Cloning of Three Cytokinin Oxidase/Dehydrogenase Genes in Bambusa oldhamii

Cytokinin oxidase/dehydrogenase (CKX) catalyzes the irreversible breakdown of active cytokinins, which are a class of plant hormones that regulate cell division. According to conserved sequences of CKX genes from monocotyledons, PCR primers were designed to synthesize a probe for screening a bamboo genomic library. Cloned results of three genes encoding cytokinin oxidase were named as follows: BoCKX1, BoCKX2, and BoCKX3. In comparing the exon-intron structures among the above three genes, there are three exons and two introns in BoCKX1 and BoCKX3 genes, whereas BoCKX2 contains four exons and three introns. The amino acid sequence of BoCKX2 protein shares 78% and 79% identity with BoCKX1 and BoCKX3 proteins, respectively. BoCKX1 and BoCKX3 genes are particularly closely related given that the amino acid and nucleotide sequence identities are more than 90%. These three BoCKX proteins carried putative signal peptide sequences typical of secretion pathway, and a GHS-motif was found at N-terminal flavin adenine dinucleotide (FAD) binding domain, suggesting that BoCKX proteins might covalently conjugate with an FAD cofactor through a predicted histidine residue.


Introduction
Cytokinin is a phytohormone of plant growth regulators, playing an important role in the control of plant division. Many developmental events, such as shoot and root branching, leaf development, delay of senescence, and chloroplast ripening are directly regulated by the physiological functions of cytokinins [1][2][3][4][5]. In plants, the homeostasis of cytokinins is balanced in many aspects [6], including rate of de novo synthesis by isopentenyl transferase enzyme [7], rate of interconversion and transport [8], and rate of cytokinin turnover, mainly by cytokinin oxidase/dehydrogenase [9].
Elevated cytokinin levels were linked with increased cytokinin degradation by CKX enzyme in many plants, such as Zea mays [25], Triticum aestivum [26], and Brassica napus [24,27], suggesting that CKX gene expression is manipulated by endogenous cytokinin extents [5]. Overexpression of the AtCKX3 [28] and OsCKX2 [9] are associated with reduced Curr. Issues Mol. Biol. 2023, 45 1903 flower numbers and grain numbers, respectively. Expression of the AtCKX1 in Nicotiana tabacum enhances drought and heat stress tolerance, indicating that cytokinin levels may have a positive effect on plant stress responses [29][30][31]. Under salinity condition, rice yield penalty is reduced by knockdown of the OsCKX2 gene as well as increased inflorescence meristem cytokinin extent [32].
Alternation in CKX enzymatic activity changes cytokinin concentrations in cells and tissues. CKX enzymes play key roles in contributing to the regulation of cytokinin-dependent processes and in controlling local cytokinin level [33]. Functions of CKX proteins can be regulated by several post-translational modifications, e.g., glycosylation, and contain FAD as a cofactor [21,34]. CKX proteins are widely detected in various subcellular compartments, such as chloroplast, mitochondria, and so on [35]. Most CKX enzymes are localized in the apoplast, e.g., ZmCKX1 [23], or vacuole, e.g., AtCKX1 and AtCKX3 [36]. Some CKX proteins are shown to be cytosolic enzymes, such as AtCKX7 [37] and ZmCKX10 [23]. AtCKX1 protein is recently reported to be an endoplasmic reticulum (ER) membrane protein [16].
Bambusa oldhamii, green bamboo, is a perennial plant in the tropics and subtropics, and bamboo shoot is an economic vegetable in far-eastern Asia [38][39][40][41][42][43][44]. Bamboo is one of the fast-growth timber plants, which are controlled by plant hormones, especially cytokinins [45]. Bamboo cytokinin biosynthesis by the BoAIPT1 isopentenyltransferase enzyme [7] and cytokinin degradation by the BoCKX enzymes may play equally important roles for fine-tuning cytokinin levels in the rapid-dividing tissues. In this study, we cloned and reported three completely sequenced CKX genes, including all exon-intron conformations and partial upstream promoter regions. Sequence analysis of the BoCKX1-3 predicts multiple asparagine residues as putative target sites of glycosylation. In addition, several cis-acting elements were discovered from the promoters regions of three BoCKX genes.

Plant Material
Edible fresh green bamboo shoot mainly harvested between April and September from Mucha mountain areas, Taipei City, Taiwan. Samples were divided into inedible shell and edible shoot and were frozen and stored at −80 • C freezer.

Reagents
DNA ladders and iProof DNA polymerase were obtained from Bio-Rad, Hercules, CA, USA. PrimeStar DNA polymerase mixture was purchased from Takara, Kusatsu, Shiga, Japan. T4 DNA ligase and restriction endonucleases were purchased from New England Biolabs, Ipswich, MA, USA. SeaKem ® LE Agarose was obtained from Lonza, Basel, Switzerland. Gel extraction/PCR cleanup kit was purchased from Biotools, New Taipei City, Taiwan. Plasmid mini-prep kit was obtained from Geneaid, New Taipei City, Taiwan. Oligonucleotides synthesis and DNA sequencing services were provided by Tri-I Biotech, New Taipei City, Taiwan.

Total RNA Extraction, cDNA Synthesis, and DIG Labeled Probe Preparation
TRIZOL reagents (Invitrogen, Waltham, MA, USA) were used to extract total RNA from bamboo etiolated shoots and utilized as materials for complementary DNA (cDNA) synthesis using MMLV reverse transcriptase (Invitrogen, Waltham, MA, USA), and used as PCR template. Degenerate primers CKX-F (5 -GGGAGATGGTGACGTGCTCCAA-3 ) and CKX-R (5 -CAGCGACACSRMGTAGAACAC-3 ) were designed according to the conserved regions of the OsCKX1 and OsCKX2 genes from Oryza sativa [4]. PCR reaction was carried out at 94 • C for 30 s, 60 • C for 30 s, and 72 • C for 1 min, and then repeated for 30 cyclereaction. The 900 bp BoCKX fragment ( Figure S1) was confirmed by DNA sequencing (Tri-I Biotech, New Taipei City, Taiwan). DIG-labeled probe was synthesized by PCR reaction using DIG DNA labeling kit supplied by Roche, Basel, Switzerland.

Genomic Library Screening, Phage DNA Preparation, and DNA Sequencing by Chromosome Walking
A genomic library of green bamboo was previously constructed by Lambda FIX ® II/XhoI Partial Fill-In Vector Kit (Strategene, San Diego, CA, USA) [7]. Nine to twenty-three kb bamboo genomic DNA fragments were conjugated with a 41.9 kb Lambda FIX ® II vector, and the overall phage DNA constructed were between 50 and 65 kb [7]. The DIG-labeled BoCKX probe was used to screen the bamboo genomic DNA library. Plaques of phages were transferred onto a Hybond-H + hybridization membrane (MilliporeSigma, Burlington, MA, USA). The processes of hybridization and detection were based on the instructions of the manufacturer (Roche, Basel, Switzerland). Phage DNA was purified from E. coli XL1-Blue MRA (P2) by traditional method [46] and digested by NotI restriction endonuclease. Phage DNA isolated from positive clones (Lambda midi kit, Qiagene, Hilden, Germany) were further confirmed by DNA sequencing (Tri-I Biotech, New Taipei City, Taiwan), and chromosome walking method was used to obtain complete genomic sequences of the three BoCKX genes.

Cloning of Three BoCKXs Genes by Screening a Bamboo Genomic DNA Library
A Lambda FIX ® II genomic library constructed genomic DNA was hybridized and screened by a DIG labeled BoCKX probe ( Figure S1) [38]. Sequence alignment results showed a high similarity of 64% between the deduced amino acid sequence of the BoCKX probe and ZmCKX1 proteins ( Figure S1). Phage DNAs were isolated from positive clones and digested by NotI restriction enzyme ( Figure 1), followed by DNA sequencing. Three different digestion patterns from four positive clones were observed, indicating that three different BoCKX genes were identified. Chromosome walking methodology was performed to obtain the total sequence information of these positive clones. Luckily, three full-length BoCKX genes with partial promoter regions were obtained, namely BoCKX1 ( Figure 1A, lane 1), BoCKX2 ( Figure 1B, lanes 1 and 2), and BoCKX3 ( Figure 1A, lane 2). Gene names, BoCKX1-3, were designated in the order of DNA sequencing accomplished. These genomic DNA sequences had been deposited at GenBank (Bethesda, MD, USA) with accession numbers, GU263785, GU263786, and GU263787, respectively.

Genomic Organization of the Three BoCKXs Genes
Phage DNA containing bamboo CKX genes were completely sequenced by chromosome walking method, and exon-intron structures of the three BoCKX genes were plotted in Figure 2A. The intron-exon organizations of various CKX genes in plants were compared and shown in Figure 2B.

Genomic Organization of the Three BoCKXs Genes
Phage DNA containing bamboo CKX genes were completely sequenced by some walking method, and exon-intron structures of the three BoCKX genes wer in Figure 2A

Genomic Organization of the Three BoCKXs Genes
Phage DNA containing bamboo CKX genes were completely sequenced by chr some walking method, and exon-intron structures of the three BoCKX genes were pl in Figure    BoCKX1 and BoCKX3 were composed of three exons and two introns (Figure 2A,B), as also observed in other CKX genes, such as OsCKX1 [4] and ZmCKX1 [13,14]. Unlike BoCKX1 and BoCKX3, the exon 2 of BoCKX2 was divided into exons 2a and 2b (Figure 2A,B).

Protein Similarity of CKXs Proteins
Protein sequence alignment research revealed that BoCKX proteins had 71-92% identities, with BoCKX1 protein being the most similar to BoCKX3 protein (92% identity; Figure 3 and Table 1).

Protein Similarity of CKXs Proteins
Protein sequence alignment research revealed that BoCKX proteins had 71-92% identities, with BoCKX1 protein being the most similar to BoCKX3 protein (92% identity; Figure 3 and Table 1).   The deduced amino acid sequences of CKXs from several plant species underwent phylogenetic analysis ( Figure 4). All CKX protein sequences contained a conserved GHS motif as the FAD-binding site and the histidine residue within the GHS motif may be responsible for covalently conjugated with FAD cofactor [34]. OsCKX1 and ZmCKX1 proteins were the closest homolog to BoCKX1 (78% identity) and BoCKX3 (76% identity) proteins, and these genes contained similar exon-intron organization ( Figure 2B, Figure 4). Although BoCKX2 and OsCKX2 shared identical genomic organization ( Figure 2B, Figure 4), the amino acid similarity was only 52% identity. The deduced amino acid sequences of CKXs from several plant species under phylogenetic analysis ( Figure 4). All CKX protein sequences contained a conserved motif as the FAD-binding site and the histidine residue within the GHS motif ma responsible for covalently conjugated with FAD cofactor [34]. OsCKX1 and ZmCKX1 teins were the closest homolog to BoCKX1 (78% identity) and BoCKX3 (76% identity teins, and these genes contained similar exon-intron organization ( Figure 2B, Figu Although BoCKX2 and OsCKX2 shared identical genomic organization ( Figure 2B, F 4), the amino acid similarity was only 52% identity.

Analysis of the cis-Acting Elements in BoCKX1
To better understand the potential transcriptional regulations, the upstream quence of the BoCKX1 was completely sequenced and limited promoter region (-262 obtained ( Figure 5). PlantProm [50] predicted the probable transcriptional start site ( BoCKX1, which was situated 83 base pairs from the translational start codon (ATG, F 5). Promoter sequence analysis was obtained from PlantPAN [49], and a putative T box and an NGATT motif [52] were addressed at −27 and −235 positions, respectively ure 5).

Analysis of the cis-Acting Elements in BoCKX1
To better understand the potential transcriptional regulations, the upstream sequence of the BoCKX1 was completely sequenced and limited promoter region (−262) was obtained ( Figure 5). PlantProm [50] predicted the probable transcriptional start site (+1) of BoCKX1, which was situated 83 base pairs from the translational start codon (ATG, Figure 5). Promoter sequence analysis was obtained from PlantPAN [49], and a putative TATA box and an NGATT motif [52] were addressed at −27 and −235 positions, respectively ( Figure 5).

Analysis of the cis-Acting Elements in BoCKX2
To better understand the potential transcriptional regulations, the upstream sequence of the BoCKX2 was completely sequenced and over 1.4 kb promoter region was obtained ( Figure 6). PlantProm [50] predicted the probable transcriptional start site (+1) of BoCKX2, which was situated 44 base pairs from the translational start codon (ATG, Figure  6). Promoter sequence analysis was performed by PlantPAN [49], and a putative TATA box and CAAT box were addressed at −27 and −40 positions, respectively ( Figure 6). Among the promoter region, nine NGATT motifs [52] and one as-1 like conserved site (−1289) [53] were identified ( Figure 6).

Analysis of the cis-Acting Elements in BoCKX2
To better understand the potential transcriptional regulations, the upstream sequence of the BoCKX2 was completely sequenced and over 1.4 kb promoter region was obtained ( Figure 6). PlantProm [50] predicted the probable transcriptional start site (+1) of BoCKX2, which was situated 44 base pairs from the translational start codon (ATG, Figure 6). Promoter sequence analysis was performed by PlantPAN [49], and a putative TATA box and CAAT box were addressed at −27 and −40 positions, respectively ( Figure 6). Among the promoter region, nine NGATT motifs [52] and one as-1 like conserved site (−1289) [53] were identified ( Figure 6).

Analysis of the cis-Acting Elements in BoCKX2
To better understand the potential transcriptional regulations, the upstream sequence of the BoCKX2 was completely sequenced and over 1.4 kb promoter region was obtained ( Figure 6). PlantProm [50] predicted the probable transcriptional start site (+1) of BoCKX2, which was situated 44 base pairs from the translational start codon (ATG, Figure  6). Promoter sequence analysis was performed by PlantPAN [49], and a putative TATA box and CAAT box were addressed at −27 and −40 positions, respectively ( Figure 6). Among the promoter region, nine NGATT motifs [52] and one as-1 like conserved site (−1289) [53] were identified ( Figure 6). Figure 6. Promoter sequence analysis of the BoCKX2 gene. The start codon of translation was shaded. The predicted transcriptional start site was indicated as "+1". The putative functional TATA box, CAAT box, as-1 like motif and ten NGATT elements were designated. Figure 6. Promoter sequence analysis of the BoCKX2 gene. The start codon of translation was shaded. The predicted transcriptional start site was indicated as "+1". The putative functional TATA box, CAAT box, as-1 like motif and ten NGATT elements were designated.

Analysis of the cis-Acting Elements in BoCKX3
To better understand the potential transcriptional regulations, the upstream sequence of the BoCKX3 was completely sequenced and~1.2 kb promoter region was obtained (Figure 7). PlantProm [50] predicted the probable transcriptional start site (+1) of BoCKX2, which was situated 79 base pairs from the translational start codon (ATG, Figure 7). Promoter sequence analysis was obtained by PlantPAN [49], a putative TATA box was addressed at −24 position, and five NGATT conserved regions [52] were boxed (Figure 7).

Analysis of the cis-Acting Elements in BoCKX3
To better understand the potential transcriptional regulations, the upstream sequence of the BoCKX3 was completely sequenced and ~1.2 kb promoter region was obtained ( Figure 7). PlantProm [50] predicted the probable transcriptional start site (+1) of BoCKX2, which was situated 79 base pairs from the translational start codon (ATG, Figure  7). Promoter sequence analysis was obtained by PlantPAN [49], a putative TATA box was addressed at −24 position, and five NGATT conserved regions [52] were boxed (Figure 7)   Figure 7. Promoter sequence analysis of the BoCKX3 gene. The start codon of translation was shaded. The predicted transcriptional start site was indicated as "+1". The putative functional TATA box and five NGATT elements were designated.

Comparison of Features in Plant CKX Proteins
CKX proteins had been reported to localize at different compartments [16,23,36,37] and some of them were secreted proteins [23]. All three BoCKX proteins were predicted to carry putative signal peptide sequences typical of secretion pathway ( Table 2). Cytosolic and vacuolar forms of CKX were identified in other species [16,36], suggesting that there still may be some unknown BoCKX genes in B. oldhamii.
CKX functions are regulated by post-translational modifications, e.g., glycosylation All CKX proteins were predicted to be glycoproteins (Table 2), and maize CKXs had been shown to be glycoproteins [14]. By using the NetNGly algorithms, BoCKX1, BoCKX2, and BoCKX3 had seven, five, and six putative glycosylation sites and at least five of them were more than 50% probability ( Table 2), implying that BoCKX1-3 may be glycoproteins. In addition to glycosylation, many phosphorylation sites were predicted in BoCKX1-3 (Table 3); however, substantial results are needed.

Comparison of Features in Plant CKX Proteins
CKX proteins had been reported to localize at different compartments [16,23,36,37], and some of them were secreted proteins [23]. All three BoCKX proteins were predicted to carry putative signal peptide sequences typical of secretion pathway ( Table 2). Cytosolic and vacuolar forms of CKX were identified in other species [16,36], suggesting that there still may be some unknown BoCKX genes in B. oldhamii.
CKX functions are regulated by post-translational modifications, e.g., glycosylation. All CKX proteins were predicted to be glycoproteins (Table 2), and maize CKXs had been shown to be glycoproteins [14]. By using the NetNGly algorithms, BoCKX1, BoCKX2, and BoCKX3 had seven, five, and six putative glycosylation sites and at least five of them were more than 50% probability ( Table 2), implying that BoCKX1-3 may be glycoproteins. In addition to glycosylation, many phosphorylation sites were predicted in BoCKX1-3 (Table 3); however, substantial results are needed.

Consensus Sequence Function References
TATA box TATAAT Core promoter cis-element of genes in eukaryotes [50] CAAT box CAAT The CAAT box is a conserved consensus sequence as the binding site of the RNA transcriptional factor [50] As-1 box TGACG Activation sequence 1 (as-1) is a salicylic acid (SA)-and auxin-responsive element [53,56,57] NGATT N=G/A/C/T GATT Arabidopsis cytokinin response regulators ARR1 binding element [52,58,59]

Discussion
A multi-gene family of CKX in plants has been reported previously [17,36]. In this study, we identified three cytokinin oxidase/dehydrogenase genes from green bamboo by screening a bamboo genomic DNA library (Figure 1). In the genomic DNA library construction, 9-23 kb bamboo genomic fragments were ligated with a phage vector [7]. Fortunately, four positive clones isolated contained three full coding regions as well as different sizes promoter regions (Figure 1). BoCKX1 and BoCKX3 genes had identical exonintron structures (Figure 2A) as well as 92% amino acid sequence identity ( Table 1). The exon-intron structure of the BoCKX2 was slightly different in comparison with the above two genes (Figure 2A). All three BoCKX genes showed similar exon-intron organizations between other CKX genes ( Figure 2B). In theory, cytoplasmic, ER-membrane, or vascularlocalized CKX should exist in green bamboo, and more screening of bamboo genomic DNA can be performed.
Several putative cis-acting elements were predicted from promoter regions of the BoCKX1-3 genes ( Figure 5, Figure 6, and Figure 7), and the functions of the motifs were listed in Table 3. All three BoCKX genes contained a conserved TATA box and a typical CAAT box was identified in the upstream site of TATA box of the BoCKX2 [50]. CKX gene expression is induced by abiotic stress and plant hormones, such as abscisic acid and cytokinins [25]. As-1 box [53,56,57] and NGATT motif [52,58,59] were present in the 5 -flanking regions of BoCKX genes, suggesting that the BoCKX gene expressions may be affected by different plant hormones and cytokinin response regulator proteins.