Genome-Wide Identiﬁcation and Expression Analysis of NCED Gene Family in Pear and Its Response to Exogenous Gibberellin and Paclobutrazol

: The 9-cis -epoxycarotenoid dioxygenase ( NCED ) is a key enzyme for the process of ABA synthesis that plays key roles in a variety of biological processes. In the current investigation, genome-wide identiﬁcation and comprehensive analysis of the NCED gene family in ‘Kuerle Xiangli’ ( Pyrus sinkiangensis Yu) were conducted using the pear genomic sequence. In total, nineteen members of PbNCED genes were identiﬁed from the whole genome of pear, which are not evenly distributed over the scaffolds, and most of which were focussed in the chloroplasts. Sequence analysis of promoters showed many cis -regulatory elements, which presumably responded to phytohormones such as abscisic acid, auxin, etc. Synteny block indicated that the PbNCED genes have experienced strong purifying selection. Multiple sequence alignment demonstrated that these members are highly similar and conserved. In addition, we found that PbNCED genes were differentially expressed in various tissues, and three PbNCED genes ( PbNCED1 , PbNCED2 , and PbNCED13 ) were differentially expressed in response to exogenous Gibberellin (GA 3 ) and Paclobutrazol (PP 333 ). PbNCED1 and PbNCED13 positively promote ABA synthesis in sepals after GA 3 and PP 333 treatment, whereas PbNCED2 positively regulated ABA synthesis in ovaries after GA 3 treatment, and PbNCED13 positively regulated ABA synthesis in the ovaries after PP 333 treatment. This study was the ﬁrst genome-wide report of the pear NCED gene family, which could improve our understanding of pear NCED proteins and provide a solid foundation for future cloning and functional analyses of this gene family. Meanwhile, our results also give a better understanding of the important genes and regulation pathways related to calyx abscission in ‘Kuerle Xiangli’.


Introduction
The plant hormone abscisic acid (ABA), considered as a universal stress hormone, plays a crucial role in the adaptation to biotic and abiotic stresses and in several physiological processes [1] such as seed development [2], root growth [3], plant responses to inhibition of water uptake, and drought stress [4,5]. The 9-cis-epoxycarotenoid dioxygenase (NCED) is a key enzyme in the ABA biosynthesis pathway [6]. NCED cleaves carotenoid C40 in plastids and produces ABA by catalysis of alcohol dehydrogenase and aldehyde oxidase [7].
The first NCED transcription factor was found in the maize vp14 mutant [8]. Recently, several studies suggested that ABA results in decreasing the carotenoid content in juice sacs of citrus fruit by inducing its own biosynthesis, but the expression of CitNCED2 and CitNCED3 increases rapidly with ABA accumulation [9,10]. Similarly, the expression of FaNCED1 plays a major role in ABA accumulation and the process of fruit ripening of strawberry [11,12]. Studies have shown that transcript levels of peach and grape NCED transcription factors regulate its fruit ripeness and senescence [13]. OsNCED5 expression was found in all tissues and has similar functions to OsNCED3 and OsNCED4 in response to abiotic stress [14]. In Arabidopsis (Arabidopsis thaliana), overexpression of AtNCED3 results in increased endogenous ABA accumulation and resistance to drought stress, as well as induced gene response of ABA [6]. Co-expression of SgNCED1 and ALO genes increased tolerance to drought and chilling and induced expression of its related gene in the transgenic tobacco and stylo plants [15]. Real-time PCR analysis revealed MpNCED2 were expressed continuously during the whole period of apple fruit development, while the expression of MpNCED1 clearly declined to a steady low level in the mid-later period of fruit development [16]. Additionally, NCED is also an important candidate gene involved in the response to cold stress in panax ginseng (P. ginseng) and can act as an endogenous signaling molecule regulating cold stress and biosynthesis of P. ginseng [17].
The NCED gene family has been identified and studied in many plant species such as Arabidopsis [18], cotton [19], avocado [20], cowpea [21], kiwifruit [22], and grape [23]. Pears (Pyrus spp.) belong to the family Rosaceae, subfamily Maloideae, and genus Pyrus [24]. Pear is a major economic crop in the world [25]. However, the role of NCED genes has not been clearly identified in pear.
The pear genome was published in 2013, which provided an opportunity to reveal genome-wide NCED gene family [26]. In the present study, we used bioinformatics methods to identify NCED genes from the pear genome, and analyze the sequence features, evolutionary traits, tissue-specific expression levels, and expression patterns in response to exogenous Gibberellin (GA 3 ) and Paclobutrazol (PP 333 ). The results provide useful information for further functional investigations of the NCED gene family in pear.

Identification and Analysis of PbNCED Genes
A total of 22 PbNCEDs were obtained by local BLASTP and Hidden Markov Model (HMM) analysis. After that, CDD and SMART were used to confirm the presence of the conserved RPE65 domain, which confirmed the existence of 19 putative PbNCED genes in the pear genome. Finally, the 19 PbNCED genes were named from PbNCED1 to PbNCED19 based on their scaffolds' locations (Table S1).
The lengths of the PbNCED proteins ranged from 474 aa to 617 aa. The molecular weights of PbNCED proteins were between 53.92 kD and 69.13 kD. The predicted pI values of PbNCED proteins ranged from 5.61 to 7.64. All PbNCED proteins had typical secondary structures including Alpha helix, Beta turn, Extended strand, and Random coil. The PbNCED proteins contained the highest proportion of extended strands and the lowest proportion of β-folded structures except PbNCED1 and PbNCED2. The result of subcellular localization showed that the PbNCED proteins were predicted to in the chloroplast, but a few PbNCED proteins possibly had a cytoplasmic, peroxisome, and mitochondria localization (Table S2).

Multiple Sequence Alignment and Phylogenetic Analysis of PbNCED Genes
To analyze the evolutionary relationships of NCED genes in pear, Arabidopsis, rice, and grape, a phylogenetic tree was constructed using MAGE-X software employing the neighbor-joining method ( Figure 1). The phylogenetic tree showed that these NCED proteins could be classified into six groups, namely, Group I, Group II, Group III, Group IV, Group V, and Group VI, with well-supported bootstrap values. Additionally, Group II, Group III, and Group IV were further divided two subfamilies including Group II-a, Group II-b, Group III-a, Group III-b, Group IV-a, and Group IV-b, respectively. Nearly all groups included PbNCEDs, with the exception of Group IV-a and Group IV-b. The numbers of PbNCEDs in Group I, Group II-a, Group II-b, Group III-a, Group III-b, Group V, and Group VI were 4, 4, 2, 3, 2, 3, and 1, respectively. The PbNCED proteins had many conserved amino acids by multiple sequence alignment of PbNCED domains (Figure 2), especially histidine (H), glycine (G). and threonine (T). The results revealed that the sequences in the PbNCED domain were highly conserved. groups included PbNCEDs, with the exception of GroupⅣ-a and GroupⅣ-b. The numbers of PbNCEDs in Group Ⅰ, Group Ⅱ-a, Group Ⅱ-b, Group Ⅲ-a, Group Ⅲ-b, Group Ⅴ, and Group Ⅵ were 4, 4, 2, 3, 2, 3, and 1, respectively. The PbNCED proteins had many conserved amino acids by multiple sequence alignment of PbNCED domains (Figure 2), especially histidine (H), glycine (G). and threonine (T). The results revealed that the sequences in the PbNCED domain were highly conserved. Figure 1. The phylogenetic relationship of the NCED gene family to Arabidopsis, rice, grape, and pear. Note: The blue solid dot represents Arabidopsis NCED protein, the purple solid pentagon represents rice NCED protein, the orange solid triangle represents grape NCED protein, and the red solid square represents pear NCED protein. The phylogenetic relationship of the NCED gene family to Arabidopsis, rice, grape, and pear. Note: The blue solid dot represents Arabidopsis NCED protein, the purple solid pentagon represents rice NCED protein, the orange solid triangle represents grape NCED protein, and the red solid square represents pear NCED protein.

Gene Structure and Motif Composition
To analyze the evolutionary relationships of NCED genes in pear, the phylogenetic tree was constructed using full length amino acid sequences. The 19 PbNCEDs were classified into five distinct subfamilies. The 15 conserved motifs were found among 19 PbNCEDs by MEME website, and the amino acid consensus sequence of each motif is listed in Supplementary Figure S1. The lengths of these conserved motifs ranged from 11 to 50 amino acids. Members of all PbNCEDs generally had more than seven motifs except for PbNCED9, which only contained six motifs. The similar motif arrangements among PbNCED proteins within subgroups indicated that the protein architecture is conserved within a specific subfamily. Further, in exploring the exon-intron structure characteristics of the NCED family in pear, we note that exons and introns existed in every PbNCED gene. The number of introns was the least in Group I and Group III, while Group II, Group IV, and Group V had a higher number of introns. These similar structural characteristics may be related to the functions of these genes in the pear genome ( Figure 3).

Chromosomal Location, Gene Duplication and Synteny Analysis
Physical locations of 19 NCED genes in pear were investigated by analysis of genomic distribution on scaffolds and were distributed on 14 pear scaffolds randomly ( Figure 4A). In addition, scaffold5, scaffold6, scaffold7, and scaffold8 had two predicted PbNCED genes, but other scaffolds had only one. To better understand the evolutionary constraints acting on PbNCED gene family, the Ks, Ka, and Ka/Ks ratios of the PbNCED gene pairs were calculated to investigate the divergence time of the duplication blocks (Table S3). The five segmental duplication gene pairs were estimated to have occurred 31.40-84.984 million years ago. These duplicated genes were identified, and the Ka/Ks of all five pairs were <1, suggesting that the PbNCED gene family had experienced strong purifying selection and the slow evolution rate in pear. The CDS of PbNCED4 and PbNCED19 were the same so that cannot be used to calculate Ka/Ks.

Gene Structure and Motif Composition
To analyze the evolutionary relationships of NCED genes in pear, the phylogenetic tree was constructed using full length amino acid sequences. The 19 PbNCEDs were classified into five distinct subfamilies. The 15 conserved motifs were found among 19 PbNCEDs by MEME website, and the amino acid consensus sequence of each motif is listed in Supplementary Figure S1. The lengths of these conserved motifs ranged from 11 to 50 amino acids. Members of all PbNCEDs generally had more than seven motifs except for PbNCED9, which only contained six motifs. The similar motif arrangements among PbNCED proteins within subgroups indicated that the protein architecture is conserved within a specific subfamily. Further, in exploring the exon-intron structure characteristics of the NCED family in pear, we note that exons and introns existed in every PbNCED gene. The number of introns was the least in Group Ⅰ and Group Ⅲ, while Group Ⅱ, Group Ⅳ, and Group Ⅴ had a higher number of introns. These similar structural characteristics may be related to the functions of these genes in the pear genome ( Figure 3).

Chromosomal Location, Gene Duplication and Synteny Analysis
Physical locations of 19 NCED genes in pear were investigated by analysis of genomic distribution on scaffolds and were distributed on 14 pear scaffolds randomly ( Figure 4A). In addition, scaffold5, scaffold6, scaffold7, and scaffold8 had two predicted PbNCED genes, but other scaffolds had only one. To better understand the evolutionary constraints acting on PbNCED gene family, the Ks, Ka, and Ka/Ks ratios of the PbNCED gene pairs

Interaction Network and Cis-Regulatory Element Analysis
The PbNCED proteins were constructed the protein-protein interaction network based on the interaction relationship of the homologous NCED proteins (AtCCD1/AtNCED3/AtNCED4/AtNCED6/AtCCD7/AtCCD8) in Arabidopsis ( Figure 5). The NCED proteins played an important role in Arabidopsis. For example, AtNCED3 played a major role in ABA synthesis under drought stress [6]; AtNCED6 were involved in ABA synthesis during seed development [27]; and AtCCD1, AtCCD7, and AtCCD8 were involved in carotenoid metabolic pathways [28]. The results illustrate that the PbNCED family proteins might have similar functions.  To reveal the orthologous relationships of NCED genes on chromosomes between Arabidopsis, pear, and apple genomes, a comparative analysis was performed between NCED genes in Arabidopsis, pear, and apple by TBtools ( Figure 4B). The four orthologous gene pairs were identified between Arabidopsis and pear. The 19 orthologous gene pairs were identified between pear and apple. The result indicated the close relationships between pear and apple, compared to Arabidopsis and pear. In addition, the three NCED genes (PbNCED2/PbNCED9/PbNCED14) were identified to have orthologous genes within the Arabidopsis, pear, and apple genomes simultaneously. These NCED genes might have evolved from the common ancestor in different plants.

Interaction Network and Cis-Regulatory Element Analysis
The PbNCED proteins were constructed the protein-protein interaction network based on the interaction relationship of the homologous NCED proteins (AtCCD1/AtNCED3/ AtNCED4/AtNCED6/AtCCD7/AtCCD8) in Arabidopsis ( Figure 5). The NCED proteins played an important role in Arabidopsis. For example, AtNCED3 played a major role in ABA synthesis under drought stress [6]; AtNCED6 were involved in ABA synthesis during seed development [27]; and AtCCD1, AtCCD7, and AtCCD8 were involved in carotenoid metabolic pathways [28]. The results illustrate that the PbNCED family proteins might have similar functions.
To investigate the transcriptional regulation and potential functions of PbNCEDs in stress responses, the cis-elements in the promoter regions (the 2000 bp sequences upstream from the translation start sites) of the PbNCED genes were submitted and analyzed in the PlantCARE database ( Figure 6). Six hormone-related response elements were displayed in Figure 6, namely, the ABA responsive element (ABRE), ethylene-responsive element (ERE), auxin-responsive elements TGA-element, the gibberellin responsive elements P-box, GARE-motif, and TATC-box, salicylic acid responsive TCA-element, and methyl jasmonate responsive element (CGTCA-motif). The number of ABRE elements was the highest among the hormone-related cis-elements. The stress-related response elements included MBS, W-box, ARE, WUN-motif, and TC-rich repeats. The ARE element was found in most of PbNCED genes. The cis-element analysis illustrated that PbNCED genes could respond to various stresses.

Interaction Network and Cis-Regulatory Element Analysis
The PbNCED proteins were constructed the protein-protein interaction network based on the interaction relationship of the homologous NCED proteins (AtCCD1/AtNCED3/AtNCED4/AtNCED6/AtCCD7/AtCCD8) in Arabidopsis ( Figure 5). The NCED proteins played an important role in Arabidopsis. For example, AtNCED3 played a major role in ABA synthesis under drought stress [6]; AtNCED6 were involved in ABA synthesis during seed development [27]; and AtCCD1, AtCCD7, and AtCCD8 were involved in carotenoid metabolic pathways [28]. The results illustrate that the PbNCED family proteins might have similar functions.  To investigate the transcriptional regulation and potential functions of PbNCEDs in stress responses, the cis-elements in the promoter regions (the 2000 bp sequences upstream from the translation start sites) of the PbNCED genes were submitted and analyzed in the PlantCARE database ( Figure 6). Six hormone-related response elements were displayed in Figure 6, namely, the ABA responsive element (ABRE), ethyleneresponsive element (ERE), auxin-responsive elements TGA-element, the gibberellin responsive elements P-box, GARE-motif, and TATC-box, salicylic acid responsive TCAelement, and methyl jasmonate responsive element (CGTCA-motif). The number of ABRE elements was the highest among the hormone-related cis-elements. The stress-related response elements included MBS, W-box, ARE, WUN-motif, and TC-rich repeats. The ARE element was found in most of PbNCED genes. The cis-element analysis illustrated that PbNCED genes could respond to various stresses.

Tissue-Specific Expression of PbNCED Genes
To further explain the functions of PbNCED genes in the development of 'Kuerle Xiangli' (Figure 7), the expression levels of thirteen selected PbNCED genes in different tissues (floral stalk, leave, new shoot, ovary, and sepal) were analyzed by qRT-PCR. PbNCED7, PbNCED11, PbNCED12, and PbNCED18 displayed relatively high expression levels in floral stalk. PbNCED2 and PbNCED10 showed high expression levels in leaves and new shoots, respectively. Moreover, PbNCED1, PbNCED3, PbNCED4, PbNCED5, PbNCED6, and PbNCED19 were specifically expressed in the sepals. PbNCED8 was highly expressed in ovaries, compared with other genes. These genes played diverse roles in 'Kuerle Xiangli' development.

Tissue-Specific Expression of PbNCED Genes
To further explain the functions of PbNCED genes in the development of 'Kuerle Xiangli' (Figure 7), the expression levels of thirteen selected PbNCED genes in different tissues (floral stalk, leave, new shoot, ovary, and sepal) were analyzed by qRT-PCR. PbNCED7, PbNCED11, PbNCED12, and PbNCED18 displayed relatively high expression levels in floral stalk. PbNCED2 and PbNCED10 showed high expression levels in leaves and new shoots, respectively. Moreover, PbNCED1, PbNCED3, PbNCED4, PbNCED5, PbNCED6, and PbNCED19 were specifically expressed in the sepals. PbNCED8 was highly expressed in ovaries, compared with other genes. These genes played diverse roles in 'Kuerle Xiangli' development.

Effect of Exogenous GA3 and PP333 on Content of GA3, IAA and ABA in 'Kuerle Xiangli'
In order to better evaluate the effect of exogenous GA3 and PP333 treatment, the GA3, IAA, and ABA content was assessed in the development stages of ovaries and sepals (Table S4). The results showed that the content of GA3 was significantly higher in GA3 and PP333-treated sepals compared with those in the control at 5 days. GA3 and ABA content in GA3 and PP333-treated ovaries decreased after rising, but IAA showed an obvious upward trend, and a significant difference was seen at 9 days compared to control. The content of IAA and ABA was significantly lower than those after PP333 treatment, but the GA3 content was higher than those after PP333 treatment in sepals at 5 days. In addition, the content of IAA and GA3 was significantly higher than those after PP333 treatment in sepals at 9 days.
The ratio of GA3/ABA and IAA/ABA was significantly higher in the GA3-treated sepals and ovaries at 5 days and 9 days than in the control. The GA3/ABA and IAA/ABA ratio was significantly lower in the PP333-treated sepals and ovaries compared to the control at 5 days and 9 days. Interestingly, the ratio of IAA/ABA had the highest value after GA3 treatment in sepals on day 5, but the ratio of IAA/ABA was the highest in ovaries on day 9 (Table S5). This result indicated that exogenous GA3 and PP333 affect the development of 'Kuerle Xiangli' flowers by changing the level of endogenous hormones, especially the content of abscisic acid. In order to better evaluate the effect of exogenous GA 3 and PP 333 treatment, the GA 3 , IAA, and ABA content was assessed in the development stages of ovaries and sepals (Table S4). The results showed that the content of GA 3 was significantly higher in GA 3 and PP 333 -treated sepals compared with those in the control at 5 days. GA 3 and ABA content in GA 3 and PP 333 -treated ovaries decreased after rising, but IAA showed an obvious upward trend, and a significant difference was seen at 9 days compared to control. The content of IAA and ABA was significantly lower than those after PP 333 treatment, but the GA 3 content was higher than those after PP 333 treatment in sepals at 5 days. In addition, the content of IAA and GA 3 was significantly higher than those after PP 333 treatment in sepals at 9 days.

Expression Analysis of PbNCED Genes under Exogenous GA3 and PP333
The ratio of GA 3 /ABA and IAA/ABA was significantly higher in the GA 3 -treated sepals and ovaries at 5 days and 9 days than in the control. The GA 3 /ABA and IAA/ABA ratio was significantly lower in the PP 333 -treated sepals and ovaries compared to the control at 5 days and 9 days. Interestingly, the ratio of IAA/ABA had the highest value after GA 3 treatment in sepals on day 5, but the ratio of IAA/ABA was the highest in ovaries on day 9 (Table S5). This result indicated that exogenous GA 3 and PP 333 affect the development of 'Kuerle Xiangli' flowers by changing the level of endogenous hormones, especially the content of abscisic acid.

Expression Analysis of PbNCED Genes under Exogenous GA 3 and PP 333
We randomly selected six PbNCEDs to study their expression levels under the effect of exogenous GA 3 and PP 333 treatment by using qRT-PCR (Figure 8). The qRT-PCR analysis revealed that the expression levels of PbNCED2 and PbNCED3 decreased after rising, while the expression of PbNCED1 rose after decreasing the GA 3 treatment. Additionally, the PbNCED6 had the highest expression level after GA 3 treatment as compared with the control and PP 333 treatment in ovaries at 9 days. The expression level of PbNCED2 decreased after increasing the PP 333 treatment in ovaries and sepals, and three genes (PbNCED1, PbNCED3, and PbNCED13) exhibited a higher expression level than the control and GA 3 treatment in sepals at 9 days. The result showed that PbNCED genes could respond to hormones. We randomly selected six PbNCEDs to study their expression levels under the effect of exogenous GA3 and PP333 treatment by using qRT-PCR (Figure 8). The qRT-PCR analysis revealed that the expression levels of PbNCED2 and PbNCED3 decreased after rising, while the expression of PbNCED1 rose after decreasing the GA3 treatment. Additionally, the PbNCED6 had the highest expression level after GA3 treatment as compared with the control and PP333 treatment in ovaries at 9 days. The expression level of PbNCED2 decreased after increasing the PP333 treatment in ovaries and sepals, and three genes (PbNCED1, PbNCED3, and PbNCED13) exhibited a higher expression level than the control and GA3 treatment in sepals at 9 days. The result showed that PbNCED genes could respond to hormones.

Correlation Analysis for PbNCED Gene Expresssion Level, Content of GA3, IAA, and ABA in 'Kuerle Xiangli' during Sepal and Ovary Development
Correlation analysis between the content of GA3, IAA, and ABA in PbNCED genes was carried out to explore the correlation between phytohormones and PbNCED genes ( Figure 9). For this analysis, among the PbNCED genes, the expression patterns of PbNCED1 and PbNCED13 had a positive correlation with content of ABA, but they had a negative correlation with content of GA3 in the sepals. Additionally, the expression pattern of PbNCED4 had a negative correlation with content of IAA in the ovaries. Taken together,

Correlation Analysis for PbNCED Gene Expresssion Level, Content of GA 3, IAA, and ABA in 'Kuerle Xiangli' during Sepal and Ovary Development
Correlation analysis between the content of GA 3 , IAA, and ABA in PbNCED genes was carried out to explore the correlation between phytohormones and PbNCED genes ( Figure 9). For this analysis, among the PbNCED genes, the expression patterns of PbNCED1 and PbNCED13 had a positive correlation with content of ABA, but they had a negative correlation with content of GA 3 in the sepals. Additionally, the expression pattern of PbNCED4 had a negative correlation with content of IAA in the ovaries. Taken together, these results indicate that the PbNCED gene could affect synthesis of GA 3 , IAA, and ABA in 'Kuerle Xiangli' during sepal and ovary development.
these results indicate that the PbNCED gene could affect synthesis of GA3, IAA, and ABA in 'Kuerle Xiangli' during sepal and ovary development.

Discussion
In the current study, a total of 19 PbNCED genes were identified in pear and those genes were classified into five subfamilies, namely, Group Ⅰ (six), Group Ⅱ (three), Group Ⅲ (five), Group Ⅳ (two), and Group Ⅴ (three). Consistent with a previous study [29], most of the PbNCED proteins were localized in the chloroplasts. Motifs of PbNCED members' arrangements were similarities in the same subgroup but obvious differences occur between different subgroups. The result implied that PbNCED genes might exhibit similar functions in the same subgroups. Moreover, multiple sequence alignments revealed that the histidine (H) of PbNCED was highly conserved in PbNCED sequence, which is similar to previous reports [18]. The Ka/Ks ratio analysis indicated that the duplicated PbNCED gene pairs evolved under negative selection, consistent with an earlier study [30].
Some valuable clues about the functional role of PbNCED genes involved in specific 'Kuerle Xiangli' physiological processes were obtained. For example, ABA synthesis played a role in the promotion of leaf senescence through BnNCED3 as a positive regulator [31]. In Citrus reshni, CrNCED1 had the highest expression in leaves and the lowest level was detected in the root [32]. The AhNCED1 transcript and endogenous ABA both accumulate predominantly in peanut leaves in response to dehydration [33]. In our study, PbNCED10 was specially expressed in leaves, with extremely low expression levels in other tissues. In addition, PbNCED1, PbNCED3, PbNCED4, PbNCED5, PbNCED6, and PbNCED19 were specifically expressed in the sepals. In the present study, expression analysis by RT-PCR of PbNCED genes implied that those genes had tissue-specific expression.
Cis-element played an important role in regulating gene expression such as in response to drought stress and the signal transduction of hormone [34]. In a previous

Discussion
In the current study, a total of 19 PbNCED genes were identified in pear and those genes were classified into five subfamilies, namely, Group I (six), Group II (three), Group III (five), Group IV (two), and Group V (three). Consistent with a previous study [29], most of the PbNCED proteins were localized in the chloroplasts. Motifs of PbNCED members' arrangements were similarities in the same subgroup but obvious differences occur between different subgroups. The result implied that PbNCED genes might exhibit similar functions in the same subgroups. Moreover, multiple sequence alignments revealed that the histidine (H) of PbNCED was highly conserved in PbNCED sequence, which is similar to previous reports [18]. The Ka/Ks ratio analysis indicated that the duplicated PbNCED gene pairs evolved under negative selection, consistent with an earlier study [30].
Some valuable clues about the functional role of PbNCED genes involved in specific 'Kuerle Xiangli' physiological processes were obtained. For example, ABA synthesis played a role in the promotion of leaf senescence through BnNCED3 as a positive regulator [31]. In Citrus reshni, CrNCED1 had the highest expression in leaves and the lowest level was detected in the root [32]. The AhNCED1 transcript and endogenous ABA both accumulate predominantly in peanut leaves in response to dehydration [33]. In our study, PbNCED10 was specially expressed in leaves, with extremely low expression levels in other tissues. In addition, PbNCED1, PbNCED3, PbNCED4, PbNCED5, PbNCED6, and PbNCED19 were specifically expressed in the sepals. In the present study, expression analysis by RT-PCR of PbNCED genes implied that those genes had tissue-specific expression.
Cis-element played an important role in regulating gene expression such as in response to drought stress and the signal transduction of hormone [34]. In a previous study, PpNCED1 and PpNCED5 gene promoted ABA biosynthesis and accelerated cell senescence by activating ROS signals in peach [35]. Overexpression of VaNCED1 results in increased endogenous ABA accumulation and induced the gene response of JA in grape [36]. Signals of ABA and IAA cross-talk each other [37,38]. ABA negatively controls hypocotyl elongation acting on GA metabolic genes, affecting GA signaling and repressing auxin biosynthetic genes [39]. The ABRE element was the most abundant among the PbNCED1, PbNCED11, PbNCED12, and PbNCED13. These genes might have functions similar to those described above. In this study, many hormone-related response elements and stress-related response elements were found, indicating that PbNCED plays an important role and has a complex function in the hormone regulation of plant biological processes and the stress response in 'Kuerle Xiangli'.
In this study, the ratio of GA 3 /ABA and IAA/ABA was GA 3 treatment > CK > PP 333 treatment, and they had significant differences to one another. The result showed that there was antagonism between GA 3 or IAA and ABA. Some valuable clues about the role of IAA (or GA 3 ) and ABA in different plants are uncovered. For example, in the development of strawberry fruits, there was synergism between IAA and GA, but IAA (or GA) and ABA had antagonism [40]. The rates of falling flowers and fruit of glycyrrhiza uralensis are negatively correlated with the IAA level and positively correlated with the ABA level [41]. IAA played a major role, while ABA had the opposite effect on stigma exertion in tomato [42]. In maize, the contents of IAA and GA were positively correlated with the maximum seed filling rate, seed weight, and mean filling rate, while ABA was negatively correlated [43]. The phytohormone content assays showed an increase in the content of abscisic acid in young wheat ears, but a decrease in the content of auxin and gibberellins [44]. Additionally, the qRT-PCR method was used to detect the relative expression of PbNCED genes in ovaries and sepals after they were sprayed with distilled water, PP 333 , and GA 3 at full bloom stage. The results showed that the expression of PbNCED1 and PbNCED13 gradually increased with the time delay, which was consistent with the ABA content change after GA 3 and PP 333 treatment in sepals. We speculated that PbNCED1 and PbNCED13 positively regulated the synthesis of ABA in sepals. In addition, the results of correlation analysis further proved the reliability of this speculation. The expression of PbNCED2 and PbNCED13 decreased after rising after GA 3 and PP 333 treatment in ovaries, consistent with the ABA content change. Further, the expression levels of PbNCED2 after GA 3 treatment were significantly higher than those after PP 333 treatment, but the expression levels of PbNCED13 were significantly lower than those after PP 333 treatment in ovaries on 5 d. We speculated that PbNCED2 positively regulated the synthesis of ABA after GA 3 treatment in ovaries, while PbNCED13 positively regulated the synthesis of ABA after PP 333 treatment in ovaries.

Plant Materials
The plant materials used in this study were selected in spring 2021 at the Shayidong Horticulture Field, Korla, Xinjiang Province. Nine uniform twenty-year-old 'Kuerle Xiangli' trees were selected and three trees in each treatment were sprayed with 50 mg/L GA3, 500 mg/L PP333, and water at the bloom stage, respectively. Flowers, mature leaves, and secondary shoots were collected from each tree on the first, fifth, and ninth days of the bloom stage, respectively. After collection, petals, stamens, pistils, and floral shoots, were manually removed, and ovaries and sepals were collected, respectively. All tissues were immediately frozen in liquid N and stored at −80 until use.

Genome-Wide Identification and Protein Properties of NCED Family in Pear
The whole genome of the Pyrus bretschneideri (taxid:225117) was downloaded from NCBI (https://www.ncbi.nlm.nih.gov/, accessed on 20 April 2021). The protein sequences of NCED in Arabidopsis were downloaded from TAIR (https://www.arabidopsis.org/index. jsp). These sequences were used to search pear protein sequence data with the BLASTP program (p < 1 × 10 −6 ). Additionally, to further identify PbNCED candidates, the Hidden Markov Model (HMM) analysis was used for the search. We downloaded HMM profile of RPE65 (PF03055) from the Pfam protein family database (http://pfam.xfam.org/), which was download and used as the query file (p < 1 × 10 −6 ) to search the pear protein sequence data. After removing the redundant hits, the presence of RPE65 domains was verified by searching the Conserved Domains Database (CDD: https://www.ncbi.nlm.nih.gov/ Structure/cdd/cdd.shtml), with ambiguous sequences being manually confirmed using the SMART (http://smart.embl-Heidelberg.de/smart/set_mode.cgi?NORMAL=1).

Sequence Analyses and Phylogenetic Tree Construction
The biochemical features, including molecular weight (MW) and isoelectric point (pI) of PbNCED proteins, were determined by the ExPASy protparam tool (https://web.expasy. org/protparam/). The protein secondary structure of PbNCED proteins was predicted with SOPMA (https://npsa-prabi.ibcp.fr/cgi-bin/npsa_automat.pl?page=npsa_sopma.html). The subcellular localizations of the PbNCED proteins were predicted with WOLF PSORT (http://www.genscript.com/wolfpsort.html). The MEME tool (http://meme-suite.org/ tools/meme) was used to predict the motifs of PbNCED proteins with an optimal width of 6-200 residues and the maximum number of motifs set to 15. The domain sequences of the PbNCED proteins aligned using Clustal Omega (https://www.ebi.ac.uk/Tools/msa/ clustalo/), and the result was displayed by Jalview software 2.11.2.6 [45]. The exon-intron structures of the PbNCED genes were identified by using the TBtools software [46]. The phylogenetic trees were constructed using the neighbor-joining (NJ) method of MEGA-X with parameters of 1000 bootstrap replicates and pairwise deletion [47]. One additional tree was constructed using the same methodology. Sequences of NCED proteins from rice and grape were downloaded from the Ensemble (http://plants.ensemble.org/species.html).

Gene Duplication and Synteny Analysis
To exhibit the synteny relationship of the orthologous NCED genes from Arabidopsis, pear, and apple, the syntenic analysis were examined using TBtools software with default parameters. Analysis of gene duplication events were examined by TBtools software. Non-synonymous (ka) and synonymous (ks) substitution of each duplicated PbNCED genes were calculated using KaKs_Calculator 2.0. The genome sequences of apple were downloaded from the GDR database (https://www.rosaceae.org/).

Interaction Network and Cis-Regulatory Element Analyses
The blastp program was used between the pear NCED proteins and the Arabidopsis NCED proteins; each pear NCED protein matched a homologous Arabidopsis NCED protein with the highest score. Six Arabidopsis NCED proteins, which represent the pear NCED proteins, were uploaded to the String website (https://string-db.org) to predict protein interactions. For promoter analysis, the upstream 2000 bp region of the transcription start site (ATG) for each PbNCED gene was defined from the pear genome and the stressrelated and hormone-related cis-elements were analyzed by the PlantCARE tool (http: //bioinformatics.psb.ugent.be/webtools/plantcare/html/).

Determination of Phytohormones
For GA3, IAA, and ABA extraction, approximately 1 g of ovaries and sepals was weighed and extracted with methanol at 4 • C for 12 h. Each mix was centrifuged for 30 min at 6000 rpm, and the supernatants were then stored at 4 • C. Each precipitate was also extracted with methanol twice. Petroleum ether was added to adsorb phenolics and pigments from the resulting supernatants. Final clean-up of the extract was conducted by extracting with ethyl acetate at pH 3.0, adjusted with formic acid. Finally, the extracts were centrifuged and evaporated, and then HPLC analysis was conducted on the dried samples. Separation of phytohormones (GA3, IAA and ABA) was performed using a ZORBAX Eclipse Plus C18 column (250 mm

RNA Extraction and Gene Expression Analysis
The total RNA from pear tissues was extracted using the Plant RNA Extraction Mini Kit (NovaBio, Shanghai, China), according to the manufacturer's instructions. All RNA were analyzed by agrose gel electrophoresis and then quantified with a Thermo Nano Drop 2000 spectrophotometer. Then, these RNAs were reverse transcribed into first-strand cDNA, following the manufacturer's procedure for the HyperScriptTM III RT SuperMix for the qPCR with the gDNA Remover Kit (TNovaBio, Shanghai, China). The qRT-PCR primers were designed by Primer3Plus website (http://www.primer3plus.com/cgi-bin/ dev/primer3plus.cgi) and are shown in Supplementary Table S6. The quantitative RT-PCR was carried out with the Bio-Rad CFX96 Real Time PCR System. The reaction program was set as follows: 95 • C for 30 s, followed by 42 cycles of 95 • C for 10 s and 60 • C for 30 s. The actin gene was used as the internal reference gene, and each sample was repeated 3 times. The data from real-time PCR amplification was analyzed using the 2 − Ct method [48].

Conclusions
In our study, we identified a total of 19 NCED genes in the pear genome and analyzed their gene structure, structural domains, and conserved motifs. The phylogenetic tree showed that pear NCED genes could be divided into five subgroups. Chromosomal localization and covariance analysis showed that the 19 NCEDs were randomly distributed on 14 scaffolds and had undergone strong purifying selection during evolution. The PbNCED genes may have been shown to participate in the regulation of abscission and the persistent calyx after spraying PP 333 or GA 3 , thus increasing ABA synthesis and carotenoid accumulation ( Figure 10). Our findings supply a better understanding of the important genes and regulation pathways related to calyx abscission in 'Kuerle Xiangli'.  The funding institution was not involved in the design of the study, collection, analysis, and interpretation of data, and in writing the manuscript.
Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.
Data Availability Statement: All data in this study can be found in public databases and supplements, as described in the Section 4 Materials and Methods.