Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

trans-Zeatin-N-glucosides have biological activity in Arabidopsis thaliana

  • H. Tucker Hallmark,

    Roles Conceptualization, Formal analysis, Methodology, Visualization, Writing – original draft, Writing – review & editing

    Affiliation Department of Biological Sciences, Auburn University, Auburn, AL, United States of America

  • Martin Černý,

    Roles Formal analysis, Methodology, Writing – original draft

    Affiliation Department of Molecular Biology and Radiobiology, Faculty of AgriSciences, Mendel University in Brno, Brno, Czech Republic

  • Břetislav Brzobohatý,

    Roles Funding acquisition, Supervision

    Affiliation Department of Molecular Biology and Radiobiology, Faculty of AgriSciences, Mendel University in Brno, Brno, Czech Republic

  • Aaron M. Rashotte

    Roles Conceptualization, Funding acquisition, Resources, Supervision, Writing – original draft, Writing – review & editing

    rashotte@auburn.edu

    Affiliations Department of Biological Sciences, Auburn University, Auburn, AL, United States of America, Department of Molecular Biology and Radiobiology, Faculty of AgriSciences, Mendel University in Brno, Brno, Czech Republic

Abstract

Cytokinin is an indispensable phytohormone responsible for physiological processes ranging from root development to leaf senescence. The term “cytokinin” refers to several dozen adenine-derived compounds occurring naturally in plants. Cytokinins (CKs) can be divided into various classes and forms; base forms are generally considered to be active while highly abundant cytokinin-N-glucosides (CKNGs), composed of a CK base irreversibly conjugated to a glucose molecule, are considered inactive. However, results from early CK studies suggest CKNGs do not always lack activity despite the perpetuation over several decades in the literature that they are inactive. Here we show that exogenous application of trans-Zeatin-N-glucosides (tZNGs, a specific class of CKNGs) to Arabidopsis results in CK response comparable to the application of an active CK base. These results are most apparent in senescence assays where both a CK base (tZ) and tZNGs (tZ7G, tZ9G) delay senescence in cotyledons. Further experiments involving root growth and shoot regeneration revealed tZNGs do not always have the same effects as tZ, and have largely distinct effects on the transcriptome and proteome. These data are in contrast to previous reports of CKNGs being inactive and raise questions about the function of these compounds as well as their mechanism of action.

Introduction

Cytokinin (CK) is an adenine derivative which acts as a phytohormone and has roles in a variety of developmental processes, including shoot growth, senescence, and root growth [1]. However, the word “cytokinin” is an umbrella term which includes dozens of compounds naturally occurring in plants [2]. In Arabidopsis thaliana (hereafter referred to as Arabidopsis), abundant isoprenoid CKs fall into one of four classes: trans-Zeatin (tZ), isopentenyladenine (iP), cis-Zeatin (cZ), or dihydrozeatin (DZ) These four classes of CK are distinguished by their N6 side chains [2]. The side chains of tZ, cZ, and DZ are hydroxylated forms of the iP side chain; tZ and cZ differ from one another in their side chain stereochemistry, and DZ lacks a double bond in its side chain [2].

Within each class of CKs, several forms exist. For example, tZ exists in a ribophosphate form (tZRP), a riboside form (tZR), a base form (tZ), an O-glucoside form (tZOG), and N-glucoside forms (tZ7G, tZ9G) [2]. Each form of CKs has a different function: generally speaking, the ribophosphate form is a precursor to bioactive CKs [2], the riboside form is a transported form with some bioactivity [3,4], the base form is the canonical active form [35], the O-glucoside form serves as an inactive but convertible storage form [69], and N-glucosides have historically been referred to as being permanently inactive and irreversibly conjugated to a glucose molecule [2,7,1012]. One exception to this is kinetin-N3-glucoside, which has been demonstrated to be reversibly conjugated [6]. Though these compounds are generally present in minuscule amounts, recent advances have allowed researchers to accurately measure CKs, and many studies in Arabidopsis have revealed that the majority of whole-plant CK is in the N-glucoside form [1319].

Early work on cytokinin-N-glucosides (CKNG) revealed mixed results about the activity of these molecules; some bioassays performed using soybean suggested these compounds lacked activity, while other studies in Raphanus and Amaranthus suggested otherwise [2022]. Perhaps the most insightful finding from these studies was that CKNGs have high metabolic stability and do not seem to be converted to bases, ribosides, or other CK forms typically considered to have biological activity [2123].

In recent years, two genes from Arabidopsis, UGT76C2 and UGT76C1, have been identified as being glucosyltransferases responsible for conversion of CK bases to CKNGs [24]. Mutation of UGT76C2 leads to a significant reduction in CKNG, though they are not fully abolished, and plants harboring this mutation have increased sensitivity to the CK benzyladenine (BA), decreased seed size, and sensitivity to drought [25,26].

Recently, a metabolic study suggested that tZNGs are converted to active tZ in Arabidopsis within minutes of exogenous application [27] but it is important to note the fact that Hošek et al. used Arabidopsis cell lines for some of their work. CKNGs are largely localized to the extracellular space [28], so using a liquid-grown cell line may have significant impacts on CK homeostasis.

In this study, we examined the biological effects of trans-Zeatin-7-glucoside (tZ7G) and trans-Zeatin-9-glucoside (tZ9G), referred to collectively as tZNGs. We show that exogenous application of two highly abundant tZNGs can mimic exogenous application of a canonically active CK base, tZ, in senescence assays, where tZ7G and tZ9G delay senescence in detached leaves. However, these compounds do not appear to function as active CK in the inhibition of root growth or as a major factor in shoot regeneration like tZ. Interestingly, tZ7G and tZ9G appear to have many distinct transcriptional and proteomic regulatory targets separate from tZ. Since CKNGs have been previously shown to be non-convertible to CK base forms, this would indicate a different mechanism is responsible for the activity of tZ7G and tZ9G in CK bioassays. Together, these data suggest that tZNGs display context-dependent CK bioactivity, as well as having their own effects on gene expression.

Materials and methods

Plant material and growth conditions

Unless otherwise stated, all plants were wildtype Col-0 Arabidopsis. Seeds were surface sterilized using 70% ethanol and 20% bleach with Tween prior to plating on full strength MS agar plates. Seeds were stratified for 2–4 days before being moved into the growth chamber, which uses a 16h/8h light cycle (100μE) with temperatures of 22°C/18°C. Mutant cypDM seed was provided by RIKEN BRC through the National BioResource Project of the MEXT/AMED, Japan.

Cotyledon senescence assay

The cotyledon senescence assays were modified for use with Arabidopsis from previously established CK bioassays [29,30] and were performed at least three times. Cotyledons (15 per treatment) of wildtype Col-0 Arabidopsis seedlings were excised 12 day after germination at the petiole and floated on 3mM MES buffer at pH 5.7. Cytokinin (tZ, tZ7G, tZ9G) was added to a final concentration of 1μM (with the exception of in the dose curve experiments in which the indicated concentration was used) or an equivalent volume of DMSO was added as a solvent control. CKNGs used in this study were obtained from OlChemIm (Olomouc, Czech Republic) as analytical standards with >95% purity guaranteed; no CK bases contaminated these standards. Previous work with CKNGs obtained from OlChemIm has revealed there is no conversion to active CKs during chlorophyll retention assays [31]. Once in solution, the cotyledons were placed in the dark at 20°C for six days, at which point chlorophyll was extracted and quantified according to a previously established protocol [32]. Briefly, cotyledons were transferred to twenty volumes of methanol and were placed at 4°C overnight. The next morning, 100μL of each solution was added to 900μL methanol and absorbance was measured at A652 and A665 to quantify chlorophyll concentration.

Root growth inhibition assay

Wildtype Col-0 Arabidopsis seedlings were germinated on full strength MS agar. Four days after germination, seedlings of uniform size were transferred to full strength MS plates supplemented with 1μM cytokinin (tZ, tZ7G, tZ9G) or an equivalent volume (0.1%) DMSO. The transferred seedlings were allowed to continue to grow under standard conditions until day 9. Root growth from day 4 to day 9 was measured using ImageJ (NIH). Three biological replicates consisting of at least ten plants per treatment were performed.

Shoot regeneration assay

Shoot regeneration assay was completed similarly to previous assays [33]. Wildtype Col-0 Arabidopsis seedlings were grown for five days in the dark for the seedlings to become etiolated. The hypocotyl was excised, avoiding tissue within 2mm of the root-shoot junction and cotyledons. Hypocotyls were then placed on full strength MS agar supplemented with 5μM cytokinin (tZ, tZ7G, tZ9G) or an equivalent volume of DMSO (0.5%) and 5μM NAA. Development was tracked over 45 days and final callus masses were measured. Three biological replicates consisting of at least four calli per treatment were performed.

Transcriptomic analysis

Whole Col-0 Arabidopsis seedlings, ten days after germination, were transferred from MS agar plates to petri dishes containing 3mM MES buffer at pH 5.7 supplemented with either 1μM cytokinin (tZ, tZ7G, tZ9G) or an equivalent volume of DMSO (0.1%). Plant were gently shaken during a two-hour treatment before seedlings were removed and snap-frozen in liquid nitrogen. RNA was extracted using a Qiagen RNeasy Plant Mini Kit, according to manufacturer’s instructions. RNA was sent to Novogene, Inc. for quality check, library preparation, and 150x paired end sequencing on an Illumina HiSeq X. At least 20 million paired end reads were generated per sample. Each treatment had three biological replicates with at least 5 seedlings per treatment. Read quality was evaluated with FastQC [34]. Trimming was performed using Trimmomatic [35]. Read mapping, transcript assembly, and differential expression analysis was performed using HISAT2, StringTie, and DESeq2 [36,37]. The significance threshold was padj < 0.05. GO analysis was performed using AgriGO 2.0 [38]. Raw sequence data is available for download at NCBI Sequence Read Archive under the BioProject ID PRJNA588257. qRT-PCR confirmation of RNAseq data was completed using SYBR-green and sequence-specific primers on an Eppendorf Realplex2 as previously described [39]. Subcellular localization analysis was performed by running DEG lists through SUBA4 and using the SUBA consensus location [40]. Full transcriptome results and primer sequences can be found in Supplementary Information.

Protein extraction and LC-MS analysis

Plants for proteomic analysis were grown and treated in a manner similar to that done for transcriptomic analysis, then flash frozen in liquid analysis for further sample preparation as detailed above. Each treatment had four biological replicates. Total protein extracts were prepared as described previously [41] with some modifications. Briefly, 100 mg of homogenized tissue was precipitated with 1 ml of methanol/methyl tert-butyl ether/water (1:3:1), pellets were solubilized [2% (w/v) SDS, 30% (w/v) sucrose, 5% (v/v) β-mercaptoethanol, 5 mM EDTA, 100 mM Tris, pH 8.0], proteins were extracted with phenol, precipitated, and the resulting pellets were solubilized. Next, aliquots corresponding to 100 μg of protein were reduced, alkylated with iodoacetamide, digested with trypsin (1:100, Promega) and desalted by C18 SPE. Finally, aliquots corresponding to 2.5 μg of peptide were analyzed by nanoflow C18 reverse-phase liquid chromatography using a 15 cm column (Zorbax, Agilent), a Dionex Ultimate 3000 RSLC nano-UPLC system (Thermo) and the Orbitrap Fusion Lumos Tribrid Mass Spectrometer. Peptides were eluted with up to a 120-min, 4% to 40% acetonitrile gradient. Spectra were acquired using the default settings for peptide identification, employing HCD activation, resolution 60000 (MS) and 15000 (MS2), and 60 s dynamic exclusion. The measured spectra were recalibrated and searched against Araport 11 protein database by Proteome Discoverer 2.3, employing Sequest HT, Mascot 2.4 and MS Amanda 2.0 with the following parameters: Enzyme—trypsin, max two missed cleavage sites; MS1 tolerance—5 ppm; MS2 tolerance—15 ppm (MS Amanda), 0.1 Da (Sequest, Mascot); Modifications—carbamidomethyl (Cys) and up to three dynamic modifications including Met oxidation, Asn/Gln deamidation, N-terminal acetylation. Only proteins with at least two unique peptides were considered for the quantitative analysis. The quantitative differences were determined by Minora, employing precursor ion quantification followed by normalization and background based t-test, and the resulting data were evaluated in Skyline 19.1 (MacCossLab Software, https://skyline.gs.washington.edu).The mass spectrometry proteomics data have been deposited to the ProteomeXchange Consortium (http://proteomecentral.proteomexchange.org) via the PRIDE partner repository (Vizcanzio et al., 2016) with the dataset identifier PXD016087.

Results

trans-Zeatin-N-glucosides delay senescence in detached cotyledons

Active cytokinins are known to delay senescence in detached cotyledons [21,29,42,43]. Chlorophyll content can be a proxy to determine if a leaf has senesced or if senescence has been delayed because chlorophyll degradation occurs during senescence [44]. In this assay, cotyledons from 12 day after germination (12dag) seedlings were floated in buffer supplemented with 1μM tZ, tZ7G, or tZ9G. Hormones were dissolved in DMSO (0.1% DMSO treatment was included as a solvent control). The treated cotyledons were placed in the dark for six days to allow senescence to take place. As expected, cotyledons treated with the known active cytokinin tZ retained significantly more chlorophyll than the solvent control (Fig 1). Interestingly, both tZ7G- and tZ9G-treated samples also retained similar levels of chlorophyll (Fig 1). To test if this effect was dose dependent, a range of hormone concentrations were tested, starting at 10pM (Fig 2). Both tZ7G and tZ9G retained more chlorophyll than the DMSO control even at 1nM concentration (Fig 2). This indicates that tZ7G and tZ9G are effective in this bioassay at similar levels as tZ.

thumbnail
Fig 1. trans-Zeatin-N-glucosides delay chlorophyll degradation in detached cotyledons.

Cotyledons from 12dag WT Arabidopsis seedlings were excised and floated abaxial side down in 3mM MES buffer (pH 5.7) supplemented with 1μM of the indicated hormone or 0.1% DMSO as a control. Solutions were placed in the dark for six days, after which chlorophyll was extracted and quantified according to Sumanta et al., 2014. Average ± SE of three biological replicates is presented. * p-value < 0.05, ** p-value < 0.01, paired two-tailed Student’s t test.

https://doi.org/10.1371/journal.pone.0232762.g001

thumbnail
Fig 2. trans-Zeatin-N-glucosides delay chlorophyll degradation in detached cotyledons in a dose-dependent manner.

Experiment performed identically to that in Fig 1 but with different concentrations of hormones. Average ± SE of three biological replicates is presented as a normalized level of chlorophyll relative to the chlorophyll content in the DMSO control cotyledons. Significance was determined by a two-way ANOVA (treatment factor, p-value = 0.0005) followed by post-hoc t-tests comparing hormone treatments to DMSO control. * p-value < 0.05, ** p-value < 0.01. Significance asterisk colors correspond to the line of the same color.

https://doi.org/10.1371/journal.pone.0232762.g002

trans-Zeatin-N-glucosides do not affect root elongation

Active cytokinins are known to inhibit root elongation [45,46]. Here, seedlings were germinated on standard media and transferred to the same media supplemented with 1μM tZ, tZ7G, tZ9G, or a 0.1% DMSO solvent control. Root growth was measured from day 4, when the seedlings were transferred to the new media, until day 9. The tZ-treated seedlings had significantly inhibited root growth relative to the DMSO control, whereas the tZNG-treated seedlings saw no difference from the control at the tested concentration (Fig 3). This suggests tZ7G and tZ9G do not significantly influence root elongation in seedlings of this stage.

thumbnail
Fig 3. trans-Zeatin-N-glucosides do not inhibit root growth.

Arabidopsis seedlings were transferred from standard MS + 1% sucrose at 4dag to the same media supplemented with 1μM of the indicated hormone or 0.1% DMSO as a control. Root growth since being transferred to the new media was measured on day 9. Average ± SE of three biological replicates is presented (n > 10 per treatment per replicate). *** p-value < 0.001, Student’s two-tailed t test.

https://doi.org/10.1371/journal.pone.0232762.g003

trans-Zeatin-7-glucoside modestly influences shoot regeneration

A classical test of cytokinin activity is the shoot regeneration assay in which hypocotyls are exposed to cytokinin and auxin. If the cytokinin is “active” and present in proper concentration, then a callus should develop, begin to turn green, and produce shoot-like structures such as leaves [47]. In this assay, hypocotyls were excised from the etiolated Arabidopsis seedlings and were subjected to a treatment with the synthetic auxin NAA and an equal amount of tZ, tZ7G, tZ9G, or DMSO as a negative control. Hormone concentrations used in this experiment were chosen based on Rashotte et al., 2006. To quantify shoot regeneration, callus weight was determined after 45 days on the hormone-containing media. Samples treated with NAA and tZ produced significantly higher callus weight than those with NAA and DMSO (Fig 4). Though the difference is modest, samples treated with NAA and tZ7G also had a slight but statistically significant increase in weight, relative to the control treatment. No significant change was seen in tZ9G samples.

thumbnail
Fig 4. Trans-Zeatin-7-glucoside mildly promotes shoot initiation.

Hypocotyls from etiolated Arabidopsis seedlings were transferred to MS + 1% sucrose supplemented with equimolar amounts of NAA and the indicated cytokinins. The negative controls were supplemented with an equal volume of DMSO in lieu of cytokinin. After 45 days, callus weight was measured. Average ± SE of three biological replicates is shown (n ≥ 4 per treatment per replicate). *** p-value < 0.001, * p-value < 0.05, Student’s two-tailed t test.

https://doi.org/10.1371/journal.pone.0232762.g004

trans-Zeatin-N-glucosides alter the transcriptome distinctly from trans-Zeatin

To determine if tZNGs affect the transcriptome of Arabidopsis, RNA sequencing was performed. Ten-day old seedlings were floated in buffer supplemented with 1μM tZ, tZ7G, tZ9G, or a 0.1% DMSO control treatment for two hours. After the two-hour treatment, tissue was flash frozen in liquid nitrogen and RNA was extracted and sequenced. The sequencing data was mapped to the Arabidopsis reference genome, and transcript level differential expression analysis was performed to determine Differentially Expressed Gene (DEG) lists for comparison. Each treatment group had hundreds of differentially expressed transcripts, but overlap between treatments was minimal (Fig 5A). Treatment with tZ, tZ7G, and tZ9G led to 340, 190, and 216 transcripts being uniquely regulated, respectively (Fig 5A). The regulation of cytokinin-related DEGs are presented in Table 1, revealing tZ to induce most of these. To validate the overall findings of the RNAseq analysis, qRT-PCR was performed and results generally mirrored transcriptome findings (Fig 5C). These results suggest tZNGs do not tend to regulate CK-related genes in a similar manner to tZ, though there are some notable exceptions such as CKX4 (Table 1, Fig 5C).

thumbnail
Fig 5. trans-Zeatin-N-glucosides alter the Arabidopsis transcriptome distinctly from trans-Zeatin.

A, Venn diagram showing number of differentially expressed transcripts after two-hour treatment of 10dag WT Arabidopsis seedlings with 1μM hormone. B, tZ-treated seedlings are enriched for CK-related Gene Ontology terms, unlike tZ7G- and tZ9G-treated seedlings. C, qRT-PCR confirmation of RNAseq results; data presented is average ± SE of three independent biological replicates. Treatment with tZ7G or tZ9G does not mimic treatment with tZ with regards to regulation of most tested CK-related genes.

https://doi.org/10.1371/journal.pone.0232762.g005

thumbnail
Table 1. List of select differentially expressed CK-related transcripts after two-hour treatment with tZ, tZ7G, and tZ9G relative to the DMSO control.

Values in bold indicate a significant difference (adjusted p-value < 0.05) between the CK-treated and control-treated samples. NA indicates read counts that were too low to establish statistical significance.

https://doi.org/10.1371/journal.pone.0232762.t001

To better understand the generated gene lists, Gene Ontology (GO) term enrichment was performed using agriGO v2.0 [38]. As expected, the tZ-treated group had several GO terms associated with cytokinin (Fig 5B). The tZ7G and tZ9G groups did not have any directly cytokinin-related GO terms, but did include “Nucleobase, nucleoside, nucleotide, and nucleic acid metabolic process,” and “Adenyl nucleotide binding” (Fig 5B). It is worth noting that tZ, tZ7G, and tZ9G are all adenine-derived compounds. Although DEGs from neither tZ7G nor tZ9G were enriched for CK-related GO terms, it is of particular interest that tZ9G induced the CK receptor AHK4, specifically a transcript isoform with an extended 5’UTR, compared to the isoform induced by tZ (Table 1). Both tZNGs induced CKH1, a histidine kinase that regulates CK response [48,49], and tZ7G induced CKX4, a cytokinin degrading enzyme which has been shown to degrade CK bases and ribosides, but likely not CKNGs [50,51] (Table 1). Several other potentially interesting transcripts were found as DEGs after tZ7G and tZ9G treatments (S2 Table). These include SEN2/AtCAT3, a catalase gene connected to senescence response, which was repressed in tZ7G and tZ9G as well as in tZ; PAC (Pale Cress), a gene involved in chloroplast and leaf development, which was repressed in tZ7G and tZ9G only; and SOD1 (superoxide dismutase 1) that was induced only in tZ9G treatment (S2 Table) [5254]. While subcellular localization of DEGs appears to be quite similar between all CK treatments, there is a slight increase in plastid localization for tZNG treatment, relative to tZ treatment (S1 Fig).

trans-Zeatin-N-glucosides alter the proteome distinctly from trans-Zeatin

In order to determine if tZNGs might be connected to gene regulation beyond the transcriptional level, a proteomic analysis was performed. Seedlings were treated with tZ, tZ7G, tZ9G, or a DMSO control in the same manner as for the transcriptome analysis, then frozen in liquid nitrogen for additional preparation and examination. Once prepared, 2119 protein families were identified, and the relative abundances of 14,412 peptides representing 1,629 proteins were quantified. The liquid chromatography/mass spectrometry (LC/MS) shotgun proteomic analysis yielded 81 proteins with a significant (P < 0.05; background based t-test) difference from DMSO-treated control samples in three biological replicates. In detail, 49, 58, and 36 proteins were identified to be regulated by tZ, tZ7G, and tZ9G, respectively (Fig 6). There was a nearly even split between induced (42) and repressed (39) proteins (S1 Table). The list of tZ-regulated proteins shares some overlap with previously noted CK-regulated proteins from other proteomic analyses (Table 2) [5560]. While there is only moderate overlap between the significantly regulated proteins between tZ7G and tZ9G (only 16 shared between the two treatments), both treatments overlap heavily with tZ (Fig 6). Each treatment had several uniquely regulated proteins.

thumbnail
Fig 6. trans-Zeatin-N-glucosides alter the Arabidopsis proteome in a partially overlapping manner with trans-Zeatin.

A, Proportional Venn diagram showing number (in parentheses) of differentially expressed proteins after two-hour treatment of 10dag WT Arabidopsis seedlings with 1μM hormone. B, Treatment with tZ7G affects amino acid metabolism, terpenoid biosynthesis, and other pathways. C, Treatment with tZ9G affects largely plastid proteins, including those involved in chloroplast biogenesis and photosystem (PS) repair.

https://doi.org/10.1371/journal.pone.0232762.g006

thumbnail
Table 2. List of select differentially expressed proteins after two-hour treatment with tZ, tZ7G, and tZ9G relative to the DMSO control.

Numbers in bold and italicized indicate a statistically significant difference in the abundance of the protein between the CK treatment and the DMSO control, background based t-test p-value < 0.05.

https://doi.org/10.1371/journal.pone.0232762.t002

Interestingly, tZ, tZ7G, and tZ9G treatment all led to the differential regulation of ACP4, ERD14, PAE2, AT3G26450, AT4G10300, which have all previously been shown to be CK responsive proteins [5558] (Table 2). The three proteins regulated by tZ, but not tZNGs, were MAPR2, AT3G15450, AT2G45180, none of which appear to be functionally characterized to date. Among the protein regulated by tZ7G are EDD1, TUB5, NADP1;1, which have been shown to be CK-regulated in previous studies [55,59,60] (Table 2). Proteins regulated by tZ9G included RPL4 and AT5G43830, which have also previously shown CK-regulation [55,57,58] (Table 2). These data indicate that while tZNGs appear to have some distinct effects on the proteome, treatment with tZNGs may affect the proteome in a manner consistent with previously tested CKs (Table 2). Novel targets of the tZNGs include multiple ribosomal proteins for tZ9G (Fig 6C) and several amino acid metabolism proteins for tZ7G (Fig 6B).

trans-Zeatin-N-glucosides do not appear to be hydrolyzed to trans-Zeatin in cotyledons

It was recently reported that within minutes of exogenous application, tZNGs were hydrolyzed to tZ base [27]. While the physiological, transcriptomic, and proteomic data above appear to refute this claim, or at least suggest tZNGs are not fully converted to tZ, we decided to also use a genetic approach to test this hypothesis. Mutants which lack functional copies of CYP735A1 and CYP735A2, the enzymes responsible for hydroxylation of iP-type cytokinins and thus production of tZ-type cytokinins, have been previously generated and described [16]. Use of these mutants in the cotyledon senescence assay, in which tZNGs display the effect most similar to that of tZ, revealed that tZNGs are not capable of delaying senescence in the cypDM mutant background, while tZ is (Fig 7). While this result has many implications, it suggests tZNGs are not converted back to tZ in an appreciable amount during this assay. Additionally, a previous study utilized tZ9G obtained from OlChemIm (same manufacturer as the CKs used in this study) to perform a chlorophyll retention assay in wheat; HPLC analysis confirmed no conversion of tZ9G was detected during the assay [31].

thumbnail
Fig 7. trans-Zeatin-N-glucosides do not delay senescence in trans-Zeatin biosynthetic mutants.

Cotyledons from 12dag cypDM mutants, which are deficient in tZ-type cytokinins, were subjected to a 6d-long cotyledon senescence assay. While exogenous tZ delayed senescence, tZNGs did not. Average ± SE of four independent biological replicates is presented. *p<0.05, Student’s two-tailed t test.

https://doi.org/10.1371/journal.pone.0232762.g007

Discussion

trans-Zeatin-N-glucosides have physiological activity in Arabidopsis

Though often over-generalized as inactivated forms of cytokinin [2,20], findings here indicate that tZNGs appear to have some cytokinin-like capabilities when exogenously applied to Arabidopsis tissues. What is particularly interesting is that the tZNGs tested do not mimic tZ in every assay (Figs 14), suggesting that while there is overlap between the activities of these compounds, they are not identical in their biological roles.

One important prerequisite to examining effects of exogenously applied tZNGs is ensuring purity of the compounds. While no synthesized compounds are 100% pure, as noted in the Methods, the tZNGs used in this study were purchased as analytical standards from OlChemIm (Olomouc, Czech Republic) and are >95% pure with no detectable tZ. Previous analysis of these standards found that no tZ was detectable in tZ9G solution before, during, or after conducting chlorophyll retention assays similar to what is described in this study [31]. This, along with the overall results of our study, strongly suggest that any effects observed were due to the exogenous application of untainted tZNGs.

Results from the senescence assay suggest exogenous application of tZ7G and tZ9G is capable of delaying senescence to a similar degree as tZ (Fig 1), and the dose curve similarly reveals it is not only at a micromolar concentration that these compounds display activity (Fig 2). The similar trend of the dose curve for tZ, tZ7G, and tZ9G may suggest the same anti-senescent pathway is being activated in each treatment, though separate mechanisms cannot be ruled out. When the same bioassay was performed using mutants which lack endogenous tZ, tZ7G and tZ9G were not capable of delaying senescence, suggesting activity of exogenously applied tZNGs relies on the presence of endogenous tZ (Fig 7).

In the root growth assay, tZ7G and tZ9G displayed effects different from that of tZ. While tZ significantly inhibited root growth in seedlings, neither tZ7G nor tZ9G altered root length (Fig 3). The difference in effects of tZ and the tZNGs suggests that conversion back to tZ, which has previously been shown to not occur [6,22], is also not a reason for the active effects seen in the senescence assay. While tZ7G and tZ9G mimic tZ during cotyledon senescence, it is possible that these compounds have no effect, or a different effect, on roots. Life stage of the plants may also play a role; the pathway in which tZ7G and tZ9G are perceived may not be active early on during seedling development.

During shoot initiation, tZ7G displayed only minor ability to increase callus mass as opposed to the robust increase in mass seen with tZ (Fig 4). Regeneration did not occur in the tZ9G treatment, nor with the negative auxin-only control. Similar to the root growth assay, these data suggest tZ, tZ7G, and tZ9G do not function identically when exogenously applied to tissue; it is possible each compound only exerts notable effects under particular conditions, i.e. during a specific developmental stage or in a particular organ or tissue.

These bioassay results are strong evidence that tZNGs are not simply converted to tZ as recently reported [27]; one would expect nearly identical results in all bioassays if this were the case. Instead, these data suggest some type of context-dependent cytokinin activity for tZNGs, though the exact nature of this is not yet understood.

trans-Zeatin-N-glucosides have differing effects on gene expression in Arabidopsis

In the transcriptomic analysis (Fig 5), tZ treatment resulted in gene expression changes quite similar to those seen in other CK transcriptome studies, as many genes from the “Golden list” were affected, and cytokinin-related GO terms were strongly enriched [61,62]. The lack of overlap between tZ with tZ7G and tZ9G, as well as the lack of overlap between the two tZNGs, suggests distinct transcriptional roles for all three compounds. There are some similar targets, however, such as CYTOKININ OXIDASE 4 (CKX4, an extracellular CK degrading enzyme) which is induced by both tZ and tZ7G (Table 1). Though it may seem that induction of CKX4 by tZ7G is possible evidence that CKX4 is capable of degrading CKNGs, previous in vitro studies suggest CKNGs are largely resistant to degradation by the AtCKX enzymes, with tZ9G only being degraded by two of seven tested CKXs under acidic conditions and at much lower rates than tZ and tZR [63]. A study in which radiolabeled tZNGs were exogenously applied to soybean tissue revealed that tZNGs do not appear to be degraded in planta [22]. Also, ARABIDOPSIS HISTIDINE KINASE 4 (AHK4, CK receptor) is regulated by both tZ and tZ9G, although the two induce different transcript isoforms which differ in their 5’UTR, but not in their protein coding region (Table 1). These data suggest some overlap between the tZ and tZNG perception and/or signaling pathways, despite the largely different transcriptional effects.

One characteristic of tZNGs that should be noted is their size. As mentioned earlier, tZNGs are composed of tZ and a glucose molecule. Therefore, tZNGs are significantly larger than tZ which is likely to impact their transportability. While few CK transporters are known and have not been studied with regards to CKNG transport [15,19,6466], it may stand to reason that tZNGs are not able to cross membranes into cells as efficiently as tZ. This may be one reason for the different transcriptional effects observed.

An interesting difference between treatments is a difference in subcellular localization of the DEGs. While the subcellular distribution of DEGs between treatments is largely the same, there is a visible increase in percentage of DEGs which localize to the plastid during tZ7G (15% of DEGs) and tZ9G (12.3%) treatment relative to tZ treatment (8.6%) (S1 Fig); this is reflected in GO Enrichment Analysis, as the term “plastid” was significantly enriched in tZ7G DEGs (FDR < 0.014) but was not significant in tZ or tZ9G DEGs. Treatment with tZ7G also has decreased extracellular-localized DEGs (3.7%) relative to treatment with tZ9G (7%) (S1 Fig).

Little overlap was seen between results of the transcriptome and proteome, though this may be unsurprising due to the short treatment time (2h), during which the compounds had to be taken up by the plants, activate signaling pathways, and activate or repress transcription before translation of induced transcripts could take place. The two hour timepoint was chosen as it has previously shown to be effective for determining CK response at the transcriptional level [61,62]. Transcriptional control is of course only one of many levels of gene expression regulation. As noted in other studies of both plant transcriptomes and proteomes, correlation between mRNA and protein levels is influenced by a large number of factors including biological reasons, such as protein turnover, as well as technical reasons, such as difficulty extracting low molecular weight and hydrophobic proteins [67,68]. Future studies will pair early transcriptomes of tZNG-treated plants with proteomes from later timepoints, possibly increasing the overlap between the two datasets, as two hours may not be enough time for the transcriptional changes observed to be observed at the protein level as well. What is striking between the transcriptome and proteome results is how neither show fully unique, nor entirely distinct, effects of the tZNGs; both sets of results show some degree of overlap as well as some unique targets (Fig 6). As noted with the subcellular localization in the transcriptome findings, there are also some differences in subcellular localization of differentially affected proteins identified as being CK-regulated; the proportion of CK-affected proteins which localize to the mitochondria is decreased relative to the total number of proteins identified in the analysis, possibly suggesting organelle-specific responses to CK treatment (S2 Fig).

Many possible mechanisms may be responsible for CKNG activity

Though CKNGs have not been shown to bind to any tested CK receptors, there may be a number of reasons for this: not all CKNGs have been tested, not all CK receptors have been tested, and the binding experiments have been carried out in microbes which may not be an ideal system for detecting these interactions [5,69,70]. These receptor binding studies have been invaluable to our understanding of CK signaling, but it is noteworthy that BA, a highly potent CK, has shown low binding potential in these assays [69,70] while showing high activity in planta. When tested, tZ7G and tZ9G showed no AHK3 or AHK4 binding but still displayed minimal activity in planta [70], perhaps pointing toward a role for AHK2 in tZNG perception. This suggests at least the possibility that CKNGs may be acting through the established CK signaling pathway despite no receptor binding assay detecting interaction. However, the lack of similar transcriptional effects between tZ, tZ7G, and tZ9G (Fig 5) may suggest tZNGs, at least at the developmental stages tested, act through different signaling pathways.

It may be possible that the mechanism of CKNG action is a yet undiscovered signaling pathway. Perhaps tZ7G and tZ9G have their own receptors separate from the canonical CK receptors (AHK2, AHK3, and AHK4), though admittedly there is no evidence supporting this hypothesis. However, other histidine kinases such as CYTOKININ INDEPENDENT 1 are known to activate the cytokinin two component signaling pathway despite not being shown to bind to the cytokinins tested [5,71,72], further suggesting our current understanding of the CK signaling pathway is incomplete.

Perhaps the most likely mechanism for the activity of CKNGs is that they act to regulate cytokinin levels. The key cytokinin-N-glucosylating enzyme UGT76C2 has been previously reported to play a role in cytokinin homeostasis [26], and it may stand to reason that exogenous application of CKNGs makes it unfavorable for UGT76C2 to convert CK bases to CKNGs, causing a buildup of CK bases which are responsible for the CK-like phenotypes observed in the senescence assays. One study utilizing an inducible cytokinin oxidase whose preferred substrate is tZ demonstrated that 6h after induction of the enzyme, tZ levels significantly decreased; however, levels of tZ7G surprisingly increased [56]. While it is not clear how increasing degradation of tZ leads to an increase of tZ7G, these data reflect the complex relationship between CK bases and CKNGs and may suggest some type of regulatory role for CKNGs. The authors hypothesize that induction of cytokinin oxidase may have led to an increase in CK biosynthesis; newly synthesized tZ would either be quickly degraded by cytokinin oxidase or converted to the degradation-resistant tZ7G form [56].This may also raise the question of whether CKNGs play some role in stimulating CK biosynthesis, which could be one explanation for the tZ-like effects seen in chlorophyll retention assays (Figs 1 and 2). One approach that could clarify the mechanism by which tZNGs exert effects would be through pairing the CK bioassays with LC-MS measurements to determine CK levels after treatment.

While it is unclear how tZNG might activate an anti-senescent pathway and lead to the delay of senescence in detached cotyledons, it should be noted that DEGs connected to senescence, chloroplast development, and oxidative stress were found to be regulated in the transcriptome analysis (S2 Table). These include some DEGs that overlap with tZ treatment like SEN2/AtCAT3 as well as unique tZNG targets such as PAC and SOD1.

Senescence assays involving the tZ-deficient cypDM mutants revealed that tZNGs are not capable of delaying senescence in cotyledons which do not produce endogenous tZ (Fig 7). This suggests that exogenous application of tZNGs alone is not sufficient to delay senescence. This may allude to the possibility that exogenous tZNG application increases endogenous levels of tZ. This possible tZ increase does not appear to be the result of hydrolysis of tZNGs to tZ base, as this would likely result in an effect similar to that of exogenous tZ application, which is not observed (Fig 7). This result also appears to provide genetic evidence countering that described in a recent metabolic study showing conversion of tZNGs to tZ [27]. It must be noted, however, that this assay only included cotyledon tissue; while it does not appear that tZNGs are hydrolyzed to tZ in cotyledons, this data cannot be generalized other organs.

It is important to note that the data presented here do not suggest that the addition of tZNGs always leads to an increase in tZ, as one would expect nearly identical results between all assays if this were the case; tZNG treatment appears to affect plants quite differently from tZ treatment in most experiments (Figs 36).

Further study of CKNG function is necessary

Whatever the mechanism, it is apparent that CKNGs merit further investigation and should not always be assumed to be inactive compounds. Past and current work in cytokinin measurement and phylogenetics has revealed CKNGs are absent or present at extremely low levels in algae and cyanobacteria [73], present at low or moderate levels in some Byrophytes [74], and present at high levels in vascular plants, suggesting an important role for these compounds in the evolution of higher plants [75]. There is almost certainly an evolutionary advantage to explain why plants expend a valuable resource like glucose in making CKNGs, but further investigation is warranted into what exactly these advantages are.

Supporting information

S1 Table. List of all proteins differentially affected by tZ, tZ7G, and tZ9G treatments.

Ratio is the relative abundance of protein in the treated sample compared to the DMSO negative control.

https://doi.org/10.1371/journal.pone.0232762.s001

(PDF)

S2 Table. trans-Zeatin-N-Glucosides regulate the transcriptome uniquely from trans-Zeatin.

All transcripts listed here are significantly (padj < 0.05) regulated by tZ7G (yellow) or tZ9G (red).

https://doi.org/10.1371/journal.pone.0232762.s002

(TIF)

S3 Table. Primer sequences used in qRT-PCR reactions.

For analysis, all genes were normalized to PDF1.

https://doi.org/10.1371/journal.pone.0232762.s003

(TIF)

S1 Fig. Subcellular distribution of DEGs is similar between tZ, tZ7G, and tZ9G treatment.

Results from the transcriptome analysis were analyzed for subcellular distribution using SUBA4. Data presented is the percent of DEGs from each treatment group which localize to the noted organelle.

https://doi.org/10.1371/journal.pone.0232762.s004

(TIF)

S2 Fig. Cytokinin treatment results in a change in localization of detectable proteins.

Results from the proteome analysis were analyzed for subcellular distribution using SUBA4. Data in blue represents the percent of differentially affected proteins showing CK responsiveness (i.e. responded to tZ, tZ7G, or tZ9G) which localize to the noted organelle. Data in black indicates the percent of all proteins identified in the analysis (i.e. both CK-responsive and non-CK-responsive) which localize to the noted organelle.

https://doi.org/10.1371/journal.pone.0232762.s005

(TIF)

S1 Data. Raw data from all presented figures are given in accordance to PLOS data policy.

https://doi.org/10.1371/journal.pone.0232762.s006

(XLSX)

Acknowledgments

We thank Lesley Turner and Valeriia Kopytko for their technical assistance throughout the project, as well as Dr. Chloe Josefson for her assistance with statistical analyses.

References

  1. 1. Kieber JJ, Schaller GE. Cytokinins. Arab Book Am Soc Plant Biol. 2014 Jan 2;12. Available from: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3894907/
  2. 2. Sakakibara H. CYTOKININS: Activity, Biosynthesis, and Translocation. Annu Rev Plant Biol. 2006;57(1):431–49.
  3. 3. Schmitz RY, Skoog F, Hecht SM, Bock RM, Leonard NJ. Comparison of cytokinin activities of naturally occurring ribonucleosides and corresponding bases. Phytochemistry. 1972 May 1;11(5):1603–10.
  4. 4. Spiess LD. Comparative Activity of Isomers of Zeatin and Ribosyl-Zeatin on Funaria hygrometrica. Plant Physiol. 1975 Mar 1;55(3):583–5. pmid:16659128
  5. 5. Yamada H, Suzuki T, Terada K, Takei K, Ishikawa K, Miwa K, et al. The Arabidopsis AHK4 Histidine Kinase is a Cytokinin-Binding Receptor that Transduces Cytokinin Signals Across the Membrane. Plant Cell Physiol. 2001 Sep 15;42(9):1017–23. pmid:11577198
  6. 6. Brzobohaty B, Moore I, Kristoffersen P, Bako L, Campos N, Schell J, et al. Release of active cytokinin by a beta-glucosidase localized to the maize root meristem. Science. 1993 Nov 12;262(5136):1051–4. pmid:8235622
  7. 7. Entsch B, Letham DS. Enzymic glucosylation of the cytokinin, 6-benzylaminopurine. Plant Sci Lett. 1979 Feb 1;14(2):205–12.
  8. 8. Falk A, Rask L. Expression of a Zeatin-O-Glucoside-Degrading [beta]-Glucosidase in Brassica napus. Plant Physiol. 1995 Aug 1;108(4):1369–77. pmid:7659745
  9. 9. Yoshida R, Oritani T. Cytokinin glucoside in roots of the rice plant. Plant Cell Physiol. 1972 Apr 1;13(2):337–43.
  10. 10. Bairu MW, Novák O, Doležal K, Van Staden J. Changes in endogenous cytokinin profiles in micropropagated Harpagophytum procumbens in relation to shoot-tip necrosis and cytokinin treatments. Plant Growth Regul. 2011 Mar 1;63(2):105–14.
  11. 11. Fox JE, Cornette J, Deleuze G, Dyson W, Giersak C, Niu P, et al. The Formation, Isolation, and Biological Activity of a Cytokinin 7-Glucoside. Plant Physiol. 1973 Dec 1;52(6):627–32. pmid:16658619
  12. 12. Spíchal L. Cytokinins–recent news and views of evolutionally old molecules. Funct Plant Biol. 2012 May 15;39(4):267–84.
  13. 13. Kiba T, Takebayashi Y, Kojima M, Sakakibara H. Sugar-induced de novo cytokinin biosynthesis contributes to Arabidopsis growth under elevated CO 2. Sci Rep. 2019 May 23;9(1):1–15.
  14. 14. Tokunaga H, Kojima M, Kuroha T, Ishida T, Sugimoto K, Kiba T, et al. Arabidopsis lonely guy (LOG) multiple mutants reveal a central role of the LOG-dependent pathway in cytokinin activation. Plant J. 2012;69(2):355–65. pmid:22059596
  15. 15. Ko D, Kang J, Kiba T, Park J, Kojima M, Do J, et al. Arabidopsis ABCG14 is essential for the root-to-shoot translocation of cytokinin. Proc Natl Acad Sci. 2014 May 13;111(19):7150–5. pmid:24778257
  16. 16. Kiba T, Takei K, Kojima M, Sakakibara H. Side-Chain Modification of Cytokinins Controls Shoot Growth in Arabidopsis. Dev Cell. 2013 Nov 25;27(4):452–61. pmid:24286826
  17. 17. Nishiyama R, Watanabe Y, Fujita Y, Le DT, Kojima M, Werner T, et al. Analysis of Cytokinin Mutants and Regulation of Cytokinin Metabolic Genes Reveals Important Regulatory Roles of Cytokinins in Drought, Salt and Abscisic Acid Responses, and Abscisic Acid Biosynthesis. Plant Cell. 2011 Jun 1;23(6):2169–83. pmid:21719693
  18. 18. Sakakibara H, Kasahara H, Ueda N, Kojima M, Takei K, Hishiyama S, et al. Agrobacterium tumefaciens increases cytokinin production in plastids by modifying the biosynthetic pathway in the host plant. Proc Natl Acad Sci. 2005 Jul 12;102(28):9972–7. pmid:15998742
  19. 19. Zhang K, Novak O, Wei Z, Gou M, Zhang X, Yu Y, et al. Arabidopsis ABCG14 protein controls the acropetal translocation of root-synthesized cytokinins. Nat Commun. 2014 Dec;5(1).
  20. 20. Letham DS, Palni LMS. The Biosynthesis and Metabolism of Cytokinins. Annu Rev Plant Physiol. 1983;34(1):163–97.
  21. 21. Letham DS, Palni LMS, Tao G-Q, Gollnow BI, Bates CM. Regulators of cell division in plant tissues XXIX. The activities of cytokinin glucosides and alanine conjugates in cytokinin bioassays. J Plant Growth Regul. 1983 Aug 1;2(1):103–15.
  22. 22. Palni LMS, Palmer MV, Letham DS. The stability and biological activity of cytokinin metabolites in soybean callus tissue. Planta. 1984 Mar 1;160(3):242–9. pmid:24258507
  23. 23. Deleuze GG, McChesney JD, Fox JE. Identification of a stable cytokinin metabolite. Biochem Biophys Res Commun. 1972 Sep 26;48(6):1426–32. pmid:5077828
  24. 24. Hou B, Lim E-K, Higgins GS, Bowles DJ. N-Glucosylation of Cytokinins by Glycosyltransferases of Arabidopsis thaliana. J Biol Chem. 2004 Nov 12;279(46):47822–32. pmid:15342621
  25. 25. Li Y, Wang B, Dong R, Hou B. AtUGT76C2, an Arabidopsis cytokinin glycosyltransferase is involved in drought stress adaptation. Plant Sci. 2015 Jul 1;236(Supplement C):157–67.
  26. 26. Wang J, Ma X-M, Kojima M, Sakakibara H, Hou B-K. N-Glucosyltransferase UGT76C2 is Involved in Cytokinin Homeostasis and Cytokinin Response in Arabidopsis thaliana. Plant Cell Physiol. 2011 Dec 1;52(12):2200–13. pmid:22051886
  27. 27. Hošek P, Hoyerová K, Kiran NS, Dobrev PI, Zahajská L, Filepová R, et al. Distinct metabolism of N-glucosides of isopentenyladenine and trans-zeatin determines cytokinin metabolic spectrum in Arabidopsis. New Phytol. 2019; Available from: https://nph.onlinelibrary.wiley.com/doi/abs/10.1111/nph.16310
  28. 28. Jiskrová E, Novák O, Pospíšilová H, Holubová K, Karády M, Galuszka P, et al. Extra- and intracellular distribution of cytokinins in the leaves of monocots and dicots. New Biotechnol. 2016 Sep;33(5):735–42.
  29. 29. Fletcher RA, McCullagh D. Cytokinin-induced chlorophyll formation in cucumber cotyledons. Planta. 1971 Mar 1;101(1):88–90. pmid:24488296
  30. 30. Letham DS. Chemistry and Physiology of Kinetin-Like Compounds. Annu Rev Plant Physiol. 1967;18(1):349–64.
  31. 31. Gajdošová S. Biological effects and metabolism of cis-Zeatin-type cytokinins in plants. Prague: Charles University; 2011. Available from: https://dspace.cuni.cz/handle/20.500.11956/47271
  32. 32. Sumanta N, Haque CI, Nishika J, Suprakash R. Spectrophotometric Analysis of Chlorophylls and Carotenoids from Commonly Grown Fern Species by Using Various Extracting Solvents. Res J Chem Sci. 2014;9(4).
  33. 33. Rashotte AM, Mason MG, Hutchison CE, Ferreira FJ, Schaller GE, Kieber JJ. A subset of Arabidopsis AP2 transcription factors mediates cytokinin responses in concert with a two-component pathway. Proc Natl Acad Sci. 2006 Jul 18;103(29):11081–5. pmid:16832061
  34. 34. Andrews S. FastQC: A quality control tool for high throughput sequence data. 2010. Available from: http://www.bioinformatics.babraham.ac.uk/projects/fastqc/
  35. 35. Bolger AM, Lohse M, Usadel B. Trimmomatic: a flexible trimmer for Illumina sequence data. Bioinformatics. 2014 Aug 1;30(15):2114–20. pmid:24695404
  36. 36. Love MI, Huber W, Anders S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol. 2014 Dec 5;15(12):550. pmid:25516281
  37. 37. Pertea M, Kim D, Pertea GM, Leek JT, Salzberg SL. Transcript-level expression analysis of RNA-seq experiments with HISAT, StringTie and Ballgown. Nat Protoc. 2016 Sep;11(9):1650–67. pmid:27560171
  38. 38. Tian T, Liu Y, Yan H, You Q, Yi X, Du Z, et al. agriGO v2.0: a GO analysis toolkit for the agricultural community, 2017 update. Nucleic Acids Res. 2017 Jul 3;45:W122–9. pmid:28472432
  39. 39. Zwack PJ, Robinson BR, Risley MG, Rashotte AM. Cytokinin Response Factor 6 Negatively Regulates Leaf Senescence and is Induced in Response to Cytokinin and Numerous Abiotic Stresses. Plant Cell Physiol. 2013 Jun 1;54(6):971–81. pmid:23539244
  40. 40. Hooper CM, Castleden IR, Tanz SK, Aryamanesh N, Millar AH. SUBA4: the interactive data analysis centre for Arabidopsis subcellular protein locations. Nucleic Acids Res. 2017 Jan 4;45(D1):D1064–74. pmid:27899614
  41. 41. Cerna H, Černý M, Habánová H, Šafářová D, Abushamsiya K, Navrátil M, et al. Proteomics offers insight to the mechanism behind Pisum sativum L. response to pea seed-borne mosaic virus (PSbMV). J Proteomics. 2017 Feb 5;153:78–88. pmid:27235724
  42. 42. Letham DS. Regulators of Cell Division in Plant Tissues XII. A Cytokinin Bioassay Using Excised Radish Cotyledons. Physiol Plant. 1971 Dec 1;25(3):391–6.
  43. 43. Zwack PJ, Rashotte AM. Cytokinin inhibition of leaf senescence. Plant Signal Behav. 2013 Jul 1;8(7). Available from: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3908980/
  44. 44. Hörtensteiner S. Chlorophyll Degradation During Senescence. Annu Rev Plant Biol. 2006;57(1):55–77.
  45. 45. Stenlid G. Cytokinins as inhibitors of root growth. Physiol Plant. 1982;56(4):500–6.
  46. 46. Werner T, Motyka V, Strnad M, Schmülling T. Regulation of plant growth by cytokinin. Proc Natl Acad Sci. 2001 Aug 28;98(18):10487–92. pmid:11504909
  47. 47. Skoog F, Miller CO. Chemical regulation of growth and organ formation in plant tissues cultured in vitro. Symp Soc Exp Biol. 1957;54:118–30.
  48. 48. Kubo M, Kakimoto T. The CYTOKININ-HYPERSENSITIVE genes of Arabidopsis negatively regulate the cytokinin-signaling pathway for cell division and chloroplast development. Plant J. 2000;23(3):385–94. pmid:10929131
  49. 49. Suzuki T, Ishikawa K, Yamashino T, Mizuno T. An Arabidopsis Histidine-Containing Phosphotransfer (HPt) Factor Implicated in Phosphorelay Signal Transduction: Overexpression of AHP2 in Plants Results in Hypersensitiveness to Cytokinin. Plant Cell Physiol. 2002 Jan 15;43(1):123–9. pmid:11828030
  50. 50. Jones RJ, Schreiber BMN. Role and function of cytokinin oxidase in plants. Plant Growth Regul. 1997 Oct 1;23(1):123–34.
  51. 51. McGaw BA, Horgan R. Cytokinin catabolism and cytokinin oxidase. Phytochemistry. 1983 Jan 1;22(5):1103–5.
  52. 52. Gepstein S, Sabehi G, Carp M-J, Hajouj T, Nesher MFO, Yariv I, et al. Large-scale identification of leaf senescence-associated genes. Plant J. 2003;36(5):629–42. pmid:14617064
  53. 53. Kliebenstein DJ, Monde R-A, Last RL. Superoxide Dismutase in Arabidopsis: An Eclectic Enzyme Family with Disparate Regulation and Protein Localization. Plant Physiol. 1998 Oct 1;118(2):637–50. pmid:9765550
  54. 54. Reiter RS, Coomber SA, Bourett TM, Bartley GE, Scolnik PA. Control of leaf and chloroplast development by the Arabidopsis gene pale cress. Plant Cell. 1994 Sep 1;6(9):1253–64. pmid:7919990
  55. 55. Černý M, Dyčka F, Bobál’ová J, Brzobohatý B. Early cytokinin response proteins and phosphoproteins of Arabidopsis thaliana identified by proteome and phosphoproteome profiling. J Exp Bot. 2011 Jan 1;62(3):921–37. pmid:20974740
  56. 56. Černý M, Kuklová A, Hoehenwarter W, Fragner L, Novák O, Rotková G, et al. Proteome and metabolome profiling of cytokinin action in Arabidopsis identifying both distinct and similar responses to cytokinin down- and up-regulation. J Exp Bot. 2013 Nov 1;64(14):4193–206. pmid:24064926
  57. 57. Černý M, Jedelský PL, Novák J, Schlosser A, Brzobohatý B. Cytokinin modulates proteomic, transcriptomic and growth responses to temperature shocks in Arabidopsis. Plant Cell Environ. 2014 Jul 1;37(7):1641–55. pmid:24393122
  58. 58. Lochmanová G, Zdráhal Z, Konečná H, Koukalová Š, Malbeck J, Souček P, et al. Cytokinin-induced photomorphogenesis in dark-grown Arabidopsis: a proteomic analysis. J Exp Bot. 2008 Oct 1;59(13):3705–19. pmid:18775952
  59. 59. Žd’árská M, Zatloukalová P, Benítez M, Šedo O, Potěšil D, Novák O, et al. Proteome Analysis in Arabidopsis Reveals Shoot- and Root-Specific Targets of Cytokinin Action and Differential Regulation of Hormonal Homeostasis. Plant Physiol. 2013 Feb 1;161(2):918–30. pmid:23209126
  60. 60. Zhang Y, Liu S, Dai SY, Yuan JS. Integration of shot-gun proteomics and bioinformatics analysis to explore plant hormone responses. BMC Bioinformatics. 2012 Sep 11;13(15):S8.
  61. 61. Bhargava A, Clabaugh I, To JP, Maxwell BB, Chiang Y-H, Schaller EG, et al. Identification of Cytokinin Responsive Genes Using Microarray Meta-analysis and RNA-seq in Arabidopsis thaliana. Plant Physiol. 2013 Jan 1;pp.113.217026.
  62. 62. Rashotte AM, Carson SDB, To JPC, Kieber JJ. Expression Profiling of Cytokinin Action in Arabidopsis. Plant Physiol. 2003 Aug 1;132(4):1998–2011. pmid:12913156
  63. 63. Galuszka P, Popelková H, Werner T, Frébortová J, Pospíšilová H, Mik V, et al. Biochemical Characterization of Cytokinin Oxidases/Dehydrogenases from Arabidopsis thaliana Expressed in Nicotiana tabacum L. J Plant Growth Regul. 2007 Sep 1;26(3):255–67.
  64. 64. Bürkle L, Cedzich A, Döpke C, Stransky H, Okumoto S, Gillissen B, et al. Transport of cytokinins mediated by purine transporters of the PUP family expressed in phloem, hydathodes, and pollen of Arabidopsis. Plant J. 2003;34(1):13–26. pmid:12662305
  65. 65. Hirose N, Makita N, Yamaya T, Sakakibara H. Functional Characterization and Expression Analysis of a Gene, OsENT2, Encoding an Equilibrative Nucleoside Transporter in Rice Suggest a Function in Cytokinin Transport. Plant Physiol. 2005 May 1;138(1):196–206. pmid:15849298
  66. 66. Sun J, Hirose N, Wang X, Wen P, Xue L, Sakakibara H, et al. Arabidopsis SOI33/AtENT8 Gene Encodes a Putative Equilibrative Nucleoside Transporter That Is Involved in Cytokinin Transport In Planta. J Integr Plant Biol. 2005;47(5):588–603.
  67. 67. Graf A, Coman D, Uhrig RG, Walsh S, Flis A, Stitt M, et al. Parallel analysis of Arabidopsis circadian clock mutants reveals different scales of transcriptome and proteome regulation. Open Biol. 7(3):160333. pmid:28250106
  68. 68. Ponnala L, Wang Y, Sun Q, Wijk KJ van. Correlation of mRNA and protein abundance in the developing maize leaf. Plant J. 2014;78(3):424–40. pmid:24547885
  69. 69. Romanov GA, Lomin SN, Schmülling T. Biochemical characteristics and ligand-binding properties of Arabidopsis cytokinin receptor AHK3 compared to CRE1/AHK4 as revealed by a direct binding assay. J Exp Bot. 2006 Dec 1;57(15):4051–8. pmid:17075078
  70. 70. Spíchal L, Rakova NY, Riefler M, Mizuno T, Romanov GA, Strnad M, et al. Two Cytokinin Receptors of Arabidopsis thaliana, CRE1/AHK4 and AHK3, Differ in their Ligand Specificity in a Bacterial Assay. Plant Cell Physiol. 2004 Sep 15;45(9):1299–305. pmid:15509853
  71. 71. Deng Y, Dong H, Mu J, Ren B, Zheng B, Ji Z, et al. Arabidopsis Histidine Kinase CKI1 Acts Upstream of HISTIDINE PHOSPHOTRANSFER PROTEINS to Regulate Female Gametophyte Development and Vegetative Growth. Plant Cell. 2010 Apr 1;22(4):1232–48. pmid:20363773
  72. 72. Kakimoto T. CKI1, a Histidine Kinase Homolog Implicated in Cytokinin Signal Transduction. Science. 1996 Nov 8;274(5289):982–5. pmid:8875940
  73. 73. Žižková E, Kubeš M, Dobrev PI, Přibyl P, Šimura J, Zahajská L, et al. Control of cytokinin and auxin homeostasis in cyanobacteria and algae. Ann Bot. 2017 Jan 1;119(1):151–66. pmid:27707748
  74. 74. Drábková LZ, Dobrev PI, Motyka V. Phytohormone Profiling across the Bryophytes. PLOS ONE. 2015 May 14;10(5):e0125411. pmid:25974061
  75. 75. Auer CA. Cytokinin conjugation: recent advances and patterns in plant evolution. Plant Growth Regul. 1997;23.