Biochemical characterization of the maize cytokinin dehydrogenase family and cytokinin profiling in developing maize plantlets in relation to the expression of cytokinin dehydrogenase genes

Plant Physiology and Biochemistry 74 (2014) 283e293

Contents lists available at ScienceDirect

Plant Physiology and Biochemistry journal homepage: www.elsevier.com/locate/plaphy

Research article

Biochemical characterization of the maize cytokinin dehydrogenase family and cytokinin profiling in developing maize plantlets in relation to the expression of cytokinin dehydrogenase genes David Zalabák a, Petr Galuszka a, Katarina Mrízová a, Kate
rina Podle
sáková b, Riliang Gu c, Jitka Frébortová b, * a Centre of the Region Haná for Biotechnological and Agricultural Research, Department of Molecular Biology,
Slechtitel u 11, Olomouc 783 71, Czech Republic b Centre of the Region Haná for Biotechnological and Agricultural Research, Department of Chemical Biology and Genetics,
Slechtitel u 11, Olomouc 783 71, Czech Republic c Key Lab of Plant Nutrition, MOA, College of Resources and Environmental Science, China Agricultural University, 100193 Beijing, China

a r t i c l e i n f o

a b s t r a c t

Article history: Received 26 September 2013 Accepted 19 November 2013 Available online 28 November 2013

The cytokinin dehydrogenases (CKX; EC 1.5.99.12) are a protein family that maintains the endogenous levels of cytokinins in plants by catalyzing their oxidative degradation. The CKX family in maize (Zea mays L.) has thirteen members, only two of which - ZmCKX1 and ZmCKX10 - have previously been characterized in detail. In this study, nine further maize CKX isoforms were heterologously expressed in Escherichia coli, purified by affinity and ion-exchange chromatography and biochemically characterized. ZmCKX6 and ZmCKX9 could only be expressed successfully after the removal of putative sequencespecific vacuolar sorting signals (LLPT and LPTS, respectively), suggesting that these proteins are localized to the vacuole. Substrate specificity analyses revealed that the CKX isoforms can be grouped into two subfamilies: members of the first strongly prefer cytokinin free bases while members of the second degrade a broad range of substrates. The most active isoform was found to be ZmCKX1. One of the studied isoforms, ZmCKX6, seemed to encode a nonfunctional enzyme due to a mutation in a conserved HFG protein domain at the C-terminus. Site-directed mutagenesis experiments revealed that this domain is essential for CKX activity. The roles of the maize CKX enzymes in the development of maize seedlings during the two weeks immediately after radicle emergence were also investigated. It appears that ZmCKX1 is a key regulator of active cytokinin levels in developing maize roots. However, the expression of individual CKX isoforms in the shoots varied and none of them seemed to have strong effects on the cytokinin pool. Ó 2013 Elsevier Masson SAS. All rights reserved.

Keywords: Zea mays L. Cytokinin Cytokinin dehydrogenase Escherichia coli IMPACT expression system Substrate preference

1. Introduction

Abbreviations: AtCKX, cytokinin dehydrogenase from Arabidopsis thaliana; CKX, cytokinin dehydrogenase; cZ, cis-zeatin; cZ9G, cis-zeatin-N9-glucoside; cZRMP, ciszeatin riboside-50 -monophosphate; DCPIP, 2,6-dichlorophenolindophenol; DHZ, dihydrozeatin; DMAPP, dimethylallyl diphosphate; EDTA, ethylenediaminetetraacetic acid; iP, N6-(D2-isopentenyl)adenine; iP9G, N6-(D2-isopentenyl)adenine-N9glucoside; iPR, N6-(D2-isopentenyl)adenosine; iPRMP, N6-(D2-isopentenyl)adenosine-50 -monophosphate; IPT, isopentenyl transferase; tZ, trans-zeatin; tZ9G, transzeatin-N9-glucoside; tZR, trans-zeatin riboside; tZRMP, trans-zeatin riboside-50 monophosphate; ZmCKX, cytokinin dehydrogenase from Zea mays. * Corresponding author. Tel.: þ420 585634871; fax: þ420 585634870. E-mail addresses: [email protected] (D. Zalabák), [email protected] (P. Galuszka), [email protected] (K. Mrízová), [email protected] (K. Podle
sáková), [email protected] (R. Gu), [email protected] (J. Frébortová). 0981-9428/$ e see front matter Ó 2013 Elsevier Masson SAS. All rights reserved. http://dx.doi.org/10.1016/j.plaphy.2013.11.020

Cytokinins are ubiquitous plant hormones that govern a wide range of plant developmental and physiological processes. Cytokinin homeostasis in plant tissues and organs is crucial for normal growth and development, so cytokinin levels within cells, tissues, and organs must be finely controlled. One of the most important mechanisms by which this is achieved involves the irreversible degradation of active cytokinins by cytokinin dehydrogenases. Cytokinins are chemical compounds that derive from adenine. The adenine moiety is substituted with either isoprenoid or aromatic side chains at the N6- position to form so-called cytokinin free bases. In planta, cytokinins are synthesized from adenosine mono-, di- and triphosphates and dimethylallyl pyrophosphate (DMAPP), which functions as a side-chain precursor. This biosynthetic process is

284

D. Zalabák et al. / Plant Physiology and Biochemistry 74 (2014) 283e293

catalyzed by isopentenyl transferase (IPT; dimethylallyldiphosphate: AMP dimethylallyltransferase; EC 2.5.1.27) enzymes and yields isopentenyladenine nucleotides (Kakimoto, 2001; Takei et al., 2001), which can be further hydroxylated by cytokininspecific cytochrome P450 monooxygenase to form trans-zeatin nucleotides (Takei et al., 2004). Nucleotides of both isopentenyladenine and trans-zeatin are activated by cytokinin-specific phosphoribohydrolases to form free bases (Kurakawa et al., 2007), which can be further modified, mainly by glycosylation of the adenine moiety or the zeatin side chain forming N-glucosides and O-glucosides, respectively (reviewed by Frébortet al. 2011). Cytokinin dehydrogenases (CKX; N6-dimethylallyladenine: acceptor oxidoreductase; EC 1.5.99.12) are flavoproteins that catalyze the irreversible cleavage of the side chains from cytokinin molecules. The final products of these cytokinin degradation reactions are adenine derivatives and aldehydes formed from the side chains. It was originally assumed that these enzymes function exclusively as oxidases, with molecular oxygen serving as the final electron acceptor (Pa
ces et al., 1971). However, they were subsequently reclassified as dehydrogenases because it was found that they strongly prefer electron acceptors such as quinones rather than molecular oxygen (Galuszka et al., 2001; Frébortová et al., 2010). The cytokinin dehydrogenases are usually encoded by a small family of genes. The number of genes within these families varies from species to species: the Arabidopsis thaliana genome has seven, but monocot species tend to have more. For example, rice and Brachypodium have 11, and maize has 13 (Mameaux et al., 2012). Even though they catalyze similar reactions, different CKXs have different rates of reaction and substrate preferences (Galuszka et al., 2007; Kowalska et al., 2010). They also usually have different spatial and temporal expression patterns. That is to say, different CKX enzymes are expressed in different tissues and organs, and at different stages of development. It has been shown that the levels of CKX gene expression can change very rapidly and dramatically, especially in response to stress signals (Vyroubalová et al., 2009). Because each CKX enzyme has unique substrate preferences and reaction rates, localized changes in the expression of specific CKX genes can cause pronounced changes in the cytokinin distribution within individual tissues and organs. This in turn affects their growth and development. The different CKX enzymes are also targeted to different subcellular compartments. Of the seven Arabidopsis CKX isoforms, one is cytosolic, two are vacuolar, and the others are secreted into the extracellular environment. Moreover, CKX overexpression studies have suggested a close correlation between the subcellular localization of the overexpressed CKX enzyme and the severity of the resulting cytokinin deficiency phenotype (Werner et al., 2003). One maize CKX isoform, ZmCKX1, is known to be localized to the apoplast while another, ZmCKX10, is
cytosolic (Smehilová et al., 2009). While the localization of the remaining maize CKX isoforms is currently unknown, it can reasonably be expected to differ somewhat from that of their counterparts in Arabidopsis, especially given the recently reported differences in cytokinin metabolite levels between monocots and dicots (Gajdo
sová et al., 2011). In the work reported herein, we characterized the entire maize CKX gene family in detail in order to understand how each member contributes to the regulation of cytokinin homeostasis. Maize was selected because it is an important model species for monocots in general. This work builds on our previous investigations into its
CKX enzymes (Vyroubalová et al., 2009; Smehilová et al., 2009), which focused on the roles of cytokinin metabolism in abiotic stress responses and the subcellular localization of two maize CKX isoforms (ZmCKX1 and ZmCKX10), respectively.

2. Results and discussion 2.1. Heterologous expression and purification of ZmCKXs ZmCKX proteins were prepared by means of intracellular heterologous expression in Escherichia coli using the pTYB12 expression vector, which enables quick and easy affinity purification of the inserted protein because it fuses a self-cleavable chitin-binding intein tag to the protein’s N-terminus. The sole exception was the ZmCKX1 enzyme, which was prepared in a Pichia pastoris expression system (Bilyeu et al., 2001), as the recombinant enzyme obtained by expression of ZmCKX1 gene in yeast was used in most biochemical studies of ZmCKX1 performed up to date and thus served as a reference. As shown in Arabidopsis, the various CKX isoforms encoded in the genome of an individual species have distinct biochemical properties, are expressed in different tissues, and have different subcellular localizations (Galuszka et al., 2007; Werner et al., 2003). The localization of each isoform is determined by an N-terminal sequence motif. In previous experiments, the presence of these signal sequences was found to impair the heterologous expression of some Arabidopsis CKX isoforms in P. pastoris; this effect could be alleviated by using expression vectors containing cloned CKX genes in which the targeting sequences had been removed (Kowalska et al., 2010). These sequences can also cause undesirable interactions between heterologous proteins and the post-translational processing machinery of E. coli, resulting in the formation of nonfunctional heterologous proteins and/or their targeting to inclusion bodies. The predictive tool SignalP 4.0 (Petersen et al., 2011) was therefore used to identify sequences encoding potential Nterminal signal peptides in the ZmCKX proteins. SignalP 4.0 was unable to identify the signal peptide sequence of ZmCKX9 so a different predictive tool, iPSORT, was used in this case (Bannai et al., 2002). Putative N-terminal signal sequences were found in all of the studied sequences (namely ZmCKX1; 2; 3; 4a; 4b; 5; 6; 8; 9 and 12) with the exception of the cytosolic ZmCKX10 (Fig. 1). The signal sequences of enzymes ZmCKX1 to 6, ZmCKX9, ZmCKX11 and ZmCKX12 seem to direct their secretion to the apoplast whereas that of ZmCKX8 was predicted to be a mitochondrial targeting sequence. However, there is no precedent for CKX proteins being localized in the mitochondria, so it seems likely that this is a case of misidentification. Notably, two Arabidopsis CKX isoforms, AtCKX1 and AtCKX3, were initially predicted to be localized to the mitochondria (Schmülling et al., 2003). However, subsequent GFP tagging experiments demonstrated that they are actually localized to the vacuoles (Werner et al., 2003). The removal of the putative secretory signals strengthened the expression of some of the recombinant CKX proteins, thus all of them were cloned without them. For example, clarified extracts of bacteria expressing the entire ZmCKX12 ORF exhibited negligible cytokinin degradation activity. However, significant activity was detected in the equivalent extracts of bacteria expressing the same gene with the signal peptide removed. Therefore, the secretory signals were removed from all of the cloned CKX genes that were inserted into the expression vectors. There are several closely homologous CKX gene pairs in the maize genome that were formed during recent chromosome duplication events (Gaut, 2001). The enzymes ZmCKX7 and ZmCKX11 are very close paralogs of ZmCKX8 and ZmCKX12, respectively, having very high levels of amino acid sequence identity (96.7 and 91.5%, respectively), and were excluded from this study because they are very weakly expressed in maize, and their abundance was below the limit of detection for the method used in this work (Vyroubalová et al., 2009). All four of the recombinant proteins corresponding to two other closely homologous pairs

D. Zalabák et al. / Plant Physiology and Biochemistry 74 (2014) 283e293

285

Fig. 1. Identification of putative N-terminal signal sequences and vacuolar sorting signals in maize CKX proteins. An alignment of maize CKX protein sequences revealed the positions of the putative N-terminal signal sequences identified using the SignalP 4.0 Server and iPSORT Server online prediction tools. Full arrows indicate the ends of the putative signal peptides that were removed to form the recombinant proteins. Boxes indicate putative sequence-specific vacuolar sorting signals and open arrows indicate the sequences that were removed to form the recombinant ZmCKX6 and ZmCKX9 proteins.

(ZmCKX4a and ZmCKX4b, ZmCKX2 and ZmCKX3) were prepared because they are all expressed relatively strongly in different tissues. In total, eight different recombinant pTYB12 vectors that yielded detectable CKX activity in cell lysates were created. No CKX activity was detected in lysates derived from bacterial cultures expressing either the complete ORFs of ZmCKX6 and ZmCKX9 or the corresponding ORFs with the predicted signal peptides removed (Fig. 1). The bacterial lysate from cells transformed with an expression vector carrying the ZmCKX6 ORFs was subjected to a Western blot analysis using an anti-chitin binding domain antibody, which revealed that the targeted CKX protein was not being expressed. The ZmCKX6 gene with the putative N-terminal signal peptide removed was also subcloned into the pGAPZaA(His) 10 expression vector (Kowalska et al., 2010) and transformed into P. pastoris strain X33. Once again, however, no expression of the target protein was detected. Further analysis of the ZmCKX6 and ZmCKX9 sequences revealed the presence of a short sequence downstream of the predicted signal peptide that was homologous to a sequence-specific vacuolar sorting signal consisting of the amino acid residues [N/L]-[P/I/L]-[I/P]-[R/N/S] (Nakamura and Matsuoka, 1993; Vitale and Raikhel, 1999). The putative vacuolar retention signals 38LLPT and 37LPTS were identified in ZmCKX6 and ZmCKX9, respectively (see Fig. 1). The removal of all N-terminal residues of ZmCKX9 up to position 42/43 resulted in the expression of a functional enzyme. In the case of ZmCKX6, no CKX activity was detected even after the deletion of this putative vacuolar sorting signal sequence. All of the active ZmCKX proteins were initially purified from clarified cell lysates by chitin affinity chromatography. Although this purification step yielded good results, further purification was required to remove some minor contaminating proteins. Therefore, the dialyzed and concentrated samples were separated by High Q ion-exchange chromatography, after which the collected protein fractions were analyzed by SDS-PAGE and the fractions containing pure ZmCKX protein were pooled and concentrated. The High Q purification step was omitted in the case of ZmCKX9 and ZmCKX12 because of the low recombinant protein yields obtained after the

first purification step in these cases. All of the purified recombinant proteins were examined by SDS-PAGE to estimate their molecular mass and to evaluate the overall performance of the purification process (see Fig. 2). The observed molecular masses of the ZmCKX proteins prepared in the bacterial heterologous expression system ranged from 55.6 to 60.9 kDa (determined using the Alpha DigDoc program). These values are consistent with the theoretical molecular masses of the recombinant proteins, which range from 55.2 (ZmCKX2 and ZmCKX4b) to 59.3 kDa (ZmCKX5). The ZmCKX1 protein generated a band corresponding to a mass of approximately 60.9 kDa rather than its predicted mass of 53.1 kDa. This was probably due to protein hyperglycosylation, which is common in the P. pastoris expression system (for a detailed discussion of this phenomenon, see (Bilyeu et al., 2001).

Fig. 2. SDS-PAGE results for the purified ZmCKX enzymes. The purified recombinant ZmCKX proteins (2ug each) were separated by gradient tricine-SDS-PAGE (4e12% polyacrylamide gel) and then stained with Coomassie Brilliant Blue. Lane contents: 1molecular mass standard, 2 e ZmCKX1, 3 e ZmCKX9, 4 e ZmCKX12, 5 e ZmCKX2, 6 e ZmCKX3, 7 e ZmCKX4a, 8 e ZmCKX4b, 9 e ZmCKX5, 10 e ZmCKX8, 11 e ZmCKX10.

286

D. Zalabák et al. / Plant Physiology and Biochemistry 74 (2014) 283e293

2.2. Site-directed mutagenesis of the ZmCKX6 gene To test the functionality of the ZmCKX6 gene, transgenic tobacco plants expressing the full length native version of ZmCKX6 were prepared as described by Galuszka et al. (Galuszka et al., 2004). It had previously been shown that the constitutive expression of CKX in both tobacco and Arabidopsis plants produces a cytokinin deficiency phenotype that is characterized by retarded shoot growth and an extremely well-developed root system (Werner et al., 2003). Approximately 10 independent transgenic lines were prepared, all of which had wild-type phenotypes in the T1 generation. The fact that heterologous expression of the ZmCKX6 protein did not result in detectable enzyme activity in E. coli lysates or in eukaryotic expression systems may indicate that its coding sequence is a pseudogene. However, ZmCKX6 is expressed strongly in intact maize tissues at the mRNA level (Vyroubalová et al., 2009). While pseudogene transcription is known to occur in some cases, they are usually expressed much less strongly than their functional counterparts (Zou et al., 2009). Furthermore, the ZmCKX6 gene sequence contains no frame shifts or premature termination codons and has a normal distribution of exons and introns. These factors, together with its strong expression in planta suggest that it is a functional gene (Gu et al., 2010) rather than a pseudogene that would be subject to negative selective pressure. It is possible that the lack of detectable CKX activity for ZmCKX6 may be due to a mutation that affects its folding, making it inactive or unstable and prone to rapid degradation. There are several conserved domains within the CKX protein family (Popelková et al., 2004). These include the GHS domain, which is responsible for the covalent binding of the enzyme’s flavin adenine dinucleotide cofactor, and the aspartate residue that functions as a catalytical base during the reaction (Malito et al., 2004). Other unique characteristic domains of the cytokinin dehydrogenases include the PHPWLN and HFG domains located in the C-terminal region. These sequences are strongly conserved, suggesting that they have some important structural or functional role. However, the nature of this role has yet to be determined. When the protein sequences of the maize CKXs were aligned with those from rice, barley, wheat, Brachypodium and Sorghum bicolor, ZmCKX6 has an unusual variation in the conserved 375PHPWLN domain: the proline 375 residue (underlined) is replaced by a histidine, changing the domain sequence to HHPWLN. Interestingly, this variation occurs in five monocotyledonous CKX proteins: ZmCKX6, OsCKX2, HvCKX3, sorghum CKX (Sb01g019000.1) and Brachypodium CKX (Bradi3g29130). These CKX proteins cluster together in the phylogenetic tree, suggesting that each monocot species probably has one isoform carrying this mutation within its CKX gene family (Fig. 3). ZmCKX6 also has a mutation in a second highly conserved sequence: the 499HFG domain; specifically, in this isoform, the phenylalanine 500 residue (underlined) is replaced with leucine. The HFG domain is very strongly conserved both among monocot CKXs and among those of Arabidopsis; ZmCKX6 is the only known monocot enzyme to carry such a mutation. In fact, there is only one other monocot cytokinin dehydrogenase (the barley enzyme HvCKX9) that carries a mutation in the HFG domain e an exchange of the phenylalanine residue for tyrosine. However, this change does not affect its CKX activity: HvCKX9 overexpression in tobacco plants produces a strong cytokinin-deficient phenotype (Galuszka et al., 2004) (note that this

Fig. 3. Phylogenetic tree of CKX families from monocotyledonous plants. The NCBI database (http://www.ncbi.nlm.nih.gov/) of plant protein sequences was searched for ZmCKX homologs in various monocotyledonous species. The BLAST homology searches on SbGDB and BdGDB (http://www.plantgdb.org/SbGDB/; http://www.plantgdb.org/BdGDB/) were

also used to search for CKX protein sequences in the sorghum and Brachypodium genome, respectively. The tree was generated by protein sequence alignment with ClustalW using the neighbor-joining method (Saitou and Nei, 1987) in program MEGA5 (Tamura et al., 2011). The sequences of maize CKX enzymes were aligned with sequences from rice, wheat, barley, Brachypodium and sorghum. The proteins in the circled cluster all have a common mutation in the PHPWLN domain but not in the HFG domain.

D. Zalabák et al. / Plant Physiology and Biochemistry 74 (2014) 283e293

gene was recently renamed from HvCKX2 to HvCKX9 according to Mameaux et al. (2012). All other known monocot CKXs and all of the Arabidopsis CKXs contain the conserved HFG domain. Based on this sequence analysis, ZmCKX6 variants were prepared carrying reverting mutations in either the HHPWLN or the HLG domain and with the N-terminal LLPT signal removed (Fig. 1). The reverting mutation in the HHPWLN domain of ZmCKX6 (H375P) did not yield a product with CKX activity. However, the reverting mutation in the HLG domain (L500F) produced a functional enzyme with CKX activity. The removal of both the signal peptide and the vacuolar sorting signal was required for CKX activity: no CKX activity was detected in cell lysates expressing a HLG reversion mutant with an intact LLPT signal. The expression of the native and mutated recombinant ZmCKX6 proteins was investigated by western blot analysis using an anti-ZmCKX6 antibody. Although no CKX activity was detected in cell lysates of bacteria expressing native ZmCKX6, western blot analysis revealed that the protein was weakly expressed (Fig. 4), suggesting that the ZmCKX6 gene either encodes a protein with very low activity or a nonfunctional protein, making it a pseudogene. This phenomenon seems to be unique since the protein-level expression of pseudogenes has not previously been reported. The reverting mutation in the HLG domain (L500F) of ZmCKX6 increased its level of expression, but no such effect was observed for the reverting mutation in the HHPWLN domain (Fig. 4). These results suggest that the HFG domain is crucial for both the activity and stability of CKX enzymes, a hypothesis that is consistent with the domain’s very strong conservation in this class of enzymes. 2.3. Biochemical properties of the ZmCKX enzymes To begin with, the specific activity of each isoform was determined by means of the dichlorophenolindophenol (DCPIP) decolorization assay, using isopentenyladenine (iP), isopentenyladenosine 50 monophosphate (iPRMP) and isopentenyladenine-N9-glucoside (iP9G) as substrates (Laskey et al., 2003). This method was chosen because it allows the use of cis-zeatin (cZ) derivatives as substrates and therefore makes it possible to determine reaction rates for all of the natural CKX substrates. As shown in Table 1, the maize CKX enzyme with the highest specific activity towards iP is ZmCKX1 (890 nkat mg1). All of the other isoforms were several orders of magnitude less active towards this substrate, with activities ranging

Fig. 4. Western blot analysis of various recombinant ZmCKX6 proteins. The clarified protein lysates (5 mg of protein) from bacterial cells expressing various ZmCKX6-intein fusions were subjected to SDSePAGE (10% polyacrylamide gel) and electro-blotted onto a nitrocellulose membrane. The fusion proteins were detected using a polyclonal antibody against the ZmCKX6-specific peptide CSANPGPEEDGDG and visualized colorimetrically. Lane contents: 1 e molecular mass standard, 2 e negative control (empty vector), 3 e ZmCKX6 wild type, 4 e ZmCKX6 HFG reversion mutant, 5 e ZmCKX6 PHPWLN reversion mutant.

287

Table 1 Specific activities of ZmCKX enzymes. Specific activities were determined in 250 mM sodium phosphate buffer (pH 7.0) containing 2.5 mM EDTA, with 250 mM iP, iPRMP, or iP9G as substrates and 0.125 mM DCPIP as an electron acceptor. Enzyme

Specific activity (nkat mg1) iP

iPRMP

iP9G

ZmCKX1 ZmCKX2 ZmCKX3 ZmCKX4a ZmCKX4b ZmCKX5 ZmCKX8 ZmCKX9 ZmCKX10 ZmCKX12

890.38 29.29 15.47 13.90 17.94 42.87 51.23 0.64 9.21 26.83

5.53 20.87 16.18 47.28 47.32 44.08 6.52 0.22 0.08 0.48

215.58 62.32 54.60 45.22 56.15 27.54 50.35 0.45 14.68 2.87

from 9.21 to 51.23 nkat mg1. The isoforms’ specific activity towards iPRMP ranged from 0.08 to 47.32 nkat mg1with ZmCKX4a, ZmCKX4b and ZmCKX5 being particularly active towards this substrate. Conversely, ZmCKX9, ZmCKX10 and ZmCKX12 all exhibited very little specific activity towards iPRMP. Similarly, the isoforms’ specific activities towards iP9G ranged from 0.45 to 215.58 nkat mg1. These results clearly show that the different CKX isoforms have very different activities towards cytokinin substrates with different substituents at the N9 position of the adenine ring. For instance, ZmCKX1 is very active towards iP but 200 times less active towards iPRMP. Next, the maize CKX enzymes’ substrate preferences with respect to the other naturally occurring cytokinins were determined. DCPIP decolorization assays were performed for all isoforms using trans-zeatin (tZ), cZ, dihydrozeatin (DHZ) and the corresponding N9-glucoside and nucleotide derivatives (iP9G, tZ9G, cZ9G, iPRMP, tZRMP and cZRMP) as substrates. With the exception of ZmCKX5, all of the isoforms exhibited relatively high activity towards cZ (Fig. 5). Some isoforms, such as ZmCKX8, ZmCKX9, ZmCKX10 and ZmCKX12 preferred cZ to all of the other tested cytokinin free bases. These results stand in contrast to those for the Arabidopsis CKX enzymes, which exhibit comparatively little activity towards cZ (Gajdo
sová et al., 2011). Only two AtCKX enzymes, the vacuolar AtCKX1 and the cytosolic AtCKX7, exhibited a level of activity towards cZ that was similar to their activity towards iP. None of the tested ZmCKXs were able to degrade dihydrozeatin. Interestingly, ZmCKX1 prefers cytokinin free bases (specifically, tZ

Fig. 5. Substrate preference profiles of the ZmCKX enzymes. Activities were determined using a substrate concentration of 250 mM and a DCPIP concentration of 0.125 mM, in 250 mM sodium phosphate buffer (pH 7.0), containing 2.5 mM EDTA. The bars show the ratio of each CKX enzyme’s activity towards the indicated substrates relative to its activity towards iP (which was arbitrarily chosen to represent 100% activity). DHZ was also tested but none of the ZmCKX enzymes were active towards this substrate. Enzyme activities were measured 5e10 times for each enzyme and substrate and the measured activities were averaged. In all cases, the standard deviation over these 5e10 runs did not exceed 20% of the absolute values.

288

D. Zalabák et al. / Plant Physiology and Biochemistry 74 (2014) 283e293

and iP) to the almost complete exclusion of other substrates, which it degrades only very slowly. ZmCKX12 exhibits a similar substrate preference pattern. In light of its specific activity, substrate preferences, and ubiquitous expression profile (see Section 2.5.), ZmCKX1 is probably the main isoform responsible for the rapid degradation of cytokinin free bases in the extracellular space. The vacuolar and cytosolic CKXs of Arabidopsis exhibit a significant preference for cytokinin-9-glucosides and cytokinin mono-, diand triphosphates relative to free bases (Kowalska et al., 2010). Interestingly, the cytosolic maize enzyme ZmCKX10 preferentially degraded cZ, followed by tZ and iP, and its ability to degrade cytokinin monophosphates and zeatin-9-glucosides was very low (Fig. 5). While ZmCKX10 was more active towards iP9G than iP, the difference between the rates at which it degrades the two substrates was much less pronounced than those for the intracellular Arabidopsis CKXs, which typically degrade free bases several orders of magnitude more slowly than cytokinin phosphates or N9glucosides. These different substrate preference patterns probably create different cytosolic cytokinin distributions in the two plant species. The substrate preferences of ZmCKX9, a putative vacuolar isoform, were very similar to those of the cytosolic CKX enzyme, and it did not exhibit any preference for N9-glucosides or phosphates. Another putative vacuolar isoform, ZmCKX6, was not characterized because it did not exhibit any detectable CKX activity in the bacterial cell lysates. Interestingly, ZmCKX8 predicted to be localized to the mitochondria had similar substrate preferences to the cytosolic and vacuolar CKXs. The substrate preference patterns of ZmCKX2 and ZmCKX3 were very similar, in keeping with the suggestion that one of them was formed by a recent duplication of the other (the two enzymes exhibit 93.5% sequence identity at the amino acid level). ZmCKX4a and ZmCKX4b also had very similar substrate profiles and exhibit around 93.4% sequence identity. Moreover, both of these two enzyme pairs had broadly similar substrate preferences (Fig. 5). However, compared to ZmCKX2 and ZmCKX3, ZmCKX4a and ZmCKX4b had a stronger preference for iP and cZ derivatives and were less capable of degrading tZ-type cytokinins. All four isoforms degraded iP9G preferentially, but unlike ZmCKX2 and ZmCKX3, ZmCKX4a and ZmCKX4b also exhibited similar rates of reaction with iPRMP. UHPLC was also used to investigate the substrate preferences of the ZmCKX enzymes in order to complement the results obtained with the DCPIP decolorization assay. In these experiments, the enzymes were probed with a mixture of cytokinins and the depletion of each substrate was monitored over a period of 120 min (Pertry et al., 2009). Three different cytokinin mixtures were tested. CKX activity was measured in both dehydrogenase (pH 5.0, with ferricyanide as an electron acceptor) and oxidase modes (pH 5.0, O2 as an electron acceptor) in each case. The first contained only free bases (iP, tZ, cZ and DHZ) while the second and third contained derivatives of iP (iP, isopentenyladenosine, iPRMP, iP9G and Fig. 6. Temporal cytokinin profiling in the organs of developing maize seedlings. The heat map shows the distribution of cytokinins in maize kernels (A), roots (B) and shoots (C) during the first 20 days after germination. Roots, shoots and kernels from maize seedlings (cv. ‘Cellux’), cultivated on hydroponic system were harvested at defined time points and analyzed to determine the concentrations of each cytokinin species. The heat map shows the average accumulation of each plant hormone over time. Red and blue correspond to higher and lower concentrations, respectively. The color scale is shown at the top. The value in each block indicates the measured concentration of the compound in question (based on the mean of two measurements). The concentration in A is expressed in units of pmol per 20 kernels, whereas those in B and C are expressed in pmol g1 of fresh weight. Total-CK represents the summed concentrations of each cytokinin group: tZ e trans-zeatin, cZ e cis-zeatin, DHZ e dihydrozeatin, and iP e isopentenyladenine. ND e value below the limit of quantification. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

D. Zalabák et al. / Plant Physiology and Biochemistry 74 (2014) 283e293

isopentenyladenine-N7-glucoside) and tZ (tZ, trans-zeatin riboside, tZRMP, tZ9G and trans-zeatin-O-glucoside), respectively. Surprisingly, all of the tested CKX isoforms preferred iP as a substrate in this experiment, indicating that iP has the lowest Km values for all of the ZmCKXs. With the exception of iP, the order of substrate preferences for each enzyme was found to be the same as in the DCPIP decolorization assays (for details, see Supplementary Fig. 1). This experiment also revealed that isopentenyladenine-N7-glucoside and trans-zeatin-O-glucoside are resistant to degradation by maize cytokinin dehydrogenases. There were no differences in substrate preference between the dehydrogenase and oxidase modes of the studied enzymes (data not shown). The rates of reaction with oxygen as the electron acceptor were significantly lower than those observed in dehydrogenase mode, indicating that all of the ZmCKXs will function as dehydrogenases if an appropriate electron acceptor is available. The putative inhibitory effect of DHZ on cytokinin dehydrogenase activity was tested using the endpoint method with p-aminophenol (Frébort et al., 2002) (data not shown). Interestingly, no inhibition was observed when the ZmCKXs were probed with DHZ either at equimolar concentrations or with a fivefold excess of the potential inhibitor relative to the substrate (100 mM iP). This suggests that DHZ cannot access the active sites of the ZmCKX enzymes. 2.4. Cytokinin profiling in developing maize plantlets The cytokinin profiles in maize plants have been studied extensively, but these investigations have either focused on the spatial rather than the temporal distribution of the cytokinins (Gajdo
sová et al., 2011; Stirk et al., 2012) or have been limited to only few cytokinin species (Cheikh and Jones, 1994; Dietrich et al., 1995). The roles of cytokinins during the early stages of maize seed imbibition and germination (during the first 36 h of the plant’s life) have been described in detail (Stirk et al., 2012). In this work, we focused on the developmental stages that follow the emergence of the radicle on day 4 of the plant’s life, and studied the cytokinin profiles in the kernels, roots and shoots of young maize seedlings over a two-week period. The prevalent cytokinins in dormant maize seeds are DHZ derivatives (Stirk et al., 2012) (Fig. 6A), of which the predominant forms are DHZ-O-glucoside and DHZ riboside-O-glucoside. The levels of these specific compounds and the overall concentration of DHZ-type cytokinins both decrease dramatically during the first 48 h of imbibition but remain high compared to other CK forms. The total concentration of DHZ-type species stabilize at around 25% of their initial levels on the fifth day (Fig. 6A). A similar trend was observed for tZ-type cytokinins; however the decrease in this case was less pronounced, and their overall levels were much lower than those of the DHZ-type compounds at all stages. The levels of iP-type cytokinins in the kernels of developing maize seedlings were stable and relatively low. After an initial fluctuation, the concentrations of cZ-type cytokinins increased on the fifth day, reaching levels approaching those of the DHZ-type compounds at day 12. The concentration of DHZ-type compounds in the shoots mirrored the trend seen in the kernels, gradually falling to 5 pmol g1 of fresh weight over the studied period (Fig. 6C). Interestingly, the opposite trend was observed in the roots: the DHZ concentration did not vary much initially but started to increase approximately two weeks after germination, primarily due to the accumulation of the corresponding O-glucoside (Fig. 6B). The levels of tZ-type cytokinins in developing maize roots fluctuated whereas those in the shoots increased slowly. In contrast, the concentration of cZ-type cytokinins increased slowly in both the roots and the shoots of the seedlings. It has previously been shown that cZ-type cytokinins (mainly cis-zeatin-O-glucoside, cis-zeatin riboside-O-

289

glucoside and cZ) account for around 80% of the total CK content of maize leaves (Gajdo
sová et al., 2011). As shown in Fig. 6, the levels of cZ-type cytokinins change very rapidly during seedling growth. It is likely that both cZ catabolism (promoted by CKXs) and glycosylation by cZ O-glucosyl transferases (UDP-glucose : cis-zeatin O-b-Dglucosyltransferase; EC 2.4.1.215) work synergistically to fine tune the cZ levels during this developmental stage. The concentrations of iP derivatives in the roots and shoots were generally low; those in the shoots were more or less constant whereas their levels in the roots decreased gradually during seedling development. The roles played by DHZ derivatives during plant development remain unclear, as do the details of their metabolism. Due to their abundance in dormant seeds, they are assumed to be inactive cytokinins that provide a substrate pool for the synthesis of active cytokinin compounds (Frébort et al., 2011). Some reports have described the conversion of tZ to DHZ by a putative enzyme, zeatin reductase (dihydrozeatin: NADPþ oxidoreductase; EC 1.3.1.69), although the enzyme itself has not yet been purified and characterized (Martin et al., 1989; Gaudinová et al., 2005). The existence of a reverse pathway could explain the dramatic depletion of the DHZ cytokinin pool observed in imbibing maize seeds (Stirk et al., 2012; Fig. 6A). However, no enzyme capable of converting DHZ into tZ, cZ or iP has yet been described. The degradation of DHZ by cytokinin dehydrogenases is excluded because none of the maize cytokinin dehydrogenases examined in this work was able to accept DHZ as a substrate. The absence of a double bond in the DHZ side chain seems to be responsible for its resistance to degradation by cytokinin dehydrogenases (Popelková et al., 2006). Although it is often assumed that DHZ is an inactive cytokinin, several reports have demonstrated that it has varied biological effects. Among other things, it delays senescence in oat leaves and is active in the phaseolus callus assay (Mok et al., 1978; Letham et al., 1983). On the other hand, DHZ does not bind to Arabidopsis cytokinin receptors at physiologically relevant concentrations. All three Arabidopsis cytokinin receptors (AHK4/CRE1, AHK2 and AHK3) have been biochemically characterized in detail (Spíchal et al., 2004; Stolz et al., 2011), revealing that their KD values for binding to DHZ range from 50 nM to 400 nM. The KD values for their best substrates ranged from 1.3 to 3.9 nM. Moreover, the levels of DHZ in Arabi

dárská dopsis tissues are usually below the limits of detection (Z et al., 2013). Interestingly, the situation is different in maize since one of the three maize cytokinin receptors, ZmHK2, has a high affinity for DHZ. The KD value of this receptor for DHZ is 2 nM, which is very close to its KD values for tZ (0.4 nM) and iP (0.6 nM) (Lomin et al., 2011). However, the ability of DHZ to activate the receptor’s downstream phosphorylation signaling pathway has not yet been tested. It is therefore not clear whether DHZ is an active cytokinin in tissues where it is abundant (e.g. kernels) or whether it functions as a sort of anti-cytokinin, competitively inhibiting the binding of active cytokinins to the ZmHK2 receptor and thereby attenuating cytokinin perception. 2.5. Expression profiles of ZmCKX genes As described in Section 2.4, the early stages of maize seedling development are accompanied by diverse changes in the cytokinin contents of the roots and shoots. To determine whether these changes in CK contents are due to changes in CKX enzyme activities, their expression was profiled using qPCR. The expression of ZmCKX genes was determined in the roots, stems, and first leaves of developing maize seedling during the first 20 days after imbibition. The changes in the expression of some ZmCKX genes in developing roots mirror the changes in the concentrations of their preferred substrates (Fig. 7A). The trend is most pronounced in the case of ZmCKX1, whose level of expression initially increases

290

D. Zalabák et al. / Plant Physiology and Biochemistry 74 (2014) 283e293

not change significantly during the studied period. On the other hand, ZmCKX6 gene expression increased gradually and peaked on day 16. However, its contribution to cytokinin homeostasis is unclear because it probably encodes a non-functional enzyme. The overall levels of CK free bases and ribosides in the developing shoot were low and mirrored the trends seen in the roots with a delay of approximately two days. While ZmCKX1 is not significantly expressed in the aerial parts of maize seedlings, it might contribute to the regulation of cytokinin levels in the leaves and stem due to acropetal cytokinin transport. The ZmCKX1 protein is known to be primarily localized in vascular bundles (Brugiere et al., 2003; Galuszka et al., 2005) and as a highly active apoplastic isoform, it will therefore influence the flow of root-borne cytokinins into the aerial parts of the plant. Other CKX genes probably contribute to the maintenance of cytokinin homeostasis in specialized cells. For instance, the peak in ZmCKX2 and ZmCKX12 expression in the stems during the 3rd week after germination may reflect the observed decrease in the levels of free bases and tZ and iP nucleotides at this point in time (Fig. 7C). However, these proposed relationships between cytokinin levels and the expression of individual CKX genes are highly speculative because only two maize isoforms have had their cellular localizations determined, and little is currently known about the subcellular distributions of various cytokinin forms. Because O-glucosides are not substrates for any maize CKX enzyme, their concentrations increase continuously as the plant ages. 3. Conclusions Fig. 7. Expression profiles of ZmCKXs in maize roots (A), first leaves (B) and stems (C) during early seedling development. Total RNA was isolated from maize tissues from two independent biological replicates. cDNA from each sample was obtained in at least two independent reactions and amplified in two separate PCRs. The b-actin and elongation factor 1 genes were used as an endogenous controls, and tissue samples from day 4 (root) or day 6 (leaf and stem), respectively were used as calibrators (blue bars) to define an expression level of 1. The bars indicate relative gene expression and the error bars indicate SD values (n ¼ 4). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

gradually, reaching a peak on the 14th day, and then starts to decline. The levels of its preferred substrates (tZ, its riboside, and iP) peak just a few days before this. The abundance of ZmCKX1 transcripts is thus clearly dependent on the levels of active cytokinins, and it seems that this enzyme plays a key role in maintaining cytokinin homeostasis in maize. A similar relationship between ZmCKX1 expression and the levels of active cytokinins has previously been observed in the pedicel tissue of developing kernels (Brugiere et al., 2003). The levels of ZmCKX2 and ZmCKX4b transcripts also decreased in six-day old roots in parallel with decreases in the levels of their preferred substrates e nucleotides and N9glucosides. The expression of the other members of the ZmCKX family did not vary greatly during the studied period of time. The concentrations of the transcripts of ZmCKX3, ZmCKX4a and ZmCKX5, ZmCKX9 and ZmCKX12 were below the limit of detection at all times and so these enzymes are unlikely to play significant roles during the development of maize primary roots. In the first maize leaf, the expression of the ZmCKX2 and ZmCKX3 genes fluctuated and no obvious trend could be discerned (see Fig. 7B). The expression of ZmCKX4b and ZmCKX6 peaked on the 8th day. ZmCKX10 expression also peaked on the 8th day, with a second peak on day 20. ZmCKX1, ZmCKX4a, ZmCKX5, ZmCKX8, ZmCKX9 and ZmCKX12 were only weakly expressed and so their transcript levels could not be quantified with any precision. As shown in Fig. 7C, the expression of ZmCKX4b, ZmCKX8, ZmCKX9 and ZmCKX10 in the stems was similar each other as it did

The maize genome contains thirteen genes that encode cytokinin dehydrogenases. Three of these had previously been shown to encode functional protein products but only two, ZmCKX1 and ZmCKX10, had been studied in detail. This work described the properties of nine further isoforms, which were obtained in pure form by expressing them as recombinant fusion proteins with inteins in E. coli and subsequently isolating them from cell lysates. In contrast to the case for most members of the CKX family in Arabidopsis, maize CKXs are generally able to degrade cZ-type cytokinins, often at rates that are higher than those for tZ-type cytokinins. This preference for cZ-type cytokinins probably reflects the fact that cZ-type cytokinins are abundant in maize tissues, whereas in Arabidopsis they predominate only in seeds. As is the case in Arabidopsis, maize contains two putative vacuolar (ZmCKX6 and ZmCKX9) and one cytosolic (ZmCKX10) CKX isoforms. In Arabidopsis, the vacuolar and cytosolic CKXs have strong preferences for cytokinin-9-glucosides and cytokinin mono-, di- and triphosphates relative to free bases, whereas the maize enzymes degrade cytokinin monophosphates and zeatin-9-glucosides more slowly than free bases. Although the ZmCKX6 gene is expressed strongly in maize roots and shoots, the protein it encodes is probably inactive. Sequence analysis and subsequent site directed mutagenesis showed that the protein does not contain the conserved HFG domain motif, which is essential for CKX activity and stability in heterologous expression systems. The physiological relevance of this finding remains to be determined. 4. Materials and methods 4.1. Cloning of ZmCKX genes Complementary DNA sequences encoding ZmCKX2, 4a, 4b, 5, 6 and 12 cloned in the pGEM-T Easy vector have been prepared previously (Gu et al., 2010). A complementary DNA sequence encoding ZmCKX3 cloned in the pINA1267 vector was kindly provided by Dr. David Kope
cný (Massonneau et al., 2004). A complementary DNA

D. Zalabák et al. / Plant Physiology and Biochemistry 74 (2014) 283e293

sequence coding for ZmCKX10 cloned in the pDrive vector was
kindly provided by Dr. Mária Smehilová et al. (2009). The coding regions of ZmCKX8 and ZmCKX9 were synthesized using the GeneArt service (Invitrogen) based on predictions generated by the maize genome project (http://www.maizesequence.org/). The cloned cDNA sequences were used as starting materials for PCR amplification. Genes were amplified without putative signal peptides in cases where such peptides were present using the Phusion DNA polymerase (Finzymes). The forward and reverse primers were designed to contain NdeI and EcoRI restriction sites, respectively; the primer sequences used in each case are listed in Table 2. The amplification program was as follows: 30 s at 98  C, followed by 30e 35 cycles of 10 s at 98  C, 30 s at 61  C and 60 s at 72  C. The program was then terminated with a final extension step of 10 min at 72  C. The PCR reactions were performed in the presence of 1.3% dimethyl sulfoxide and 1.3 M betaine. PCR products were purified using QIAquick PCR Purification Kit (Qiagen) and subsequently digested with NdeI and EcoRI (NEB) overnight at 37  C. The pTYB12 plasmid was treated in the same way. Gel-extracted PCR fragments and linearized samples of the pTYB12 vector were ligated using the T4 DNA ligase (NEB) overnight at 16  C. The ligation mixtures were electroporated into competent E. coli cells of the TOP10 (NEB) strain. The positive clones were then sequenced to confirm the nucleotide sequences of the cloned genes (GATC Biotech, Germany). 4.2. Heterologous production and purification of recombinant ZmCKX enzymes All of the ZmCKX enzymes with the exception of ZmCKX1 were purified from E. coli BL21 STAR (DE3) cell cultures transformed with pTYB12:ZmCKX constructs. Bacteria were cultivated overnight at 37  C in 50 mL of liquid LB media (ampicillin 100 mg L1) in a 250 mL Erlenmeyer flask, with orbital shaking at 160 rpm. The bacterial culture was then used to inoculate fresh 50 mL LB media aliquots supplemented with glucose (10 g L1) and carbenicillin (mg L1) to a final OD600 value of 0.1. These aliquots were cultivated at 37  C on an orbital shaker (160 rpm) until their OD600 value rose Table 2 Primer sequences used to clone ZmCKX genes into the pTYB12 vector. The underlined sequences correspond to the NdeI (fw primers) and EcoRI (rev primers) restriction sites, respectively. Primer name

Position of signal peptide

Sequence 50 to 30 -end direction

ZmCKX2_fw ZmCKX2_rev ZmCKX3_fw ZmCKX3_rev ZmCKX4a_fw ZmCKX4a_rev ZmCKX4b_fw ZmCKX4b_rev ZmCKX5_fw ZmCKX5_rev ZmCKX6_fw ZmCKX6_rev ZmCKX6_fw ZmCKX6_rev ZmCKX8_fw ZmCKX8_rev ZmCKX9_fw ZmCKX9_rev ZmCKX9_fw ZmCKX9_rev ZmCKX10_fw ZmCKX10_rev ZmCKX12_fw ZmCKX12_rev

19/20

GGAATTCCATATGAGGCTGACCATGCACGTCC CCGGAATTCTCACAAAGACAATGGGAGTGAC GGAATTCCATATGCACGTCCCCGACGAG CCGGAATTCTCACAAAGGCAATGGCAGTGAC GGAATTCCATATGCTCCCCGTCGAGCCGC CGGAATTCCAGCTGGATTACGTTACGAGTC GGAATTCCATATGCTGCCGGTGGAGCCGC CCGGAATTCTCACGAGTCGGCGACGAG GGAATTCCATATGACCCCTTCCACGCTCGCC CGGAATTCGATCATGAGGCAAGTAGCGGG GGAATTCCATATGGACCTGGGCCCCCTG CCGGAATTCCTAGCTCCCGTACGCAGC GGGAATTCCATATGGCGGCCGCGTCCAGCG CCGGAATTCCTAGCTCCCGTACGCAGC GGAATTCCATATGCACCCCCGGCCACTGC CGGAATTCCTCAGGCAGAAGCAATGCCAG GGGAATTCCATATGCTCATCGTCACCAGCTTCCTCC CCGGAATTCTTAAAGAACATTTCGAGAATCAGAGCCC GGGAATTCCATATGCATTCCCATGTGCCTGCTGTAAC CCGGAATTCTTAAAGAACATTTCGAGAATCAGAGCCC GGAATTCCATATGATGCTCGCGTACATGGAC CCGGAATTCCTACACGGCGACGGACG GGAATTCCATATGGTGAATTTCATACAGAGCCC CCGGAATTCTCACATCGGGCTCGAGG

23/24 21/22 22/23 21/22 30/31 44/45 23/24 30/31 42/43 e 20/21

291

to 0.5. Expression of the recombinant protein was induced by adding IPTG to a final concentration of 0.5 mM. Overnight cultures incubated at 18  C on an orbital shaker (160 rpm) were then centrifuged at 4500 g to harvest bacterial cells. The recombinant proteins were purified according to the manufacturer’s protocol with some modifications (IMPACT, NEB). The bacterial pellet from 5 to 10 L of expression culture was dissolved in lysis buffer (20 mM Tris/HCl, pH 8.0, 1 M NaCl, 0.1% Triton X-100, 5% glycerol, 1 mM PMSF) and then disrupted using a French press (Thermo Scientific) at 25,000 psi. The cell lysates were clarified by centrifugation at 20,000 g, after which the supernatants were loaded onto an equilibrated chitin column at a flow rate of 0.5 mL min1. The column was then washed with column buffer (which was identical to the lysis buffer without glycerol), flushed with DTT (50 mM in column buffer) and incubated for 64 h at þ4  C to induce intein cleavage. The protein was then eluted with 60 mL of column buffer and dialyzed against 2 L of 50 mM Tris/HCl, pH 8.0 (three times, with the second dialysis being performed overnight). The dialyzed sample was then concentrated using an ultrafiltration cell (10 kDa cutoff membrane, Amicon) and loaded onto an ionexchange chromatographic column High Q (BioRad) to remove residual contaminating bacterial proteins. The purified proteins were eluted with a linear gradient of KCl (0e1 M), collecting 2 mL fractions. The purity of each fraction was tested by SDS-PAGE. Pure fractions were subsequently pooled and concentrated by ultrafiltration to a final volume of about 2 mL. The ZmCKX1 enzyme was expressed in P. pastoris as described by Bilyeu et al. (2001) and purified by ion-exchange chromatography on hydroxypatite. 4.3. Site-directed mutagenesis of ZmCKX6 The open reading frame of ZmCKX6 was modified in two separate positions using a QuikChange II XL Site-Directed Mutagenesis Kit according to the manufacturer’s protocol (Stratagene). The following primers were used to change the CAC sequence encoding residue 375 to CCC (histidine to proline): ZmCKX6_PHPWLN_f (forward; 50 - GG GAG GAG GTG CCC CAC CCG TGG CTC e 30 ) and ZmCKX6_PHPWLN_r (reverse; 50 - G AGC CAC GGG TGG GGC ACC TCC TCC C e 30 ). The CTG sequence encoding the residue at position 500 was converted to TTT (leucine to phenylalanine) using the following primer pair: ZmCKX6_HLG_f (forward; 50 - GG CAG CAG CAC TTT GGC CGG CGC TGG e 30 ) and ZmCKX6_HLG_r (reverse; 50 CCA GCG CCG GCC AAA GTG CTG CTG CC e 30 ). Bold letters indicate the positions of the introduced site mutations. As the template for site-directed mutagenesis, 50 ng of two constructs carrying long and short version of the ZmCKX6 gene were used per reaction, respectively. The long version contained the ORF amplified without the putative secretory signal (corresponding to the first 30 amino acids). The short version contained the ORF amplified without the putative vacuolar sorting signal (corresponding to the first 44 amino acids). Each construct was repaired in two separate reactions of 18 cycles each, with annealing at 68  C. The successful formation of the desired mutated constructs was initially verified by restriction analyses since the changes resulted in the deletion of an ApaLI (PHPWLN domain) restriction site and the creation of an EaeI restriction site (HFG domain), respectively. All of the repaired constructs were then analyzed by sequencing. 4.4. Western-blot analysis A polyclonal rabbit antibody raised against the ZmCKX6-specific peptide CSANPGPEEDGDG was prepared by a commercial firm (GenScript) and used for western-blot analysis. SDSePAGE was performed on a slab gel (10%) in Triseglycine running buffer. The PageRuler Unstained Protein Ladder (Thermo Scientific) was used

292

D. Zalabák et al. / Plant Physiology and Biochemistry 74 (2014) 283e293

as a marker. Protein samples were heated before application at 95  C for 10 min in the presence of 1% SDS and 1% 2mercaptoethanol. Proteins were blotted onto a nitrocellulose membrane (0.45 mm) in the MiniTrans blot system (Bio-Rad, Hercules, CA, USA). Membranes were stained with Ponceau S blue to visualize proteins. After decolorization in water, the membrane was blocked with 3% gelatine in 20 mM Tris/HCl (pH 7.5), containing 500 mM NaCl (TBS buffer), for 2 h. After the blocking, the membrane was washed twice for 10 min in TBS buffer containing 0.05% of Tween-20, and then probed with rabbit anti-ZmCKX6 antibody diluted in 1% gelatine in TBS buffer (1:1000). The membrane was subsequently rinsed twice with Tween-TBS, and then incubated for 2 h with anti-rabbit IgG containing conjugated alkaline phosphatase (Sigma) in 1% gelatine in TBS buffer (1:5000). After rinsing in Tween-TBS buffer, the membrane was washed quickly in a buffer for the alkaline phosphatase reaction (100 mM Tris/HCl, 100 mM NaCl, 5 mM MgCl2, pH 9.5) and then developed in the same buffer with the addition of nitrotetrazolium blue chloride (NBT) and 5bromo-4-chloro-3-indolyl phosphate, disodium salt (BCIP). 4.5. CKX activity assay 4.5.1. DCPIP decolorization assay The cytokinin dehydrogenase enzyme assay was based on the decolorization of the electron acceptor 2,6-dichlorophenolindophenol (DCPIP) according to Laskey et al. (2003). The reaction mixture contained a substrate at a concentration of 250 mM (iP, tZ, cZ, DHZ, iP9G, tZ9G, cZ9G, iPRMP, tZRMP or cZRMP) and 0.125 mM DCPIP in 250 mM sodium phosphate buffer at pH 7.0, with 2.5 mM EDTA. The reaction was initiated by adding of 20 mL of a suitably diluted enzyme to 2 mL of reaction mixture. The decrease in absorbance at 590 nm was recorded at 15 s intervals for 3 min in a spectrophotometer equipped with a multi-cell holder (Agilent). A reference sample was also analyzed in which the substrate was replaced with the solvent dimethyl sulfoxide. iP was used as a reference substrate for all of the tested compounds, enabling the determination of relative activities. Reactions were conducted in duplicate and each substrate was analyzed in triplicate at least. In addition to the determination of relative activity, the specific activity of each enzyme towards iP was measured. 4.5.2. UHPLC method The activity of each enzyme towards a mixture of cytokinin substrates was determined by measuring the depletion of each cytokinin over a period of 120 min. The reaction mixtures contained various cytokinins at concentrations of 1 mM each and 0.5 mM ferricyanide or O2 in 100 mM McIlvaine buffer at pH 5.0. The first cytokinin mixture consisted of iP, tZ, cZ and DHZ; the second contained iP, iPR, iPRMP, iP9G and iP-N7G; and the third contained tZ, tZR, tZRMP, tZ9G and tZ-O-glucoside. In each case, the reaction was initiated by adding 20 mL of suitably diluted enzyme to 400 mL of reaction mixture at 37  C. At preselected time points (0, 5, 10, 20, 60 and 120 min) aliquots of 50 mL were collected, heated to 95  C for 5 min to inactivate the enzyme, and then chilled on ice. The aliquot was then combined with 50 mL of mobile phase A (15 mM ammonium formate, pH 4.0) and filtered through 0.22 mm nylon filters. The resulting samples were injected onto a C18 reverse-phase column (ZORBAX RRHD Eclipse Plus 1.8 mm, 2.1  50 mm, Agilent) connected to a Nexera (Shimadzu) ultra performance liquid chromatograph. The column was eluted with a linear gradient of mobile phase A and methanol (B) using the following solvent mixtures: 0 min, 20% B; 3e8 min, 20e100% B. The flow rate was 0.4 mL min1 and the column temperature was 40  C. The depletion of various cytokinins from the reaction mixture at each time point was determined by peak area subtraction using the LabSolutions software package (Shimadzu).

4.6. Plant material (hydroponic cultivation) Dry kernels of Zea mays cv. ‘Cellux’ were imbibed in tap water and germinated on wetted filter paper. After 3 days, the seedlings were transferred to aerated hydroponic tanks filled with Hoagland’s nutrient solution. The plants were grown in an environmental chamber with 16 h light periods (250 mmol m2 s1) at 27  C and 8 h dark periods at 20  C. Samples of roots and shoots (or stems and first leaves) were collected at the defined time-points from at least 10 maize seedlings and immediately frozen in liquid nitrogen. 4.7. qPCR expression profiling The expression of CKX genes was profiled and analyzed as described previously (Vyroubalová et al., 2009). Briefly, total RNA was isolated using the RNAqueousÒ kit and treated twice with the TURBO DNase-freeÔ kit (Life Technologies) and then used in first strand cDNA synthesis with the RevertAidÔ H Minus M-MuLV reverse transcriptase and oligo-dT mixture (Fermentas). Diluted cDNA samples were used as templates in real-time PCR reactions containing the POWER SYBR Green PCR Master mix or TaqManÒ Gene Expression Master Mix (Life Technologies), 300 nM of each primer, and 250 nM of specific 50 6-carboxyfluorescein and 30 tetramethylrhodamine labeled probes, respectively. RNA from each biological replicate was transcribed in two independent reactions and each cDNA sample was analyzed in at least two technical replicates on a StepOnePlusÔ Real-Time PCR System using one of the default programs (Life Technologies). Ct values were normalized with respect to b-actin and elongation factor 1 genes. Expression values were determined and statistically evaluated using the DataAssist v3.0 Software package (Life Technologies). 4.8. Cytokinin analysis Cytokinins were purified from maize tissues and subsequently analyzed according the procedure described in detail by Vyroubalová et al. (2009). Acknowledgments This work was supported by the Czech Science Foundation grant P501/10/1141 and by the OP RD&I grant No. ED0007/01/01 Centre of the Region Haná for Biotechnological and Agricultural Research. P.G. and J.F. were partly supported by the Operational Program Education for Competitiveness e European Social Fund (project CZ.1.07/2.3.00/20.0165). We thank Marta Greplová for preparation of ZmCKX1. Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.plaphy.2013.11.020 References Bannai, H., Tamada, Y., Maruyama, O., Nakai, K., Miyano, S., 2002. Extensive feature detection of N-terminal protein sorting signals. Bioinformatics 18, 298e305. Bilyeu, K.D., Cole, J.L., Laskey, J.G., Riekhof, W.R., Esparza, T.J., Kramer, M.D., Morris, R.O., 2001. Molecular and biochemical characterization of a cytokinin oxidase from maize. Plant Physiol. 125, 378e386. Brugiere, N., Jiao, S.P., Hantke, S., Zinselmeier, C., Roessler, J.A., Niu, X.M., Jones, R.J., Habben, J.E., 2003. Cytokinin oxidase gene expression in maize is localized to the vasculature, and is induced by cytokinins, abscisic acid, and abiotic stress. Plant Physiol. 132, 1228e1240. Cheikh, N., Jones, R.J., 1994. Disruption of maize kernel growth and development by heat stress. Role of cytokinin/abscisic acid balance. Plant Physiol. 106, 45e51.

D. Zalabák et al. / Plant Physiology and Biochemistry 74 (2014) 283e293 Dietrich, J.T., Kamínek, M., Blevins, D.G., Reinbott, T.M., Morris, R.O., 1995. Changes in cytokinins and cytokinin oxidase activity in developing maize kernels and the effects of exogenous cytokinin on kernel development. Plant. Physiol. Biochem. 33, 327e336.
Frébort, I., Sebela, M., Galuszka, P., Werner, T., Schmülling, T., Pe
c, P., 2002. Cytokinin oxidase/dehydrogenase assay: optimized procedures and applications. Anal. Biochem. 306, 1e7. Frébort, I., Kowalska, M., Hluska, T., Frébortová, J., Galuszka, P., 2011. Evolution of cytokinin biosynthesis and degradation. J. Exp. Bot. 62, 2431e2452. Frébortová, J., Novák, O., Frébort, I., Jorda, R., 2010. Degradation of cytokinins by maize cytokinin dehydrogenase is mediated by free radicals generated by enzymatic oxidation of natural benzoxazinones. Plant J. 61, 467e481. Gajdo
sová, S., Spíchal, L., Kamínek, M., Hoyerová, K., Novák, O., Dobrev, P.I.,

Galuszka, P., Klíma, P., Gaudinová, A., Zi zková, E., Hanu
s, J., Dan
cák, M.,
ková, R., Strnad, M., Motyka, V., 2011. Trávní
cek, B., Pe
sek, B., Krupi
cka, M., Van Distribution, biological activities, metabolism, and the conceivable function of cis-zeatin-type cytokinins in plants. J. Exp. Bot. 62, 2827e2840.
Galuszka, P., Frébort, I., Sebela, M., Sauer, P., Jacobsen, S., Pe
c, P., 2001. Cytokinin oxidase or dehydrogenase? Mechanism of cytokinin degradation in cereals. Eur. J. Biochem. 268, 450e461. Galuszka, P., Frébortová, J., Werner, T., Yamada, M., Strnad, M., Schmülling, T., Frébort, I., 2004. Cytokinin oxidase/dehydrogenase genes in barley and wheat: cloning and heterologous expression. Eur. J. Biochem. 271, 3990e4002. Galuszka, P., Frébortová, J., Luhová, L., Bilyeu, K.D., English, J.T., Frébort, I., 2005. Tissue localization of cytokinin dehydrogenase in maize: possible involvement of quinone species generated from plant phenolics by other enzymatic systems in the catalytic reaction. Plant Cell Physiol. 46, 716e728. Galuszka, P., Popelková, H., Werner, T., Frébortová, J., Pospí
silová, H., Mik, V., Köllmer, I., Schmülling, T., Frébort, I., 2007. Biochemical characterization and histochemical localization of cytokinin oxidases/dehydrogenases from Arabidopsis thaliana expressed in Nicotiana tabaccum L. J. Plant Growth Regul. 26, 255e267.
Gaudinová, A., Dobrev, P.I., Solcová, B., Novák, O., Strnad, M., Friedecký, D., Motyka, V., 2005. The involvement of cytokinin oxidase/dehydrogenase and zeatin reductase in regulation of cytokinin levels in pea (Pisum sativum L.) leaves. J. Plant Growth Regul. 24, 188e200. Gaut, B.S., 2001. Patterns of chromosomal duplication in maize and their implications for comparative maps of the grasses. Genome Res. 11, 55e66. Gu, R., Fu, J., Guo, S., Duan, F., Wang, Z., Mi, G., Yuan, L., 2010. Comparative expression and phylogenetic analysis of maize cytokinin dehydrogenase/oxidase (CKX) gene family. J. Plant Growth Regul. 29, 428e440. Kakimoto, T., 2001. Identification of plant cytokinin biosynthetic enzymes as dimethylallyl diphosphate: ATP/ADP isopentenyltransferases. Plant Cell Physiol. 42, 677e685.
Kowalska, M., Galuszka, P., Frébortová, J., Sebela, M., Béres, T., Hluska, T.,
Smehilová, M., Bilyeu, K.D., Frébort, I., 2010. Vacuolar and cytosolic cytokinin dehydrogenases of Arabidopsis thaliana: heterologous expression, purification and properties. Phytochemistry 71, 1970e1978. 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. Laskey, J.G., Patterson, P., Bilyeu, K., Morris, R.O., 2003. Rate enhancement of cytokinin oxidase/dehydrogenase using 2,6-dichloroindophenol as an electron acceptor. Plant Growth Regul. 40, 189e196. Letham, D.S., Palni, L.M.S., Tao, G.-Q., Gollnow, B.I., Bates, C.M., 1983. Regulators of cell division in plant tissues XXIX. The activities of cytokinin glucosides and alanine conjugates in cytokinin bioassays. J. Plant Growth Regul. 2, 103e115. Lomin, S.N., Yonekura-Sakakibara, K., Romanov, G.A., Sakakibara, H., 2011. Ligandbinding properties and subcellular localization of maize cytokinin receptors. J. Exp. Bot. 62, 5149e5159. Malito, E., Coda, A., Bilyeu, K.D., Fraaije, M.W., Mattevi, A., 2004. Structures of Michaelis and product complexes of plant cytokinin dehydrogenase: implications for flavoenzyme catalysis. J. Mol. Biol. 341, 1237e1249. Mameaux, S., Cockram, J., Thiel, T., Steuernagel, B., Stein, N., Taudien, S., Jack, P., Werner, P., Gray, J.C., Greenland, A.J., Powell, W., 2012. Molecular, phylogenetic and comparative genomic analysis of the cytokinin oxidase/dehydrogenase gene family in the Poaceae. Plant Biotechnol. J. 10, 67e82. Martin, R.C., Mok, M.C., Shaw, G., Mok, D.W.S., 1989. An enzyme mediating the conversion of zeatin to dihydrozeatin in phaseolus embryos. Plant Physiol. 90, 1630e1635. Massonneau, A., Houba-Hérin, N., Pethe, C., Madzak, C., Falque, M., Mercy, M., Kope
cný, D., Majira, A., Rogowsky, P., Laloue, M., 2004. Maize cytokinin oxidase genes: differential expression and cloning of two new cDNAs. J. Exp. Bot. 55, 2549e2557.

293

Mok, M.C., Mok, D.W.S., Armstrong, D.J., 1978. Differential cytokinin structureactivity relationships in Phaseolus. Plant Physiol. 61, 72e75. Nakamura, K., Matsuoka, K., 1993. Protein targeting to the vacuole in plant cells. Plant Physiol. 101, 1e5. Pa
ces, V., Werstiuk, E., Hall, R.H., 1971. Conversion of N6-(D2-Isopentenyl)adenosine to adenosine by enzyme activity in tobacco tissue. Plant Physiol. 48, 775e778. Pertry, I., Václavíková, K., Depuydt, S., Galuszka, P., Spíchal, L., Temmerman, W., Stes, E., Schmülling, T., Kakimoto, T., Van Montagu, M.C.E., Strnad, M., Holsters, M., Tarkowski, P., Vereecke, D., 2009. Identification of Rhodococcus fascians cytokinins and their modus operandi to reshape the plant. Proc. Natl. Acad. Sci. U. S. A. 106, 929e934. Petersen, T.N., Brunak, S., von Heijne, G., Nielsen, H., 2011. SignalP 4.0: discriminating signal peptides from transmembrane regions. Nat. Methods 8, 785e786. Popelková, H., Galuszka, P., Frébortová, J., Bilyeu, K.D., Frébort, I., 2004. Cytokinin dehydrogenase: characterization and structure homology modeling of the flavoprotein catabolizing plant hormones cytokinins. In: Pandalai, S.G. (Ed.), Recent Research Developments in Proteins, vol. 2Transworld Research Network, Kerala, pp. 63e81. Popelková, H., Fraaije, M.W., Novák, O., Frébortová, J., Bilyeu, K.D., Frébort, I., 2006. Kinetic and chemical analyses of the cytokinin oxidase/dehydrogenase catalysed reaction: correlations with the crystal structure. Biochem. J. 398, 113e124. Saitou, N., Nei, M., 1987. The neighbor-joining method: a new method for reconstructing phylogenetic trees. Mol. Biol. Evol. 4, 406e425. Schmülling, T., Werner, T., Riefler, M., Krupková, E., Bartrina y Manns, I., 2003. Structure and function of cytokinin oxidase/dehydrogenase genes of maize, rice, Arabidopsis and other species. J. Plant Res. 116, 241e252. Spíchal, L., Rakova, N.Y., Riefler, M., Mizuno, T., Romanov, G.A., Strnad, M., Schmülling, T., 2004. Two cytokinin receptors of Arabidopsis thaliana, CRE1/ AHK4 and AHK3, differ in their ligand specificity in a bacterial assay. Plant Cell Physiol. 45, 1299e1305. Stirk, W.A., Václavíková, K., Novák, O., Gajdo
sová, S., Kotland, O., Motyka, V., Strnad, M., van Staden, J., 2012. Involvement of cis-zeatin, dihydrozeatin, and aromatic cytokinins in germination and seedling establishment of maize, oats, and lucerne. J. Plant Growth Regul. 31, 392e405. Stolz, A., Riefler, M., Lomin, S.N., Achazi, K., Romanov, G.A., Schmülling, T., 2011. The specificity of cytokinin signalling in Arabidopsis thaliana is mediated by differing ligand affinities and expression profiles of the receptors. Plant J. 67, 157e168.

Smehilová, M., Galuszka, P., Bileyu, K.D., Jaworek, P., Kowalska, M., Sebela, M., Sedlá
rová, M., English, J.T., Frébort, I., 2009. Subcellular localization and biochemical comparison of cytosolic and secreted cytokinin dehydrogenase enzymes from maize. J. Exp. Bot. 6, 2701e2712. Takei, K., Sakakibara, H., Sugiyama, T., 2001. Identification of genes encoding adenylate isopentenyltransferase, a cytokinin biosynthesis enzyme, in Arabidopsis thaliana. J. Biol. Chem. 276, 26405e26410. Takei, K., Yamaya, T., Sakakibara, H., 2004. Arabidopsis CYP735A1 and CYP735A2 encode cytokinin hydroxylases that catalyze the biosynthesis of trans-zeatin. J. Biol. Chem. 279, 41866e41872. Tamura, K., Peterson, D., Peterson, N., Stecher, G., Nei, M., Kumar, S., 2011. MEGA5: molecular evolutionary genetics analysis using maximum likelihood, evolutionary distance, and maximum parsimony methods. Mol. Biol. Evol. 28, 2731e 2739. Vitale, A., Raikhel, N.V., 1999. What do proteins need to reach different vacuoles? Trends Plant Sci. 4, 149e155.
Václavíková, K., Ture

Vyroubalová, S., cková, V., Novák, O., Smehilová, M., Hluska, T., Ohnoutková, L., Frébort, I., Galuszka, P., 2009. Characterization of new maize genes putatively involved in cytokinin metabolism and their expression during osmotic stress in relation to cytokinin levels. Plant Physiol. 151, 433e447. Werner, T., Motyka, V., Laucou, V., Smets, R., van Onckelen, H., 2003. Cytokinin deficient transgenic Arabidopsis plants show multiple developmental alterations indicating opposite functions of cytokinins in the regulation of shoot and root meristem activity. Plant Cell 15, 1e20. Zou, C., Lehti-Shiu, M.D., Thibaud-Nissen, F., Prakash, T., Buell, C.R., Shiu, S., 2009. Evolutionary and expression signatures of pseudogenes in Arabidopsis and rice. Plant Physiol. 151, 3e15.




sil, D., Novák, O., Zdárská, M., Zatloukalová, P., Keinrad, M.B., Sedo, O., Pote Sva
cinová, J., Pe
sek, B., Malbeck, J., Va
sí
cková, J., Zdráhal, Z., Hejátko, J., 2013. Proteome analysis in Arabidopsis reveals shoot- and root-specific targets of cytokinin action and differential regulation of hormonal homeostasis. Plant Physiol. 161, 918e930.