Analysis of cytokinin content and associated genes at different developmental stages in pak choi (Brassica rapa ssp. chinensis Makino)

Abstract Pak choi (Brassica rapa ssp. chinensis Makino) is a typical seed vernalization vegetable. Because premature bolting in spring causes significant economic losses, it is very important to clarify the flower formation mechanism to prevent this. Cytokinins are important plant hormones involved in regulating plant growth and development. To understand the relationship between cytokinin metabolism and flowering of pak choi, in this study, we determined the cytokinin trans-zeatin content in shoot apices of pak choi at different developmental stages by enzyme-linked immunosorbent assay. The results showed that cytokinin levels increased significantly after low temperature treatment at 4 °C, and continued to increase with vegetative growth after transplanting, reaching a peak at the critical period which is immediately prior to flower buds’ differentiation (S0), then decreased thereafter. The levels for low temperature treatment were consistently higher than controls. To explore the molecular mechanism underpinning the cytokinin changes, expression of homologous genes encoding cytokinin metabolic enzymes was analysed by transcriptome sequencing. The expression levels of nine genes (Bra004037, Bra023701, Bra002204, Bra014968, Bra028326, Bra028182, Bra034022, Bra009143 and Bra005869) were consistent with the changes in cytokinin content. The correlation between differentially expressed genes and cytokinin content in different developmental stages of apexes was analysed, and six closely related genes (Bra023701, Bra002204, Bra014968, Bra028182, Bra005869 and Bra009143) were identified. The results help to illuminate the molecular mechanisms controlling flowering of pak choi.


Introduction
Pak choi (Brassica rapa ssp. chinensis Makino) is a cruciferous vegetable that requires vernalization at low temperature before flowering under high temperature and long sunshine conditions. Weak winterness cultivars are more prone to bolting, and premature bolting in spring causes significant economic losses [1]. Therefore, it is of great significance to clarify the flowering mechanism to prevent premature bolting.
Phytohormones play an important role in the flowering process of plants. Among them, cytokinins play a key role in long-distance signalling from root to shoot. This regulates root and shoot growth, photomorphogenesis, flowering time, senescence and seed development [2]. High levels of cytokinins can increase the activity of shoot apical meristem (SAM) [3] and promote the growth of axillary buds [4]. Cytokinins can also promote cell division in Chinese cabbage sprouts [5]. In Arabidopsis thaliana, the SAM at the early stage of floral transition was found to contain more cytokinins [6], the external application of cytokinin can promote the transition of flower formation [7,8], increase the number of female flower and inhibit stamen development [9]. Treatment of Brassica napus with cytokinins increases the ovule and seed numbers [10]. Some scholars found that cytokinins can induce cabbage tip callus to form shoots [11]. In B. napus, cytokinins increase the trans-zeatin content during flower bud differentiation [12], and overexpression of the isopentenyl-transferase (IPT) gene increases the inflorescence branching in plants [13]. In A. thaliana, cytochrome P450 monooxygenase, family 735, subfamily A (CYP735As) has been shown to catalyse the synthesis of trans-zeatin [14], overexpression of uridine diphosphate glucose glycosyltransferase 76Cs (UGT76Cs) reduces cytokinin content [15]. In addition, cytokinin can also delay the senescence of Chinese cabbage leaves during storage [16], and the shedding of Chinese cabbage petals will lead to the reduction of cytokinins [17], drought stress conditions increase the cytokinin content of rapeseed [18]. These findings indicated that cytokinins play a role in different stages of plant growth and development, but the role in the process of flower bud differentiation is particularly important. However, for pak choi, the process of cytokinin metabolism during flower formation is poorly understood.
At present, more than 30 kinds of cytokinins have been reported, including isopentenyl adenine (ip), dihydro-zeatin (dZ), cis-zeatin (cZ) and trans-zeatin (tZ), among which tZ is considered to be the most important cytokinin [19]. Therefore, in the present study, the content of cytokinin in the apexes of pak choi were measured at different developmental stages, and expression levels of associated genes were analysed by transcriptome sequencing to explore the molecular mechanism of cytokinins regulating the flowering process of pak choi.

Material handling and sampling
The pak choi inbred line 75 # was used in the experiment, and seeds with full grains were soaked in water and germinated at 24 °C in an incubator. At 3 days after seed germination, seedlings were placed at low temperature (4 °C) for 20 days and traditional management was carried out. Among them, 800 plants were treated with low temperature, and 500 plants were control.
on days 0 and 10 after transplanting, the shoot apexes of plants subjected to low temperature treatment and control (CK) plants were sampled and recorded as DAV0, CK-DAV0 and DAV10, CK-DAV10, respectively. Flower bud differentiation was observed under a stereo microscope. The shoot apex was sampled when it is at flower buds differentiated stage 0 (S0), CK-S0 and stage 1 (S1), according to Song et al. [20]. A 0.2 g sample was used to determine the cytokinin content, and RnA extraction was performed from 0.1 g samples. Three biological replicates were included. After sampling, samples were snap-frozen in liquid nitrogen and stored at −80 °C for later use.

Cytokinin content determination
The content of cytokinin in the shoot apexes of pak choi at different developmental stages (DAV0, CK-DAV0, DAV10, CK-DAV10, S0, CK-S0 and S1) was determined by enzyme-linked immunosorbent assay at China Agricultural university. The determined component was tZ, and the method was performed as described previously [21].

Transcriptome analysis
Total RnA was extracted using an RnA Prep Pure Plant Kit (Tiangen, DP432) according to the manufacturer's instructions. extracted RnA was stored at −80 °C. Transcriptome sequencing was performed on samples DAV10, S0 and S1 by Biomarker Technologies Co. Ltd (Beijing, China), and sequencing steps, expression analysis and functional annotation were carried out as described previously [22].

Gene expression and functional annotation analysis
In order to avoid the influence of gene length and sequencing depth on the expression level, the method of Fragments Per Kilobasesper Millionmapped Reads (FPKM) was used to normalize the sequencing data. Differentially expressed genes (Degs) at different developmental stages were identified using DeSeq2 software. With FDR < 0.01 and Fold Change (FC) ≥ 2 as the threshold, pairwise comparisons were made in DAV10 vs. S0 and S0 vs. S1, and all Degs were annotated by gene ontology (go).

Real-time quantitative PCR validation
To validate the transcriptome sequencing result, go annotations to cytokinin-related genes were screened in the self-tested transcriptome data, and six Degs that related to cytokinin were randomly selected and used specific primers for real-time quantitative PCR (RT-qPCR). According to the general principles of primer design, Primer-BLAST in nCBI (https://www.ncbi. nlm.nih.gov/tools/primer-blast/) was used to design and select specific primers with amplified fragments of about 200 bp (Table 1), with ACTIN as an internal reference gene. RT-qPCR amplifications were performed on a Quant Studio 3 real-time PCR machine with an initial denaturation step at 94 °C for 5 min, followed by 40 cycles at 94 °C for 30 s, 49.5 °C for 30 s and 72 °C for 30 s, with three biological replicates. Relative expression levels were calculated using the 2 -ΔΔCT method to normalize against the internal reference gene [23], and the specific method was applied described previously [24].

Comparison of cytokinin content in shoot apexes at different developmental stages
The cytokinin(tZ)content of the shoot apexes at different developmental stages was measured for pak choi subjected to low temperature treatment and untreated controls ( Figure 1). The results showed that in the low temperature treatment group, the cytokinin contents were increased during the transition from the vegetative growth phase to the reproductive growth phase, and the levels reached a peak value of 6.92 ng/g FW when the flower bud differentiation was at stage 0 (S0). That is to say, once flower bud differentiation started, the cytokinin content decreased. our results are consistent with these of prior observations, indicating that increased cytokinin content could promote flower bud differentiation in pak choi. The content of cytokinin in the control group showed the same trend, but the levels were consistently lower than in the low temperature treatment group throughout the growth period. This indicated that, within a certain range, higher accumulation of cytokinin contributes to flower bud differentiation of pak choi.
The cytokinin content of the low temperature treatment and control groups at 0 days after transplanting were compared, and the cytokinin levels of the low temperature treatment group were higher, which indicates that low temperature treatment can increase the accumulation of cytokinin and thereby promote the differentiation of flower buds.

Principal component analysis of samples
Principal component analysis (PCA) was performed on nine samples of shoot apexes at different developmental stages, and the contribution rate of the first principal component (PC1) was 56.9%, compared with 29.5% for PC2 and 5.3% for PC3 ( Figure 2). For PC1, 2 and 3, samples with the same colour were clustered together, and samples with different colours were separate, indicating that the repeatability within the group was good, and there were differences between the groups, hence subsequent gene mining could be carried out.

Verification of transcriptome results by RT-qPCR
In order to verify the reliability of transcriptome data, six genes were randomly selected for RT-qPCR verification, and the results were consistent with the sequencing results, confirming that the transcriptome data were accurate and reliable ( Figure 3).
FPKM, fragments per kilobase of transcript per million mapped reads. Different letters indicate a significant difference (p < 0.05).
In order to further analyse the reasons for the changes in cytokinin content in different developmental stages of pak choi, genes encoding key enzymes involved in cytokinin metabolic processes were statistically analysed ( Table 3). The results revealed 13 genes that encode IPT enzymes and CYP735As enzymes, which are involved in cytokinin biosynthesis. Three genes encode ugT76Cs enzymes, which are involved in cytokinin catabolism.
The results of transcriptome sequencing were used to analyse the expression changes of genes encoding enzymes related to cytokinin metabolism. Bra028182 and Bra034022 encoding CYP735A2 were upregulated in DAV10 vs. S0 and downregulated in S0 vs. S1; Bra009143 encoding ugT76C1 and Bra005869 encoding ugT76C4 in catabolism were downregulated in DAV10 vs. S0 and S0 vs. S1. The results of hormone content analysis showed that the cytokinin levels in S0 were higher than in DAV10 and S1, and the changes in gene expression were consistent with the changes in hormone content, indicating that these nine genes may be related to the rate and extent of cytokinin synthesis.

Identification of cytokinin metabolism-related genes
According to FDR < 0.01 and Fold Change (FC) ≥ 2, DAV10 vs. S0, S0 vs. S1 were compared and screened for Degs (Table 4). Based on the go functional annotation of Degs, seven genes (Bra023701, Bra002204, Bra014968, Bra028182, Bra034021, Bra009143 and Bra005869) and three genes (Bra002204, Bra014968 and Bra009142) related to cytokinin metabolism were found in DAV10 vs. S0 and S0 vs. S1, respectively. Among them, the expression levels of six Degs (Bra023701, Bra002204, Bra014968, Bra028182, Bra005869 and Bra009143) were consistent with the changing trend of cytokinin content.  total reads, number of all reads; clean reads, number of reads after filtering joined sequences, contaminated sequences and low-quality sequences; mapped reads, the number of reads mapped to the reference genome and the percentage of clean reads. unique mapped reads, number of reads mapped to unique locations in the reference genome and the percentage of clean reads mapped. DaV10, 10 days after transplanting; S0, stage immediately prior to flower bud differentiation; S1, flower bud differentiation stage 1; 1, 2 and 3 represent three biological replicates.
Bra023701, Bra002204 and Bra014968 encode the IPT enzyme, which catalyses the first and rate-limiting step in cytokinin biosynthesis, using ATP, ADP and AMP as DAMPP receptors to ultimately form isopentenyl adenine (iP) and tZ (Table 4 and Figure 5). Bra028182 and Bra034022 encode the enzyme CYP735As, which utilizes isopentenyl adenine nucleotides to generate tZ nucleotides, In DAV10 vs. S0, the expression of the above genes was increased; we thought that this is beneficial to the synthesis of cytokinin, which promotes the flowering of pak choi. upregulation of CYP735A1 in pak choi can increase the content of cytokinin, but the expression of Bra034021, which encodes CYP735A2, was   (Table 4 and Figure 5). The reasons for this need to be further analysed. Bra005869 and Bra009143 are involved in the degradation of cytokinins and were downregulated in DAV10 vs. S0 (Table  4 and Figure 5). Their homologous genes in Arabidopsis are ugT76C4 and ugT76C1, and downregulation of these two genes at this stage reduces cytokinin decomposition, which promotes flower bud differentiation of pak choi.
In S0 vs. S1 cytokinin anabolism, Bra002204 encoding IPT5 and Bra014968 encoding IPT7 were downregulated, and the transcript levels of AtIPT1, AtIPT5 and AtIPT7 are negatively regulated by CTK in Arabidopsis [38]. Thus, higher cytokinin content inhibits IPT transcription to reduce cytokinin content, consistent with our results. In cytokinin catabolism, Bra009142, which encodes ugT76C2, was downregulated, which may be related to the initiation of flower bud differentiation.

The relationship between cytokinins and flower bud differentiation and development
Cytokinins play an important role in flower bud differentiation for plant flower formation. Previous studies have shown that bud activation and growth depend on local cytokinin synthesis in axillary buds or stems [30], and cytokinins move toward the shoot end to promote Arabidopsis bud growth [31]. During lychee flower bud differentiation, cytokinin activity in flower buds increases [32]. In rapeseed, cytokinins can regulate stem and shoot development [33], increase in cytokinin content is linked with the start of flower bud differentiation [34]. our results were consistent with theirs, suggesting that increased cytokinin content can promote flower bud differentiation.

Expression of some important genes involved in cytokinin metabolism
In order to further elucidate the reasons for the change of cytokinin content in different stages of flower bud differentiation in pak choi, the metabolic pathway of cytokinin was analysed. The results of RnA sequencing revealed that the expression levels of nine genes  [35], increase flowers [36] and increase the number of seeds [37]. In Arabidopsis, the transcript levels of AtIPT1, AtIPT5 and AtIPT7 are negatively regulated by CTK [38], suggesting that higher cytokinin content inhibits IPT transcription to reduce cytokinin content, which is also confirmed by our results. CYP735As utilize isopentenyl adenine nucleotides to generate trans-zeatin nucleotides [14] and promote the transcriptional levels of trans-zeatin, which are lower during dormancy and increase during flowering [39]. our results verified this, which indicates that the flower bud differentiation of pak choi requires a higher content of cytokinins. ugT76C4 and ugT76C1 can catalyse the glycosylation at the 7-, 9-position of the free-state cytokinin ring to generate n-glycosides [29], inactivating cytokinin to reduce cytokinin content, and in our results, the expression of ugT76Cs decreased, especially the expression change of ugT76C1 reached a significant level, which would reduce the decomposition of cytokinin, and then promote the flower bud differentiation of pak choi. However, further studies are needed to confirm the targeting effect of genes in cytokinin metabolism on flower bud differentiation of pak choi.

Conclusions
In this study, the cytokinin contents in the shoot apexes in pak choi at different developmental stages were measured. The results showed that low temperature treatment increased the content of cytokinin, and higher cytokinin levels could promote flower bud differentiation of pak choi. To elucidate the molecular mechanism of changes in cytokinin content, the expression levels of the genes encoding cytokinin metabolic enzymes were analysed based on RnA sequencing results. The results showed that the expression patterns of most genes were consistent with the changes in cytokinin content, especially Bra004037, Bra023701, Bra002204, Bra014968 and Bra028326 encoding IPT enzymes, Bra028182 and Bra034022 encoding CYP735As enzymes, Bra009143 and Bra005869 encoding ugT76Cs enzymes. By comparing the content of tZ and expression of Degs between different developmental stages in pak choi, six genes (Bra023701, Bra002204, Bra014968, Bra028182, Bra005869 and Bra009143) whose expression levels were consistent with changes in cytokinin content were identified. Among them, Bra023701, Bra002204, Bra014968 and Bra028182 involved in cytokinin synthesis were upregulated in DAV10 vs. S0 and downregulated in S0 vs. S1, while Bra005869 and Bra009143 involved in cytokinin breakdown, were downregulated in DAV10 vs. S0. These result in accelerating synthesis and slowing down degradation of cytokinin at a specific stage. Because the trends in expression changes were consistent with the cytokinin content in different developmental stages, six genes were predicted as most closely related to the flowering process of pak choi. The findings can help us to understand the molecular mechanism underlying cytokinin regulation of flower bud differentiation in pak choi.

Data availability
The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found below: https://www.ncbi. nlm.nih.gov/sra/PRJnA821155