|Home | About | Journals | Submit | Contact Us | Français|
In yeasts and animals, cyclin-dependent kinases are key regulators of cell cycle progression and are negatively and positively regulated by WEE1 kinase and CDC25 phosphatase, respectively. In higher plants a full-length orthologue of CDC25 has not been isolated but a shorter gene with homology only to the C-terminal catalytic domain is present. The Arabidopis thaliana;CDC25 can act as a phosphatase in vitro. Since in arabidopsis, WEE1 plays an important role in the DNA damage/DNA replication checkpoints, the role of Arath;CDC25 in conditions that induce these checkpoints or induce abiotic stress was tested.
arath;cdc25 T-DNA insertion lines, Arath;CDC25 over-expressing lines and wild type were challenged with hydroxyurea (HU) and zeocin, substances that stall DNA replication and damage DNA, respectively, together with an abiotic stressor, NaCl. A molecular and phenotypic assessment was made of all genotypes
There was a null phenotypic response to perturbation of Arath;CDC25 expression under control conditions. However, compared with wild type, the arath;cdc25 T-DNA insertion lines were hypersensitive to HU, whereas the Arath;CDC25 over-expressing lines were relatively insensitive. In particular, the over-expressing lines consistently outgrew the T-DNA insertion lines and wild type when challenged with HU. All genotypes were equally sensitive to zeocin and NaCl.
Arath;CDC25 plays a role in overcoming stress imposed by HU, an agent know to induce the DNA replication checkpoint in arabidopsis. However, it could not enhance tolerance to either a zeocin treatment, known to induce DNA damage, or salinity stress.
The cell cycle is controlled by cyclin-dependent kinases (CDKs), an evolutionarily conserved complex. Their kinase activity is dependent, in part, on regulatory subunits called cyclins, and, in part, on reversible phosphorylation (Francis, 2008). In fission yeast, the mitotic CDK, Cdc2, is positively regulated by Cdc25 phosphatase that removes a phosphate on tyrosine 15 and is negatively regulated by WEE1 kinase that phosphorylates the same tyrosine residue (Russell and Nurse, 1987; Moreno et al., 1990). During the human cell cycle, three CDC25 isoforms were identified: CDC25A acts at the G1/S transition (Molinari et al., 2000) while Cdc25B and C function at the G2/M transition (Nilsson and Hoffmann, 2000).
The first plant CDC25 phosphatase was identified in the green unicellular alga, Ostreococcus tauri. This gene was capable of rescuing the yeast S. pombe cdc25-22 conditional mutant and its protein product dephosphorylated a CDK/CyclinB complex in vivo (Khadaroo et al., 2004). In higher plants, a CDC25-like protein was identified in arabidopsis, Arath;CDC25, that was able to activate CDKs in vitro (Landrieu et al., 2004) and it could induce a short cell length when over-expressed in fission yeast (Sorrell et al., 2005). However, it only consists of a C-terminal catalytic domain while animal and yeast CDC25s also have an N-terminal regulatory domain. Also it is unable to complement the temperature-sensitive cdc25-22 yeast mutant and, unlike other cell cycle regulatory genes, Arath;CDC25 is not up-regulated in rapidly dividing tissues (Landrieu et al., 2004; Sorrell et al., 2005; Dissmeyer et al., 2009). A plant WEE1 gene has also been identified (Sorrell et al., 2002) but T-DNA insertion mutants indicate that it is not an essential gene for normal cell division (De Schutter et al., 2007).
In yeast and animals, CDC25 phosphatase activity is negatively regulated in stress conditions that induce either the DNA replication or DNA damage checkpoint pathways in response to a perturbation of DNA replication or chromosomal damage, respectively. These pathways delay entry into mitosis until DNA replication is normalized or until single- or double-strand breaks are repaired (Hartwell and Weinert, 1989). In yeast and animals, these insults to DNA are detected by the RAD3 and ATM/ATR proteins, respectively (Elledge, 1996; Abraham et al., 2000), resulting in phosphorylation and activation of the CHK1/2 kinases. The CHK kinases are conserved in yeast and animals (al-Khodairy et al., 1994; Sanchez et al., 1997; Zachos et al., 2003) and under checkpoint conditions and through phosphorylation are responsible for the in vivo repression of CDC25 phosphatase activity (Gabrielli et al., 1997; Rothblum-Oviatt et al., 2001; Chen et al., 2003). This checkpoint pathway also stabilizes WEE1 kinase activity to prevent cells from dividing (Branzei and Foiani, 2008). In animals, CDC25C binds to a 14-3-3 protein, which leads to cytoplasmic sequestration of the phosphatase blocking its access to mitotic CDK (Dalal et al., 1999; Zeng et al., 1999; Donzelli and Draetta, 2003). In arabidopsis, the fission yeast cdc25 could also bind to a plant 14-3-3 in yeast two-hybrid assays (Sorrell et al., 2005).
In plants, the DNA replication and damage checkpoints have only now started to be understood. Deploying arabidopsis knockout mutants deficient in ATM or ATR kinase, atm plants were hypersensitive to DNA-damaging agents, such as γ-irradiation, but rather insensitive to replication blocking agents, such as hydroxyurea (HU) or aphidicolin (Garcia et al., 2003; Culligan et al., 2004). Conversely, atr mutants were hypersensitive to replication blocking agents but only mildly sensitive toward γ-irradiation (Culligan et al., 2004). These findings strongly indicate that ATM is the primary sensor of chromosomal damage whereas ATR senses stalled DNA replication, in mechanisms that seem to be well-conserved in higher eukaryotes. However, the absence of both CHK1-like kinases and a full-length CDC25 in the arabidopsis genome suggests that under checkpoint-induced conditions higher plants have a different signalling pathway compared with other eukaryotes. Nevertheless, and just like the DNA replication checkpoint in animals, WEE1 is highly expressed in arabidopsis treated with HU, and arath;wee1 T-DNA knockout lines are hypersensitive to HU, leading to the conclusion that in plants WEE1 has a more restricted role in the plant cell cycle, perhaps confined to checkpoint pathways (De Schutter et al., 2007).
Arath;CDC25 can also function as an arsenate reductase. Under arsenate stress conditions, CDC25oe plants are less affected in root growth compared with wild-type controls and knockout mutants (Dhankher et al., 2006; Bleeker et al., 2006). Even though Arath;CDC25 can also exhibit phosphatase activity (Landrieu et al., 2004), the case for its role in checkpoint control is uncertain. To address this question, root growth characteristics of arabidopsis lines either lacking Arath;CDC25 or over-expressing Arath;CDC25 and their response to HU or zeocin were examined. HU stalls DNA replication through inhibition of the enzyme, ribonucleotide reductase, thereby inducing the DNA replication checkpoint (Eklund et al., 2001) whereas zeocin is a radiomimetic drug known to induce breaks in plant chromosomes (Trastoy et al., 2005) and has been used to induce the DNA damage checkpoint (De Schutter et al., 2007). To test whether Arath;CDC25's role in arsenate stress can be widened to other types of abiotic stress, all lines exposed to NaCl treatment were also tested.
Neither over-expression nor mutation of the Arath;CDC25 gene affected growth and development in arabidopsis under control conditions. However, the T-DNA insertion lines, lacking CDC25 expression, were hypersensitive to HU compared with wild type (WT), and, conversely, the Arath;CDC25oe lines were relatively insensitive. All genotypes were equally sensitive to zeocin and salt. Hence, the data reported here suggest a role for Arath;CDC25 in plants challenged specifically with HU but not other forms of abiotic stress.
Two homozygous T-DNA insertion lines (GABI-Kat 772G06 and SALK_143282) and two homozygous over-expression lines of At5G03455 driven by the CaMV 35S promoter (Benfey and Chua, 1990) were kindly provided by Professor Henk Schat, Vrije Universiteit Amsterdam (Bleeker et al., 2006). Seeds were surface sterilized and grown on M&S medium supplemented with 1 % (w/v) sucrose and grown vertically in 90-mm Petri dishes in long days (16 h light/8 h dark; 300–400 µmol m−2 s−1). For stress-related experiments, either 1 or 2 mm HU (Sigma Aldrich, UK), 10 or 50 µM zeocin (Invitrogen, UK), or between 0 and 200 mm NaCl was added to the medium. For phenotyping, the seeds were sown 1·5 cm apart along a horizontal line toward the top of the plates.
For genotyping, leaf material was harvested and genomic DNA was extracted according to Edwards et al. (1991). Total RNA was extracted from arabidopsis seedlings using TRI Reagent (Sigma Aldrich) essentially as described by the manufacturer. Residual genomic DNA was removed by DNase treatment (Promega). cDNA was synthesized using MLV-RT and oligo dT (Promega). Specific primers were designed to genotype the insertion mutants (SALK 143282: Atcdc25tDNA1 GCGTGGACCGCTTGCTGCAACT and Atcdc25/Rho2 GGGAATTCGTTTAGGCGCAATCGCCCTTGCAAG, melting temperature (Tm) 59 °C; GABI-Kat772G06: Atcdc25tDNA2 CTGGGAATGGCGAAATCAAGGCATC, and Atcdc25/Rho2 Tm 59 °C) and WT Arath;CDC25, At5G03455 (Atcdc25F 5′-CGTCTCGAGATGGGGAGAAGCATATTTTCC-3′ and Atcdc25R TACGACTAGTTTAGGCGCAATCGCCCTT; Tm 55 °C). Semi-quantitative RT-PCR was used to analyse expression levels: PCR cycle number was optimized for each of the specific primer pairs and cDNA batch by carrying out PCR on a set of standards (cDNA dilutions) over a range of cycles to ensure a linear relationship between template and product concentration. 18S primers (PUV2 TTCCATGCTAATGTATTCAGAG and PUV4 ATGGTGGTGACGGGTGAC) were used to normalize the data, relying on the high AT content of rRNA. This approach has been used successfully for a range of experimental systems in the Cardiff laboratory (including Orchard et al., 2005; Price et al., 2008). All RT-PCR reactions were repeated at least three times and standards included in each PCR run. Ethidium bromide fluorescence was quantified using the Gene Genius Bioimaging System and Gene Tools software package (Syngene Ltd).
Seedlings were fixed in situ on the Petri dish in 3 : 1 (v/v) absolute ethanol : glacial acetic acid fixative for at least 24 h at 5 °C. Fixed seedlings were then rinsed thoroughly in distilled water, treated with 5 m hydrochloric acid at room temperature for 30 min, rinsed three times with ice cold distilled water (each rinse for 5 min), before Feulgen stain was added and the plates were incubated at room temperature for 24 h. Root measurements were made using a BH-2 Olympus research microscope using either the ×10 or ×20 objectives. Images of the seedlings were captured using a FUJI 1X digital camera HC-300Z or an Epson digital scanner.
Approximately 0·5 cm2 of the mature 5th leaf of the rosette of plants (approx. 4 weeks old) were chopped with a sharp razor blade in 500 µL of ice cold extraction buffer (Partec). The nuclear suspension was filtered through CellTrics 30-μm filters (Partec) directly into the sample tube and 1 mL of staining buffer (DAPI) was added. For the analysis of the nuclei, a Partec Ploidy Analyser PA II with UV excitation by mercury arc lamp was used. The data analysis was carried out with PA II's Partec FlowMax 2·4b (XP) software.
Expression levels of Arath;CDC25 were measured in 14-d-old seedlings of two T-DNA insertion homozygous lines of Arath;CDC25: GABI-Kat arath;cdc25 carrying the T-DNA insertion in the second exon (line ID772G06; T-DNA start position 863262) and SALK arath;cdc25 carrying the T-DNA insertion in the second intron (SALK_143282; T-DNA start position 863366) (Fig. 1). Transcript was not detectable in either insertion line (Fig. 2A) confirming that these are indeed knockout lines. Over-expression of Arath;CDC25 was confirmed by semi-quantitative RT-PCR in two transgenic lines (#18 and #22) (Fig. 2B) and was in agreement with previously published expression levels (Bleeker et al., 2006).
Primary root length and the total number of lateral roots and lateral root primordia (referred to from here on as laterals) were analysed in lines in which expression of Arath;CDC25 had been perturbed, and compared with WT (Fig. 3A and B).
Differences were not evident in root phenotype for the GABI-Kat T-DNA insertion line (referred to from here on as gabi) compared with WT (P > 0·05; Fig. 3A and B) and, although the SALK T-DNA insertion line (referred to from here on as salk) showed a slight reduction in the number of laterals compared with WT, it was not significant (P > 0·05). The primary root length of Arath;CDC25oe line 18, but not line 22, was significantly longer compared with WT (P < 0·001) in 10-d-old seedlings (Fig. 3A and B). However, there were no significant differences in the number of laterals formed in this line compared with WT (Fig. 3A and B; P > 0·05). Indeed, the rate of lateral initiation per millimetre of primary root (referred to from here on as rate of lateral initiation) of 0·18 in Arath;CDC25oe line 18 was not significantly different from that of 0·21 in WT (Fig. 3). Overall, the clustering of data in Fig. 3B is consistent with a null effect of perturbation of Arath;CDC25 expression on either root length or the total number of laterals formed under control conditions.
Final phenotypic considerations under control conditions were of cell size and the extent of endoreduplication in the genotypes perturbed in Arath;CDC25 expression, traits known to alter in response to abiotic stress. Flow cytometry was used to measure the nuclear DNA content of Arath;CDC25oe line 18 and the salk T-DNA insertion line compared with WT. There was no significant change in the DNA content between the lines used and WT (see Fig. S1A in Supplementary data, available online). Epidermal cell length along the primary root apical meristem was also measured for Arath;CDC25oe line 18 and the salk T-DNA insertion line and compared with WT in 10-d-old seedlings. There were no significant changes in cell length in either line compared with WT (see Fig. S1B). These results further illustrate a null effect of the perturbation of Arath;CDC25 in arabidopsis roots grown under control conditions.
In arabidopsis, WEE1 has a role that seems to be restricted to the DNA replication and damage pathways (De Schutter et al., 2007; see Introduction). If, similarly, Arath;CDC25 has functions restricted to checkpoint pathways, it might be expected that genotypes with perturbed levels of CDC25 expression would react differentially to treatments that induce checkpoints compared with WT. Therefore, the effect of inducing the DNA replication checkpoint using HU was examined. Ten-day-old seedlings of both salk and gabi T-DNA insertion lines, exhibited a significant decrease (1·11- to 1·27-fold) in primary root length compared with WT when grown in the presence of 1 mm HU (Fig. 4C; P < 0·05). In contrast, the over-expression lines, Arath;CDC25oe 18 and 22, showed a highly significant increase (2·25- to 2·5-fold) in primary root length compared with WT (P < 0·001) (Fig. 4A and C). The salk T-DNA insertion line made significantly more laterals than WT (P < 0·001) (Fig. 4A) but this was not consistent between the two T-DNA insertion lines. However, in both of the over-expressing lines there was a 4-fold increase in the total number of laterals compared with WT (P < 0·001) with a maximal 2·7-fold increase in the rate of lateral root initiation compared with WT (P < 0·001; Fig. 4A and C).
The 2 mm HU treatment resulted in a reduction of primary root growth in all genetic backgrounds (Fig. 4B and D). However, whilst the data for both T-DNA insertion lines clustered tightly with WT data, Arath;CDC25oe lines 18 and 22 showed a 2- to 2·2-fold increase in primary root length relative to WT (P < 0·001). Also there was a 2·3- to 2·6-fold increase in the number of laterals in the over-expressing lines compared with WT (P < 0·001). Hence in 10-d-old seedlings, both T-DNA insertion lines were hypersensitive to HU at the 1 mm treatment, whilst both over-expressing lines were relatively insensitive at both HU concentrations.
To test whether the effects of Arath;CDC25 perturbation persisted in older seedlings, a new experiment spanning 17 d was established in which root length was measured daily. A highly significant differential primary root growth response was elicited by 1 mm HU in lines perturbed in Arath;CDC25 expression compared with WT (Fig. 5). Primary roots of both T-DNA insertion lines grew more slowly than WT but the over-expressing lines grew faster than WT (Fig. 5A–D). Moreover, the T-DNA insertion lines initiated fewer lateral roots compared with WT whilst the over-expressing lines produced significantly more laterals with a 1·3- to 1·6-fold increase in the rate of lateral initiation compared with WT (Fig. 5E). Hence, in this longer-term experiment, lines deficient in Arath;CDC25 expression continued to be hypersensitive to HU whilst the over-expressing lines grew faster and initiated more laterals compared with WT.
Since arath;wee1 seedlings are hypersensitive to HU and Arath;WEE1 expression is induced by HU (De Schutter et al., 2007), the effect of HU treatment on Arath;CDC25 expression was tested in 10-d-old WT seedlings grown in the presence of 1 or 2 mm HU. There were no significant changes in the transcript level of Arath;CDC25 in the treatment with 1 or 2 mm HU compared with the control (P > 0·05 in all cases; Fig. 6).
Zeocin induces nuclear chromosomal DNA breaks in arabidopsis and, hence, it can be used instead of ionizing radiation to induce the DNA damage checkpoint (Trastoy et al., 2005; De Schutter et al., 2007). Both 10 and 50 µm zeocin inhibited primary root growth to a remarkably similar extent in all genotypes (Fig. 7A and B). Highly significant within-genotype differences were recorded for root length (Fig. 7A–C). In all cases, there was a commensurate decrease in the number of lateral root primordia with respect to primary root length (data not shown). Hence all genotypes were equally sensitive to zeocin in terms of both primary root elongation and the initiation of lateral roots.
To test whether the differential response of these genotypes was more specific to HU treatment, another form of abiotic stress, salt, was tested. An initial trial of WT seedlings on 0–200 mm NaCl indicated 50 mm as a suitable LD50 (Fig. 8A). One Arath;CDC25 over-expressor (line 18) and one T-DNA insertion line (gabi) were tested for root growth and all seedlings were equally sensitive to 50 mm NaCl (Fig. 8B). Ten-day-old seedlings of all genotypes were also equally sensitive to 75 µm cadmium (data not shown)
The major finding presented here is that two T-DNA insertion lines for Arath;CDC25 were hypersensitive to HU whilst two Arath;CDC25 over-expressing lines grew faster and produced more laterals than WT. In an extended HU treatment, a lack of, or over-expression of, Arath;CDC25 resulted in a 30 % decrease and a 15 % increase in final root length, respectively. In other words, on HU, the T-DNA insertion lines grew more slowly whilst the over-expression lines grew faster compared with WT. Thus, HU did not block growth permanently in the arath;cdc25 T-DNA insertion lines as one might expect if CDC25 were the sole positive regulator of the HU-induced response.
In a range of unrelated organisms, HU has been used to induce a DNA replication checkpoint (see Rhind and Russell, 2000; De Schutter et al., 2007); it stalls DNA replication (Boddy and Russell, 2001). The HU-induced differential response of genotypes perturbed in Arath;CDC25 expression suggests strongly that more proliferative cells by-pass this checkpoint when Arath;CDC25 is over-expressed, fewer do so in WT, and even fewer in the T-DNA insertion lines. However, the data do not support a hypothesis that Arath;CDC25 is the sole positive regulator in a plant DNA replication checkpoint because, if so, the T-DNA insertion lines should simply stop growing on 1 mm HU. Hence, the current consensus view of the plant DNA replication checkpoint, that HU activates WEE1 to suppresses CDK activity thereby blocking the G2/M transition, cannot be extended to include Arath;CDC25 as the sole opponent of this negative regulation as CDC25 most certainly can in animal cells (Donzelli and Draetta, 2003; Karlsson-Rosenthal and Millar, 2006). However, the present data implicate Arath;CDC25, perhaps operating redundantly with other plant phosphatases, as an important component in recovery from perturbed DNA replication.
In contrast to expression of fission yeast, Spcdc25 in plant cells (Orchard et al., 2005) there is a null phenotypic effect of over-expression of Arath;CDC25 under control conditions. However, its absence renders the plant hypersensitive to HU. It could be hypothesized that, Arath.;CDC25's target in ameliorating a DNA replication checkpoint is the dephosphorylation of CDKs that then drive cells into mitosis (see Introduction). However, the maximal 15 % increase in root length of the over-expressing lines over WT and the maximal 30 % decrease in root length in the T-DNA insertion lines represent subtle and hitherto unreported phenotypic effects induced by HU. This makes it highly unlikely that current techniques would quantify a preferential dephosphorylation of CDKs in the over-expressing lines compared with WT or, indeed, quantify diminished levels of dephosphorylation in the T-DNA insertion lines because all genotypes showed respectable rates of root elongation even after 17 d on HU.
The failure of Arath;CDC25 over-expression to ameliorate DNA damage stress imposed by zeocin shows that this gene is unable to oppose inhibition of root growth imposed by Arath;WEE1 in the DNA damage checkpoint (De Schutter et al., 2007). This reveals a distinct difference in the plant DNA damage checkpoint pathway in animals where WEE1 and CDC25 are the targeted adversaries acting on CDK1 (see Introduction). However, in animals both ATM and ATR can sense DNA damage (Rhind and Russell, 2000) whereas in plants ATR is the sensor for perturbation of DNA replication whilst ATM senses DNA damage (Culligan et al., 2004). Hence, plant cell-cycle checkpoints are beginning to look quite different from animal checkpoints and recovery from DNA damage in plants must invoke an as yet unresolved pathway that does not require Arath;CDC25.
The Arath;CDC25 protein shows sequence similarity to the Saccharomyces cerevisiae arsenate reductase and it can show arsenate reductase activity when challenged with arsenate. Because Arath;CDC25 can function as an arsenate reductase this should not necessarily exclude it from also being a plant CDC25 phosphatase, since Homo sapiens CDC25A can be suppressed by arsenite (Lehmann and McCabe, 2007) and CDC25B and CDC25C have arsenate reductase activity when challenged by arsenate in vitro (Bhattachargee et al., 2010). Thus CDC25 protein has dual specificity flipping between phosphatase and arsenate reductase activity depending on substrate availability or on regulatory or target proteins. However, the role established for Arath;CDC25 in making plants tolerant to exogenous arsenate (Bleeker et al., 2006; Dhankher et al., 2006) cannot be extended to a general stress-protection response, e.g. enhancing salinity tolerance, since all genotypes used in the current study were equally sensitive to NaCl. Finally, because neither 1 nor 2 mm HU affected the expression of Arath;CDC25 in WT, suggests that this inducer of the DNA replication checkpoint probably affects Arath;CDC25 post-transcriptionally.
In conclusion, T-DNA insertion lines for Arath;CDC25 are hypersensitive to HU, suggesting that CDC25 is a component but not an absolute requirement for recovery from a DNA replication checkpoint pathway. Further evidence for this is provided by the enhanced growth and developmental response of two independent Arath;CDC25 over-expressing lines compared with WT when challenged with HU. Thus it appears that Arath;CDC25's functionally redundant role is restricted to the DNA replication checkpoint since all genotypes were equally sensitive to zeocin and NaCl.
Supplementary data are available online at www.aob.oxfordjournals.org and consist of a figure showing the epidermal cell length along the primary root meristem and the nuclear DNA content in the 5th leaf of the rosette for the WT, Arath;CDC25oe 18 and arath;cdc25 T-DNA insertion line SALK.
We thank Cardiff, Worcester and Calabria Universities (MIUR-ex 60 % to M.B.B.) and the John Innes Centre for funding this work.