SEARCH

SEARCH BY CITATION

Keywords:

  • cell cycle;
  • DOF transcription factor;
  • Arabidopsis;
  • OBP1;
  • D-type cyclin

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

In contrast to animal growth, plant growth is largely post-embryonic. Therefore plants have developed new mechanisms to precisely regulate cell proliferation by means of internal and external stimuli whilst the general core cell cycle machinery is conserved between eukaryotes. In this work we demonstrate a role for the Arabidopsis thaliana DNA-binding-with-one-finger (DOF) transcription factor OBP1 in the control of cell division upon developmental signalling. Inducible overexpression of OBP1 resulted in a significant overrepresentation of cell cycle genes among the upregulated transcripts. Direct targets of OBP1, as verified by chromatin immunoprecipitation, include at least the core cell cycle gene CYCD3;3 and the replication-specific transcription factor gene AtDOF2;3. Consistent with our molecular data, short-term activation of OBP1 in cell cultures affected cell cycle re-entry, shortening the duration of the G1 phase and the overall length of the cell cycle, whilst constitutive overexpression of OBP1 in plants influenced cell size and cell number, leading to a dwarfish phenotype. Expression during embryogenesis, germination and lateral root initiation suggests an important role for OBP1 in cell cycle re-entry, operating as a transcriptional regulator of key cell cycle genes. Our findings provide significant input into our understanding of how cell cycle activity is incorporated into plant growth and development.


Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

Cell proliferation plays a prominent role in plant growth and development, sharing its basic mechanism with other eukaryotes. The cell cycle consists of the replication (S) and mitosis (M) separated by two gap phases, G1 and G2. Additionally, some cells can skip mitosis, in a process called endoreduplication, leading to increased cell ploidy. Progression of the cell cycle is tightly regulated and ensured by deploying transcriptional, post-transcriptional and post-translational control mechanisms that are fairly well characterised (Inze and De Veylder, 2006). However, much less is known about how the cell cycle is integrated with plant growth and development, cell cycle onset being one of possible control points. At the level of the core cell cycle machinery the commitment of cells to the mitotic cycle is related to the activity of the D-type cyclin (CYCD)/retinoblastoma (RB) pathway. Binding of CYCD activates PSTAIRE-containing cyclin-dependent kinase A (CDKA) leading to phosphorylation of RB protein and release of E2F transcription factors (TFs) (Healy et al., 2001; Nakagami et al., 1999). Conversely cell cycle onset can be arrested by the interaction of CDKA and D-type cyclins with Kip-related (KRP) or SIAMESE (SIM)/EL2 proteins (Churchman et al., 2006; De Veylder et al., 2001; Peres et al., 2007; Riou-Khamlichi et al., 1999; Wang et al., 1998). D-type cyclins, KRP and SIM/EL2 proteins are all encoded by multigene families and are believed to link cell cycle activity with developmental and environmental stimuli. Expression of several CYCD genes was shown to be regulated by nutritional and hormonal signals, whilst certain KRP and SIM/EL2 genes are induced by abiotic stress (Inze and De Veylder, 2006; Riou-Khamlichi et al., 1999). For example the expression of Arabidopsis CYCD3;1 is induced by both sucrose and cytokinins and, in agreement with this, CYCD3;1 overexpression in cell cultures overcomes G1 phase arrest in sucrose-starved cells (Menges et al., 2006; Oakenfull et al., 2002; Riou-Khamlichi et al., 2000). In a separate study, CYCD3;1, CYCD3;2 and CYCD3;3 were shown to be essential for cell cycle activation in the root apical meristem during germination, and their expression was induced shortly after imbibition (Masubelele et al., 2005). However, knowledge about the signaling pathway up-stream of D-type cyclins or KRP proteins is limited. For most of the TFs that affect cell proliferation either spatially, temporally or quantitatively during development or environmental responses, it is not clear if they regulate core cell cycle genes directly (Gegas and Doonan, 2006). Exceptions include a membrane-bound NAC TF, NTM1, acting upstream of CYCD3;1, KRP2 and KRP7 (Kim et al., 2006); an AP2 TF, ENHANCED SHOOT REGENERATION 2 (ESR2), whose direct targets include CYCD1;1 and KRP4 (Ikeda et al., 2006); and the AP2 TF AINTEGUMENTA (ANT) that maintains the meristematic competence of cells by sustaining expression of CYCD3;1 (Mizukami and Fischer, 2000).

In Arabidopsis, the DNA-binding-with-one-finger (DOF) TF family contains 37 members playing divergent roles in plant-specific processes. Their highly conserved N-terminal DOF region acts as a DNA-binding domain and corresponds to a conserved DNA cis-element (A/T)AAAG or its complementary inverse sequence (Yanagisawa, 2004). The DOF TF OBF BINDING PROTEIN 1 (OBP1) was, as its name suggests, previously identified in a screen for protein partners of the bZIP transcription factor OBF4 (OCS BINDING FACTOR 4) (Zhang et al., 1995). In vitro, OBP1 stimulates OBF4 binding to auxin and salicylic acid-responsive ocs elements found in the cauliflower mosaic virus (CaMV) 35S and plant glutathione S-transferase (GST6) promoters (Chen et al., 1996). Subsequently two other unrelated DOF transcription factors, named OBP2 and OBP3, were shown to have similar qualities. They all bind OBF4 and enhanced its binding to the ocs element (Kang and Singh, 2000). These initial in vitro-based findings have been followed by more detailed characterisation of OBP2 and OBP3. OBP3 was characterised as a novel component of light signalling (Ward et al., 2005), whilst OBP2 is involved in the regulation of indole glucosinolate metabolism (Skirycz et al., 2006). Significantly, none of these studies provided in vivo evidence of GST6 regulation.

During this study we highlight an unexpected role for DOF TFs in cell cycle regulation. OBP1 was found to control cell cycle progression by targeting the expression of core cell cycle genes and components of the replication machinery in a developmentally specific manner. Correspondingly, OBP1 overexpression results in a shortening of the cell cycle. Based upon its expression profile and effects on cell cycle progression, we propose that OBP1 is part of a signal transduction cascade working upstream of the cell cycle.

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

Identification of OBP1-regulated biological processes by profiling technologies

Co-expression analysis is a proven tool for elucidating possible biological function (Gachon et al., 2005; Persson et al., 2005). To identify biological processes that may be regulated by OBP1 we utilised the Arabidopsis co-expression tool (ACT) (Manfield et al., 2006). Out of the top 48 genes having expression that positively correlates with OBP1, the majority can be assigned to DNA/RNA metabolism, cell division and organization (Table S1).

To further characterise OBP1, we analysed transcript profiles using Affymetrix near-full genome ATH1 arrays. For this approach we used a dexamethasone (DEX)-inducible system in which OBP1 fused to a glucocorticoid receptor (GR) domain was placed under the control of the CaMV 35S promoter (Lloyd et al., 1994). The DEX treatment induces nuclear targeting of the OBP1–GR fusion protein which thus causes activation of target genes (Figure S1).

To identify candidate processes regulated by OBP1, a single inducible overexpression line of OBP1 (the IOE-OBP1 line) was compared with an empty-vector control 10 h after DEX induction. In total 631 transcripts were upregulated and 842 downregulated (two-fold cut-off) (Tables S2 and S3). Over-representation analysis using PageMan (Usadel et al., 2006) revealed significant enrichment of cell cycle and cell organization genes among upregulated transcripts and cell wall modification and signalling receptor kinases among downregulated transcripts (Tables S2 and S3). We also compared our results with publicly available datasets. In total, 190 of the upregulated transcripts (about six-fold more than expected by chance) were previously shown to oscillate during the cell cycle (Menges et al., 2003) with the majority peaking during the G2 and M phases (Figure S2, Table S4). Importantly, the same study showed that OBP1 expression was highest during the G1 phase. Of 131 genes assigned as high-confidence proliferation genes (Beemster et al., 2005), 85 were upregulated in our study (Table S5). Furthermore, out of the 87 core cell cycle genes (Menges et al., 2005; Vandepoele et al., 2002; Wang et al., 2004) represented on the Affymetrix ATH1 array, 24 were induced (Figure S1, Table S6).

The induction of 14 out of 17 selected upregulated genes was confirmed using quantitative (Q)-RT-PCR in two independent IOE-OBP1 lines in three biologically independent experiments. Comparison was made between mock- and DEX-treated plants and the effect of DEX was evaluated by including an empty-vector control line that was sprayed with DEX (Table 1). Together, our data imply the involvement of OBP1 in cell proliferation.

Table 1.   Expression changes of selected genes in dexamethasone (DEX)-inducible OBP1 lines
Gene name Q-RT-PCR (10 h) ΔΔCt (log2)Affy (10 h) (log2)
ControlIOE#4IOE#7IOE#7
  1. Transcript levels were measured using quantitative RT-PCR in IOE-OBP1 (lines IOE#4, IOE#7) and empty-vector control plants (Control) 10 h after DEX induction. Each sample was pooled from three plants and the experiment was repeated three times under long-day conditions (data are means ± SD, = 3). Asterisks indicate values that are significantly different from the respective controls (**< 0.05; *< 0.1; Student’s t-test). ΔΔCt = ΔCt (mock) − ΔCt (DEX). For reference, the last column shows expression changes as determined by Affymetrix expression profiling (Ct refers to the number of cycles at which SYBR Green fluorescence reaches an arbitrary value during the exponential phase of DNA amplification).

MCM2−0.09 ± 0.241.41 ± 0.36**2.16 ± 0.95**1.88
MCM5−0.25 ± 0.420.62 ± 0.13*0.15 ± 0.411.39
SNC1−0.58 ± 0.321.57 ± 0.36**−0.39 ± 2.532.44
Unknown At5g17160−0.91 ± 0.542.71 ± 0.42**2.90 ± 0.11**3.82
Kinesin At5g33300−0.37 ± 0.241.21 ± 0.77*2.76 ± 0.11**2.82
Kinesin At3g44050−1.47 ± 1.161.46 ± 0.60**2.24 ± 0.24**4.01
CKS2−0.56 ± 0.401.71 ± 0.69**2.88 ± 0.75**2.78
CDKB2;1−0.72 ± 0.382.02 ± 0.35**4.48 ± 1.65**3.21
CYCA2;2−0.70 ± 0.611.84 ± 1.352.94 ± 0.56**2.62
CYCB1;3−0.49 ± 0.291.68 ± 0.37**2.93 ± 0.23**2.83
CYCB2;4−0.95 ± 0.901.09 ± 0.49*3.08 ± 0.30**3.11
CYCD3;2−0.46 ± 0.450.82 ± 0.05**2.00 ± 1.06*1.48
CYCD3;3−0.70 ± 0.292.60 ± 0.30**2.22 ± 0.76**2.56
CYCD4;1−0.38 ± 0.400.67 ± 0.22**1.05 ± 0.50*1.51
DEL3−1.12 ± 0.821.71 ± 0.22**2.39 ± 0.68**2.20
E2Fa−1.03 ± 0.691.00 ± 0.47**0.81 ± 0.10*1.29
AtDOF2;3−0.54 ± 0.582.55 ± 0.32**3.11 ± 0.61**1.56

Identification of OBP1 target genes

The large number of expression changes 10 h after DEX induction suggested these might be secondary targets of OBP1. Therefore, to identify direct targets we profiled selected genes 5 h after induction. Given the probable involvement of OBP1 in cell cycle regulation, we designed Q-RT-PCR primers for 82 core cell cycle genes, 90 cell cycle-specific TFs, GST6 and the two replication licensing factors MCM2 and MCM5. Among these we identified eight genes consistently upregulated 5 h and 10 h after induction: four TFs (MYB88, TAZ-At5g67480, MYB-like At2g40970 and AtDOF2;3), two core cell cycle proteins (CYCD3;3 and KRP7) and MCM2 and MCM5 (Table 2a). Subsequently we measured expression of these genes in 35S-OBP1 and RNAi-OBP1 lines (Table 2b). In general, fewer genes were regulated in these lines than we identified using IOE-OBP1. As the 35S-OBP1 lines showed a strong phenotype we decided to focus our molecular studies on CYCD3;3 and AtDOF2;3 as we detected lower expression in RNAi-OBP1 plants (Table 2b).

Table 2.   Expression changes of selected genes in plants with modified OBP1 expression. (a) Transcript levels were measured using quantitative (Q)-RT PCR in (a) IOE-OBP1 (lines IOE#4, IOE#7) and empty-vector control plants (Control) 5 h after dexamethasone (DEX) induction and (b) RNAi-OBP1 (lines #1, #2 and #3) and 35S-OBP1 (lines #1, #6 and #9) plants
GeneQ-RT-PCR; short day (5 h)Q-RT-PCR; long day (5 h)Affy (10 h)
ΔΔCt (log2)ΔΔCt (log2)(log2)
ControlIOE#4IOE#7ControlIOE#4IOE#7IOE#7
(a)
CYCD3;30.54 ± 0.801.58 ± 0.96**1.45 ± 0.71**−0.24 ± 1.141.87 ± 0.64**2.332.56
KRP70.07 ± 0.901.02 ± 0.66**1.07 ± 0.17**−0.34 ± 1.412.33 ± 0.77**1.841.67
MYB880.63 ± 0.991.53 ± 0.62**0.97 ± 0.47**−0.12 ± 0.472.75 ± 0.77**1.711.94
DOF2;30.05 ± 0.651.21 ± 0.84**1.01 ± 0.45**0.13 ± 0.542.03 ± 0.65**0.801.56
MYB-like0.54 ± 0.720.53 ± 0.530.55 ± 0.19**−0.24 ± 0.440.59 ± 0.32**0.810.7
TAZ 0.11 ± 0.350.79 ± 0.68**0.64 ± 0.32**−0.35 ± 0.502.16 ± 2.061.561.36
MCM50.22 ± 0.960.82 ± 0.28**0.89 ± 0.22**−0.22 ± 0.380.95 ± 0.18**1.951.39
MCM20.11 ± 1.461.27 ± 0.55**1.14 ± 0.57**−0.22 ± 0.451.77 ± 0.48**1.171.88
GeneRNAi-OBP135S-OBP1
ΔΔCt (log2)ΔΔCt (log2)
RNAi#1 (= 1)RNAi#2 (= 2)RNAi#3 (= 3)35#1 Ex135#6 Ex135#9 Ex135S#1 Ex235S#6 Ex2
  1. (a) Data are means ± SD, = 3, or means, = 2. Biological replication was from individual plants within a single experiment for the short-day grown plants and between three experiments for the long-day grown plants. For reference, the last column shows expression changes as determined by Affymetrix expression profiling. Asterisks indicate values that are significantly different from the respective controls (**< 0.05; *< 0.1; Student’s t-test). ΔΔCt = ΔCt (mock) − ΔCt (DEX), or ΔΔCt = ΔCt (control) − ΔCt (transgenic) (Ct refers to the number of cycles at which SYBR Green fluorescence reaches an arbitrary value during the exponential phase of DNA amplification).

  2. (b) Data are means ± SD, = 3, or means, = 1 or 2. For RNAi-OBP1 biological replication was between independent experiments. For experiment 2 (Ex2) biological replications was within the experiment from individual 35S-OBP1 plants. Unless otherwise stated, samples were pooled from three plants. Asterisks indicate values that are significantly different from the respective controls (**< 0.05; *< 0.1; Student’s t-test). ΔΔCt = ΔCt (mock) − ΔCt (DEX), or ΔΔCt = ΔCt (control) − ΔCt (transgenic) [Ct as for part (a)].

(b)
CYCD3;3−1.72−0.40−1.09 ± 0.70**0.43−0.231.26−0.32 ± 0.980.07 ± 1.42
KRP7−0.02−0.010.03 ± 0.100.731.421.160.84 ± 0.47**0.88 ± 0.26**
MYB88−0.23−0.01−0.27 ± 0.20**−0.241.872.971.15 ± 0.33**1.21 ± 0.33**
DOF2;3−2.62−0.89−1.57 ± 1.39*1.652.821.901.45 ± 0.54*1.97 ± 0.60**
MYB - like−0.86−0.240.41 ± 0.340.872.291.790.22 ± 0.110.06 ± 0.14
TAZ −1.31−0.200.48 ± 0.670.730.751.26−0.45 ± 0.50−1.07 ± 1.29
MCM5−0.19−0.170.14 ± 0.42−1.030.760.960.51 ± 0.460.41 ± 0.15
MCM2−0.540.280.15 ± 0.380.291.401.560.59 ± 0.600.86 ± 0.10*
OBP1−2.95−0.93−1.11 ± 0.86**3.978.288.226.99 ± 0.22**7.14 ± 0.23**

Using chromatin immunoprecipitation (ChIP) coupled to Q-PCR detection we demonstrated direct binding of OBP1 to CYCD3;3 and AtDOF2;3 promoter sequences (Figure 1a). Importantly, for both CYCD3;3 and AtDOF2;3, promoter scanning illustrated that the bound DNA regions contain DOF binding sites [CTTT(T/A) or its complement (Figure S3)]. Using an electrophoretic mobility shift assay (EMSA) we confirmed binding of OBP1 to the consensus DOF element (Chen et al., 1996) and demonstrated that, at least in vitro, OBP1 has a higher affinity for DNA fragments with tandem DOF sites (Figure 1b). The latter is also in agreement with previous work on other members of the DOF family (Yanagisawa and Schmidt, 1999).

image

Figure 1.  OBP1 binds to promoter fragments of CYCD3;3 and AtDOF2;3. (a) Dexamethasone (DEX)-induced (5 h) IOE-OBP1 (IOE#4 and IOE#7) and empty-vector control plants (dE) were used for chromatin immunoprecipitation (ChIP) experiments. The abundance of CYCD3;3 (left) and AtDOF2;3 (right) promoter fragments was determined using quantitative (Q)-PCR. Samples were from three (line IOE#7 fragments: AtDOF2;3: 826–674, and CYCD3;3: 380–328, 157–8) or two independent experiments additionally replicated with three individual plants inside each experiment. We discarded Ct values >40 (Ct refers to the number of cycles at which SYBR Green fluorescence reaches an arbitrary value during the exponential phase of DNA amplification). Data were normalised using genomic primers for UBQ10 and mean of the control samples in each experiment (ΔCt IOE – mean ΔCt dE; IOE = inducible overexpression). Data are means ± SD, = 3–8. Asterisks indicate values that are significantly different from the respective controls (< 0.05, Student’s t-test). (b) Electrophoretic mobility shift assay (EMSA) was performed using OBP1-His6 protein, AtDOF2;3 promoter fragment labelled with fluorescent probe IRD800 and unlabelled competitor sequences with or without mutated DNA-binding-with-one-finger (DOF) binding sites. Lanes: 1, labelled DNA fragment; 2, OBP1 protein and labelled DNA fragment. Note the distinct shift indicating binding; 3, 4, OBP1 protein, labelled DNA fragment and excess competitor. Note proportional shift disappearance; 5, 6, OBP1 protein, labelled DNA fragment and excess of competitor containing one mutated DOF element. Note lower proportional shift disappearance compared with lanes 3 and 4, indicating higher OBP1 affinity to tandem DOF sites; 7, 8, OBP1 protein, labelled DNA fragment and excess of competitor containing two mutated DOF elements. Note that result are similar to lane 2, indicating the requirement of DOF elements for competition.

Download figure to PowerPoint

OBP1 – cell proliferation and cell expansion

As molecular data indicated the likely involvement of OBP1 in cell cycle regulation we decided to use cell cultures as a model to gain more functional insight. We selected Arabidopsis MM2d cells as they can be easily synchronised (Menges and Murray, 2002), transformed them with the IOE-OBP1 construct and selected two independent lines with high OBP1 expression. Initially, we analysed cell number, wet weight and cell size during standard culture growth, for 7 days after subculture. All the comparisons were made between DEX-induced and mock-treated lines and the possibility of a DEX effect was evaluated using wild-type cultures. There were no significant effects of DEX on wild-type cells. The DEX activation of IOE-OBP1 lines caused a significant reduction in both wet weight and cell size within 24 h (Figure 2a,b). In addition, there was an increase in the number of cells, but only in the first 24 h following induction (Figure 2c). These data indicated that the first 24 h after DEX induction was most informative, with OBP1 activation leading to more but smaller cells.

image

Figure 2.  Cell size and number are affected by OBP1 overexpression. (a)–(c) Early stationary phase cells of MM2d wild- type and IOE-OBP1 transgenic lines (IOE#1 and IOE#3) were transferred into fresh medium and sampled each day. For induction experiments a freshly prepared 200-ml culture was split into two 100-ml cultures, with one dexamethasone (DEX) induced and the other mock induced. Wet weight (means ± SD, = 3 independent experiments) (a), cell size (means ± SD, = 23–25 cells) (b) and cell number (means ± SD, = 3 independent experiments) (c) were determined. (d) The ratio of cell number calculated between day 2 and day 3 of exponential growth. MM2d wild-type and IOE-OBP1 transgenic cell cultures lines (IOE#1 and IOE#3) were induced 2 days after subculture (data are means ± SD, = 3 flasks within one experiment). Asterisks indicate values that are significantly different from the respective controls (< 0.05, Student’s t-test).

Download figure to PowerPoint

To investigate if the increased rate of cell division was dependent on the culture growth phase we induced cultures with DEX straight after subculture and during the exponential growth phase. Increased cell division in comparison to mock-treated controls was observed in both phases (Figure 2d). To more precisely determine the timing of the observed differences we took measurements every 4 h for 20 h after DEX induction. These data revealed that cell number increased approximately 12 h after induction (Figure 3a,b). As these data indicated that OBP1 induction may shorten the duration of the cell cycle, we used flow cytometry to investigate the distribution of cell cycle phases. Within 4 h of DEX induction we measured a higher percentage of S-phase nuclei in induced samples at the expense of G1 nuclei. This suggests that the increase in cell number may be due to a shortening of the G1 phase or faster commitment of cells to enter the cell cycle (Figure 3c). To gain further evidence we used sucrose starvation to synchronise cells in G1 phase followed by re-addition of sucrose to allow cell cycle progression. To probe OBP1 function, re-addition of sucrose was accompanied by DEX induction and we compared the phase distribution of nuclei in samples harvested every 2–3 h for 15 h. In DEX-induced cultures, there was a steeper decline in the proportion of G1 nuclei and the S-phase peak occurred earlier, after 6 h instead of 8 h. Consistent with the previous experiment, this indicates that OBP1 activation caused an earlier transition from G1 to S phase that again may also be due to faster onset of the cell cycle. In addition, induction also caused the proportion of G2-phase nuclei to decline earlier, and this was accompanied by a sharp increase in G1-phase nuclei after 12–15 h. This indicates that the effect of OBP1 on the speed of G1/S transition is perpetuated through the G2 phase, resulting in earlier division (Figure 4).

image

Figure 3.  OBP1 affects the rate of cell proliferation. Exponentially growing (2 days after subculture) MM2d wild-type and IOE-OBP1 transgenic cell cultures (lines IOE#1 and IOE#3) were either dexamethasone (DEX) induced (5 μm) or mock induced (0.005% ethanol). Data for IOE-OBP1 transgenic cell cultures are means ± SD from four independent flasks in one experiment (two flasks for each line: IOE#1 and IOE#3). Asterisks indicate values that are significantly different from the respective controls (< 0.05, Student’s t-test). (a), (b) Cell number was counted every 4 h for 20 h. (c) Phase distribution of nuclei measured by flow cytometry.

Download figure to PowerPoint

image

Figure 4.  OBP1 shortens the length of the cell cycle. A sucrose starvation/re-addition experiment was performed as described by Menges and Murray (2002). Briefly cells in their mid-exponential growth phase were depleted from sucrose by washing and resuspending in sucrose-free media. Following a 24-h starvation period during which the majority of the cells stopped dividing and synchronised in G0/G1 phase sucrose was re-added (time 0) initiating cell cycle progression. Sucrose re-addition was accompanied by mock or dexamethasone (DEX) induction. Data for IOE-OBP1 transgenic cell cultures are means ± SD from four independent flasks in one experiment (two flasks for each line: IOE#1 and IOE#3). Phase distribution of nuclei was measured by flow cytometry. Asterisks indicate values that are significantly different from the respective controls (< 0.05, Student’s t-test).

Download figure to PowerPoint

To determine how the effects of OBP1 on cell proliferation and expansion influence growth and development, we performed phenotypic analysis of OBP1 transgenic plants. Constitutive overexpression resulted in drastically smaller plants compared with wild type. Seedlings were characterised by shorter hypocotyls (especially pronounced when grown in the dark) and stunted roots (Figure 5a–c). Later in development, 35S-OBP1 plants developed smaller, curly leaves that were occasionally lobed (Figure 5d) and characterised by unbranched trichomes (data not shown). Main roots were shorter with fewer lateral roots being formed (Figure 5f). Flowering was delayed and the main inflorescence was much shorter (Figure 5d). We could re-create an identical seedling phenotype for IOE-OBP1 plants grown on DEX. Also, rosettes sprayed with DEX develop similar phenotypes within 3–4 days of induction; developing leaves became curly and remained smaller whilst older leaves often became yellow and occasionally died (Figure 5e). At the cellular level, similar to the situation in cell cultures, epidermal cell size measured in leaves and cotyledons of 35S-OBP1 lines was approximately 50–70% of that of the control (Figure 5f). Interestingly, cell number calculated from cell size and total cotyledon/leaf area was increased in cotyledons but decreased in true leaves, further contributing to the dwarfish phenotype (Figure 5f). As cotyledons are initiated in the embryo, observed discrepancies may be due to developmental differences. In contrast, we observed no obvious phenotype for RNAi-OBP1 plants, which may be due to the incomplete silencing (Table 2b). Alternatively OBP1 function may be redundant with its Arabidopsis paralogue AtDOF5;6, that like OBP1 (see below) is also highly expressed in the shoot apical meristem (SAM). To distinguish between these possibilities future work will concentrate on obtaining complete loss-of-function mutants for both genes.

image

Figure 5.  Phenotypic analysis of OBP1 overexpression (OE) plants. (a) Seedlings of control (left) and IOE-OBP1 (right) plants grown on 5 μm dexamethasone (DEX) and in the dark. Hypocotyl of (b) control and (c) IOE-OBP1 dark-grown seedlings (bar = 200 μm), with green indicating an individual cell. (d) Flowering 35S-OBP1 plant. (e) IOE-OBP1 plant a week after DEX application. (f) Analysis of epidermal cells of cotyledons and the sixth rosette leaf of empty-vector control and 35S-OBP1 plants (data are mean ± SD, = 4–6). Root length and lateral root number were calculated for 10-day-old seedlings (data are means ± SD, = 30–50). Shading indicates values that are significantly different from the respective control plants (< 0.05; Student’s t-test).

Download figure to PowerPoint

As alterations in cell size can be associated with changes in endoreduplication we additionally analysed ploidy levels in leaves and dark- and light-grown hypocotyls. This revealed a significant reduction concomitant with decreased cell size (Table S7).

OBP1 expression correlates with cell division activity

To learn more about the biological role of OBP1 we examined its expression pattern. Mining publicly available microarray data was supported by the analysis of OBP1-GUS reporter lines. During embryogenesis, strong OBP1 expression was found in the embryo from the globular to the torpedo stage (Figure 6a–d). No GUS staining was seen in dormant embryos but it appeared again a few hours after the start of seed imbibition and stayed strong in the root apical meristem (RAM) and SAM and the cotyledon tips of germinating seedlings (Figure 6e,f). Expression in the RAM disappeared later in development but was visible in lateral root initials (Figure 6g–i). In contrast, OBP1 expression in the SAM was high throughout development, with the strongest GUS staining observed in stipules (Figure 6j). Moderate GUS staining was also observed for the whole leaf in very young leaves, but later first became restricted to the leaf tip and occasionally hydathodes (Figure 6k,l) and then disappeared completely at the senescence stage. In the stem, GUS staining appeared in the vasculature (procambial cells) and was strongest at the cut ends (Figure 6m,n). High OBP1 expression was also characteristic for axillary nodes, flower nectaries, callus and cell cultures during exponential phase of growth (Figure 6o,p).

image

Figure 6. OBP1 expression pattern defined by the analysis of OBP1-GUS lines. Embryo at the early (a) and late (b) globular stage, and at the early (c) and late (d) heart stage (bar = 50 μm). (e) Embryo at the end of stratification (bar = 75 μm). (f) Seedling 2 days after germination (bar = 1 mm). (g)–(i) Initiation of lateral root (bar = 100 μm). (j) Cross section of shoot apical meristem (SAM) (bar = 200 μm), (k) SAM of 1-week-old seedling (bar = 1 mm), (l) Hydathode (bar = 1 mm), (m) cut stem segment with axillary node (bar = 2 mm), (n) hand cut stem section (bar = 200 μm), (o) pedicle of flower (bar = 1 mm), (p) callus; MM2d cells transformed with OBP1-GUS construct (bar = 250 μm).

Download figure to PowerPoint

Using the Genevestigator meta-analysis tool (Zimmermann et al., 2004) we compared the tissue expression pattern of OBP1 and its eight early target genes. The expression patterns of these genes overlapped well, with relatively high expression in cell cultures, callus, SAM, radicle and root (Figure S4). Considering the expression of OBP1 in the RAM of germinating seedlings and a recent publication implicating activation of the cell cycle in radicle protrusion (Masubelele et al., 2005), we investigated the expression of these genes in seed germination expression profiling data. The expression of several D-type cyclins, including CYCD3;3, peaked shortly after the end of seed stratification, followed by other components of the core cell cycle machinery, and was preceded by increased OBP1 expression (Masubelele et al., 2005). This was confirmed by two other experimental data sets (AtGenExpress project) in which expression of OBP1 increased 3 h after imbibition, followed by its early targets (CYCD3;3, MYB-like TF, AtDOF2;3, MCM2 and MCM5) (Figure S5). Importantly, transcripts of CYCD3;3, MYB-like TF and AtDOF2;3 are relatively highly abundant in the radicles of young seedlings.

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

Influence of OBP1 on the cell cycle

The increasing availability of public microarray data has fuelled recent advances in investigating the function of genes via their digital expression pattern and through their co-expression with other genes of known function. A similar analysis for OBP1 suggested its potential involvement in cell proliferation. OBP1 is highly expressed in continuously proliferating tissues/organs (callus, cell cultures) as well as in cells that re-enter (G0/G1) cell division (e.g. lateral root initials). As expected from its expression pattern among the genes most co-expressed with OBP1 (ACT; Manfield et al., 2006), the majority can be related to proliferation, including several that are specific for G1/S transition and replication. Together these data may support a role for OBP1 during G0/G1 and G1/S transitions. In agreement, six of the eight early OBP1 target genes identified using IOE-OBP1 lines can be associated with G1 or S phase. CYCD3;3 and KRP7 are important regulators of cell cycle onset, whilst MCM2 and MCM5 encode for subunits of the mini-chromosome maintenance (MCM) complex, a major DNA helicase in eukaryotic organisms that is essential for replication (Inze and De Veylder, 2006). Among four induced transcription factors, the expression of two, AtDOF2;3 and MYB-like (At2g40970), was associated with the S phase in aphidicolin block/release experiments (Menges et al., 2003). In addition, the R2R3-type MYB88 TF was shown to restrict divisions late in the Arabidopsis stomatal cell lineage (Lai et al., 2005).

These molecular data indicating the involvement of OBP1 in cell division prompted us to transform MM2d cell cultures with our IOE-OBP1 construct to investigate the function of OBP1. As OBP1 is G1 specific (Menges et al., 2003) and targets the expression of genes associated with G0/G1 and G1/S transitions we hypothesized that OBP1 activation may affect onset of the cell cycle. Indeed, activation of OBP1 resulted in an increased number of cells within 12 h, related to a shortening of the G1 phase or faster commitment of cells to the cell cycle. Sucrose starvation/re-addition experiments further confirmed these results. These data, together with the fact that only the initial cell division was positively affected by OBP1 activation, may indicate that OBP1 is mainly involved in the control of cell cycle entry. Interestingly, and in accordance with induction of a large subset of mitotic genes in IOE-OBP1 plants 10 h after induction, the effect of OBP1 on the speed of G1/S transition in synchronisation experiments was perpetuated resulting in earlier division. The existence of a mechanism by which the activation of DNA replication stimulates mitotic entry was previously suggested and possibly involves the engagement of D-type cyclins (Inze and De Veylder, 2006). Other support for this hypothesis comes from the observation that CYCD3;3-associated CDKA activates both the G1/S and G2/M boundaries (Nakagami et al., 2002).

As mentioned previously, only the initial cell division was positively affected by OBP1 activation and subsequently the rate of cell division slowed to below control levels, eventually resulting in fewer cells. Importantly, a similar effect on cell division was also found in planta: short-term activation of OBP1 in fully developed rosettes led to the induction of cell proliferation-associated marker genes whilst fully developed 35S-OBP1 plants had leaves with fewer cells and lateral roots, suggesting cell cycle repression. This subsequent reduction in the rate of cell division might be related to the induction of negative cell cycle regulators such as KRP7 and MYB88 by OBP1. Simultaneous activation of both cell cycle activators and inhibitors was also shown for other TFs involved in the control of cell proliferation and is believed to provide a feedback mechanism for the tight control of cell division. A good example is the previously mentioned NTM1 TF, that controls expression of KRP2 and KRP7 as well as CYCD3;1 (Kim et al., 2006).

In summary, OBP1 modulates cell cycle activity by affecting the expression of CDKA regulators, S-phase specific transcription factors and components of the replication machinery, and is therefore important for cell cycle onset. The oscillation of OBP1 expression during the cell cycle (Menges et al., 2003), in combination with its effect on both cell cycle activators and inhibitors, confers a feedback mechanism assisting in the tight control of cell division (Figure 7).

image

Figure 7.  Model of the OBP1 regulatory network. OBP1 regulates expression of both activators (red font) and repressors (blue font) of the cell cycle, as well as other cell cycle-related genes (black font). We proved that CYCD3;3 and AtDOF2;3 are direct OBP1 targets (solid lines). We could also show that OBP1 negatively affects cell expansion. In addition OBP1 expression is regulated by the transcription factor ENHANCED SHOOT REGENERATION 2 (ESR2).

Download figure to PowerPoint

Influence of OBP1 on cell expansion

Reduced cell number may be compensated for by increased cell size and vice versa, phenomena described by the ‘compensated cell enlargement theory’ (Tsukaya and Beemster, 2006). As OBP1 affects cell number we also decided to look at cell size. Consistently, between different tissues and in cell cultures we measured a significant reduction of cell size in response to short- and long-term OBP1 activation. However, as leaves of 35S-OBP1 plants were also characterised by fewer cells, reduced cell size cannot be explained by the compensation theory and may suggest that OBP1 targets both cell expansion and cell proliferation. In contrast, reduced cell size correlated well with reduced level of endoreduplication that is often described in the literature. There have been similar findings for plants constitutively overexpressing CYCD3;1 (Dewitte et al., 2003). To look for possible molecular mechanisms that could explain changes in cell size we checked the expression of genes encoding growth-related proteins among transcripts downregulated in our profiling data and identified a large number of genes encoding for cell wall loosening enzymes essential for vacuolar cell growth, expansins (8 out of 36) and xyloglucan hydrolases (6 out of 33) (Li et al., 2003). Therefore, OBP1 may affect growth by targeting the expression of genes encoding for cell wall loosening enzymes (Figure 7).

OBP1 molecular interactions and their biological significance

We provide in vivo evidence for regulation of cell cycle genes by OBP1 and demonstrate that at least for two of these, CYCD3;3 and AtDOF2;3, this regulation is due to direct binding of OBP1 to their promoter sequences. Previously, in vitro evidence indicated that the ocs element in the GST6 promoter was bound by OBP1 (Chen et al., 1996; Zhang et al., 1995). As we demonstrated that 5 h was sufficient time for both binding to the promoter and regulating expression, we also performed these experiments for GST6. However, its expression was not changed and there was no enrichment of the promoter fragment containing the ocs element (data not shown). Importantly, in vivo proof of GST6 regulation is also lacking for the other OBPs, OBP2 and OBP3 (Kang et al., 2003; Skirycz et al., 2006; Ward et al., 2005). This highlights the limitations of in vitro-based studies in distinguishing between putative and functional interactions. Alternatively OBP1-mediated GST6 regulation may require stress conditions that were not tested in our study.

To identify putative regulatory factors upstream of OBP1 we searched the literature and available expression data. Interestingly, OBP1 was among 51 targets of the AP2/ERF TF ESR2 identified using expression profiling of estradiol-inducible ESR2 lines (Ikeda et al., 2006). We searched the OBP1 promoter for potential ESR2-binding sites and identified the [A/T]GCCGAC element which was previously pulled-down in a screen for possible binding sites of the ESR2 homologue ESR1 (Banno et al., 2006). Significantly, this element is also conserved in the promoters of OBP1 orthologues from Brassica oleracea and poplar (Figure S6). Furthermore, similarities in the tissue-specific expression pattern and the 35S phenotypes strengthen a possible connection between OBP1 and ESR2. This is particularly interesting, as ESR2, similarly to OBP1, is associated with cell proliferation (Chandler et al., 2007; Ikeda et al., 2006; Marsch-Martinez et al., 2006).

To establish the biological context in which OBP1 may regulate the cell cycle we looked for overlap in expression patterns between OBP1, its early targets and ESR2. As CYCDs and KRPs act as sensors that integrate developmental and environmental signals with the cell cycle machinery, we were particularly interested in processes that involve onset of the cell cycle. Considering OBP1, ESR2 and D-type cyclin expression in the root apex of germinating seedlings and a recent publication implicating cell cycle activation in radicle protrusion (Masubelele et al., 2005), we hypothesized that OBP1 may be involved in the activation of cell division in the RAM during seed germination. Consistent with this, the expression of OBP1 early target genes follows increases in OBP1 expression shortly after imbibition. Another possibility is the involvement of OBP1 in the activation of cell division during early cotyledon development. It was recently shown that ESR2 and its homologue ESR1 play a role in both cotyledon initiation and boundary maintenance (Chandler et al., 2007). Whilst the latter depends on ESR2-mediated activation of CUP-SHAPED COTYLEDON 1 (CUC1) (Ikeda et al., 2006), the molecular mechanism of ESR1/2 regulation of cotyledon initiation remains to be elucidated. Considering the involvement of OBP1 in the cell cycle and that, similarly to ESR2, OBP1 is expressed in the embryo it is possible that OBP1 may be involved in ESR2-mediated cotyledon initiation. In support of a role in cotyledon development, and in contrast to the situation in true leaves, cotyledons of 35S-OBP1 plants are characterised by increased cell number.

In summary, OBP1 regulates the expression of several cell cycle-associated genes, directly binding to the promoters of CYCD3;3 and TF AtDOF2;3 whilst itself being regulated by ESR2. Expression pattern and cell culture data strongly suggest that OBP1 is involved in cell cycle initiation in a developmentally specific manner (Figure 7). Importantly, the OBP1 orthologue from poplar is also expressed in proliferating tissues (Figure S7), suggesting that its function is conserved between different plant species. Future work will focus on gaining a better understanding of the biological context of the action of OBP1.

Experimental procedures

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

General methods

Standard molecular techniques were performed as described in Sambrook et al. (2001). Primers were obtained from MWG (http://www.mwg-biotech.com/). DNA sequencing was performed by AGOWA (http://www.agowa.de/). Unless indicated, chemicals were purchased from Roche (http://www.roche.com/), Merck (http://www.merck.com/), Invitrogen (http://www.invitrogen.com/) or Sigma (http://www.sigmaaldrich.com/). For sequence analyses the tools provided by the National Center for Biotechnology Information (http://www.ncbi.nlm.nih.gov/), the Arabidopsis Information Resource (TAIR; http://www.arabidopsis.org/) and the Plant Transcription Factor Database (http://plntfdb.bio.uni-potsdam.de/v2.0/) were used.

Plant material

Long-day conditions were as follows: a 16-h day length (200 μmol m−2 sec−1), a day/night temperature of 20/16°C and relative humidity of 60/75%. Short-day conditions were as follow: 8-h day (150 μmol m−2 sec−1), a day/night temperature of 20/18°C and relative humidity of 60/75%. Agrobacterium tumefaciens strain GV3101 (pMP90) was used to transform Arabidopsis thaliana (L.) Heynh. cv. C24 (Clough, 1998). In vitro, seedlings were grown in half-strength Murashige and Skoog medium (Murashige and Skoog, 1962) supplemented with 1% sucrose under a 16-h day (140 μmol m−2 sec−1)/8-h night regime. For root experiments plants were grown vertically under the same conditions. If not indicated differently, measurements were performed on fully developed rosettes harvested from randomized 5- to 6-week-old plants prior to bolting. For induction experiments plants were sprayed with 25 μm DEX solution or grown in media containing 5 μm DEX.

Cell cultures

The MM2d cell cultures were kindly provided by Professor James Murray (University of Cambridge, UK). Cultures were maintained as described by Menges and Murray (2002). For induction experiments cells were treated with 5 μm DEX or with 0.005% ethanol (EtOH; mock). For cell number and size analysis cell clumps were eliminated by 30-min incubation with equal volumes of 1% cellulase and 0.6% macerase in 0.6 m sorbitol at 37°C. Cells were counted using a counting chamber and Olympus BX51 microscope whilst cell size was analysed using cell^P Software (Olympus, http://www.olympus-global.com/en/global/). Sucrose starvation was performed as described by Menges and Murray (2002). Transformation of MM2d cells was performed as described by Menges and Murray (2004).

Constructs

Primer sequences are listed in Table S8. 35S-OBP1: PCR amplified OBP1 coding region was PCR amplified, inserted into pUni/V5-His-TOPO and cloned via the PmeI/PacI sites into a modified pGreen0229 plant transformation vector (http://www.pgreen.ac.uk/) containing a CaMV 35S promoter. OBP1-GUS: PCR amplified ∼2-kb fragment upstream of the ATG start codon was inserted into plasmid pCR-Blunt II-TOPO and cloned via SalI and XbaI sites into pGPTV-Kan vector (Becker et al., 1992). IOE-OBP1: PCR amplified OBP1 coding region was inserted into TOPO-TA and cloned via the XbaI sites into a d143 (pBI-GR) vector (Lloyd et al., 1994). RNAi-OBP1: PCR amplified 3′-UTR OBP1 fragment was cloned into pENTR/D-TOPO and transferred to the destination vector pJawohl8-RNAi (provided by Dr Imre Somssich, MPI for Plant Breeding, Cologne) by LR recombination reaction. All clones were confirmed by sequencing.

Quantitative real-time PCR

Total RNA was isolated using TRIzol as described in the manual. Synthesis of cDNA and Q-RT-PCR were performed as described in Czechowski et al. (2004). Primer sequences are given in Tables S9 and S10. Data were normalized to ACTIN2 (At3G18780), UBIQUITIN10 (At4G05320), GAPDH (At1G13440), PDF2 (At1G13320) and expressed protein (At2G32170); ΔCt = Ct (gene) − Ct [mean (house-keeping genes)]. Ct refers to the number of cycles at which SYBR Green fluorescence reaches an arbitrary value during the exponential phase of DNA amplification (set at 0.2).

Microarray experiments

Samples were pooled from three plants in a single biological experiment. Extraction, labelling, hybridization and scanning were done as described by Redman et al. (2004). Each sample was hybridized to a single Affymetrix ATH1 Genome array (ATH1) at the German Resource Center for Genome Research, Berlin. Data were Robust Multi-Chip Average normalized as described in Irizarry et al. (2003). Microarray data are deposited in ArrayExpress (accession number E-MEXP-1360).

Chromatin immunoprecipitation

Plant material was processed as described by Leibfried et al. (2005). One microgram of polyclonal anti-GR antibody (Abcam, http://www.abcam.com/) was used per sample. Quantitative PCR was performed as described in Czechowski et al. (2004). Primer sequences are given in the Table S11. Data were normalized to genomic fragment of UBIQUITIN10 (At4G05320), ΔCt Ct (promoter fragment) − Ct (UBQ fragment).

Electrophoretic-mobility shift assay

Protein purification.  The expression vector in Escherichia coli strain SCS1/pSE111 was kindly provided by Dr Birgit Kersten (Kersten et al., 2003). For the induction 1 mm IPTG was used, whilst protein purification was performed under native conditions using Ni2+-agarose according to the Qiagen protocol (http://www.qiagen.com/).

Electrophoretic-mobility shift assay.  The IRD-800 labelled DNA fragment (5′-AAATAATCATAAAGTATTAAAGTAATATATAAC), unlabeled competitor fragments (5′-AAATAATCATAAAATATTAAAGTAATATATAAC and 5′-AAATAATCATAAAATATTAAAATAATATATAAC) and their complementary strands were ordered from MWG. Binding reaction was performed at ∼21°C for 20 min in 10 μl volume {2 μl 5× binding buffer [100 mm HEPES, pH 7.6; 5 mm EDTA; 50 mm (NH4)2SO4; 25 mm DTT; 1% (w/v) Tween 20; 150 mm KCl]; 1 μl 0.1 mm labelled DNA fragment; 1 μl of BSA (10 mg ml−1); 2 μl poly d(I-C) (0.35 μg μl−1); approximately 70 ng of OBP1 protein and 10× or 100× molar excess of unlabelled competitor sequence}. DNA–protein complexes were separated on 6% DNA retardation gels (Invitrogen) whilst the IRD-800 signal was detected using the Odyssey Infrared Imaging System from LI-COR Biosciences (http://www.licor.com/).

GUS assays

β-Glucuronidase activity was determined histochemically as described (Skirycz et al., 2006). Pictures were taken using a MZ 12.5 stereomicroscope (Leica, http://www.leica.com/) or a BX51 light microscope (Olympus). Embryos were visualized as described in Chandler et al., (2007).

Microscopy

Epidermal cell sizes were analysed following removal of chlorophyll with 100% ethanol using a BX51 light microscope and cell^P Software (Olympus).

Flow cytometry

Frozen cell culture samples were processed using Cystain UV Precise P kit (Partec, http://www.partec.com/). Subsequently nuclei were analysed with the CyFlow flow cytometer with FloMax Software (Partec) and the G1/S/G2 distribution was estimated using the DNA cell cycle analysis software Multicycle AV for Windows (Phoenix Flow Systems, http://www.phnxflow.com/).

Acknowledgements

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

We thank Dr Imre Somssich (MPI for Plant Breeding, Cologne) for providing the pJawohl8-RNAi vector, and Professor James Murray (University of Cambridge, UK) for providing the MM2d cells. We thank Dr Eugenia Maximowa for the help with microscopy, Dr Karin Koehl and her MPI Green Team for plant care and Josef Bergstein for photography (all MPI, Molecular Plant Physiology, Potsdam-Golm). AS thanks the Ernst Schering Foundation, Berlin, for providing a doctoral fellowship; she is also a member of the International PhD Programme ′Integrative Plant Science′ (IPP-IPS) funded by the DAAD (Deutscher Akademischer Austauschdienst) and the DFG (Deutsche Forschungsgemeinschaft) under no. DAAD Az. D/04/01336. Funding of this research was provided by BMBF (GABI Program, FKZ 0312276M) and Bayer Crop Science. Further support was provided by the Interdisciplinary Research Centre ′Advanced Protein Technologies′ (IZ-APT) of the University of Potsdam. BM-R thanks the Fonds der Chemischen Industrie for funding (no. 0164389).

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

Supporting Information

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

Figure S1.OBP1 over-expression regulates many core cycle genes. (a) Schematic representation of dexamethasone (DEX) inducible system used to uncover OBP1 target genes. Upon DEX binding the glucocorticoid (GR) receptor domain undergoes conformational changes that trigger nuclear targeting of OBP1-GR fusion protein. (b) Up or downregulated core cell cycle genes as revealed by transcript profiling of IOE-OBP1 lines in comparison to empty-vector control plants 10 h after DEX induction. Data are from one biological experiment; each sample was pooled from three fully developed Arabidopsis rosettes.

Figure S2. Cell cycle related genes are enriched among transcripts up-regulated in IOE-OBP1 lines. We calculated fold enrichment of genes that expression was shown to oscillate during the cell cycle (Menges et al., 2003) among transcripts up- and down-regulated by OBP1 activation.

Figure S3. Approximately 1-kb promoter fragments of AtDOF2;3 (upper panel) and CYCD3;3 (lower panel) genes. Sequence features are highlighted: ATG, green; 5′ untranslated region (UTR), bold; putative DOF elements (AAAG, CTTT, [A/T]AAAG, CTTT[A/T]), grey, and enriched fragment, blue.

Figure S4. Meta-analysis of organ-specific expression of OBP1, its early targets and GST6 was performed using the Meta-Analysis tool available at the Genevestigator website (Zimmermann et al., 2004; Schmid et al., 2005). Intensity of the blue shading corresponds to expression level.

Figure S5. Expression of OBP1 and its early targets is associated with germination. Graphs represent fold change (log2) of expression of OBP1 and its early targets during seed imbibition. Data were retrieved from two germination experiments of the AtGenExpress project (Expression profiling of early germinating seeds and Effect of ABA during seed imbibition, Kamiya and Nambara).

Figure S6. Promoter analysis of OBP1 and its orthologs from poplar and Brassica oleracea. In an attempt to identify conserved factors up-stream of OBP1 we looked for orthologous genes in the closely related species Brassica oleracea. TIGR (The Institute for Genome Research) and the CSHL/WU (Cold Spring Harbor Laboratory/Washington University) consortium are currently performing genome shotgun sequencing with a goal of reaching 0.5–1× coverage of the Brassica oleracea genome. In parallel they are fully sequencing six Brassica BAC contigs identified by O’Neill and Bancroft (2000) as being homologous to a region of the Arabidopsis genome that is duplicated between chromosomes 4 and 5. A BLAST server at TIGR allows retrieval of sequences from the preliminary contigs and we used it to look for sequences similar to OBP1 and its up-stream region. We found a 819 bp long fragment homologous to OBP1 ORF, 5′ UTR and approximately 400 bp of promoter region. Information on the OBP1 ortholog from poplar was retrieved from Plant Transcription Factor Databases website (http://planttfdb.cbi.pku.edu.cn/). Sequence comparison was performed by ClustalW (http://www.ebi.ac.uk/Tools/clustalw/). Red shading indicates the putative ESR2 binding site; green shading indicates the ATG start codon.

Figure S7. Expression pattern of OBP1 ortholog from poplar. Information on the OBP1 ortholog from poplar was retrieved from the Plant Tanscription Factor Databases website (http://planttfdb.cbi.pku.edu.cn/). (a) Expression across wood-forming meristems. Data are from two independent experiments (Schrader et al., 2004). X1–X10, sections across wood-forming meristem. The red frame indicates cambial cells. C, cambium; E, expansion zone; SW, secondary wall formation zone; LM, late maturation zone. (b) Expression in different meristems (A, apical; C, cambial; R, root) and in leaves (L) (Schrader et al., 2004).

Table S1. Top 48 genes co-expressed with OBP1. Co-expression analysis was performed using the Arabidopsis Co-expression Tool (Manfield et al., 2006). Over-representation analysis was performed using PageMan (Usadel et al., 2006).

Table S2. Up-regulated transcripts. Genes differentially expressed between DEX treated empty-vector control and IOE-OBP1 plants. Genes were considered to exhibit an altered expression level when the hybridization signal was >2-fold increased in comparison between the two sets of plants. Over-representation analysis was performed using PageMan (Usadel et al., 2006).

Table S3. Down-regulated transcripts. Genes differentially expressed between DEX treated empty-vector control and IOE-OBP1 plants. Genes were considered to exhibit an altered expression level when the hybridization signal was >2-fold decreased in comparison between the two sets of plants. Over-representation analysis was performed using PageMan (Usadel et al., 2006).

Table S4. Overlap between genes affected in IOE-OBP1 plants and genes that expression was shown to oscillate during cell cycle in aphidicolin block/release experiments (Menges et al., 2003).

Table S5.. Overlap between genes affected in IOE-OBP1 plants and proliferation-associated genes identified by (Beemster et al., 2005).

Table S6. List of core cell cycle genes and genes encoding for cell wall loosening enzymes that expression was changed in IOE-OBP1 plants 10 h after DEX application.

Table S7. Ploidy measurements. (a) Ploidy distribution in hypocotyls of dark or light grown one week old seedlings. Data are means ± SD (n = 3); no SD (n = 1). Each sample was pooled from approximately 30–50 seedlings grown on separate agar plates. (b) Ploidy distribution measured in fully developed leaf six. Data are means ± SD (n = 3–5).

Table S8. List of primers used for cloning. Added restriction sites (RS) are underlined.

Table S9. List of Q-RTPCR primers.

Table S10. Q-RTPCR primer platform for core cell cycle genes and TFs that expression oscillates during cell cycle (Menges et al., 2003). The majority of the TF primer sequences were taken from Czechowski et al. (2004).

Table S11. Q-PCR primers used for ChIP.

Please note: Wiley-Blackwell are not responsible for the content or functionality of any supporting materials supplied by the authors. Any queries (other than missing material) should be directed to the corresponding author for the article.

FilenameFormatSizeDescription
TPJ_3641_sm_figsS1-S7.pdf95KSupporting info item
TPJ_3641_sm_tablesS1-S11.xls489KSupporting info item

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.