The APC/C subunit 10 plays an essential role in cell proliferation during leaf development

Authors

  • Nubia B. Eloy,

    1. Department of Plant Systems Biology, VIB, 9052 Gent, Belgium
    2. Department of Plant Biotechnology and Bioinformatics, Ghent University, 9052 Gent, Belgium
    Search for more papers by this author
  • Marcelo de Freitas Lima,

    1. Laboratório de Biologia Molecular de Plantas, Instituto de Bioquímica Médica, CCS, Cidade Universitária – Ilha do Fundão, CEP 21941-590, Rio de Janeiro, RJ, Brasil
    2. Departamento de Química, Universidade Federal Rural do Rio de Janeiro, CEP 23890-000, Seropédica, RJ, Brasil
    Search for more papers by this author
  • Daniël Van Damme,

    1. Department of Plant Systems Biology, VIB, 9052 Gent, Belgium
    2. Department of Plant Biotechnology and Bioinformatics, Ghent University, 9052 Gent, Belgium
    Search for more papers by this author
  • Hannes Vanhaeren,

    1. Department of Plant Systems Biology, VIB, 9052 Gent, Belgium
    2. Department of Plant Biotechnology and Bioinformatics, Ghent University, 9052 Gent, Belgium
    Search for more papers by this author
  • Nathalie Gonzalez,

    1. Department of Plant Systems Biology, VIB, 9052 Gent, Belgium
    2. Department of Plant Biotechnology and Bioinformatics, Ghent University, 9052 Gent, Belgium
    Search for more papers by this author
  • Liesbeth De Milde,

    1. Department of Plant Systems Biology, VIB, 9052 Gent, Belgium
    2. Department of Plant Biotechnology and Bioinformatics, Ghent University, 9052 Gent, Belgium
    Search for more papers by this author
  • Adriana S. Hemerly,

    1. Laboratório de Biologia Molecular de Plantas, Instituto de Bioquímica Médica, CCS, Cidade Universitária – Ilha do Fundão, CEP 21941-590, Rio de Janeiro, RJ, Brasil
    Search for more papers by this author
  • Gerrit T. S. Beemster,

    1. Department of Plant Systems Biology, VIB, 9052 Gent, Belgium
    2. Department of Plant Biotechnology and Bioinformatics, Ghent University, 9052 Gent, Belgium
    3. Department of Biology, University of Antwerp, 2020 Antwerpen, Belgium
    Search for more papers by this author
  • Dirk Inzé,

    Corresponding author
    1. Department of Plant Systems Biology, VIB, 9052 Gent, Belgium
    2. Department of Plant Biotechnology and Bioinformatics, Ghent University, 9052 Gent, Belgium
      (fax +32 9 3313809; e-mail dirk.inze@psb.vib-ugent.be).
    Search for more papers by this author
    • These authors contributed equally to this work.

  • Paulo C. G. Ferreira

    1. Laboratório de Biologia Molecular de Plantas, Instituto de Bioquímica Médica, CCS, Cidade Universitária – Ilha do Fundão, CEP 21941-590, Rio de Janeiro, RJ, Brasil
    Search for more papers by this author
    • These authors contributed equally to this work.


(fax +32 9 3313809; e-mail dirk.inze@psb.vib-ugent.be).

Summary

The largest E3 ubiquitin-ligase complex, known as anaphase-promoting complex/cyclosome (APC/C), regulates the proteolysis of cell cycle regulators such as CYCLIN B and SECURIN that are essential for sister-chromatid separation and exit from mitosis. Despite its importance, the role of APC/C in plant cells and the regulation of its activity during cell division remain poorly understood. Here, the Arabidopsis thaliana APC/C subunit APC10 was characterized and shown to functionally complement an apc10 yeast mutant. The APC10 protein was located in specific nuclear bodies, most probably resulting from its association with the proteasome complex. An apc10 Arabidopsis knockout mutant strongly impaired female gametogenesis. Surprisingly, constitutive overexpression of APC10 enhanced leaf size. Through kinematic analysis, the increased leaf size was found to be due to enhanced rates of cell division during the early stages of leaf development and, at the molecular level, by increased APC/C activity as measured by an amplification of the proteolysis rate of the mitotic cyclin, CYCB1;1.

Introduction

In contrast to mammals, most plant development is post-embryonic. During embryogenesis, the root–shoot axis is established, but all organs characteristic for plants result from iterative organ initiation and development at the different meristems. The sessile lifestyle of plants requires a highly flexible control of cell division and growth to achieve the final architecture. Two effector systems, cell proliferation and cell expansion, determine the size and shape of mature organs. In many plants, the onset of cell expansion coincides with endoreduplication (Beemster et al., 2005; De Veylder et al., 2007), a variant of the normal cell cycle in which cells stop dividing although they keep growing and replicating their DNA (Gutierrez, 2005; Inzé and De Veylder, 2006; Ishida et al., 2009).

Progression through the cell division cycle needs the precise temporal and spatial control of regulatory proteins to duplicate the DNA correctly and to distribute the duplicated genomes to the two daughter cells during mitosis. In plants, as in other eukaryotes, regulation of cell cycle progression depends on the activity of cyclin-dependent kinases (CDKs) (De Veylder et al., 2007). The activity of CDKs is crucial at both the G1 to S and the G2 to M phase transitions and it is regulated through association with regulatory subunits, designated cyclins, by phosphorylation and dephosphorylation, interaction with inhibitory proteins, and proteolysis (Inzé and De Veylder, 2006; Sullivan and Morgan, 2007). Rapid proteolysis of cell cycle regulatory proteins occurs through the ubiquitin-mediated pathway that ensures the irreversibility of the cell cycle progression. The specificity of ubiquitin-dependent proteolysis is achieved at the level of substrate ubiquitination, which provides the E3 ligase enzymes with key roles in several cellular processes, especially in the cell cycle. The two principal E3 ubiquitin ligases involved in cell cycle control are the Skpl-cullin-F-box-protein (SCF) and the anaphase-promoting complex/cyclosome (APC/C) (Vodermaier, 2004). The SCF complex plays an essential role at the G1 to S phase checkpoint by degrading cell cycle-dependent kinase inhibitors (CKIs), such as SIC1 in yeast (Dirick et al., 1995) and Kip-related proteins (KRPs) in plants (Verkest et al., 2005; Ren et al., 2008). In contrast, APC/C functions essentially at the G2 to M transition. For example, to exit from mitosis, mitotic cyclins have to be destroyed by the APC/C in all organisms (Morgan and Roberts, 2002; Marrocco et al., 2010). The APC/C is one of the most complex molecular machines known to catalyze ubiquitination reactions that direct proteins for degradation by the 26S proteasome (Glotzer et al., 1991) and, depending on the organism, can contain more than a dozen subunits (Peters, 2002; Yoon et al., 2002; Capron et al., 2003a; Marrocco et al., 2010). This complexity is peculiar because many other ubiquitin ligases, including SCF, only consist of one or a few subunits (Harper et al., 2002). The targets of APC/C are selected through recognition of destruction motifs, predominantly the destruction (D) box and KEN box (Glotzer et al., 1991; Pfleger and Kirschner, 2000).

APC10, a subunit of the APC/C, was first identified in a genetic screen in yeast for mutants defective in the degradation of mitotic cyclins (Hwang and Murray, 1997) and, later, in budding yeast (Saccharomyces cerevisiae), fission yeast (Schizosaccharomyces pombe) and mammals (Kominami et al., 1998; Grossberger et al., 1999; Kurasawa and Todokoro, 1999). Substrate specificity of the APC/C complex is achieved by activator proteins, such as CDH1 and CDC20. Whereas CDC20 is needed for the transition from metaphase to anaphase, CDH1 regulates the exit from mitosis (Zachariae et al., 1998; Harper et al., 2002).

In budding yeast, APC10 is required for the interaction of substrates with CDH1, implying that it is a potential regulatory subunit that determines the APC/C activity by directing substrate recognition (Passmore et al., 2003). APC10 has recently been found to be part of a combined catalytic and substrate recognition module located within the central cavity of the APC/C that consists of CDH1, APC10 and the cullin domain of APC2 (da Fonseca et al., 2011).

The identification of the complete set of genes encoding the APC/C subunits in Arabidopsis suggests that the basic processes controlled by ubiquitin-mediated proteolysis in plants are similar to those in other eukaryotes (Capron et al., 2003a; Marrocco et al., 2010). The functional conservation of the APC/C complex has been strengthened by the demonstration that A-type and B-type cyclins are degraded in a D box-dependent manner, which is a structural nine-amino-acid sequence motif identified in the N-terminal domain of the targeted proteins (Genschik et al., 1998; Criqui et al., 2000).

Mutations in the APC/C subunit-encoding Arabidopsis genes CDC16 (APC6), APC2 and CDC27b (APC3b) have led to gametophytic and embryonic lethality as a consequence of the inability to degrade mitotic cyclins (Blilou et al., 2002; Capron et al., 2003b; Kwee and Sundaresan, 2003), confirming the functional importance of APC/C in plants. In addition, CDC27a (APC3a) and CDC27b (APC3b) double mutants are essential for female gametophyte development, and the CDC27b (APC3b) subunit has also been implicated in post-embryogenic differentiation at the meristems (Pérez-Pérez et al., 2008). Additionally, reduced levels of the genes encoding the APC6 and APC10 subunits in Arabidopsis provoke several defects in the vascular development (Marrocco et al., 2009). The plant CDH1 ortholog CCS52A has been shown to control endoreduplication in leaf petioles and nodules of Medicago truncatula (Cebolla et al., 1999; Vinardell et al., 2003) and in rosette leaves of Arabidopsis and to be essential for meristem maintenance in Arabidopsis roots (Lammens et al., 2008; Vanstraelen et al., 2009).

To unravel the functional role of individual APC/C subunits, we analyzed loss-of-function and gain-of-function mutants of APC10. Knockout apc10 mutants are shown to be female gametophytically lethal. In contrast, plants overexpressing APC10 develop larger leaves than the wild type due to enhanced cell division. The increase in proteolysis of the mitotic cyclin CYCB1;1 in the APC10-overexpressing plants indicates that improved growth is associated with enhanced APC/C activity.

Results

Identification of APC10 as a functional APC/C subunit

The Arabidopsis genome contains one gene (At2g18290) putatively encoding APC10 (Eloy et al., 2006). To verify whether this APC10 gene effectively encodes a functional protein, the coding sequence was inserted into a yeast expression vector and transformed into a temperature-sensitive apc10ts fission yeast (apc10-27) strain (Kominami et al., 1998).

The expression of the Arabidopsis APC10 was able to rescue the apc10ts phenotype at the restrictive temperature of 35°C, while the negative control containing the antisense construct was unable to grow under the same conditions (Figure 1). Additionally, when APC10 was used as a bait in a tandem affinity purification (TAP) experiment to identify associated proteins, the entire APC/C core complex was pulled down, supporting that APC10 is part of the APC/C complex in plants (Van Leene et al., 2010). Hence, from the yeast complementation and the TAP results, we can conclude that At2g18290 encodes a functional APC10 subunit.

Figure 1.

 Complementation of the temperature-sensitive fission yeast strain apc10ts (apc10-27) with the coding region of the Arabidopsis APC10.
Growth at permissive (25°C, left) or restrictive (35°C, right) temperature of an apc10ts strain carrying a vector with the antisense cDNA as control (APC10-pREPHAWG) or expressing the APC10 gene (APC10-pREPHAGW). The original mutant yeast strain apc10ts (apc10-27) was also grown on the plate.

APC10 expression and subcellular localization

The analysis of APC10 expression in a wide range of microarrays suggested that it is constitutively expressed during plant development and throughout the cell cycle. To analyze the expression pattern in plant tissues, a 2.9-kb fragment upstream of the ATG codon of APC10 was fused to the β-glucuronidase (GUS) reporter gene by means of a FAST vector (Shimada et al., 2010) and introduced into Arabidopsis plants through Agrobacterium tumefaciens-mediated transformation (Clough and Bent, 1998). Thirty independently transformed plants were obtained and six representative lines expressing the APC10 promoter–reporter construct were analyzed in detail. Staining for GUS activity revealed a strong and more or less constitutive expression in young leaves, hypocotyls, female organs and roots with the remarkable exception of the root apical region (Figure 2). A similar expression pattern was obtained when the 2.9-kb promoter fragment and the entire coding sequence, including the five introns, was fused at the stop codon to the GUS reporter and analyzed in transgenic plants (Figure S1). When leaf development progressed, the APC10-GUS expression gradually diminished, but remained high in vascular tissues (Figure 2). A cross-section through the shoot apical meristem (Figure 2b) showed that APC10 was expressed in all cells. However, reduced expression was observed at the tip of developing true leaves, most probably reflecting reduced expression of APC10 in cells that start to differentiate (Donnelly et al., 1999).

Figure 2.

 Expression of the pAPC10–GUS reporter gene at different developmental stages.
(a) Seedling 8 days after stratification (DAS).
(b) Transverse section of the shoot apical meristem (dark field). The pink dots represent GUS expression. The arrow points out the absence of expression.
(c) Mature leaf.
(d) Seedling 12 DAS.
(e) Ovules.
(f) Main root.
(g) Lateral root.

To determine the subcellular localization of the APC10 protein, it was stably produced as a C-terminal protein fused to the green fluorescent protein (GFP), driven by the cauliflower mosaic virus (CaMV) 35S promoter in Arabidopsis seedlings. The APC10–GFP fusion protein was functional because it was able to complement the fission yeast apc10ts strain at a non-permissive temperature (data not shown). In the root meristem, the APC10–GFP fluorescence was predominantly located in the nucleus, excluded from the nucleolus and, to a minor extent, also present in the cytoplasm. Several small fluorescent dots were observed inside the nucleus as well (Figure 3a, arrows). These punctuations resembled nuclear bodies, probably representing the association of APC10 with the proteasome complex (Tao et al., 2005). In the root elongation zone, the nuclear bodies could also be visualized, although to a lesser extent than in the root tip (Figure 3b,c). In leaf pavement cells and trichomes, the APC10–GFP fluorescence had the same pattern as in root cells, including the presence of nuclear bodies (Figure 3d,e).

Figure 3.

 Subcellular localization of APC10-GFP in Arabidopsis cells.
(a) Representative single optical section through an Arabidopsis root tip showing the predominantly nuclear localization and the association of APC10–GFP with nuclear bodies (arrows).
(b) Representative single optical section through the Arabidopsis root elongation zone showing the nuclear localization of APC10–GFP.
(c) High magnifications of (b) showing the association of APC10–GFP with nuclear bodies (arrows).
(d) Representative leaf pavement cell showing the APC10–GFP localization.
(e) Representative trichome showing the APC10–GFP localization in nuclear bodies (arrows). Bars = 10 μm (a–e) and 20 μm (b).

APC10 is an essential component of the Arabidopsis APC complex

To investigate the function of APC10 in plants, we analyzed the loss-of-function effect in Arabidopsis. Two T-DNA insertions lines located in the large fourth intron of APC10 were requested from the Nottingham Arabidopsis Stock Centre (NASC): apc10-2 (SALK_083989) and apc10-3 (SALK_043367) (Figure 4a). However, no homozygous plants could be obtained among the 60 plants analyzed for each line by genotyping. Upon examination, we found that immature siliques obtained by self-fertilization of the two heterozygous mutant lines analyzed, apc10-2+/− and apc10-3+/−, contained approximately 50% (145 out of 278 and 125 out of 257, respectively) aborted ovules, which were small, whitish and shrunken (Figure S2).

Figure 4.

 Characterization of apc10 mutant plants.
(a) Exon (black boxes) and intron (lines) structure of APC10. Untranslated regions are shown by grey boxes. The triangles indicate T-DNA insertion sites.
(b) Analysis of APC10 mRNA levels in 8- and 10-day-old apc10-1 mutant and wild-type plants. Total RNA was prepared from the first two leaves and amplified by reverse-transcription PCR with APC10-specific primers. All values were normalized against the expression level of housekeeping genes (see Experimental Procedures) and the expression levels were compared to the relative expression levels in Col-0. Data are means ± standard deviation (= 3).
(c) Leaf area calculation of 22-day-old plants grown in soil: Col-0 (black line) and apc10-1 (grey rectangles). The leaves with the higher numbers are the youngest. Cot = cotyledon. Data are means ± standard deviation (= 10).

Genetic analysis revealed that after self-fertilization the offspring of the apc10-2/+ heterozygous plants segregated with a statically significant 1:1 ratio of antibiotic-resistant (776) versus antibiotic-sensitive (825) seedlings, which strongly hints at a gametophytically lethal mutation. To determine whether the observed segregation was caused by male or female transmission defects, apc10-2/+ heterozygous and wild-type plants were crossed reciprocally. The transmission of the T-DNA insertion in the APC10 gene was followed by testing growth on kanamycin, for which the resistance is encoded by the T-DNA. When the apc10-2/+ mutant was used as the male donor, approximately 45% of the offspring seedlings were resistant to kanamycin (Table 1), indicating that transmission through the male gametophyte is unaffected; when the apc10-2+/− mutant was pollinated with wild-type pollen no kanamycin-resistant offspring could be found (Table 1). In conclusion, the above genetic analysis points to an essential role for APC10 in female gametogenesis.

Table 1.   Genetic analysis and reciprocal crosses of the apc10-2/+ mutant with the wild type
GenotypeKanRKanSR/ST-DNA transmission (%)
  1. Kan, kanamycin; R, resistant; S, sensitive.

apc10-2+/− self-fertilization7768250.9448
Wild type × apc10-2+/−871080.8045
apc10-2+/− × wild type0960.000

In contrast to the apc10-2 and apc10-3 alleles, a homozygous mutant was found harboring a T-DNA insertion at 49 bp upstream to the ATG of APC10 (SALK_123967), hereafter referred to as apc10-1 (Figure 4a). Interestingly, in the apc10-1 line, APC10 expression was not completely abolished. During early development (8 days after stratification, DAS), the APC10 mRNA levels in the first leaf pair of apc10-1 were on average 2.5-fold reduced (Figure 4b), but 2 days later, at 10 DAS, the mRNA levels were almost two-fold higher than those of the wild-type plants (Figure 4b). After 22 days, the apc10-1 plants strongly overexpressed APC10 (Figure 4b). Unexpectedly, analysis of 22-day-old apc10-1 plants grown on soil revealed that the leaf area was significantly larger than that of the wild-type plants, particularly in younger rosette leaves (Figure 4c). In the apc10-1 mutant, other phenotypes were also observed, such as somewhat abnormal cotyledons and altered flowers, but most probably not associated with the APC10 overexpression (see below).

In conclusion, our data show that the T-DNA insertion in the APC10 promoter enhances gene expression and suggest that the increased APC10 expression results in a positive effect on leaf growth and size.

Overexpression of APC10 increases cell division rate

To confirm that, based on the phenotype of apc10-1, the APC10 overexpression increased leaf size, transgenic Arabidopsis plants were generated that ectopically overexpressed APC10 under the control of the CaMV 35S constitutive promoter. Fifteen independent transformants harboring a single insertion locus and with increased expression levels of APC10 were selected (data not shown) and grown on soil to obtain homozygous plants. Quantitative (q) PCR analysis of two independent homozygous lines (APC10OE2.3 and APC10OE5.3) grown for 15 DAS (Figure 5a) confirmed the increased APC10 mRNA levels and a protein gel blot with an APC10 antibody revealed enhanced APC10 protein levels (Figure 5b). The leaf phenotypes of the APC10-overexpressing plants were analyzed and were shown to be very similar; hence, only the analysis of APC10OE5.3, hereafter designated APC10OE, is presented.

Figure 5.

 Expression analysis of APC10OE plants.
(a) Quantitative reverse transcription (RT)-PCR transcript analysis of the APC10 expression in wild-type control and two APC10OE plants. Total RNA prepared from 15-day-old plants was amplified by RT-PCR. All values were normalized against the expression level of the housekeeping genes (see Experimental Procedures). Data are means ± standard deviation (= 3).
(b) Protein gel blot analysis in plants of two different APC10OE lines and Col-0 15 days after stratification (DAS). The APC10 proteins levels were visualized with an APC10 antibody.
(c) Phenotypic characterization of APC10OE (grey bars) compared to Col-0 (black line) Arabidopsis plants. Data obtained with line APC10OE5.3 were similar to those of another line (APC10OE2.3; data not shown). Leaves were measured from 22-day-old plants grown in vivo.
(d) Representative picture from the measurements shown in (c).
(e) Representative picture from plants grown in vitro for 22 DAS.

A comparative phenotypic analysis of APC10OE and wild-type (Col-0) plants showed an increase in the leaf area without effect on the temporal progression of leaf development. No other phenotypes, such as abnormal cotyledons or abnormal flowers, were observed in any of the APC10OE plants. To verify the positive effect of APC10 overexpression on leaf growth, the area of all leaves from the APC10OE line and Col-0 control were determined from 3-week-old soil-grown plants. The area of almost all leaves (except the first two and the two cotyledons) was significantly larger in the transgenic plants compound to the Col-0 controls (Figure 5c,d) and similar to the phenotype observed for soil-grown apc10-1 plants. Also, fresh and dry weights of transgenic APC10OE plants were approximately 16 and 25%, respectively, higher than those of Col-0 plants (Figure S3a,b).

To investigate the cellular basis of the observed phenotype, leaf development of APC10OE and wild-type plants grown in vitro was analyzed at the cellular level and as a function of time (De Veylder et al., 2001). In contrast to plants grown in soil, increased leaf size could already be observed in the first leaf pair of APC10OE plants. To understand the cellular basis of this increase, the leaf blade area, cell number and cell size of the abaxial epidermis were determined by daily quantitative image analysis of the first leaf pairs from 5 to 23 DAS. During the initial stages of leaf development, the leaf size between APC10OE and wild-type lines did not differ, but from day 10 onward, the difference became significant and, ultimately, the mature leaves of APC10OE were approximately 35% larger than those of the control, the difference first becoming visible at 11 DAS (Figure 6a). Cell area increased from approximately 100 to 1500 mm2 and followed an identical trend in both the wild-type and APC10OE plants (Figure 6b), suggesting that the coordination between cell division and cell expansion was not affected by the overexpression of APC10. By means of measurements of leaf area and average cell area, it is possible to estimate accurately the number of cells per leaf throughout development (De Veylder et al., 2001). On days 5 and 6, the number of cells per leaf did not vary, but from day 7 onward, APC10OE leaves contained more cells and, at maturity, the difference was approximately 40% (Figure 6c). Based on the evolution of the leaf cell number, the cell division rates were consistently higher in the APC10OE leaves than those in the wild type between days 5 and 11, albeit not always significantly (Figure 6d). The average cell cycle duration, calculated as the inverse of the cell division rates (De Veylder et al., 2001) during the initial phase of the nearly constant cell expansion between days 5 and 8, decreased from 21 ± 2 h in the control to 19 ± 2 h in APC10OE. During early leaf development, all cells in the leaf primordium divide, but later in development cell division ceases and further growth occurs by cell expansion (Donnelly et al., 1999). The timing of the transition between cell proliferation and cell expansion was seemingly not affected: in both lines, the cell division in the abaxial epidermis of leaf 1 and 2 ceased at day 14. The inference that the timing of development remained the same was supported by the identical timing of the appearance of stomata (Figure 6e). Taken together, the kinematic analysis indicates that the increased final leaf size in APC10OE is due to enhanced rates of cell division during the early stages of leaf development.

Figure 6.

 Kinematics analysis of the first leaf pair of APC10OE (open circle) and wild-type (black circle) plants grown in vitro.
(a) Leaf blade area.
(b) Number of cells on the abaxial side of leaves.
(c) Cell area.
(d) Cell division rate.
(e) Stomatal index.
The first leaf pairs were harvested at the indicated time points (days after stratification, DAS). The analysis was done on the APC10OE5.3 line from 5 to 23 DAS.

Overexpression of APC10 inhibits endoreduplication

To study the effect of APC10 overexpression on cell cycle activity, the evolution of the DNA content was evaluated during development by flow cytometry. To this end, the ploidy distribution of leaves 1 and 2 was determined every second day throughout development. The distribution of the 2C, 4C, 8C and 16C peaks allowed the calculation of the endoreduplication index that gives the mean number of endoreduplication cycles per nucleus of an average leaf cell (Boudolf et al., 2009).

The ploidy analysis of APC10OE plants showed that overexpression of APC10 resulted in a shift to low ploidy levels: the fraction of cells with 2C and 4C DNA contents was higher in APC10OE plants (for example, at 14 DAS; Figure 7a) and, conversely, the proportion of cells with 8C and 16C DNA contents was lower than that of Col-0 plants. As a consequence, the endoreduplication index in APC10OE was reduced when compared with wild-type plants starting at 10 DAS (Figure 7b). Taken together, the detailed growth and ploidy analyses indicate that APC10OE enhances cell division, while inhibiting endoreduplication.

Figure 7.

 Ploidy distribution of the first leaf pair of APC10OE and wild-type plants grown in vitro.
(a) Ploidy level measured by flow cytometry of the first leaves from 14-day-old APC10OE and Col-0 plants.
(b) Endoreduplication index (2C × 1 + 4C × 2 + 8C × 4 + 16C × 8) calculated from the complete data during leaf development. Data are means ± standard deviation.
(c) Quantitative RT-PCR transcript analysis of cell cycle gene expression markers in Col-0 (black) and APC10OE (grey) plants. Total RNA prepared from the third leaf harvested at 12 days after stratification (DAS) and amplified by RT-PCR. All values were normalized against the expression level of the housekeeping genes and expression compared to the expression data in the Col-0 control. Data are means ± standard deviation (= 3). *Significantly different P < 0.05.

To investigate which phase of the cell cycle is affected by APC10OE leading to an increase in cell proliferation, we analyzed the expression levels of selected S and M phase marker genes. Plants were grown in vitro for 12 DAS, the stage at which leaf 3 is still proliferating (Skirycz et al., 2010). Leaf 3 was harvested and analyzed by qPCR. Consistent with the stimulation of G2/M, the data showed that the transcript level of the two G2/M phase-specific genes, CDKB2;1 (Segers et al., 1996), and BUB3.1 (Caillaud et al., 2009), had increased in the APC10OE plants. Also two G2/M phase-specific proliferation genes (Prol1_At5g16250 and Prol2_At3g42660) (Beemster et al., 2005) were up-regulated, while one G1/S phase-specific gene, CYCD3;1 was down-regulated (Figure 7c). Additionally, CYCB1;1 and Histone H4 did not differ at the transcriptional level.

APC10 overexpression enhances the protein degradation of the CYCB1-D box

Furthermore, we analyzed whether overexpression of APC10 affected the levels of cell cycle regulatory proteins by altering the APC/C-mediated protein degradation. To this end, we crossed transgenic plants carrying the promoter (p) CYCB1;1:D-box–GUS construct with APC10OE plants to monitor the mitotic activity in the meristems. As control, wild-type plants (Col-0) were crossed with pCYCB1;1:D-box–GUS plants.

Expression of the pCYCB1;1:D-box–GUS construct allowed us to estimate the number of G2/M cells and the rate of CYCB1;1 degradation in the apical meristem and leaf primordia (Genschik et al., 1998; Colón-Carmona et al., 1999).

Leaves of 20 F1 plants from each cross (APC10OE × pCYCB1;1:D-box–GUS and Col-0 × pCYCB1;1:D-box–GUS) were harvested, stained for GUS activity and cleared with lactic acid. Photographs were taken and the GUS staining intensity quantified with ImageJ software (see Experimental Procedures). First, in plants at 7 DAS, in which leaves 1 and 2 were entirely proliferating (Figure 6), the GUS staining intensity was clearly reduced (Figure 8a,c). In leaf 3 at 12 DAS, in which the transition between cell proliferation and cell expansion is clear, a reduced GUS staining intensity was visible (Figure 8b,d). As this reduced GUS staining was not caused by reduced CYCB1;1 expression (Figure 7c), the data strongly suggest that APC10 overexpression enhances the D-box-dependent proteolysis of CYCB1;1.

Figure 8.

 Anaphase-promoting complex/cyclosome (APC/C) activity in APC10OE plants.
(a) Picture of the pCYCB1;1:D-box–GUS staining on leaves 1 and 2 at 7 days after stratification (DAS) of the APC10OE and wild-type lines.
(b) Picture of the pCYCB1;1:D-box–GUS staining on leaf 3 at 12 DAS of the APC10OE and wild-type lines.
(c) Measurement of the pCYCB1;1:D-box-GUS intensity in leaves 1 and 2 of the APC10OE and wild-type lines. *Significantly different P < 0.05.
(d) Measurement of the pCYCB1;1:D-box–GUS intensity in leaf 3 of the APC10OE and wild-type lines. Numbers indicate the intensity of the D-box CYCB1;1 GUS staining in leaves 1 and 2 (= 20) and leaf 3 (arbitrary units were used). **Significantly different P < 0.001.

Discussion

Most APC/C subunits are conserved in all eukaryotes and their function is tightly associated with cell cycle events (Jin et al., 2008; Herzog et al., 2009; Marrocco et al., 2010). Previously, the requirement of the APC/C subunits CDC27b/HOBBIT (APC3b), APC2, and CDC16 (APC6) subunits for cell division, meristem formation, and gametogenesis have been reported (Blilou et al., 2002; Capron et al., 2003b; Kwee and Sundaresan, 2003). Defects in APC2 and CDC16 (APC6) affected the female gametophyte (Capron et al., 2003b; Kwee and Sundaresan, 2003), whereas a knockout mutation of APC8 primarily had an impact on male gametogenesis (Zheng et al., 2011). In concert, we also found that knockout mutations of APC10 specifically interfered with female gametogenesis.

Based on the analysis of the GUS reporter lines, APC10 is expressed in almost all tissues with the notable exception of root meristems, in agreement with previous qPCR analyses that demonstrated the occurrence of APC10 in all organs studied, including tissues with low cell proliferation rates (Eloy et al., 2006). Also the expression of the gene encoding the APC6/NOMEGA subunit is seemingly absent in root meristems (Kwee and Sundaresan, 2003). In addition, expression profiles of APC10 with the Bio-Array Resource (http://www.bar.utoronto.ca) and Genevestigator (https://www.genevestigator.com) revealed constitutive expression in all tissues analyzed, albeit very weak in roots. The weak expression in the root meristems is also consistent with the lack of a positive effect on root growth in APC10OE plants (data not shown). The very low APC10 expression in the root meristem compared with the high expression in the shoot apical meristem suggests the existence of divergent mechanisms for the control of the APC activity in root meristems, namely less dependent on the APC10 subunit. Nevertheless, ectopic production of an APC10–GFP fusion protein in roots results in a nuclear localization very similar to that observed in leaf pavement cells and trichomes. Consistent with the female gametogenesis mutant phenotype, the expression of the APC10 gene in ovules is very strong, whereas an APC10 knockout mutation has no effect on male gametogenesis. APC10 is also not expressed in anthers and pollen. Recently, the APC10 subunit has been found to be required for proper vascular development (Marrocco et al., 2009). Accordingly, GUS staining in leaves is very intense, especially in vascular tissues.

We showed that increased APC10 expression positively affected leaf growth and final leaf organ size. Ectopic overexpression of APC10 increased cell number, without any impact on the cell area, whereas overproduction of an APC10–GFP fusion protein also enhanced leaf growth, similarly to plants in which APC10 alone was overexpressed (data not shown). A T-DNA insertion in the promoter of the APC10 gene (apc10-1) also intensifies the APC10 expression and leaf organ size. A similar situation in which a T-DNA insertion in the promoter enhances gene expression has been found for STRUWWELPETER (SWP) (Autran et al., 2002). However, in contrast to the leaf growth-enhancing phenotype observed by constitutive overexpression of APC10 under the control of the 35S promoter, the apc10-1 mutant also shows additional phenotypes, such as altered flowers and cotyledons (data not shown). Although these phenotypes have been noted (Lindsay et al., 2011), they are most probably not due to the overexpression of APC10, because no abnormal cotyledons, flowers, or other organs were observed in analyzed lines overexpressing either APC10 or APC10-GFP under the control of the 35S promoter. Possibly, the T-DNA insertion in apc10-1 also affects other genes or other closely linked mutations affect cotyledon and flower morphology. Except for being larger, the APC10OE plants are phenotypically indistinguishable from the wild type, suggesting that enhanced APC10 expression has a direct and specific impact on cell division.

Interestingly, APC10 overexpression reduces endoreduplication levels. Modifications of a number of cell cycle genes – including the APC/C subunits – affect the cellular DNA content. In Medicago, partial suppression of the CCS52A gene reduces the number of endoreduplication cycles and the cellular enlargement in nodules (Cebolla et al., 1999). In Arabidopsis, both CCS52A1 and CCS52A2 knockout plants display a decreased endoreduplication index in rosette leaves, revealing that both are involved in the control of endoreduplication (Lammens et al., 2008). During tomato (Solanum esculentum) fruit development, down-regulation of the CCS52A gene reduces endoreduplication levels and the cell size as well (Mathieu-Rivet et al., 2010). Similarly, decreased endocycle levels were observed in Arabidopsis leaves from APC10 RNA interference (RNAi) lines (Marrocco et al., 2009). The observation that lessening the APC/C activity by silencing individual components leads to reduced endocycle levels is intuitively contradictory to our results in which the endocycle levels decreased when overexpressing APC10. However, the ectopic overexpression of CCS52A has been shown to induce an early delay in endoreduplication between 5 and 20 days post-anthesis (dpa), which is later resumed and even enhanced from 20 dpa until the end of development (Mathieu-Rivet et al., 2010), indicating a developmental stage-dependent outcome of the mis-regulated APC/C function. In light of these data, the reduced endocycle levels observed in young leaves overexpressing APC10 could indicate that a similar mechanism might be at play during leaf development in Arabidopsis. Although in the APC10OE plants the endoreduplication is reduced, it does not correlate with differences in cell size as reported by Melaragno et al. (1993) and Sugimoto-Shirasu and Roberts (2003).

The mechanisms that control the cell division rate are still largely unknown in plants. A detailed kinematic analysis revealed that the increased growth rate observed in APC10OE plants is a consequence of increased cell division rates: the average duration of the cell cycle decreased from approximately 21 h in the wild type to 19 h in APC10OE, most probably by a faster degradation of mitotic regulators, such a CYCB1;1 as shown here. It is well known that during every cell cycle, mitotic cyclins need to be degraded to ensure proper progression at the metaphase-to-anaphase transition (Criqui et al., 2000). In agreement, overexpression of a non-degradable CYCB1 delays exit from mitosis and cause malformation of plants (Weingartner et al., 2004). Our data suggest that APC10 is rate-limiting for the APC/C activity, at least in developing shoots. We speculate that an increased APC/C activity causes a more rapid progression through mitosis. Overexpression of the CDC27a gene in tobacco (Nicotiana tabacum) also increases leaf size, but the mechanism is still unknown (Rojas et al., 2009). In contrast, overexpression of genes encoding the APC subunits APC3b and APC11 provokes no obvious growth-related phenotype (data not shown). In addition to its effect on CYCB1;1 degradation, overexpression of APC10 also results, by an as yet unknown mechanism, in up-regulation of the steady-state levels of CDKB2;1 and three other genes specific to the proliferation stage (Segers et al., 1996; Beemster et al., 2005; Caillaud et al., 2009). However, expression levels of CYCB1;1 and Histone 4 remain unaltered, whereas the CYCD3;1 expression is down-regulated. The effect of the APC10 overexpression on transcription of cell cycle genes might be indirect. Recently, the APC8 subunit of APC/C has been reported to be involved in the regulation of miR159 that targets DUO POLLEN1, a transcriptional regulator of CYCB1;1 during male gametophyte development (Zheng et al., 2011).

Besides APC10, other genes involved in proteolysis have been shown to enhance plant growth. A mutation in the 19S proteasome subunit RTP2a increases leaf size and endoreduplication (Kurepa et al., 2009; Sonoda et al., 2009). Furthermore, a dominant negative mutation in the putative ubiquitin receptor DA1 increases leaf, flower and seed organ size (Li et al., 2008) and a mutation in E3 ligase encoded by BIG BROTHER extends the timing of cell proliferation during organ development, resulting in leaves and petals with more cells (Disch et al., 2006).

Many plant genes have been identified that enhance leaf growth (for a review, see Gonzalez et al., 2009). A predominant mechanism that drives the formation of enlarged leaves is the extension of the timing of cell proliferation during leaf development (Mizukami and Fischer, 2000; Horiguchi et al., 2009; Gonzalez et al., 2010). Additionally, increased proliferation of meristemoid cells can cause the formation of enlarged leaves (White, 2006). Hence, the overexpression of APC10 increases leaf size by prolonging the duration of the cell cycle possibly through a faster degradation of cell cycle regulatory proteins, such as B-type cyclins.

In conclusion, we have shown that the cell division rate controlled through the regulation of the APC/C activity by means of the APC10 subunit seems to play an important role in leaf growth and development.

Experimental procedures

Cloning

The genome of Arabidopsis thaliana (L.) Heyhn. (http://www.arabidopsis.org/) was searched for homologs of the APC10 gene with a BLAST program. The coding region of APC10 (At2G18290) was used to design specific primers to isolate the respective cDNA and it was amplified directly by PCR from tissues of Arabidopsis ecotype Columbia-0 (Col-0).

The PCR fragment, referring to the complete cDNA of the APC10 gene was introduced by attBattP recombination sites into pDONr 201 with the Gateway system (Invitrogen, http://www.invitrogen.com). The APC10-coding region was transferred to a binary vector containing the CaMV 35S promoter and the kanamycin selection marker in plants (pK7WG2) (35S:APC10) and to the 35S:GFP fusion vector (pK7WGF2) for the subcellular localization of APC10 (GFP:APC10) (Karimi et al., 2007). For analysis of the APC10 promoter, a 2928-bp genomic fragment (upstream of the ATG start codon) containing the putative APC10 promoter was amplified from the genomic DNA of Arabidopsis plants (ecotype Col-0), cloned into the pDONR201 vector (Invitrogen), and then transferred to the GUS::GFP-containing binary vector pFASTG04, generating the plasmid ProAPC10:GUS.

The genomic fragment comprised 1148 bp of putative promoter and coding sequences corresponding to the N-terminal 116 amino acids of CYCB1;1, including that a mitotic Destruction Box was amplified from the genomic DNA of Arabidopsis plants (ecotype Col-0), cloned into the pDONR201 vector (Invitrogen), and then transferred to the GUS–GFP vector (pKGWFS7) for the CYCB1;1promoter-D-box analysis.

Yeast complementation

The APC10-coding region was transferred to the yeast vectors pREPHAWG and pREPHAGW (M. Karimi, VIB-Ghent University, Gent, Belgium, unpublished data), containing the LEU2 selection marker, and transformed into S. pombe strain apc10-27 (Kominami et al., 1998). Leu+ colonies were grown in selective minimal media (SD) without leucine at 25 and 35°C.

Plant material and transgenic plants production

apc10-1 mutant plants (seed code, SALK_083989, SALK_043367 and SALK_123967) were obtained from the Salk collection (http://signal.salk.edu/). The presence of the T-DNA insertion and absence of the wild-type gene was confirmed by genomic PCR from leaves of 15-day-old plants. These plants were selected to produce more seeds and for subsequent analysis.

The Arabidopsis plants were grown on agar plates or soil under long-day conditions (16 h light, 8 h darkness) at 23°C under standard greenhouse conditions. All analyses in planta (proAPC10:GUS, APC10:GFP, and APC10OE lines) were carried out in the Arabidopsis accession Col-0 background. For the production of Arabidopsis transgenic plants, Agrobacterium tumefaciens strain C58C1 harboring the plasmid pMP90 was used to transform the plants by the floral dip method (Clough and Bent, 1998).

Growth analysis

Plants were grown in vitro in half-strength Murashige and Skoog medium (Murashige and Skoog, 1962) supplemented with 1% sucrose at 21°C, under a 16- h day/8-h night regime and at a density of one plant per 4 cm2. Plants were also grown in rock wool, a semi-hydroponic inert porous substrate that transports water and fertilizers to the roots by capillary action, and in soil under the same day-length regime.

For the measurements of rosette leaf area, 8–12 seedlings were grown on rock wool or under in vitro conditions for 22 days. Individual leaves (cotyledons and rosette leaves) were dissected, and their area was measured with imagej software (http://rsb.info.nih.gov/ij/).

For the biomass measurement, the vegetative part of a 22-day-old plant was harvested. For fresh weight measurement, approximately 20 plants of each line were weighed; for dry weight, the same plants were placed on Petri plates, allowed to dry for 1 week, and weighed again.

Kinematic analysis

The complete kinematics was analyzed as described (De Veylder et al., 2001) on leaf 1 and 2 of eight to ten APC10OE and wild-type plants grown in vitro harvested daily from 5 to 23 DAS. The leaves were cleared with 100% ethanol, mounted in lactic acid on microscope slides, and photographed. The leaf area was determined with imagej software. Abaxial epidermal cells (40–100 cells) for four to five leaves of blades of leaf 1 and 2 were drawn with a microscope (Leica; http://www.leica-microsystems.com/) fitted with a drawing tube and a differential interference contrast objective. Photographs of leaves and drawings were used to measure the leaf area and to calculate the average cell area, respectively, with imagej software. Leaf and cell areas were subsequently used to calculate cell numbers.

RNA extraction and cDNA preparation

Total RNA was extracted from the frozen material with TRIzol reagent (Invitrogen). To eliminate the residual genomic DNA present in the preparation, the RNA was treated by RNAse-free DNase I according to the manufacturer’s instructions (GE Healthcare, http://www.gehealthcare.com) and purified with the RNeasy Mini kit (Qiagen, http://www.qiagen.com). Complementary DNA was made with the SuperScript III first-strand synthesis system (Invitrogen) with the oligo (dT) primer solution according to the manufacturer’s instructions.

GUS staining

Different tissues and organs of Arabidopsis were harvested and incubated in heptane for 10 min and, subsequently, incubated in 5-bromo-4-chloro-3-indolyl-β-glucuronide (X-Gluc) buffer [100 mm 2-amino-2-(hydroxymethyl)-1,3-propanediol (TRIS)-HCl, 50 mm NaCl buffer (pH 7.0), 2 mm K3[Fe(CN)6], and 4 mm X-Gluc] at 37°C for 5 h or 24 h. Seedlings were washed in 100 mm TRIS–HCl, 50 mm NaCl (pH 7.0) and cleared overnight in ethanol 95%, then kept in 90% lactic acid. Samples were photographed under a differential interference contrast microscope (Leica).

Real-time quantitative reverse-transcription PCR

For cDNA synthesis, 500 ng–1 μg of RNA was used with the SuperScript Reverse III reagent (Invitrogen) according to the manufacturer’s instructions. Primers were designed with QuantPrime (Arvidsson et al., 2008). The primer sequences used in the qPCR experiments are listed in Table S1. The cDNA was amplified on a LightCycler 480 (Roche Diagnostics, http://www.roche.com) in 384-well plates with LightCycler 480 SYBR Green I Master (Roche) according to the manufacturer’s recommendations. Melting curves were analyzed to check the specificity of the primer. Normalization was done against the average of the housekeeping genes UBQ10, GAPDH, and CBP20: DCt = Ct (gene) − Ct (mean [housekeeping genes]) and DDCt = DCt (control) − DCt. DCt values for the three biological replicates were used for statistical analysis; Ct refers to the number of cycles at which SYBR Green fluorescence reaches an arbitrary value during the exponential phase of the cDNA amplification. The data were first normalized to the expression level of the housekeeping genes for each RNA sample and then scaled to the maximum expression per gene that was fixed to 1. We used BAR-Utoronto (http://bar.utoronto.ca/) and Genevestigator (https://www.genevestigator.com) for analysis of the APC10 expression in silico.

Protein gel blotting

Whole seedlings (15 DAS) were collected and quickly frozen in liquid nitrogen, ground and mixed with an equal volume of extraction buffer (50 mm TRIS–HCl, 150 mm NaCl, 15 mm EGTA, 15 mm MgCl2, 1 mm DTT, and 0.1% (v/v) Tween20, pH 7.5) containing a protease-inhibitor cocktail according to the manufacturer’s instructions (Sigma-Aldrich, http://www.sigmaaldrich.com/), then centrifuged for 20 min at 18 000 g at 4°C. Total proteins were quantified with the Bio-Rad Protein Assay (http://www.bio-rad.com/). Then, approximately 30 μg of extract was separated on a 10% SDS-PAGE gel and transferred onto Immobilon-P membranes according to the manufacturer’s instructions (Millipore, http://www.millipore.com/). Membranes were blocked and incubated with an anti-APC10 (1/1000) antibody. Secondary anti-rabbit conjugated to horseradish peroxidase (GE Healthcare) (1/5000) was used and blotted with enhanced chemiluminescence detection according to the manufacturer’s instructions (GE Healthcare). The APC10 antibody was produced by cloning the APC10 open reading frame into the pDEST17 (Invitrogen) and expressed in Escherichia coli (BL21) with 0.5 mm isopropyl β-d-1-thiogalactopyranoside. The protein was purified with a His-Trap HP column (GE Healthcare) and concentrated to 10 mg ml−1 with Centricon (Millipore). The protein was lyophilized and sent to Covance (http://www.covance.com/), where the proper antibodies were produced in rabbits.

Flow cytometry

The first leaf pairs of 8- to 24-day-old seedlings were chopped with a razor blade in 500 μl buffer (45 mm MgCl2, 30 mm sodium citrate). The leaf tissues in 200–400 μl buffer (45 mm MgCl2, 30 mm sodium citrate, 20 mm MES, pH 7, and 1% Triton X-100), filtered over a 30-μm mesh, and supplemented with 1 μl of 1 μg mg−1 of 4,6-diamidino-2-phenylindole. The nuclear DNA content distribution was analyzed with a Cyflow ML flow cytometer (Partec, http://www.partec.com/).

Detection of D-box CYCB1;1–GUS activity

Seeds were plated on MS agar growth medium. After 3 DAS at 4°C, the plates were placed in a growth chamber (22°C; 16 h photoperiod) for 8 or 12 days to measure leaves 1 and 2 or leaf 3, respectively. Histochemical GUS activity was detected with X-Gluc as described above. For observation, the seedlings were mounted in lactic acid on glass microscope slides and pictures were taken. To quantify the intensity of the GUS staining, the stained area of each leaf was marked and its intensity was measured and quantified with the imagej program with the values given in arbitrary units. Approximately 20 plants were analyzed for each experiment. Numbers indicate the average intensity (> 20 ± SE) of D-box CYCB1;1–GUS staining.

Confocal imaging

Arabidopsis seedlings were imaged between slide and cover slip on a LSM710 inverted confocal microscope (Zeiss, http://www.zeiss.com/) equipped with the ZEN software package and a C-Apochromat 40 × /1.20 W Korr M27 water-corrected lens. Excitation was done with a multi-argon laser (458, 488, and 514 nm) with 5% laser line-attenuated transmission and a MBS488 dichroic mirror. Fluorescence was detected through a spectral emission window ranging from 493 to 598 nm. Digital gain was set at 1.00; signal averaging at 8; and optical slice thickness at 1 airy unit.

Acknowledgements

We thank Dr Takashi Toda for providing the yeast mutant strain and Dr Martine De Cock for help in preparing the manuscript. This work was supported by Ghent University (‘Bijzonder Onderzoeksfonds Methusalem project’ grant no. BOF08/01M00408 and Multidisciplinary Research Partnership ‘Biotechnology for a Sustainable Economy’ project no. 01MRB510W), the Belgian Science Policy Office (BELSPO) [Interuniversity Attraction Poles Programme (IUAP VI/33) and post-doctoral fellowship to NBE], the Research Foundation-Flanders (post-doctoral fellowship to DVD), the Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq), for graduate fellowships (MdeFL) and research grants (to ASH and PCGF).

Ancillary