The cellular physiology of loquat (Eriobotrya japonica Lindl.) fruit with a focus on how cell division and cell expansion processes contribute to pome morphogenesis

Scientia Horticulturae 224 (2017) 142–149

Contents lists available at ScienceDirect

Scientia Horticulturae journal homepage: www.elsevier.com/locate/scihorti

The cellular physiology of loquat (Eriobotrya japonica Lindl.) fruit with a focus on how cell division and cell expansion processes contribute to pome morphogenesis Wenbing Su, Yunmei Zhu, Ling Zhang, Xianghui Yang, Yongshun Gao, Shunquan Lin

MARK

⁎

College of Horticulture, South China Agricultural University, Guangzhou, 510642, China

A R T I C L E I N F O

A B S T R A C T

Keywords: Eriobotrya japonica Lindl Fruit development Cell division Cell expansion Cell cycle exit Gene regulatory

Develop from flower tube, loquat (Eriobotrya japonica Lindl.) fruit differs from apple and other pome on size and shape greatly. To date, cell regulators of loquat fruit morphogenesis are still unknown. Fruit growth, histological observation and regulatory genes expression were investigated for 22 stages in loquat. Growth measurements reveal that loquat contains larger carpel volume and thinner flesh compared to apple. Cellular analyses demonstrate that cell number contributes to size varieties among loquat and other pome. Low cell production rate confers to weak cell proliferation capacity during early development. The correlation of cell number/size changes and several regulatory gene expression implies that: loquat fruit maintained cell division under regulation of EjFWLs and other genes from anthesis to 42 days past anthesis; unique temporal expression of EjWEE1 and EjKRP3 at 42 and 63 days past anthesis participate in cell cycle exit and polyploidy establishment; complementary expression of EjCCS52 isoforms promotes cell growth in an endoreduplication dependent pathway like EjWEE1 and EjKRP3 may be involved; EjEXPAs involved in acid cell growth play crucial role in fruit cell size enlargement. Together, these data indicate that cell division and expansion under complicated regulation, and weak cell proliferation capacity result to less cell layer which confers to thin cortex.

1. Introduction Multicellular organisms rely on the coordinated progression of cell proliferation and cell differentiation (growth) for organ morphogenesis. Cell proliferation and differentiation are regulated in a coordinated manner through organ development. Generally, cell division activity would gradually decrease as organogenesis proceeds, and most, cells eventually exit division and enter differentiation. The scheduled cessation of cell division and cell growth initiation are critical for the formation of organs with genetically defined sizes and morphologies (Conlon and Raff 1999; Hepworth and Lenhard, 2014). DREAM complexes (containing the retinoblastoma protein (RB), E2F and its dimerization partner DP, and MYB homologues) have been identified in animals and plants, and they link several distinct transcription factors to coordinate gene expression throughout the cell cycle (Fischer and DeCaprio 2015). (Kobayashi et al., 2015) identified distinct roles for the MYB3R family members during Arabidopsis cell cycle. The results revealed that complex bonds to MYB3R4 might activate the expression of G2/M target genes (e.g., KNOLLE, EDE1, CYCB1;2, CDKB1;2 and CDKB2;2), whereas combinations with MYB3R3 would

⁎

Corresponding author. E-mail address: [email protected] (S. Lin).

http://dx.doi.org/10.1016/j.scienta.2017.06.012 Received 10 January 2017; Received in revised form 5 May 2017; Accepted 10 June 2017 0304-4238/ © 2017 Elsevier B.V. All rights reserved.

repress cell division. At a given time in development, plant cells at the meristem will exit mitotic cell cycle and start differentiating. To date, the mechanisms that control cell cycle are still not well understood, and the exit from cell cycle also needs more investigation. In Arabidopsis, the exit from cell division requires a decrease in the CDK activity (De Veylder et al., 2007). The inhibition of CDK activity by ICK/KRP proteins results in premature cell cycle exit. WEE1 expression is rapidly induced by DNA stress, resulting in the tyrosine phosphorylation of CDKs and the controls of cell cycle arrest in Arabidopsis (De Schutter et al., 2007). CCS52 (cell cycle switch) proteins inhibit the mitotic cycle and drive cells into endocycle for post-mitotic cell growth (Cebolla et al., 1999). Plant cells also arrest cell cycle in response to environmental cues (De Veylder et al., 2007). In developing from the ovary (ovaries) or other floral organs, fruits protect developing seeds and contribute to seed dispersal for species propagation. In addition, fruits provide humans with nutrition, culinary diversity, and great pleasure (Tanksley 2004). A true fruit develops from the ovary, and the ovary wall becomes the pericarp. For pseudocarpic fruit, organs such as receptacle bracts (or the floral tube and other tissues) participate in fruit formation(Gillaspy et al., 1993).

Scientia Horticulturae 224 (2017) 142–149

W. Su et al.

Fig 1. Fruit growth of ‘Zaozhong-6’ loquat. (A) A time series of loquat receptacle size. Bar = 2 cm. (B) Changes in fruit diameter. The inset presents a magnified view of fruit diameter during early pome development. (C) Changes in the cortex. The inset presents a magnified view of the cortex thickness during early pome development. For each time point fruit diameter and cortex thickness were measured with 5 biological replicates.

expansion in VvCEB1-overexpressing embryos (Nicolas et al., 2013). In pears (Hiwasa et al., 2003), grapes (Ishimaru et al., 2007; Suzuki et al., 2015) and peaches (Cao et al., 2016), the expression of EXPAs has been analysed during fruit late development; some of these genes are associated with cell division or expansion. In addition, a SNP marker for peach fruit size control was identified in the promoter region of one of the Expansin genes by (Cao et al., 2016) recently. However, the molecular mechanisms of cell division and cell expansion regulation in woody fruit trees still requre further study. The molecular mechanisms involved in regulating fruit morphogenesis and final size in loquats are not yet understood. The growth pattern of the loquat has been studied by (Blumenfeld 1980), (Ding and Zhang 1988) and (Cuevas et al., 2003), and it was divided into 3 stages by (Ding and Zhang 1988). To date, no information has been reported about quantitative changes in gene expression with cell division or cell enlargement during loquat fruit morphogenesis. The primary goal of this work is to establish exhaustive cytological kinematics for loquat fruit, and to determine the key cellular program shift points to illustrate how cell division and cell expansion coordinate and relate to gene expression profiles during pome formation. To understand how cell division and cell growth coordinate with one another throughout fruit development, the receptacles or cortexes from 22 stages were collected and sectioned, and numerous genes associated with cell division, cell cycle exit and cell expansion were selected for quantitative analysis.

Different types of fruit display various developmental programmes. For example, the avocado pericarp continues its cell division until shortly before ripening (Schroeder, 1953) while most fruits such as peaches (Ognjanov et al., 1995), tomatoes (Joubes et al., 1999) and apples (Bain and Robertson 1951; Malladi and Hirst 2010) continue to engage in cell proliferation during early fruit development stage, and later, the cells enlarge for a long period. GS3, GS5, GW8 and GLW7 have been demonstrated to play important roles in the grain width and length regulation of rice and other grass (Mao et al., 2010; Li et al., 2011; Wang et al., 2012; Si et al., 2016), and many genes have been functionally identified to act as pivotal regulators during fruit size and shape formation in dicots. FW2.2 was first cloned in tomato as a negative regulator of ovary wall cell division during tomato fruit weight evolution (Frary et al., 2000). Another QTL (quantitative trait loci) is FW3.2, which promotes tomato pericarp cell division (Chakrabarti et al., 2013). Additionally, natural variations in POS1 and AtARF18 were successively shown to have diverse capacities for cell size regulation during tomatillo and silique development, respectively (Wang et al., 2014; Liu et al., 2015). Moreover, two cloned loci called Fascinated (FAS) and Locule number (LC) were discovered to have functions in tomato shape domestication by controlling the locule number in the ovary (Cong et al., 2008; Munos et al., 2011). The natural variations in the above genes all contribute to fruit size or shape evolution. Unfortunately, few genes have been determined to specifically regulate woody perennial fruit development. The expression patterns of two MdANTs during early fruit growth were found to coincide with cell production period, and they were positively correlated with that of the cell cycle regulatory genes. The data implied that MdANTs may participate in cell production in apples (Dash and Malladi 2012). PpKNOPE1 represses GA3ox1 expression during peach mesocarp cell differentiation and ectopic KNOPE1expression in Arabidopsis, resulting in decreased cell size (Testone et al., 2015). A bHLH (basic helix loop helix) transcription factor called VvCEB1 has been determined to be grape berry-specific, and it strongly stimulates cell

2. Materials and methods 2.1. Plant materials In this study, five mature ‘Zaozhong-6’ trees under normal management in the loquat germplasm resource preservation garden (South China Agricultural University, Guangzhou, China) were used. Fully open flowers at full-bloom stage were tagged. Measurements and samples were collected before or after tagging. Samples were collected 143

Scientia Horticulturae 224 (2017) 142–149

W. Su et al.

3. Results

starting 21 days before anthesis (−21 DPA) until 126 days past anthesis (126 DPA) when the fruits were commercially ripened as shown in Fig. 1. For each measurement point before blooming, 10–15 flowers were selected for diameter measurements and further studies. Starting from anthesis, 30 of the tagged fruits were used for growth kinematics inspection while the others were used for sampling. The receptacle (or hypanthium) next to the ovules/seeds at the equatorial region was excised from the flowers (or fruits) and used for histological and gene expression analyses as described by (Denne 1960). These tissues were either saved in FAA solution (5% formalin-5% acetic acid-90% alcohol) or frozen in liquid nitrogen and stored at −80 °C.

3.1. Fruit growth pattern of ‘Zaozhong-6′ loquat The equatorial diameters of the 30 tagged fruits from 5 trees were measured weekly as shown in Fig. 1A. The fruit diameter growth reveals that the loquat fruits grew exponentially during the early stages and linearly during later stages following a ‘single-sigmoid-curve’ pattern. From 28 DPA on, the fruit sizes increased rapidly until ripening, and the flower bud size increased in a similar pattern before anthesis (Fig. 1B). By contrast, the receptacle sizes at anthesis and the days closely following flower blooming were distinguishable from each other by sight-seeing (Fig. 1A). The phenotypic trait nonconformity drove us to explore its origin and cortex growth may contribute to some extent according to former observation. We then tried to observe the growth model of the cortex. Other than the diameter, the cortex initiated thickening to a slight extent from the first week after anthesis. Then, it entered the first dramatic thickening period followed by a 4-week thickening cessation. Afterwards, the cortex thickness grew more dramatically during almost all the later fruit development periods, except that there was a deceleration from 91 to 98 DPA (Fig. 1C). A comparison of diameter and cortex thickness reveals that the flesh part occupies a smaller portion while an ovary with multiple seeds or ovules takes up a larger volume of the entire fruit (Fig. 1B and C, Fig. A.1 in Supplementary material).

2.2. Methods 2.2.1. Measurement of fruit diameter and cortex thickness For growth inspection, fruit diameters were measured for the 30 tagged fruits every week. Ten to 15 fruits/flowers were collected from 5 trees each week; they were cut into halves, with one half being frozen immediately and the other half used for cortex thickness measurement. 2.2.2. Cortex section preparation and cell measurement After the measurements, cortex sections were prepared as described by (Chen et al., 2016) with modifications. The sections were photographed with bright-field illumination using AxioVision LE64 software (Carl Zeiss, German). The cell area was measured with Images J software (http://rsb.info.nih.gov/vj/). The cell number was counted from the epidermis across the cortex to the pith of each section, and this counn was set as cell layer according to (Denne 1960) and (Malladi and Hirst 2010). The relative cell proliferation rate and relative cell expansion rate were determined from the cell layer and cell area data as follows: relative cell proliferation rate = (Layer2-Layer1)·(Time2Time1)−1 and relative cell expansion rate = (Cell size2-Cell size1)·(Time2-Time1) −1. The period from −21 to −14 DPA was defined as P1 (period 1), and in the same manner, the period from −14 to −4 DPA and −4 to 0 DPA were set as P2 and P3.

3.2. Cell proliferation and cell expansion in developing fruit The receptacle region next to the ovule was fixed and cut into 10 μm sections for cell observation and measurement. Some sections representing the four typical stages are shown in Fig. A.1 in Supplementary material. Before bloom, the receptacle cell exhibited a high proliferation rate from −21 to −4 DPA, and then, cell division was arrested when it was close to flower blooming. After anthesis, the cell number sustained a higher proliferation rate than the former period till 42 DPA. After that, the total cell layers fluctuated at approximately 62. In contrast to the 43 cell layers at anthesis, a mature pome has approximately 62 layers, which means that the loquat fruit produced one-third of its total cell number across the cortex after fruit set (Fig. 2A).The relative cell proliferation rate showed that most cell layers were produced during early development from 0 to 42 DPA (Fig. 2B). The cells began a slow enlargement appoximately 2 weeks after anthesis until 42 or 49 DPA, and afterwards, the cell size undulated by approximately 2.6 × 103 μm2 for weeks. The cells then entered a fastcell expansion period with a quick cell area increase from 80 to 119 DPA, and they grew into the final size at 119 DPA (Fig. 2C). The analysis of the relative cell expansion rate in Fig. 2D reveals that there are two cell size enlargement phases, that is, a slow growth phase from P6 to P11 and a fast growth phase including P15, P17, P18 and P20 during mid and late fruit development. Compared to cell division, loquat fruits take more time for cell expansion throughout fruit development. The longer time needed for cell enlargement may contribute to the finding that the cells in the middle cortex of mature pomes are almost tens to hundreds of times larger than those at anthesis and before bloom (Fig. 2A and C).

2.2.3. Identification of cell process regulatory genes in loquats The amino acid sequences of cell program-related genes from tomato (Frary et al., 2000), Arabidopsis (Cebolla et al., 1999; Ormenese et al., 2004; Esmon et al., 2006; De Schutter et al., 2007; Lammens et al., 2008; De Almeida Engler et al., 2012; Kobayashi et al., 2015), apple (Malladi and Hirst 2010; Malladi and Johnson 2011) and pear (Hiwasa et al., 2003) were used to BLAST against the loquat whole genome database (SQ Lin, South China Agricultural University, China, Personal communication). For most homologues we focus for fruit development in apple and pear, there were not functionally verification or high-density expression detected, or even cloned yet. In this case, orthologous gene in model plant like tomato and Arabidopsis were used for gene query as well. For example, fw2.2 protein sequence from tomato was selected for the FW2.2-like genes were named after their similar roles to that of fw2.2 in cell division regulation and none rosaceous plant FW2.2-like has been functional verified. Accession numbers of the loquat genes were listed in Table A.1. 2.2.4. RNA extraction and qPCR Total RNA extraction and first strain cDNA synthesis were performed as previously described (Zhang et al., 2016). Quantitative real time PCR (qPCR) was performed in a LightCycler480 (Roche) using iTaq™ universal SYBR Green Supermix according to the manufacturer’s protocol (Bio-Rad, USA). The qPCR primers listed in Table A.2 were designed using a BatchPrimer3 program (You et al., 2008). The relative expression levels with three biological replicates of the target genes were calculated as reported previously (Livak and Schmittgen 2001) by normalizing them against the loquat endogenous reference gene EjACT (AB710173.1).

3.3. Developmental expression of cell proliferation regulation genes The Slfw2.2 amino acid sequence (Frary et al., 2000) was used for alignment within the loquat genome database (SQ Lin, South China Agricultural University, China, Personal communication). Two homologues were found in loquats and they were named EjFWL1 and EjFWL2. Quantitative PCR revealed that both EjFWL transcript levels peaked near fertilization, declined dramatically at 7 DPA and maintained stable transcript levels until 21 DPA. Very low expression levels 144

Scientia Horticulturae 224 (2017) 142–149

W. Su et al.

Fig. 2. Cell proliferation and cell expansion during loquat fruit morphogenesis. (A) Cell layers were determined as the cell number between the ovary wall and the peel. The inset presents a magnified view of the cell layers during early pome development. (B) The relative cell proliferation rate was determined during pome development. (C) The cell area of the loquat cortex was measured using a light microscope. The inset presents a magnified view of the cell size during early pome development. (D) The relative cell expansion rate was determined during pome development. The internal periods were set according to (B). Error bars represent the standard errors of the means of 5 sections.

anthesis, and it gradually decreased to its lowest level at 35 DPA. This expression level fluctuated up and down three times, with the sharpest increase occurring at P20 when the cell size dramatically increased (Fig. 2D). The expression of EjCCS52A2 remained rather low throughout fruit development, and EjCCS52 B showed an extremely low expression level at athesis. The highest expression level occured at −21 and 21 DPA, and it sharply decreased to the basal level during late fruit development (Fig. 3E). EjEXPAs were highly expressed during late fruit development stages while EjEXPA1expression level was remarkably lower than that of EjEXPA15 (Fig. 3F).

were maintained for both genes in the later developmental phase, and it is interesting that these genes showed a slight increase near fruit ripening (Fig. 3A). Five G2/M-specific genes were chosen to reveal the cell division intensity of the loquat cortex (Kobayashi et al., 2015). qPCR analyses over the course of fruit development indicated that all these genes were highly expressed during early fruit development (during 7–35 DPA), and the transcript level of EjCYCB1;2 peaked earliest with that of EjCDKB2;2 peaking latest (Fig. 3B). For other stages after anthesis, all the G2/M-specific genes showed low transcript levels. However, high expression level also appeared before bloom when the receptacle size changed rapidly.

4. Discussion 3.4. Developmental expression of cell cycle exit regulation genes 4.1. Low cell division capacity results in a thin cortex and small size in loquat fruit

The expression level of EjWEE1 was relatively low during early development, and elevated at 42 DPA, when the cell layers were almost stable, and then it remained at a higher level throughout the later period than what it was at the early developmental stages (Figs. 2 A, 3 C). For EjKRP3 expression, there were two peaks throughout fruit development. The first peak emerged at 7 DPA (at the beginning of rapid cell production) and the later one emerged at 63 DPA when the cortex reached its ultimate cell number (Figs. 2 B, 3 D).

As a pome (Maloideae, Rosaceae), loquat fruit (Fig. A.1 in Supplementary material) may present a few seeds of similar size as in stone fruits (Amygdaloideae, Rosaceae). For this reason, (Blumenfeld 1980) once believed that there is a question as to whether loquat fits better into a single or double S model. Other than the pattern followed by peaches (Ognjanov et al., 1995), the diameter inspection in this work indicates that loquat fruit grows in a single S model (Fig. 1B) similar to that of other pomes such as apples (Bain and Robertson 1951; Malladi and Johnson 2011) and pears (Zhang et al., 2006) and in accordance with the results reported by (Ding and Zhang 1988) and (Cuevas et al., 2003).

3.5. Developmental expression of cell growth regulation genes CCS52 s displayed distinguishable expression patterns during fruit development. A high EjCCS52A1 expression level appeared soon after 145

Scientia Horticulturae 224 (2017) 142–149

W. Su et al.

Fig. 3. Expression analyses of cell cycle and cell expansion regulation genes in fruit. (A) EjFWL1 and EjFWL2; the inset presents a magnified view of EjFWL1. (B) Five G2/M-specific genes: EjKNOLLE, EjCDKB1;2, EjCYCB1;2, EjCDKB2;2 and EjEDE1. (C) EjWEE1; (D) EjKRP3. (E) EjCCS52A1, EjCCS52A2 and EjCCS52B; and (F) EjEXPA1 and EjEXPA15, the inset presents a magnified view of EjEXPA1.

total cell number across the cortex after fruit set (Fig. 2A). Comparisons showed that cell number increase after anthesis in loquats (Ding and Zhang 1988) and apples (Bain and Robertson 1951; Malladi and Hirst 2010; Malladi and Johnson 2011). We discovered that the loquat only proliferated by one-third of its total cell number during fruit development while mature apples could increase their cortex cell numbers to 5 or more times that of the anthesis stage receptacle. A large amount of variation in the cell number of cortex might be an important reason why apple sizes (especially the fleshy parts) were generally larger than those of loquats. For Malus species, a combination of a greater cell division capacity and an enhanced degree of cell enlargement are involved in the increase in the fruit size (Harada et al., 2005). However, shifts in the ability of

Cellular observations by (Cuevas et al., 2003) and (Ding and Zhang 1988) showed that there was little cell layer increase after fruit set during early development, and there was a rapid increase months later. By contast, our work suggested that the cell division began a rapid increase soon after anthesis. The weather maybe one of the most important factors leading to cell division delay, because it was really cold in Wuhan (−7.5 °C, 11th December 1985, Hubei Province) when the loquats bloomed, whereas in Guangzhou, it was much warmer (20–30 °C, 19th November 2015). In addition, unclear sample collection stages after anthesis in the prior study may also contribute to the discrepancy in the cell division pattern. Compared to the early stages, there were few cell increases in ripen loquat in both (Ding and Zhang 1988) and our work, i.e., the loquat fruit produced a small portion of its 146

Scientia Horticulturae 224 (2017) 142–149

W. Su et al.

Fig. 4. Cellular scenario for loquat fruit development. The time axis is represented by numerous boxes to show the time interval during two measurement points. Diverse line widths after anthesis represent cell division or cell expansion intensity. Sub-stages of cell division are presented with different coloured lines.

after fertilization from 7 to 28 or 35 DPA over high cell production period (Fig. 3B). The advanced high transcript of EjFWLs to G2/Mspecific genes indicated that EjFWLs might be cell division repressors of loquat, and conservative to PbFWLs in pear, whose expression levels were shown to be negatively correlated to fruit sizes (Tian et al., 2016). However, it is interesting that there was a large shift in timing of the peak expression between tomatoes and loquats, with the peak FW2.2 of tomatoes occurring approximately 6–8 DPA or later (Cong et al., 2002), while the EjFWLs peak expression level occurred at blossom. To the best of our knowledge, the ovary wall would restart cell division after pollination and fertilization, but pome receptacles continued to change along with flower opening even without fertilization. Similar cell process may emerge during apple fruit development, as genes like MdCYCB3;1 and MdCYCD3;1 exhibited peak expression at bloom or immediately after bloom (Malladi and Johnson 2011). This finding implies that EjFWLs may participate in other cellular processes, in addition of negative cell division regulatinon, such as cell division orientation regulation during pome morphogenesis. The additional functions of EjFWLs in pomes required further study in vivo.

cells to divide rather than enlarge have been reported, and they primarily contribute to fruit size during Pyrus pyrifolia evolution (Zhang et al., 2006). The investigations above show that cell division and cell enlargement might function individually or may cooperate with one another to determine the fruit size. Cultivated loquat fruit cells were of similar sizes as domesticated apples, which demonstrates that the size variation may primarily be related to the cell number (Malladi and Johnson 2011). A further study of the cell division capacity throughout apple fruit development illuminated that cell division in the cortex ceased by three weeks after full blossom, and afterwards, there was very little cell proliferated (Bain and Robertson 1951; Malladi and Hirst 2010; Malladi and Johnson 2011). In loquat fruits, cell division was initiated after blossoming and rapid division continued for 5–6 weeks (Fig. 2A). The comparisons indicated that it is the cell division rate rather than the division duration that inhibits the loquat cortex cell proliferation capacity, resulting in a thinner cortex. Additionally, the cell size contributes less than the cell number on cortex width. According to the cell observations in this work, a cellular scenario of loquat fruit morphogenesis was established and shown in Fig. 4. The fruit development could be divided into a cell division phase and a cell expansion phase with an overlap between 14 and 42 DPA. According to the cell proliferation activity and morphogenesis phase, this period might be divided into 4 sub-stages, namely the flowerbud swelling stage, anthesis preparation stage, fertilization and cell division initiation stage and peak cell proliferation stage. Cell division overlaps cell expansion from 14 to 42 DPA. The cell expansion phase covers a long period that includes the cell growth stage, expansion initiation stage and peak cell expansion stage based on the cell expansion rate at different stages.

4.3. EjWEE1 and EjKRP3 participate in cell cycle exit regulation A study in Arabidopsis showed that the induction of WEE1 expression would repress plant growth by arresting the cell cycle at the G2 phase (De Schutter et al., 2007). Additional data showed that removal at mitosis by WEE1 occurs via the 26S proteasome machinery (Cook et al., 2013). During fruit development, EjWEE1 expression increased dramatically at 42 DPA when cell division was almost stopped, and afterwards it stayed at a higher level during late fruit development when cell division stopped and cell sizes increased (Fig. 3C).This was identical to MdWEE1 as it peaked expression around cell division ending (Malladi and Johnson 2011) and implies that EjWEE1 might participate in cell cycle exit. There are seven KRPs in Arabidopsis, and KRP3 might be involved in both the establishment of polyploidy and cell progression regulation (Ormenese et al., 2004). This cell cycle controlling ability has also been demonstrated in rice, with OsKRP3 inhibited rice CDK in the syncytial endosperm (Mizutani et al., 2009). The expression pattern of EjKRP3 in loquat showed that it was strongly expressed at 7 DPA and decreased significantly at 14 DPA according to the same pattern of OsKRP3 in rice caryopsis (Mizutani et al., 2010) and MdKRP5 during apple fruit development (Malladi and Johnson 2011). For loquat fruit, there was another slightly sharp increase of EjKRP3 expression level at 63 DPA (Fig. 3D). In addition, expression of several cell-cycle inhibitors, such as MdWEE1 and MdKRPs was sharply

4.2. EjFWLs act as negative regulators of cell proliferation in early fruit development FW2.2 is a negative regulator of cell division early in fruit development (Frary et al., 2000). The transcript level was very low at anthesis in tomatoes and a high expression level appeared around the peak cell division phase. In large- and small-fruited tomatoes from nearly isogenic lines, the heterochronic FW2.2 transcript levels are associated with the fruit weight variations (Cong et al., 2002). Both EjFWLs were highly expressed from 0 to 21 DPA, the period with the highest cell proliferation capacity in this study, which is consistent with that of tomatoes (Figs. 3 A; 2 A–B). Consistent with the expression pattern of MdCDKB1;2, a positive cell cycle regulator (Malladi and Johnson 2011), the G2/M-specific genes showed high transcript levels 147

Scientia Horticulturae 224 (2017) 142–149

W. Su et al.

Acknowledgements

enhanced to block the cell-cycle cascade at the G1/S and G2/M transition points, as fruits response to heat stress (Flaishman et al., 2015). In this study, the EjWEE1expression level sharply increased near the end of cell division, and the EjKRP3 expression pattern was strongly correlated with cell division re-entry and cell expansion initiation in the loquat cortex. These expression profiles suggest that EjWEE1may be involved in cell cycle exit and EjKRP3 would participate in both cell progression control and polyploidy establishment.

We thank Dr. Yixun Yu and Dr. Biao Lai for their critical comments on the manuscript, and Yuanyuan Jiang for assistance with the cell measurements. This work was supported by the State Key Laboratory for Conservation and Utilization of Subtropical Agro-Bioresources [No. 201504010028，from the Guangzhou municipal government] and the Innovation and Utilization for Germplasm Resources of Guangdong [No. 2015A030303015].

4.4. Endoreduplication independent- and endoreduplication dependent- cell expansion cooperate during fruit cell enlargement

Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.scienta.2017.06.012.

Cells are known to exit the mitotic cycle upon differentiation or in response to stress cues. EjWEE1 and EjKRP3 not only participate in cell cycle exit but also participate in polyploidy establishment for cell differentiation (De Veylder et al., 2007). Mechanisms such as endoreduplication-independent cell expansion and endoreduplicationdependent cell expansion contain many sub-clusters for regulating cell size enlargement (Chevalier et al., 2014).Vacuole enlargement alters cell size according to the acid growth balloon theory; CCS52s induced DNA polyploidy establishment and cell homeostasis followed endoreduplication-dependent cell expansion (i.e. the karyoplasmic ratio theory). In our study, EjCCS52s showed diverse expression patterns, and the mRNA level of EjCCS52A1 peaked at anthesis and late development. EjCCS52B highly expressed at 21 DPA and was positive correlated with the pattern of cell division, while a low EjCCS52A2 transcript level was maintained throughout fruit development (Fig. 3E). The distinguishable expression patterns imply that EjCCS52s may act redundantly or complementarily with one another. Consistent with our speculation, previous work by (Baloban et al., 2013) showed that AtCCS52A isoforms control endoreduplication and plant size in a complementary manner. In both apple (Malladi and Hirst 2010) and pear (Wang et al., 2015), DNA content was showed to be related to fruit size diversity, and PcCCS52A may play roles in the occurrence and persistence of DNA reduplication in the receptacles (Hanada et al., 2015). Moreover, endoreduplication was initiated by CCS52A, playing a crucial role in tomato fruit growth (Mathieu-Rivet et al., 2010). These findings propelled us to believe that endoreduplication-dependent cell expansion might be involved in loquat fruit growth to some extent. Expansins are key regulators of wall extension (they are involved in the acid cell growth theory) during growth (Chevalier et al., 2014). A transcription analysis in pear (Hiwasa et al., 2003) and grape (Ishimaru et al., 2007) revealed that the expression levels of EXPAs are strongly correlated with fruit cell size enlargement. Both EjEXPA1 and EjEXPA15 showed higher transcription levels during the cell expansion phase than the cell division phase in loquat fruit (Fig. 3F). The strong positive coincidence of EjEXPA expression and cell size enlargement suggests that endoreduplication-independent cell expansion (acid cell growth) plays a crucial role in loquat cell size regulation.

References Bain, J.M., Robertson, R.N., 1951. The physiology of growth in apple fruits. I. Cell size, cell number, and fruit development. Aus. J. Sci. Res. 4, 75–107. Baloban, M., Vanstraelen, M., Tarayre, S., Reuzeau, C., Cultrone, A., Mergaert, P., Kondorosi, E., 2013. Complementary and dose-dependent action of AtCCS52A isoforms in endoreduplication and plant size control. New Phytol. 198, 1049–1059. Blumenfeld, A., 1980. Fruit growth of loquat. J. Am. Soc. Hortic. Sci. 105, 747–750. Cao, K., Zhao, P., Zhu, G., Fang, W., Chen, C., Wang, X., Wang, L., 2016. Expansin genes are candidate markers for the control of fruit weight in peach. Euphytica 210, 441–449. Cebolla, A., Vinardell, J.M., Kiss, E., Olah, B., Roudier, F., Kondorosi, A., Kondorosi, E., 1999. The mitotic inhibitor ccs52 is required for endoreduplication and ploidy-dependent cell enlargement in plants. Embo. J. 18, 4476–4484. Chakrabarti, M., Zhang, N., Sauvage, C., Munos, S., Blanca, J., Canizares, J., Diez, M.J., Schneider, R., Mazourek, M., McClead, J., Causse, M., van der Knaap, E., 2013. A cytochrome P450 regulates a domestication trait in cultivated tomato. Proc. Natl. Acad. Sci. U. S. A. 110, 17125–17130. Chen, Z., Zhao, J., Qin, Y., Hu, G., 2016. Study on the graft compatibility between ‘Jingganghongnuo’ and other litchi cultivars. Sci. Hortic. 199, 56–62. Chevalier, C., Bourdon, M., Pirrello, J., Cheniclet, C., Gevaudant, F., Frangne, N., 2014. Endoreduplication and fruit growth in tomato: evidence in favour of the karyoplasmic ratio theory. J. Exp. Bot. 65, 2731–2746. Cong, B., Liu, J.P., Tanksley, S.D., 2002. Natural alleles at a tomato fruit size quantitative trait locus differ by heterochronic regulatory mutations. Proc. Natl. Acad. Sci. U. S. A. 99, 13606–13611. Cong, B., Barrero, L.S., Tanksley, S.D., 2008. Regulatory change in YABBY-like transcription factor led to evolution of extreme fruit size during tomato domestication. Nat. Genet. 40, 800–804. Conlon, I., Raff, M., 1999. Size control in animal development. Cell 96, 235–244. Cook, G.S., Gronlund, A.L., Siciliano, I., Spadafora, N., Amini, M., Herbert, R.J., Bitonti, M.B., Graumann, K., Francis, D., Rogers, H.J., 2013. Plant WEE1 kinase is cell cycle regulated and removed at mitosis via the 26S proteasome machinery. J. Exp. Bot. 64, 2093–2106. Cuevas, J., Salvador-Sola, F.J., Gavilan, J., Lorente, N., Hueso, J.J., Gonzalez-Padierna, C.M., 2003. Loquat fruit sink strength and growth pattern. Sci. Hortic. 98, 131–137. Dash, M., Malladi, A., 2012. The AINTEGUMENTA genes, MdANT1 and MdANT2, are associated with the regulation of cell production during fruit growth in apple (Malus × domestica Borkh.). BMC Plant Biol. 12, 98. De Almeida Engler, J., Kyndt, T., Vieira, P., Van Cappelle, E., Boudolf, V., Sanchez, V., Escobar, C., De Veylder, L., Engler, G., Abad, P., Gheysen, G., 2012. CCS52 and DEL1 genes are key components of the endocycle in nematode-induced feeding sites. Plant J. 72, 185–198. De Schutter, K., Joubes, J., Cools, T., Verkest, A., Corellou, F., Babiychuk, E., Van Der Schueren, E., Beeckman, T., Kushnir, S., Inze, D., De Veylder, L., 2007. Arabidopsis WEE1 kinase controls cell cycle arrest in response to activation of the DNA integrity checkpoint. Plant Cell 19, 211–225. De Veylder, L., Beeckman, T., Inz, D., 2007. The ins and outs of the plant cell cycle. Nat. Rev. Mol. Cell Bio. 8, 655–665. Denne, M.P., 1960. The growth of apple fruitlets, and the Effect of Early Thinning on Fruit Development. Ann. Bot. 24, 397–406. Ding, C., Zhang, H., 1988. The effect of plant hormones on fruit development of loquat. Acta Horti. Sinica. 15, 148–154. Esmon, C.A., Tinsley, A.G., Ljung, K., Sandberg, G., Hearne, L.B., Liscum, E., 2006. A gradient of auxin and auxin-dependent transcription precedes tropic growth responses. Proc. Natl. Acad. Sci. U. S. A. 103, 236–241. Fischer, M., DeCaprio, J.A., 2015. Does Arabidopsis thaliana DREAM of cell cycle control? EMBO J. 34, 1987–1989. Flaishman, M.A., Peles, Y., Dahan, Y., Milo-Cochavi, S., Frieman, A., Naor, A., 2015. Differential response of cell-cycle and cell-expansion regulators to heat stress in apple (malus domestica) fruitlets. Plant Sci. 233, 82–94. Frary, A., Nesbitt, T.C., Frary, A., Grandillo, S., van der Knaap, E., Cong, B., Liu, J., Meller, J., Elber, R., Alpert, K.B., Tanksley, S.D., 2000. fw2.2: A quantitative trait locus key to the evolution of tomato fruit size. Science 289, 85–88. Gillaspy, G., Ben-David, H., Gruissem, W., 1993. Fruits: a developmental perspective.

5. Conclusion Loquat fruit grows in a single S model under complicated regulation of cell division and cell expansion. Cellular observations demonstrate that weak cell division capacity appeared after anthesis and results into small cell layer increase across cortex during fruit development. Cell proliferation capacity seems to be an important factor which affects large-size fruit formation. Fruit size increased more greatly during cell expansion phase compared to that during cell division phase. Gene expression analyses revealed that G2/M-specific genes peaked mRNA levels during high cell division phases are markers of fruit cell division, and EjFWLs may be negative regulators of cell division. EjWEE1 and EjKRP3 are suggested to participate in both cell cycle exit and polyploidy establishment. Highly expressed during peak cell expansion phase, EjEXPAs are candidate marker genes which would present fruit cell expansion intensity. 148

Scientia Horticulturae 224 (2017) 142–149

W. Su et al.

Mizutani, M., Naganuma, T., Tsutsumi, K., Saitoh, Y., 2010. The syncytium-specific expression of the Orysa;KRP3 CDK inhibitor: implication of its involvement in the cell cycle control in the rice (Oryza sativa L.) syncytial endosperm. J. Exp. Bot. 61, 791–798. Munos, S., Ranc, N., Botton, E., Berard, A., Rolland, S., Duffe, P., Carretero, Y., Le Paslier, M.C., Delalande, C., Bouzayen, M., Brunel, D., Causse, M., 2011. Increase in tomato locule number is controlled by two single-nucleotide polymorphisms located near WUSCHEL. Plant Physiol. 156, 2244–2254. Nicolas, P., Lecourieux, D., Gomes, E., Delrot, S., Lecourieux, F., 2013. The grape berryspecific basic helix-loop-helix transcription factor VvCEB1 affects cell size. J. Exp. Bot. 64, 991–1003. Ognjanov, V., Vujanic-Varga, D., Misic, P.D., Veregbaranji, I., Macet, K., Tesovic, Z., Krstic, M., Petrovic, N., 1995. Anatomical and biochemical studies in persisting and shedding seeds during the development of peach fruits. Sci. Hortic. 64, 33–48. Ormenese, S., De Almeida Engler, J., De Groodt, R., De Veylder, L., Inzea, D., Jacqmard, A., 2004. Analysis of the spatial expression pattern of seven kip related proteins (KRPs) in the shoot apex of Arabidopsis thaliana. Ann. Bot. 93, 575–580. Schroeder, C.A., 1953. Growth and development of the fuerte avocado fruit. Proceedings of the American Society For Horticultural Science 61, 103–109. Si, L., Chen, J., Huang, X., Gong, H., Luo, J., Hou, Q., Zhou, T., Lu, T., Zhu, J., Shangguan, Y., Chen, E., Gong, C., Zhao, Q., Jing, Y., Zhao, Y., Li, Y., Cui, L., Fan, D., Lu, Y., Weng, Q., Wang, Y., Zhan, Q., Liu, K., Wei, X., An, K., An, G., Han, B., 2016. OsSPL13 controls grain size in cultivated rice. Nat. Genet. 48, 447–456. Suzuki, H., Oshita, E., Fujimori, N., Nakajima, Y., Kawagoe, Y., Suzuki, S., 2015. Grape expansins, VvEXPA14 and VvEXPA18 promote cell expansion in transgenic Arabidopsis plant. Plant Cell Tissu. Organ Cult. 120, 1077–1085. Tanksley, S.D., 2004. The genetic, developmental, and molecular bases of fruit size and shape variation in tomato. Plant Cell 16, S181–S189. Testone, G., Condello, E., Di Giacomo, E., Nicolodi, C., Caboni, E., Rasori, A., Bonghi, C., Bruno, L., Bitonti, M.B., Giannino, D., 2015. The KNOTTED-like genes of peach (Prunus persica L: Batsch) are differentially expressed during drupe growth and the class 1 KNOPE1 contributes to mesocarp development. Plant Sci. 237, 69–79. Tian, J., Zeng, B., Luo, S., Li, X., Wu, B., Li, J., 2016. Cloning: localization and expression analysis of two fw2.2-like genes in small- and large-fruited pear species. J. Integr. Agr. 15, 282–294. Wang, S., Wu, K., Yuan, Q., Liu, X., Liu, Z., Lin, X., Zeng, R., Zhu, H., Dong, G., Qian, Q., Zhang, G., Fu, X., 2012. Control of grain size, shape and quality by OsSPL16 in rice. Nat. Genet. 44, 950–954. Wang, L., He, L., Li, J., Zhao, J., Li, Z., He, C., 2014. Regulatory change at Physalis Organ Size 1 correlates to natural variation in tomatillo reproductive organ size. Nat. Commun. 5, 4271. Wang, X., Wang, H., Shi, C., Zhang, X., Duan, K., Luo, J., 2015. Morphological, cytological and fertility consequences of a spontaneous tetraploid of the diploid pear (pyrus pyrifolia nakai) cultivar ‘cuiguan’. Sci. Hortic. 189, 59–65. You, F.M., Huo, N., Gu, Y.Q., Luo, M., Ma, Y., Hane, D., Lazo, G.R., Dvorak, J., Anderson, O.D., 2008. BatchPrimer3: A high throughput web application for PCR and sequencing primer design. BMC Bioinf. 9, 253. Zhang, C., Tanabe, K., Wang, S., Tamura, F., Yoshida, A., Matsumoto, K., 2006. The impact of cell division and cell enlargement on the evolution of fruit size in Pyrus pyrifolia. Ann. Bot.-London. 98, 537–543. Zhang, L., Yu, H., Lin, S., Gao, Y., 2016. Molecular characterization of FT and FD homologs from Eriobotrya deflexa nakai forma koshunensis. Front. Plant Sci. 7, 8.

Plant Cell 5, 1439–1451. Hanada, T., Nashima, K., Kato, M., Takashina, T., Ikeda, K., Sakamoto, Y., Takahashi, H., Nakazono, M., Oikawa, A., Shiratake, K., Isuzugawa, K., 2015. Molecular cloning and expression analysis of the wee1 and ccs52a genes in european pear (pyrus communis l.) and their possible roles in a giant fruit mutant. J. Hortic. Sci. Biotech. 90 (5), 511–517. Harada, T., Kurahashi, W., Yanai, M., Wakasa, Y., Satoh, T., 2005. Involvement of cell proliferation and cell enlargement in increasing the fruit size of Malus species. Sci. Hortic. 105, 447–456. Hepworth, J., Lenhard, M., 2014. Regulation of plant lateral-organ growth by modulating cell number and size. Curr. Opin. Plant Biol. 17, 36–42. Hiwasa, K., Rose, J.K.C., Nakano, R., Inaba, A., Kubo, Y., 2003. Differential expression of seven a-expansin genes during growth and ripening of pear fruit. Phyiol. Plant. 117, 564–572. Ishimaru, M., Smith, D.L., Gross, K.C., Kobayashi, S., 2007. Expression of three expansin genes during development and maturation of Kyoho grape berries. J. Plant Physiol. 164, 1675–1682. Joubes, J., Phan, T., Just, D., Rothan, C., Bergounioux, C., Raymond, P., Chevalier, C., 1999. Molecular and biochemical characterization of the involvement of cyclin-dependent kinase A during the early development of tomato fruit. Plant Physiol. 121, 857–869. Kobayashi, K., Suzuki, T., Iwata, E., Nakamichi, N., Suzuki, T., Chen, P., Ohtani, M., Ishida, T., Hosoya, H., Mueller, S., Leviczky, T., Pettko-Szandtner, A., Darula, Z., Iwamoto, A., Nomoto, M., Tada, Y., Higashiyama, T., Demura, T., Doonan, J.H., Hauser, M., Sugimoto, K., Umeda, M., Magyar, Z., Boegre, L., Ito, M., 2015. Transcriptional repression by MYB3R proteins regulates plant organ growth. Embo. J. 34, 1992–2007. Lammens, T., Boudolf, V., Kheibarshekan, L., Zalmas, L.P., Gaamouche, T., Maes, S., Vanstraelen, M., Kondorosi, E., La Thangue, N.B., Govaerts, W., Inze, D., De Veylder, L., 2008. Atypical E2F activity restrains APC/C-CCS52A2 function obligatory for endocycle onset. Proc. Natl. Acad. Sci. U. S. A. 105, 14721–14726. Li, Y., Fan, C., Xing, Y., Jiang, Y., Luo, L., Sun, L., Shao, D., Xu, C., Li, X., Xiao, J., He, Y., Zhang, Q., 2011. Natural variation in GS5 plays an important role in regulating grain size and yield in rice. Nat. Genet. 43, 1266–1269. Liu, J., Hua, W., Hu, Z., Yang, H., Zhang, L., Li, R., Deng, L., Sun, X., Wang, X., Wang, H., 2015. Natural variation in ARF18 gene simultaneously affects seed weight and silique length in polyploid rapeseed. Proc. Natl. Acad. Sci. 112, E5123–E5132. Livak, K.J., Schmittgen, T.D., 2001. Analysis of relative gene expression data using realtime quantitative PCR and the 2−△△CT method. Methods 25, 402–408. Malladi, A., Hirst, P.M., 2010. Increase in fruit size of a spontaneous mutant of ‘Gala' apple (Malus × domestica Borkh.) is facilitated by altered cell production and enhanced cell size. J. Exp. Bot. 61, 3003–3013. Malladi, A., Johnson, L.K., 2011. Expression profiling of cell cycle genes reveals key facilitators of cell production during carpel development, fruit set, and fruit growth in apple (Malus × domestica Borkh.). J. Exp. Bot. 62, 205–219. Mao, H., Sun, S., Yao, J., Wang, C., Yu, S., Xu, C., Li, X., Zhang, Q., 2010. Linking differential domain functions of the GS3 protein to natural variation of grain size in rice. Proc. Natl. Acad. Sci. U. S. A. 107, 19579–19584. Mathieu-Rivet, E., Gevaudant, F., Sicard, A., Salar, S., Do, P.T., Mouras, A., Fernie, A.R., Gibon, Y., Rothan, C., Chevalier, C., Hernould, M., 2010. Functional analysis of the anaphase promoting complex activator CCS52A highlights the crucial role of endoreduplication for fruit growth in tomato. Plant J. 62, 727–741.

149