Identification, classification, and transcription profiles of the B-type response regulator family in pear

Type-B response regulators (B-RRs) are transcription factors that function in the final step of two-component signaling systems. In model plants, B-RRs have been shown to play important roles in cytokinin signal transduction. However, the functions of B-RRs in pear have not been well studied. In this report, we conducted a genome-wide analysis and identified 11 putative genes encoding B-PpRR proteins based on the published genome sequence of Pyrus bretschneideri. A phylogenetic tree of the B-PpRR family was constructed, and the motif distribution, chromosome localization, and gene structure of B-PpRR family genes were determined. Gene transcript profiles, which were determined from transcriptome data, indicated that B-PpRR genes potentially function during pear fruit development, bud dormancy, and light/hormone-induced anthocyanin accumulation. Treatment of the fruitlets of ‘Cuiguan’ pear (Pyrus pyrifolia), which never accumulates anthocyanin, with the cytokinin N-(2-chloro-4-pyridyl)- N′-phenylurea (CPPU) clearly induced anthocyanin accumulation. Anthocyanins accumulated in the skin of fruitlets by 16 days after CPPU treatment, along with the significant activation of most anthocyanin biosynthetic genes. Analyses of B-PpRR transcript levels suggested that B-PpRR genes mediated this accumulation of anthocyanins. These findings enrich our understanding of the function of B-PpRR genes in the physiological processes of pear.


Introduction
Cytokinins are N 6 -substituted adenine derivatives that play essential roles in regulating plant growth and development. They have been shown to control cell division, vascular development, chloroplast biogenesis, stress tolerance, senescence, and differentiation of the shoot meristem, leaf, and root [1]. Cytokinin signal transduction is mediated by a two-component PLOS  system based on a phosphotransfer cascade, which refers to phosphorylation on a nitrogen atom of a histidine (His) residue and on an acyl group of an aspartate (Asp) residue [2]. The two-component system was originally found in bacteria [3]. In Arabidopsis, there are three cytokinin receptors known as Arabidopsis histidine kinases (AHK), AHK2, AHK3, and AHK4/CRE1/WOL. The autophosphorylation of a conserved His residue in the kinase domain occurs after binding to cytokinins. The phosphate group is then transferred to a conserved Asp residue in the receiver domain of the receptors [1], and further to Arabidopsis histidine-phosphotransfer proteins (AHP). Finally, the phosphate group is transferred to the Arabidopsis response regulators (ARR) and releases their activities. The ARR is the essential player in cytokinin signal transduction. Two types of ARRs (Atype and B-type) have been identified based on structure and function. Both the A-type and Btype RRs have a receiver domain at the N-terminal, but they differ at the C-terminal. The Cterminal of the B-type RRs (B-RR) has a long extension with a MYB-like DNA binding domain [4]. In total, 10 A-ARRs and 12 B-ARRs have been identified in the Arabidopsis genome. B-ARRs function as transcription factors that directly bind to the promoter of target genes and activate their transcription [2]. A-ARRs are direct targets of B-ARRs and negatively regulate the cytokinin signaling pathway [2]. Specifically, A-ARRs inhibit the activity of B-ARRs, forming a negative feedback loop in the cytokinin signaling pathway [1]. Overexpression of ARR1, one of the most studied B-ARRs, resulted in enhanced sensitivity to cytokinin [5], while the triple mutant arr1, 10,12 showed almost complete insensitivity to cytokinin. Single mutants were not significantly different from wild type, indicating that there is functional overlap among B-ARRs [6].
Studies on model plants such as Arabidopsis have revealed the important roles of B-RRs in regulating multiple biological processes, including cell division, differentiation in the root meristem, organ size [7][8][9], and anthocyanin accumulation [10]. Cytokinins have been shown to regulate a wide array of downstream responses through ARR1/10/12, indicating that they play the most important roles in cytokinin signal transduction. The triple mutant arr1,10,12 with decreased cell division showed significant lower expression of a D-type cyclin, CycD3;1, which is implicated in the regulation of cell division by cytokinins [6,11]. The inflorescence stems of the triple mutant arr1 arr10 arr12 were narrower than those of wild type, and their siliques were much shorter [6]. Together, these findings demonstrated that B-ARRs regulate cell division and organ size via the cytokinin signaling pathway. It was also reported that the triple mutant arr1 arr10 arr12 accumulated less anthocyanin than did wild type when treated with cytokinin. Thus, cytokinins may regulate fruit development and anthocyanin biosynthesis at least partly through B-type RRs.
Cytokinins are known to regulate fruit size in pear [12]. However, less is known about their functions in dormancy and coloration. To unravel the functions of cytokinins in pear development, in this study, B-type RR genes were isolated based on the published genome sequence of Pyrus bretschneideri. Phylogenetic and structural analyses classified the pear B-RRs into four main groups. Then, transcript profile analyses using RNA-seq data indicated that members of Group II play important roles in pear fruit development, bud dormancy, and light-induced anthocyanin accumulation. Experimental analyses revealed that B-RRs were involved in N-(2-chloro-4-pyridyl)-N 0 -phenylurea (CPPU)-mediated fruitlet coloration of P. pyrifolia cv. Cuiguan, providing further evidence that cytokinin stimulates anthocyanin accumulation. These findings enrich our understanding of the potential functions of B-RR genes during pear fruit development, and provide basic information for further studies on the molecular mechanism of cytokinins in anthocyanin accumulation and floral bud dormancy.

Fruit bagging treatment
Fruit of red Chinese sand pear 'Meirensu' (P. pyrifolia Nakai) (MRS) were sampled from a commercial orchard in Kunming City, Yunnan Province, China. And the owner of the orchard gave permission to conduct our study on this site. We stated that no specific permissions were required for these locations/activities, because our field studies did not involve endangered or protected species. The MRS fruit were bagged at 40 days after full bloom (DAFB) with double layers of yellow-black paper bags. Half of the fruit were re-exposed to sunlight when the bags were removed 10 days before harvest, and the remaining fruit served as control samples. Thirty fruit were randomly sampled with three biological replicates at 0 H (hour), 6 H, 24 H, and 144 H after bag removal, while control fruit were sampled at the same times and were designated as 6 HC, 24 HC, and 144 HC (C, control; the 0 H sample did not require a control). Fruit peel was collected and immediately frozen in liquid nitrogen and stored at −80˚C until use.

Plant materials of bud dormancy
The bud materials used for the transcriptome analysis of 'Suli' pear (P. pyrifolia white pear group) and Japanese pear 'Kosui' (P. pyrifolia) have been described previously [13].

Cytokinin treatment of fruitlets
Six mature 'Cuiguan' pear (P. pyrifolia Nakai) trees with similar tree vigor and loading were selected from the orchard of Zhejiang University, Hangzhou, Zhejiang, China. Receptacles of 'Cuiguan' flowers were sprayed with 30 mg/L CPPU/0.02% Tween-20 as the cytokinin treatment and with distilled water/0.02% Tween-20 as the negative control during the full-blossom period. Receptacles were collected at 0 d before treatment, and fruitlets were collected at 3 d, 7 d, 11 d, 16 d, 21 d, and 31 d after treatment. The pear peel was carefully collected and immediately frozen in liquid nitrogen and stored at −80˚C until use.
Identification of B-type RR genes in pear B-type RR genes (B-PpRR) were isolated from the published genome sequence of P. bretschneideri (http://gigadb.org/dataset/100083, Wu et al. 2013). The hidden Markov model (HMM) profiles of the response regulator receiver domain (PF00072) and Myb-like DNA-binding domain (PF00249) were used as queries to search the white pear database using HMMER 3.0 (http://hmmer.org/) with default parameters [14][15]. The sequences containing both two domains were selected and compared with the B-ARR genes in the Plant Transcription Factor Database (http://planttfdb.cbi.pku.edu.cn/index.php) [16].

Mapping B-PpRR genes on pear chromosomes
We searched for the locations of B-PpRR genes in the sequences of pear chromosomes from GigaDB (http://gigadb.org/dataset/100083) [17]. The figure was drawn using JoinMap4.0 software.

Sequence alignment and phylogenetic analyses
The B-PpRR family sequences for Arabidopsis, apple, and strawberry were retrieved from the Plant Transcription Factor Database. The B-PpRR protein sequences were aligned using MEGA5.0 (http://www.megasoftware.net/download_form) with default settings [18].
Phylogenetic trees were constructed using MEGA 5.0 software with the neighbor-joining (NJ) method and 1000 bootstrap replicates.

Gene structure and conserved motif identification
The gene structure of each B-PpRR gene was drawn using the online software Gene Structure Display Server (http://gsds.cbi.pku.edu.cn/) [19]. Motifs in pear B-PpRR protein sequences were identified using the motif-based sequence analysis tool, MEME Suite Version 4.11.2 [20]. The optimum width of amino acid sequence was set from 6 to 30. The maximum number of motifs was set to 30.

RNA extraction and cDNA synthesis
Total RNA was extracted using a modified CTAB method [25]. After digestion of genomic DNA by DNase I, the concentration of total RNA was measured using BioDrop (Biochrom). First-strand cDNA was synthesized from 4 μg DNA-free RNA using the Revert Aid™ First-Strand cDNA Synthesis Kit (Fermentas, Glen Burnie, MD, USA) according to the manufacturer's instructions. The cDNA was diluted 10-fold and 2 μl was used as the template for gene cloning and real-time quantitative PCR (Q-PCR) analysis.
The efficiency and specificity of primers were determined by melting curves and by sequencing PCR products. Real-time PCR was performed using a CFX96 instrument (Bio-Rad). Each Q-PCR reaction mixture contained 7.5 μl SYBR Premix Ex TaqTM (Takara, Ohtsu, Japan), 0.5 μl each primer (10 μM), 1 μl cDNA, and 5.5 μl RNase-free water in a total volume of 15 μl. The reaction started with a preliminary step of 95˚C for 3 min, followed by 40 cycles of 95˚C for 10 s and 60˚C for 30s. A template-free control for each primer pair was included for each run. For each gene, the relative transcript level was determined with the 2 -ΔΔCt algorithm [26] with normalization against the pear Actin gene.

Statistical analysis
The LSD values (α = 0.05) were calculated for mean separations using Statistical Product and Service Solutions software v. 19 (SPSS Inc., Chicago, IL, USA).

Identification, phylogenetic analysis, and structural analysis of B-PpRR proteins
In total, 13 genes putatively encoding B-PpRR proteins were isolated from the published genome sequence of P. bretschneideri. After manually removing duplicated sequences, the remaining 11 genes were further analyzed.
To evaluate the phylogenetic relationships of pear B-PpRR genes and classify them within established subfamilies, we compared the B-PpRR sequences from pear, apple (Malus domestica), strawberry (Fragaria × ananassa), and Arabidopsis (Arabidopsis thaliana). The phylogenetic analysis clustered the B-PpRR family into four groups, among which group II was further divided into three subgroups (Fig 1a). The MEME motif analysis revealed that some of the motifs were conserved, while others were distributed among some of the subfamilies (Fig 1b). All the B-PpRR proteins had the REC signal receiver domain (motifs 7, 5, 4/17, 6, 2) at the N-terminal for signal reception in the two-component system. Groups I and III contained an OmpR domain consisting of REC and winged-helix (wHTH) domains, while the type of REC domain in group III differed from that in other groups. All of the B-PpRR genes contained the MYB domain (including motifs 1 and 3) so that all the B-PpRR genes were confirmed to be MYB transcription factors. However, the type of MYB domain in Group IV differed from that in other groups. Specifically, Group II contained a REC domain like that in Group IV and an MYB_SHAQKYF domain (Fig 1b). Some other motifs were found only in some members. For example, motif 8 was present in only five genes and motif 13 in only four genes. Notably, ARR1/10/12, which play essential (or general) roles in cytokinin signal transduction, were included in Group II.

Chromosomal locations and structure of B-PpRR genes
We identified the locations of nine B-PpRR genes on pear chromosomes (Fig 2a). The remaining two genes were located on scaffolds that have not been mapped to chromosomes.  As gene structure is a typical imprint of evolution within a gene family, the structure of B-PpRR genes was displayed using tools at the GSDS website. Full-length cDNA sequences were compared with genomic DNA sequences to determine the numbers and positions of exons and introns in each B-PpRR gene. All of the B-PpRR genes in pear contained introns, the number ranging from four to 11 (Fig 2b). The number of introns was conserved within each group.

Transcript profiles of B-PpRR genes during fruit development
To investigate the potential functions of B-PpRR genes during fruit development, the transcript profiles of B-PpRR genes in large fruit of pear 'Xueqing' (XQ, average weight >400 g per fruit) (P. pyrifolia) in different developmental stages were determined from transcriptome data (Fig  3). Six B-PpRR genes, B-PpRR5/9/3/6/7/8, showed differences in their transcript levels among different developmental stages. Among them, B-PpRR5/6/8 were transcribed in the early stages of fruit development and their transcript levels peaked at 7d after full blossom, probably because of the requirement to trigger and maintain cell division at anthesis until the ovary reached full maturity. There were high transcript levels at B-PpRR9 at all time points and peak transcript levels at maturity, while the transcript level of B-PpRR7 peaked at 85 d. Notably, all these B-PpRR genes belonged to group II, suggesting that these B-PpRR genes primarily participate in regulating fruit development.

Transcript profiles of B-PpRR genes during floral bud dormancy
The transcript profiles of B-PpRR genes were investigated in the floral buds of 'Suli' pear (P. pyrifolia white pear group) during dormancy [13] (Fig 4). B-PpRR5/9/3/6/8 were differentially transcribed among different dormancy stages. Among these five genes, B-PpRR9 showed the highest transcript levels, which gradually increased during chilling accumulation and peaked towards endodormancy release. The transcript levels of B-PpRR5 and B-PpRR6 were low during chilling accumulation but peaked during endodormancy release, while the transcript levels of B-PpRR8 showed the opposite trend. The transcript level of B-PpRR3 decreased to minimum levels during chilling accumulation and then increased.
The transcript profiles of B-PpRR genes were also investigated in the dormant floral buds of Japanese pear 'Kosui' (P. pyrifolia) [21]. Four B-PpRR genes, B-PpRR9/6/7/8, showed significant differences in transcript levels among different stages. Among these genes, B-PpRR9/6/7 showed increased transcript levels during endodormancy release, while B-PpRR8 showed the opposite transcript profile. The transcript profiles of these genes in 'Kosui' pear were similar to those in 'Suli', suggesting their conserved functions in regulating bud dormancy in Chinese pear and Japanese pear.

Transcript profiles of B-PpRR genes during anthocyanin accumulation
Cytokinins have been reported to enhance anthocyanin content and transcript levels of anthocyanin biosynthetic genes in the presence of light in various plants. To verify the function of B-PpRR genes in regulating anthocyanin accumulation in pear, the transcript profiles of B-PpRR genes were investigated in the peel of bagged fruit of MRS using transcriptome data (under submission) (Fig 5). After removing the bags from fruit, anthocyanins accumulated in pear skin, but no anthocyanins accumulated in control fruit skin. Among the B-PpRR genes, B-PpRR7 showed high transcript levels in all samples, particularly in the 6 H sample after bag  removal, suggesting that B-PpRR7 might participate in regulating light-induced anthocyanin accumulation. In contrast, the transcript levels of other B-PpRR genes were very low or even undetected.
To further verify the functions of cytokinin and B-PpRRs in regulating fruit coloration, the green/yellow pear cultivar 'Cuiguan' was treated with the cytokinin CPPU. Unlike most non-red apples whose fruit accumulate anthocyanin at the early developmental stage, most Asian pears never accumulate anthocyanin. After the CPPU treatment of receptacles during the full-blossom period, the fruitlets accumulated anthocyanin after 16 days, and the anthocyanin content peaked at 21 d after treatment (Fig 6a) while the control fruit remained green. The relative transcript levels of the anthocyanin biosynthesis genes PpCHS, PpCHI, PpF3H, PpDFR, and PpUFGT, Analysis of B-type response regulator family in pear peaked at 16 d after treatment, while that of the upstream gene PpPAL peaked at 3 d after treatment. The transcript level of PpANS peaked at 21 d after treatment. Among the regulator genes, PyMYB10 was up-regulated from 16 d to 31 d after CPPU treatment and PpbHLH3 transcript levels peaked at 16 d after CPPU treatment (at the same time as PpWD40), while the highest mRNA level of PpbHLH33 was at 3 d after treatment. In general, the peak transcript levels of anthocyanin biosynthesis genes preceded the peak of visible color accumulation, indicating that CPPU triggered anthocyanin accumulation by activating anthocyanin biosynthetic genes.
We also analyzed the transcript levels of genes within the cytokinin signaling pathway. The transcript levels of PpHK1 (cytokinin receptor gene) increased at 7 d after treatment with CPPU and then gradually declined, while the transcript levels of PpHK3 peaked at 7 d and fluctuated until 16 d after treatment (Fig 6c). These results indicated that CPPU activation of the cytokinin signaling pathway mainly depended on PpHK1.
Finally, the transcript levels of B-PpRR genes in group II, B-PpRR5/9/3/6/7/8 were analyzed by q-PCR assays. For most of these genes, their transcript levels first increased and then decreased (Fig 6d). Specifically, CPPU induced transcription of B-PpRR3 and B-PpRR7. Their transcript profiles were similar to that of PpHK1, suggesting that they played important roles in regulating anthocyanin biosynthesis under these conditions. [4], rice [27], and poplar [28]. In this study, 11 B-RR genes were identified from the genome sequence of P. bretschneideri (Fig 1a), approximately equivalent to those in Arabidopsis (12 genes) [4], rice (13 genes) [27] and poplar (11 genes) [28], suggesting that no B-RR gene family expansion in these species although some genome duplication events happened in some species, such as pear (Wu et al, 2013). The importance of cytokinins in plant growth and development has probably led to the conserved family size of B-RR genes. This may reflect selection against substantial changes in the stoichiometry of B-RR genes in the cytokinin signaling pathway [28].

B-RR genes have been characterized in Arabidopsis
The REC signal receiver domain at the N-terminal and the MYB domain were present in all of the pear B-type RR genes (Fig 1b), indicating that these genes are MYB transcription factors and that their N-terminal is conserved. The B-PpRR genes were further classified into four subgroups according to the phylogenetic and structural analyses (Fig 1a). In Arabidopsis, B-ARRs formed three groups based on a phylogenetic analysis [29]. The extra group in pear was group III, whose members had a different REC domain from that in other groups, which haven't been distinguished in Arabidopsis (Fig 1b).
Interestingly, members of Group II showed differences in their transcript levels during fruit development (Fig 3), bud dormancy (Fig 4), and anthocyanin accumulation induced by light (Fig 5) or cytokinins (Fig 6d), suggesting their involvement in these physiological processes. In Arabidopsis, the well-studied ARR1/10/12, were also clustered in Group II. The arr1 arr10 arr12 triple mutant showed almost complete insensitivity to exogenously applied cytokinins, which exhibited decreased cell division in the shoot, enhanced seed size, increased sensitivity to light, altered chlorophyll and anthocyanin concentrations, and an aborted primary root with with protoxylem but no metaxylem [6]. Our results also showed that that Group II members played dominant roles in cytokinin signal transduction in pear. The functions of the other three groups are yet to be determined.

B-PpRR links cytokinin signaling pathway and fruit development
In most plants, fruit development can be divided into four phases. The first phase is ovary development and fruit set; the second phase is fruit growth via cell division, which determines the number of cells; the third phase begins after cell division ceases, and consists of cell expansion until the fruit reaches its final size; and the last phase is fruit ripening [30]. We analyzed the transcript profiles of B-PpRR genes in large XQ pear fruit (>400 g) during its development that contains the four phases (Fig 3). We found that five B-PpRR genes showed changes in their transcript levels during fruit development, among which B-PpRR5/6/8 showed high transcript levels at the early stage, when there was rapid cell division (Fig 3). As cytokinins induced the cell division in the early development of pear fruit, the spatial and temporal synthesis and action of cytokinins were required for the early development of fruit growth [31], the high expression of B-PpRR genes may contribute the active cytokinin responses in these stages. Indeed, the involvement of B-RR genes in the regulation of cell division [32], which is directly related to the cell number of fruits, has been well characterized. Specifically, in Arabidopsis, ARR1 and ARR12 activate SHY2, a repressor of auxin signaling, which then affects the cell differentiation and division of the root meristem [8][9]. In addition, the triple mutant of arr1/10/12 also had far fewer cells in the root, compared with that in wild type [7]. These findings suggest that B-RR genes participate in regulating cell number and organ size. In our study, it is likely that B-PpRR5/6/8 were the main B-PpRR genes mediating cytokinin signals in the early stage of pear fruit growth (Fig 3), which indicated that B-PpRR5/6/8 may play important roles in the fruit size of XQ pear. In contrast, B-PpRR9 showed higher transcript levels at the later stage of pear fruit development, and its transcript levels were higher than those of all other B-PpRR genes throughout fruit development (Fig 3). This transcript profile suggested that B-PpRR9 may have much more diverse functions during fruit development and may also be involved in seed development during fruit ripening, as reported in tea and soybean [33][34] Cytokinin regulates bud dormancy through B-PpRR genes Dormancy is a beneficial biological property which is essential for survival in a hostile environment. It is a complex process that is regulated by photoperiod, temperature, and phytohormones. In pear, plant hormones are important factors in regulating dormancy [35]. According to the transcription profiles, we found that several B-PpRR genes showed changes in their transcript levels during dormancy in both Chinese white pear and Japanese pear (Fig 4). Specifically, B-PpRR9/6 in both cultivars, B-PpRR5/3 in 'Suli', and B-PpRR7 in 'Kosui', showed similar trends in their transcription profiles during the transitional state of endodormancy and ecodormancy, and their transcript levels increased after endodormancy release (Fig 4). These results suggested that B-RR-mediated pathway probably contributed to the release of endodormancy. Although the effects of cytokinin on the regulation of endodormancy of deciduous trees have not been well described, the positive function of cytokinin on the termination of potato tuber dormancy has been reported [36]. Furthermore, Rousselin et al. (1992) also found that a cytokinin-resistant mutant of N. plumbaginifolia exhibited reduced seed dormancy [37], indicating that cytokinins play vital roles in dormancy. Similarly, our findings suggested that the cytokinin-signaling pathway is involved in regulating pear bud dormancy. However, B-PpRR8 showed the opposite trend in its transcript profile during the transitional state of endodormancy and ecodormancy, and its transcript levels decreased after endodormancy release (Fig 4). Together, these results illustrated the spatio-temporal and tissue-specific transcription patterns of B-PpRRs in pear bud dormancy.

Cytokinins enhance anthocyanin biosynthesis via B-PpRR genes in pear
Anthocyanin accumulation is has been known as affected by various hormones, including cytokinin [38]. Exogenous cytokinin has also been shown to induce anthocyanin accumulation in strawberry suspension cells [39], A. thaliana [40], and in callus cultures of red-fleshed apple [41]. In this work, we confirmed that anthocyanin biosynthesis was enhanced by application of cytokinin in the pear fruitlet (Fig 6a) in which B-PpRR genes were involved. Among the B-PpRRs, B-PpRR7 showed peak transcript levels immediately after bag removal (Fig 5), indicating that it is an early light-responsive gene that is possibly involved in light-induced anthocyanin accumulation. However, it seemed that B-PpRR7 was not able to directly regulate anthocyanin biosynthesis because the peak transcript level of B-PpRR7 occurred much earlier than that of anthocyanin accumulation (Fig 5). Interestingly, the transcriptional pattern of B-PpRR7 was similar to that of PpHY5, the key transcriptional factor in light signal transduction, suggesting a potential interaction between these two genes. In Arabidopsis, cytokinin induced anthocyanin accumulation in a light-dependent manner and resulted in the increased accumulation of HY5 [42]. It is possible that B-PpRR7 is involved in light-induced anthocyanin biosynthesis via directly or indirectly inducing PpHY5.
Furthermore, the application of CPPU on 'Cuiguan fruitlet' induced the expression of most anthocyanin biosynthetic genes as well as the regulator gene PyMYB10 (Fig 6b), indicating that cytokinin induced anthocyanin accumulation via PyMYB10. Similarly, in Arabidopsis, exogenous cytokinin also enhanced anthocyanin accumulation via up-regulation of PAP1 and bHLHs [10] through the up-regulation of a subset of B-ARR genes, which was also observed in our results (Fig 6d). Taken together, cytokinin enhance the anthocyanin accumulation through the activation of anthocyanin-biosynthesis genes via B-PpRR genes in pear.
Interestingly, the CPPU-induced anthocyanin accumulation was only observed in fruitlets (Fig 6a) but not in mature fruit, probably because anthocyanin biosynthesis in fruitlets and ripe fruit is controlled by different regulatory mechanisms. In grape, two kinds of MYB transcription factors that regulated phenylpropanoid biosynthesis showed different expression patterns in fruitlets and ripe fruit; for example, VLMybA1-1 was detected in ripening grape, and VvMYB5a was detected in the fruitlets [43][44]. Similarly, therefore, the B-PpRR family of MYB transcription factors might directly regulate the biosynthesis of anthocyanin in 'Cuiguan' pear fruitlets. It is also an interesting finding that CPPU induced coloration in a non-red pear cultivar. This may be a useful strategy to alter fruit color.

Conclusion
The results of this study showed that cytokinins are involved in regulating pear fruit size and development, and pear bud dormancy, via B-PpRR genes. The results also showed that cytokinins play positive roles in light-or hormone-induced anthocyanin accumulation. Notably, all the differentially transcribed B-PpRR genes belong to group II, indicating that this group of B-PpRR genes is involved in the cytokinin signaling pathway. Our results also indicated that the functions of B-RR genes are conserved among different species. Further research is required to explore the molecular mechanism of cytokinin-induced anthocyanin accumulation.