WO2006027310A2 - Modulation of plant cell number - Google Patents

Abstract

The present invention relates to the use of the ANG4 gene, or a variant thereof, to modulate the cell number of a plant organ. Said modulation can be used to increase the plant biomass, or to adapt the plant architecture.

Description

MODULATION OF PLANT CELL NUMBER

The present invention relates to the use of the ANG4 gene, or a variant thereof, to modulate the cell number of a plant organ. Said modulation can be used to increase the plant biomass, or to adapt the plant architecture.

Plant architecture, especially leaf and root morphology, is an important factor in the determination of the plant productivity. Therefore, the study of genes involved in plant architecture and their regulation has drawn a lot of attention by several research groups. The isolation, identification, characterization and manipulation of genes that are candidates for controlling leaf development is a key in understanding how plant leaves are constructed. Several methods have been used to study genes and their functions that regulate leaf development such as forward or reverse genetics. During leaf development processes, there are at least two factors that affect the leaf phenotype, at first cell division, that results in a given cell number, and second is cell expansion, which is required for the establishment of the cell size and shape. The length and width of leaves are regulated by cell division and cell expansion according to a gradient (Pyke et al., 1991; Van Lijsebettens and Clarke, 1998). In addition, the leaves are also modulated by environmental factors such as water, nutrients, light and CO2 concentration. Berna et al. (1999) gives an overview of mutations and phenotypic classes that influence leaf morphology in Arabidopsis. Some of those mutations were characterized on gene level. Genes that regulate cell number along the width axis are DRL1 and SWP1 genes that act mainly on lateral growth of the lamina (Nelissen et al., 2003 and Autran et al 2002).

Although these genes might be used to modulate the plant biomass, there is still a further need for genes controlling plant architecture, especially for genes capable of controlling the cell number in specific plant organs.

In this invention, we studied a mutant with narrow leaves, angusta4, from the seed collection of Berna et al, (1999) and identified the causal gene, which we called ANG4. The mutant was originally created by EMS method (Figure 1A). Molecular analysis surprisingly showed that the causal gene for the angusta4 mutation, which is located on chromosome 2, is a RING finger protein (Anami, 2004; Stone et al., 2005) with E3 ligase activity. This activity is related to protein degradation, but has never been linked to altered leaf morphology. The width and length of angusta4 laminas was compared to wild type (Landsberg erecta) (Figure 1A). The data showed that total length lamina in the angusta4 leaves is significantly reduced compared to Ler. angusta 4 had narrow first leaf and shorter petioles than Landsberg. The epidermal and palisade cell area in the angusta4 (11mm2) is smaller than in wild type (19mm2) as well. Even more surprisingly, we found that that the phenotype of the leaf is due to a drastic reduction in the number of palisade cells. Moreover, we found that the same mutation has a dramatical effect on root growth too, making the gene an interesting tool for biomass modulation. A first aspect of the invention is the use of a gene encoding a protein comprising SEQ ID N°2 (TAIR_At2g44950, Figure 10), or a functional fragment or variant thereof, to modulate the cell number of a plant organ, or a part thereof. Preferably, said gene is encoding a protein consisting of Seq ID N°2. Gene as used here, refers to the coding sequence, which may be linked to its own promoter, but is preferably operably linked to a promoter which is not its own. Said promoter can be any promoter suitable for expression in plants. Preferably, said promoter is a strong promoter, such as, but not limited to the 35S promoter. "Gene" refers both to the genomic sequence (including possible introns) as well as to the cDNA derived from the spliced messenger, operably linked to a promoter sequence. Coding sequence is a nucleotide sequence, which is transcribed into mRNA and/or translated into a polypeptide when placed under the control of appropriate regulatory sequences. The boundaries of the coding sequence are determined by a translation start codon at the δ'-terminus and a translation stop codon at the 3"-terminus. A coding sequence can include, but is not limited to mRNA, cDNA, recombinant nucleotide sequences or genomic DNA, while introns may be present as well under certain circumstances.

Operably linked refers to a juxtaposition wherein the components so described are in a relationship permitting them to function in their intended manner. A promoter sequence "operably linked" to a coding sequence is ligated in such a way that expression of the coding sequence is achieved under conditions compatible with the promoter sequence. A variant as used here is a plant gene comprising a ring finger, with a homology with SEQ ID N° 2 of at least 25% identities and/or 45% positives, preferably at least 35% identifies and/or 55% positives, more preferably at least 45% identities and/or 65% positives, even more preferably at least 55% identities and/or 75% positives, most preferably at least 65% identities and/or 85% positives, as measured by a protein-protein Blast search. Preferably, said variant has E3 ubiquitin protein ligase activity. Preferred variants are the Oryza sativa ANG4 homologues CAD41603 and NP922769, as listed in figure 10. Plant organs, as used here, comprise roots, stem, leaves and flowers. Preferably, said plant organ is a plant leaf and/or a plant root. Parts of a plant organ are, as a non-limiting example, the palisade cells of the leaves, or the lateral roots. One preferred embodiment is the use according to the invention, whereby the modulation of the cell number is used to modulate the leaf morphology. A functional fragment, as used here, is any fragment that still has the E3 ubiquitin-protein ligase activity.

Still another aspect of the invention is the use of a gene encoding a protein comprising, preferably consisting of SEQ ID N°2, or a functional fragment or variant thereof, to modulate the root length. Preferably said gene, variant of functional fragment is overexpressed and said modulation is an increase in root length. Preferably, said gene comprises SEQ ID N°1 (genbank NM_130060).

Another aspect of the invention is the use of a gene encoding a protein comprising, preferably consisting of SEQ ID N°2, or a functional fragment or variant thereof, to increase biomass. Preferably, said increase of biomass is obtained by an overexpression of said gene, variant or functional fragment.

BRIEF DESCRIPTION OF THE FIGURES

Fig. 1 : Leaf phenotype of angusta4 and wild type. A: In vivo condition, fully grown rosettes of

Ler and angusta4. B: Juvenile and adult fully expanded leaves of Ler. C: Juvenile and adult fully expanded leaves of angusta4.

Fig. 2: In vitro leaf phenotype of wild type and angusta4 mutant plant (26days after germination)

Fig. 3: Leaf phenotype of angusta4

A: Transversal sections at the widest location of expanded lamina of first leaves of wild type(

Ler)-(top) and angusta4 (bottom). B: Mean value of palisade cell number in first leaves of angusta4 and the wild type (asterisk indicates statically significant difference). C: Cross section of wild type at the midvein (Adaxial surface is up). D: Close up of wild type vascular tissue. E:

Cross section of angusta4 at the midvein (Adaxial surface is up). F: Close up of angusta4 vascular tissue. (V: vascular bundle; P: palisade cell ; x: xylem; ph: phloem ; I : inter cellular space). G, H, I: Morphological data of expanded leaves of ang4-1 mutant (A, B, C). Bars represent mean values and standard deviation. *** means statistical difference at p<0.001 from the t test. Histological observations of expanded leaves of ang4 mutant and Ler

Fig. 4: A: In vitro root growth after 15 days of germination.

B: Biomass of ang4-1, ang4-2, ang4-3 mutants, Ler and CoI in soil conditions. Average of 2 assays, 4 blocks per assay and 8 plants per block. The bars correspond to the standard deviation.

Fig. 5: A: Root growth kinetics of angusta4 in comparison with wild type (Ler). B and C:

Longitudinal sections by confocal microscopy through the root apical meristem in the root tip in wild type and the angusta4 mutants. ).

Fig. 6: Fine mapping of the ANG4 gene: The Figure indicates the map-based cloning strategy where a set of eight AFLP primer combinations was applied to 20 F2 individual mutants that indicated that the ANG4 mutation was located on chromosome 2 (blue and white pattern).

Further application of AFLP marker SM33_202 and SM26_495 narrowed the ANG4 interval to

293 kb. Finally, InDeI and SNP markers were used on a total of 1062 recombinants that delimited the ANG4 area to 27 kb at the bottom of chromosome 2 flanked by SNP markers CER458218 and CER458367. The 27 kb region contained 4 genes, one of which was the ANG4 candidate gene.

Fig. 7: Separation of PCR products of At2g44960 gene following amplification with two primer sets. Lane 1 contains a 1 kb molecular weight marker. Lane 2-7 contains PCR products from Ler At2g44960 gene while lane 8-13 contains PCR products from ANG4 At2g44960 gene. PCR products of lane 2, 3, 4, 8, 9, and 10 were amplified with primer combinations Defle 12 and Defle 13 while PCR products of lane 5, 6, 7, 11, 12 and 13 were amplified with primer combination: Defle 14 and Defle 15.

Fig. 8: Example of an alignment performed by CLUSTALW 1.8. This alignment is between 2652bp and 3873bp part of the At2g44950 gene that was amplified by 5OTCGCCCATTGTTGTTTCAG3' and 5'AATTGCGGAAACCATGTTCC 3' primer combination. It clearly demonstrates the point mutation induced by EMS as a C was changed to a T generating a stop codon UAG.

Fig. 9: ANG4 gene structure. Shown are the ANG4 candidate genes covering a 27 kb region on chromosome 2 and linked by CER458218 and CER458367 SNP markers. The unspliced mRNA of ANG4 has 19 exons and 18 introns covering a region of 6298 bp while the full length cDNA covers a region of 2637 bp. EMS mutagen ization caused a C to change to a T generating a stop codon at the end of exon 16 hence truncating the protein from 878 amino acids to 844 amino acids (Exons in blue boxes, and introns in orange boxes). Figure drawn to scale; for the candidate genes structure, 1 cm = 2 kb and for the unspliced mRNA and the spliced mRNA, 1 cm = 1 kb

Fig. 10: An alignment of the ANG4 homologues in different species. The orange underlined sequence indicates the conserved RING finger motif. Conserved cystein and histidine residues are colored with red and blue colors respectively. At2g44950 is ANG4 sequence with 878 amino acid residues. At1g55250 is the ANG4 homologue on chromosome 1 in Arabidopsis with 899 amino acids. NP_55586 and AAK58539 are ANG4 homologues in human genome with 1001 and 975 amino acids respectively. CAD41603 and NP922769 are the ANG4 homologues in Oryza sativa with 883 and 789 amino acids respectively. Figure 11 : RT-PCR analysis of ANG4 gene expression in different Ler organs. The expression pattern was visualized on acrylamide gel. 4 μl samples were loaded an acrylamide gel in 1X Tris-Boric acid- EDTA buffer and electrophoresed at 3000V. Primers Defle 44 and syana_01 were labeled with P³³.Numbers on the gel indicate different Ler organs as follows: 1- Ler apex, 2- Ler shoot apex, 3- Ler roots, 4- Ler cotyledons, 5- Ler young leaves, 6-Ler Expanded leaves, 7- Ler flowers and 8- water as a control sample.

Fig. 12: Summary of Arabidopsis genes with altered mRNA expression in ang4 and two other leaf development mutants, elo2 and drl1-2. RNA was extracted from shoot apex of young plants and expression measured using ATH 1 microarrays (Affymetrix) method in triplicates. Comparisons of expression level were done between each mutant and the wild-type Ler following the Bayesian test of linear model performed with Bioconductor programs. Values without parenthesis are the number of DE genes equally expressed in different mutants, and values in parenthesis, the number of DE genes up-regulated in one mutant and down- regulated in another.

Fig. 13: Kinematic analysis of leaf growth of the first leaf pair of the wild-type Ler and the ang4-1 mutant. (A) leaf blade area, (B) epidermal cell number on the abaxial side of the leaf, (C) relative leaf expansion rate, (D) average cell division rates of the epidermal cells on the abaxial side of the leaf, (E) average epidermal cell size on the abaxial side of the leaf, (F) stomatal index on the abaxial side of the leaf. Error bars correspond to the standard deviation (n=5).

Fig. 14: Flow cytometry analysis of nuclear DNA content of the Ler (A) and ang4-1 mutant (B). Fig. 15: Ler wild type and OE-ANG4 (T1) plants two weeks after transfer to soil.

EXAMPLES

Materials and methods according to the invention

Plant material and growth conditions

Seeds of the Arabidopsis thaliana (L.) Heynh. Landsberg erecta (Ler) and the ang4-2 mutant (SALK_122512) were obtained from the Nottingham Arabidopsis Stock Centre. The ang4-1 homozygous mutant was provided by J. L. Micol (Universidad Miguel Hernandez, Alicante, Spain) (Berna et al., 1999). The T-DNA insertion line ang4-3 (GABI_276D08) was supplied from GABI-Kat.

Plants of the wild-type Landsberg erecta (Ler) and angusta4 (ang4) were grown in in vitro conditions with following conditions: 16/8 hrs (d/n) with white light (Neon tubes, cool white), 100 μEm-²h-1 PAR and 20 °C. The medium was 2.15 g/l MS salts (micro and macro elements), 1 g/l sucrose, 0.5 g/l MES, pH 6.0, 6 g/l plant tissue culture agar. Seeds were sowed in 150 x 25 mm round dishes, sealed with Urgopore tape. Sixty seeds were sowed per plate. The vernalization period was 3 days after sowing.

For the root growth experiment, one lane of 5 plants were sowed in square plate in vertical position. The homozygous ang4-2 and ang4-Z lines were selected in in vitro medium containing kanamycin 25 mg/l for the ang4-2 or sulfadiazine 11.25 mg/l. for the ang4-3 line. The phenotype of the T-DNA insertion lines was scored in soil growth conditions.

Standard leaf analysis

Eight to twelve expanded first and third leaves of 30-days-old and 40-days-old Ler and ang4 in vitro grown plants have been harvested, treated with 100% methanol O/N, cleared with 90% lactic acid for 2-3 days O/N and put on a slide for image analysis. Petiole, lamina and leaf length, lamina width and area of first and third leaves have been measured with the Scion Image software (version β-3b; Scion Corp., Frederick, MD) from digital pictures directly taken from binocular observations.

The statistical significance of the mean differences (p≤0.05) was analyzed by t-test using the SPSS (Statistical Package for the Social Sciences, version 10.0.5, SPSS, Inc.; Chicago, IL) software on normally distributed data.

Root growth kinetics

15 seeds of each angusta4 line was sown out (made only or one row per plates) in the square plates with GM medium contain vitamin. The plates were oriented in a vertical position. By using scalpel, roots of these lines were marked every two days until 14 days.

Differential Interference Contrast (DIC) optic analysis

The cleared first and third Ler and ang4 leaves prepared for the imaging analysis have been used to perform DIC (Differential Interference Contrast) optics analysis. This technique allows counting the number of cells of a determinate histological tissue layer and most importantly measuring the cell area from the adaxial side using a Scion Image.

Leaf histology: determination of Palisade Cell Number (PCN)

26 day-old fully expanded first and third leaves of Ler and angusta4 plants were harvested and immediately fixed in FAA (90% EtOH, 5% acetic acid, 5% formaldehyde) at 4°C overnight. The process of dehydratation was done by increasing concentrations of EtOH as followed: 2 x 30 min EtOH 50%, 2h EtOH 50%, 2 h EtOH 70%, 2h EtOH 80%, O/N EtOH 80%, 2 x 2h EtOH 90% and ultimately O/N EtOH 95%. Tissue infiltration was realized in a gradually permeation of Historesin and was achieved by first putting the leaves for 4h in a mix of 50% EtOH and 50% Historesin, followed by another mix of 30% EtOH and 70% Historesin for 4h and finally in 100% Historesin for 4h. During that time, the samples were always kept for 30 min in vacuum. During the last step, the leaves were shacking at room temperature for 3 days. The leaves were then immerged in a new basic resin solution containing a 1% temperature-sensitive activator and left shaking ON. Leaves were finally oriented in beds which were half-filled with the resin solution, covered with new resin and left polymerizing at 45°C for at least 2 h. The histology analysis has been performed on 5μm sections collected on glass slides by using a Reichert Jung Ultracut Microtome using homemade glass knives. The Historesin leaf- containing blocks obtained after polymerization were oriented on a plastic cube and fixed with super-glue. The plastic cubes were holder by the micro tube climb.

Cytoplasm were stained in each sections by toluidin blue following the process below: The treated glass slides were stained for 8 min in 0.05% Teledyne blue and 0.1 M phosphate buffer, pH 6.8 for 10 minutes. After two washes (5 tolOminutes each) in sterilize water, the slides were dried and mounted with DePex. Photographs were taken by using an Olympus CAMEDIA C-3040 digital camera zoom 3.3 mega pixel at the same magnification and pictures image were performed by Adobe Photoshop 6.0 program.

5 μm transversal sections of 28-day-old Ler and ang4 first and third full-expanded leaves have been made with a Reichert Jung Ultracut microtome in order to determine with the aid of a binocular microscope the Palisade Cell Number (PCN) present at the widest part of the lamina. This parameter is an indicator of leaf blade lateral growth (Tsuge et al., 1996). Several leaves have been entirely sectioned from tip to petiole: one section every ten has been collected and put on a glass slide. The glass slides were subsequently stained with toluidine blue and mounted with DePex.

Confocal microscopy observations of root meristem

7 days old seedling of angusta4 were stained with 100 ng/ml propidium iodide solution for 3 minutes and washed 3 times by sterilized water. Stained root were observed under a MRC600 Biorad confocal microscope using 543nm excitation 560 LB light.

Flow cytometry

The flow cytometry analysis was performed as described by De Veylder et al. (2001). The first two leaves were chopped with a razor blade in 300 μl of buffer (45 mM MgCI₂, 3OmM sodium citrate, 2OmM 3-[Λ/-morpholino]propanesulphonic acid, pH 7, and 1 % Triton X-100) (Galbraight et al., 1991). To the supernatant, which was filtered over a 30-μm mesh, 1 μl of 4,6-diamidino-2-phenylindole (DAPI) from a stock of 1 mg/mL was added. The nuclei were analyzed with the BRYTE HS flow cytometer, using Win-Bryte software (Bio-Rad, Hercules, CA). Of each time point, two biological and three technical replicates were taken.

Leaf growth kinematic analysis

Leaf growth was analyzed kinematically from 5 to 28 days after sowing as described (De

Veylder et al., 2001). The wild-type and ang4-1 plants were germinated and grown in in vitro conditions in GM+V medium. The following parameters were determined: total area of all cells in the drawing, total number of cells, and number of guard cells. From these data, we calculated the average cell area and estimated the total number of cells per leaf by dividing the leaf area by the average cell area (averaged between the apical and basal positions). Finally, average cell division rates for the whole leaf were determined as the slope of the log₂- transformed number of cells per leaf, which was done using five-point differentiation formulas

(Erickson, 1976).

Map based cloning procedure The DNA extraction, AFLP, insertion/deletion (InDeI) and single-nucleotide polymorphism (SNP) analysis were done according to (Peters et al., 2004) and (Cnops et al., 2004). A standard set of 8 AFLP markers were analyzed on 20 F₂ mutants and identified the mutation in a 493 kb interval on chromosome 2. The fine-mapping of the ANG4 locus was done using the InDeI and SNP markers described in Table Sl Recombinants were used for fine-mapping and delineated the locus to 97 and 27 kb regions flanked by SNP markers. The last interval covered a 27 kb region between CER458218 and CER458367 SNP markers and contained 4 genes that were sequenced.

The candidate genes identified in the last mapping interval were amplified from DNA and cDNA, and fully sequenced in at least 3 replicates to identify the base exchange in the ang4-1 mutant compared to Ler.

Microarrays experiment

The studied organs, shoot apex of plants (comprising shoot apex meristem, first and second rosette leaf primordia at petiole-less stage), were harvested removing the cotyledons and the hypocotyls. The harvesting was done in the laboratory conditions under additional light and 20°C between 11 am and 6 pm. The age of the plants at the harvesting step were between 8 and 15 days after germination depending of the delay of mutant development. RNA were extracted with TRIzol Reagent (Life Technologies, Breda, The Netherlands). The experimental design comprised 3 replicates of Ler and ang4, one replicate corresponding to one RNA extraction and about 150 apexis.

Microarrays experiment was done by the VIB Microarrays Facility lab (Paul van Hummelen, Leuven, Belgium; http://www. microarrays. be/) using ATH1 Affymetrix chips of 23,800 probes sets for Arabidopsis thaliana. The raw data were normalized and summarized using Robust Multi-Array average method from affy package of Bioconductor statistical R programs (Wu and Irizarry, 2004). The genes were ranking in order of evidence for differential expression DE between mutant and wild type using an empirical Bayes method performed with the limma package of Bioconductor. This method consists to combine at the gene level with means and standard deviation from the 3 replicates to form a statistic B which is a Bayes log posterior log- odds that each gene is DE (Lonnstedt and Speed, 2002; Smyth et al. 2003). The p value calculating from B data was corrected by Holm's method and the cut-off value of p was 0.01.

Alleles characterization

The ang4-2 and ang4-Z mutants with T-DNA insertion respectively in the exon 6 and the exon 19 of ANG4 gene were studied (http://www.arabidopsis.org). The T-DNA insertion was checked by PCR on F2 plants using primers designed before (P1) and after (P3) the putative position of the T-DNA and a primer specific of the left border of the T-DNA (P2). A positive amplification between P1 and P2 validates the position of the T-DNA insertion. A coincident positive or negative amplification using P1 and P3 shows that the line is respectively heterozygous or homozygous.

Over-expression construct and plant transformation

To obtain overexpression lines of ANG4 the open reading frame (including ATG and stop codon) of ANG4 (2637 bp) was amplified by Pfu polymerase and cloned into the pDONRT221 vector using the GATEWAY recombination strategy (Invitrogen) to obtain ENTRY clones. The ENTRY clone was recombined with the pK7WG2 vector (Karimi et al., 2002) to obtain a DESTINATION vector with the ORF under the control of a 35S promotor. This construct was introduced into Agrobacterium tumefaciens and subsequently Ler or ang4-1 plants were transformed with the Agrobacterium tumefaciens suspension through floral dip. The T₀ seeds were grown in high density on growth medium containing Kanamycin (50 μg/ml), Nystatin (50 μg/ml) and Carbenicillin (250 μg/ml) to select the transformants. These T₁ transformants were transferred to soil to obtain T₂ seeds.

Example 1: Histological analysis of the Ang4 mutation

We performed an anatomical analysis in the first leaf by using light microscopy to identify phenotypic functions of the ANGUSTA 4 gene. Our interest was to focus on the number of palisade cells, structure of vascular tissue in the leaf as well as primary root development of mutant plants in comparison with that of wild type.

We looked at the anatomy of angusta4 leaves to determine whether cell division or cell expansion was affected and to check polarity and studied root growth kinetic as a measure for root apical meristem activity in the mutants. In plant, cell expansion and cell division are key parameter in the determination of organ shape.

Major trait of the angusta class of mutants is narrow leaf lamina (Berna et al., 1999; Figure 1). The reduced leaf size in ang4-1 mutant was confirmed by morphological measurements of expanded leaves. The measurements showed a significant decrease of lamina length and width, petiole length and total lamina and petiole length (Figure 3G). The lamina area of ang4- 1 first and second expanded leaves was 10.7 ± 2.4 mm² i.e. 55 % of Ler lamina area which was 19.3 ± 2.5 mm² (Figure 3H). The length/width ratio of the lamina was significantly increased in ang4-1 mutant showing a modification of the lamina proportions and to a narrower shape (Figure 3I). The fresh weight of the rosette leaves at the flower emergence stage of development were significantly reduced in ang4 mutants: ang4-λ biomass was 40 % of Ler biomass, and ang4-2 and ang4-3 fresh weight were respectively 51 % and 55 % compared to CoI (Figure 4B). The dry weight was also strongly affected by the mutation in ANG4 with 39 % for the ang4-1 plants compared to Ler, and respectively 45 % and 49 % for the ang4-2 and ang4-3 plants compared to CoI.

Serial sections through historesin embedded expanded first leaves (26 day old seedlings) were taken (Figure 3A). The number of palisade cells was counted in a number of serial sections at the widest width of the leaf to be used as a measure for lateral growth of the leaf lamina (Tsuge et al 1996). The number of palisade cells of angusta4 was smaller than wild type. The data showed that the number of palisade cells is 30 in the angusta4, and about 66 cells in Ler (Figure 3B). Thus, palisade cells were reduced by about 50% in angusta4 compared to wild type. The structure of palisade cell was larger and distributed more irregularly than in Landsberg (Figure 3C & E).

The vascular tissue of Ler wild type and angusta4 mutants was also visualized under the microscope. The polarity was correct in the mutant: xylem at the dorsal side and phloem at the ventral side. The midvein of wild type and mutant are shown (Figure 3D - F). In the angusta4 mutants, cells surrounding xylem and phloem were bigger than in Ler. The number of cells is also higher in the vascular bundle in the angusta4 midvein (Figure 3 E and F). These data show that the ANGUSTA4 gene is involved in the regulation of cell number during leaf growth; it has no function in leaf polarity.

To investigate in more detail the function of the ANGUSTA4 gene, primary root growth was analysed. 60 seedlings of angusta4 and Ler were germinated in the square plates and kept in vertical position in the tissue culture room. The root tip was marked every 2 days with a scalpel blade. The mean value was calculated for each time point. A graphical representation of these mean values is shown in Figure 5A. After 15 days, the length of angusta4 reached 1 cm, which is much shorter than the 5 cm of the Ler line. In addition, angusta4 roots started to form adventitious roots after four days germination; each angusta4 plant had 2 to 3 adventitious roots.

Moreover, apical sections from in planta Arabidopsis roots (7 days old seedling and n=20) were visualized under confocal microscope to investigate the structure of the root apical meristem. Figure 5B and C showed the meristem zone of the primaty root in angusta4 and wild type. Longitudinal section of root meristem region of angusta4 showed no difference in cell division and cell expansion. It indicates no defective root meristem activity. The flower organization is also altered by ANG4 mutation. The floral diagrams of ang4-1 showed an asymmetric position of the petals and missing anther or carpel. The flower of ang4- 2 and ang4-3 plants was not modified but the inflorescence stem appeared thinner as compared to CoI. To verify if ang4 mutation only affected aerial organs, the root growth rate was analysed and compared to CoI alleles and wild types. The root growth was strongly decreased in ang4-1 plants compared to the wild type Ler. However, the root growth of ang4-2 and ang4-3 was similar to that of the wild-type CoI suggesting that the mutation of ANG4 gene does not alter the root growth in the genetic background of CoI.

Thus, the ANG4 gene has a function in leaf and flower development and root growth.

Example 2: mapping of ANG4 leaf form mutation

The mutant, ang4, was obtained from the collection of 255 mutant lines induced by EMS mutagenesis (Berna et al., 1999). The aim of this work was to verify the ANG4 region delimited by AFLP, InDeI and SNPs markers and by recombinant analysis. The Ler mutant was crossed with CoI-O wild type and the resulting FVs were allowed to self in order to produce F2 mapping populations (Robles and Micol., 2001). 320 F2 mutants together with their Ler and CoI-O parents were analyzed using a standard set of eight AFLP primer combinations shown in Table 1 in order to visualize 85 AFLP markers on the genome (Peters et al., 2004). After scoring the resulting 85 AFLP markers, linkage to chromosome 2 and non-linkage to other chromosomes was observed. Table 2 shows the genotypic scoring that was done using AFLP, InDeI and SNP markers. Presence of the AFLP marker signifies that the marker behaves as the CoI parent and is represented in Table 2 as number 1. For the F2 individuals this means that the marker is either homozygous or heterozygous. Absence of the AFLP marker indicates that the marker is homozygous Ler and it is indicated as number zero (0) in Table 2.

Initially, as shown in Table 2, F3 recombinants 670, 227, and 1389 were scored as homozygous mutants (100%ar/g4) while recombinants 635,1472, 1747 and 387, 1607,1716 were scored as heterozygote (1 ang4: 3 wild type) and homozygote (100% Ler) respectively. During meiosis, for recombinant 1747, a cross-over event took place between markers CER458218 and CER442324. This recombinant was used to delimit ANG4 mutation from the top of chromosome 2 and hence marker CER442324 was taken as the top marker that limited the ANG4 interval. In contrast, a cross over event occurred between markers CER458218 and CER458219 for recombinant 670 and 227, markers CER442324 and CER458218 for recombinant 1389 and markers CER442323 and CER458367 for recombinants1472. All these markers delimited the ANG4 mutation from the bottom of chromosome 2. Delimiting the ANG4 region became rather difficult because the mutant was phenotypically very clear in the Ler background and less clear after crossing (i.e., it was difficult to score the phenotype in the F2 derived F3 populations).

In order to verify the ANG4 interval of 27 kb, and probably reduce this region to about 10 to 20 kb, phenotypic scores of the F3 of nine recombinants that were not very informative in the previous scoring were repeated. In vitro, thirty seeds of each recombinant were planted on GM medium in 150x25mm Petri dishes in replicate. 200 seeds of each recombinant were planted in vivo on trays containing 52 wells in which one seed was planted in each well. Phenotypic scores were done at four time points over a period of 4 weeks to determine whether the F3 was homozygous mutant (100% ang4), heterozygous (1 ang4: 3wild type) or homozygous wild type (100% wild type Ler) and these scores are summarized in Table 3, and compared to the previous less extensive scoring. Recombinants 635, 670, and 1389 were scored differently compared to previous scoring.

Earlier phenotypic scores had shown that recombinant 635 was heterozygote while recombinants 670, 1385, and 1472 were homozygote mutant. Table 3, which indicates new phenotypic scoring, revised these earlier scores. Recombinant 227 was very difficult to score in the second round of phenotype scoring, as it was less clear in vitro. Scores from in vivo growth conditions indicated that it was homozygous mutant.

For instance, recombinant 670 was scored as a homozygous mutant before and from Table 3, it was scored as heterozygous (1 mutant: 3 wild type). It was therefore decided that a number of recombinants that were not clearly scored and therefore not very informative, including recombinants 387,670,1389,1607 and 1716 would be ignored and that recombinants that were clearly scored as shown in Table 4 will be used to delimit the ANG4 mutation. As indicated in Table 4, the SNP marker that delimited ANG4 mutation from the top of chromosome 2 was CER458218 based on recombinant 227 while marker CER458367 delimited ANG4 mutation from the bottom of chromosome 2 based on recombinant 1472. These markers are within a 27 kb (26,647 mb) region. This region was the minimal region delimited by markers while the maximal ANG4 region was between CER458219 as the top marker based on recombinants 377and 1775 and CER458367 as the bottom marker based on recombinant 1472. Recombinant lines that were most informative were those with Ler scoring because CoI-O is a recombinant inbreed line (RIL) and as such any cross over event in it does not necessarily indicate linkage to the mutation of interest as shown in Table 4 for recombinant 635.

Verification of the ANG4 mutation after the phenotypic scores showed that, indeed the ANG4 mutation was within the 27 kb region delimited by genotypic scoring as indicated in Table 2 and Figure 6. Within this 27 kb region; there are 4 intact genes one of which has to be ANG4 gene.

The ANG4 interval was determined at 27 kb and flanked by CER458218 and CER458367 markers. This was based on the recombinant analysis of 1062 F2 plants. We checked the phenotypic region of the remaining recombinants in the F3 generation both in vivo and in vitro at 4 time points over a period of 4 weeks. The ANG4 region was determined and allowed to deduce the F2 genotypes. This F2 genotypic information was integrated in Table 4 and the ANG4 interval delineated to a 27 kb region containing 4 intact genes one of which has to be ANG4 gene.

Example 3: Sequencing of candidate ANG4 genes

Four candidate genes are situated in the 27 kb interval delimited by the recombinant analysis and are listed in Table 5 with their respective functions. For each gene, the genomic DNA was amplified from the ang4 mutant and compared to the wild type Ler in order to determine the single base change.

Total genomic DNA from ang4 mutant and Ler were extracted using the CTAB method and DNeasy Plant mini kit. Ler DNA acted as a control. For ang4 mutant, four candidate genes were amplified by performing three independent PCR for each of the primer combinations (Table 9). The same primer combinations were used to three independent PCR to amplify the four genes present in the genomic DNA of Ler. Primer pairs were designed for all the candidate genes that amplified overlapping segments of 800 bp-1200 bp spanning the entire 27 kb region (Figure 6; Table 9). Three independent PCR reactions of these segments were sequenced. An example of this PCR amplification is shown in Figure 7 where each band indicates DNA amplified with two primer sets. Sequence alignment was performed by CLUSTALW 1.8 software and compared with that of the wild type plant Ler. An example of sequence alignment is shown in Figure 8 with the gene At2g44950. Sequencing of these fragments and comparison with the wild type Ler sequence identified a mis-sense change in the candidate gene At2g44950 generating a stop codon UAG instead of the CAG codon corresponding to amino acid glutamine in the predicted exon 16 (Figure 9). Sequence alignment of other candidate genes, At2g44940, At2g44970 and At2g44980 genes did not show any mutation.

The At2g44950 gene is within the 27 kb region on chromosome 2 together with At2g44940, At2g44970 and At2g44980 genes flanked with CER458218 marker from the top of chromosome 2 and CER458367 marker from the bottom of chromosome 2 as shown in Figure 9. Amongst these candidate genes, ANG4 is the largest covering a region of 6298 bp with an open reading frame (ORF) of 5245 bp; while the At2g44940, At2g44970 and At2g44980 genes covers 1157 bp with an ORF of 887 bp, 3337 bp with an ORF of 3020 bp and 4230 bp with the same number of base pairs as its ORF respectively. ANG4 gene has two untranslated regions, one at the 5' end covering a region of 344 bp and the other at the 3' end with 307 bp. It consists of 19 exons and 18 introns. Once the introns have been spliced, the exons form the full length cDNA that consists of 2637 bp and this is translated in a protein of 878 amino acids Tg )- The mutation that was found in the At2g44950 gene (Figure 8), truncated the protein from 878 to 844 amino acids. This was as a result of the stop codon, UAG that was created at position 5183 in the unspliced mRNA and at position 2134 in the spliced mRNA when cytosine nucleotide changed to a tyrosine nucleotide that was caused by an EMS mutagenesis.

The At2g44950 gene has a RING-finger motif that begins with the amino acid cystein at position 826 in the amino acid sequence and ends with amino acid cystein at position 864 (CKACNDR-PKEWITKCYHLFCNPCVQK-LTGTRQKKCPTC) as shown in Figure 10. 18 amino acids of the RING finger motif are part of the 844 amino acids that makes a protein after the mutation and 23 amino acids of the RING finger motif are lost (Figure 9). This means that the RING finger motif that functions as part of the E3 ligase was inactivated in the ang4 mutant and that this might have lead to defect in the degradation of a number of proteins in the proteasome.

Molecular cloning of ANG4 demonstrates that map-based cloning using AFLP markers is a reliable strategy for accessing genes from the genome of Arabidopsis thaliana. Cloning of ANG4 will facilitate studies on its function for crop improvement.

Example 4: ANG4 homologues and functional domain

Data base searches revealed the presence of At2g44950 homologues as uncharacterized cDNA or open reading frames obtained from genome projects in a number of organisms including Arabidopsis thaliana, humans, and rice. ANG4 has a close homologue in Arabidopsis thaliana located on chromosome 1 (At1g55250). Sequence comparison analysis indicates that NP_055586 is the human orthologue of the Arabidopsis ANG4.The human genome also contains a second ANG4 homologue, AAK58539 (RING finger protein 20), which is encoded by a gene that is distinct from the NP_055586 gene (RING finger protein 40). In Oryza sativa (japonica cultivar-group), there appears to be two ANG4 homologues with accession numbers CAD41603 and NP922769. Figure 10 shows an alignment of the amino acids of ang4 mutant and its homologues in humans, Arabidopsis and rice which revealed a conserved Really Interesting New Gene motif (RING finger) at the end of the sequences indicating that ANG4 is an evolutionary conserved protein. The RING finger domain has been classified into 20 different subgroups in Arabidopsis thaliana (Stone et al., 2005). In this sub groups, ANG4 was classified as having an ATP binding domain. We searched for this ATP binding domain (the P- loop) using Prosite (www.expasy.org/cgi-bin/prosite/S and was not found though the ANG4 homology to ATPases involved in chromosome segregation and cell division was found. Search for other functional motifs was done but no other functional domain was found besides the RING finger. Example 5: alleles in ANG4

A number of alleles for At2g44950 gene with T-DNA insertions are available from Signal (http://signal.salk.edu/cgi-bin/tdnaexpress?GENE=at2g44950&FUNCTION=&TDNA=) and GABI ((http://www.mpiz--koeIn.mpg.cle/GAB!-Kat/clb/search.php?type===seq&term===60-K015154-- 022-276-D08-8409) collections. T-DNA insertion lines are also available for the ANG4 homologue in Arabidopsis (At1g55250) (Table 6 A and B).

Example 6: ANG4 expression patterns in different Arabidopsis organs

To examine the expression pattern of ANG4 gene in Ler organs, we performed semi¬ quantitative reverse transcriptase (RT)-PCR analysis with different tissues including shoot apex, flower, young leaves, expanded leaves, cotyledon and roots. RNA was extracted from different frozen ground Ler organs using TRIZOL reagent (Life Technologies, Paisley, UK) according to manufacturer's protocol. The cDNA samples were standardized on actin transcript (At3g 18780) amount using primers Defle 44 and Defle 45 with the following sequences: TGCTGGACGTGACCTTACTG as a forward primer and GGGCTGGAACAAGACTTCTG as a reverse primer. The melting temperature for these standard primers that acted as control in this experiment was 59°C for both. For ANG4 gene, the following gene specific primers were used: syana_01 as a forward primer and syana_02 as a reverse primer with the following sequences: TG CTCG AATCAG ATG GAAGA and AGCTAGCTGACCGCACAAAT respectively. The melting temperature for syana_01 was 59°C while for syana_02 was 60⁰C. Actin is a fundamental component of the cytoskeleton in all eukarγotes and directs the spatial organization of many crucial sub cellular processes. Hightower and Meagher (1986) proposed that the six subclasses of actin have been conserved during vascular plant evolution and hence it can be used as a reference for expression analysis of other plant genes. Figure 11 shows the result of a typical RT-PCR analysis of the expression pattern of ANG4 in different Ler organs. Primers Defle 44 and Defle 45 amplified a single 253 bp actin PCR product while primers syana_01 and syana_02 amplified a predicted single 164 bp ANGA PCR product. This analysis shows that the ANG4 gene is expressed in all organs studied. The expression pattern of ANG4 gene in all Ler organs studied could indicate that it may play a basic role in all these organs. The understanding of whether ANG4 gene may be involved in other possible roles, it would be important to investigate its expression levels in response to hormone and stress treatment. In addition, the expression analysis at the cellular level will be analyzed using the GFP marker line. Expression of At2g44950 gene in all organs means that it is required for fundamental or basic processes in all plant organs and throughout the life cycle. Cellular experimental analysis would also indicate whether ANG4 gene function is related to cell division processes. Example 7: Genome wide expression in ang4 shoot apex

A total of 1821 genes were differentially expressed (DE) in the apex of young plants of ang4 compared to Ler, that represents 8 % of the Arabidopsis genome. Comparing these results with those obtained in other narrow leaf mutants (elo2 and drl1-2 involved in independent process to ang4), 1314 genes appeared differentially expressed specifically in ang4 (Figure 12). Considering the level of expression, 494 genes are DE at a 2-fold change expression threshold. The number of genes regulated by ANG4 is higher than those regulated by the DRL1 and ELP1 genes, respectively mutated in drl1-2 and elo2 mutants, showing the general function of ANG4 in the development.

Most of the genes regulated by ANG4 are involved in cytokinesis and cell cycle. A partial list of the DE genes in ang4 shows that 24 cell cycle genes and 27 microtubule and myosin related genes, are regulated in ang4 mutant (Table 7). Among these, one finds 8 genes related to E2F-DP complex regulating the G1 to S transition in plants (De Veylder et al., 2003). Eight A- and B-type cyclins genes and 3 B-type cyclin-dependent kinase genes involved in G2 to M transition in cell cycle are down-regulated in ang4 genotype. Kinesins represent a super-family of microtubule motor proteins involved in the transport of vesicles and organelles, spindle formation and elongation, chromosome segregation, microtubule dynamics and morphogenesis (Reddy and Day, 2001). Among the 61 kinesin genes identified in Arabidopsis genome, 19 are down-regulated in ang4, that TETRASPORE involved in the formation of tetrad of microspores after meiosis (Yang et al., 2003). The HINKEL gene, another kinesin, plays a role in the reorganization of phragmoplast microtubules during cell plate formation (Strompen et al., 2002). Other cytokinesis related genes are also DE in ang4, as the cytoskeletal components actin 8, tubulins, myosin like proteins and microtubule-associated proteins. The PLEIADE gene that has a function in the stabilization of cytokinetic structures of cell plate during cytokinesis is also down-regulated in ang4 mutant (Muller et al., 2002). The KNOLLE gene, a cell-cycle-regulated syntaxin involved in membrane fusion in cytokinesis, is also repressed in ang4 (Muller et al., 2003). The SIAMESE gene, required for coordinating cell division and cell differentiation during the development of trichomes and may function as a repressor of mitosis in the endoreduplication cell cycle, is up-regulated in ang4. These results suggest an implication of ANG4 gene in cell cycle regulation.

Some genes related to plant development are also regulated by ANG4 gene expression (Table 8). The GLABRA1 gene is a MYB transcription factor that specify the primary cell fate during development of epidermal hairs in Arabidopsis (Schiefelbein, 2003). The homeobox genes KNAT2 and KNAT6 have a role in meristem initiation and maintenance (Tsiantis and Hay, 2003). The genes NAM and AINTEGUMENTA are known to be involved in organ initiation and separation (Traas and Vernoux, 2002). In Arabidopsis, SCARECROW (SCR) is essential for the asymmetric division of the cortex/endodermis progenitor cell in the root (Kamiya et al 2003). Two genes related to auxins are DE in ang4: a putative ARF1 auxin responsive transcription factor and a putative AUX 1 -like permease, a regulator of root gravitropism (Liscum and Reed, 2002).

Example 8: Effect of ang4 mutation on endoreduplication and cell expansion

The effect of the ang4 mutation on leaf development and cell cycle duration was analyzed by a kinematic analysis on the first leaf pair of in vitro grown plants. Leaf blade area was similar in Lerand the ang4-1 mutant at the earliest observations. However, the increase in leaf area was slower in ang4-1 compared to Ler between 5 and 8 DAS (Figure 13A). At maturity, the leaf blade area of ang4-1 was about 47 % of those of Ler, with respectively 11 and 24 mm². During the same period, the number of cells per leaf also increased quicker in Ler than in ang4-1 (Figure 13B). So, at maturity (after 18 DAS), the ang4-1 leaves contained only 48 % the number of epidermal cells of Ler. Differences in the rate of increase of leaf area and cell number must be due to effects on cell expansion and division respectively. Indeed, between 5 and 10 DAS, the cell division rate and the relative leaf expansion rate (RLE) were lower in ang4-1 compared to Ler, but they decreased more slowly in the mutant (Figure 13 C and D). Consequently, the cell division rate and the RLE rate became similar in ang4-1 and Ler from the 10 DAS and along the expansion phase until no cell was dividing anymore at the 15 DAS. The expansion continued in both Ler and ang4-1 until the 18 DAS when the leaf reached the maturity. So, the ANG4 mutation alters the cell division and the leaf expansion only during the early stage of leaf development.

At this stage, the cells of ang4-1 were bigger than in Ler with an average cell area respectively of 82 μm² and 54 μm² at day 5 (Figure 13E). After 7 DAS, no difference of the cell area could be observed at the later stages between ang4-1 and Ler showing that the balance between division and expansion rates is the same in ang4-1 and in Ler.

Because the final divisions give rise to stomata, the stomata index (Sl) indicates the exit from cell cycle and the end of proliferation activity, which starts from the tip to the base of the leaf in Arabidopsis [De Veylder, 2001]. The SI also increased slower in ang4-1 compared to Ler between 5 and 8 DAS, resulting in the final SI in mature leaves being lower with 0.23 in average for ang4-1 and 0.35 for Ler (Figure 13F). These data validate the previous data showing that the ANG4 mutation decreases the cell division activity at the early stage of leaf growth without modifying the duration of the proliferation and expansion phases. At 5 DAS, the average cell cycle duration, which is the inverse of cell division rate, was almost 50 % longer in ang4-1 (20.6 h) than in Ler (14.1 h), and it was longer until 11 DAS where the cell cycle duration was the same in both genotypes (respectively 48.4 h and 50.7 h for Lerand ar/g4-1). To investigate deeper the effect of ang4 mutation on cell cycle progression, we analyzed wild- type and mutant leaves by means of flow cytometry. The ploidy level of the first leaf pair was determined throughout the development of wild-type and mutant leaves to reveal the changes in relative duration of G1/G2 phase during mitotic cell division and timing and amount of endoredu plication in the ang4-1 mutant.

At 8 DAS when leaves could first be harvested, a shift is seen in the G1-to-G2 populations in the ang4-1 mutant compared to Ler (Figure 14). In the mutant, the population of cells in 4C is similar to that in 2C (ang4-1; 2C = 46.2 %, 4C = 44.0%), while in wild type, the number of cells in 4C is only half of those in 2C (Ler, 2C = 66.2 %, 4C = 33.8%), suggesting that increased cell cycle duration is associated with a block at the G2-to-M transition point of the cell cycle in ang4-1. The exit from mitotis coincided with the start of the endocycles and could be seen by the increase of 4C content and the appearance of higher ploidy levels (8C, 16C). Cell cycle activity ended as evidenced by a stable DNA distribution around 18 DAS, coinciding with the end of growth. The endocycle was enhanced in the ang4-1 mutant from earliest stage with already 10 % of the cells in 8C at 8 DAS, while this level of 8C was only reached at 13 DAS in Ler. The consequence was a higher ploidy levels in ang4-1. In the mature leaves, more than 4 % of the cells contained a ploidy level of 32C in the ang4-1 mutant, while the ploidy level in mature Ler leaves only reached 16C. The exit from the endocycle occurred at the same date for both ang4- 1 and Ler, at 18 DAS. So, when ANG4 is mutated, cells arrest in the G2/M phase of the cell cycle and proceed into endocycles instead. We postulate that the ANG4 protein has a function in the degradation of a cell cycle regulator(s) working at the G2-M transition of the cell cycle during early organ growth.

To confirm that these effects were not specific for the leaves, flow cytometry was done on roots, hypocotyls and first leaves at one time point in development (12 DAS). The ploidy levels obtained for the root and hypocotyls were comparable to those of the first leaves, indicating that ANG4 affects the cell cycle throughout plant development.

The flow cytometry profile of the angl allele, GABI_634H04, differs from that of the CoI control and is similar but weaker to that of ang4: more endopolyploidy (presence of 32C), slight shift in the G1-to-G2 cell populations (reduced 2C cell number and increased 4C cell number). The mutational analysis of the angl allele indicates that ANGL (At1g55250) is also functional and might have functional redundancy with the ANG4 gene (At2g44950).

Example 9: ANG4 overexpression increases leaf size

Photographical observations of ANG4 overexpression plants (T1) clearly indicate that the plants have improved growth performance compared to wild type plants. For example, the rosette leaf size of the overexpression plants are considerably increased as can be seen in Figure 15. TABLES

Table 1. Standard set of eight AFLP primer combinations used to detect linkage between 85 Col/Ler AFLP markers and ANG4 locus. Table obtained from Peters et al., 2004.

Table 2: Genotypic scores of 9 recombinants using AFLP, InDeI and SNP markers and L- indicates co-dominant marker, 1- dominant marker, 0- No marker H- heterozygous. Numbers in top row indicate the F3 individual recombinants. Recombinants indicated in blue was scored as ang4 mutants, green as wild type and turquoise as heterozygote.

Table 3: Phenotypic scores of 9 recombinants: The scores were done at four time points over a period of 4 weeks both in soil and in vitro. In both growth conditions the scores were the same. Heterozygous indicates that the wild type and the mutants were observed while homozygous mutant implies only mutants were observed. Homozygous wild type indicates no mutant was observed in those recombinants.

Table 4. Recombinant used to delimit the ang4 mutation. Only 7 recombinants shown on the top row of the Table from recombinant 227 through recombinant 1747 were selected because they were the most informative recombinants while the other recombinants were ignored. The SNP markers and their position on continuous sequence are indicated indicates heterozygozity after a cross over event during meiosis. L- Ler, and C- CoI ecotypes. The (?) in the Table means that the scoring of the recombinants was not clear.

F3 RECOMBINANTS

Table 5. ANG4 candidate genes: The 4 candidate genes in the 27kb region and their functions based on TAIR annotation. At- Arabidopsis thaliana, g- genomic.

Table 6. Alleles for At2g44950 and At1g55250 genes: A, ANG4 alleles. Two SALK lines, SALKJ 22512 and SALK_044415 from SIGnAL collections and two GABI line, GABI_276D08, and GABI_306H08. B, alleles for the ANG4 homologues in Arabidopsis (At1g55250); SALK_071289 and SAKLJ41948 from SIGnAL collections and GABI_634H04 and GABI_529603 from GABI collections.

A -At2g44950

SALK LINES GABI LINES INSERTION SITE POSITION IN CHROMOSOME 2

SALK 122512 EXON 1 18549684

SALK 044415 INTRON 3 18550469

GABI 306H08 INTRON 2 18550191

GABI 276D08 INTRON 13 18553269

B -Atlg55250

Table 7: Differentially expressed genes in ang4 mutant compared to Ler and related to cell cycle and cytokinesis. Data were performed on microarrays ATH 1 experiment with RNA from shoot apex of young plants grown in in vitro conditions. The p values are calculated according a Bayesian test of linear model and corrected by Holm's method.

P. value probes name fold change Sequence gene descriptions process related expression derived from

1 29E-05 248413_at 021 At5g51600 PLEIADE gene cytokinesis

5 78E-06 258098_at 023 At3g23670 hypothetical protein similar to kinesin like protein cytokinesis

6 79E-06 248057_at 026 At5g55520 putative myosin heavy chain protein cytokinesis

2 18E-06 261660_at 026 At1g18370 HINKEL, kinesin heavy chain isolog cytokinesis

323E-06 264802_at 026 At1g08560 KNOLLE, putative syntaxin-related protein cytokinesis

542E-07 261159_s_at 028 At1g34460 putative cyclin cell cycle

2 89E-05 252691_at 0 30 At3g44050 kinesin -like protein KLP2 protein cytokinesis

1 31E-04 257115_at 030 At3g20150 kinesin-like protein cytokinesis

1 74E-06 247039_at 0 31 At5g67270 putative microtubule-associated protein cytokinesis

2 97E-05 245607_at 0 32 At4g14330 kinesin like protein cytokinesis

340E-06 255265_at 0 32 At4g05190 kinesin like protein A cytokinesis

940E-05 253978_at 0 32 At4g26660 putative kinesin cytokinesis

548E-05 263441_at 0 33 At2g28620 putative kinesin-like spindle protein cytokinesis

1 12E-04 266009_at 0 33 At2g37420 putative kinesin heavy chain cytokinesis

4 74E-05 261780_at 0 36 At1g76310 CYCB2_4 (cychn) cell cycle

2 53E-05 259151_at 0 36 At3g10310 kinesin-like protein similar to carboxy-terminal kinesin 2 cytokinesis

8 84E-06 265349_at 0 36 At2g22610 putative kinesin heavy chain cytokinesis

5 35E-06 257008_at 0 38 At3q14210 Myrosinase-associated protein cell cycle

545E-04 253148_at 0 38 At4g35620 CYCB2_2 (cyclin) cell cycle

241E-05 267618_at 0 38 At2g26760 CYCB1_4 (cychn) cell cycle

3 51E-05 254400_at 040 At4g21270 kinesin-related protein katA cytokinesis

6 86E-05 245739_at 040 At1g44110 CYCA1_1 (cychn) cell cycle

1 80E-06 258573_at 041 At3g04260 BC010 (E2Fb binding protein) cell cycle

2 96E-05 259851_at 042 At1g72250 putative kinesin cytokinesis

1 53E-06 259978_at 043 At1g76540 CDKB2_1 (Cyclin dependent kinase) cell cycle

2 71E-04 266401_s_at 043 At2g38620 CDKB1_2 (Cychn-dependent kinase) cell cycle

3 12E-04 262802_at 045 At1g20930 CDKB2_2 (Cychn-dependent kinase) cell cycle

2 30E-04 257267_at 046 At3g15030 TCP family (E2Fa-DPa induced Transcπption factor) cell cycle

1 69E-05 257524_at 046 At3g01330 DEL3 (E2F-DP-hke protein) cell cycle

3 53E-06 248150_at 046 At5g54670 kinesin-like protein cytokinesis

3 35E-04 262081_at 047 At1g59540 kinesin motor protein (kιn2) cytokinesis

6 18E-05 245259_at 047 At4g14150 kinesin like protein cytokinesis

2 72E-04 261605_at 048 At1g49580 CDPK-related protein kinase cell cycle

1 59E-05 260329_at 049 At1g80370 CYCA2_4 (cychn) cell cycle

5 13E-05 263017_at 049 At2g17620 CYCB2_1 (cychn) cell cycle

4 54E-04 262752_at 049 At1g16330 CYCB3_1 (cychn) cell cycle

2 60E-05 266295_at 049 At2g29550 tubulin beta-7 chain cytokinesis

3 78E-04 261765_at 0 51 At1g15570 CYCA2_3 (cychn) cell cycle

448E-04 252736_at 0 52 At3g43210 TETRASPORE (TES), kinesin-hke protein ZCF125 cytokinesis

7 37E-04 262494_at 0 54 At1g21810 myosin-like protein cytokinesis

843E-04 265464_at 0 54 At2g37080 putative myosin heavy chain cytokinesis

7 93E-04 250386_at 0 55 At5g11510 MYB3R4 (transcπption factor) cell cycle

1 06E-03 264061_at 0 55 At2g27970 CKS2 (CDK binding protein) cell cycle

1 06E-03 246683_at 0 56 At5g33300 putative protein chromokinesin KIF4 cytokinesis

8 76E-03 250685_at 0 56 At5g06670 kinesin heavy chain-like protein cytokinesis

823E-04 261639_at 0 57 At1g50010 putative tubulin alpha-2/alpha-4 chain cytokinesis

3 16E-03 245576_at 0 57 At4g14770 CPP1 -related transcπption factor family (E2Fa-DPa cell cycle induced TF)

4 07E-03 249095_at 1 44 At5g43900 myosin heavy chain MYA2 cytokinesis

1 08E-03 251052_at 1 54 At5g02470 DPA transcπption factor cell cycle

1 45E-04 250923_at 1 69 At5gO3455 GTPV2 (putative CDC25 homolog) cell cycle

4 59E-04 260765_at 1 79 At1g49240 actin 8 cytokinesis

2 13E-04 250844_at 1 97 At5g04470 SIAMESE gene (SIM) cell cycle

4 89E-03 264006_at 2 02 At2g22430 homeodomain (ATHB-6) (E2Fa-DPa induced TF) cell cycle

1 93E-06 253217_at 2 58 At4g34970 actin depolymeπzing factor - like protein cytokinesis

1 17E-03 250666_at 325 At5g07100 WRKY family (E2Fa-DPa induced Transcπption factor) cell cycle

1 65E-09 253890_s_at 3 84 At5g54100 Putative protein contains similarity to stomatin like protein cell cycle Table 8: Differentially expressed genes in ang4 mutant compared to Ler and related to plant development. Data were performed on microarrays ATH 1 experiment with RNA from shoot apex of young plants grown in in vitro conditions. The p values are calculated according a Bayesian test of linear model and corrected by Holm's method.

P. value probes name fold change Sequence gene descriptions expression derived from

3.28E-06 259686_at 0.26 At1g63100 transcription factor SCARECROW development

2.15E-03 257221_at 0.46 At3g27920 GLABRA1 (GL1), MYB family transcription factor development

1.26E-05 265454_at 0.50 At2g46530 putative ARF1 family auxin responsive transcription factor development

1.86E-03 263013_at 1.92 At1g23380 KNAT6, knotted-like homeobox protein development

9.50E-05 260334_at 1.96 At1g70510 KNAT2, homeotic protein (ATK1) development

7.91 E-04 263194_at 2.58 At1g36060 AP2 domain transcription factor development

5.10E-07 259680_at 2.86 At1g77690 putative AUX1-like permease development

3.21 E-08 265813_at 3.47 At2g 18060 putative NAM (no apical meristem)-like protein development

3.23E-10 245173_at 8.79 At2g47520 putative AP2 domain transcription factor, aintegumenta-like development protein

Table 9: Primers used for ANG4 candidate genes amplification and sequencing

REFERENCES

- Anami, S. (2004). Cloning and functional analysis of genes controlling organ growth and development in Arabidopsis thaliana. Masters thesis for International Post-graduate course on Molecular Biology (VUB, Brussels).

- Autran D, Jonak C, Belcram K, Beemster GTS, Kronenberger J, Grandjean O, Inze D, Traas J. 2002. Cell numbers and leaf development in Arabidopsis: a functional analysis of the STRUWWELPETER gene. The EMBO Journal 21, 6036-6049.

- Berna, G., Robles, P and Micol, J. L. (1999) A mutational analysis of leaf morphogenesis in Arabidopsis thaliana. Genetics, 152, 729-742

- De Veylder L, Joubes J, Inze D. (2003) Plant cell cycle transitions. Curr Opin Plant Biol., 6, 536-543.

- Cnops G, Jover-Gil S, Peters J, Neyt P, De Block S, Robles P, Ponce M, Gerats T, Micol J, Van Lijsebettens M (2004) The rotunda2 mutants identify a role for the LEUNIG gene in vegetative leaf morphogenesis. Journal of Experimental Botany 55: 1529-1539

De Veylder L, Beeckman T, Beemster GTS, Krols L, Terras F, Landrieu I, Van Der Schueren E, Maes S, Naudts M, Inze D (2001) Functional analysis of cyclin-dependent kinase inhibitors of Arabidopsis. Plant Cell 13: 1653-1667

- Erickson RO (1976) Modeling of plant growth. Annu. Rev. Plant Physiol. 27: 407-434 Galbraight DW, Harkins KR, Knapp S (1991) Systemic endopolyploidy in Arabidopsis thaliana. Plant Physiol. 96: 985-989

Hightower, R.C. and Meagher, R. B. (1986). The molecular evolution of actin. Genetics 114, 315-332.

- Kamiya N, ltoh J, Morikami A, Nagato Y, Matsuoka M. 2003. The SCARECROW gene's role in asymmetric cell divisions in rice plants. Plant J, 36, 45-54.

Karimi, M., Inze, D. And Depicker, A. (2002). GATEWAY vectors for Agrobacterium- mediated plant transformation. Trends in Plant Sciences 7, 193-195.

- Liscum E, Reed JW. 2002. Genetics of Aux/IAA and ARF action in plant growth and development. Plant MoI Biol, 49, 387-400.

- Lonnstedt I, Speed T 2002. Replicated microarray data. Statistica Sinica 12, 31-46.

- Muller S, Fuchs E, Ovecka M, Wysocka-Diller J, Benfey PN, Hauser MT. 2002. Two new loci, PLEIADE and HYADE, implicate organ-specific regulation of cytokinesis in Arabidopsis. Plant Physiol., 130, 312-324.

Muller I, Wagner W, Volker A, Schellmann S, Nacry P, Kuttner F, Schwarz-Sommer Z, Mayer U, Jurgens G. 2003. Syntaxin specificity of cytokinesis in Arabidopsis. Nat Cell Biol., 5, 531-534.

- Nelissen H, Clarke JH, De Block M, De Block S, Vanderhaeghen R, Zielinski RE, Dyer T, Lust S, Inze D, Van Lijsebettens M. 2003. DRL1, a homolog of the yeast TOT4/KTI12 protein, has a function in meristem activity and organ growth in plants. The Plant Cell 15,

639-654.

Peters JL, Cnops G, Neyt P, Zethof J, Comelis K, Van Lijsebettens M, Gerats T. 2004. An

AFLP-based genome-wide mapping strategy: a practical approach to positional cloning.

Theoretical and Applied Genetics 108, 321-327.

Pyke KA, Marrison JL, Leech RM. 1991. Temporal and spatial development of the cells of the expanding first leaf of Arabidopsis thaliana (L.) Heynh. Journal of Experimental Botany

42, 1407-1416.

Reddy AS, Day IS. 2001. Kinesins in the Arabidopsis genome: a comparative analysis among eukaryotes. BMC Genomics. 2, Epub 25.

Robles, P. and Micol, J. L. (2001). Genome-wide linkage analysis of Arabidopsis genes required for leaf development. MoI. Genet. Genomics. 266, 12-19.

Schiefelbein J. 2003. Cell-fate specification in the epidermis: a common patterning mechanism in the root and shoot. Curr Opin Plant Biol., 6, 74-78.

Smyth GK, Yang YH, Speed T 2003. Statistical issues in cDNA microarray data analysis.

Meth. MoI. Biol. 224, 111-136.

Stone S, Hauksdόttir H, Herschleb J, Kraft E, Callis J (2005) Functional analysis of the

RING-type ubiquitin ligase family of Arabidopsis. Plant Physiol. 137: 13-30.

Strompen G, El Kasmi F, Richter S, Lukowitz W, Assaad FF, Jurgens G, Mayer U. 2002.

The Arabidopsis HINKEL gene encodes a kinesin-related protein involved in cytokinesis and is expressed in a cell cycle-dependent manner. Curr Biol. ,12, 153-158.

Traas J, Vernoux T. 2002. The shoot apical meristem: the dynamics of a stable structure.

Philos Trans R Soc Lond B Biol Sci., 357, 737-747.

Tsiantis M, Hay A. 2003. Comparative plant development: the time of the leaf? Nat Rev

Genet, 4,169-180.

Tsuge T, Tsukaya H, Uchimiya H. 1996. Two independent and polarized processes of cell elongation regulate leaf blade expansion in Arabidopsis thaliana (L.) Heynh. Development

122, 1589-1600.

Van Lijsebettens M, Clarke J. 1998. Leaf development in Arabidopsis. Plant Physiology and

Biochemistry 36, 47-60.

Wu Z, Irizarry KA 2004. Preprocessing of oligonucleotide array data. Nat. Biotechnol. 22,

656-658.

Yang CY, Spielman M, Coles JP, Li Y, Ghelani S, Bourdon V, Brown RC, Lemmon BE,

Scott RJ, Dickinson HG. 2003. TETRASPORE encodes a kinesin required for male meiotic cytokinesis in Arabidopsis. Plant J., 34, 229-240.

Claims

1. The use of a gene encoding a protein comprising SEQ ID N°2, or a functional fragment or variant thereof, to modulate the cell number of a plant organ, or a part thereof.

2. The use according to claim 1 , whereby said plant organ is a plant leaf or a plant root.

3. The use according to claim 1 or 2, whereby said plant organ part are the leaf palisade cells.

4. The use according to any of the preceding claims, to modulate leaf morphology.

5. The use of a gene encoding a protein comprising SEQ ID N°2, or a functional fragment or variant thereof, to modulate root length.

6. The use of a gene encoding a protein comprising SEQ ID N°2, or a functional fragment or variant thereof, to increase plant biomass.