Deliverables
Many cell cycle genes are expressed at very low levels and it was unclear at the start of the project whether the sensitivity of microarrays would be sufficient for detecting modulations in expression of low-abundant cell cycle genes. The results that have been obtained here demonstrate the suitability of this method and data on cell cycle gene expression in different organs and in four cell cycle mutant lines have been generated. The current array is focused on cell cycle genes, which has the advantage that results can be analysed without the need for extensive bioinformatics support and at relative low cost. Complementary to this cell cycle array, a genome-wide array is accessible to the consortium (Affymetrix, Gruissem's group).
In the course of the project, Meyer's group has optimised individual steps of microarray preparation and sample labelling, which allows the collection of statistically sound data and has improved the sensitivity of the microarray assay using a linear amplification method. Inflorescence tissue and suspension culture cells were identified as suitable material for microarray analysis of cell cycle genes.
Meyer's group used inflorescence RNA to characterise the expression profile of cell cycle genes in 8 homozygous insertion lines, which were obtained from CropDesign. These 8 lines have T-DNA insertions in E2F5, KRP6, FKH2, KRP1, KRP3, S6b, DPa, S6kinase respectively. It is noteworthy that HAP genes are often upregulated in KRP altered TDNA lines; HAPs are transcription factors of the NF-Y family known to control several important processes such as cell cycle (animals), embryo development and flowering in plants. This finding corroborates results obtained in the animal field. It seems that for only 1 line (Dp1a) the decrease of target gene expression is visible on the macroarray. The other lines could be upregulation lines (as suggested by the increased expression of the target genes on the macro).
Bergounioux's group has submitted the CDC45 RNAi construct to Meyer's group to monitor the expression of selected cell cycle genes in its plants. Results indicate that 5 of the 12 clones that were down-regulated are cyclin-dependent kinase inhibitors (KRPs). Other clones that were down-regulated include CDC6 of the prereplication complex, DP1a and E2F-like factors of transcription involved in activating genes of the G1/S transition, and a wee1 homolog, a negative regulator of entry into mitosis. Genes that were upregulated include HAP transcription factors (homologs of yeast transcription factors that bind the CCAAT box in many eukaryotic promoters and that exists in multiple forms in Arabidopsis [Edwards et al., 1998]) and genes involved in the G1/S transition or S-phase, such as CDC2 or histone H4, respectively (Stevens et al., 2002).
In addition Meyer's group has performed some analysis of salt stressed arabidopsis cell cultures using the cell cycle macroarray. This study pointed out some new and interesting data.
S6ribosomal 2 and histone H4 are upregulated in two different situations of salt stress. This corroborates earlier data by Mizoguchi et al (1995) on the induction of S6 kinases after cold and salt stress treatment indicating that protein synthesis is a way to adapt plants to different stresses (need for new proteins as a stress response and an adaptation mechanism). By adapting the protein translation machinery the cell can modulate the transcripts that are preferentially expressed. Translational control would thus permit a rapid switch from cell division to stress defence, even before new transcripts are produced. Along the same line, it has also been shown that the over-expression of the sugar beet eIF1A specifically increased the sodium and lithium salt tolerance of yeast (Rausell et al., 2003). In addition, KRP1, Fbox, cyclin C2 and various CDKs were down regulated after salt treatment with 250mM salt whereas Y2H GT-1 transcription factor, cyclin D3, CDC20 are down-regulated after mild treatment. Several cyclins, CDKs are down regulated after both mild and strong stress treatment which is in line with a model predicting a switch in transcription from cell division genes to stress defence genes. The down-regulation of KRP1, a CDK inhibitor seems contradictory at first sight. Yet KRPs have a function during cell division as a modulator of CDK activity. When CDKs are not expressed, there is no role for KRPs either.
Bergounioux's group has studied the regulation of the plant pre-replication complex. This is important for understanding both G1/S transition as endoreduplication. The screen of the tagged mutant library of INRA did not give any knockout mutants for the genes of the pre-replication complex. Therefore, the strategy was changed to the construction and analysis of anti-sense, over-expressing and RNAi plants for cdc6-1, cdc6-2, cdc45, MCM2 and MCM3.
For cdc6-1 and cdc6-2 over-expression, second and third generation plants have been analysed. Cdc6-1 over-expression did not induce endoreplication suggesting that, opposite to S. pombe but, similarly to S. cerevisiae and other eukaryotes re-initiation of the replication should require simultaneous down-regulation of several pathways.
Cdc45 is required for the initiation of DNA replication in yeast, cell proliferation in mammals and functions as a DNA polymerase alpha loading factor in Xenopus. Arabidopsis cdc45 is up-regulated at the G1/S transition and in young meiotic flower buds. Arabidopsis cdc45 RNAi lines are partially to completely sterile. The severity of the phenotype is correlated to levels of the cdc45 transcript and small RNA fragments. Severe chromosome fragmentation arising during meiosis leads to abnormal chromosome segregation and unequal distribution of meiotic products resulting in defective pollen and ovule development. Microarray analysis as been done with Meyer's group on 35S::RNAiCDC45 plants. Meyer's group reported an increased expression of HAP5A, 2A, 4A and CDC2a and b as well as down regulation of cdc6, and E2F1 (E2Fa).
The commitment to DNA replication is a key step in cell division control. Consistent with its role at the G1/S transition the AtMCM3 gene is transcriptionally regulated at S phase. The 5' region of this gene contains several E2F consensus-binding sites, two of which match the human consensus closely.
Furthermore the promoter is activated by AtE2F-a and AtDP-a factors in transient expression studies while mutating either E2F binding site decreases the activity of the promoter. One of the E2F binding sites is shown to be responsible for the G2-specific repression of the promoter in synchronised cell suspension cultures. The second E2F binding site has a role in meristem-specific expression in planta as deletion of this site eliminates the expression of a reporter gene in root and apical meristems. Thus two highly similar E2F binding sites in the promoter of the MCM3 gene are responsible for different cell cycle regulation or developmental expression patterns depending on the cellular environment (Stevens et al., 2002).
Bergounioux's group showed that MCM2 or MCM3 anti-sense constructs did not give any phenotypes. CropDesign showed that plant over-pressing At MCM2 or 3 had reduced growth.
The SCF complex is a protein degradation machinery that consists of at least three different types of proteins, being SKP1, cullin and F-box protein. In contrast to yeast that encodes single SKP1 and cullin/CDC53 SCF subunits and only a few F-box proteins, the Arabidopsis genome contains 19 SKP1, 6 CULLIN and several hundreds of potential F-box protein genes. This result was totally unexpected and a first and major objective of this study is therefore to identify those members of the SKP1, cullin and F-box protein families that truly have a function in cell cycle control.
Dudits' group has searched for those F-box genes that are expressed in a cell cycle-dependent fashion, as this is indicative of a role in the cell division process. For this purpose, construction of an Arabidopsis F-box gene array was initiated. Based on the alignment of all Arabidopsis F-box proteins and by similarity searches, further 22 F-box cDNAs were cloned. To generate probes for array hybridisation, an Arabidopsis (ecotype Ler) cell culture (obtained from L.Bogre) was synchronised successfully by aphidicolin block-release. RNA samples prepared from cells in different cell cycle phases were then converted to labelled cDNA via reverse transcription.
A second approach followed by Dudits' group aimed at investigating the substrates of F-box proteins, as it is anticipated that F-box proteins with a role in cell cycle control would have known cell cycle proteins as substrates. One strategy to identify such substrates is through 2-hybrid screens in yeast. It should be noted however that substrate proteins of F-box proteins are post-translationally modified before being degraded by these F-box proteins. Since it is unclear whether interaction between substrate and F-box protein occurs without such modifications and since it is unknown whether these modifications would occur in yeast, applying a 2-hybrid screen for identifying potential substrates of F-box proteins is an approach of which the success is unpredictable. So far, this approach did not yield any known cell cycle proteins as interactors. Yet, the 2-hybrid screen was experimentally validated by the fact that one of the identified interactors was an SKP1 protein, a known partner of F-box proteins in the SCF complex.
An alternative strategy for looking at substrates of F-box proteins has also been initiated. This approach should be more reliable than the yeast 2-hybrid screen as it looks for protein interactions directly in plants. Yet, it is at the same time more laborious and technically more demanding. Wild type and truncated (F-box deleted) versions of F-box proteins were fused to HA or Myc epitopes and introduced into Arabidopsis cells via Agrobacterium-mediated transformation. Expression of epitope-tagged proteins was analysed by western blotting. From transformed cell lines protein extracts were prepared and after a prefractionation step complexes were purified by immunoaffinity chromatography. Purified complexes were fractionated further via two-dimensional gel electrophoresis and subunits are currently being identified by MALDI-TOF spectrometry. At the same time, these constructs have been transformed into transgenic plants, in order to investigate the phenotypic effects of over-expression of specific F-box proteins on overall plant development.
The interaction of FB2 with Arabidopsis D-type cyclins 3.2; 4.1 and 5.1 suggests that FB2-SCF complexes might have a role in the regulation of CDK kinase activity by controlling cyclin stability. The interaction between FB10 and two different RING proteins is confirmed. The C terminal CUE domain, thought to bind Ub-conjugating enzymes is indeed required for binding. Its HRD1-RING-CUE domain structure is similar to that of human gp78 tumor autocrine motility factor receptor that is involved in ER associated degradation of proteins confirming the role of the RING finger candidate genes in protein degradation.
A collaboration was started between Murray's group and the lab of P Genschick in Strasbourg, France to study one particular RING finger involved in the SCF complex.
Gruissem's group has focused his work on a better understanding of cellular and molecular mechanisms leading to plant cell differentiation. Organogenesis is the result of fine tuned balance of cell proliferation and cell differentiation. Amongst the processes of transcriptional control that link cell cycle and development, the ECCO project has focused on two key molecules in this integration, being the Retinoblastoma-related protein RB and its interacting partner MSI. In addition to interactions with RB, MSI forms a complex together with FAS1 and FAS2 that has been identified as the chromatin assembly factor CAF-1. While RB, FAS1 and FAS2 are single copy genes in Arabidopsis; there are five AtMSI genes.
Gruissem's group has identified T-DNA insertion mutants in most of them In addition, Gruissem's group has constructed transgenic lines over-expressing cDNAs of RB and AtMSI1 in both sense and antisense orientation as well as RNAi-constructs under the control of different promoters. Analysis of these mutants at the molecular level and for phenotypic alterations has been carried out. Also, several Affymetrix GeneChip experiments have been performed. Results indicate that deficiency in AtMSI1 affects preferentially genes of certain functional classes - notably cell cycle and DNA repair as well as response to pathogens and formation of cell wall and cytoskeleton.
WD40 repeat proteins similar to yeast MSI1 are conserved in animals and plants, in which they participate in complexes involved in chromatin metabolism. Although MSI1-like proteins are well-characterised biochemically, their function in the development of multicellular eukaryotes is not well understood. Gruissem's group has constructed Arabidopsis plants in which the AtMSI1 protein level was altered. Strong ectopic expression of AtMSI1 produced no visible altered phenotype, but reduction of AtMSI1 dramatically affected development. The primary shoot apical meristem was unable to develop organs after the transition to flowering. Flowers that developed on floral shoots from axillary meristems experienced a progressive loss of floral morphology, including a reduction in size of the petals and stamens and the development of carpel-like sepals. Ovule development was disrupted in all flowers, resulting in complete female sterility. Molecular analysis of the mutant plants revealed that AtMSI1 is required to maintain the correct temporal and organ-specific expression of homeotic genes, including AGAMOUS and APETALA2.
In contrast, Gruissem's group could show that FAS1 and FAS2, which together with AtMSI1 form the chromatin assembly complex CAF-1, are not required for repression of these genes. Therefore, AtMSI1 has specific functions in addition to CAF-1-mediated chromatin assembly. Efficient formation of heterochromatin, but not methylation of centromeric DNA repeats, depends on AtMSI1 presence demonstrating a key role of AtMSI1 in maintenance of chromatin structure (Hennig et al., 2003).
In addition, seed development in angiosperms initiates after double fertilization, leading to the formation of a diploid embryo and a triploid endosperm. The active repression of precocious initiation of certain aspects of seed development in the absence of fertilization requires the Polycomb group proteins MEDEA (MEA), FERTILIZATION-INDEPENDENT ENDOSPERM (FIE) and FERTILIZATION-INDEPENDENT SEED2. Gruissem's has shown that the Arabidopsis WD-40 domain protein MSI1 is present together with MEA and FIE in a 600kDa complex and interacts directly with FIE. Mutant plants heterozygous for msi1 show a seed abortion ratio of 50% with seeds aborting when the mutant allele is maternally inherited, irrespective of a paternal wild type or mutant MSI1 allele. Furthermore, msi1 mutant gametophytes initiate endosperm development in the absence of fertilization at a high penetrance. After pollination, only the egg cell becomes fertilized, the central cell starts dividing prior to fertilization, resulting in the formation of seeds containing embryos surrounded by diploid endosperm. Gruissem's group concluded that MSI1 has an essential function in the correct initiation and progression of seed development (Kohler et al., 2003).
Inze's group was one of the most productive groups within the ECCO project. Together with CropDesign they have conducted an important gene discovery task. Instrumental for the genomic analysis of cell cycle control in Arabidopsis was the identification and classification of the core cell cycle constituents (Vandepoele et al., 2002). The highlight of the Gene Discovery research was however the genome-wide analysis of cell cycle phase specific transcripts. Using cDNA-AFLP technology, 1340 different transcripts have been identified that show differential expression during the plant cell cycle (Breyne et al., 2002). In parallel to this transcript-profiling project in tobacco BY-2 cell cultures, significant progress was made in the synchronization of Arabidopsis cell cultures. This has resulted in the successful deployment of Arabidopsis microarrays for the study of phase-specific transcripts during the cell cycle. As a spin-off from these projects, Inze's and Dudits's groups have jointly initiated a similar cDNA-AFLP approach for the study of cell cycle phase specific transcripts in rice.
Several proteins have been identified as key control points for cell cycle progression. Seven members of the KRP family of cell cycle inhibitor proteins have been characterized (CropDesigna anf Inze's group). KRPs bind and inactivate cyclin-dependent kinases (CDKs), which are considered to be the central regulators of cell cycle progression. Each of these KRP protein members is able to partially block cell division, leading to plants with much less but larger cells (De Veylder et al., 2001). Interestingly, knockout mutants in single KRP proteins did not show any apparent phenotype indicating functional redundancy within this protein family.
Activation of cyclin-dependent kinases by cyclins releases Rb and thus leads to activation of the E2F/DP complex. Over-expression studies have now revealed the dramatic effects of E2F/DP deregulation on cell division activity and overall plant development, thus confirming the key role of this complex in the cell cycle (De Veylder et al., 2002). To gain a better insight into the phenotypic behaviour of E2Fa-DPa transgenic plants and to identify E2Fa-DPa target genes, a transcriptomic microarray analysis was performed (Vlieghe et al., 2003). Out of 4,390 unique genes, a total of 188 had a twofold or more up- (84) or down-regulated (104) expression level in E2Fa-DPa transgenic plants compared to wild-type lines. Detailed promoter analysis allowed the identification of novel E2Fa-DPa target genes, mainly involved in DNA replication. Secondarily induced genes encoded proteins involved in cell wall biosynthesis, transcription and signal transduction or had an unknown function.
A large number of metabolic genes were modified as well, among which, surprisingly, many genes were involved in nitrate assimilation. Data suggest that the growth arrest observed upon E2Fa-DPa over-expression results at least partly from a nitrogen drain to the nucleotide synthesis pathway, causing decreased synthesis of other nitrogen compounds, such as amino acids and storage proteins. While over-expression of E2F or DP alone had only minor effects on plant development, increasing the expression levels of both proteins simultaneously caused a reiteration of G1/S transitions and the production of either very small or very large cells. This dichotomy in cell size was attributed to either the presence or absence of a mitosis-inducing factor. When present, G1/S reiteration would be ensued by mitosis, leading to enhanced rates of cell division and thus to more but smaller cells. In the absence of such mitosis-inducing factor, G1/S reiteration would cause consecutive rounds of DNA replication without cell division and thus result in large, endoreduplicating cells. The combination of both in a single plant clearly disrupted normal plant development, as illustrated by the fact that all plants over-expressing both E2F and DP exhibited an irreversible premature arrest in development. This mitosis¿inducing factor was subsequently identified as being a special type of CDK, named CDKB1;1, that is unique to plants (Boudolf et al., 2004).
With respect to the control of the G2/M transition, the picture is still more fragmentary. CDKs and cyclins clearly also control this phase transition, but the nature of the CDKs and cyclins is different for G1/S and G2/M. Degradation of mitotic cyclins, such as B2-type cyclins, will block G2/M transition and is essential for the onset of endoreduplication, a process of consecutive DNA replication rounds without intermittent mitosis. Interestingly, also KRP proteins appear to play a role in the control of endoreduplication and a simulation model has been proposed to explain how KRPs can arrest mitosis while G1/S transition is still maintained (Verkest et al., 2005).
Inze's group has published more that 13 papers in international peer reviewed journals.
Analysis of endoreduplication was focused mostly on the role of the ccs52 protein family and its interactors. CCS52 is an activator of the anaphase promoting complex APC in all organisms including plants. CCS52 triggers the destruction of M phase specific cyclins, thus pushes cells into endoreduplication. Several genes (E2F/DP like proteins, ORCc, MCMs) that play a role in G1/S transition were investigated for their role in endoreduplication. Inze's group has shown that co-expression of E2Fa/Dp1a in transgenic Arabidopsis enhances endoreduplication levels in certain tissues. This important finding sheds light on the relationship between G1/S transition and endoreduplication and the control mechanisms that regulate both processes.
In synchronised cell cultures of Medicago truncatula expression of ccs52A was constitutive in the cell cycle, in contrast to that of ccs52B, which was observed only in G2-M. The protein level of ccs52B correlated with the transcript levels. For ccs52A, a decrease in the protein level was observed in the S-G2 phase. As a prototype for the ccs52 proteins, different point and deletion mutations were generated in the Mt/MsCcs52A protein. The effect of the mutations was studied in fission yeast where over-expression of the wild type protein elicited growth arrest, cell enlargement and endoreduplication. This analysis revealed that deletion of the two, conserved N-terminal oligopeptide motifs and the C-terminal IR residues are essential for the biological activity. In contrast, elimination of the CDK phosphorylation sites rendered the protein constitutively active.
In Arabidopsis, GUS expression patterns of the ccs52A and ccs52B genes was only partially overlapping, most often the two genes exhibited complementary activities (most strikingly during flower development). The expression was linked to development of certain cell types (e.g.trichome, hairs), cell layers (e.g. tapetum/A/, pollen tubes /B/), tissues (e.g.vascular tissues, meristems) or organs (e.g. primordia, different parts of the flower), indicating non-redundant functions of the ccs52A and ccs52B genes. Since no ccs52 mutants could be found in the mutant collection of INRA, France, CropDesign has made RNAi constructs for the ccs52A and ccs52B genes. Various overall growth phenotypes have been observed. Plants over-expressing the Arabidopsis CCS52A1 gene under the control of the strong constitutive promoter 35S showed smaller leaves, decreased flower and seed production; Over-expression of CCS52A1 under the control of a mild constitutive promoter led to much more positive phenotypes such as rounded leaves, increased stem diameter, altered trichome architecture, bigger seeds. Additional effects have been found at the microscopic level on cell size, cell number, and endoreduplication. Several contracts have been made by CropDesign with the Arabidopsis CCS52 genes using different promoters (p35S; pUBI; p2S2) and the different phenotypes have been compared. Over-expression under the control of p35S of the 3 Arabidopsis CCS52 genes conducted to the similar drastic growth reduction, accumulation of anthocyanins in the leaves and strong reduction of fertility and seed production.
The yeast two-hybrid screens were performed with both the ccs52A and B proteins from Arabidopsis. In the case of both the Arabidopsis, the strongest interaction was found with a CCA3 homologous protein. In addition, several new, predominantly Destruction box-containing interacting partners were identified including a putative homeodomain transcription factor, as well as a known one (ATHB-14), an unknown TRAF-homologue protein, a hypothetical, carotenoid biosynthesis regulator homologue, an autocrine motility factor receptor-homologue as well as an EREBP4-like protein. Several CCS52 interactors constructs have been produced by CropDesign. One known gene (Athb14) is involved in leaf symmetry and dorsoventrality, which could explain the effects of over-expression of CCS52 on leaf development.
The anaphase-promoting complex (APC/C) is an essential ubiquitin protein ligase that regulates mitotic progression and exit by controlling the stability of cell cycle regulatory proteins, such as the mitotic cyclins. In plants, the function, regulation, and substrates of the APC/C are poorly understood. The plant APC2 gene is able to partially complement a budding yeast apc2 ts mutant. By yeast two-hybrid assays, interaction of APC2 with two other APC/C subunits: APC11 and APC8/CDC23 has been shown and corroborates the function of this subunit in the APC complex. A reverse-genetic approach identified Arabidopsis plants carrying T-DNA insertions in the APC2 gene. Apc2 null mutants are impaired in female megagametogenesis. The APC2 gene is expressed in various plant organs and does not seem to be cell cycle regulated. This work suggests a conserved function of the APC/C in plants but a different mode of regulation. (Capron et al., 2003).
Gutierrez's group has focused his studies within the ECCO project on DEL cell cycle genes. E2F and DP form dimeric protein complexes with different functions in the cell cycle. One of the major functions of E2F/DP complexes described in animals is the control of G1/S transition. Plants also contain E2F and DP-like proteins. Based on their sequence, they cannot be unequivocally categorized as being either E2F or DP, since they share sequence similarity to both types of animal proteins. There is increasing evidence now for the existence of at least two different types of E2F. The first type is canonical E2Fs, which activate G1/S transition and DNA replication. E2Fa and E2Fb are of this type. The second group may actually function as repressors. They are also called DEL (DP-E2F-Like) proteins, because they are structurally quite divergent from the classical E2Fs. E2Fc is also an atypical E2F, as revealed by the studies of Gutierrez's group.
To investigate the role of some E2F/Dp proteins in different plant organs, Partner 3 has made transgenic plants in which the promoter sequences of E2F/Dp genes are fused to the GUS reporter gene to compare gene expression profiles. AtE2F2 (Acc.No.AF242581, E2Fc) is highly expressed in meristems and cotyledons in 1.5 day-old seedlings and in apical meristems and vascular tissues in 5 day-old seedlings. E2Fc is expressed in the basal part of developing leaves, roughly coincident with cell division activity in trichomes and in early stages of flower development.
At E2Ff (At3g01330; AtE2F4, AtELP1) is highly expressed in young cotyledons and leaves, hypocotyls and roots (but not in meristems), moderately expressed in flowers and not detectable in siliques (Ramirez-Parra et al., 2004). Arabidopsis E2Ff TDNA mutant (SALK Institute, USA) showed a strong growth reduction whereas 35S::E2Ff arabidopsis plants showed no phenotype. The authors concluded that growth is compatible with an altered expression of E2Ff. There was no clear function of E2Ff in cell proliferation. Rather it was concluded that E2Ff would act as a repressor in differentiated cells and could repress E2F target genes previously activated by another E2F from the activator class (E2Fa). Interestingly it was found that E2Ff binds to the promoter if several genes involved in cell wall modelling (Ramirez-Parra et al., 2004).
Juergens' group focused on the map-based cloning of genes that are deficient in cytokinesis. A total of 38 putative cytokinesis mutants that have been analysed. Complementation tests define 6 different genes. Two of the genes, KNOLLE and KEULE, were previously isolated by map-based cloning. Mapping populations have been established for 4 additional genes.
The KIS gene and 4 PILZ group genes have been isolated by map-based cloning and shown to encode proteins required for microtubule formation and cell division but not for actin filaments and cell growth (Steinborn et al., 2002). Specifically, KIS encodes tubulin-folding cofactor A. The mutant kis has a strong trichome phenotype indicating a defect in proper cell division. The HIK gene has also been isolated by map-based cloning and shown to encode a plant-specific cell cycle-regulated kinesin motor protein involved in phragmoplast microtubule dynamics (Strompen et al., 2002).
The RUK gene has been mapped to a 250 kb interval in the top arm of chromosome 5. Fine mapping of RUK was impeded by low recombination frequency within the relevant interval. New mapping populations have been generated for two different alleles with the aim to identify more relevant recombinants in a different genetic background. In addition, cosmid and transformation-competent BAC clones from the critical genomic interval have been transformed into ruk/ RUK plants and will be further analysed for rescuing activity. The RUK gene encodes a ca. 150kDa protein that appears to be plant-specific and, except for an N-terminal putative kinase domain, has no distinctive domain features. Experiments have been initiated (i) to localise a myc-tagged version of the protein subcellularly and (ii) to examine whether cell cycle-regulated expression of the RUK protein rescues ruk mutant embryos.
To identify a KNOLLE-interacting synaptobrevin (Syb), representatives of 4 subgroups of Syb coding sequences have been tested for their ability to interact with KNOLLE in a GST-fusion pull-down assay. One candidate, AtSyb4, interacts strongly with KNOLLE in the presence of a KNOLLE-interacting t-SNARE, the SNAP25 homologue SNAP33, which has been shown to co-localise with KNOLLE during cell-plate formation.
Recombinant AtSyb4 protein has been used to generate rabbit polyclonal antiserum, which, however, needs to be purified for immuno-localisation studies. Screening of a T-DNA insertion library has yielded syb4 candidate lines, which will be further characterised genetically and phenotypically. Screening of a T-DNA insertion library has yielded two syb4 candidate lines. In both lines, the insertions are poorly transmitted, suggesting that SYB4 is required for gametophyte development. The mutants are being characterised genetically and phenotypically in more detail. The mutant line is gametophyte lethal.
Traas' group has focused his work on the understanding of the role of cell cycle genes on meristem organisation and plant growth. A first requirement for studying the effects of cell cycle genes on meristem function is the development of a reproducible method for studying cell division dynamics in the shoot apical meristem. A method was developed that allows following individual cells in the living shoot apex. Plants are germinated on NPA, an inhibitor of auxin transport. These plants form naked inflorescence stems, without any primordia. When transferred to a medium without inhibitor, these naked meristems will subsequently form primordia.
These regenerating meristems can be directly stained with vital dyes and viewed in the confocal microscope. Using this technique, Traas's group was able to follow cell divisions throughout the shoot apex during several days (Grandjean et al., 2004). This method, combined with green fluorescent protein marker lines and vital stains, allows us to follow the dynamics of cell proliferation, cell expansion, and cell differentiation at the shoot apex. Using primordium promoters (ANT and LFY) driving GFP expression, the recruitment of meristematic cells in the incipient flower primordia was followed. This suggested that cells preferentially activated the reporter genes just after cell division, i.e. while they are in early G1. Using this approach, the effects of several mitotic drugs on meristem development were studied. Oryzalin (depolymerising microtubules and blocking the cells in G2/M) very rapidly caused cell division arrest. Nevertheless, both cell expansion and cell differentiation proceeded in the treated meristems.
Interestingly, DNA synthesis was not blocked, and the meristematic cells went through several rounds of endoreduplication in the presence of the drug. We next treated the meristems with two inhibitors of DNA synthesis, aphidicolin and hydroxyurea. In this case, cell growth and, later, cell differentiation was inhibited, suggesting an important role for DNA synthesis in growth and patterning. In addition, cells at the periphery of the meristem and in the young primordia expanded much faster than those at the meristem centre. This showed that differential cell expansion rates and cell differentiation do not necessarily depend on the cell cycle. Traas's group has been also working on a model in which auxin gradient drives the expression of certain genes such as Aintegumenta, and PIN1 as well as certain cell division events (use of cell cycle blockers).
Traas's group has also concentrated on the analysis of the expression of 7 CDK inhibitors, called CKI 1-7 (or KRP1-7), using in situ hybridisation in order to determine the localisation of CKI 1 to 7mRNAs in the Shoot Apical Meristem. From the 7 genes studied, P8 showed that CKI 2 and CKI 3mRNAs accumulate in the SAM at different developmental stages. CKI 2mRNA accumulates in young primordia at the vegetative stage and in the SAM at the inflorescence stage. CKI 3mRNA accumulates in the central cylinder of mature embryos and under the rib zone in vegetative and inflorescence SAMs. The other CKIs were below the detection level. At this stage, we do not intend to continue the in situ hybridisation studies.
A total of 25 T-DNA insertion mutants have been identified in cell cycle genes in collaboration with CropDesign. Subtile phenotype has been observed on petal cell size in KRP mutants. In addition, another cell cycle mutant in E2F5 gene was shown to have a leaf phenotype. The analysis of these mutants by Meyer's group has already identified target genes that could be related with the observed phenotypic defects.
The struwwelpeter (swp) mutant in Arabidopsis shows reduced cell numbers in all aerial organs (Autran et al., 2002). In certain cases, this defect is partially compensated by an increase in final cell size. Although the mutation does not affect cell cycle duration in the young primordia, it does influence the window of cell proliferation, as cell number is reduced during the very early stages of primordium initiation and a precocious arrest of cell proliferation occurs. In addition, the mutation also perturbs the shoot apical meristem (SAM), which becomes gradually disorganized. SWP encodes a protein with similarities to subunits of the Mediator complex, required for RNA polymerase II recruitment at target promoters in response to specific activators. To gain further insight into its function, we over-expressed the gene under the control of a constitutive promoter. This interfered again with the moment of cell cycle arrest in the young leaf. Our results suggest that the levels of SWP, besides their role in pattern formation at the meristem, play an important role in defining the duration of cell proliferation.
The ECCO project has helped CropDesign, the SME partner in this project, in building a highly competitive position in this field of research. Cell cycle technologies are important assets for a company that targets yield enhancement and yield stability in crop plants as major commercial objectives. During the project, 174 putative cell cycle genes have been identified and assembled in a database. For 77 of these, a loss-of-function mutant screen has been performed. 25 mutants were identified and 22 cell cycle genes were inactivated through an RNAi approach. Phenotypic analysis of those mutant lines has been conducted and only subtitle phenotypes could be observed. Mutant lines have been sent to the ECCO partners for a deeper analysis.
Over-expression of cell cycle gene was done for 80 different genes. Phenotypic analysis indicated strong negative phenotypes and also more positive effects of the constructs tested. Again, several lines have been sent to different ECCO partners for deeper analysis. For example a thorough analysis has been done concerning the genes CCS52 in arabidopsis in collaboration with Kondorosi's team. Altogether, this has led to a much better understanding of the cellular functions of specific cell cycle components. In addition, 4 patent applications were filed that were based on results from the ECCO consortium.
Currently, these patent applications are being pursued internationally and applications of these cell cycle technologies are being pursued in different important crops.
In Arabidopsis, the D-type cyclin CYCD3 is rate-limiting for transition of the G(1)/S boundary, and is transcriptionally unregulated at this point in cells re-entering the cell cycle in response to plant hormones and sucrose. However, little is known about the regulation of plant cell-cycle regulators at the protein level. We show here that CYCD3;1 is a phosphoprotein highly regulated at the level of protein abundance, whereas another D-type cyclin CYCD2;1 is not. The level of CYCD3;1 protein falls rapidly on sucrose depletion, correlated with the arrest of cells in G(1) phase, suggesting a rapid turnover of CYCD3;1. Treatment of exponentially growing cells with the protein synthesis inhibitor cycloheximide (CHX) confirms that CYCD3;1 is normally a highly unstable protein, with a half-life of approximately 7 min on CHX treatment. In both sucrose-starved and exponentially growing cells, CYCD3;1 protein abundance increases in response to treatment with MG132 (carbobenzoxyl-leucinyl-leucinyl-leucinal), a reversible proteasome inhibitor, but not in response to the cysteine protease inhibitor E-64 or the calpain inhibitor ALLN (N-acetyl-leucyl-leucyl-norleucinal).
The increase on MG132 treatment is because of de novo protein synthesis coupled with the blocking of CYCD3;1 degradation. Longer MG132 treatment leads to C-terminal cleavage of CYCD3;1, accumulation of a hyper-phosphorylated form and its subsequent disappearance. It can be concluded that CYCD3;1 is a highly unstable protein whose proteolysis is mediated by a proteasome-dependent pathway, and whose levels are highly dependent on the rate of CYCD3;1 protein synthesis. (Planchais et al., 2004; Differential stability of Arabidopsis D-type cyclins: CYCD3;1 is a highly unstable protein degraded by a proteasome-dependent mechanism (Planchais et al., Plant J. 2004 May; 38(4):616-25.)
In addition, tools have been developed for studying the role of eIF4E/ S6/ S6kinase signaling pathwayin cell cultures and intact plants. This was done in collaboration with CropDesign, Heberle-Bors's group and Murray's groups. CropDesign has found T-DNA insertion mutants in S6K, S6II, and CBP. Murray's group has over-expressed eIF4E and eIF4E2(iso) under control of the 35S promoter. T2 seeds of the 35S: eIF4E are available. Interestingly, leaves and siliques of these plants were purple. T0 seeds of 35S: eIF4E2(iso) did not germinate suggesting that over-expression of this gene perturbs embryogenesis and/or germination processes. Murray's group has made tobacco BY-2 cells, in which eIF4E1 and eIF4E2(iso) can be over-expressed in an inducible manner.
This construct induced aberrant phenotype in tobacco with abnormal cell morphology and more elongated cells. In Arabidopsis, plant lines have been produced with EiF4E1 and 2iso but not analysed yet. CropDesign's data on S6kinase indicate also negative phenotypes in arabidopsis.
Heberle-Bors' group has studied a MAPkinase pathway that is involved in cytokinesis. They identified this pathway first in tobacco and are now also isolating the corresponding genes from Arabidopsis. The following genes were isolated:
- NtMEK1: a tobacco MAP kinase, which activates the tobacco MAP kinase Ntf6,
- ATMPK16: a MAP kinase, which is the putative Arabidopsis homologue of Ntf6,
- AtMKK6: a MAP kinase, which is the putative Arabidopsis homologue of NtMEK1.
A transposon insertion line was found for ATMPK16. Screening of selfed plants has been done to identify homozygous lines. Furthermore, GFP-ATMPK16 transgenic lines were produced for visualization of MAPK localization to the phragmoplast. Three GFP-alone transgenics showed GFP expression. Four GFP-ATMPK16 transgenics only show fluorescence at the leaf tip (not seen in WT). GFP-ATMPK16 fusion is functional in yeast assays. The tobacco homologe Ntf6 was also cloned as a GFP fusion (fluorescent after particle bombardment of pollen), the NtMEK1 as a RFP fusion, so that co-localisation of Ntf6 and NtMEK1 can be observed.
A Guanine nucleotide dissociation inhibitor (GDI) was shown to physically interact with Ntf6 and to alter kinase activity of Ntf6 and NtMEK1. This raises the possibility that GDI is an Ntf6 substrate (first substrate of a plant MAPK demonstrated). The Arabidopsis MAP kinase ATMPK13 is the putative orthologue of the tobacco MAP kinase Ntf6. This latter kinase has been implicated in the process of cytokinesis in tobacco cells. Attempts were therefore made to elucidate the role of ATMPK13 in Arabidopsis.
A transgenic plant line with a transposon insertion in the ATMPK13 gene was identified. Although this line displayed an interesting phenotype (viz. the absence of petals), it was shown genetically that this was not due to the transposon insertion in the ATMPK13 gene, and was probably due to an insertion in another gene. Indeed, Southern analysis showed the presence of at least 6 transposons in this particular line. Because of the difficulty of purifying the ATMPK13 insertion from the other transposons, together with the mobile nature of the transposon, work on this line was discontinued. A T-DNA insertion line is presently being analysed.
In an attempt to identify the subcellular localization of ATMPK13, a GFP-ATMPK13 fusion was constructed. GFP fluorescence of the fusion protein and the functionality of the kinase as a fusion protein were demonstrated in yeast using functional assays. Four transgenic GFP-ATMPK13 transformants were obtained, but none of them showed GFP fluorescence (the control lines with GFP alone were strongly fluorescent).
A peptide antibody was raised against the N-terminus of ATMPK13, but the quality of the antibody was poor. In Western analysis it recognised a band only in flowers. By contrast an antibody against the C-terminus (provided by L. Bogre) recognised a band in flowers, roots, stems, and seedlings. The reason for this discrepancy is unclear. (Qualitative) RT-PCR demonstrated the presence of an ATMPK13 transcript in all tissues except cauline leaves and siliques. The MAP kinase MEK1 is an activator of Ntf6, and has been shown to function in cytokinesis. The putative Arabidopsis orthologue of MEK1 is MKK6. An MKK6 cDNA was isolated. Studies in yeast showed that MKK6 can activate ATMPK13 - a yeast strain mutated in the MPK1 MAP kinase is only complemented when MKK6 and ATMPK13 are co-expressed, and higher kinase activity can be immunoprecipitated from the double transforming.
Therefore the MKK6/ATMPK13 pair of kinases show similar behaviour to the tobacco MEK1/Ntf6 pair. Indeed MEK1 can activate ATMPK13 in yeast functional assays. While it has been possible to show that ATMPK13 encodes a functional kinase molecule in yeast, it has proved to be recalcitrant to studies in bacteria and after re-introduction into the plant. Possibly post-translation modifications and/or a particular sub-cellular localization are required for proper ATMPK13 function.
Searching for OpenAIRE data...
There was an error trying to search data from OpenAIRE
No results available