (A) aleurone layer, (B) embryo, (C) pericarp, (D) seed coat, (E) maternal tissue. (F) Maize seedling (2 weeks after germination), stem phloem cells. Right and left panels show sam- ple staining and control experiments without the primary antibody, respectively. The location of ZmCKX1 is outlined by 10 nm gold particles. en, endosperm; em, embryo; pc, pericarp; al, aleurone layer; ma, maternal tissue; ph, phloem.

Fig. 5 Immunohistochemical detection of ZmCKX1 in maize kernel using confocal microscopy. (A) Aleurone layer, gold labeling of ZmCKX1 with specific antibody; (B) aleurone layer, control experiment without the antibody; (C) embryo close-up, gold labeling of ZmCKX1 with spe- cific antibody; (D), embryo, control experiment without the antibody. Right and left panels show images of the same section using confocal and inverted optical microscopy, respectively. The location of ZmCKX1 is outlined by 10 nm gold particles. em, embryo; pc, pericarp; al, aleurone layer. Localization of cytokinin dehydrogenase in maize 721

Table 2 Changes in turnover rates of ZmCKX1 in the presence 2). Kaempferol showed a lesser effect with tyrosinase than with of electron acceptors generated by other enzymatic systems a peroxidase, probably due to the fact that it can partly inhibit b c tyrosinase by chelating the copper (Kubo and Kinst- Source compound Relative turnover rate Hori 1999). Peroxidase + H2O2 Tyrosinase Rosmaric acid 0.0 0.0 Co-localization of laccase and ZmCKX1 expression Scopoletin 0.0 0.8 Among the plant phenolic compounds that could be con- Caffeic acid 0.0 0.8 verted to electron acceptors useable by ZmCKX1 was 4-meth- Daphnoretin 0.0 0.5 ylcatechol. This aromatic compound was of particular interest o-Coumaric acid 0.0 0.4 because it is oxidized by laccase, an enzyme that is secreted Luteolin 0.0 0.0 into the apoplast and is associated with the cell wall (Mayer Hydrocaffeic acid 1.1 2.0 and Staples 2002). Consequently, 4-methylcatechol and lac- Umbelliferone 0.5 0.6 case may be well positioned in plant tissues to interact locally p-Coumaric acid 1.2 3.0 with ZmCKX1 in the apoplast. Guaiacol 0.2 0.8 We confirmed the activity of laccase in kernel tissues by Acetosyringone 6.6 2.7 visualizing the distribution of orange-colored reaction prod- ′ Kaempferol 2.7 1.3 ucts generated by laccase oxidation of 3,3 -diaminobenzidine 4-Methylcatechol 1.0 9.7 (Binnington and Barrett 1988). In whole longitudinal and trans- Tyrosine 0.1 1.1 verse kernel sections, laccase activity was detectable in mater- Ferullic acid 0.0 0.0 nal tissues, as well as in the aleurone cell layer and embryo Indolyl-3-acetic acid 0.0 0.0 (Fig. 6). Laccase staining was also found in the stem phloem a cells of seedlings. These are the same tissues in which CKX 3-Methyl-2-butenal produced by oxidative cleavage of the side-chain activity and ZmCKX1 protein were detected in earlier experi- of iP was detected by the 4-aminophenol assay (Frébort et al. 2002) after 30 min incubation at 37°C. Reaction mixtures (0.6 ml) contained ments. 4.5 nM ZmCKX1, 0.15 mM iP and 0.1 mM of the studied reagents in Laccases are encoded by multiple genes in diverse plant 75 mM potassium phosphate buffer, pH 7.2, with 0.08 U ml–1 of species (Ranocha et al. 2002). We were able to confirm the –1 peroxidase and 0.1 mM hydrogen peroxide, and 17 U ml of tyrosi- expression of at least two putative maize laccase genes using nase, respectively. reverse transcription–polymerase chain reaction (RT–PCR). b The source compound serves as substrate of either peroxidase or tyro- sinase. Fragments of the two laccase genes, Lac1 and Lac2, were c The observed rates are expressed as the ratio to the rate with 0.1 mM amplified using primers derived from maize expressed Q0, under the same conditions without peroxidase/hydrogen peroxide sequence tag (EST) clones prepared from stressed roots or tyrosinase added. (GenBank AI855307) and seedling cDNA (GenBank BM078912), respectively. After PCR, expression of both Lac1 and Lac2 was detected in kernel tissues. In addition, Lac2 was oxidases. Consequently, we tested the potential for plant phe- expressed in seedling tissues including root, while Lac1 was nolic compounds to serve as electron acceptors for ZmCKX1 limited to stem and leaf tissues. after enzymatic oxidation. Of the five known maize CKX genes (Massonneau et al. For these in vitro experiments, we tested two commer- 2004), Zmckx1 was the predominant gene, being expressed in cially available enzymes, horseradish peroxidase and fungal developing kernels 3 weeks after pollination. During seedling tyrosinase, as reagents capable of generating quinone or radi- development, Zmckx1 expression was detected in young roots cal species (Doerge et al. 1997, Kubo and Kinst-Hori 1999). and the shoot apex, as well as the upper part of the seedling in Candidate phenolic compounds represent components of meta- later developmental stages. The expression patterns of CKX bolic pathways involved in normal housekeeping functions, genes were in agreement with the previous results (Bilyeu et al. including pigmentation, signaling, fertility and lignin forma- 2003, Massonneau et al. 2004). Expression of CKX and laccase tion (Lee et al. 1996, Winkel-Shirley 2001). genes in dissected maternal tissue from the base of maize ker- Many of the compounds tested did not contribute to nels (Fig. 7) further confirmed the results of previous experi- enhancement of ZmCKX1 activity in the presence of oxidases ments showing activity staining for both enzymes in this tissue. (Table 2). Other compounds, including hydrocaffeic acid, umbelliferone and p-coumaric acid, induced relative reactions Detection of electron acceptors for ZmCKX1 in vascular fluids rates that were similar to the reference quinone, Q0. In con- Although ZmCKX1 protein and enzymatic activity were trast, in the presence of tyrosinase, 4-methylcatechol increased detected predominantly in maize reproductive tissues, subsets ZmCKX1 reaction rates nearly 10-fold compared with Q0. In of cells in the phloem (probably phloem companion cells) also addition, acetosyringone increased ZmCKX1 relative reaction contain CKX activity (Fig. 3). Whether ZmCKX1 in vascular rates >6-fold in the presence of horseradish peroxidase (Table tissues is able to catalyze cytokinin degradation depends on 722 Localization of cytokinin dehydrogenase in maize

Fig. 6 Histochemical detection of lac- case activity in maize. The activity is expressed as an orange precipitate of 3,3′-diaminobenzidine oxidation prod- uct (A–F); experiments without the sub- strate are shown as negative control (G– I). (A) Longitudinal and (B) transverse section of maize kernel (3 weeks after pollination) through the central part; (C) transverse section through maternal tissue; (D) kernel aleurone layer close- up; (E) maternal tissue close-up; (F) maize seedling (2 weeks after germina- tion), close-up of phloem cells on stem cross-section; (G) transverse section of the maize kernel (control); (H) mater- nal tissue close-up (control); (I) kernel aleurone layer close-up (control). en, endosperm; em, embryo; pc, pericarp; al, aleurone layer; ma, maternal tissue; vb, vascular bundles of maternal tissue; pe, pedicel; ph, phloem; xy, xylem.

whether a suitable electron acceptor for activity is also present in those tissues. To address this possibility, we assessed ZmCKX1 activity using iP as a substrate in the presence of maize xylem or phloem exudates. Minimal ZmCKX1 activity was detected in control treat- ments that included the enzyme and substrate without vascular fluids or electron acceptor. We did, however, detect significant ZmCKX1 activity in the presence of xylem and phloem fluids, which suggested the occurrence of effective electron acceptors in vascular tissues (Table 3). ZmCKX1 activity in the presence of phloem fluids was about three times greater than in xylem fluids; however, the enzyme activity was still about 50% less than activity in the presence of the reference quinone, Q0. ZmCKX1 activity was enhanced further when the sam- ples contained tyrosinase as a quinone-generating agent. ZmCKX1 reaction rates in phloem and xylem exudates were increased about 3.5- and 1.3-fold over rates in fluids without tyrosinase, respectively. These additional increases in reaction rates suggest the presence of additional quinone or radical pre- cursors in these fluids. Values of pH in the xylem and phloem are around 6 and 8, Fig. 7 Expression of CKX and laccase in maize kernel and seed- respectively (Gerendas and Schurr 1999). As the above experi- lings. Co-expression of Zmckx1 and Lac1 genes in basal maternal tis- sue of immature maize kernels by RT–PCR using total RNA as ment was performed at the optimum for fungal tyrosinase, pH template. The respective cDNA gene amplification products were 230 7, different physiological pH values may affect the quinone for- and 222 for Zmckx1 and Lac1, respectively. The presence of contami- mation in vivo. nating DNA results in the amplification of intron sequences corre- sponding to the faint bands of 324 bp for Zmckx1 (intron 94 bp) and Discussion ~400 bp for Lac1 (the genomic sequence of the maize laccase gene is not known). mat, RNA isolated from maternal tissue; neg, control reaction with digested genomic DNA and non-reverse transcribed Zmckx1 is one of at least five CKX genes known to occur RNA; MW, DNA size markers of the indicated length. in maize. As an important member of this gene family, it is Localization of cytokinin dehydrogenase in maize 723

Table 3 Presence of a natural electron acceptor of CKX in maize vascular fluids Sample (electron acceptor) Treatment None Tyrosinase Activity (nkat ml–1) Fold Activity (nkat ml–1) Fold Xylem exudates (15 µl) 0.19 4.8 0.26 6.5 Phloem exudates (15 µl) 0.57 14.3 1.96 49.0 Control (no acceptor added) 0.04 1.0 0.04 1.0

Q0 (80 µM) 1.18 29.5 – – 3-Methyl-2-butenal produced by oxidative cleavage of the side chain of iP was detected by the 4-aminophe- nol assay (Frébort et al. 2002) after 30 min incubation at 37°C. Reaction mixtures (0.6 ml) contained the studied samples, 0.45 nM ZmCKX1, 0.15 mM iP and 500 U ml–1 fungal tyrosinase in 0.1 M potassium phos- phate buffer, pH 7. transcribed predominantly in developing maize kernels. Brugière et al. (2003) recently described significant Zmckx1 encodes a protein that is predicted to occur in the apo- Zmckx1 transcript accumulation in the vasculature of kernels plast of plant tissues where it probably plays a significant role and roots. Based on patterns of accumulation, they suggested in regulating cytokinin concentrations. that the dehydrogenase is transcriptionally controlled by local Evidence for protein secretion has been based on previ- concentrations of cytokinins. These authors did not character- ous Zmckx1 sequence analysis (Morris et al. 1999). In this ize the distribution of ZmCKX1 protein in vegetative or repro- report, we establish that the ZmCKX1 protein is indeed ductive tissues. However, our observation that ZmCKX1 was secreted into the apoplast of maize kernel tissues. Using immu- restricted to the apoplast of kernel tissues provides additional nochemical staining, we consistently detected ZmCKX1 in the evidence that the protein is ideally situated to regulate trans- apoplast in close proximity to the cell wall, regardless of tissue located cytokinins. type. The highest density of staining occurred in the aleurone A recent study in Arabidopsis (Takei et al. 2004) shows layer, embryo and pedicel tissues. This is also the first spatial that phloem companion cells also contain the cytokinin bio- characterization of a CKX family member in maize tissues. synthetic enzyme, isopentenyl . Our observations Although polyclonal antibody has been used, the localization suggest that cytokinins may also be degraded in phloem of should be relatively specific for ZmCKX1, since the maize seedlings. However, ZmCKX1 can efficiently degrade ZmCKX1-derived peptide CF400 used for antibody prepara- cytokinins in the apoplast only when a suitable electron accep- tion (Morris et al. 1999) shares no more than 42% amino acid tor is available to promote the two-substrate dehydrogenase sequence identity with other maize CKXs known to date. As reaction. We showed previously that oxygen is a poor electron the maize genome is not known, however, the immuno- acceptor for ZmCKX1 relative to other components of the localization might also detect unknown CKX if they plant metabolic pool, including, in particular, quinones (Frébor- were closely related to ZmCKX1. tová et al. 2004). The low efficiency of oxygen as an electron The patterns of ZmCKX1 protein antibody staining coin- acceptor was clarified by the recently determined three-dimen- cided closely with dehydrogenase activity detected by dis- sional structure of ZmCKX1 (Malito et al. 2004) that showed sected tissue, tissue homogenate or in situ dye precipitation assays. Both activity assays and immunohistochemistry experi- an absence of the tunnel suitable for molecular oxygen trans- ments revealed the presence of high levels of CKX in aleu- port to the active site. Similarly, there is no room for a quinone rone, embryo and maternal tissues. In tissue homogenate to bind in the proximity of the ZmCKX1 flavin . It is assays, compartmentalization of CKX family members, sub- possible that an electron acceptor such as quinone may bind to strates, pH and other factors influencing the catalytic activity is the protein surface and reoxidize the reduced cofactor via elec- destroyed in the blending process. Consequently, detectable tron transfer through the protein matrix. This hypothesis is in enzyme activity represents contributions of not only ZmCKX1 agreement with the wide specificity of the enzyme for structur- but also other CKX family members. ally different electron acceptors. Activity localizations based on visualization of dye pre- Quinones are known as widespread secondary metabo- cipitates produced by CKX activity proved useful in confirm- lites that function as signal molecules, particularly important in ing dissected tissue and tissue homogenate activity estimations. the exchange of chemical signals between plant roots, a phe- Dye precipitation methods proved particularly useful because nomenon classically termed allelopathy. The bioactivity of qui- the assay is conducted with intact tissues and there is less con- nones is due in large part to radical intermediates formed fusion of homogenizing metabolic components from differing during redox cycling between quinone and hydroquinone states cells or organelles. (Wrobel et al. 2002). 724 Localization of cytokinin dehydrogenase in maize

dation of the enzyme with a quinone electron acceptor pro- duced from a phenolic compound such as 4-methylcatechol by laccase enzymes at the site of cell wall biosynthesis (Fig. 8). A primary of enzymatic oxidation of 4-methylcatechol by apple polyphenol oxidase is 4-methyl-1,2-benzoquinone that is rapidly converted to 2-hydroxy-5-methyl-1,4- benzoquinone in acidic pH, whereas in alkaline pH it undergoes polymerization (Richard-Forget et al. 1992). Other quinone-generating sys- tems located in the apoplast such as widely occurring peroxi- dases (Tognolli et al. 2002) could also serve to generate electron acceptors for ZmCKX1. Our previous data showed that the phenolics themselves generally have no effect on CKX activity (Frébortová et al. 2004) unless they are converted by oxidases to reactive qui- none species that serve as electron acceptors in the dehydrogenase reaction of CKX (Table 2). It is therefore very Fig. 8 Proposed connection of cytokinin and metabolism of plant probable that the results of earlier studies with partially puri- phenolics in maize. fied CKX from bean (Armstrong 1994) and tobacco (Wang and Letham 1995) tissue cultures that showed activation of the enzyme with caffeic acid, acetosyringone and 2,6-dimethoxy- We have hypothesized that quinones can be generated phenol were affected by the presence of polyphenol oxidase or from phenolic precursors by modification of appropriate oxi- a similar enzyme in the preparations. dases such as peroxidase (EC 1.11.1.7) and polyphenol oxi- Some indices from the literature for the connection dases, including catechol oxidase (EC 1.10.3.1), laccase (EC between cytokinin metabolism and plant phenolics come from 1.10.3.2) and tyrosinase (EC 1.14.18.1). Natural substrates of the observation that the expression of cytokinin biosynthesis in these enzymes include compounds such as ferulic acid and Agrobacterium tumefaciens is induced by plant phenolics auxins, catechols, flavonoids and other phenolic compounds (Powel et al. 1988). Also a recent proteomic study on cell wall that are known to occur in maize (Galletti et al. 1996). Peroxi- proteins describing an up-regulation of peroxidase and laccase dases, most of which are heme-dependent enzymes, consume in tobacco with highly increased endogenous cytokinin content hydrogen peroxide and participate in many physiological proc- after transformation with Tcyt of Agrobacterium tumefacians esses such as lignification, wound and stress responses, and (Blee et al. 2001) should be noted. auxin metabolism (Tognolli et al. 2002). Polyphenol oxidases The results presented in this work reveal a fundamental all contain copper bound by histidyl ligands in the active site connection of diverse metabolic processes in subcellular com- and differ only marginally in substrate specificity. Catechol partments of plant cells. Since ZmCKX1 is a member of a oxidase and tyrosinase act mostly on monophenols or o-diphe- broad family of oxido-reductase enzymes capable of using oxy- nols that are present in the cytosol or chloroplasts and are gen as an electron acceptor, it remains to be explored how the responsible for darkening of fruits after peeling (Rompel et al. secreted CKX enzymes evolved preferentially to utilize elec- 1999, Gowda and Paul 2002). Laccase acts preferably on p- tron acceptors generated from other enzymatic systems. In this diphenols and is mainly a cell wall or xylem enzyme that par- context, it may be particularly revealing to compare the reac- ticipates in lignification (Mayer and Staples 2002). tion mechanism for non-secreted CKXs with ZmCKX1 Of these enzymes, we focused on laccase as a potentially (Schmülling et al. 2003, Popelková et al. 2004). significant oxidase because of its occurrence in the apoplast in association with the cell wall, a distribution pattern that Materials and Methods matches that of ZmCKX1. Our results confirmed the potential for laccase to generate effective electron acceptors. In particu- Plant material Three weeks after pollination, immature kernels and cob parts lar, our model enzyme for laccase, fungal tyrosinase, oxidized were harvested from a field-grown (Columbia, Missouri, 2003) hybrid 4-methylcatechol, acetosyringone and p-coumaric acid which maize line (Asgrow RX730 RRYGAF) Developing ears were brought accelerated ZmCKX1 reaction rates 3- to 9-fold relative to Q0 to the laboratory where the aleurone layer, embryo, endosperm, (Table 2). pericarp and maternal tissues were dissected manually from the ker- The evidence for apoplastic localization of ZmCKX1 and nels (Fig. 1). Dissected tissues were used immediately for assessment of CKX activity and protein localization studies. overlapping expression of the polyphenol oxidase enzyme lac- Maize seedlings from the same hybrid line were also produced case in kernel tissues has led us to propose a mechanistic model for assessment of enzyme activity. Seedlings were grown for 14 d in a for in planta ZmCKX1 cytokinin that includes reoxi- controlled-environment growth chamber I24L (Conviron, Winnipeg, Localization of cytokinin dehydrogenase in maize 725

Canada) with a 14 h day length (220 µmol photons m–2 s–1), 27/20°C activity, protein content was assayed according to Bradford (1976) day/night temperatures and 60% relative humidity. Additional maize with bovine serum albumin (BSA) as standard. Each tissue-specific seedlings grown for the histochemical staining were maintained in the reaction assay was replicated six times. dark (etiolated) at 24°C to prevent chlorophyll formation. Histochemical localization of enzymatic activities Enzymes and chemicals Seven-day-old etiolated maize seedlings were cut into 2 cm seg- The recombinant enzyme ZmCKX1, produced by a Pichia pas- ments. Each segment was sandwiched between two pieces of a longi- toris expression system, was purified as described previously (Bilyeu tudinally cut Sambucus nigra pith and sliced into 20–40 µm cross- et al. 2001) and stored as a concentrated stock solution (87.7 µM) in sections by a clamp-on hand, sliding microtome (AllMikro; Braunsch- TE buffer, pH 8, containing 1 M KCl, at 4°C. The enzymes per- weig, Germany). Sectioned samples were then immersed in 0.2 M oxidase (EC 1.11.1.7) from horseradish (100 U mg–1), tyrosinase (EC Tris–HCl, pH 8.0 (Ruzin 1999). –1 1.14.18.1), from mushroom (2,000 U mg ) and bovine liver catalase Whole maize kernels were first embedded in 4% agar as –1 (2,000–5,000 Bergmeyer units mg protein ) were obtained from described by Laurenzi et al. (1999) and then sliced into 50–200 µm Sigma (St Louis, MO, USA). cross-sections using a Lancer Vibratome 1000 (The Vibratome Com- We evaluated diverse compounds as candidate substrates of per- pany, St Louis, MO, USA). oxidase and tyrosinase. These included, rosmaric acid, scopoletin, caf- Seedling and kernel sections were immediately stained to detect feic acid, daphnoretin, o- and p-coumaric acid, luteolin, hydrocaffeic CKX activity using a protocol adapted from Šebela et al. (2001). In the acid and umbelliferone (donated by Dr. Lubomír Opletal, Charles Uni- original protocol, aminoaldehyde dehydrogenase activity was detected versity, Hradec Králové, Czech Republic). Indole-3-acetic acid was by the conversion of substrate to a purple nitroblue tetrazolium-forma- from LOBA Feinchemie (Fischamend, Austria). Acetosyringone (3,5- zan. In our experiments, we immersed fresh tissue sections in a stain- dimethoxy-4-hydroxyacetophenon) and 3,3′-diaminobenzidine were ing solution containing 5 mM iP (substrate), 0.15 mM phenazine from Aldrich (Millwaukee, WI, USA) and ferulic acid was from methosulfate and 0.75 mM nitroblue tetrazolium in 0.2 M Tris–HCl Theodor Schuchardt company (Munich, Germany). DCPIP, iP, iPR, N- buffer, pH 8.0. Tissues were incubated in the dark for 0.5–24 h at 37°C methyldibenzopyrazine methyl sulfate (phenazine methosulfate), 3,3′- and examined for purple product formation. To prevent precipitation of (3,3′-dimethoxy-4,4′-biphenylene)-bis[2-(4-nitrophenyl-5-phenyl)-2H- reagents during tissue incubation, the sections were washed with tetrazolium chloride] (nitroblue tetrazolium), 4-aminophenol, 0.2 M Tris–HCl buffer, pH 8.0, and transferred to a fresh staining solu- coenzyme Q0 (2,3-dimethoxy-5-methyl-1,4-benzoquinone), kaemp- tion after every 6 h period. A control treatment was included in which ferol [3,5,7-trihydroxy-2-(4-hydroxyphenyl)-4H-1-benzopyran-4-one], tissues were immersed in the staining solution without substrate. 4-methylcatechol, guaiacol (2-methoxyphenol), tyrosine and other Kernel and stem tissues were also stained to detect catechol chemicals were from Sigma (St Louis, MO, USA). oxidase (tyrosinase)/laccase activity using a protocol adapted from Binnington and Barrett (1988). The modification included addition of Measurement of CKX activity catalase to the reaction mixture to decompose endogenously formed CKX activity was measured using the method of Frébort et al. hydrogen peroxide thus avoiding co-staining of peroxidase activity. (2002). The reaction mixture (total volume of 0.6 ml in an Eppendorf Staining solution contained 0.05% 3,3′-diaminobenzidine and 0.1% tube) containing the sample, 0.5 mM DCPIP, and either 0.15 mM iP or catalase in 0.1 M potassium phosphate, pH 7.0. The sections were iPR as substrate for the enzymatic reaction, in 75 mM Tris–HCl buffer incubated for 30–90 min at 37°C, in the dark, and then examined for (pH 8) or 75 mM imidazole–HCl (pH 6), was incubated for 0.5–2 h at orange product formation. Control experiments were carried out with 37°C. The reaction was halted by the addition of 0.3 ml of 40% (w/v) the same staining solution without 3,3′-diaminobenzidine. trichloroacetic acid. The reaction mixture was then centrifuged at After incubation, the sections were washed with reaction buffers 12,000×g for 5 min, and 0.2 ml of 4-aminophenol [2% (w/v) solution as above without staining reagents or substrate, and placed in a drop of in 6% (w/v) trichloroacetic acid] was added to the supernatant. CKX 50% glycerol (w/v) on a glass mount. Microscopy was carried out activity was determined spectrophotometrically (photodiode array using a Model BX50 light system microscope (Olympus Optical Co., spectrophotometer 8453 with UV-Visible Chem Station software Tokyo, Japan) equipped with a CoolSNAP Camera System (Roper A.05.02; Hewlett-Packard, Foster City, CA, USA, and DU 7500; Scientific, Tucson, AR, USA) connected to a PC or Diaplan micro- Beckman, Fullerton, CA, USA) from the amount of the cytokinin- scope (Leitz, Wetzlar, Germany) equipped with a CoolPix E4500 dig- derived aldehyde product Schiff base with 4-aminophenol at 352 nm ital camera (Nikon, Tokyo, Japan). Whole-kernel sections were viewed –1 –1 (ε352 = 15.2 mM cm ). For background elimination, the whole spec- through a dissecting scope Stemi SV 8 (Zeiss, Oberkochen, Germany) trum in the range 300–700 nm was scanned against a blank without the and documented by a Camedia digital camera (Olympus, Tokyo, substrate. Japan). Image analysis was performed using the CoolSNAP software Samples for total CKX activity estimates (homogenized) were (Roper Scientific). Due to light sensitivity of the staining reagents prepared as follows. Maize kernel and seedling for measurement of used, it was important to avoid prolonged light exposure of the viewed CKX activity were cut into small pieces, immersed in liquid nitrogen, samples. and pulverized using a hand mortar. Pulverized tissue was extracted with a 3- to 9-fold excess (v/w) of 0.1 M Tris–HCl buffer, pH 8.0, con- Immunohistochemical localization of CKX protein taining 1 mM phenylmethylsulfonyl fluoride and 1% Triton X-100. Root and shoot segments were sampled from 7-day-old etiolated Cell debris was removed by centrifugation at 12,000×g for 10 min and maize seedlings for protein localization studies. Immature kernels the supernatant aliquots were used for activity assays. were also sampled at 3 weeks after pollination. Kernel tissues were CKX activity was also measured in dissected kernel tissues with- prepared for immunohistochemical localization by first trimming their out homogenization (presumably the apoplastic enzyme activity). Ker- sides parallel to the embryonic axis to facilitate penetration of the fixa- nel tissues of interest were dissected by scalpel and forceps and added tive. Trimmed kernels were vacuum infiltrated at room temperature for immediately, without homogenization, to a reaction mixture. 2 h in 0.1 M sodium phosphate buffer (PBS), pH 7, containing 4% (w/ To standardize activity estimates among tissues, activities of v) paraformaldehyde and then incubated overnight at 4°C in the same CKX were expressed per g of fresh weight. For calculating specific medium at normal pressure. After that, the fixative solution was 726 Localization of cytokinin dehydrogenase in maize removed, and the segments were washed with 0.1 M PBS (pH 7). Tis- pared with the previous experiments with phenolic compounds, in order sues were then embedded in 4% agar and sectioned into 30–150 µm to overcome any inhibition effect of EDTA used for the exudation. cross-sections using a Lancer Vibratome 1000. Before staining, embedded tissue sections were blocked by mild RT–PCR detection of Zmckx and laccase homologs shaking with 3% (w/v) BSA and 0.1% Tween-20 in PBS buffer for 2 h. Total RNA was extracted from different plant tissues using a After blocking, the sections were incubated overnight with purified Plant RNA Purification Reagent (Invitrogen, Carlsbad, CA, USA). + goat polyclonal antibody raised against the ZmCKX1 peptide, CF400 Poly(A) RNA was purified from the total amount of RNA using an (Morris et al. 1999), diluted 1 : 400 in PBS buffer containing 1% BSA Oligotex Suspension (Qiagen, Hilden, Germany). The first-strand + and 1% heat-inactivated normal rabbit serum (Sigma). Unbound cDNA was reverse transcribed from 0.5–1.0 µg of poly(A) RNA primary antibody was removed by four 10 min washes with PBS using a Reverse Transcriptase RAV-2 (TaKaRa Shuzo, Shiga, Japan) containing 1% BSA and 0.1% Tween-20. Bound antibody was de- and oligo-dT(20) primer (Promega, Madison, WI, USA) according to tected by incubating tissue with rabbit anti-goat IgG gold conjugate, the manufacturer’s recommendation. 10 nm gold particle size (Sigma), diluted 1 : 40 in PBS buffer contain- Genomic DNA was isolated from 7-day-old maize leaves by an ing 1% BSA and 1% heat-inactivated normal rabbit serum. After 1.5 h improved phenol/chloroform method (Peterson et al. 1997). Aliquots incubation, tissue sections were washed in PBS and in several changes of 10 µg of the isolated DNA were digested by selected restriction of distilled water. Control treatments included tissues that were proc- endonucleases: EcoRI, EcoRV and XhoI (New England BioLabs, essed through all stages of staining, but in the absence of the primary Beverly, MA, USA). antibody. Detection of secondary antibody staining was enhanced by Primers for amplification of Zmckx2, Zmckx3 and Zmckx4 gene use of the IntenSE™ M Silver Enhancement Kit (Amersham, Buck- fragments were designed as described earlier (Bilyeu et al. 2003). For inghamshire, UK). Zmckx1, the following primers that encompass the gene third intron ′ ′ ′ Antibody-treated tissues were examined by both light and scan- were used: zmckx1f 5 -atcctgcagggcaccgacat-3 and zmckx1r 5 - ′ ning laser confocal microscopy (Radiance 2000 system; BioRad, ccgcgccaggtaggtcttgt-3 . Primers for two laccase genes were derived Hercules, CA, USA, coupled to an IX70 inverted microscope; Olympus, from maize EST clones prepared from stressed root (Lac1, GeneBank accession number AI855307: lac1f 5′-ctacaacctggtggaccc-3′; lac1r 5′- Tokyo, Japan). gtcgcatactcccacccaat-3′) and seedling and silk (Lac2, BM078912: lac2f 5′-accaagctgtaccccgctca-3′; lac2r 5′-aactggatgaccgccca-3′) show- Generation of CKX electron acceptors from phenolics and other com- ing significant homology to laccase genes from other plants. pounds The PCR was carried out with denaturation at 94°C for 3 min Additional experiments were performed to assess the potential followed by 45 cycles of amplification. The usual cycle consisted of for enzymatic generation of CKX electron acceptors from phenolics melting at 94°C for 30 s, annealing at 53–60°C for 30 s and extension and other compounds (Table 2). Candidate source compounds (final at 72°C for 1 min. The PCR mixture was prepared in a total volume of concentration 0.1 mM) were dissolved in dimethylsulfoxide (DMSO) 50 µl using Taq polymerase (TaKaRa Shuzo) as recommended by the and added to the CKX assay mixture (0.6 ml, 0.45 nM ZmCKX1) with supplier. In case of amplification of Zmckx2 and 3 gene fragments, 0.15 mM iP, buffered with 75 mM potassium phosphate buffer, pH 7.2, standard buffer contained an addition of 10% DMSO. Ten percent and supplemented with either horseradish peroxidase (0.08 U ml–1) –1 aliquots of every reverse transcription reaction were used as tem- and 0.1 mM hydrogen peroxide, or fungal tyrosinase (17 U ml ). plates. For control reactions, aliquots (1 µg) of digested genomic DNA After 30 min incubation, CKX activities were assayed as above. and non-reverse transcribed RNA (0.1 µg) were used as templates. PCR products were resolved on 1.5% (w/v) agarose gels. Collection of xylem and phloem exudates For the collection of xylem exudates (Smith-Becker et al. 1998), Acknowledgments 2-week-old maize plants were cut with a razor blade at the stem base. Residual phloem exudates were blotted from the cut stump for 10 min, after which time the root pressure exudates started to accumulate on The authors wish to thank Carole Gedenberg for technical assist- the surface and were collected by the tip of a micropipet within the ance in the early phase of the CKX staining experiments, Ivo Fry- next 2 h. drych for some CKX activity assays, and Kimberly Crouch for laboratory help. Technical help with tissue sectioning by vibratome An EDTA exudation method (King and Zeevaart 1974) was used and operating the confocal microscope by Mayandi Sivaguru from the to collect phloem exudates from maize leaves. Five segments of 1 cm Molecular Cytology Core Facility, University of Missouri, is grate- were excised from the base of leaves at the fully expanded stage and fully acknowledged. This work was supported by the following grants: immediately immersed in 0.15 ml of 5 mM Na2EDTA in 2 mM sodium IBN 0210160 from NSF USA, KONTAKT ME515 from the Ministry phosphate, pH 7 (adjusted with NaOH). Segments were then incu- of Education, Youth and Physical Education, Czech Republic, and the ° bated overnight under high humidity in a light-proof box at 25 C. To grants 522/03/0979 and 204/03/P103 from the Grant Agency, Czech estimate the volume of phloem exudates obtained, a control experi- Republic. ment was done according to Cervelli et al. (2001). Leaf segments of the same fresh weight were cut and soaked in 0.1 M sodium phos- phate buffer, pH 7, under vacuum for 30 min. Leaves were then dried References on a filter paper and immediately spun down for 10 min at 12,000×g in a dry Eppendorf tube. The volume of the obtained supernatant was Armstrong, D.J. (1994) Cytokinin oxidase and the regulation of cytokinin degra- measured and taken as the amount of phloem fluids. dation. In Cytokinins: Chemistry, Activity and Function. Edited by Mok, D.W.S. and Mok, M.C. pp. 139–154, CRC Press, Boca Raton, FL. Aliquots of exudates were added to the CKX reaction mixture –1 Auer, C.A. (1997) Cytokinin conjugation: recent advances and patterns in plant (0.6 ml; 0.45 nM ZmCKX1, 0.15 mM iP and 500 U ml fungal tyrosi- evolution. Plant Growth Regul. 23: 17–32. nase in 0.1 M potassium phosphate buffer, pH 7) to substitute for arti- Bilyeu, K.D., Cole, J.L., Laskey, J.L., Riekhof, W.R., Esparza, T.J., Kramer, ficial electron acceptor, and the effect on CKX activity was studied. M.D. and Morris, R.O. (2001) Molecular and biochemical characterization of The reaction mixture contained a 30-fold excess of tyrosinase com- a cytokinin oxidase from maize. Plant Physiol. 125: 378–386. Localization of cytokinin dehydrogenase in maize 727

Bilyeu, K.D., Laskey, J.G. and Morris, R.O. (2003) Dynamics of expression and Kakimoto, T. (2003) Biosynthesis of cytokinins. J. Plant Res. 116: 233–239. distribution of cytokinin oxidase/dehydrogenase in developing maize ker- King, R.W. and Zeevaart, J.A.D. (1974) Enhancement of phloem exudation nels. Plant Growth Regul. 39: 195–203. from cut petioles by chelating agents. Plant Physiol. 56: 96–103. Binnington, K.C. and Barrett, F.M. (1988) Ultrastructural localization of phe- Kubo, I. and Kinst-Hori, I. (1999) Flavonols from saffron flower: tyrosinase noloxidases in cuticle and haemopoietic tissue of the blowfly Lucilia cup- inhibitory activity and inhibition mechanism. J. Agric. Food Chem. 47: 4121– rina. Tissue Cell 20: 405–419. 4125. Blee, K.A., Wheatley, E.R., Bonham, V.A., Mitchell, G.P., Robertson, D., Laskey, J.G., Patterson, P., Bilyeu, K.D. and Morris, R.O. (2003) Rate enhance- Slabas, A.R., Burrell, M.M., Wojtaszek, P. and Bolwell, G.P. (2001) ment of cytokinin oxidase/dehydrogenase using 2, 6-dichloroindophenol as Proteomic analysis reveals a novel set of cell wall proteins in a transformed an electron acceptor. Plant Growth Regul. 40: 189–196. tobacco cell culture that synthesizes secondary walls as determined by bio- Laurenzi, M., Rea, G., Federico, R., Tavladoraki, P. and Angelini, R. (1999) De- chemical and morphological parameters. Planta 212: 404–415. etiolation causes a phytochrome-mediated increase of Bradford, M.M. (1976) A rapid and sensitive method for the quantitation of expression in outer tissues of the maize mesocotyl: a role in the photomodula- microgram quantities of protein utilizing the principle of protein–dye binding. tion of growth and cell wall differentiation. Planta 208: 146–154. Anal. Biochem. 72: 248–254. Lee, Y.W., Jin, S., Sim, W.S. and Nester, E.W. (1996) The sensing of plant sig- Brugière, N., Jiao, S., Hantke, S., Zinselmeier, C. and Roessler, J.A. (2003) nal molecules by Agrobacterium: genetic evidence for direct recognition of Cytokinin oxidase gene expression in maize is localized to the vasculature, phenolic inducers by the VirA protein. Gene 179: 83–88. and is induced by cytokinins, abscisic acid, and abiotic stress. Plant Physiol. Malito, E., Coda, A., Bilyeu, K.D., Fraaije, M.W. and Mattevi, A. (2004) Struc- 132: 1–13. tures of Michaelis and product complexes of plant cytokinin dehydrogenase: Bürkle, L., Cedzich, A., Döpke, C., Stransky, H., Okumoto, S., Gilissen, B., implications for flavoenzyme catalysis. J. Mol. Biol. 27: 1237–1249. Kühn, C. and Frommer, W.B. (2003) Transport of cytokinins mediated by Massonneau, A., Houba-Herin, N., Pethe, C., Madzak, C., Falque, M., Mercy, purine transporters of the PUP family expressed in phloem, hydathodes, and M., Kopecný, D., Majira, A., Rogowsky, P. and Laloue, M. (2004) Maize pollen of Arabidopsis. Plant J. 34: 13–26. cytokinin oxidase genes: differential expression and cloning of two new Cervelli, M., Cona, A., Angelini, R., Polticelli, F., Federico, R. and Mariottini, P. cDNAs. J. Exp. Bot. 55: 2549–2557. (2001) A barley polyamine oxidase isoform with distinct structural features Mayer, A.M. and Staples, R.C. (2002) Laccase: new functions for an old and subcellular localization. Eur. J. Biochem. 268: 3816–3830. enzyme. Phytochemistry 60: 551–565. Corbesier, L., Prinsen, E., Jacqmard, A., Lejeune, P., Van Onckelen, H., Mok, D.W. and Mok, M.C. (2001) Cytokinin metabolism and action. Annu. Rev. Périleux, C. and Bernier, G. (2003) Cytokinin levels in leaves, leaf exudate Plant Physiol. Plant Mol. Biol. 52: 89–118. and shoot apical meristem of during floral transition. Morris, R.O., Bilyeu, K.D., Laskey, J.G. and Cheikh, N.N. (1999) Isolation of a J. Exp. Bot. 54: 2511–2517. gene encoding a glycosylated cytokinin oxidase from maize. Biochem. Dietrich, J.T., Kamínek, M., Blevins, D.G., Reinbott, T.M. and Morris, R.O. Biophys. Res. Commun. 255: 328–333. (1995) Changes in cytokinins and cytokinin oxidase activity in developing Peterson, D.G., Boehm, K.S. and Stack, S.M. (1997) Isolation of milligram maize kernels and the effects of exogenous cytokinin on kernel development. quantities of nuclear DNA from tomato (Lycopersicon esculentum), a plant Plant Physiol Biochem. 33: 327–336. containing high levels of polyphenolic compounds. Plant Mol. Biol. 15: 148– Doerge, D.R., Divi, R.L. and Churchwell, M.I. (1997) Identification of the 153. colored guaiacol oxidation product produced by peroxidases. Anal Biochem. Popelková, H., Galuszka, P., Frébortová, J., Bilyeu, K.D. and Frébort, I. (2004) 250: 10–17. Cytokinin dehydrogenase: characterization and structure homology modeling Frébort, I., Šebela, M., Galuszka, P., Werner, T., Schmülling, T. and Pec, P. of the flavoprotein catabolizing plant hormones cytokinins. In Recent (2002) Cytokinin oxidase/dehydrogenase assay: optimized procedures and Research Developments in Proteins, Vol. 2. Edited by Pandalai, S.G. pp. 63– applications. Anal. Biochem. 306: 1–7. 81.Transworld Research Network, Kerala, India. Frébortová, J., Fraaije, M.W., Galuszka, P., Šebela, M., Pec, P., Hrbác, J., Powel, G.K., Hommes, N.G., Kuo, J., Castle, L.A. and Morris, R.O. (1988) Bilyeu, K.D., English, J.T. and Frébort, I. (2004) Catalytic reaction of cytoki- Inducible expression of cytokinin biosynthesis in Agrobacterium tumefaciens nin dehydrogenase: preference for quinones as electron acceptors. Biochem. by plant phenolics. Mol. Plant Microbe Interact. 1: 235–242. J. 380: 121–130. Ranocha, P., Chabannes, M., Chamayou, S., Danoun, S., Jauneau, A., Boudet, Galletti, G.C., Bocchini, P., Smacchia, A. and Reeves, J.B., III (1996) Monitor- A.M. and Goffner, D. (2002) Laccase down-regulation causes alterations in ing phenolic composition of maturing maize stover by high performance liq- phenolic metabolism and cell wall structure in poplar. Plant Physiol 129: uid chromatography and pyrolysis/gas chromatography/mass spectrometry. J. 145–155. Sci. Food Agric. 71: 1–9. Richard-Forget, F.C., Rouet-Mayer, M.A., Goupy, P.M., Philippon, J. and Galuszka, P., Frébort, I., Šebela, M. and Pec, P (2000) Degradation of cytokinins Nicolas, J.J. (1992) Oxidation of chlorogenic acid, catechins, and 4-methyl- by cytokinin oxidases in plants. Plant Growth Regul. 32: 315–327. catechol in model solutions by apple polyphenol oxidase. J. Agric. Food Galuszka, P., Frébort, I., Šebela, M., Sauer, P., Jacobsen, S. and Pec, P. (2001) Chem. 40: 2114–2122. Cytokinin oxidase or dehydrogenase? Mechanism of the cytokinin degrada- Rompel, A., Fischer, H., Meiwes, D., Büldt-Karentzopoulos, K., Magrini, A., tion in plants. Eur. J. Biochem. 268: 450–461. Eicken, C., Gerdemann, C. and Krebs, B. (1999) Substrate specificity of cate- Galuszka, P., Frébortová, J., Werner, T., Yamada, M., Strnad, M., Schmülling, T. chol oxidase from Lycopus europaeus and characterization of the bioprod- and Frébort, I. (2004) Cytokinin oxidase/dehydrogenase genes in barley and ucts of enzymatic caffeic acid oxidation. FEBS Lett. 445: 103–110. wheat: cloning and heterologous expression. Eur. J. Biochem. 271: 3990– Ruzin, S.E. (1999) Plant Microtechniques and Microscopy. Oxford University 4002. Press, New York. Gerendas, J. and Schurr, U. (1999) Physicochemical aspects of ion relations and Schmülling, T. (2002) New insights into the functions of cytokinins in plant pH regulation in plants—a quantitative approach. J. Exp. Bot. 50: 1101–1114. development. J. Plant Growth Regul. 21: 40–49. Gowda, L.R. and Paul, B. (2002) Diphenol activation of the monophenolase and Schmülling, T., Werner, T., Riefler, M., Krupková, E. and Bartrina y Manns, I. diphenolase activities of field bean (Dolichos lablab) polyphenol oxidase. J. (2003) Structure and function of cytokinin oxidase/dehydrogenase genes of Agric. Food Chem. 50: 1608–1616. maize, rice, Arabidopsis and other species. J. Plant Res. 116: 241–252. Haberer, G. and Kieber, J.J. (2002) Cytokinins: new insights into a classic Šebela, M., Luhová, L., Brauner, F., Galuszka, P., Radová, A. and Pec, P. (2001) phytohormone. Plant Physiol. 128: 354–362. Light microscopic localisation of aminoaldehyde dehydrogenase activity in Heyl, A. and Schmülling, T. (2003) Cytokinin signal perception and trans- plant tissues using nitroblue tetrazolium-based staining method. Plant duction. Curr. Opin. Plant Biol. 6: 480–488. Physiol. Biochem. 39: 831–839. Houba-Herin, N., Pethe, C., d’Alayer, J. and Laloue, M. (1999) Cytokinin oxid- Smith-Becker, J., Marois, E., Huguet, E., Midland, S.L., Sims, J.J. and Keen, ase from Zea mays: purification, cDNA cloning and expression in moss N.T. (1998) Accumulation of salicylic acid and 4-hydroxybenzoic acid in protoplasts. Plant J. 17: 615–626. phloem fluids of cucumber during systemic acquired resistance is preceded Hutchison, C.E. and Kieber, J.J. (2002) Cytokinin signaling in Arabidopsis. by a transient increase in phenylalanine ammonia- activity in petioles Plant Cell 14: s47–s59. and stems. Plant Physiol. 116: 231–238. 728 Localization of cytokinin dehydrogenase in maize

Takei, K., Ueda, N., Aoki, K., Kuromori, T., Hirayama, T., Shinozaki, K., Winkel-Shirley, B. (2001) Flavonoid biosynthesis. A colorful model for genetics, Yamaya, T. and Sakakibara, H. (2004) AtIPT3 is a key determinant of nitrate- biochemistry, cell biology, and biotechnology. Plant Physiol. 126: 485–493. dependent cytokinin biosynthesis in Arabidopsis. Plant Cell Physiol. 45: Wrobel, R.L., Matvienko, M. and Yoder, J.I. (2002) Heterologous expression 1053–1062. and biochemical characterization of an NAD(P)H:quinone Tognolli, M., Penel, C., Greppin, H. and Simon, P. (2002) Analysis and expres- from the hemiparasitic plant Triphysaria versicolor. Plant Physiol. Biochem. sion of the class III peroxidase large gene family in Arabidopsis thaliana. 40: 265–272. Gene 288: 129–138. Wang, J. and Letham, D.S. (1995) Cytokinin oxidase—purification by affinity Yamada, H., Suzuki, T., Terada, K., Takei, K., Ishikawa, K., Miwa, K., chromatography and activation by caffeic acid. Plant Sci. 112: 161–166. Yamashino, T. and Mizuno, T. (2001) The Arabidopsis AHK4 histidine Werner, T., Motyka, V., Strnad, M. and Schmülling, T. (2001) Regulation of kinase is a cytokinin-binding receptor that transduces cytokinin signals across plant growth by cytokinin. Proc. Natl Acad. Sci. USA 98: 10487–10492. the membrane. Plant Cell Physiol. 42: 1017–1023.

(Received December 14, 2004; Accepted February 14, 2005)