Peptide signalling in plant development:

Transcription

1 Peptide signalling in plant development: Functional analysis of CLE ligands in Arabidopsis Martijn Fiers

2 Promotor: Prof. Dr. W. J. Stiekema Hoogleraar Genoom Informatica, Laboratorium voor Bioinformatica, Wageningen Universiteit Co-promotor: Prof. Dr. C. M. Liu Center for Signal Transduction & Metabolomics, Institute of Botany, Chinese Academy of Sciences, Beijing, China Promotiecommissie: Prof. Dr. S. C. de Vries, Wageningen Universiteit Prof. Dr. G. C. Angenent, Radboud Universiteit Nijmegen Dr. R. Heidstra, Universiteit Utrecht Dr. R. Offringa, Universiteit Leiden

3 Martijn A. Fiers Peptide signalling in plant development: Functional analysis of CLE ligands in Arabidopsis Proefschrift ter verkrijging van de graad van doctor op gezag van de rector magnificus van Wageningen Universiteit, Prof. dr. M. J. Kropff, in het openbaar te verdedigen op woensdag 10 januari 2007 des namiddags te uur in de Aula.

4 Martijn Fiers (2007) Peptide signalling in plant development: Functional analysis of CLE ligands in Arabidopsis PhD Thesis Wageningen University, The Netherlands with summaries in English and Dutch ISBN

5 Contents Chapter 1 CLV3/ESR signals: Ligands in plant development 7 Chapter 2 Mis-expression of the CLV3/ESR-like gene CLE19 in 35 Arabidopsis leads to a consumption of root meristem Chapter 3 Secreted proteins in Brassica napus microspore cultures 63 Chapter 4 The 14-Amino Acid CLV3, CLE19 and CLE40 Peptides 71 Trigger Consumption of the Root Meristem in Arabidopsis through a CLAVATA2-Dependent Pathway Chapter 5 The CLE motif of CLAVATA3 is functionally independent 99 from the non-conserved flanking sequences Chapter 6 Concluding remarks and future prospects 121 Summary 127 Samenvatting 131 Dankwoord 135 List of publications 137 Curriculum vitae 139 Appendix colour figures 141

6

7 Chapter 1 CLV3/ESR signals: Ligands in plant development Martijn Fiers and Mark Fiers Plant Research International, B.V., P.O. Box 16, 6700 AA Wageningen, The Netherlands

8 Chapter 1 8

9 General Overview GENERAL OVERVIEW In multicellular organisms, cell to-cell communication is essential for co-ordinating their growth and differentiation. In animals, peptide ligands are well known to be the major players in such communication (eg. insulin, secretin and tachykinin; Alberts et al., 1994). This is in contrast to plants, in which most intercellular communication has been explained on the basis of phytohormones such as auxin, cytokinin, gibberillin, abscisic acid, ethylene and brassinosteroid (Kende et al., 1997). Recently several putative peptide ligands have been identified in plants that are involved in plant-pathogen interaction, cell division, anther-stigma interaction and stem cell maintenance. These plant-specific peptides and their role in cell-tocell communication will be discussed in this review. Attention will be given to recent understanding on CLV3 and related genes. Peptide ligands in plants The first peptide ligand isolated from plants was SYSTEMIN (Pearce et al., 1991). It was isolated as a systemic signal induced upon wounding in tomato. The active 18-AA peptide is processed from a 200 amino acid (AA) precursor called PROSYSTEMIN, which is able to activate defense genes such as those encoding proteinase inhibitors. Using a radioactive photoaffinity-labeled SYSTEMIN analogue, a 160 kd protein with high affinity to the peptide was isolated and called SR160 (Scheer et al., 2002). This protein was identified as a leucine rich receptor (LRR) like kinase (RLK) with, surprisingly, the highest homology to Arabidopsis BRASSINOSTEROID (BR) INSENSITIVE1 (BRI1; Scheer et al., 2002). BRI1 is involved in the perception of BR and forms a receptor complex with BRI1-ASSOCIATED RECEPTOR KINASE1 (BAK1) (Nam and Li, 2002). SR160 was also independently found to be mutated in a tomato mutant, cu-3, which is dwarf and does not respond to brassinosteroids as does the Arabidopsis bri mutant (Montoya et al., 2002). The fact that the same receptor seems to be involved in steroid and SYSTEMIN signaling brings in a possible intriguing link between steroid hormone and peptide hormone perception in plants. The only other receptor known to perceive two types of ligands is the mammalian oxytocin receptor (OTR), a member of the G-protein-coupled receptor family. OTR is able to bind to two different peptide hormones, oxytocin and progesterone. These two hormone ligands 9

10 Chapter 1 compete for binding to the OTR receptor; with progesterone binding to OTR leading to a decrease in the sensitivity to oxytocin (Grazzini et al., 1998). However, SYSTEMIN and brassinosteroids (BRs) elicit very different physiological responses. The possible dual function of the BRI1 receptor is more similar to that of the Drosophila TOLL receptor, which also possesses an extracellular LRR domain. The TOLL receptor is involved in two different physiological processes, namely the dorsoventral patterning in embryos and the congenital immune defense against fungi and bacteria in adult flies (Lemaitre et al., 1996). In both cases the ligand SPAETZLE induces the dimerisation and activation of the receptor complex. Whether SR160 forms a homo-dimer like the TOLL receptor, or a hetero-dimer to sense SYSTEMIN as in BR signalling remains to be investigated. Complementation of the bri1 mutant with the SR160 gene could shed some light on the question if SR160 is the functional equivalent of BRI1. Another family of signaling peptides is involved in self-incompatibility (SI). Many flowering plants are self-incompatible, which prevents self-pollination thereby enforcing outbreeding. The SI system is encoded by a single locus named the sterility locus (S-locus). SI in the Brassicaceae is regulated by two components, namely the secreted S-locus Cysteine Rich protein (SCR) and the S-locus Receptor like Kinase (SRK). The SRK is a receptor-like kinase consisting of an S-locus glycoprotein (SLG) as the extracellular domain and a cytoplasmic Ser/Thr kinase domain. The SRK gene is specifically expressed in the stigmatic papillae (Stein et al., 1991). The ligand SCR is expressed in pollen, encoding a highly polymorpic protein with a putative secretion signal, which is able to interact with the SRK receptor (Schopfer et al., 1999; Kachroo et al., 2001). While several findings have confirmed that the SCR proteins are the pollen determinant of the SI, the signalling cascade downstream of the receptor, which leads to the pollen rejection, is still unknown. Most plant cells have the ability, even when fully differentiated, to dedifferentiate and proliferate in vitro as callus upon treatment with plant hormones such as auxin and cytokinin. The density of the cell suspension is a critical factor for the growth; high cell density promotes proliferation, while cells may not be able to divide when diluted to a very low density. Cell division in a low density culture can be enhanced greatly when conditioned medium is used (Bellincampi et al., 1987). Using this phenomenon the sulfated pentapeptide 10

11 General Overview PHYTOSULFOKINE (PSK) was isolated and characterized from medium of cultured mesophyll cells of Asparagus (Matsubayashi et al., 1996). PSK is the only known plant peptide ligand that is post-translationally modified with a sulfate group attached to two tyrosine residues within the peptide. A rice PSK (OsPSK) was isolated that comprises an 89- AA preprotein,a signal sequence at the N-terminus and the peptide ligand near the C-terminus (Yang et al., 1999). In Arabidopsis four PSK-like proteins were identified in which the N- terminus is not conserved but with a PSK domain at its C-terminus simular to OsPSK (Yang et al., 2001). A PSK receptor was identified from carrot microsomal fractions using a photo-activatable PSK analogue (Matsubayashi et al., 2002). The receptor encodes a LRR receptor kinase, with 21 extracellular LRR domains, one transmembrane region and an intracellular kinase domain, which shares both structural and sequence identity with BRI1 and CLV1 (Matsubayashi et al., 2002). Although the processing and the in vitro function of PSK are well documented, the downstream signaling cascade and role of PSK in normal plant growth and development remains unknown. CLV3 signalling in the shoot apical meristem of Arabidopsis The shoot apical meristem (SAM) in plants provides the founder cells for the formation of new organs and can be divided into different zones and layers. The central zone (CZ) at the middle of the meristem is responsible for meristem maintenance, with an upper stem cell cluster, marked with CLV3 expression, that provides cells for leaf formation and stem growth,and a lower cell cluster, the rib zone (RZ). The descendants of the CZ cells move into the peripheral zone (PZ) where the different organs, leaves in the case of the SAM, are initiated (Figure 1). In addition to its division into zones, the SAM can also be separated into different cell layers. The L1 and L2 layers represent the tunica layers where L3 represents the internal layers or corpus (Figure 1). 11

12 Chapter 1 Figure 1. Schematic representation of the shoot apical meristem. The SAM can be divided into three clonally distinct cell layers (L1, L2 and L3). L1 and L2 represent the epidermal and sub epidermal layer (tunica layers) whereas L3 represents the corpus. The central zone (CZ) harbors the stem cells that are specified by CLV3 expression and the organizing centre that is marked by WUS expression. The CZ is surrounded by the peripheral zone (PZ) where organ primordia are initiated. CLE19 is specifically expressed in the periphery of the SAM where the new primordia will be formed, whereas CLE40 and CLV2 are expressed throughout the SAM. The slow-dividing multipotent stem cells are located in a micro-environment (niche) in the CZ of the SAM where new cells are formed for the PZ, but also for its own replenishment. The stem cells maintain simultaneously two antagonistic events, cell propagation and cell differentiation, where the decision is made by a population-based mechanism in which signals from neighbouring cells play the most important role to determine the fate of the progeny cells (Weigel et al., 2002). CLAVATA3 (CLV3) is expressed in the stem cells of the Arabidopsis SAM and encodes a secreted precursor protein, which is processed into a 12-amino acid (AA) peptide hormone (Kondo et al., 2006). CLV3 is thought to interact with the underlying CLV1/CLV2 receptor complex to restrict the number of stem cells in the SAM (Figure 1, Fletcher et al., 1999; Rojo et al., 2002). CLV1 is a membrane-bound leucine-rich repeat (LRR)-receptor kinase and CLV2 is an LRR-receptor-like protein lacking a kinase domain (Clark et al., 1997; Jeong et al., 1999). While the stem cells are marked by CLV3 expression, the underlying SAM organizing centre (OC) is marked by the expression of the stem cell-promoting WUSCHEL (WUS) gene (Figure 1, Schoof et al., 2000). WUS, a homeobox transcription factor, provides a positive signal to maintain an undifferentiated state, this in contrast to the CLV pathway which restricts stem cell development by negatively regulating WUS expression (Brand et al., 12

13 General Overview 2000; Schoof et al., 2000). CLV3 in turn is positively regulated by WUS, creating a feedback regulatory loop between CLV3 and WUS that regulates the number of stem cells in the SAM. As such, clv1, clv2 and clv3 mutants have enlarged SAMs, while the wus mutation or CLV3 over-expression results in the differentiation of the stem cells and subsequently the termination of SAM development (Laux et al., 1996; Brand et al., 2000). Induction of CLV3 expression in WT Arabidopsis results in a decrease of endogenous CLV3 and WUS expression as soon as three hours after induction (Müller et al., 2006). The size reduction of the SAM, due to over expression of CLV3, is accompanied by a size increase of the primordia (Müller et al., 2006). Muller et al showed that over expression of CLV3, and subsequent down-regulation of WUS, caused a decrease in cell division in the meristem center and a shift of the CZ-PZ boundary by allowing cells from the CZ to be recruited in organ primordia. The combination of these two processes resulted in termination of the SAM. In the opposite experiment, expression of an inducible CLV3 RNAi (CLV3i) in WT Arabidopsis, results in an increase of the WUS gene expression because of the release of the negative feedback by CLV3 (Reddy and Meyerowitz, 2005). This increase of WUS expression is followed by an increase in cell division and the re-specification of PZ cells to the CZ, resulting in an enlargement of the SAM as seen in the clv mutants (Reddy and Meyerowitz, 2005). The balance between CLV3 and WUS expression is therefore essential for a proper specification of the CZ and PZ domains; when this balance is disturbed it affects the rate of cell division accompanied by a respecification of cells from CZ to PZ or vice versa. Although two receptors and a putative ligand have been identified in the CLV pathway, the downstream signalling cascade is largely unknown. One of the downstream components is KINASE-ASSOCIATED PROTEIN PHOSPHATASE (KAPP), which comprises a type 2C protein phosphatase with a membrane anchor and was identified as negative regulator of the CLV1 signal transduction pathway (Williams et al., 1997; Stone et al., 1998). KAPP, as CLV1, is expressed in apical and young floral meristems (Williams et al., 1997). Over expression of KAPP in WT Arabidopsis results in a clavata phenotype and mimics a clv1 mutant phenotype, while upon co-suppression of KAPP in clv1-1 and clv1-6 backgrounds a restoration of the clavata phenotype to WT is observed (Williams, 1997; Stone, 1998). CLV1 13

14 Chapter 1 and KAPP are able to interact in vitro, while in vivo they are part of a 450 kd receptor complex comprising a Rho GTPase-related protein and CLV1 (Williams et al., 1997; Stone et al., 1998; Trotochaud et al., 1999). Screening Arabidopsis clavata mutants for suppressors of the CLV signalling pathway provided another downstream signalling factor in the CLV1 pathway, POLTERGEIST (POL) (Yu et al., 2000). POL encodes a protein phosphatase 2C and is part of a family of six members, namely POL and POLTERGEIST LIKE (PLL) 1 to 5 (Yu et al., 2000; 2003). While both KAPP and POL are 2C type protein phosphatases, there are some differences between the two proteins: KAPP is membrane anchored, whereas POL has a putative nuclear localisation signal and is not associated with the membrane (Yu et al., 2003; Stone et al., 1994). Additionally, KAPP is expressed in apical and young floral meristems along with CLV1, POL is broadly expressed within the meristem and throughout the plant (Williams et al., 1997; Yu et al., 2003). Both of these proteins are able to bind to and modulate CLV1 activity and are negative regulators of the CLV pathway. CLV3/ESR gene family CLV3 belongs to the CLV3/ESR (CLE) family of genes and is named after the first two founder genes CLV3 and Endosperm Surrounding Region (ESR). ESR genes were first identified in maize as being expressed in the endosperm regions surrounding the embryo (Opsahl et al., 1997). All three ESR genes encode extra-cellular proteins with unknown function but with a conserved CLE motif similar to CLV3, near their C-termini (Bonello et al., 2002; Cock and McCormick, 2001). In Arabidopsis the CLE family consists of 31 known members of which the majority are transcribed in one or more tissues (Cock and McCormick, 2001; Hobe et al., 2003; Sharma et al., 2003; Fiers et al., 2004; Strabala et al., 2006; Figure 2). All CLE genes share three characteristics with CLV3 and the ESR proteins: they all encode a small protein (<10kD) with a putative secretion signal at their N-termini and contain a conserved 14-AA motif (CLE-motif) at or near their C-termini (Cock and McCormick, 2001). The importance of the CLE motif is demonstrated by a single point mutation in the CLE motif 14

15 General Overview present in the clv3-1 and clv3-5 mutant, which is sufficient to abolish the CLV3 function (Fletcher et al., 1999). Figure 2. Alignment of CLE proteins. Arabidopsis CLE proteins are aligned, together with several related proteins from other species including those from parasitic nematode (HgSYV46), Brassica (BnCLE19) and maize (ZmEsr3). The CLE domain is framed and the mature CLV3 peptide is underlined. 15

16 Chapter 1 Domain deletion or replacement analysis in CLV3 revealed the critical importance of the CLE motif (Fiers et al., 2006; Ni et al., 2006). The region between SS and CLE motif can be removed or replaced by unrelated ERECTA sequences and the C-terminal sequence after the CLE motif can be deleted without interfering with the function of CLV3. In contrast, the removal of the CLE-motif and the sequence C-terminal to the CLE-motif abolished its function completely (Fiers et al., 2006; Ni et al., 2006). Recently, the mature CLV3 (MCLV3) peptide was determined and was shown to consist of 12 AA comprising the CLE motif (Figure 2), of which the first two prolines were modified to hydroxyproline (Kondo et al., 2006). Hydroxylation of these prolines is not required for MCLV3 activity but maybe serves another function such as stability of the peptide. It is surprising to note that untill now mutation of CLE genes does not appear to result in clear mutant phenotypes as in the case CLV3, i.e. an enlarged SAM and increased number of carpels (Clark et al., 1995). The knockout phenotype of CLE40 showed a very subtle root waving phenotype while no phenotype was observed upon knockout of CLE19 (Hobe et al., 2003; Fiers et al., 2004). CLV3 is only expressed in the CZ of the SAM while CLE40 is expressed in all tissues at low levels including the CLV3 expression domain. The presence of endogenous CLE40 could not complement the CLV3 defect, since the clv3/cle40 double mutant exhibited only a clv3 phenotype and no enhancement of meristem defects was observed (Hobe et al., 2003). Interestingly, when CLE40 is placed under the control of the regulatory elements of CLV3, it can complement clv3 completely (Hobe et al., 2003). One possible explanation that CLE40 does not restore a CLV3 defect could be that the expression level of CLE40 is too low to replace CLV3 signalling. Exchanging the C-terminal CLE domain of CLV3 with those from other CLE proteins resulted in variable degrees of complementation in clv3-1 background, from almost complete complementation in the case of CLE1 and CLE6, to hardly any effect with CLE25 and CLE26 (Ni et al., 2006). Experiments using in vitro-applied peptides showed that the clv3 defect in the SAM could be restored by CLE40p, a 14-AA peptide corresponding to the CLE-motif of CLE40, while CLE19p and CLE5p could only partially complement the mutant phenotype, and CLE22p was not able to restore the enlarged meristem in clv3 mutants (Fiers et al., 2006). These domain swapping experiments revealed an interesting redundancy among the CLE 16

17 General Overview family members, in which the receptors seem to be able to tolerate a certain degree of CLE sequence variants. Several CLEs are able to replace CLV3 in vitro or when placed under the CLV3 regulatory elements, but there is no evidence that this also occurs in planta. Since the secreted CLE molecules are not able to travel over long distances (Lenhard and Laux, 2003), a biologically relevant redundancy requires an over-lapping or adjacent expression, and a sufficiently high expression level. One feature that several CLE genes have in common is the termination of the root meristem upon over-expression. CLV3, CLE40 and CLE19 over-expression results, among other phenotypes, in the termination of the root meristem (Hobe et al., 2003, Fiers et al., 2004; Casamitjana et al., 2003). When CLE19 was placed under control of a root specific promoter, it also caused termination of the root meristem. This transgenic line was used as basis for a mutagenesis approach to find key regulators in the CLE19 signalling machinery in the root (Casamitjana et al., 2003), and led to the identification of two mutants that were able to restore the root meristem phenotype to WT namely SUPRESSOR OF LLP1 (sol1) and sol2 (Casamitjana et al., 2003). Only SOL1 was cloned and encodes a Zn 2+ carboxypeptidase which is expressed in different tissues of the plant (Casamitjana et al., 2003). Remarkably the sol1 mutant does not display any phenotype beside the suppression of CLE19 over expression in the root (Casamitjana et al., 2003). SOL1 is thought to play a role in ligand processing, although this role has not been experimentally validated (Casamitjana et al., 2003). Beside CLV3, CLE19 and CLE40, several other CLEs can cause a short root phenotype, as shown by over-expression of transgenes or by treatment with the corresponding peptides (Strabala et al., 2006; Fiers et al., 2005; Ito et al., 2006). Like CLE1, CLE6 and CLE40 can replace CLV3 in the SAM, there is also a certain degree of redundancy in root meristem consumption due to excess CLE protein or CLE peptides in root development (Fiers et al 2005; Ni et al., 2006, Strabala et al., 2006, Fiers et al., 2006). Seedlings grown on media containing a CLE peptide comprising the CLE motif of CLE40, CLE19 or CLV3 caused termination of the root meristem, which mimics the over-expression phenotype of the corresponding CLE gene (Fiers et al., 2005, Strabala et al., 2006). The pericycle, endodermis and cortex cell-layers are all misexpressed upon incubation of roots with CLV3, CLE19 and CLE40 peptide (Fiers et al., 2005). Furthermore, the cortical daughter cells are affected. These cells increase in number but lose their stem cell identity and obtain a cortex identity (Fiers et al., 2005). 17

18 Chapter 1 To identify components of the signal transduction pathway involved in perception of the CLE signal, clv1, clv2 and clv3 mutants were grown on media containing one of the CLE peptides (Fiers et al., 2005). The results showed that the CLV2 receptor-like protein is involved in CLE signal perception, suggesting the presence of a CLV2-dependent signalling pathway in the root meristem (Fiers et al., 2005). Over-expression of several CLE genes such as CLV3, CLE19 and CLE40 seems to interact with or block an unknown cell identity-maintaining CLV2 receptor complex in the roots. This blockage results in a mis-communication between cells and cell layers that finally results in termination of the root meristem. Beside the identification of CLE peptides in Arabidopsis, a CLE-like peptide, Tracheary Element Differentiation Inhibitory Factor (TDIF), was isolated from Zinnia Elegans mesophyll cell culture medium and identified as a suppressor of xylem development (Ito et al., 2006). This 12-AA CLE peptide, with two hydroxylated prolines, as seen with MCLV3, is identical to the CLE domain of CLE41/44. Three CLE peptides, CLE41, 42 and 44 containing two hydroxyproline residues, were shown to independently suppress tracheary element differentiation, but were unable to terminate development of the root apical meristem (Ito et al., 2006). While CLV3 is involved in the suppression of stem cell development, TDIF suppresses the differentiation of xylem cells from stem cell like procambial cells and promotes cell division. Interestingly a functional CLV3 peptide was able to promote xylem cell differentiation in a Zinnia cell culture revealing two counteracting pathways in CLE signalling in vascular development, one that promotes and one that inhibits stem cell differentiation (Ito et al., 2006). It will be interesting to determine whether CLE peptides, beside their involvement in the suppression of stem cell development as shown with CLV3, are also involved in the promotion of stem cell development. CLE proteins outside the plant kingdom The only known CLE gene outside the plant kingdom is the CLV3-like gene, HgSYV46, of the parasitic soybean cyst nematode Heterodera glycines (Wang et al., 2001; Gao et al., 2001; Olsen et al., 2003; Figure 2). The juvenile nematode hatches from eggs laid in the soil, penetrates the root directly behind the root tip at the elongation zone, using its stylet to breach cell walls, and then moves towards a site near the vascular tissue (Davis et al., 2005). The oesophageal gland cells of nematodes actively synthesize secretions which are injected 18

19 General Overview through the stylet (oral spear) into cells to alterate the cell identity into specific feeding cells (Davis et al., 2005). The HgCLE gene was isolated from an oesophageal gland cell-specific library and contains a putative signal sequence at its N-terminus and a CLE domain near its C-terminus and was shown to be specifically expressed within the dorsal oesophageal gland cell (Wang et al., 2001; 2005; Gao et al., 2005; Olsen et al., 2003). When this gene was placed under control of the cauliflower mosaic virus 35S (CaMV 35S) promoter it was able to complement the clv3-1 mutant and in WT Columbia resulted in the termination of the root meristem similar to CLV3 (Brand et al., 2000; Wang et al., 2005). The function and origin of the HgCLE protein is still not known but the gene may have been adapted from plants and imitating an endogenous CLE peptide and so promoting the differentiation of root cells into specific feeding cells. CLE genes in rice CLE genes have been identified in a wide variety of plant species ranging from dicots (Arabidopsis, Brassica, Zinnia and Populus) to monocots (maize, rice and wheat; Cock and McCormick., 2001; Olsen et al., 2003; Fiers et al., 2004; Ito et al., 2006). The sequenced rice genome provides a good opportunity to identify new CLE genes and to compare their structure with the Arabidopsis CLEs. To this end, all genes from the rice genome, as annotated by TIGR (v4), in the rice genome, were analyzed using a Hidden Markov Model (HMMer) domain search using a model trained with all known CLE domains from Arabidopsis. As a cut-off a HMMer score of 8 was used, corresponding to the lowest scoring Arabidopsis CLE gene. All Open Reading Frames (ORFs) with a putative CLE domain were determined using the TIGR annotation except for four Oryza sativa (Os) CLEs with a putative intron (OsCLE4, OsCLE18, OsCLE33 and OsCLE34), which were changed according to the genomic sequence and consist of one ORF. To find a rice CLE close to CLV3 the CLE domain of CLV3 was compared with the rice genomic database. The best scoring sequence was extended and analyzed in NetPlantGene ( because of the lack of a methionine. Two putative introns were found, as in CLV3, and the resulting protein was named OsCLV3. All putative CLE proteins were checked with signalp for the presence of a secretion signal ( 19

20 Chapter 1 Figure 3. Alignment of rice CLE proteins. Rice CLE proteins are aligned, the CLE domain is framed and the extra CLE domains in OsCLE32 and OsCLE33 are boxed with a dotted line. Data mining of the rice genome resulted in a total of 34 putative CLE genes compared to 31 in Arabidopsis (figure 3). Of these 34 rice CLEs, 32 contained a putative secretion signal (including OsCLE32) while the remaining two CLE genes, OsCLE22 and OsCLE25, may 20

21 General Overview contain an intron in the sequence somewhere before the CLE domain, which in Arabidopsis is reported for CLE40 and CLV3 (Fletcher et al., 1999; Hobe et al., 2003). Interestingly, there are two rice CLEs with multiple CLE domains, five in the case of OsCLE34 and even six in the case of OsCLE33 (Figure 3). The only other known CLE protein with multiple CLE domains is a wheat EST (BE401912) that encodes a putative CLE protein with three putative CLE domains (Olsen et al., 2003). Although both rice and wheat are monocots, the rice genome does not contain a homologue of the wheat CLE BE If the CLE domain is the functional part of CLE proteins, then the wheat and the two rice CLE proteins may generate multiple CLE peptides. This has also been reported in the case of another peptide ligand, namely the tobacco SYSTEMIN peptides (Tob Sys1 and Tob Sys2) which are both derived from one polypeptide precursor (Pearce et al., 2001). OsCLE TIGR annotation OsEst OsCLE TIGR annotation OsEst OsCLE m04676 OsCLE m02702 OsCLE m04679 OsCLE m03069 Yes OsCLE m05972 Yes OsCLE m00304 Yes OsCLE m00905 Yes OsCLE m05516 OsCLE m03497 OsCLE m04155* OsCLE m03665 Yes OsCLE m00698 Yes OsCLE m05431 OsCLE m00749 Yes OsCLE m04868 Yes OsCLE m05577* Yes OsCLE m04667 OsCLE m02111 Yes OsCLE m05429 OsCLE m04179 OsCLE m04668 Yes OsCLE m04089 OsCLE m03315 Yes OsCLE m02540 Yes OsCLE m01423 OsCLE m05749 Yes OsCLE m03370 OsCLE m03300 OsCLE m05558 Yes OsCLV3 OsCLE m04772 OsCLE m02761* OsCLE m00508 OsCLE m00185* Table 1. Rice CLE genes for which an EST is available. The TIGR annotation is shown for the rice CLEs. Genes modified from the original TIGR annotation are marked with a *. CLE genes with a complete or partial corresponding EST are depicted. 21

22 Chapter 1 Figure 4 *. Similarity matrix of a comparison between rice and Arabidopsis CLE proteins. The bar depicted on the right represents the similarity (%) as calculated by ClustalW. *) Colour figures in appendix, page I Search of the NCBI database ( for rice EST sequences resulted in the confirmation of 15 CLE genes by complete or partial ESTs sequences but no EST was found for OsCLE32 and OsCLE33, both of which contain multiple CLE domains (Table 1). In order to determine which rice CLE protein is the orthologue of which Arabidopsis CLE protein, all CLEs from rice and Arabidopsis have been aligned and identity between them scored (Figure 4). This comparison reveals that CLE proteins have diverged too much to identify any rice orthologue in the Arabidopsis CLE family. The only similarity that could be 22

23 General Overview detected is within the Arabidopsis or rice CLE family, as seen in the case of CLE5 and CLE6 (Figure 4). Figure 5 *. Similarity matrix of a comparison between rice and Arabidopsis CLE domains. The bar depicted on the right represents the similarity (%) as calculated by ClustalW. *) Colour figures in appendix, page II In a final attempt to find orthologues we used the conserved 14AA CLE domain of all Arabidopsis and rice CLE proteins and compared them with ClustalW (Figure 5). This comparison resulted in the identification of several groups of related CLE domains, as seen in the case of CLE45, which groups together with OsCLE5, 6 and 14. Also OsCLV3 and CLV3 cluster together making OsCLV3 the most likely orthologue of CLV3 based on the comparison of the CLE domains (Figure 5). 23

24 Chapter 1 While there are some overlapping CLE domains between rice and Arabidopsis, most of the clustering with CLE domains is observed either among Arabidopsis or rice sequences, i.e. not between sequences from the two species, and is probably due to the short length of the CLE domain (14AA). This comparison resulted in identification of some putative orthologues between the rice and Arabidopsis CLE family, but in the absence of functional analysis this assigned orthology remains speculative. Receptors used for transmitting the CLE signal The only known receptors to be involved in CLE-signaling are the CLV1 and CLV2, which are thought to be part of a receptor complex in the meristem (Jeong et al., 1999; Trotochaud et al., 1999). Both CLV1 and CLV2 belong to a large family of receptors. CLV1 belongs to the LRR receptor kinases (LRR-RKs), which comprise the largest subfamily of transmembrane receptor-like kinases in plants, with over 200 members in Arabidopsis (Torii, 2005). CLV2 belongs to the LRR-receptor like proteins (LRR-RLPs), of which there are 51 members in Arabidopsis (Fritz et al., 2005). LRR-RKs and LRR-RLPs regulate a wide variety of processes ranging from stem cell maintenance, defense response, hormone perception and symbiosis between bacteria and plant (Tori, 2005; Fritz et al., 2005; Krucell et al., 2002; Nishimura et al., 2002). CLV1 is specifically expressed in the SAM, inflorescence and floral meristems in Arabidopsis, where it is expressed in the L2 layer and in the CZ and the RZ, while CLV2 is more ubiquitiusly expressed in different tissues, including the SAM (Clark et al., 1997; Jeong et al., 1999; Figure 1). The strong link between CLV1 and CLV2 in the SAM is shown by several independent approaches. Genetic data showed that double mutants of clv1 and clv2 gave the same phenotypes as the single mutant, while biochemical data showed that CLV1 is part of a covalently linked receptor complex, and that in the case of clv2 mutants, there was almost no CLV1 protein present when CLV1 transcripts are elevated, which suggests that CLV2 is needed to form a stable receptor complex to interact with CLV3 (Jeong et al., 1999; Trotochaud et al., 1999; Kayes et al., 1998). The weakness in this simple model appeared when a null mutant of CLV1 displayed a very weak phenotype, while the strong clv1 alleles were shown to be dominant-negative (Dievart et 24

25 General Overview al., 2003). This result, together with the fact that the clv3-2 mutant displays a much stronger phenotype than clv1-1, suggests that CLV1 has a functional overlap with another receptor(s). This observation could also explain the differences in strength of the phenotypes between different clv mutants (Clark et al., 1995; Dievart et al., 2003). It is proposed that CLV2 can form a receptor dimer with CLV1 to perceive the CLV3 signal, but direct biochemical evidence is lacking. Another possibility is that a receptor complex is formed, upon CLV3 induction, as a tetramer including CLV1/2 and possible also unknown receptor(s). An interesting feature of CLV2 is its involvement in root apical meristem (RAM) development, as shown by testing functional CLE peptides with clv2-1 mutant (Fiers et al., 2005). The root meristem of the clv2 mutant did not collapse upon CLE peptide treatment, suggesting the involvement of the CLV2 receptor in the perception of the CLE peptides in the root, and thereby providing a link between root suggesting that CLV2 interacts with an unknown LRR-RLK receptor in the root meristem to perceive and transmit the CLE signal. Three CLV1-like LRR-RLKs, expressed during meristem development, Barely Any Meristem capitalize (BAM) 1, 2 and 3, are of particular interest in regard to the CLV pathway (De Young et al., 2006). The BAM receptors have a developmental role in stem cell regulation that is opposite to that of CLV1. Loss of function of the triple bam1/bam2/bam3 mutant results in the loss of stem cells at the shoot and flower meristem, while clv1 mutants show an increase in the size of the shoot and flower meristem (Clark et al., 1993; DeYoung et al., 2006). CLV1 can complement BAM1 and BAM2 function in developing organs and BAM1 and BAM2 are able to partially complement CLV1 function within the meristem, revealing some redundancy among these proteins (DeYoung et al., 2006). Whether the BAM receptors can interact with the CLV1 or CLV2 receptor or are part of another signalling cascade remains to be investigated. The next logical step would be to combine the bam and clv mutants to determine if the bam mutants can enhance or rescue the mutant phenotype in one of the clv mutants. More and more receptors, as well as downstream signalling components, like transcription factors, are being identified that play a role in meristem development. One of the major challenges for future research on CLE proteins is the identification of CLE protein complex components and the identification of downstream signalling components. 25

26 Chapter 1 Outline of this thesis Chapter 1 provides a general overview of peptide signaling in plants. Four known peptide ligands have been described that are variously involved in plant-pathogen interaction, cell division, anther-stigma interaction. One of these signalling peptides, CLV3, which is involved in stem cell maintenance in Arabidopsis thaliana, is discussed in detail. Chapters 2 to 5 describe experimental contributions to the research field of peptide signaling in plants. Chapter 2 is a short survey on the biochemical aspects of haploid embryogenesis and in particular the identification of extra cellular signalling molecules in microspore embryo development. Differences were examined in protein profiles present in the medium from high-yielding microspore derived embryos cultures and low-yielding cultures that did not produce any embryos. Several proteins were isolated and identified by de-novo protein sequencing. Chapter 3 describes DD3-12, one of the genes that were isolated in a screen to identify markers for embryo development in Brassica napus haploid embryo development. DD3-12, also known as BnCLE19, encodes a small 74 amino acid secreted protein that belongs to the CLAVATA3/ESR (CLE) like family of proteins. The over-expression phenotypes of BnCLE19 and CLE19, its Arabidopsis orthologue are described. Chapter 4 describes research aimed at the analysis of the functional domain of CLE proteins. The only conserved domain among all CLE proteins, beside the secretion signal, is a 14 amino acid C-terminal CLE domain. To test if this domain is the functional part of the CLE proteins and if this domain can act independently from the rest of the protein an in-vitro system was developed using chemically synthesized 14 amino acid CLE peptides. We show that these CLE peptides can cause an over-expression phenotype in the roots of Arabidopsis, namely termination of the root meristem, by causing a misspecification in several cell layers and a premature differentiation of the cortical daughter cells. 26

27 General Overview Chapter 5 describes the functional analysis of the CLV3 peptide in SAM development through complemention of the clv3 mutant and by examining the potential redundancy of CLE proteins using different CLE peptides. The in-vitro peptide approach is combined with a deletion analysis of the CLV3 gene to prove that the CLE domain, in addition to the secretion signal, is sufficient to rescue the clv3 mutant phenotype. Chapter 6 concludes this thesis with perspectives of future research in receptor-ligand interaction. 27

28 Chapter 1 REFERENCES Bellincampi, D. and Morpurgo, G. (1987). Conditioning factor affecting growth in plant cells in culture. Plant Sci. 51, Bonello, J.F., Sevilla-Lecoq S., Berne A., Risueno M.C., Dumas C., and Rogowsky, P.M. (2002). Esr proteins are secreted by the cells of the embryo surrounding region. J. Exp. Bot. 53, Brand, U., Fletcher, J.C., Hobe, M., Meyerowitz, E.M., and Simon, R. (2000). Dependence of stem cell fate in Arabidopsis on a feedback loop regulated by CLV3 activity. Science 289, Casamitjana-Martinez, E., Hofhuis, H.F., Xu, J., Liu, C.M., Heidstra, R., and Scheres, B. (2003). Root-specific CLE19 overexpression and the sol1/2 suppressors implicate a CLVlike pathway in the control of Arabidopsis root meristem maintenace. Curr. Biol. 13, Cock, J.M., and McCormick, S. (2001). A large family of genes that share homology with CLAVATA3. Plant Physiol. 126, Clark, S.E., Running, M.P., and Meyerowitz, E.M. (1993). CLAVATA1, a regulator of meristem and flower development in Arabidopsis. Development 119, Clark, S.E., Running, M.P., and Meyerowitz, E.M. (1995). CLAVATA3 is a specific regulator of shoot and floral meristem development affecting the same processes as CLAVATA1. Development 121, Clark, S.E., Williams, R.W., and Meyerowitz, E.M. (1997). The CLAVATA1 gene encodes a putative receptor kinase that controls shoot and floral meristem size in Arabidopsis. Cell 89, DeYoung, B.J., Bickle, K.L., Schrage, K.J., Muskett, P., Patel, K., and Clark, S.E. (2006). The Clavata1-related BAM1, BAM2 and BAM3 receptor kinase-like proteins are required for meristem function in Arabidopsis. Plant Journal 45, 1-16 Diévart, A., Dalal, M., Tax, F.E., Lacey, A.D., Huttly, A., Li, J.M., and Clark, S.E. (2003). CLAVATA1 dominant-negative alleles reveal functional overlap between multiple receptor kinases that regulate meristem and organ development. Plant Cell, 15, Davis, E.L., and Mitchum, M.G. (25). Nematodes, Sophisticated Parasites of Legumes. Plant Phys. 137,

29 General Overview Fiers, M., Hause, G., Boutilier, K., Casamitjana-Martinez, E., Weijers, D., Offringa, D., van der Geest, L., van Lookeren Campagne, M., and Liu, C.M. (2004). Mis-expression of the CLV3/ESR-like gene CLE19 in Arabidopsis leads to a consumption of root meristem. Gene 327, Fiers, M., Golemiec, E., Xu, J., van der Geest, L., Heidstra, R., Stiekema, W., and Liu, C,M. (2005).The 14-amino acid CLV3, CLE19 and CLE40 peptides trigger consumption of the root meristem in Arabidopsis through a CLAVATA2-dependent pathway. Plant Cell 17, Fiers, M., Golemiec, E., van der schors, R., van der Geest, L., li, K.W., Stiekema, W., and Liu, C,M. (2006). The CLV3/ESR motif of CLV3 is functionally independent from the non-conserved flanking sequences. Plant Physiol. 141, Fletcher, J.C., Brand, U., Running, M.P., Simon, R., and Meyerowitz, E.M. (1999). Signaling of cell fate decisions by CLAVATA3 in Arabidopsis shoot meristems. Science 283, Fritz-Laylin, L.K., Krishnamurthy, N., Tor, M., Sjolander, K.V., and Jones, J.D. (2005). Phylogenomic analysis of the receptor-like proteins of rice and arabidopsis. Plant Physiol. 138, Gao, B., Allen, R., Maier, T., Davis, E.L., Baum, T.J. and Hussey, R.S. (2001). Identification of putative parasitism genes expressed in the esophageal gland cells of the soybean cyst nematode Heterodera glycines. Mol. Plant Microbe Interact. 14, Hobe, M., Muller, R., Grunewald, M., Brand, U., and Simon, R. (2003). Loss of CLE40, a protein functionally equivalent to the stem cell restricting signal CLV3, enhances root waving in Arabidopsis. Dev. Genes Evol. 213, Grazzini, E., Guillon, G., Mouillac, B., and Zingg, H.H. (1998). Inhibition of oxytocin receptor function by direct binding of progesterone. Nature 392, Ito, Y., Nakanomyo, I., Motose, H., Iwamoto, K., Sawa, S., Dohmae, N., and Fukuda, H. (2006). Dodeca-CLE Peptides as Suppressors of Plant Stem Cell Differentiation. Science 313, Jeong, S., Trotochaud, A.E., and Clark, E. (1999). The Arabidopsis CLAVATA2 gene encodes a receptor-like protein required for the stability of the CLAVATA1 receptor-like kinase. Plant Cell 11,

30 Chapter 1 Kachroo, A., Schopfer, C.R., Nasrallah, M.E., and Nasrallah, J.B. (2001). Allelespecific receptor-ligand interactions in Brassica selfincompatibility. Science 293, Kayes, J.M., and Clark, S.E. (1998). CLAVATA2, a regulator of meristem and organ development in Arabidopsis. Development 125, Kende, H. and Zeevaart, J.A.D. (1997). The five classical plant hormones. Plant Cell 9, Kondo, T., Sawa, S., Kinoshita, A., Mizuno, S., Kakimoto, T., Fukuda, H., and Sakagami, Y. (2006). A Plant Peptide Encoded by CLV3 Identified by in Situ MALDI-TOF MS Analysis. Science 313, Krusell, L., Madsen, L.H., Sato, S., Aubert, G., Genua, A., Szczyglowski, K., Duc, G., Kaneko, T., Tabata, S., de Bruijn, F., Pajuelo, E., Sandal, N., and Stougaard, J. (2002). Shoot control of root development and nodulation is mediated by a receptor-like kinase. Nature 420, Laux, T., Mayer, K.F., Berger, J., and Jürgens, G. (1996). The WUSCHEL gene is required for shoot and floral meristem integrity in Arabidopsis. Development 122, Lemaitre, B., Nicolas. E., Michaut, L., Reichhart, J.M., and Hoffmann, J.A. (1996). The dorsoventral regulatory gene cassette spatzle/toll/cactus controls the potent antifungal response in Drosophila adults. Cell 86, Li, J., Wen, J., Lease, K.A., Doke, J.T., Tax, F.E., and Walker, J.C. (2002). BAK1, an Arabidopsis LRR receptor-like protein kinase, interacts with BRI1 and modulates brassinosteroid signaling. Cell 110, Matsubayashi, Y. and Sakagami, Y. (1996). Phytosulfokine, sulfated peptides that induce the proliferation of single mesophyll cells of Asparagus officinalis L. Proc. Natl. Acad. Sci. USA 93, Matsubayashi, Y., Ogawa, M., Morita, A. and Sakagami, Y. (2002). An LRR receptor-like kinase involved in perception of a peptide plant hormone, phytosulfokine. Science 296, Montoya, T., Nomura, T., Farrar, K., Kaneta, T., Yokota, T., Bishop, G.J. (2002). Cloning the tomato curl3 gene highlights the putative dual role of the leucine-rich repeat receptor kinase tbri1/sr160 in plant steroid hormone and peptide hormone signaling. Plant Cell 14,

31 General Overview Müller, R., Borghi, L., Kwiatkowska, D., Laufs, P., and Simon, R. (2006). Dynamic and compensatory responses of Arabidopsis shoot and floral meristems to CLV3 signaling. Plant Cell 18, Nam, K. H., and Li, J. (2002). BRI1/BAK1, a receptor kinase pair mediating brassinosteroid signaling. Cell 110, Nishimura, R., Hayashi, M., Wu, G.J., Kouchi, H., Imaizumi-Anraku, H., Murakami, Y., Kawasaki, S., Akao, S., Ohmori, M., Nagasawa, M., Harada, K., and Kawaguchi, M. (2002). HAR1 mediates systemic regulation of symbiotic organ development. Nature 420, Ni, J., and Clark, S.E. (2006). Evidence for functional conservation, sufficiency, and proteolytic processing of the CLAVATA3 CLE domain. Plant Physiol. 140, Opsahl-Ferstad, H.G., Le Deunff, E., Dumas, C. and Rogowsky, P.M., (1997). ZmEsr, a novel endosperm-specific gene expressed in a restricted region around the maize embryo. Plant J. 12, Olsen, A.N. and Skriver, K. (2003). Ligand mimicry? Plant-parasitic nematode polypeptide with similarity to CLAVATA3. TIPS, Pearce, G., Strydom, D., Johnson, S., and Ryan, C.A. (1991). A polypeptide from tomato leaves induces wound-inducible inhibitor proteins. Science 253, Pearce, G., Moura, D.S., Stratmann, J., and Ryan, C.A. (2001). Production of multiple plant hormones from a single polyprotein precursor. Nature 411, Reddy, G.V., and Meyerowitz, E.M. (2005). Stem-cell homeostasis and growth dynamics can be uncoupled in the Arabidopsis shoot apex. Science 310, Rojo, E., Sharma, V.K., Kovaleva, V., Raikhel, N.V., and Fletcher, J.C. (2002). CLV3 is localized to the extracellular space, where it activates the Arabidopsis CLAVATA stem cell signaling pathway. Plant Cell 14, Scheer, J.M. and Ryan, C.A. (2002). The systemin receptor SR160 from Lycopersicon peruvianum is a member of the LRR receptor kinase family. Proc. Natl. Acad. Sci. U.S.A. 99, Schoof, H., Lenhard, M., Haecker, A., Mayer, K.F.X., Jürgens, G., and Laux, T. (2000). The stem cell population of Arabidopsis shoot meristems is maintained by a regulatory loop between the CLAVATA and WUSCHEL genes. Cell 100,

32 Chapter 1 Schopfer, C.R., Nasrallah, M.E., and Nasrallah, J.B. (1999).The male determinant of selfincompatibility in Brassica. Science 286, Sharma, V.K., Ramirez, J. and Fletcher, J.C. (2003). The Arabidopsis CLV3-like (CLE) genes are expressed in diverse tissues and encode secreted proteins. Plant Mol. Biol. 51, Stein, J. C., Howlett. B., Boyes, D.C., Nasrallah, M.E., and Nasrallah, J.B. (1991). Molecular cloning of a putative receptor protein kinase gene encoded at the selfincompatibility locus of Brassica oleracea. Proc. Natl. Acad. Sci. USA 88, Stone, J.M., Collinge, M.A., Smith, R.D., Horn, M.A., and Walker, J.C. (1994). Interaction of a protein phosphatase with an Arabidopsis serine-threonine receptor kinase. Science 4, Stone, J.M., Trotochachaud, A.E., Walker, J.C., and Clark, S.E. (1998). Control of meristem development by CLAVATA1 receptor kinase and Kinase-Associated Protein Phosphatase interactions. Plant Physiol. 117, Strabala, T.J., O Donnell, P.J., Smit, A.M., Ampomah-Dwamena, C., Jane Martin, E., Netzler, N., Nieuwenhuizen, N.J., Quinn, B.D., Foote, H.C.C., and Hudson, K.R. (2006). Gain-of-function phenotypes of many CLAVATA3/ESR Genes, including four new family members, correlate with tandem variations in the conserved CLAVATA3/ESR domain. Plant Physiol. 140, Torii, K.U. (2005). Leucine-rich repeat receptor kinases in plants: structure, function, and signal transduction pathways. Int. Rev. of Cyt. 234, 1-46 Trotochaud, A.E., Hao, T., Wu, G., Yang, Z., and Clark, S.E. (1999). The CLAVATA1 receptor-like kinase requires CLAVATA3 for its assembly into a signaling complex that includes KAPP and a Rho-related protein. Plant Cell 11, Wang, X., Allen, R., Ding, X., Goellner, M., Maier, T., de Boer, J.M., Baum, T.J., Hussey, R.S. and Davis, E.L. (2001). Signal peptide-selection of cdna cloned directly from the esophageal gland cells of the soybean cyst nematode Heterodera glycines. Mol. Plant Microbe Interact. 14, Wang, X., Mitchum, M.G., Gao, B., Li, C., Diab, H., Baum, T.J., Hussey, R.S., and Davis, E.L. (2005). A parasitism gene from a plant-parasitic nematode with function similar to CLAVATA/ESR (CLE) of Arabidopsis thaliana. Mol. Plant Path. 6, Weigel, D. and Jürgens, G. (2002). Stem cells that make stems. Nature 14,

33 General Overview Williams, R.W., Wilson, J.M., and Meyerowitz, M. (1997). A possible role for kinaseassociated protein phospatase in the Arabidopsis CLAVATA1 signaling pathway. Proc. Natl. Adad. Sci. U.S.A. 94, Yang, H., Matsubayashi, Y., Nakamura, K. and Sakagami, Y. (1999). Oryza sativa PSK gene encodes a precursor of phytosulfokine-α, a sulfated peptide growth factor found in plants. Proc. Natl. Acad. Sci. USA 96, Yang, H., Matsubayashi, Y., Nakamura, K. and Sakagami, Y. (2001). Diversity of Arabidopsis genes encoding precursors for phytosulfokine, a peptide growth factor. Plant Physiol. 127, Yu, L.P., Simon, E.J., Trotochaud, A.E., and Clark, S.E. (2000). Poltergeist functions to regulate meristem development downstream of CLAVATA loci. Development 127, Yu, L.P., Miller, A.K., and Clark, S.E. (2003). Poltergeist encodes a protein phosphatase 2C that regulates CLAVATA pathway controlling stem cell identity at Arabidopsis shoor and flower meristems. Curr. Biol. 13,

34 Chapter 1 34

35 Chapter 2 Mis-expression of the CLV3/ESR-like gene CLE19 in Arabidopsis leads to a consumption of root meristem Martijn Fiers a, Gerd Hause b, Kim Boutilier a, Eva Casamitjana-Martinez c, Dolf Weijers d, Remko Offringa d, Lonneke van der Geest a, Michiel van Lookeren Campagne a,1, and Chun-Ming Liu a a Plant Research International, B.V., P.O. Box 16, 6700 AA Wageningen, The Netherlands b Biocenter, University of Halle, D Halle, Germany c Developmental Genetics, Utrecht University, Padualaan 8, 3584 CH Utrecht, The Netherlands. d Institute of Molecular Plant Sciences, Leiden University, Wassenaarseweg 64, Leiden, The Netherlands 1 Current Address: Bayer CropScience N.V., Lyon, France Published in Gene 327, (2004)

36 Chapter 2 36

37 Mis-expression of CLE19 Abstract Mild heat shock treatment (32 o C) of isolated B. napus microspores triggers a developmental switch from pollen maturation to embryo formation. This in vitro system was used to identify genes expressed in globular to heart-shape transition embryos. One of the genes isolated encodes a putative extra-cellular protein that exhibits high sequence similarity with the in silico identified CLV3/ESR-related19 polypeptide from Arabidopsis (AtCLE19) and was therefore named BnCLE19. BnCLE19 is expressed in the primordia of cotyledons, sepals and cauline leaves, and in some pericycle cells in the root maturation zone. Mis-expression of BnCLE19 or AtCLE19 in Arabidopsis under the control of the CaMV 35S promoter resulted in a dramatic consumption of the root meristem, the formations of pin-shaped pistils and vascular islands. These results imply a role of CLE19 in promoting cell differentiation or inhibiting cell division. Abbreviations: CaMV, cauliflower mosaic virus; CLV, CLAVATA; CLE, CLV3/ESR-related; BnCLE19, Brassica napus CLE19; GUS, β-glucuronidase; DD-RT-PCR, differential display reverse transcription PCR; Mr, relative molecular mass; kda, kilodalton; LLP: ligand-like protein; RT-PCR, reverse transcriptase PCR; SOL, SUPPRESSOR OF RCH1-CLE19; aa, amino acid; ORF, open reading frame; QC, quiescent center. 37

38 Chapter 2 1. Introduction Embryogenesis establishes the basic body plan for the adult plant, with the cotyledon(s), shoot apical meristem, hypocotyl and root meristem as the primary morphological domains along the apical-basal axis, and the epidermis, cortex and vascular bundles as the fundamental tissue structures along the radial axis (Jürgens et al., 1994). Several approaches have been used to identify genes involved in embryogenesis. Screening for Arabidopsis mutants defective at either the embryo development or the seed germination stage has identified genes that are essential for embryo development (Jürgens et al., 1994; Liu and Meinke, 1998). Another gene identification method that has been widely used is differential gene expression analysis. Molecular analysis on early zygotic embryos is, however, complicated by the fact that zygotic embryos are very small and encased by a mass of endosperm and maternal tissues. Microspore-derived haploid embryos (Heck et al., 1995; Custers et al., 2001; Boutilier et al., 2002) provide efficient alternatives to obtain a large amount of relatively synchronized embryo material. These studies have identified a number of genes involved in embryo development. The expression of some of these genes, such as AGL15, BBM1 and PEI1 etc., lack apparent organ or tissue specificity but these genes appear to play general roles in embryo development (Heck et al., 1995; Li and Thomas, 1998; Boutilier et al., 2002). Other embryo regulatory genes are only expressed in certain cells of the embryo, for example, the WUS, CUC1, 2 and 3, STM, UFO, CLV1 and CLV3 in the shoot apical meristem (Vroemen et al., 2003; for review, see Sharma and Fletcher, 2002; Groß-Hardt and Laux, 2003) and the PID and ANT gene in the cotyledons (Long and Barton, 1998; Christensen et al., 2000). Functional analysis of these genes has greatly improved our understanding of discrete aspects of plant embryo development. Nevertheless, we are still far away from a complete picture of the complex molecular machinery behind this process. In this report, we describe the use of the B. napus microspore embryogenesis system to identify genes expressed in embryos at the transition from the globular to heart-shape stage. One of the genes identified, named BnCLE19, is the orthologue of the in silico identified gene AtCLE19 from Arabidopsis (Cock and McCormick, 2001). AtCLE19 is a member of the 38

39 Mis-expression of CLE19 CLV3/ESR-related (CLE) genes that encode small putative extra-cellular proteins with a conserved C-terminal box shared by the CLV3 peptide ligand (Fletcher et al., 1999; Cock and McCormick, 2001). Northern blotting showed that BnCLE19 is expressed in globular to heartshaped embryos and young pistils. Detailed expression analysis was performed in transgenic Arabidopsis plants carrying either a BnCLE19 promoter::β-glucuronidase A (GUS) fusion construct or GAL4-UAS based transactivation constructs with a GFP-GUS fusion gene as a reporter, which showed that BnCLE19 is expressed in certain differentiating cells such as primordia of cotyledons, sepals and cauline leaves, and in some pericycle cells in the root maturation zone. Mis-expression of BnCLE19 or AtCLE19 in Arabidopsis under the control of the CaMV 35S promoter resulted in the formations of pin-shaped pistils and vascular islands, and a dramatic consumption of the root meristem without affecting lateral root induction. 2. Materials and methods 2.1 Plant materials and microspore embryogenesis culture Double-haploid Brassica napus L. cv. Topas DH4079 plants were maintained, and the isolation and culture of microspores were performed as previously described (Custers et al., 1994). Arabidopsis thaliana (ecotype C24 and Columbia) were grown in a greenhouse at 22 ± 3 o C, 15 hr daylight. 2.2 Differential display reverse transcription PCR (DD-RT-PCR) DD-RT-PCR was performed using the RNAmap Kit B (GeneHunter, Frederick, MD, USA) according to the manufacturer s recommendations. Total RNAs from freshly isolated microspores, non-embryogenic microspores cultured at 18 C (8 hours), embryogenic microspores cultured at 32 C (8 hours, 10 days, 16 days), and leaf tissue of B. napus were isolated as described by Ausubel et al. (1990) and DNase I treated using the MessageClean Kit (GeneHunter). Control material was obtained by heat shocking microspores at 41 C, a condition that does not lead to embryogenesis (Custers et al., 1994). The expression pattern of differentially expressed DD-RT-PCR clones was confirmed using Northern blot analysis. 39

40 Chapter RNA gel blot analyses, cdna isolation and RT-PCR For RNA gel blot analysis in B. napus, 10 µg of total RNAs from various tissues of were denatured with glyoxal prior to electrophoresis and blotted onto a Hybond-N + membrane, which was then hybridized overnight at 65 C and washed twice for 30 minutes at 65 C with 0.2xSSC and 0.5% (W/V) SDS. Equal loading was based on ethidium bromide staining. For Northern blot analysis of transgenic Arabidopsis, total RNA was isolated from roots excised from 2-week old seedlings grown on 1/2MS salts with 1% sucrose and 1.5% agar. One microgram of total RNA was loaded on gel and blotted as described above and hybridised with radioactively labelled BnCLE19 cdna. Actin was used as a control. For the isolation of the BnCLE19 cdna, an Uni-ZAP XR cdna library (Stratagene) was constructed using poly(a) + RNA from globular to heart-shape B. napus microspore-derived embryos. Approximately 10 6 plaques were screened under high-stringency conditions with the cdna fragment isolated from the DD-RT-PCR as a probe. The First Choice RLM-RACE Kit from Ambion (Cambridgeshire, UK) was used to amplify the full-length AtCLE19 transcript from Arabidopsis inflorescence. The obtained cdnas were cloned and sequenced. For RT-PCR, cdnas were prepared from total RNA of Arabidopsis isolated from cauline leaves, inflorescence, stems, roots, heart-shaped zygotic embryos, flower buds, petals and anthers using the RNeasy Plant Mini Kit (Qiagen, Valencia, USA), and then treated with DNase I (Invitrogen, Breda, NL). RT-PCR on actin (5 -GCGGTTTTCCCCAGTGTTGTTG- 3 ; 5 -TGCCTGGACCTGCTTCATCATACT-3 ) was used to quantify the cdnas. Two primers (5 -TCCCCATCAAACAAA-CAAAAAC-3 ; 5 -GATACACATATAATTGTTCT- TC-3 ) located at 5 and 3 UTR of CLE19 cdna, were used to determine the expression pattern of AtCLE19 in Arabidopsis. After optimisation, 35 and 41 cycles were used to amplify actin and AtCLE19, respectively. 2.4 Isolation of the BnCLE19 promoter and construction of GUS and GFP reporter construct The Universal Genome Walker Kit (Clontech, Palo Alto, USA) was used to isolate genomic DNA fragments upstream of the BnCLE19 coding region. The nested PCR was carried out using the adapter primer 1 supplied by the manufacturer and BnCLE19 specific primers: 5 - CCATTCTTCATCAGCAAACTCCGAAATGA-3 and 5 -CAGAAAAGAGGAAGCC- AATATCAAACTC-3. A fusion construct (pbncle19::gus) was made by the insertion of 40

41 Mis-expression of CLE19 BnCLE19 promoter (from 0 to 1,086 bp, GenBank accession no. AF343658) into a vector that has an intron-containing GUS gene and a nopaline synthase terminator. The expression cassette was then excised and inserted into the binary vector pbinplus (van Engelen et al., 1995). The derived construct was confirmed by sequencing and transferred to Agrobacterium tumefaciens C58C1PMP90, and then transformed to A. thaliana ecotype C24 using the floral dip method (Clough and Bent, 1998). The same promoter sequence was fused to an artificial transcription factor GAL4-VP16 gene (pbncle19::gal4-vp16) and delivered to Arabidopsis (Columbia) using the floral dip method. After homozygous lines were obtained, they were crossed to homozygous effector lines carrying a GFP-GUS fusion gene under the control of GAL4-VP16 binding sequence UAS (puas::gfp-gus). This is the so-called transactivation system (Benjamins et al., 2001). 2.5 Ectopic expression of BnCLE19 and AtCLE19 in Arabidopsis The full-length ORFs of BnCLE19 and AtCLE19 were cloned behind the double-enhanced CaMV 35S promoter (from -395 to 90 and from 525 to 1) and an AMV translational enhancer, as described by Datla et al. (1993). Transgenic plants of Arabidopsis (C24) were made and root development and geotropism were studied by growing the progeny seedlings on vertically cultured plates with the same medium mentioned above. 2.6 T-DNA insertion knockout of AtCLE19 in Arabidopsis Three T-DNA insertion populations, the Wisconsin Knockout Center ( the SAIL population at Syngenta ( and the GABI-Kat in Germany ( were searched either by PCR-based analysis or database mining to identify putative insertion lines. Insertions in the AtCLE19 gene were confirmed by PCR analysis and sequencing. Homozygous progeny plants were identified through plating seeds from each progeny plant on selection media in combination of PCR analyses using primers for the T-DNA and the AtCLE19 gene. 41

42 Chapter GUS assay, whole-mount clearing and cryo-electron microscopy A modified GUS assay was carried out by following the method described by Jefferson et al (1987) with 2 mm ferricyanide and ferrocyanide each in the reaction buffer. At least five independent transgenic lines were used for the GUS or GFP assay. To define the precise expression pattern of BnCLE19 during embryogenesis, zygotic embryos from transgenic plants were excised from seed and then stained for GUS activity. For the whole-mount clearing and observation of root and flower, samples was as prepared as described by Sabatini et al (1999). Dark-field microscopy was used to observe the vascular pattern in cleared flower samples. For detailed observation, GUS stained materials were fixed and embedded in paraffin and sectioned to 7µm before observed under a Nomarski microscope. For cryo-electron microscopy, plant materials were glued to copper stubs using conductive carbon glue and freeze immediately in liquid nitrogen. The samples were then transferred to a low temperature field emission scanning electron microscope (LT-FESEM, JSM 6300F, JEOL, Japan) equipped with an Oxford cryo-chamber. After a light coating with argon gas the samples were observed and pictures were taken with a digital camera. For confocal analysis of BnCLE19 promoter activity in Arabidopsis, F1 embryos carrying both transactivation constructs (pbncle19::gal4-vp16 and puas::gfp-gus) were excised from ovules at different developmental stages, and transferred directly to a glass slide with 5% glycerol solution, and observed under a confocal microscope. 3. Results 3.1 Isolation of differentially expressed genes from B. napus microspore-derived embryos Mild heat shock treatment (32 o C) of isolated B. napus microspores triggers a developmental switch from pollen maturation to embryo formation, while culturing at 18 o C leads to pollen maturation (Fig. 1A). Embryo samples at the globular to heart-shaped transition stage (10 days after culture) were analyzed using DD-RT-PCR for genes that are either up- or downregulated during this developmental change. Embryos at this stage change from a relatively unorganised globe to a mini-plant with bilateral symmetry, in which the major tissue and 42

43 Mis-expression of CLE19 organ primordia are established. Torpedo staged embryos (16 days after culture), leaf material and microspores treated at 18 o C or 41 o C (a condition that does not trigger embryogenesis) were used as controls. Figure 1. Isolation of BnCLE19 from microspore-derived embryos of B. napus by DD-RT-PCR. (A) Schematic presentation of the B. napus microspore culture system used to obtain a large amount of relatively synchronised embryos. Immature microspores at the late uni-cellular and early bi-cellular stages were isolated and cultured at different temperatures, leading to either embryogenesis (32 o C) or pollen maturation (18 o C). (B) A portion of the DD-RT-PCR auto-radiograph showing the expression of BnCLE19 (arrow) in B. napus embryos after 10 days (globular to heart-shaped stage) and 16 days (torpedo stage) of culture at 32 o C. BnCLE19 mrna was not detectable in freshly isolated microspores (T=0), in microspores cultured for 8 hr at 18 o C or 32 o C, or at 41 o C for 45 min, or in leaves. Genes showing developmentally regulated expression profiles were studied further (Custers et al., 2001). Here we present the characterization of one of these genes named BnCLE19 (originally named LLP1), which is up-regulated in embryos at the globular stage and onwards (Fig. 1B, indicated by an arrow). 3.2 cdna Isolation The BnCLE19 DD-RT-PCR fragment was used to obtain a full-length transcript from a cdna library prepared from 10-day old microspore embryo cultures. A 417 bp cdna clone encoding a 74-aa putative extra-cellular peptide was identified (Fig. 2A). Two stop codons located upstream of the longest open reading frame (ORF), suggested that a full-length ORF was obtained (GenBank accession no. AF343656). Queries with the BnCLE19 sequence to 43

44 Chapter 2 the Arabidopsis genome revealed similarity to AtCLE19, which was identified in silico through database mining (Fig. 2B; Cock and McCormick, 2001). However, isolation of fulllength capped cdnas from Arabidopsis showed that the transcription start site of AtCLE19 (GenBank accession no. AF343657) begins 72 bp after the previously predicted translational start codon (ATG), resulting in a polypeptide with the same length as the BnCLE19, but 58 amino acids shorter than the previously annotated AtCLE19 (Cock and McCormick, 2001). The new annotation of AtCLE19 led to an increased probability that the predicted protein contains a signal peptide (from 37.5% to 99.8%, Fig. 2A, in bold). Figure 2. BnCLE19 encodes a CLV3-like extra-cellular protein. (A) The BnCLE19 cdna was isolated from a cdna library prepared from globular to heart-shape embryos obtained from B. napus microspore culture. The Two stop codons before the longest ORF are marked with triple asterisks (***). The DD-RT-PCR fragment (underlined) represents a partial BnCLE19 cdna sequence. The signal peptide is shown in bold. An asterisk indicates the functional stop codon. (B) Alignment of BnCLE19 with CLE19, CLE21, CLV3, ZmESR3, CLE26, and HgCLE. The conserved identical amino acids are shaded in black and similar amino acids in grey. The putative signal peptide cleavage site for BnCLE19 and CLE19 is marked with an arrowhead and the conserved CLE box is framed. Neither BnCLE19 nor AtCLE19 contain introns. Within the 225 bp coding region, they share 83% and 68% sequence identity at the DNA and protein levels, respectively. Therefore, AtCLE19 in Arabidopsis should be the orthologue of BnCLE19 in B. napus. 44

45 Mis-expression of CLE19 CLE19 is a member of the CLV3/ESR-related (CLE) family of genes that encode polypeptides with several common features. All of the CLE genes encode small proteins (average Mr 7.7 kda), that have an N-terminal putative signal peptide or a membrane anchor, and contain a conserved 14-AA motif (KRXVPXGPNPLHNR) located at or near their C-termini (Fig. 2B, termed CLE box accordingly; Cock and McCormick, 2001; Sharma et al., 2003). There are 26 CLE members in Arabidopsis genome (Cock and McCormick, 2001; Sharma et al., 2003). RT-PCR has shown that all of them except one (CLE26) are expressed in one or more tissues during development (Sharma et al., 2003). Some CLEs, such as CLV3, ZmESR3, CLE26 and HgCLE, have a 10 to 53-amino acid extension after the CLE box (Fig. 2B; Cock and McCormick, 2001). CLV3 is an extra-cellular peptide ligand and is expressed in the putative stem cells of shoot and floral meristems, and functions in restricting the number of stem cells through its interaction with the CLV1/CLV2 receptor complex and the WUS transcription factor (Fletcher at al., 1999; Schoof et al., 2000; Brand et al., 2000). ZmESR3 encodes a secreted protein and is expressed in a small region of endosperm surrounding the maize embryo (Opsahl-Ferstad et al., 1997). Recently, a CLE protein (HgCLE, Fig. 2B) has also been identified in a soybean cyst nematode, suggesting that the parasitic nematode may have co-opted the plant signalling peptide to enable interaction with the host cells (Olsen and Skriver, 2003). No matching AtCLE19 cdna was found among 113,330 ESTs available in the Arabidopsis database. An explanation could be that most cdna libraries are constructed using sizefractionated cdna. Genes such as AtCLE19 with short transcripts may only be present in very low abundance in these libraries. 3.3 The expression of BnCLE19 Northern blot analysis of RNA from B. napus tissues demonstrated relatively high levels of BnCLE19 expression in globular to heart shape embryos and in young flower buds (1-4 mm in length). In 5 mm long flower buds, the expression was only detected in pistils (Fig. 3A). To determine the detailed spatial pattern of BnCLE19 expression, a 1086 bp genomic sequence (GenBank accession no. AF343658) upstream of the BnCLE19 ORF was isolated from B. napus and fused to the GAL4-VP16 transcription factor gene (pbncle19:gal4- VP16) that consequently drove the expression of the puas::gfp-gus fusion construct after 45

46 Chapter 2 crossing homozygous lines carrying these individual constructs. F1 embryos were analysed for GFP pattern under a confocal microscope to determine the BnCLE19 promoter activity during embryogenesis. As showed in Fig. 3B-E, GFP expression was first detected in triangular embryos, in a single layer of protoderm cells covering the cotyledon primordia and the shoot apical meristem (Fig. 3B, indicated by a curved line along the cells). From the heartshape to the early torpedo stage, GFP was only expressed in the epidermal cells covering the newly formed cotyledons, but not in those ones along the shoot apical meristem (Fig. 3, C to E). At the bent-cotyledon stage the expression was shifted to the basal region of the cotyledons and switched off completely in cotyledon staged embryos (data not shown). Figure 3 *. Expression analysis of BnCLE19. (A) Northern blot analysis of BnCLE19 expression in B. napus. Ten µg of total RNA isolated from various organs and tissues was separated, blotted and probed with the BnCLE19 cdna. BnCLE19 is expressed in developing embryos (globular to heart stage embryos from microspore culture), flower buds and pistils. Petals, anthers and pistils were obtained from 5 mm flower buds containing tri-nucleate pollen. (B-E) Expression of BnCLE19 in zygotic embryos, as shown by the confocal microscopic observation of F1 embryos carrying both pbncle19:gal4-vp16) and puas::gfp-gus constructs. In tri-angular stage embryos, GFP expression was observed in the epidermal cell layer that covers the cotyledon primordia and the shoot apical meristem (B). In the heart- to torpedo-shaped embryos, the GFP signal was only observed in the epidermal cells covering the cotyledon primordia (C-E). The scale bars in (B) represents 10 µm for (B and C), and in (C) represents 25 µm for (D and E). *) Colour figure in appendix, page III 46

47 Mis-expression of CLE19 Figure 4 *. Histological analysis of post-embryonic GUS expression in pbncle19-gus transgenic Arabidopsis. The photographs correspond to whole-mount materials cleared with Hoyer s solution (A-C, F and H) and paraffin sections (D, E and G). (A-D) In roots, pbncle19::gus is expressed in the root hair region and the differentiation zone above (B-D), but not in the root meristem (D), nor in the newly formed lateral root (B), nor in older roots with secondary thickenings (A). The scale bar in (D) represents 100 µm for A-D. (E) Transverse section of a root in the root hair region, showing GUS expression in 2 to 3 pericycle cells (arrowheads) facing the protoxylem poles. The tissue deformation was caused by the acetone pre-fixation used in the GUS assay. The scale bar represents 100 µm. (F) pbncle19::gus expression was seen in the periphery of meristems in the axillary bud, where the cauline leaves will form. The scale bar represents 25 µm. (G and H) During flower development, GUS expression was seen in the sepal primordia in stage 2-5 flower buds, but not in the main inflorescence meristem (marked with an asterisk). The scale bars represent 40 µm. (I) In a stage 10 flower bud, GUS expression was seen at the top of the pistil, where the stigma hairs will form. The scale bar represents 150 µm. *) Colour figure in appendix, page IV 47

48 Chapter 2 GUS staining was not observed in the roots with secondary thickening, nor in the root meristem (Fig. 4, A and D), hypocotyl and cotyledons (data not shown). In radial sections, GUS expression was observed in a few pericycle cells facing the protoxylem poles (Fig. 4C and E). The expression in lateral roots was comparable to that in the main root and could only be observed after root hairs emerged (data not shown). In above ground tissues, the first detectable GUS expression was seen in the periphery of the axillary meristems (Fig. 4F). In expanding leaves faint GUS expression was observed in the abaxial side of the petioles (data not shown), which vanished in fully expanded leaves (data not shown). No GUS expression was observed in the central domain of the meristem (Fig. 4F- H). During floral development, GUS was expressed in the sepal primordia during floral stage 2 to 5 (Fig. 4G and H; see Smyth et al., 1990, for the floral stage definitions). GUS expression was restricted to the stigma in flower buds between stage 7 and stage 10 (Fig. 4I), and switched off completely shortly before the flower opened. Among 6 pbncle19::gus and 5 F1 transactivation lines tested, the expression pattern was consistent among different lines although the intensity varied slightly. 3.4 AtCLE19 is expressed in a similar manner as BnCLE19 A conserved regulatory function for the AtCLE19 and BnCLE19 promoters is expected based on the nucleotide similarity between these two promoters. BnCLE19 and AtCLE19 exhibit 81% sequence similarity in a 415 bp region upstream of the start codon, and a TATA box (TATAAA) was identified for both genes at 128 bp before the start of the ORF. We used RT- PCR to determine the expression pattern of AtCLE19. The results showed that, AtCLE19 in Arabidopsis is strongly expressed in heart-shape embryos and young flower buds, weakly expressed in inflorescence, faintly expressed in leaves and roots (visible on gel only when 5 times more cdna was used, data not shown), and is not detectable in stems, petals and anthers (Fig. 5). This expression pattern is consistent with the expression pattern of BnCLE19, but not the same as reported by Sharma et al (2003), who have showed that AtCLE19 is also expressed in pollen. 48

49 Mis-expression of CLE19 Figure 5. Expression analysis (A-B) AtCLE19 is expressed in a similar manner as BnCLE19, as shown by RT-PCR analysis (top panel). AtCLE19 (A) is strongly expressed in heart-shape embryos (e) and flower buds (f), intermediate in inflorescence (i) and weakly expressed in leaves (l) and roots (r) (leaves and roots only visible when 5 times more cdna was used), and not detectable in stems (s), petal (p) and anther (a), c: chromosomal DNA, m: molecular weight marker. Actin was used as a control (B). The numbers at the side denote the molecular size in Kb. (C-D) Expression of BnCLE19 (C) and actin (D) in the roots excised from T2 transgenic and non-transgenic Arabidopsis seedlings (C24). One microgram of total RNAs was used for the Northern blot analysis. 1L: seedlings with long roots segregated from transgenic line #1; 1S: seedlings with short roots from line #1; 2S: seedlings with short roots from line #2; 3L: wildtype-looking seedlings (long roots) segregated from transgenic line #3; 3S: seedlings from line #3, with short roots and pin-shaped pistils; 4S: seedlings from line #4 with short roots; 5L: seedlings from transgenic line #5, with long roots. WT: wildtype C24 seedlings. Note that elevated expression of BnCLE19 in transgenic Arabidopsis roots correlates with the short root phenotype. Line #3 and #4 showed pin-shaped pistils in flowers but line #1 and #2 did not. 3.5 Ectopic mis-expression of BnCLE19 and AtCLE19 in Arabidopsis under the control of CaMV 35S promoter A double enhanced CaMV 35S promoter with an AMV translational enhancer (Datla et al., 1993) was used to drive the expression of the BnCLE19 (and AtCLE19 cdna (p35s::atcle19) in Arabidopsis (C24). Thirteen out of 75 p35s::bncle19 transformants and 2 out of 24 p35s::bncle19 transformants exhibited short roots, slow growth, late flowering and pin-shaped pistil phenotypes. Many other lines showed short root phenotype 49

50 Chapter 2 but without pin-shaped pistils. The frequency of flowers with pin-shaped pistils varies from 5% to 50% in different transgenic lines, which is relatively consistent though generations. Bolting did not occur until days after seeds were planted, instead of after 20 days in the wildtype. In contrast to one paraclade normally produced from each axillary bud in wildtypes, multiple ones (up to 7) were often formed sequentially in the mis-expression lines, particularly in the axils of cauline leaves. No embryo lethals were observed in these lines. Genetic analysis indicated that their phenotypes were inherited as a dominant trait in Mendelian fashion and linked with the transgene (Fig. 6A). The phenotype persists through generations (4 generations tested). Northern blot analysis of 5 transgenic lines carrying the BnCLE19 construct showed that the short root phenotype is linked to the elevated expression of BnCLE19 gene in the roots. As showed in Fig. 5C, all seedlings with a short root phenotype (1S, 2S, 3S and 4S) showed high level of expression of BnCLE19. Neither the wildtype (lane WT), nor the transgenic plants without any phenotype (lane 5L), or the normal long root plants (lanes 1L and 3L) segregated from hemizygous parental plants gave detectable expression when 1µg of root total RNA was analyzed on gel (Fig. 5, C and D). As compared to those lines with only short root phenotypes (line #1 and #2), lines exhibited both the short root and the pin-shaped pistil phenotypes (line #3 and #4) seems do not have higher expression of BnCLE19 in roots (Fig. 5C). Root geotropism (Fig. 6A) and lateral root initiation did not seem affected by mis-expression of BnCLE19 and AtCLE19 genes. Root hairs were formed almost to the tip of the roots (Fig. 6B). Tissue clearing, followed by Nomarski microscopy of the roots from p35s::bncle19 transgenic plants showed that root meristematic tissue was gradually consumed during root growth and development (Fig. 6C-G). As compared to the wildtype (Fig. 6C), in BnCLE19 mis-expression plants the root meristem zone became shorter, and was followed immediately by the formation of highly vacuolated cells that were typically seen in the root hair region (Fig. 6D, 7 days after plating). At this stage the quiescent center (QC) was still recognizable (indicated by an arrow). Ten days after germination, only a small number of meristematic cells were present in the root tip, but the QC was still visible (Fig. 6E). Tissue sections of roots at this stage and staining with toluidine blue revealed the existence of well differentiated cells in the meristem region (Fig. 6G). 50

51 Mis-expression of CLE19 Figure 6. Ectopic expression of BnCLE19 in Arabidopsis led to a consumption of root meristems. The root meristems are marked with a square bracket in C to E, and the QC cells are marked with an arrow. (A) Two-week old progeny seedlings from a hemizygous p35s::bncle19 parental plant were growing on a vertically culture plate, showing segregation for transgenic seedlings with a dominant short root phenotype and three seedlings being wildtype. Note that root geotropism was not affected in these short root seedlings. (B) A close-up observation of a p35s::bncle19 root with the short root phenotype, showing the formation of root hairs to the root tip. (C) A whole-mount, 7-day old wildtype root of Arabidopsis observed under Nomarski optics. Note the gradual cell size changes from the root meristem to the elongation zone. No obvious pattern change occurred in the following week in the wildtype. (D) A primary root from a 7-day old p35s::bncle19 Arabidopsis line showing the reduced length of the root meristem. Note the evident cell size increase above the short meristem. (E) Primary root from a 10-day old p35s::bncle19 Arabidopsis line. Note the greatly reduced root meristem and the highly vacuolated cells above the meristem and the xylem elements (indicated by an arrowhead). The QC was still recognisable at this stage. (F) Primary root from a 2-week old p35s::bncle19 seedling showing that the root meristem has disappeared completely. Note that the xylem elements (indicated by an arrowhead) reached the central cell zone. No QC can be found at this stage. (G) Longitudinal section of a root from a 10-day old p35s::bncle19 seedling, showing the enlarged and differentiated cells in the root meristem. The scale bar in (C) represents 100 µm for (C), 150 µm for (D to F), 200 µm for (G). 51

52 Chapter 2 Both the root meristematic cells and the QC disappeared in 2-week old p35s::bncle19 seedlings (Fig. 6F). All the cells in this region became highly vacuolated and exhibited a thickening of their cell walls. Staining of starch showed that the columella identity was present even after the root meristem was fully differentiated (data not shown). Xylem elements reached the central cell region (Fig. 6F, indicated by an arrowhead). The expression of BnCLE19 under the control of the 35S promoter appears to have no influence on embryonic root formation and lateral root induction. The same phenotype was observed in the transgenic plants carrying the p35s::atcle19 construct (data not shown), and when AtCLE19 was expressed in Arabidopsis under the control of a root meristem-specific promoter, RCH1 (Casamitjana-Martinez et al., 2003). Figure 7. The formation of pin-shaped pistils in Arabidopsis mis-expressing BnCLE19. (A) and (B) are photos from cryo-electron microscopy. (A) Wildtype open flower with some outer whorls removed to show the well-developed pistil. (B) An open flower from p35s::bncle19 plants treated in the same way as in (A) to show the pin-shaped pistil. (C) Whole-mount clearing of an open flower from p35s::bncle19 plant showing the pin-shaped pistil. Neither ovules, nor xylem were observed in such pistils. Note the stigma formed at the top of the pistil. The scale bars represent 200µm for (A) and (B), 100µm for (C). 52

53 Mis-expression of CLE19 The pin-shaped pistils observed in BnCLE19 and AtCLE19 mis-expression plants have a filamentary structure that did not contain carpels and ovules (Fig. 7). In wild-type plants ca. 150 bulbous cells formed at the top of the stigma (Fig. 7A; Sessions and Zambryski, 1995), whereas only such cells could be observed on the pin-shaped pistil (Fig. 7, B and C). The region below the stigma, most likely corresponding to the style, had 6-cell layers across the median section, which is much narrower than wild-type styles that have more than 30 cell layers. No vascular bundle was observed within these pin-shaped pistils. These pistils are quite different from those of ettin mutants in which the ovary is reduced but the style is expanded longitudinally. Occasionally, flowers without pistils were also observed in some transgenic lines. This phenotype could be the consequence of a consumption of floral meristem in whorl 4. Furthermore, defective vascular development was observed in flowers of BnCLE19 and AtCLE19 mis-expression lines. In the wildtype inflorescence, vascular bundles are formed at stage 9 by extension from the main stem up to pedicels and then to floral organs (Fig. 8A). Xylem elements in the flower buds are established first in sepals and followed sequentially by pistils, stamens and petals, resulting in a complete vascular network (Fig. 8C). In p35s::bncle19 flower buds, regional vascular formation without connecting to the main stems was often observed (Fig. 8D, indicated by arrowheads). These xylem elements ended at the receptacle region of the flower. The vascular connection failure seems to be associated with the formation of pin-shaped pistils, since this phenomenon was not observed in flowers with normal pistils. However, local xylem formation in sepals and petals, as vascular islands, was observed in normal flowers and flowers with a pin-shaped pistil (Fig. 8D, indicated by arrows). The same results were observed in p35s::atcle19 transgenic plants (data not shown). 53

54 Chapter 2 Figure 8. Mis-expression of BnCLE19 in Arabidopsis leads to a failure of xylem connections in flower buds. (A) Nomarski observation of a whole mount stage 9 wild-type flower bud from the wildtype, showing the extension of xylem elements (indicated by arrowheads) from the main stem towards the flower bud. (B) GUS staining of flower buds from a hybrid plant obtained from a cross between homozygous DR5::GUS and p35s::bncle19 lines. The GUS staining was seen only in the anthers with uni-cellular microspores, which is the same as that in the DR5::GUS parental line. (C) Dark-field observation showing the well-developed xylem network in a wild-type flower bud. (D) Dark-field observation showing that mis-expression of BnCLE19 leads to the formation of disconnected xylem elements in flower buds (indicated by arrowheads). Note also the vascular islands formed (indicated by arrows) in these flower buds. The scale bars represent 100 µm for (A), 150µm for (B) and 400µm in (C) and (D). Since vascular differentiation is known to be associated with auxin flux (Aloni, 1995), we examined if the formation of vascular islands was caused by local accumulation of auxin. As shown previously, the DR5::GUS reporter construct can be used to monitor auxin distribution (Sabatini et al., 1999). In wild-type Arabidopsis roots, the highest GUS staining is observed in the QC cells of the root meristem (Sabatini et al., 1999). In the upper part of the plant, we observed that GUS staining was mainly in the anthers (Fig. 8B), in particular, the pollen 54

55 Mis-expression of CLE19 grains after the uni-nucleate stage. We crossed the DR5::GUS line with the p35s::bncle19 lines, and analyzed the progeny plants for DR5::GUS expression. In roots we observed that the high level of GUS staining in the QC cells persists right before the meristematic cells disappeared (data not shown), which is the same as the expression of AtCLE19 under the control of RCH1 promoter (Casamitjana-Martinez et al., 2003). In the flowers the same GUS pattern was also observed, regardless of whether the flower had a normal or pin-shaped pistil. This suggested that either BnCLE19 functions downstream of the auxin signaling or that it acts through an auxin-independent pathway for promoting vascular development. 3.6 T-DNA insertion knockout of AtCLE19 To further analyze the function of the CLE19 genes, T-DNA knockout lines were identified in different Arabidopsis insertion populations. In total three T-DNA insertions were obtained in the AtCLE19 locus, with the insertion sites located at 218 bp (Wisconsin line), -40 bp (GABI-Kat line) and +130 bp (SAIL line, these numbers are in relation to the ATG of AtCLE19). Homozygous insertion lines were obtained from each transgenic line, however no visible phenotypes were observed as compared to the sibling heterozygous plants. The reasons for a lack of phenotype could either be that AtCLE19 plays a minor role in plant development or that other CLE genes function redundantly with CLE19. Given the strong phenotypes observed in plants mis-expressing BnCLE19 gene under the control of 35S promoter and root-specific expression of AtCLE19 under the RCH1 promoter (Casamitjana- Martinez et al., 2003) in Arabidopsis, the first possibility is unlikely. With respect to the redundancy of AtCLE19, AtCLE21 shows the highest overall similarity with AtCLE19. These two polypeptides exhibited 36.5% overall sequence identity, with only one amino acid difference in the CLE box. It is possible that AtCLE21 may complement the AtCLE19 function in the insertion lines. 4. Discussion In the present research, we are interested in identifying genes expressed in embryos during the transition from globular to heart-shape stage of development. This is the period during which the embryo changes from a relatively unorganised globe to a mini-plant with bilateral 55

56 Chapter 2 symmetry in which the major tissue and organ primordia are established. BnCLE19 gene was isolated from B. napus microspore-derived embryos using the DD-RT-PCR technique. It encodes a putative extra-cellular protein with sequence similarity to the CLE19 polypeptide from Arabidopsis (Cock and McCormick, 2001). The CLE family consists of 26 genes in the Arabidopsis genome (Cock and McCormick, 2001; Sharma et al., 2003). All of them except one (CLE26) are expressed in diverse tissues of Arabidopsis (Sharma et al., 2003). The encoded CLE polypeptides are characterized by their small sizes, an N-terminal secretion signal and a conserved C-terminal CLE motif shared by CLV3 and ZmESR3 (Cock and McCormick, 2001). CLV3, expressed in the stem cells of the shoot apical meristem, encodes an inter-cellular peptide ligand that acts through the CLV1/2 receptor complex to impose a negative signal to the stem cells, in balance with the stem cell-promoting signal generated by the WUS transcription factor expressed in the underlying organizing center (Fletcher et al., 1999; Brand et al., 2000; Schoof et al., 2000; Rojo et al, 2002; Lenhard and Laux, 2003). The balance between CLV3 and WUS signals permits the onset of cell differentiation in the periphery and, at the same time, maintains a stable number of stem cells in the center of the shoot meristem (Lenhard and Laux, 2003). ZmESR3 of maize is expressed in a small region of endosperm surrounding the embryo, but its function is not clear yet (Opsahl-Ferstad et al., 1997). CLE40 can functionally complement CLV3 when expressed under the control of CLV3 promoter (Hobe et al., 2003). During embryogenesis, the expression of BnCLE19 is associated with cotyledon development, which was first seen at the top of the late globular embryos prior to cotyledon initiation, at about the same time as the PINOID gene (Christensen et al., 2000). At the heart-shape and early torpedo stage, BnCLE19 expression is restricted to the cotyledon primordia. Later the expression is narrowed down to the edge of the cotyledons and switched off completely in mature embryos. The expression pattern contrasts with CLV3, which is expressed in the shoot apical meristem (Fletcher et al., 1999) and several other cotyledon-expressed genes, such as FIL (Siegfried et al., 1999), ANT (Long and Barton, 1998), PID (Christensen et al., 2000) and REV (Otsuga et al., 2001). Like all known embryo development-related genes, BnCLE19 was also expressed in post-embryo development. After seed germination, BnCLE19 is expressed in the periphery of the axillary meristem, in sepal primordia, young stigma and in some pericycle cells at the root hair region. The common feature among these cells is the 56

57 Mis-expression of CLE19 intermediate state of differentiation, which implies a role of BnCLE19 in organogenesis or cell differentiation. Mis-expression of BnCLE19 and AtCLE19 in Arabidopsis leads to a premature cell differentiation in several tissues. Firstly, p35s::bncle19 and p35s::atcle19 plants exhibited a short root phenotype, in which the primary root meristem was fully differentiated in 12 day-old seedlings. Root hairs were formed to the tip of the roots. The QC cells disappeared at about the same time when the meristem was fully differentiated, suggesting that the consumption of meristem may not be caused by the differentiation of QC. This is a phenotype shared by CLE40 and CLV3 mis-expressions (Hobe et al., 2003). Secondly, pinshaped pistils (pistil without carpel and ovule) were formed in plants mis-expressing BnCLE19 and AtCLE19, which could be the consequence of a pre-mature consumption of meristematic cells in the whorl 4 of these flowers. This phenotype is the opposite of the clv3 mutant phenotype in which an increased number of carpels were observed. The third, flowers with disconnected vascular bundles were observed in p35s::bncle19 and p35s::atcle19 plants, which could be a premature differentiation of the xylem. As such, we hypothesize that, as a differentiation signal, the CLE19 polypeptide may be perceived by a receptor kinase complex in roots and floral organs, which then triggers pre-mature cell differentiation. Such a receptor complex may not be available in many meristematic cells, for example, the initiation phase of the root meristem since both embryonic root formation and the lateral root induction were not affected by the transgene. Based on the current evidence we cannot exclude a second possibility that the primary function of CLE19 is to inhibit cell division rather than to promote cell differentiation or organogenesis. Although the mis-expression of CLE19 led to a strong phenotype in root and flower development in Arabidopsis, T-DNA insertions of in the coding and the 5 UTR regions showed no phenotype. This implies that either CLE19 plays a minor role in normal development, or additional genes such as CLE21 that has the highest sequence similarity with CLE19 can complement the CLE19 mutation. Using the short root phenotype generated by mis-expression of BnCLE19, Casamitjana- Martinez et al. (2003) carried out a mutant screen for repressors of the root meristems consumption phenotype. A root meristem-specific promoter prch1 was used to drive the 57

58 Chapter 2 expression of CLE19 and a phenotype identical to that seen in p35s::bncle19 roots was observed in the transgenic lines. Two genetic loci, SOL (SUPPRESSOR OF RCH1-LLP1) 1 and SOL2, were identified. It is interesting to note that the sol2 mutant, similar to clavata, also exhibits an increased number of carpels which is the opposite to the pin-shaped pistil phenotype, suggesting that SOL2 might be involved in the CLV signaling pathway to restrict the meristem size. The SOL1 gene was cloned by chromosome walking, and found to encode a putative Zn 2+ -carboxypeptidase. Casamitjana-Martinez et al. (2003) proposed that the CLE19 polypeptide may be further processed to produce a functional peptide ligand. If further processing of BnCLE19 does indeed occur, then it would also explain why our immunological studies of BnCLE19 using polyclonal antiserum have failed to detect any BnCLE19 signal in the inflorescence of cauliflower and B. napus (data not shown). The same may apply to the CLV3 ligand, since to date no one has been able to reproducibly detect the CLV3 protein in Arabidopsis (Nishihama et al., 2003; Lenhard and Laux, 2003). Sytemin and phytosulfokine are examples of peptide ligands derived from larger pre-proteins (Pearce et al., 1991; Matsubayashi and Sakagami, 1996). The question that then remains to be answered is whether CLE proteins are indeed processed further into even smaller peptides, and if so, which peptide fragments are the functional ones. Acknowledgements We thank Ben Scheres, Mark Aarts, and Jan Custers for fruitful discussions; Sylvia Climent for providing the over-expression vector; Jan Cordewener and Hans Jansen for assisting with the DD-RT-PCR analysis, Adriaan van Aelst for assisting cryo-electron microscopy, and Haiying Zhang for transactivation analysis. We also thank Trevor Wang (John Innes Centre) for critical reading the manuscript. We appreciate the excellent services from the Wisconsin Arabidopsis Knockout Center, the Salk Institute, the TMRI (Syngenta), GABI-Knat (Germany) and the Nortingham Arabidopsis Stock Centre for providing the T-DNA insertion lines. This work was supported in part by the Dutch Ministry of Agriculture, Nature Management and Fisheries (DWK281/392) and the EU-SIME project (Bio4-CT ). 58

59 Mis-expression of CLE19 References Aloni, R., Differentation of vascular tissues. Annu. Rev. Plant Physiol. 38, Ausubel, F.M., Brent, R., Kingston, R.E., Moore, D.D., Seidman, J.G., Smith, J.A., Struhl, K., Current Protocols in Molecular Biology (New York: John Willey and Sons). Boutilier, K., Offringa, R., Sharma, V.K., Kieft, H., Ouellet, T., Zhang, L., Hattori, J., Liu, C.M., Van Lammeren, A.A., Miki, B.L., Custers, J.B., Van Lookeren Campagne, M.M., Ectopic expression of BABY BOOM triggers a conversion from vegetative to embryonic growth. Plant Cell 14, Brand, U., Fletcher, J.C., Hobe, M., Meyerowitz, E.M., Simon, R., Dependence of stem cell fate in Arabidopsis on a feedback loop regulated by CLV3 activity. Science 289, Casamitjana-Martinez, E., Hofhuis, H.F., Xu J., Liu, C.M., Heidstra, R., Scheres, B., Root-specific CLE19 over-expression and the sol1 and sol2 suppressors implicate a CLV-like pathway in the control of Arabidopsis root meristem maintainance. Curr. Biol., in press. Christensen, S.K., Dagenais, N., Chory, J., Weigel, D., Regulation of auxin response by the protein kinase PINOID. Cell 100, Clough, S.J., Bent, A.F., Floral dip: a simplified method for Agrobacterium-mediated transformation of Arabidopsis thaliana. Plant J. 16, Cock, J.M., McCormick, S., A large family of genes that share homology with CLAVATA3. Plant Physiol. 126, Custers, J.B.M., Cordewener, J.H.G., Fiers, M.A., Maassen, B.T.H., van Lookeren Campagne M.M., Liu, C.M., Androgenesis in Brassica: A model system to study the initiation of plant embryogenesis. In Current Trends in the Embryology of Angiosperm, S.S. Bhojwani and W.Y. Soh (Eds), pp Custers, J.M.B. Cordewener, J.H.G., Nollen, Y., Dons, H.J.M., van Lookeren Campagne M.M., Temperature controls both gametophytic and sporophytic development in microspore cultures of Brassica napus. Plant Cell Rep. 13, Datla, S.S.R., Bekkaoui, F., Hammerlindl, J.K., Pilate, G., Dunstan, D.I., Crosby, W.L., Improved high-level constitutive foreign gene expression in plants using an AMV RNA4 untranslated leader sequence. Plant Sci. 94,

60 Chapter 2 Fletcher, J.C., Brand, U., Running, M.P., Simon, R., Meyerowitz, E.M., Signaling of cell fate decisions by CLAVATA3 in Arabidopsis shoot meristems. Science 283, Groß-Hardt, R., Laux, T., Stem cell regulation in the shoot meristem. J. Cell Sci. 116, Heck, G.R., Perry, S.E., Nichols, K.W., Fernandez, D.E., AGL15, a MADS domain protein expressed in developing embryos. Plant Cell 7, Hobe M., Muller R., Grunewald M., Brand U., Simon R., Loss of CLE40, a protein functionally equivalent to the stem cell restricting signal CLV3, enhances root waving in Arabidopsis. Dev Genes Evol. 213, Jefferson, R.A., Kavanagh, T.A., Bevan, M.W., GUS fusions: p-glucuronidase as a sensitive and versatile gene fusion marker in higher plants. EMBO J. 6, Jürgens, G., Torres, R.A., Berleth, T., Embryogenic pattern formation in flowering plants. Annu. Rev. Genet. 28, Lenhard, M., Laux, T., Stem cell homeostasis in the Arabidopsis shoot meristem is regulated by intercellular movement of CLAVATA3 and its sequestration by CLAVATA1. Development 130, Li, Z., Thomas, T.L., PEI1, an embryo-specific zinc finger protein gene required for heart-stage embryo formation in Arabidopsis. Plant Cell 10, Liu, C.M., Meinke, D.W., The titan mutants of Arabidopsis are disrupted in mitosis and cell cycle control during seed development. Plant J. 16, Long, J.A., Barton, M.K., The development of apical embryonic pattern in Arabidopsis. Development 125, Matsubayashi, Y., Sakagami, Y., Phytosulfokine, sulfated peptides that induce the proliferation of single mesophyll cells of Asparagus officinalis L. Proc. Natl. Acad. Sci. USA. 93, Nishihama, R., Jeong, S., DeYoung, B., Clark S.E., Retraction. Science 300, Olsen, A.N., Skriver, K. (2003) Ligand mimicry? Plant-parasitic nematode polypeptide with similarity to CLAVATA3. Trends Plant Sci. 8, Opsahl-Ferstad, H.G., Le Deunff, E., Dumas, C. and Rogowsky, P.M., ZmEsr, a novel endosperm-specific gene expressed in a restricted region around the maize embryo. Plant J. 12,

61 Mis-expression of CLE19 Otsuga, D., DeGuzman, B., Prigge, M.J., Drews, G.N., Clark. S.E., REVOLUTA regulates meristem initiation at lateral positions. Plant J. 25, Pearce, G., Strydom, D., Johnson, S., Ryan, C.A., A polypeptide from tomato leaves induces wound-inducible inhibitor protein. Science 253, Rojo, E., Sharma, V.K., Kovaleva, V., Raikhel, N.V., Fletcher, J.C., CLV3 is localized to the extracellular space, where it activates the Arabidopsis CLAVATA stem cell signaling pathway. Plant Cell 14, Sabatini, S., Beis, D., Wolkenfelt, H., Murfett, J., Guilfoyle, T., Malamy, J., Benfey, P., Leyser, O., Bechtold, N., Weisbeek, P., Scheres, B., An auxin-dependent distal organizer of pattern and polarity in the Arabidopsis root. Cell 99, Schoof H., Lenhard M., Haecker A., Mayer K.F., Jürgens G., Laux T., The stem cell population of Arabidopsis shoot meristems is maintained by a regulatory loop between the CLAVATA and WUSCHEL genes. Cell 100, Sharma, V.K., Fletcher, J.C., Maintenance of shoot and floral meristem cell proliferation and fate. Plant Physiol. 129, Sharma, V.K., Ramirez, J., Fletcher, J.C., The Arabidopsis CLV3-like (CLE) genes are expressed in diverse tissues and encode secreted proteins. Plant Mol. Biol. 51, Siegfried, K.R., Eshed, Y., Baum, S.F., Otsuga, D., Drews, G.N., Bowman, J.L., Members of the YABBY gene family specify abaxial cell fate in Arabidopsis. Development 126, Smyth, D.R., Bowman, J.L., Meyerowitz, E.M., Early flower development in Arabidopsis. Plant Cell 2, van Engelen, F.A., Molthoff, J.W., Conner, A.J., Nap, J.P., Pereira, A., Stiekema, W.J., PBINPLUS, an improved plant transformation vector based on pbin19. Transgenic Res. 4, Vroemen, C.W., Mordhorst, A.P., Albrecht, C., Kwaaitaal, M.A., De Vries, S.C., The CUP-SHAPED COTYLEDON3 gene is required for boundary and shoot meristem formation in Arabidopsis. Plant Cell 15,

62 Chapter 2 62

63 Chapter 3 Secreted proteins in Brassica napus microspore cultures Modified published in Crop Improvement II as a part of chapter 5, Biochemical and Molecular aspects of Haploid Embryogenesis in Haploids. Boutilier, K., Fiers, M., Liu, C.-M., and Geest A.H.M. (2004). Springer-Verlag, Heidelberg. pp Picture by R.V.L Joosen

64 Chapter 3 64

65 Secreted proteins in microspore cultures Introduction Brassica napus (B. napus) is considered a good model system to study Microspore Derived Embryo (MDE) development, in part due to the high efficiency of embryo production. However, embryo yield in B. napus MDE cultures can vary from 0 up to 10%, possibly due to variability in the condition of the donor plants, the differences in developmental stages of the microspores used as starting material, toxic factors released by dead or dying microspores or maybe even due to the secretion of proteins and/or peptides in the media by developing embryos (Pechan and Smykal 2001). To study the biochemical differences in conditioned medium of high and low-yielding embryogenic cultures a proteomics-based approach was used. We are focussing our analysis on small proteins, as a number of small proteins, and even peptides, have been shown to play a role in plant cell proliferation and meristem growth (Takayama and Sakagami 2002). One example of a growth stimulating peptide is phytosulfokine (PSK). In Asparagus cell culture, mesophyll cell division only occurs when the cells are cultured at sufficiently high density. A sulphated penta- (PSK-α) and a sulphated tetra-peptide (PSK-β) released by Asparagus cells cultured at high density was shown to stimulate mesophyll cell division in low density cell cultures (Matsubayashi et al. 1997). PSKs were also identified as the compound that contribute to the growth stimulating effect of conditioned medium from carrot somatic embryo cultures (Hanai et al. 2000). 65

66 Chapter 3 Results We examined the differences in protein profiles present in the conditioned medium from high-yield MDE cultures (HEC) in which 2% of microspores developed into embryos, and a low-yield MDE cultures (LEC) that did not produce any embryos. Both cultures were examined at about 10 days after culture initiation. HEC cultures contained a mix of developing embryos at the globular to torpedo stages, together with some arrested, but viable microspores, as well as dead microspores. LEC contained only viable-arrested and dead microspores. An isolation protocol was established in which the culture medium was acidified and protein fractions were isolated based on the differences in hydrophobicity and charge on C18-Solid Phase Extraction (SPE) columns (Fig. 1). After this first crude pre-purification the purified proteins were separated using a High Pressure Liquid Chromotography (HPLC) apparatus equipped with a C18 column (Fig. 1). Figure 1. Isolation scheme of proteins present in the media of Brassica MDE and LEC cultures. Schematic representation of the B. napus microspore culture system used to obtain embryos and conditioned medium. Media from 10 days old cultures were pre-purified using a SPE column and fractionated with a HPLC. Fractions were analyzed using MALDI-TOF and interesting fractions were analyzed using LC-MS MS/MS. This separation yielded 60 fractions, of which around 40 fractions showed detectable signals at 214 nm (mainly for proteins and peptides) and 254 nm (mainly for metabolites). Individual fractions were analyzed using Matrix Assisted Laser Desorption/ionisation Time-Of-Flight Mass Spectrometry (MALDI-TOF-MS). 66

67 Secreted proteins in microspore cultures Clear differences between the HEC and LEC medium could be detected using MALDI-TOF- MS (Fig. 2). Two large non-protein peaks (metabolites) were specific for the HEC medium. Figure 2. HPLC analysis between proteins present in the media from a High-Embrogenic Culture (HEC) and a Non-Responsive Culture (NRC). Pre-purified and cinncentrated media extracts were separated using HPLC equipped with a C18 column. From fraction 12 the MALDI-TOF MS spectra is depicted containing a specific peptide with a mass of 3994 Dalton present in the HEC culture and not in the NRC. From the 60 fractions isolated we were able to identify 12 proteins in the LEC conditioned medium, and 16 proteins in the HEC conditioned medium. Only two of these 28 proteins were found in the conditioned medium from both cultures. Fractions representing the most dramatic differences between HEC and LEC conditioned medium were subjected to trypsin digestion and de novo sequencing, from which six small proteins from different fractions of the HEC and LEC media were identified (Table 1). All of the small proteins listed in Table 1 are encoded by genes with predicted secretion signal peptides, which is consistent with their presence in the culture medium. Two of these small proteins are homologous to Bp4 and BAN54, which are known to be pollen-specific (Albani et al. 1990; Kim 1997). 67

68 Chapter 3 Mass 1 Identified sequences % of identity 2 Protein description Localisation 3 Fraction 12 (HEC) 1105 VECDAICKPK 100 unknown floral gene, B.napus Extracellular 1377 GCKVECDAICKPK 92 same as above same as above 1505 LPNSNWCCNTTPR 100 same as above same as above 1503 LPNSNWC:CNTTPR same as above same as above 1535 LPNSNWC X C X NTTPR same as above same as above Fraction 23 (HEC) 1237 TYPYKLPLDK Unknown 1253 PYKLPLDK Unknown Fraction 24 (HEC) 1336 YCXDEQQLPVNK 73 BAN54, B.rapa Extracellular 1392 LMDEQQLPVNK 73 same as above same as above 1426 YXDEQQLPVNK 64 same as above same as above 1520 LFSCDEQQLPVNK 62 same as above same as above 1292 IPITGSYCLPTK 100 Bp4-like, B.oleracea same as above 1382 IPITGSYXLPTK 92 same as above same as above Fraction 31 (LEC) 2606 CIGYLTQNGPLPR 100 LTP, B. napus Extracellular 1451 TRTNLNNMAR 80 same as above same as above Table 1. Tryptic peptides identified by mass spectrometry in the media of high embryogenic- (HEC) and low embryogenic cultures (LEC). 1 Peptide mass in Dalton; 2 Sequence similarity with known proteins in GenBank; 3 Localization based on sequence analysis; 4 C:C = internal SS-bridge; 5 C X = extra group(s) on cysteine; X: unknown The Bp4 gene was present on a dedicated B. napus MDE microarray (Boutillier et al., unpublished results) and expression analysis also supports a late pollen-specific rather than embryo expression pattern for this gene. The presence of these proteins in the medium from 10-day MDE cultures, which no longer contain any viable pollen, suggests that these proteins are derived from developing pollen grains from earlier stages of culture, and thus are very stable. Fraction 12 contained a protein that was abundantly present in HEC but absent in the LEC medium. This protein corresponds to an mrna sequence identified in B. napus flowers and flower buds (Genbank acc. CX189373, Figure 3). 68

69 Secreted proteins in microspore cultures Figure 3. Comparison between the putative protein derived from B. napus mrna CX and 5 sequenced peptide fragments obtained after trypsin digestion of the protein present in fraction 12 after HPLC separation. The theoretical masses are indicated and the putative cleavage site of the secretion signal is depicted with an arrow. Mass spectrometry data suggested that there is a potential internal disulfide bond connecting two cysteines within the peptide. The protein encoded by the cdna carries a putative signal sequence, suggesting that it is an extracellular protein. No clear homologue of this protein could be found in the Arabidopsis genome. In summary, a proteomics approach has been used as an efficient and sensitive way to analyse secreted proteins in B. napus culture media. Additional assays will be needed to elucidate the functions of these small proteins, and to determine if any of these proteins are responsible for the increased or decreased embryogenic capacity observed in different microspore cultures. 69

70 Chapter 3 References Albani D, Robert LS, Donaldson PA, Altosaar I, Arnison PG, Fabijanski SF (1990). Characterization of a pollen-specific gene family from Brassica napus which is activated during early microspore development. Plant Mol Biol 15: Hanai H, Matsuno T, Yamamoto M, Matsubayashi Y, Kobayashi T, Kamada H, Sakagami Y (2000). A secreted peptide growth factor, phytosulfokine, acting as a stimulatory factor of carrot somatic embryo formation. Plant Cell Physiol 41:27-32 Kim HUC, T.Y. (1997). Characterization of three anther-specific genes isolated from Chinese cabbage. Plant Mol Biol. 33: Matsubayashi Y, Takagi L, Sakagami Y (1997). Phytosulfokine-alpha, a sulfated pentapeptide, stimulates the proliferation of rice cells by means of specific high- and lowaffinity binding sites. Proc Natl Acad Sci USA 94: Pechan PM, Smykal P (2001). Androgenesis: Affecting the fate of the male gametophyte. Physiol Plant 111:1-8 Takayama S, Sakagami Y (2002). Peptide signalling in plants. Curr Opin Plant Biol 5:

71 Chapter 4 The 14-Amino Acid CLV3, CLE19 and CLE40 Peptides Trigger Consumption of the Root Meristem in Arabidopsis through a CLAVATA2- Dependent Pathway Martijn Fiers a,b, Elzbieta Golemiec c, Jian Xu d, Lonneke van der Geest a, Renze Heidstra b,d, Willem Stiekema b, and Chun-Ming Liu a,b a Plant Research International, B.V. P.O. Box 16, 6700 AA Wageningen, The Netherlands b Centre for BioSystems Genomics, PO Box 98, 6700 AB Wageningen, The Netherlands c Institute of Plant Physiology, Polish Academy of Sciences, Niezapominajek 21, Kraków, Poland d Department of Molecular Genetics, Utrecht University, Padualaan 8, 3584 CH, Utrecht, The Netherlands Modified published in The Plant Cell 17, (2005) Cover Picture in colour in appendix, page V

72 Chapter 4 72

73 CLE Peptides in Root Development ABSTRACT CLV3, CLE19 and CLE40 belong to a family of 26 genes in Arabidopsis that encode putative peptide ligands with unknown identity. It has been shown previously that ectopic expression of any of these 3 genes leads to a consumption of the root meristem. Here we show that in vitro application of synthetic 14-AA peptides, CLV3p, CLE19p and CLE40p, corresponding to the conserved CLE motif, mimics the over-expression phenotype. The same result was observed when CLE19 protein was applied externally. Interestingly, clv2 failed to respond to the peptide treatment, suggesting that CLV2 is involved in the CLE peptide signaling. Crossing of the CLE19 over-expression line with clavata mutants confirms the involvement of CLV2. Analyses using tissue-specific marker lines revealed that the peptide treatments led to a premature differentiation of the ground tissue daughter cells and mis-specification of cell identity in the pericycle and endodermis layers. We propose that these 14-AA peptides represent the major active domain of the corresponding CLE proteins, which interact with or saturate an unknown cell identity-maintaining CLV2 receptor complex in roots, leading to consumption of the root meristem. INTRODUCTION In multicellular organisms, cell to cell communication is essential for coordinating growth and differentiation. In animals, peptides are known to be the major players in cell-cell communication (for review, see Alberts et al., 1994). This is in contrast to plants in which most intercellular communication is mediated by phytohormones such as auxin, cytokinin, GA, ABA, ethylene and brassinosteroids (Kende et al., 1997; Mandave et al., 1988). However, in recent years, several putative peptide ligands have been identified in plants (for review, see Ryan et al., 2002; Lindsey et al., 2002) and have been shown to mediate signalling events during plant-pathogen interactions (Pearce et al., 1991), cell division (Matsubayashi et al., 1996) and anther-stigma interactions (Schopfer et al., 1999; Kim et al., 2003). CLAVATA3 (CLV3) is a putative peptide ligand of Arabidopsis that interacts with a disulphide-linked CLAVATA1/CLAVATA2 (CLV1/CLV2) receptor complex to restrict the stem cell population in the shoot apical meristem (SAM) in a non-cell autonomous manner 73

74 Chapter 4 (Fletcher et al., 1999). CLV1 is a membrane-bound leucine-rich repeat (LRR)-receptor kinase while CLV2 is an LRR-receptor-like protein lacking a kinase domain (Jeong et al., 1999; Clark et al., 1997). The stem cells are marked by CLV3 expression, while the SAM organizing centre (OC) is marked by the expression of the WUSCHEL (WUS) stem cell-promoting transcription factor (Laux, 2003). A feedback regulation loop between CLV3 and WUS maintains the number of stem cells in the SAM (Brand et al., 2000; Schoof et al., 2000). As such, clv1, clv2 and clv3 mutants have enlarged SAMs, while the wus mutant or CLV3 overexpression terminates SAM development (Laux et al., 1996; Hobe et al., 2003). Although biochemical studies showed that CLV3 is required for the formation of a 450 kd functional CLV1/CLV2 receptor complex with several associated proteins (Trotochaud et al., 1999), the biochemical nature of the active ligand encoded by CLV3 has not yet been elucidated. CLV3 belongs to the CLV3/ESR (CLE) family of 26 genes in Arabidopsis, of which 25 are transcribed in one or more tissues (Cock and McCormick, 2001; Hobe et al., 2003; Sharma et al., 2003; Fiers et al., 2004). These genes encode small proteins which contain a putative secretion signal at their N-termini and a conserved 14-amino acid (AA) motif (CLE-motif) at or near their C-termini (Cock and McCormick, 2001). In the clv3-1 and clv3-5 mutants, a single AA change (from G to A) in the CLE motif is enough to disrupt the function of CLV3 (Fletcher et al., 1999), indicating the importance of this motif. First identified in maize as being expressed in endosperm regions surrounding the embryo, ESR genes encode extracellular proteins with unknown functions (Bonello et al., 2002). Such CLE genes have not only been found in many plant species, but also in parasitic nematodes (Wang et al., 2005). One interesting feature of these CLE proteins is that they share very little sequence similarity outside the CLE motif (Cock and McCormick, 2001). Remarkably, CLE40 from Arabidopsis and Hg-SYV46 from nematode can fully complement the clv3 mutant phenotype when expressed under the control of the CLV3 and CaMV 35S (35S) promoter, respectively, illustrating the functional redundancy of these genes (Hobe et al., 2003; Wang et al., 2005). The redundancy could also be seen from the T-DNA insertonal mutants of CLE19 and CLE40, which gave no and a very subtle phenotype, respectively (Hobe et al., 2003; Fiers et al., 2004). In contrast, over-expression of several CLE genes such as CLV3, CLE19, CLE40 or Hg- SYV46 under the control of the 35S promoter induce striking developmental phenotypes in the root and shoot development in Arabidopsis (Hobe et al., 2003; Fiers et al., 2004; Wang et al., 74

75 CLE Peptides in Root Development 2005). Termination of root meristem development has been observed in transgenic plants over-expressing any of these four genes (Hobe et al., 2003; Fiers et al., 2004; Wang et al., 2005). These studies point to the existence of common and redundant signaling machinery in roots that responds to different CLE genes to trigger the differentiation of the root meristem. Here, we report the development of a novel in vitro root assay to elucidate how CLE genes trigger the consumption of the root meristem. Application of chemically synthesized 14-AA peptides, CLV3p, CLE19p and CLE40p (hereafter referred to as CLE peptides), corresponding to the conserved CLE motif of the related CLE proteins, led to consumption of the root meristem in a manner that closely resembles the over-expression phenotypes mentioned above. The same effect was achieved by application of heterologously produced full-length CLE19 protein. The clv2 mutation abolished the sensitivity of roots to these peptides, suggesting that CLV2 is involved in the CLE signaling pathway. Using different marker lines, we demonstrate that the peptide treatments led to an inward shifting of cell identity in several cell layers and to a premature differentiation of the ground tissue daughter cells. We propose that the CLE peptides represent the major functional domain of the CLE proteins, and that this domain interacts with CLV2 to trigger the termination of the root meristem. RESULTS In vitro application of CLE peptides causes a short root phenotype It has been reported previously that over-expression of CLV3, CLE19, CLE40 or Hg-SYV46 in Arabidopsis under the control of 35S promoter resulted in termination of root meristem development (Hobe et al., 2003; Fiers et al., 2004; Wang et al., 2005). The same phenotype was observed when CLE19 was expressed under the control of the root meristem-specific promoter RCH1 (Casamitjana-Martinez et al., 2003). The AA sequences of CLV3, CLE19 and CLE40 were compared to identify common motifs. The only conserved sequence among these proteins resides in the CLE motif (Figure 1A). We therefore examined whether a chemically synthesized 14-AA peptide of CLE19p (Figure 1B), corresponding to the conserved CLE motif of CLE19, is also able to trigger the consumption of the root meristem. 75

76 Chapter 4 As a control, we used a 16-AA peptide corresponding to the C-terminus of AGAMOUS that has no similarity with the CLE peptides (AGp, Figure 1B). Arabidopsis seeds were germinated on vertical plates with media containing different concentrations of the individual peptides. The length of the primary root was measured after 7 days (Figure 2A). Figure 1. The CLE motif and the peptides used in the root assays. (A) Alignment of CLV3, CLE5, CLE19, Bn CLE19 and CLE40. The signal sequences are underlined, and the CLE motif is framed. (B) Chemically synthesized peptides used in the root assays. Identical AAs in 5 or more proteins/peptides are shaded in black, and AAs that are similar or identical in at least 3 proteins are highlighted in grey. Although no clear effect on root growth was observed when CLE19p was applied at the 0.1 to 1 µm concentration range, a dramatic inhibitive effect was observed at 10 µm or higher (Figure 2A). Further increase of the CLE19p concentration from 10 µm to 100 µm gave only a slight decrease in root length (Figure 2A). AGp had no significant effect on root length. As such, we used a concentration of 10 µm for the subsequent analyses. 76

77 CLE Peptides in Root Development Figure 2. Treatment with CLE peptides gave a short root phenotype in Arabidopsis (A) Effect of different concentrations of CLE19p and AGp (control) on WT (Col-0) root development. The lengths of the main roots were measured after 7 days growth on peptidecontaining media (n=8 for each treatment). (B) Effects of different peptides on the root lengths of Arabidopsis (Ler). The roots were measured after 14 days of growth on media with 10 µm of various peptides. Note that CLE5p, CLV3m, CLV3t and CLE19t did not inhibit root growth (n=10 for each treatment). Data and error bars represent mean ± sd. These results gave us the first indication that the conserved motif of CLE19 is sufficient to mimic the short root phenotype generated by over-expression of the CLE19 gene. Further, CLV3p and CLE40p (Figure 1B) corresponding to the CLE motifs of CLV3 and CLE40, respectively, were tested for their effect on root growth. Unlike CLV3, CLE40 is expressed in roots (Hobe et al., 2003). Several peptides including a mutant CLV3 peptide (CLV3m: with a G to A conversion as in clv3-1 and clv3-5 mutants), 2 truncated peptides (CLV3t and CLE19t: with 5 AA removed the C-termini of CLV3p and CLE19p, respectively), and a shuffled CLV3 peptide (CLV3s: the same AA composition as CLV3p but with a shuffled sequence) were chemically synthesized as controls (Figure 1B). The peptides were added individually to medium at a concentration of 10 µm. Seedlings treated with CLV3p and CLE40p, similar to those treated with CLE19p, showed a clear reduction in root length, as compared to the controls with either no peptide, or treated with CLV3m, CLV3s, CLV3t or CLE19t, which did not have a significant effect on root length (Figure 2B). Statistical analysis showed no significant differences among treatments with CLV3p, CLE19p and CLE40p. 77

78 Chapter 4 To investigate if all peptides derived from the CLE motif of the CLE proteins are able to produce the short root phenotype, we synthesized a 14-AA peptide (CLE5p) of CLE5. The CLE motif of CLE5 differs from that of CLV3, CLE19 and CLE40 (Figure 1B). CLE5 is expressed in roots (Sharma et al., 2003). CLE5p shares between 43-50% identity with the CLE peptides. Treatment with CLE5p did not cause any reduction in root length (Figure 2B), indicating that the sequence of the peptide is critical for the function of the CLE peptides in triggering the consumption of root meristem, and not all peptides with a CLE motif from on of the 26 CLE genes results in a similar response. To examine if CLV3p application in solid media could complement the clv3 phenotype, we measured the size of the clv3 SAMs using Nomarski optics after 4, 8 and 14 days of treatment. We failed to detect any changes, suggesting that either CLV3p cannot be transported from the root to the SAM, or the peptide is not able to function in the SAM. The CLE peptides trigger the consumption of the root meristem in a manner similar to CLE19 over-expression We next investigated if the short root phenotype generated after the CLE peptide treatment resembles the phenotype observed after over-expression of CLE19 (Fiers et al., 2004). Using Nomarski optics, we observed that roots treated for 14 days with CLV3s, CLV3m and CLE5p were morphologically indistinguishable from roots grown on plates without peptide (Figure 3A to 3D). These roots have a cell division zone which consists of a population of cytoplasmdense cells, followed by gradually enlarged elongated cells In contrast, roots treated with the active CLE peptides were much thinner, with a significantly decreased number of meristematic cells (Figures 3E to 3G). These roots seem to have an equal number of cell layers along the radial axis, as in the wild type (WT), except for a region above the quiescent center (QC) where the formation of ground tissue appeared to be delayed. The thinner root phenotype seems to be caused by reduction in cell expansion. In CLE peptide-treated roots, only a few cytoplasm-dense cells could be recognized, which were immediately followed by elongated and highly vacuolated cells above this region. This consumption of the root meristem closely resembles the phenotype observed in roots of CLE19 over-expression lines (Casamitjana-Martinez et al., 2003; Fiers et al., 2004). 78

79 CLE Peptides in Root Development Figure 3. Effect of CLE peptides on root meristems (A) to (G) The morphology of the primary roots of Arabidopsis (Ler), 14 days after treatments with different peptides. Note the consumption of the root meristem (E-G) after treatments with CLE peptides. The bar in (G) represents 50 µm for (A) to (G). (H) Number of meristematic cells in roots of Ler seedlings after 14 days of treatment with different peptides. Data and error bars represent mean ± sd (n=10). The numbers of meristematic cells in the primary root of WT and CLE peptide-treated roots were quantified by counting the number of non-elongated, cytoplasm-dense cells along the cortex layer, starting from the QC after 14 days of treatment. In WT seedlings, the number of meristematic cells was on average 81, while only 1 to 4 such cells could be found from seedlings germinated in the presence of a CLE peptide (n=10; Figure 3E to 3G, 3H). Previous reports showed that ectopic expression of CLV3, CLE19 or CLE40 caused consumption of the root meristem without immediately disturbing auxin distribution in the 79

80 Chapter 4 columella initial cells or QC function (Casamitjana-Martinez et al., 2003; Hobe et al., 2003; Fiers et al, 2004). Figure 4. Effects of CLE peptides on the QC. GUS staining was performed in the QC25 roots after a 5- day treatment with or without peptides. The roots were cleared and observed under Nomarski optics, showing no visible changes in the QC cells. Seeds were germinated on media containing (A) no peptide, (B) CLV3s, (C) CLV3m or (D) CLV3p. To determine if this was also the case after the treatments with CLE peptides, seedlings of DR5:GUS (a reporter line showing auxin distribution, Sabatini et al., 1999) and QC25 (QCspecific marker, Casamitjana-Martinez et al., 2003) marker lines were examined. Although a clear reduction in root length and a decreased number of meristematic cells were observed, no visible difference was observed in the DR5:GUS and QC25 (Figure 4) expression patterns. This observation revealed that neither the peptide treatment nor the over-expression acts primarily on the QC. CLV2 is involved in perception of the CLE peptides in roots In order to identify components of the signal transduction pathway involved in perception of the CLE peptide signal in roots, we determined whether the CLAVATA signalling pathway is involved in the short root phenotype by treating clv1, clv2 and clv3 mutants with the CLE peptides. Seedlings of WT (Ler), clv1-1 and clv3-2 grown on plates with any of the CLE peptides showed a significantly shorter root length and decreased number of meristematic cells (Figure 5A and 5B). For example, the number of meristematic cells in clv1-1 roots 80

81 CLE Peptides in Root Development grown on peptide-free media was 47±12 mm, which was reduced to 2±2 mm after treatment with CLE19p (Figure 5B). clv2-1, however, showed significantly less sensitivity to the CLE peptides (Figure 5A and 5B). In the absence of peptide application we did not detect any defects in the root of the clv2, using both Nomarski optics and confocal microscopy. The primary roots of clv2 mutants treated with CLE peptides also maintained a normal meristem morphology (Figure 5E), in contrast to fully differentiated root meristems of WT, clv1-1 and clv3-2 after the peptide treatments (Figures 5C, 5D and 5F). Seedlings of clv2-1 treated with CLV3p, CLE19p or CLE40p had meristematic cells, as compared to 2-4 such cells for clv1-1 seedlings, and 1-3 for clv3-2 (Figure 5B). Heterologously produced CLE19 protein also triggers a consumption of the root meristem To examine if the full-length CLE protein functions like the CLE peptides, a construct was made in which the sequence encoding the Brassica napus (Bn) CLE19 protein without the putative secretion signal peptide was fused to the C-terminus of glutathione S-transferase (GST) and expressed in E. coli. After cleaving with thrombin and purification (Figure 6A), the Bn CLE19 protein was evaluated for its effect on Arabidopsis roots. Application of the Bn CLE19 protein at a concentration of 10 µm strongly inhibited root growth of Arabidopsis seedlings, as compared to the control GST protein, which did not have any detectable effect (Figure 6B). Figure 6. Heterologous production of CLE19 protein and its effect on root development (A) SDS-PAGE gel of thrombin-cleaved GST-Bn CLE19 and GST protein produced in E. coli with a mass of about 5 kd and 24 kd, respectively. The molecular mass (in kd) is shown on the right. (B) Effect of purified the Bn CLE19 or GST proteins on root development. WT (Col-0) seedlings were grown for 10 days on plates containing 10 µm of Bn CLE19 or GST. Data and error bars represent mean ± sd (n=8). 81

82 Chapter 4 Figure 5. Effect of peptides on the root growth in different mutants All roots were measured after 2 weeks growing on media containing different peptides. (A) The length of the primary roots of clv1-1, clv2-1 and clv3-2 grown on media with different peptides. Note that the CLE peptides did not give a significant inhibition in clv2-1 (n=10). (B) Number of meristematic cells in roots of clv1-1, clv2-1 and clv3-2 grown on media with different peptides (n=10). Data and error bars in (A) and (B) represent mean ± sd. (C) to (F) The morphology of roots of Ler and clavata mutants treated with CLE19p, showing that clv2-1 is not sensitive to the peptide. The bar in (C) represents 50 µm for (C) to (F). 82

83 CLE Peptides in Root Development After 10 days treatment, the average root length in the GST-treated seedlings was about 34 mm, while for the Bn CLE19-treated seedlings it was about 7 mm (Figure 6B). Nomarskibased observations showed that Bn CLE19 protein-treated roots exhibited a similar phenotype as the CLE peptide-treated or CLE19 over-expression roots. No additional phenotypes were observed as compared to the treatment with CLE19p, demonstrating a functional similarity between the CLE peptides and the full-length protein. Genetic analysis confirms that CLV2 is involved in perceiving the CLE19 signal in roots After establishing the role of CLV2 in perception of CLE peptides in roots, we studied if CLV2 is also needed for the perception of the ectopically expressed CLE19 in transgenic Arabidopsis. We crossed a P 35S :Bn CLE19 over-expression line (as described in Fiers et al., 2004) with Ler, clv1-1, clv2-1 and clv3-2, respectively. In the F1 generation, we identified seedlings with short roots and transferred them to the greenhouse to obtain F2 seeds. Plants with long roots were selected from each family of F2 seedlings. We reasoned that plants with long roots should not carry the P 35S :Bn CLE19 transgene, unless the line carries a mutation that can suppress the function of the transgene. Table 1. Analysis of the F2 plants with long roots from crosses between P 35S :Bn CLE19 and Ler or different clavata mutants (clv1, clv2 and clv3). F2 plants Plants with multiple carpels a Plants with WT carpels Cross combinations with long roots without P 35S :Bn CLE19 with P 35S :Bn CLE19 without P 35S :Bn CLE19 with P 35S :Bn CLE19 P 35S :Bn CLE19 x Ler P 35S :Bn CLE19 x clv1-1 P 35S :Bn CLE19 x clv2-1 b P 35S :Bn CLE19 x clv a multiple carpels in the pistil is the typical phenotype for all clavata mutants. b two independent experiments were performed, and the data were combined in sum. As shown in Table 1, all plants from the F2 populations of P 35S :Bn CLE19 x Ler, P 35S :Bn CLE19 x clv1-1 and P 35S :Bn CLE19 x clv3-2 with a long-root phenotype did not carry the transgene. Plants with both long roots and the transgene were only identified in the F2 population of P 35S :Bn CLE19 x clv2-1 (Table 1). All these 11 plants also displayed the clv2 83

84 Chapter 4 mutant phenotype in siliques. A similar result was obtained with the P 35S :At CLE19 transgenic line (data not shown), demonstrating that the clv2 mutation also suppressed the short-root phenotype produced by the At CLE19 transgene. Consistent with this conclusion, none of the short-root plants obtained from the F2 population of P 35S :Bn CLE19 x clv2-1 were homozygous for the clv2 mutation (data not shown), further confirming that the homozygous clv2 mutation was needed to suppress the short-root phenotype induced by Bn CLE19 overexpression. Treatment with CLE peptides leads to a mis-specification of cell identities To study the effect of CLE peptides on root development, three tissue-specific marker lines were selected, treated and inspected by confocal microscopy. These lines were 1) J0571, which shows GFP expression in cortex and endodermis layers including the ground tissue initials and their daughter cells, and occasionally in the QC; 2) P SCR :GFP, which reflects the expression of the SCARECROW (SCR) transcription factor (Wysocka-Diller et al., 2000) and shows GFP expression in the endodermis layer, QC and ground tissue initials; and 3) P CO2 :YFP-H2B, which shows YFP expression in cortex cells, but not in their initials, nor in the QC (Heidstra et al., 2004). These reporter lines were grown on media containing either CLE peptides or controls (CLV3s, CLV3m or without peptide). What is the identity of this single file of cells located at the position of the ground tissue daughter cell? The GFP expression in the J0571 marker line (Figure 7C and 7D, indicated by arrowheads) suggested that they still maintained their ground tissue identity, despite the fact that the division pattern was altered. Similarly, when P SCR :GFP seedlings were treated with CLE peptides, GFP expression was observed in this single file, suggesting that it still maintains an endodermis and ground tissue identity (insert in Figure 7G, marked by arrowheads). Since the SCR promoter is not active in the cortex layer (Figure 7E) we could exclude the possibility that these cells obtained a full cortex identity. To further clarify the identity of these cells, the cortex-specific marker line P CO2 :YFP-H2B (Figure 7I) was examined upon peptide treatment. Interestingly, the single-file of cells showed YFP expression (Figure 7J to 7M; indicated by arrowheads). We also noticed that, the YFP expression was excluded from the ground tissue initial cells (Figure 7J and 7M, marked by arrows). As such, we concluded that the ground tissue daughter cells had obtained a cortex/endodermis double identity before the asymmetrical periclinal division occurs. 84

85 CLE Peptides in Root Development This phenotype partially resembles the phenotype of the scr mutant in which P CO2 :YFP-H2B and SCR are expressed in the single-layered ground tissue (Heidstra et al., 2004). In all these lines, GFP expression was localized in the expected cell layer in the absence of peptides (Figure 7A, 7E and 7I) or with control peptides (Figure 7B, 7F and 7H). In all samples examined a maximum of 1 ground tissue daughter cell was observed above the single ground tissue initial cell (Figure 7A; labelled with an arrowhead). In a number of samples neither visible ground tissue daughter cell nor ground tissue initial cell could be detected (see Figure 7I). When seedlings of J0571 were treated with CLE peptides for 4 days, a single-file, with up to 4 ground tissue daughter cells, was observed (Figure 7C, 7D; labelled with arrowheads). Over 50% of the plants treated with CLE peptides exhibited this phenotype. This is an early defect since at this stage the reduced root growth was not yet evident. 85

86 Chapter 4 Figure 7 *. Effects of CLE peptides on the cell identity of roots. The ground tissue initial cells are marked by arrows, and the ground tissue daughter cells, including cells at this position, are labelled with arrowheads. Abnormal expression of GFP or YFP in other cell layers is marked by asterisks. 86

87 CLE Peptides in Root Development (A) to (D) Confocal analysis after 4-day treatment of the roots of J0571. Note that treatments with CLV3p (C) or CLE19p (D) lead to a delayed separation of cortex and endodermis and expression of GFP in pericycle cells. (A) No peptide; 10 µm of (B) CLVm; (C) CLV3p and (D) CLE19p. (E) to (H) Confocal analyses of P SCR :GFP seedlings after 4-day treatments. (E) No peptide; (F) 10 µm CLV3s and (G) 10 µm CLV3p. The insert shows the delayed separation of the cortex and endodermis cells (with GFP expression). (H) 10 µm CLE19p. (I) to (M) P CO2 :YFP-H2B marker line treated with different peptides. Note the YFP expression in the ground tissue daughter cells and the pericycle cells, but not in ground tissue initial cells. (I) No peptide, 4 days after germination. The insert shows the absence of YFP expression in the QC, ground tissue initial and ground daughter cells. (J) 10 µm CLE19p, 4 days after treatment; (K) 10 µm CLE40p, 5 days after treatment; (L) and (M) 10 µm CLE19p, 5 days after treatment. The scale bars represents 30 µm. *) Colour figure in appendix, page VI After an 8-day treatment with CLE peptides, these single-file ground tissue daughter cells were enlarged and elongated, but the number of cells in the file did not increase further (Figure 8), suggesting that the division of these cells was impaired. Examination of the cell layers across the root meristem showed that, beside the changes in the ground tissue, all other cell layers were morphologically recognizable. However, in both the J0571 and P SCR :GFP marker lines treated with CLE peptides for 4 days we observed the expression of GFP in the pericycle layer (Figures 7C, 7D, 7G and 7H, indicated by asterisks). This ectopic GFP expression was observed in over 70% of roots treated with CLE19p, and around 40% treated with CLV3p and CLE40p. The pericycle initial cells (located next to QC) and the adjacent few cells in the same layer often showed strong GFP signal, while GFP levels decreased in the above cells (Figure 7C, 7D, 7G and 7H). In a few cases, noncontinuous GFP expression was observed in this layer (Figure 7C). The number of pericycle cells with GFP expression at one focal plane ranged from 1 to 50. These cells were often distributed asymmetrically along the two sides of the root (Figure 7C, 7D, 7G and 7H). Interestingly, in 8-day old seedlings treated with CLE peptides, GFP expression in the pericycle cells was no longer detectable (Figure 8). 87

88 Chapter 4 Figure 8 *. Confocal observation of J0571 roots, after an 8-day treatment with CLE19p. (A) No peptide. (B) CLE19p. Note that at this stage the number of ground tissue daughter cells (marked with arrowheads) was not increased further, and no GFP expression was detectable in the pericycle layer. Ground tissue daughter cells are marked with arrowheads. The scale bars represents 30 µm. *) Colour figure in appendix, page VII Similarly, upon the treatment with CLE peptides, we observed the expression of the cortex marker in the endodermis layer (Figure 7J, 7K and 7L, indicated by asterisks). About 70% of roots treated with the CLE19p showed YFP expression in the endodermal layer, while the frequency of YFP expression was ca. 40% for CLV3p- and CLE40p-treated roots. Penetration of CLE19p into roots To determine whether the CLE peptides enter the root or function on the root surface, a labelled CLE19p (R-CLE19p) was synthesized with a lissamine rhodamine fluorophorecoupled to the N-terminal lysine. WT (Col-0) seedlings were incubated with 10 µm of R-CLE19p, free lissamine rhodamine or propidium iodide (PI), and analysed by confocal microscopy. PI is a vital dye that is widely used for staining of the cell wall. It can only enter the cells if the membrane integrity is disrupted (Robinson et al., 2002). R-CLE19p was able to enter the root. R-CLE19p fluorescence was observed in all cell layers of the roots within 1 min after peptide application. The fluorescence was located predominantly in the intercellular spaces (Figure 9A). After 4 min the signal was more intense, but still only located in the intercellular spaces (Figure 9D). Interestingly, the free lissamine rhodamine entered the roots with a lower efficiency, as indicated by the weak fluorescence (Figure 9B and 9E). PI entered the roots with the highest 88

89 CLE Peptides in Root Development efficiency (Figure 9C and 9F), taking only 1 min to reach the saturation level (Figure 9C and 9D). Figure 9. Penetration of fluorescence-labelled CLE19p in roots. The roots of Arabidopsis (Col-0) were incubated for 1 min (A to C) or 4 min (D to F) with 10 µm R-CLE19p (A and D), 10 µm lissamine rhodamine (B and E) or 10 µm PI (C and F). The scale bar in (F) represents 30 µm for (A) to (F). DISCUSSION Previous studies suggested that CLV3 acts as a secreted protein that is required for the formation of the active CLV1/CLV2 signalling complex (Trotochaud et al., 1999; Fletcher et al., 1999; Rojo et al., 2003). However, the precise molecular identity of the functional CLV3 protein is still unknown. Several attempts have been made to identify the mature CLV3 protein using antibodies (Nishihama et al., 2003.; Rojo et al., 2002; Laux et al., 2003), but so far only a CLV3-T7 fusion protein could be detected (Rojo et al., 2002). Recently we and other labs have demonstrated that ectopic expression of CLV3, CLE19, CLE40 or HgSYV46 leads to a consumption of the root meristem (Casamitjana-Martinez et al., 2003; Hobe et al., 89

90 Chapter ; Fiers et al., 2004; Wang et al., 2005). Sequence alignment showed that the similarity between the putative proteins encoded by these genes was restricted to the CLE motif. Both CLV3 and CLE19 appeared to be not expressed in root meristems (Fletcher et al., 1999; Fiers et al., 2004), suggesting that the phenotype of ectopic expression represents a gain-of-function phenotype. We developed an in vitro root assay in Arabidopsis to examine how these proteins might generate the same phenotype in roots. The in vitro assay showed that three chemically synthesised 14-AA CLE peptides, corresponding to the conserved CLE motif of CLV3, CLE19 and CLE40, were able to mimic the over-expression phenotype when applied in vitro. Three CLE peptides share only 50% (between CLV19p and CLE3p or CLE40p) to 64% (between CLV3p and CLE40p) sequence identity. The similar phenotype observed suggests that a common signalling pathway is involved and that the receptor(s) is able to recognise multiple ligands with certain sequence variations. It is possible that in the endogenous situation the specificity of CLV3, CLE19 and CLE40 is defined by the regulation of expression. The control peptides, including CLV3s, CLV3m, CLV3t and CLE19t did not affect root development. CLE5p, a peptide corresponding to a distant member of the CLE genes expressed in roots, was also unable to trigger the consumption of the root meristem. These observations suggest that the CLE peptides act in a sequence-specific manner. Experiments with a labelled peptide showed that this peptide is able to efficiently enter the intercellular spaces of the root. The fluorescence signal was located predominately in the cell wall, suggesting that active transport is not likely to be involved. Although the labelled peptide penetrated into roots faster than the free fluorescence dye, PI, which cannot enter the intact plasma membrane (Robinson et al., 2002), enters into the roots even faster. The differences in penetration efficiency can be explained either by the differences in hydrophobicity or molecular masses. By using a QC marker and the DR5:GUS line, we observed that application of the CLE peptides resulted in a consumption of the root meristem without directly interfering with the QC function or auxin distribution in roots. This conclusion is further supported by the observation that the ground tissue initial cells located next to the QC and tightly regulated by the QC (van den Berg et al., 1995, 1997) were not affected. These data are in agreement with previous observation that over-expression of CLV3, CLE19 or CLE40 in Arabidopsis causes a 90

91 CLE Peptides in Root Development termination of root meristems without acting primarily on the QC (Casamitjana-Martinez et al., 2003; Fiers et al., 2004; Hobe et al., 2004). To determine if the sequence before the CLE motif could influence the activity, Bn CLE19 protein (with its N-terminal secretion signal removed) was produced in E. coli and tested in the root assay. The full-length CLE19 protein appeared to be functional in Arabidopsis in a similar fashion as CLE19p, suggesting that the extra sequence before the CLE motif does not affect its function. This is consistent with what is known for other peptide ligands in plants, such as PSK and flagellin (Yang et al., 1999; Felix et al., 1999). Although the pre-proteins for PSK in rice, Arabidopsis and asparagus are different in sequence, the functional peptides are believed to be the same (Yang et al., 1999).. Since no endogenous CLE ligand has been identified yet, we can not exclude the possibility that the endogenous peptides are longer or shorter than the CLE peptides tested. For the identification of signalling components involved in perception of these CLE peptides, we examined whether components of the Arabidopsis CLV signalling pathways, specifically CLV1, CLV2 and CLV3, are involved in CLE peptide signalling. Interestingly, the clv2 mutant was insensitive to CLE peptide treatment, suggesting that CLV2 is functionally involved in perception of the CLE peptides in roots. Genetic analysis demonstrated that CLV2 is also involved in the perception of the P 35S :Bn CLE19 transgene, suggesting that the same machinery is involved in the perception of the transgenic Bn CLE19 and CLE peptides. Among these CLAVATA genes, CLV2 is the only one expressed in Arabidopsis roots (Birnbaum et al., 2003). It seems that CLV2 not only participates in the CLV1/CLV2 receptor complex to transmit the CLV3 signal in the SAM (Jeong et al., 1999), but also functions in an unidentified CLV1-like receptor kinase complex in roots to perceive the CLE peptides. The fact that clv2 does not show any root phenotype suggests that either CLV2 is redundant in roots while the redundant protein(s) can not sense the CLE peptides, or CLV2 is not functional in roots and the CLE peptides act ectopically to generate a gain of function phenotype. Three cell-layer identity markers were examined to to further study the effects of the peptide treatment,. Firstly we observed that the treatments led to the accumulation of multiple cells at the position of the ground tissue daughter cell. This is striking since in the wild-type situation an asymmetrical division always occurs immediately after the ground tissue daughter cell is 91

92 Chapter 4 produced, to form an outer cortex cell and an inner endodermis cell (van den Berg et al., 1995, 1997). This tightly regulated cell division pattern is controlled by SCR in a cell autonomous manner, and the function of SHORT ROOT (SHR) is required for the activation of SCR which in turn induces the rotation of the division plane (Helariutta et al., 2000; Heidstra et al., 2004). Detailed analyses showed that two ground tissue markers (J0571 and P SCR :GFP) were expressed in the row of ground tissue daughter cells, suggesting these cells still maintain their ground tissue identity. In the absence of CLE peptide application YFP expression in the cortex-specific marker line (P CO2 :YFP-H2B) was excluded from the ground tissue initials and the ground tissue daughter cells (see insert in Figure 7I; Heidstra et al., 2004), However, upon treatment with CLE peptides YFP expression was observed in the single file of cells located at the position of the ground tissue daughter cell (Figure 7J to 7M), indicating that these cells obtained a cortex identity prior to the asymmetrical division. Since SCR was expressed in these cells, we conclude that they have obtained a cortex/endodermis double identity. This single-cell layer of ground tissue with a double identity is similar to that in scr but different from the shr mutant in which the single layer has only cortex identity (Di Laurenzio et al, 1996; Helariutta et al., 2000). Interestingly, both scr and shr mutants have a short-root phenotype in which the root meristem gradually differentiates after germination, suggesting that interrupted cell layer formation could have a direct effect on the maintenance of meristematic activity in roots. Whether CLE peptides interfere with SCR signalling components remains to be studied. Additionally, it is also not known if the CLE peptides first inhibit the division of the ground tissue daughter cells, which then leads to the differentiation of these cells, or vice versa. Another striking effect observed after treatment with CLE peptides is the confusion in cell layer identity. The ground tissue genes (P SCR :GFP and J0571) were transiently activated in the pericycle layer, and a cortex gene was active in the endodermal layer. The common feature is that the inner cell layers takes the identity of the outer cell layer. Such a confusion or mis-specification of layer identity could be a consequence of the failure of cells to sense the presence of neighbouring cells. It is well known that positional signals are used by plant cells to define their identity (van den Berg et al., 1995; 1997). As such, we believe that these CLE peptides and the CLV3, CLE19 and CLE40 transgenes act in a dominant-negative fashion to interact with or saturate an unknown cell identity-maintaining CLV2 receptor 92

93 CLE Peptides in Root Development complex. This in turn blocks intercellular communication among different cells and cell layers in the root, which leads to consumption of the root meristem. METHODS Plant growth conditions and plant strains The marker lines J0571 (made by Jim Haseloff), P SCR :GFP (Wysocka-Diller et al., 2000) and the mutants of clv1-1, clv2-1 and clv3-2 were provided Nottingham Arabidopsis Stock Center. P CO2 :YFP-H2B, DR5:GUS and QC25 were used as previously described (Casamitjana- Martinez et al., 2003; Heidstra et al., 2004) Root assay Seeds were gas-sterilized in a desiccator for 1 hr with 100 ml of bleach (4% NaClO) mixed with 3 ml of HCl in a beaker. For peptide treatments, the sterilized seeds were plated at a distance of 0.5 cm on media containing different concentrations of peptides, 0.5 Murashige and Skooge micro- and macro-elements (MS, Duchefa, Haarlem, NL), 1% (W/V) sucrose, 0.5 g/l 2-(N-morpholino) ethane-sulfonic acid (MES, ph 5.8) with 1.5% (W/V) of agar. Peptides and proteins were added to the sterilized media before the medium was solidified. Plates were first incubated at 4 C in the dark for 2 days and then transferred to a room with a temperature of 23 C, 16 hr light per day, and cultured nearly vertical. The root length was measured from the base of the hypocotyl to the tip of the primary root. GUS analysis was as described by Fiers et al. (2004), and GFP analyses were performed after 4, 6 and 8 days of growth. Peptides were ordered from Mimopopes (Clayton Victoria, Australia) with a purity of >70%, and dissolved in a filter-sterilized sodium phosphate buffer (50 mm, ph 6). To examine the penetration of the peptide into roots, CLE19p with a lissamine rhodamine fluorophore coupled to the N-terminal lysine (R-CLE19p) was ordered (Mimopopes). Fourday old seedlings of Arabidopsis (Col-0) were incubated in either 10 µm R-CLE19p, 10 µm lissamine rhodamine or 10 µg/ml propidium iodide by incubating the roots for different time periods after which they were examined using confocal microscopy. 93

94 Chapter 4 Microscopy Roots and dissected SAMs were cleared following the protocol of Sabatini et al. (1999) and analyzed using a Nikon microscope equipped with Nomarski optics. For confocal microscopy, roots were counter-stained with 10 µg/ml PI (Sigma) and analyzed with a Leica SP2 inverted confocal microscope following the protocol of Heidstra et al. (2004). Protein purification Two primers were designed (5 -TATGGATCCGCTTCATTTCGGAGTTTG-3 and 5 - ATACTCGAGTTACCTGTTGTGAAGTGGA-3 ) to amplify Bn CLE19 (without the signal sequence) from the cdna. The PCR fragment was digested with BamHI and HindIII (Invitrogen, Breda, NL) and cloned into the pgex4t-2 vector, and the fusion protein was purified as described by the manufacturer (Amersham Biosciences, Roosendaal, NL). The Bn CLE19 and the GST control were tested using SDS-PAGE and quantified with Coomassie plus protein assay reagent (Pierce, Etten-Leur, NL). Genetic analysis The F1 of the crosses between a P 35S :Bn CLE19 plant and individual clavata mutants (clv1-1, clv2-1 and clv3-2) or WT (Ler) were examined using the root assay mentioned above for a normal or short root phenotype. F2 seeds were harvested from plants with short roots and assayed again for individuals with a long or short root phenotype. These two groups were transplanted separately to soil and checked for the presence of the transgene with a PCR for the NPTII gene (5 -TGGGCACAACAGACAATCGGCTGC-3 and 5 -TGCGAATCGGGA- GCGGCGATACCG-3 ). Homozygous clavata mutants were identified from each group by their carpel phenotype. Accession Numbers Sequence data from this article can be found in the EMBL/GenBank data libraries under accession numbers AT2G27250 (CLV3), AT2G31082 (CLE5), At3g24225 (CLE19), AF (Bn CLE19), and AT5G12990 (CLE40). Acknowledgements We thank Kim Boutilier and Gerco Angenent for critical reading of the manuscript. We also thank the Wageningen University MicroSpectroscopy Centre for technical support, the 94

95 CLE Peptides in Root Development Wisconsin Arabidopsis Knockout Center, the Salk Institute, the TMRI (Syngenta) and the Nottingham Arabidopsis Stock Centre for providing the T-DNA insertion lines. This work was financially supported in part by the Dutch Ministry of Agriculture, Nature Management and Fisheries (DWK281/392) and by the Centre for BioSystems Genomics (CBSG), which is part of the Netherlands Genomics Initiative / Netherlands Organization for Scientific Research. E. G. was supported by the EU CROPSTRESS project (QLAM ). REFERENCES Alberts, B., Bray, D., Lewis, J., Raff, M., Roberts, K., and Watson, J. D. (1994) Molecular Biology of the Cell (Garland, New York). van den Berg, C., Willemsen, V., Hage, W., Weisbeek, P., and Scheres, B. (1995). Cell fate in the Arabidopsis root meristem determined by directional signalling. Nature 378, van den Berg, C., Willemsen, V., Hendriks, G., Weisbeek, P., and Scheres, B. (1997). Short-range control of cell differentiation in the Arabidopsis root meristem. Nature 390, Birnbaum, K., Shasha, D.E., Wang, J.Y., Jung, J.W., Lambert, G.M., Galbraith, D.W., Benfey, P.N. (2003) A gene expression map of the Arabidopsis root. Science 302, Bonello, J.F., Sevilla-Lecoq S., Berne A., Risueno M.C., Dumas C., and Rogowsky, P.M. (2002) Esr proteins are secreted by the cells of the embryo surrounding region. J. Exp Bot. 53, Brand, U., Fletcher, J.C., Hobe, M., Meyerowitz, E.M., and Simon, R. (2000). Dependence of stem cell fate in Arabidopsis on a feedback loop regulated by CLV3 activity. Science 289, Casamitjana-Martinez, E., Hofhuis, H.F., Xu, J., Liu, C.M., Heidstra, R., and Scheres, B. (2003). Root-specific CLE19 over-expression and the sol1 and sol2 suppressors implicate a CLV-like pathway in the control of Arabidopsis root meristem maintenances. Curr. Biol. 13, Clark, S.E., Williams, R.W., and Meyerowitz, E.M. (1997). The CLAVATA1 gene encodes a putative receptor kinase that controls shoot and floral meristem size in Arabidopsis. Cell 89,

96 Chapter 4 Cock, J.M., and McCormick, S. (2001). A large family of genes that share homology with CLAVATA3. Plant Physiol. 126, Di Laurenzio, L., Wysocka-Diller, J., Malamy, J.E., Pysh, L., Helariutta, Y., Freshour, G., Hahn, M. G., Feldmann, K. A., and Benfey, P.N. (1996). The SCARECROW gene regulates an asymmetric cell division that is essential for generating the radial organization of the Arabidopsis root. Cell 86, Felix, G., Duran, J. D., Volko, S., Boller, T. (1999). Plants have a sensitive perception system for the most conserved domain of bacterial flagellin. Plant J. 18, Fiers, M., Hause, G., Boutilier, K., Casamitjana-Martinez, E., Weijers, D., Offringa, D., van der Geest, L., van Lookeren Campagne, M., and Liu, C.M. (2004). Mis-expression of the CLV3/ESR-like gene CLE19 in Arabidopsis leads to a consumption of root meristem. Gene 327, Fletcher, J.C., Brand, U., Running, M.P., Simon, R., and Meyerowitz, E.M. (1999). Signaling of cell fate decisions by CLAVATA3 in Arabidopsis shoot meristems. Science 283, Heidstra, R., Welch, D., and Scheres, B. (2004). Mosaic analyses using marked activation and deletion clones dissect Arabidopsis SCARECROW action in asymmetric cell division. Genes Dev. 18, Helariutta, Y., Fukaki, H., Wysocka-Diller, J., Nakajima, K., Jung, J., Sena, G., Hauser, M.T., and Benfey, P.N. (2000). The SHORT-ROOT gene controls radial patterning of the Arabidopsis root through radial signaling. Cell 101, Hobe, M., Muller, R., Grunewald, M., Brand, U., and Simon, R. (2003). Loss of CLE40, a protein functionally equivalent to the stem cell restricting signal CLV3, enhances root waving in Arabidopsis. Dev. Genes Evol. 213, Jeong, S., Trotochaud, A.E., and Clark, E. (1999). The Arabidopsis CLAVATA2 gene encodes a receptor-like protein required for the stability of the CLAVATA1 receptor-like kinase. Plant Cell 11, Kim, S., Mollet, J.C., Dong, J., Zhang, K., Park, S.Y., and Lord, E.M. (2003). Chemocyanin, a small basic protein from the lily stigma, induces pollen tube chemotropism. Proc. Natl. Acad. Sci. USA 100, Kende, H., and Zeevaart, J.A.D. (1997). The five classical plant hormones. Plant Cell 9,

97 CLE Peptides in Root Development Laux, T., Mayer, K.F., Berger, J., and Jürgens, G. (1996). The WUSCHEL gene is required for shoot and floral meristem integrity in Arabidopsis. Development 122, Laux, T. (2003). The stem cell concept in plants: a matter of debate. Cell 113, Lindsey, K., Casson, S., and Chilley, P. (2002). Peptides: new signalling molecules in plants. Trends Plant Sci. 7, Mandave, N.B. (1988). Plant growth-promoting brassinosteriods. Annu. Rev. Plant Phys. Plant Mol. Biol. 39, Matsubayashi, Y., and Sakagami, Y. (1996). Phytosulfokine, sulfated peptides that induce the proliferation of single mesophyll cells of Aspargus officinalis L. Proc. Natl. Adac. Sci. USA 93, Nishihama, R., Jeong, S., DeYoung, B., and Clark, S.E. (2003) Retraction. Science 300, Pearce, G., Strydom, D., Johnson, S., and Ryan, C.A. (1991). A polypeptide from tomato leaves induces wound-inducible inhibitor proteins. Science 253, Pearce, G., Moura, D.S., Stratmann, J., and Ryan, C.A. Jr. (2001) RALF, a 5-kDa ubiquitous polypeptide in plants, arrests root growth and development. Proc. Natl. Acad. Sci. USA 98, Robinson, J.P., Darzynkiewicz, Z., Dean, P.N., Hibbs, A.R., Orfao, A., Rabinovitch, P.S., and Wheeless, L.L. (2002). Current Protocols in Cytometry. (New York: John Wiley & Sons). Rojo, E., Sharma, V.K., Kovaleva, V., Raikhel, N.V., and Fletcher, J.C. (2002). CLV3 is localized to the extracellular space, where it activates the Arabidopsis CLAVATA stem cell signaling pathway. Plant Cell 14, Ryan, C.A., Pearce, G., Scheer, J., and Moura, D.S. (2002). Polypeptide hormones. Plant Cell 14, S251-S264. Sabatini, S., Beis, D., Wolkenfelt, H., Murfett, J., Guilfoyle, T., Malamy, J., Benfey, P., Leyser, O., Bechtold, N., and Weisbeek, P. (1999). An auxin-dependent distal organizer of pattern and polarity in the Arabidopsis root. Cell 99, Schoof, H., Lenhard, M., Haecker, A., Mayer, K.F.X., Jürgens, G., and Laux, T. (2000). The stem cell population of Arabidopsis shoot meristems is maintained by a regulatory loop between the CLAVATA and WUSCHEL genes. Cell 100,

98 Chapter 4 Schopfer, C.R., Nasrallah, M.E., and Nasrallah, J.B. (1999). The male determinant of selfimcompability in Brassica. Science 286, Sharma, V.K., Ramirez, J., Fletcher, J.C. (2003). The Arabidopsis CLV3-like (CLE) genes are expressed in diverse tissues and encode secreted proteins. Plant Mol. Biol. 51, Trotochaud, A.E., Hao, T., Wu, G., Yang, Z., and Clark, S.E. (1999). The CLAVATA1 receptor-like kinase requires CLAVATA3 for its assembly into a signalling complex that includes KAPP and a Rho-related protein. Plant Cell 11, Wang, X., Mitchum, M.G., Gao, B., Li, C., Diab, H., Baum, T.J., Hussey, R.S., and Davis, E.L. (2005). A parasitism gene from a plant-parasitic nematode with function similar to CLAVATA/ESR (CLE) of Arabidopsis thaliana. Mol. Plant Path. 6, Wysocka-Diller, J. W., Helariutta, Y., Fukaki, H., Malamy, J.E., and Benfey, P.N. (2000). Molecular analysis of SCARECROW function reveals a radial patterning mechanism common to root and shoot. Development 127, Yang, H., Matsubayashi, Y., Nakamura, K., and Sakagami, Y. (1999). Oryza sativa PSK gene encodes a precursor of phytosulfokine-α, a sulfated peptide growth factor found in plants. Proc. Natl. Acad. Sci. USA 96,

99 Chapter 5 The CLE motif of CLAVATA3 is functionally independent from the non-conserved flanking sequences Martijn Fiers a,b, Elzbieta Golemiec c, Roel van der Schors d, Lonneke van der Geest a, Ka Wan Li d, Willem J. Stiekema b and Chun-Ming Liu d a Plant Research International, B.V. P.O. Box 16, 6700 AA Wageningen, The Netherlands b Centre for BioSystems Genomics, PO Box 98, 6700 AB Wageningen, The Netherlands c Institute of Plant Physiology, Polish Academy of Sciences, Niezapominajek 21, Kraków, Poland d Free University, Department of Molecular & Cellular Neurobiology, De Boelelaan 1085, 1081 HV Amsterdam, The Netherlands e Center for Signal Transduction & Metabolomics, Key Laboratory of Photosynthesis and Environmental Molecular Physiology, Institute of Botany, Chinese Academy of Sciences, Beijing , China Published in Plant Physiology 141, (2006) Cover Picture in colour in appendix, page VIII

100 Chapter 5 100

101 The Critical Role of the CLE Motif in CLV3 ABSTRACT It is believed that CLAVATA3 (CLV3) encodes a peptide ligand that interacts with the CLV1/CLV2 receptor complex to limit the number of stem cells in the shoot apical meristem (SAM) of Arabidopsis (Arabidopsis thaliana), however, the exact composition of the functional CLV3 product remains a mystery. A recent study on CLV3 shows that the CLV3/Embyo Surrounding Region (CLE) motif together with the adjacent C-terminal sequence is sufficient to execute the CLV3 function when fused behind a N-terminal sequence of ERECTA. Here we show that most of the sequences flanking the CLE motif of CLV3 can be deleted without affecting CLV3 function. Using a liquid culture assay, we demonstrate that CLV3p, a synthetic peptide corresponding to the CLE motif of CLV3, is able to restrict the size of the SAM in clv3 seedlings, but not in clv1 seedlings. In accordance with this decrease in meristem size, application of CLV3p to in vitro grown clv3 seedlings restricts the expression of the stem-cell promoting transcription factor WUSCHEL (WUS). Thus, we propose that the CLE motif is the functional region of CLV3 and that this region acts independently of its adjacent sequences. INTRODUCTION Stem cells positioned in the central zone of the plant shoot apical meristem (SAM) are the source of totipotent cells, which continuously give rise post-embryonically to new organs (Steeves and Sussex, 1989; Weigel and Jürgens, 2002). These slow-dividing cells simultaneously maintain two antagonistic events, cell proliferation and cell differentiation, in a similar manner to animal stem cells (Groß-Hardt and Laux, 2003; Ohlstein et al., 2004; Scheres, 2005). The determination of the fate of the meristem progeny cells occurs by a population-based mechanism in which signals from neighboring cells play the most important role (Spradling et al., 2001; Weigel and Jürgens, 2002). Genetic experiments have shown that, as part of a feedback regulatory loop, the stem cell-promoting homeodomain transcription factor WUSCHEL (WUS), which is expressed in the stem cell organizing center (OC), provides a positive signal to maintain an undifferentiated state, while CLAVATA3 (CLV3) interacts with the underlying CLV1/CLV2 receptor complex to generate a negative signal that 101

102 Chapter 5 limits WUS expression, and in this way is able to restrict the number of stem cells (Brand et al., 2000; Schoof et al., 2000). Thus, mutation in WUS leads to a termination of the SAM, whereas mutations in any of the CLV genes results in the expansion of WUS expression and subsequently an enlarged SAM with an increased number of stem cells (Clark et al.,1997; Mayer et al., 1998; Fletcher et al., 1999; Jeong et al., 1999). CLV3 is able to restrict its own expression by preventing the differentiation of the cells in central zone (Reddy and Meyerowitz, 2005). Genetic data suggest that CLV3 encodes a mobile ligand that acts in a non-cell autonomous fashion in intercellular communication (Trotochaud et al., 1999; Lenhard and Laux, 2003; Rojo et al., 2002). CLV3 belongs to a family of small proteins, named CLV3/Embryo Surrounding Region (CLE) that is found in plants and parasitic nematodes (Opsahl-Ferstad et al., 1997; Fletcher et al., 1999; Cock and McCormick, 2001; Hobe et al., 2003; Olsen and Skriver., 2003; Wang et al., 2005). CLE proteins share an N-terminal secretion signal (SS) and a conserved 14-amino acid (AA) CLE motif at or near their C-termini. The internal sequence between the SS and the CLE motif is generally not conserved, and a C-terminal extension is only found in few CLE members, including CLV3. Over-expression of several CLE genes, such as CLV3, CLE19 and CLE40 from Arabidopsis (Arabidopsis thaliana) and HgSYV46 from the nematode Heterodera glycines, causes a termination of root development (Casamitjana-Martinez et al., 2003; Hobe et al., 2003; Fiers et al., 2004; Wang et al., 2005; Strabala et al., 2006). In vitro application of synthetic peptides, CLV3p, CLE19p or CLE40p corresponding to the CLE motif of their encoding CLE proteins phenocopies the overexpression phenotype in a CLV2-dependent manner (Fiers et al., 2005). At the cellular level, these treatments lead to a mis-specified cell identity and the premature differentiation of ground tissue daughter cells, suggesting that these CLE peptides are able to interact with or saturate an unknown cell identity-maintaining receptor complex in roots (Birnbaum et al., 2003; Fiers et al., 2005). Moreover, several CLE genes could complement clv3 mutants when expressed under the control of CLV3 regulatory elements, indicating the functional conservation among different CLE members (Hobe et al., 2003, Wang et al., 2005, Ni and Clark, 2006). A recent study also showed that the CLE motif and its downstream sequence alone can complement the clv3-1 mutation when fused behind the secretion machinery (an N- terminal SS fragment and downstream 46 AA) of ERECTA, demonstrating the critical 102

103 The Critical Role of the CLE Motif in CLV3 importance of these regions (Ni and Clark, 2006). Whether the CLE motif of CLV3 itself is sufficient to perform CLV3 function remains to be examined. In the present study several deletion constructs of CLV3 were made and used to determine which domains are essential for CLV3 function by complementing the clv3 mutant. Our results show that, beside the SS (Rojo et al., 2002), the only other essential domain is the CLE motif. Removal of the sequence between the SS and the CLE motif, or the sequence after the CLE motif did not affect the complementation of the clv3 mutant, while the removal of the CLE box abolishes CLV3 function completely. Using an in vitro assay, we showed that the synthetic peptide CLV3p is able to restrict the size of the SAM in clv3 seedlings, but not that in clv1, by restricting WUS expression, suggesting that the peptide acts in a similar manner as the endogenous gene. Peptides from different CLE genes confer various degrees of complementation. As such, we provide evidence that the CLE motif of CLV3 as well as several other CLEs acts independent of their flanking sequences. RESULTS Deletion Analysis of the CLV3 Gene Except the CLE motif (Fig. 1, in frame) very little AA similarity could be identified among the various CLE proteins (Cock and McCormick, 2001). Figure 1. Alignment of CLV3 and the synthetic peptides used for the in vitro assay. Identical AAs are shaded in black and similar ones in grey, the signal sequences is underlined, and the CLE box is framed with the consensus sequence underneath. 103

104 Chapter 5 Figure 2. Functional analysis of CLV3. A, Full-length CLV3 genomic sequence (construct #1) including 5 upstream and 3 downstream regulatory sequences (line with breaks) and 4 deletion constructs (#2 to #5) were made by removing the sequence between SS and CLE motif (#2); part of this sequence (#3); the sequence after the CLE motif (#4), or the CLE motif and the sequence after (#5). 104

105 The Critical Role of the CLE Motif in CLV3 The SS is shown in gray, CLE motif in black, the sequence between SS and CLE motif in white, and the sequence after the CLE motif with vertical lines. The positions of two introns are depicted with arrowheads. Data and error bars on the right represent the mean number of carpels per sillique from transgenic clv3-2 (for each construct, minimal 14 T 0 -transgenic plants with 30 silliques per plant were counted). B to E, The inflorescence and meristem of WT (Ler) Arabidopsis (B, C) and clv3-2 (D, E). F to O, The infloresence and meristem of transgenic clv3-2 carrying construct #1 (F, G), #2 (H, I), #3 (J, K), #4 (L, M) and #5 (N, O), respectively. Note the restoration to WT in transgenic clv3-2 plants carrying constructs #1 to #4. The bar in C represents 50 µm for all SAM pictures. The arrowheads indicate the lower edges of the SAM and the area of the SAM was measured above a straight line between these two arrowheads. P, The area of the SAM from clv3-2, WT and different transgenic lines. The data and error bars represent the mean area in median sections ± SD (T 1, n 14). To examine if these non-conserved sequences were critical for CLV3 function, several deletion constructs were made in which different domains of CLV3 were removed (constructs #2 to #5; Fig. 2A) and compared with the full-length CLV3 gene (construct #1). In construct #2, the sequence between the SS and the CLE motif, including the two introns, was removed, whereas in construct #3 only the sequence after the second intron and the CLE motif was removed, leaving the two introns intact. In construct #4, all of the coding sequence downstream of the CLE motif was deleted; and in construct #5, both the CLE motif and its downstream sequence were removed. These constructs, which were fused to the CLV3 5 and 3 regulatory elements (Hobe et al., 2003), were used to transform clv3-2 to examine if they could restore the phenotype to wild type (WT). The clv3-2 mutant has a typical cone-shaped SAM in which the leaf primordia appear only at the lower half of the structure (Fig. 2E). Due to the enlargement of the floral meristem as well as the SAM, the number of carpels per silique was increased from two in the WT to five in clv3-2 (Clark et al., 1995). Carpel number was used as an indirect measure of meristem size (Clark et al., 1995; Ni and Clark, 2006). Restoration of the wild-type carpel number was observed in all transgenic plants carrying constructs with a the CLE box (Figs. 2F-2M), while no restoration was observed in plants carrying a construct in which both CLE domain and C-terminus were deleted (Figs. 2N and 2O). Occasionally, some silliques in transgenic plants carrying a construct #2 or #3 had three carpels, resulting in an average carpel number of slightly more than two (Fig. 2A, T 0 plants n 14). 105

106 Chapter 5 Figure 3. Effect of different peptides on the SAMs of clv3-2. A to H, SAM of clv3-2 seedlings incubated with no peptide (A) or 10 µm of CLV3m (B), CLVt3 (C), CLE5p (D), CLE22p (E), CLE19p (F), CLE40p (G) and CLV3p (H) for 8 days and examined with Nomarski optics. Note a nearly full restoration of the size of the SAM to WT upon incubations with CLE40p (G) or CLV3p (H). 106

107 The Critical Role of the CLE Motif in CLV3 I, SAM of an 8-day old Ler seedling. J and K, SAM of clv1-1 seedlings incubated with no peptide (J) or CLV3p (K). L, Area of the SAM of clv3-2 after different peptide treatments for 8 days, in comparison with Ler. M, Area of the SAM of clv1-1 after peptide treatments for 8 days, in comparison with Ler. The area of the SAM was measured on a median plane by calculating the area above the straight line from the upper edges of two opposite leaf primordia (arrowheads). The data and error bars represent the mean ± SD (n 20 in minimal 2 independent experiments). The bar in A represents 50 µm for A to K. The size of SAM in the transgenic plants was measured in median sections obtained through Normaski optics (Figs. 2C, 2E, 2G, 2I, 2K, 2M and 2O). The results showed that all constructs containing the CLE domain restored the SAM more or less to the size of the WT ( µm 2 for Ler), ranging from µm 2 for plants carrying construct #4 to µm 2 for construct #3. In contrast, plants carrying a construct without CLE domain (#5) had a SAM size of over 14,233+2,829 µm 2 (Fig. 2P), which is significantly different from transgenic plants carrying constructs #1 to #4 (p > 0.01). Why the SAM in these plants was slightly larger than that in clv3-2 (10,520+2,812 µm 2 ) is not known. It is possible that the RNA or protein produced by the construct #5 might affect the stability of endogenous CLEs, causing an enhanced phenotype. If this is the case, the effect must be specific for the SAM in the seedlings since the carpel number was not increased. No other phenotypes such as the termination of the SAM, as seen in plants over-expressing CLV3, were observed in any of the transgenic lines. In Vitro Peptide Assay In a previous report, we showed that addition of the CLV3p to solidified medium could not rescue the enlarged SAM phenotype of the clv3-2 mutant (Fiers et al., 2005), suggesting that the peptide is either non-functional in the SAM, or that it can not be transported efficiently from the roots to the SAM. To ensure sufficient access of the peptide to the SAM, we developed a protocol in which seedlings were grown in a liquid culture to which peptides were added. Besides CLV3p, five peptides derived from CLV3p were synthesized and used as controls (Fig. 1). They were: 1) CLV3m that is similar to CLV3p but with a single AA change from G to A, as in clv3-1 and clv3-5 mutants (Fletcher et al., 1999); 2) three truncated peptides of 107

108 Chapter 5 CLV3p in which a certain AAs were deleted from the ends of the peptides (CLV3t1-3); 3) CLV3s, a peptide with the same AA composition as CLV3p but randomized in sequence. To examine the sequence conservation among different CLEs, four additional peptides, CLE5p, CLE19, CLE22p and CLE40p, were made based on the CLE motif of CLE5, CLE19, CLE22 and CLE40, respectively. These four CLE members are all expressed in reproductive shoot apices (Sharma et al., 2003; Hobe et al., 2003; Fiers et al., 2004). All peptides were individually applied to liquid media in the same concentration (10 µm) as used in the previous root assay (Fiers et al., 2005). The seedlings were examined at successive time points for changes in the size of the SAM. Seedlings of clv3-2 treated with any of the CLV3p derivatives; CLV3m, CLV3s, CLV3t1, CLV3t2 or CLV3t3 displayed a large cone-shaped SAM, as seen in the control treatment without peptide (Figs. 3A-C), while SAMs of seedlings treated with CLV3p were much smaller and resembled those of the WT (Figs. 3H and 3I). Consecutive samplings showed that the effect of CLV3p in restricting the SAM was most evident after 8 days of treatment (data not shown). No termination of the SAM was observed. The size of individual cells in the SAM was the same as in the non-treated seedlings, which excluded the possibility that the reduction in meristem size was caused by reduced cell expansion. We digitally measured the surface area of the median section of cleared SAMs (Fig. 3L). Strikingly, the SAM of clv3-2 treated with CLV3p was only about 1, µm 2, which resembles that of Ler seedlings, but not that of clv3-2. None of the tested truncated peptides promoted a reduction in SAM size, demonstrating that sequence integrity of the CLE motif is required for its function. The specificity of CLV3p is further illustrated by the fact that CLV3m did not restore the clv3-2 phenotype, and further illustrates the concordance between the genetic mutation and in vitro data. Similarly, peptides of CLE5p, CLE22p, CLE19p and CLE40p were tested in the liquid culture. The results showed that CLE40p could fully complement the clv3-2 phenotype (Fig. 3G), while application of CLE5p and CLE19p both resulted in a partial reduction of around 50% in the SAM size (Figs. 3D, 3F and 3L). This is in agreement with previously reported complementation data (Hobe et al, 2003; Ni and Clark, 2006). However, CLE22p only led to a minor reduction on clv3-2 SAM (Figs. 3E and 3L). Statistical analysis showed that this 108

109 The Critical Role of the CLE Motif in CLV3 reduction is not significantly different from the control without peptide (p > 0.1). This result differs from that obtained by a transgenic approach in which PCLV3:CLE22 is able to partially complement clv3-1 (Ni and Clark, 2006). Although an arrest of SAM development was observed upon over-expression of the CLV3 gene (Brand et al., 2000), a similar arrest was not observed after peptide treatment in liquid culture. This is different from the results obtained in roots, where both over-expression of CLV3 and treatment with CLV3p led to a termination of the root meristem (Hobe et al., 2003; Fiers et al., 2005). One possibility could be the presence of a cuticle layer in the SAM, which may form a barrier for efficient peptide uptake. To elucidate whether CLV3p acts through the CLV1/CLV2 complex in the same manner as the endogenous CLV3, we incubated the clv1-1 mutant with CLV3p, CLV3m, CLV3s and CLE40p. No significant differences were observed in the SAM (Figs. 3J-K and 3M), suggesting that a functional CLV1 receptor is required for the perception of the peptide signal. Stability of the Peptide Our data above showed that the treatments with CLV3p or CLE40p did not lead to termination of the SAM, a result which differs from those obtained by over-expressing CLV3 or CLE40 (Brand et al., 2000; Hobe et al., 2003). Furthermore, although the size of the clv3-2 SAM reverted almost to that of the WT (Ler) after 8-days of treatment, it expanded again after a prolonged incubation (data not shown). Therefore, we speculated that these peptides may have been degraded during incubation. To test this, we examined the peptide stability using matrix-assisted laser-desorption ionization time-of-flight (MALDI-TOF) mass spectrometry (MS). The results showed that CLV3p in medium without seedlings was rather stable and no significant breakdown products were detected during a culture period of two weeks (data not shown). In contrast, the first sign of peptide degradation was observed in peptide-containing media incubated with seedlings after 2 days (Fig. 4A), after which CLV3p was gradually degraded and was no longer present on the 8 th day of culture (Fig. 4C-D). This seems to contradict the observation that the most evident effect of the peptides in restoring the clv3-2 phenotype was observed on the 8 th day of treatment. We believe that the delayed response reflects the time course required for the signal to be incorporated into the developmental program of the SAM. 109

110 Chapter 5 One of the major truncated degradation products of the CLE peptides had a molecular mass of 1,156 Da and was consistently found during incubation. MALDI-TOF analysis of the fragment showed that one AA was removed from the N-terminus and two from the C- terminus (Fig. 4C) and this truncated peptide therefore corresponds to the synthesized CLV3t3 used in this study. Figure 4. CLV3p is gradually degraded during incubation with Arabidopsis seedlings. A to D, MALDI-TOF MS analysis of the media during the course of incubation with seedlings of Arabidopsis (Ler) after 2 days (A), 5 days (B), 7 days (C) and 8 days (D). The CLV3p and CLV3t3 peak are indicated. Further experiments with CLV3t3 showed that it was unable to restrict the size of the SAM in vitro, suggesting that this degradation might be non-specific. The rapid degradation of the peptides in the media may explain partially why 1) the size of the SAM of clv3 enlarges gradually after 8 days; 2) a relative high concentration of peptide is needed for complementation in the in vitro culture experiments; and 3) no termination of the SAM was observed. WUS Expression Pattern is Restored upon CLV3p Peptide Treatment We addressed whether CLV3p, like the full-length CLV3 protein, acts by restricting the size of SAM via the WUS transcription factor (Brand et al., 2000; Schoof et al., 2000). In situ hybridisation was performed to examine WUS expression in the clv3-2 mutant after peptide treatments. WUS transcripts in clv3-2 were observed in a large span of cells in the upper L3 and occasionally in the L2 layers of the SAM, but excluded from the pith region (Fig. 5A), which is consistent with results observed previously (Schoof et al., 2000; Brand et al., 2000). 110

111 The Critical Role of the CLE Motif in CLV3 Figure 5. WUS expression in clv3-2 upon incubation with CLV3p. A to C, In situ hybridization using an antisense probe of WUS cdna, showing the WUS expression in the SAM of clv3-2 after 8 days of treatment. Treatment with no peptide (A) or CLV3m (B) resulted in an expanded WUS expression as compared to clv3-2 treated with CLV3p (C) in which the WUS expression is restricted to the OC as seen in WT. D, The control sample treated with CLV3m and hybridized with the WUS sense probe. The bar in A represents 50 µm for all samples. After treatment with CLV3m, the WUS expression pattern was almost identical to that of the non-treated seedlings (Figs. 5A and 5B). In contrast, seedlings treated with CLV3p for 8 days had a much smaller and condensed WUS domain (Fig. 5C), as seen in WT seedlings (Mayer et al., 1998; Schoof et al., 2000), and the number of cells expressing WUS was also reduced significantly (Figs. 5A and 5C). DISCUSSION CLV3 is believed to function as a mobile ligand that binds to the CLV1/CLV2 receptor complex (Fletcher et al., 1999; Brand et al., 2000; Schoof et al., 2000). Although the SS of CLV3 has been shown to be functional, it has been proven to be difficult to detect the presence of the endogenous CLV3 product (Rojo et al., 2002; Fiers et al., 2004). The only endogenous CLE protein that has been detected in plants is the Endosperm Related (ESR) 111

112 Chapter 5 protein in maize (Bonello et al., 2002). A recent report using domain swap experiments and complementation analysis suggested that the CLE motif is of critical importance for CLV3 function (Ni and Clark, 2006). In the present article, we provide direct evidence to show that all sequences except the SS and the CLE motif can be removed without affecting the CLV3 function. Furthermore, treatment of seedlings in a liquid culture with peptides corresponding to the CLE motif of CLV3 and several other CLEs restricted the size of the SAM via a CLV1- depedent pathway. The CLE Motif and the SS of CLV3 are Sufficient to Complement clv3 Deletion analysis showed that only the SS and the CLE motif are necessary for CLV3 function, which allows us to exclude the possibility that the flanking sequences contribute to the function of CLV3. This is in agreement with the results from domain swap experiments, in which the SS and the variable sequence between SS and the CLE motif could be replace by a similar fragment from ERECTA (Ni and Clark, 2006). Whether processing is required to generate a CLV3 ligand, and how such processing might take place and the nature of the recognition site remain to be investigated. Research in this direction is of general interest since several peptide ligands, such as phytosulphokine and systemin, have been identified biochemically in plants (Pearce et al., 1991; Yang et al., 1999), but no cleavage recognition sites have been found yet. CLV3p is Functional in vitro, as a Short Distance Diffusible Signal Our previous work shows that synthetic peptides corresponding to the CLE motif of CLV3, CLE19 and CLE40, when applied to solid media with germinating seedlings, are able to mimic the over-expression phenotype of these three genes, leading to a termination of root meristem development (Hobe et al., 2003; Fiers et al., 2004; Fiers et al., 2005). Here we demonstrate using a liquid culture system that the 14-AA CLV3p can complement the meristem phenotype of clv3-2 seedlings, restoring the size of the SAM to that of WT. The extracellularly supplied 14-AA peptide (CLE motif of CLV3) is therefore necessary and sufficient for CLV3 function (Fig. 3H). The reversion of the SAM after the peptide treatment suggests a reversible interaction of CLV3p with its receptors. Considering our earlier experiments in which no complementation was observed in the SAM when the peptide was applied to the roots in solid medium, we believe that the peptide may act as a short-distance 112

113 The Critical Role of the CLE Motif in CLV3 diffusible signal which moves through the intercellular spaces, as was observed in the root tip (Fiers et al., 2005). Different CLE Genes May Act in a Similar Manner Examination of peptides corresponding to several other CLEs in the in vitro culture assay revealed a high functional conservation among different members of the CLE family. The differences between the CLE genes in complementing the clv3 phenotype were also observed using a genetic approach (Ni and Clark, 2006). A major inconsistency between the peptide assay and the transgenic analysis is for CLE22, which is partially functional in complementing clv3-1 in the transgenic approach (Ni and Clark, 2006), but was not able to restore the clv3-2 mutation in the in vitro peptide assay. Two reasons could explain this discrepancy. Firstly, it may result from the difference in alleles used in each study; clv3-1 (with a point mutation in the CLE motif and an intermediate phenotype) used by Ni and Clark (2006) is weaker than clv3-2 (a deletion in the coding region, resulting in a strong phenotype) used in this study. Secondly, the differences in developmental stages may also contribute to the differences. The peptide assay was performed at the seedling stage and was quantified using the size of the primary meristem; while the transgenic assay is based on the phenotype at the mature stage (carpel numbers in the fruits; Ni and Clark, 2006). It is possible that the unknown receptor involved in perception of CLE22, as pointed out by Ni and Clark (2006), is not available at the seedling stage. The observations that not all CLE genes are inter-changeable and that they might therefore interact with different receptors point out an additional complexity in the signal perception of the CLE proteins. It helps to explain why transgenic plants over-expressing CLE19 and several other CLE genes don t show termination of the SAM as observed upon overexpression of CLV3, CLE40 or even HgSYV46 (Brand et al., 2000; Hobe et al., 2003; Fiers et al., 2004; Wang et al., 2005; Strabala et al., 2006). CLV3p is Unstable in Culture Medium CLV3p applied to the culture medium of developing seedlings restricted the expansion of the SAM only for a limited period of time. This is in contrast to the over-expression of the CLV3 gene under the control of the CaMV 35S promoter, where a termination of the SAM was observed as a consequence of down regulating WUS expression (Brand et al., 2000; Hobe et 113

114 Chapter 5 al., 2003). We believe that this difference can be explained by the observation that peptides applied to the medium were degraded in the presence of the seedlings in addition to the presence of a cuticle layer in the SAM, which may form a barrier for efficient peptide uptake. One major processed product observed in the seedling assay was identical to CLV3t3, which removed two AA from the N- and one from the C-terminus. Peptide assay showed that CLV3t3 could not complementclv3, suggesting that the processing may be not functionally relevant. Similarly, Ni and Clark (2006) have also observed N-terminal processing of CLV3 using a cauliflower extract, which cleaved heterologously produced GST-CLV3 fusion protein on at least two positions, one after M41, and another one after R70 (identical to CLV3t2 at the N-terminus). Since a GST antibody was used to detect potential processing, any additional processing occurred downstream of the CLE motif in combination with the observed cleavage may not be detected. CLV3p Application In Vitro Mimics the Endogenous CLV3 Gene In the clv3 mutant, the expression domain of the stem cell-promoting homeodomain transcription factor WUS is expanded to include a large number of cells in the L2 and L3 layers of the cone-shaped SAM, rather than the a compact cell cluster in the L3 layer of WT SAMs (Brand et al., 2000). CLV3p applied in the culture medium appears to acts in a similar fashion as the endogenous CLV3 protein, as WUS expression was restricted to a small cluster of cells in the OC. How the CLV3p peptide restricts the size of the SAM remains unknown. Most likely it triggers the meristematic cells in the periphery into the differentiation and organogenesis program through a signal transmitted by the CLV1/CLV2 receptor complex, after activation by CLV3, towards WUS (Schoof et al., 2000; Brand et al., 2000). To determine if this is indeed the case, clv1-1 seedlings were incubated with different CLE peptides. The clv1-1 allele was selected since it carries a dominant negative point mutation in the intracellular kinase domain of CLV1 that is thought to interfere with the function of both CLV1 and other overlapping receptor kinases (Clark et al., 1997; Diévart et al., 2003; DeYoung et al., 2006). The size of the SAM in clv1-1 was not significantly changed upon incubation with CLV3p or any of the control peptides. Therefore, we propose that CLV3p, like endogenous CLV3, acts in a CLV1-114

115 The Critical Role of the CLE Motif in CLV3 dependent manner (Trotochaud et al., 1999), although we could not exclude the possibility that redundant receptor kinases are also involved. In summary, our results showed that the CLE motif of CLV3 is sufficient to mimic CLV3 function both in vivo (together with the SS) and in vitro in a sequence-dependent manner. The data obtained so far support the hypothesis that the 14-AA CLE motif is the functional part of CLV3, in restricting the stem cell population in the SAM of Arabidopsis. The consistency between the in vitro peptide treatment and the in vivo transgenic approach with different CLE genes performed by Ni and Clark (2006) showed that the mode of action of the CLE peptides and the corresponding genes is the same and led us to propose that in the case of CLV3 the CLE motif is the functional part and can act independent of the non-conserved flanking sequences. MATERIALS AND METHODS Peptide Assay WT (Ler), clv1-1 and clv3-2 (all in a Ler background) were provided by the Nottingham Arabidopsis Stock Center (Nottingham, UK). Seeds were gas-sterilized in a desiccator for 1 hr with a mixture of 100 ml of kitchen bleach (containing 4% sodium hypochlorite) and 3 ml of concentrated HCl. These seeds were incubated in liquid medium containing 0.5x Murashige & Skoog salt mixture (Duchefa, Haarlem, NL), 1 % sucrose, 0.5 g/l 2-(Nmorpholino) ethane-sulfonic acid (MES, ph 5.8) for 3 days at 4 C. Afterwards the medium was refreshed and peptides were added at a concentration of 10 µm. Incubation was performed in a 50 ml Falcon tubes (20 seeds/tube) with 6 ml of media on a roller bank, 23 C, 16 hr light per day. All peptides were synthesized by MIMOTOPES (Roseville, Australia) with a purity of >80%. Whole-Mount Examination Shoot apices of 8-day old seedlings were excised under a dissecting microscope, cleared following the protocol of Sabatini et al. (1999) and analyzed under a Nikon microscope equipped with Nomarski optics. From each SAM a picture was taken from the median optical section. The surface area was defined by measuring the area of the meristem dome above a 115

116 Chapter 5 straight line between the top edges of the smallest leaf primordia. All areas were measured using the ImageJ program ( MALDI-TOF MS Analysis During the incubation of the seedlings in peptide-containing media, samples were taken once a day and frozen immediately for later analysis. These samples were diluted 200 times with 0.1% tri-fluoric acid, and spotted (0.5 µl each) on a stainless steel target, together with 0.5 µl of α-cyano-4-hydroxycinnamic acid (dissolved until saturation in 60% acetonitrile and 1% tri-fluoric acid,) for analysis in a MALDI-TOF mass spectrometer (Proteomics Analyzer, ABI4700). As a control, the stability of the CLV3p was measured by adding the peptide to the same medium without seedlings and incubating under the same conditions, before analyzing as above. In Situ Hybridization Eight-day old seedlings with or without peptide treatment were fixed for 2 hr in a modified formaldehyde acetic acid solution (Liu et al., 1993), and embedded in paraffin. Sections with a thickness of 10 µm were prepared and used for non-radioactive in situ hybridization using the mrnalocater kit (Ambion, Cambridgeshire, UK), following the protocol of the manufacturer. A fragment of the WUS cdna was amplified from Arabidopsis (Arabidopsis thaliana) cdna, using two sets of primers (one with a T7 promoter), namely 5 - ATATAATACGACTCACTATAGCTCGTGAGCGTCAGA-AG-3, 5 -GAAGCGTACGTCGATGTTC-3 and 5 -ATATAATACGACTCACTATAGAAGCGTACG-TCGATGTTC-3, 5 -GCTCGTGAG-CGTCAGAAG-3. The first pair of primers was used to produce the sense and the latter one was used to produce the anti-sense RNA probes through in vitro transcription using T7 polymerase, labeled with digoxigenin (Roche, Almere, NL). In situ hybridization and immunological detection were performed as previously described (Cañas et al., 1994). Complementation Analysis The full-length CLV3 (construct #1) including the CLV3 coding region (including two introns) plus a 1.8 kb promoter and 1.5 kb terminator (Hobe et al., 2003), was generated using the primers MF1 (5 -GACAAGTTTGTACAAAAAAGCAGGCT CAGTCTCTTGTCGCTT- AACG- 3) and MF2 (5 -GACCACTTTGTACAAGAAA-GCTGGGTGATCAATTACT- 116

117 The Critical Role of the CLE Motif in CLV3 AACTACAATGG- 3) including the Gateway cloning sites (in bold). The promoter and terminator fragments for the deletion constructs (construct #2 to #5) were generated using MF1/MF3 (5 -ACAGTCCTTAACTCTTCATGAGAAGCATCATGAAGGAACA- 3) and MF2/MF4 (5 -TGTTCCTTCATGATGCTTCTCATGAAGAGTTAAGGACTGT- 3) for construct #2, MF1/MF5 (5 -ACAGTCCTTAACTCTTCATGCATCTGCCAATTGA- ACAAC- 3) and MF2/MF6 (5 -GTTGTTCAATTGGCAGATGCATGAAGAGTTAAG- GACTGT- 3) for construct #3, MF1/MF7 (5 -AGCAACAAGAGATTAGGTCAATGAT- GGTGCAACGGGTCAG- 3) and MF2/MF8 (5 -CTGACCCGTTGCACCATCATTGACCT- AATCTCTTGTTGCT- 3) for construct #4, and MF1/MF9 (5 -AGCAACAAGAGA- TTAGGTCACTCTTCATGTAGTCCTAAAC- 3) and MF2/MF10 (5 -GTTTAGGACTAC- ATGAAGAGTGACCTAATCTCTTGTTGCT- 3) for construct #5, respectively. The promoter and terminator fragments were combined using PCR with overlapping primers after which the complete deletion construct was generated via PCR using the primers MF1 and MF2. The PCR fragment was recombined into pdonr207 (BP reaction) following the protocol of the supplier (Invitrogen, Breda, NL). The binary vector was obtained with an LR-reaction using the plasmids from the BP reaction mixed with the pkgw vector following the protocol of the supplier (Invitrogen). The binary vectors, containing the different CLV3 fragments, were sequenced before transformation to Agrobacterium tumefaciens C58pmp90. The clv3-2 plants were transformed using the flower-dip method as described (Clough and Bent, 1998). ACKNOWLEDGEMENTS We thank Kim Boutilier for critical reading of the manuscript, X. Yin and Jan Kodde for statistical analysis, A. Shchennikova for advice on in situ hybridization and NASC for providing clv1 and clv3. This work was supported in part by the Dutch Ministry of Agriculture, Nature Management and Fisheries (DWK281/392) and by the Centre for BioSystems Genomics (CBSG), which is part of the Netherlands Genomics Initiative/Netherlands Organization for Scientific Research. E. G. was supported by the EU CROPSTRESS project (QLAM ). 117

118 Chapter 5 LITERATURE CITED Brand U, Fletcher JC, Hobe M, Meyerowitz EM, Simon R (2000) Dependence of stem cell fate in Arabidopsis on a feedback loop regulated by CLV3 activity. Science 289: Birnbaum K, Shasha DE, Wang JY, Jung JW, Lambert GM, Galbraith DW, Benfey PN (2003) A gene expression map of the Arabidopsis root. Science 302: Bonello JF, Sevilla-Lecoq S, Berne A, Risueno MC, Dumas C, Rogowsky PM (2002) Esr proteins are secreted by the cells of the embryo surrounding region. J Exp Bot 53: Cañas LA, Busscher M, Angenent GC, Beltran JP van Tunen AJ (1994) Nuclear localization of the petunia MADS box protein FBP1. Plant J 6: Casamitjana-Martinez E, Hofhuis HF, Xu J, Liu CM, Heidstra R, Scheres B (2003). Root-specific CLE19 over-expression and the sol1 and sol2 suppressors implicate a CLV-like pathway in the control of Arabidopsis root meristem maintenances. Curr Biol 13: Clark SE, Running MP, Meyerowitz EM (1995) CLAVATA3 is a specific regulator of shoot and floral meristem development affecting the same processes as CLAVATA1. Development 121: Clark SE, Williams RW, Meyerowitz EM (1997) The CLAVATA1 gene encodes a putative receptor kinase that controls shoot and floral meristem size in Arabidopsis. Cell 89: Clough SJ, Bent AF (1998) Floral dip: a simplified method for Agrobacterium-mediated transformation of Arabidopsis thaliana. Plant J 16: Cock JM, McCormick S (2001) A large family of genes that share homology with CLAVATA3. Plant Physiol 126: DeYoung BJ, Bickle BL, Schrage KJ, Muskett P, Patel K, Clark SE (2006) The CLAVATA1-related BAM1, BAM2 and BAM3 receptor kinase-like proteins are required for meristem function in Arabidopsis. Plant J. 45: 1-16 Diévart A, Dalal M, Tax FE, Lacey AD, Huttly A, Li J, Clark SE (2003) CLAVATA1 dominant-negative alleles reveal functional overlap between multiple receptor kinases that regulate meristem and organ development. Plant Cell 15: Fiers M, Golemiec E, Xu J, van der Geest L, Heidstra R, Stiekema W, Liu CM (2005) The 14-amino acid CLV3, CLE19 and CLE40 peptides trigger consumption of the root meristem in Arabidopsis through a CLAVATA2-dependent pathway. Plant Cell 17:

119 The Critical Role of the CLE Motif in CLV3 Fiers M, Hause G, Boutilier K, Casamitjana-Martinez E, Weijers D, Offringa D, van der Geest L, van Lookeren Campagne M, Liu CM (2004) Mis-expression of the CLV3/ESRlike gene CLE19 in Arabidopsis leads to a consumption of root meristem. Gene 327: Fletcher JC, Brand U, Running MP, Simon R, Meyerowitz EM (1999) Signaling of cell fate decisions by CLAVATA3 in Arabidopsis shoot meristems. Science 283: Groß-Hardt R, Laux T (2003) Stem cell regulation in the shoot meristem. J.Cell Sci 116: Hobe M, Muller R, Grunewald M, Brand U, Simon R (2003) Loss of CLE40, a protein functionally equivalent to the stem cell restricting signal CLV3, enhances root waving in Arabidopsis. Dev Genes Evol 213: Jeong S, Trotochaud AE, Clark E (1999) The Arabidopsis CLAVATA2 gene encodes a receptor-like protein required for the stability of the CLAVATA1 receptor-like kinase. Plant Cell 11: Lenhard M, Laux T (2003) Stem cell homeostasis in the Arabidopsis shoot meristem is regulated by intercellular movement of CLAVATA3 and its sequestration by CLAVATA1. Development 130: Liu CM, Xu Z, Chua NH (1993) Auxin Polar Transport Is Essential for the Establishment of Bilateral Symmetry during Early Plant Embryogenesis. Plant Cell 5: Mayer KF, Schoof H, Haecker A, Lenhard M, Jurgens G, Laux T (1998) Role of WUSCHEL in regulating stem cell fate in the Arabidopsis shoot meristem. Cell 95: Ni J, Clark SE (2006) Evidence for functional conservation, sufficiency, and proteolytic processing of the CLAVATA3 CLE domain. Plant Physiol. 140: Ohlstein B, Kai T, Decotto E, Spradling A (2004) The stem cell niche: theme and variations. Curr Opin Cell Biol. 16: Olsen, AN, Skriver, K (2003) Ligand mimicry? Plant-parasitic nematode polypeptide with similarity to CLAVATA3. Trends Plant Sci 8: Opsahl-Ferstad HG, Le Deunff E, Dumas C, Rogowsky PM (1997) ZmEsr, a novel endosperm-specific gene expressed in a restricted region around the maize embryo. Plant J 12: Pearce G, Strydom D, Johnson S, Ryan CA (1991) A polypeptide from tomato leaves induces wound-inducible inhibitor proteins. Science 253,

120 Chapter 5 Reddy GV, Meyerowitz EM (2005) Stem-Cell Homeostasis and Growth Dynamics Can Be Uncoupled in the Arabidopsis Shoot Apex. Science 310: Rojo E, Sharma VK, Kovaleva V, Raikhel NV, Fletcher JC (2002) CLV3 is localized to the extra cellular space, where it activates the Arabidopsis CLAVATA stem cell signaling pathway. Plant Cell 14: Sabatini S, Beis D, Wolkenfelt H, Murfett J, Guilfoyle T, Malamy J, Benfey P, Leyser O, Bechtold N, Weisbeek P (1999) An auxin-dependent distal organizer of pattern and polarity in the Arabidopsis root. Cell 99: Scheres B (2005) Stem cells: A plant biology prospective. Cell 122: Schoof H, Lenhard M, Haecker A, Mayer KF, Jürgens G, Laux T (2000) The stem cell population of Arabidopsis shoot meristems in maintained by a regulatory loop between the CLAVATA and WUSCHEL genes. Cell 100: Sharma VK, Ramirez J, Fletcher JC (2003) The Arabidopsis CLV3-like (CLE) genes are expressed in diverse tissues and encode secreted proteins. Plant Mol Biol 51: Spradling A, Drummond-Barbosa D, Kai T (2001) Stem cells find their niche. Nature 414: Steeves TA, Sussex IM (1989) Patterns in Plant Development. Cambridge, UK Cambridge University Press Strabala TJ, O'donnell PJ, Smit AM, Ampomah-Dwamena C, Martin EJ, Netzler N, Nieuwenhuizen NJ, Quinn BD, Foote HC, Hudson KR (2006) Gain-of-function phenotypes of many CLV3/ESR genes, including four new family members, correlate with tandem variations in the conserved CLE domain. Plant Physiol. 140: Trotochaud AE, Hao T, Wu G, Yang Z, Clark SE (1999) The CLAVATA1 receptor-like kinase requires CLAVATA3 for its assembly into a signalling complex that includes KAPP and a Rho-related protein. Plant Cell 11: Wang X, Mitchum MG, Gao B, Li C, Diab H, Baum TJ, Hussey RS, Davis EL (2005) A parasitism gene from a plant-parasitic nematode with function similar to CLAVATA/ESR (CLE) of Arabidopsis thaliana. Mol Plant Pathol 6: Weigel D, Jürgens G (2002) Stem cells that make stems. Nature 415: Yang H, Matsubayashi Y, Nakamura K, Sakagami Y (1999) Oryza sativa PSK gene encodes a precursor of phytosulfokine-α, a sulfated peptide growth factor found in plants. Proc Natl Acad Sci USA 96:

121 Chapter 6 Concluding Remarks and future prospectives

122 Chapter 6 122

123 Concluding remarks and future prospectives While the CLAVATA/ESR (CLE) family consists of 31 members in Arabidopsis there is only limited knowledge about this family of putative ligands. In Arabidopsis only 3 members CLAVATA3 (CLV3), CLE19 and CLE40 have been described in more detail where CLV3 is the only CLE family member of which the receptors and downstream target has been identified (Fletcher et al., 1999; Hobe et al., 2003; Fiers et al., 2004). Beside Arabidopsis, CLE like genes have also been identified in a wide variety of plants like Zinnia, maize and Brassica (Chapter 2). A survey of the rice genome (Chapter 1) resulted in the identification of 33 putative rice CLE proteins including 2 CLE proteins with multiple CLE domains. The recent discovery of the CLE domain as the endogenous ligand places the finding of a wheat EST and 2 putative rice CLEs with multiple CLE domains in a new light (Olsen et al, 2003; Kondo et al., 2006; Chapter 1). Combined, these two findings suggest that a single CLE protein is processed into multiple active peptide ligands as seen in the case of the tobacco SYSTEMIN peptides (Tob Sys1 and Tob Sys2) which are both derived from one polypeptide precursor (Pearce et al., 2001). Untill now the clv3 mutant is the only mutated CLE protein which displays a strong phenotype unlike CLE19 which displays no phenotype upon knockout (Fletcher et al., 1999; Fiers et al., 2004). To obtain more insight into the level of redundancy and processes in which CLE genes are involved the availability and subsequent combination of knockout mutants would be very helpful. However, T-DNA knockout lines are difficult to obtain due to the small size of the CLE genes and therefore it might be better to use a strategy such as RNAi to specifically silence individual CLE genes, or maybe even groups of related CLE genes. Beside the small size of the CLE genes, also the difficulty to detect the endogenous CLE ligand has been shown to be problematic. Except for the Embryo Surrounding Region (ESR) proteins (Bonello et al., 2002; Lenhard et al., 2003) only CLE ligands fused to a reporter protein could be detected in plants that is, only the reporter could be detected. Even over-expression of CLE genes did not result in detection of a CLE protein which is probably due to processing of the CLE protein into a peptide ligand comprising only the CLE domain (Kondo et al., 2006). Polyclonal antibodies are of little use for detection because the CLE domain is only a minor part of the CLE protein while peptide anti-bodies raised against a specific CLE domain very likely will cross react with other closely related CLE domains. 123

124 Chapter 6 In Arabidopsis all clv mutants, while different in strength, display simular phenotypes. Crosses between clv1, clv2 or clv3 mutants did not unveil any additional phenotypes beside an enlarged meristem and increased number of floral organs. This redundancy in phenotype suggests a communal pathway in stem cell signalling in the meristem and has been the basis for later models (Clark et al., 1995; Kayes et al., 1998; Fletcher et al., 1999). The CLV pathway is mainly build on genetics and biochemical data are hardly available. While CLV3 has been shown to act in the same signalling pathway as CLV1 and CLV2 this is based on genetic data, and while undisputed, there is no evidence that the CLV3 ligand binds directly to the CLV1/2 receptor complex. Beside CLV1 and CLV2, also the involvement of additional receptors which are supposed to be involved in the CLV receptor complex is still unclear (Diévart et al., 2003). The next challenge will be to identify the receptors involved in the perception of the different CLE-proteins. Results with the CLE peptides as described in chapter 4 and 5 of this thesis may provide an opportunity to circumvent problems due to processing of the CLE protein and subsequently the removal of any purification tag or protein. One of the possibilities to circumvent these problems is the use of a photo-affinity label at the C or N terminus of the peptide ligand. In this way the ligand can be covently coupled to its receptor by ultra violet light treatment as shown in the case of PHYTOSULFOKINE (PSK) and SYSTEMIN (Matsubayashi et al., 2002; Scheer et al., 2002). The CLE peptides might be the ideal substrate for tagging with a photo affinity label to covently bind the CLE peptide with any interacting proteins. This approach can combined with a purification tag, like biotin, for detection and purification of the ligand-receptor complex. Such an approach might provide the experimental proof that CLV3 is directly interacting with the CLV1/2 receptor complex. Several Leucine Rich Repeat (LRR)-receptor complexes have been identified and characterised that still lack a ligand. Examples are the ERECTA family, the Somatic Embryogenesis Receptor Kinase (SERK) receptors and the CLV1-like LRR receptor involved in nodulation. If these receptors are targets for CLE ligands remains to be investigated but CLE ligands might be interesting candidates for several types of ligand-receptor signalling processes in the life cycle of the plant. 124

125 Concluding remarks and future prospectives REFERENCES Bonello, J.F., Sevilla-Lecoq S., Berne A., Risueno M.C., Dumas C., and Rogowsky, P.M. (2002). Esr proteins are secreted by the cells of the embryo surrounding region. J. Exp. Bot. 53, Clark, S.E., Running, M.P., and Meyerowitz, E.M. (1995). CLAVATA3 is a specific regulator of shoot and floral meristem development affecting the same processes as CLAVATA1. Development 121, Diévart, A., Dalal, M., Tax, F.E., Lacey, A.D., Huttly, A., Li, J.M., and Clark, S.E. (2003). CLAVATA1 dominant-negative alleles reveal functional overlap between multiple receptor kinases that regulate meristem and organ development. Plant Cell 15, Fiers, M., Hause, G., Boutilier, K., Casamitjana-Martinez, E., Weijers, D., Offringa, D., van der Geest, L., van Lookeren Campagne, M., and Liu, C.M. (2004). Mis-expression of the CLV3/ESR-like gene CLE19 in Arabidopsis leads to a consumption of root meristem. Gene 327, Fiers, M., Golemiec, E., Xu, J., van der Geest, L., Heidstra, R., Stiekema, W., and Liu, C.M. (2005).The 14-amino acid CLV3, CLE19 and CLE40 peptides trigger consumption of the root meristem in Arabidopsis through a CLAVATA2-dependent pathway. Plant Cell 17, Fiers, M., Golemiec, E., van der schors, R., van der Geest, L., li, K.W., Stiekema, W., and Liu, C.M. (2006). The CLV3/ESR motif of CLV3 is functionally independent from the non-conserved flanking sequences. Plant Physiol. 141, Fletcher, J.C., Brand, U., Running, M.P., Simon, R., and Meyerowitz, E.M. (1999). Signaling of cell fate decisions by CLAVATA3 in Arabidopsis shoot meristems. Science 283, Hobe, M., Muller, R., Grunewald, M., Brand, U., and Simon, R. (2003). Loss of CLE40, a protein functionally equivalent to the stem cell restricting signal CLV3, enhances root waving in Arabidopsis. Dev. Genes Evol. 213, Ito, Y., Nakanomyo, I., Motose, H., Iwamoto, K., Sawa, S., Dohmae, N., and Fukuda, H. (2006). Dodeca-CLE Peptides as Suppressors of Plant Stem Cell Differentiation. Science 313,

126 Chapter 6 Kayes, J.M., and Clark, S.E. (1998). CLAVATA2, a regulator of meristem and organ development in Arabidopsis. Development 125, Kondo, T., Sawa, S., Kinoshita, A., Mizuno, S., Kakimoto, T., Fukuda, H., and Sakagami, Y. (2006). A Plant Peptide Encoded by CLV3 Identified by in Situ MALDI-TOF MS Analysis. Science 313, Lenhard, M. and Laux, T. (2003). Stem cell homeostasis in the Arabidopsis shoot meristem is regulated by intercellular movement of CLAVATA3 and its sequestration by CLAVATA1. Development 130, Matsubayashi, Y., Ogawa, M., Morita, A. and Sakagami, Y. (2002). An LRR receptor-like kinase involved in perception of a peptide plant hormone, phytosulfokine. Science 296, Olsen, A.N. and Skriver, K. (2003). Ligand mimicry? Plant-parasitic nematode polypeptide with similarity to CLAVATA3. TIPS, Pearce, G., Moura, D.S., Stratmann, J., and Ryan, C.A. (2001). Production of multiple plant hormones from a single polyprotein precursor. Nature 411, Scheer, J.M. and Ryan, C.A. (2002). The systemin receptor SR160 from Lycopersicon peruvianum is a member of the LRR receptor kinase family. Proc. Natl. Acad. Sci. U.S.A. 99,

127 Summary Summary In multicellular organisms, cell to cell communication is essential for coordinating growth and differentiation. In plants most known intercellular communication is mediated by phytohormones such as auxin, cytokinin, GA, ABA, ethylene and brassinosteroids. However, in recent years, several putative peptide ligands have been identified in plants and were shown to mediate signalling events. One of these ligands is CLAVATA3 (CLV3) which is a peptide ligand in Arabidopsis and is involved in stem cell maintenance. CLV3 is part of the CLV3/ESR (CLE) gene family which consists of 31 members in Arabidopsis. These genes encode small proteins which contain a putative secretion signal at their N-termini and a conserved CLE-motif at or near their C-termini. Recently the mature CLV3 was identified and shown to consist of 12 amino acids comprising the CLE motif with two hydroxylated prolines. This thesis aims to better understand signalling by the CLE gene family and in particular CLE19 and CLV3. This research started with the characterization of one of the CLE members namely CLE19 in Brassica napus embryos. This was followed by a small proteomic study of the secretome of Brassica microspore embryos to identify differences in low and high yielding embryo cultures and concludes with a newly developed peptide based approach to better understand CLE signalling in shoot and root meristems. Chapter 1 is an introduction on CLE signalling in Arabidopsis. This chapter combines all the data that is published related to the CLE gene family. The Arabidopsis CLE family was used as a starting point for making a genomic survey of the rice genome to identify rice CLE genes. This survey resulted in the identification of 33 putative rice CLE genes. All rice CLEs were compared to the ones from Arabidopsis in order to find orthologues based on the amino acid sequence of the complete protein or only the CLE domain. The comparison, based on the complete protein did not result in the identification of CLE orthologues between rice and Arabidopsis. Comparing the 14 amino acid CLE domains resulted in the identification of some putative orthologues but most similar CLE domains were within the Arabidopsis or rice gene family. The CLE domain is probably too short to identify any orthologues between the 127

128 Summary monocotyledon (rice) and dicotyledon (Arabidopsis) but can be used as a starting point for further research. This chapter is concluded with an outline of this thesis. In Chapter 2 a short survey on the biochemical aspects of haploid embryogenesis is described and in particular the identification of extra cellular signalling molecules in microspore embryo development of Brassica napus. Differences were examined in protein profiles present in the medium from high-yielding microspore derived embryos cultures and low-yielding that did not produce any embryos. Several proteins were isolated and identified by de-novo protein sequencing including some unknown proteins with no obvious orthologue in the Arabidopsis genome. Chapter 3 describes DD3-12, one of the genes that were isolated in a screen to identify markers for embryo development in Brassica napus. DD3-12 encodes BnCLE19, a small 74 amino acid secreted protein that belongs to the CLE like family of proteins. With the use of northerns, a pbncle19::gus construct and RT-PCR analysis the expression pattern of CLE19 was established in Brassica and Arabidopsis. Interestingly CLE19 is expressed in the L1 and L2 layers of the periphery of shoot and inflorescence meristems as compared to CLV3 which is expressed in the first three layers in the top of the meristem. But also in embryos CLE19 expression was observed from globular stage onwards in the epidermal cells covering the developing cotyledons. BnCLE19 and CLE19, its orthologue in Arabidopsis, were over expressed in Arabidopsis which resulted, among other phenotypes, in the consumption of the root meristem. Interestingly this consumption of the root meristem was also observed upon over expression of other members of the CLE family, namely CLV3 and CLE40. The only conserved domain among all CLE proteins, beside the secretion signal, is a 14 amino acid CLE domain at or near the C-terminus. To test if this domain is the functional part of the CLE proteins and if this domain can act independently from the rest of the protein an in-vitro system was developed using chemically synthesized 14 amino acid CLE peptides. Chapter 4 describes the research aimed at the analysis of the functional domain of CLE proteins. We show that the CLE peptides can cause an over-expression phenotype in the roots of Arabidopsis, namely termination of the root meristem, as previously shown in chapter 3 upon over expression of CLE19. Already after 4 days of growth in CLE peptide containing 128

129 Summary media the first signs of a general miscommunication between cells and cell layers were observed. In early root development the CLE peptides caused a misspecification in the pericycle, endodermis and cortex cell layers and a premature differentiation of the cortical daughter cells. In the search for receptors involved in the perception of the CLE ligands we challenged the clv mutants with several CLE peptides. Only the clv2 mutant, when incubated with CLE peptides, resulted in normal root development with no defects in the meristematic tissue. This result was revealing a CLV-like signalling pathway in the roots with CLV2 as one of the receptors which was involved in the perception of the CLE peptides during root development. Chapter 5 pictures the functional analysis of the CLV3 peptide in shoot apical meristem (SAM) development of the clv3 mutant which results in the reduction of the enlarged SAM to wild type (WT) proportions. Different CLE peptides were used to determine the redundancy between different CLE genes in SAM development. The functional analysis of different CLE peptides in clv3 resulted in a variable degree of complementation. The CLE40p and CLV3p peptides caused an almost complete restoration of the SAM to WT while CLE19p and CLE5p resulted in a partial complementation. This in contrast with CLE22p which was not able to replace CLV3 resulting in a greatly enlarged meristem as seen in clv3. These results clearly showed that not all CLE ligands were freely interchangeable and a certain degree of sequence specificity is required which was also shown with a mutated and several truncated peptides. The in-vitro peptide approach is combined with a deletion analysis of the CLV3 gene to proof that the CLE domain, beside the secretion signal, is essential and sufficient in rescuing the clv3 mutant. This thesis finishes with some concluding remarks and future prospectives in Chapter 6. This chapter describes the future of research in the field of CLE signalling with a particular interest in understanding ligand-receptor signalling and the pitfalls and drawbacks in identifying the receptors involved in the perception of the CLE ligands. 129

130 Summary 130

131 Samenvatting Samenvatting Cel tot cel communicatie is essentieel in multicellulaire organismen voor het coördineren van groei en differentiatie. In planten is meest van de bekende intercellulaire communicatie geregeld door hormonen zoals auxine en cytokinine. In de laatste jaren zijn er verschillende peptide liganden geïdentificeerd die, naast de bekende hormonen, betrokken zijn bij de signaaltransductie in planten. Een van deze liganden is CLAVATA3 (CLV3), een peptide ligand in Arabidopsis welke betrokken is bij de instandhouding van de stamcellen. CLV3 is onderdeel van de CLV3/ESR (CLE) genfamilie die bestaat uit 31 leden in Arabidopsis. Deze genen coderen voor kleine eiwitten, welke een secretiesignaal bezitten aan de N-terminus en een geconserveerd CLE-motief aan of vlakbij de C-terminus. Recentelijk is het functionele CLV3 peptide geïdentificeerd dat bestaat uit een peptide van 12 aminozuren, met 2 gehydroxyleerde prolines, welke het CLE motief vormt. De studie in deze doctoraalscriptie heeft als doel om inzicht te verschaffen in de signaaloverdracht door de CLE genfamilie en in het bijzonder CLE19 en CLV3. Het onderzoek startte met de karakerisatie van een van de CLE genen namelijk CLE19 in Brassica napus embryo s. Dit werd gevolgd door een studie naar eiwitten die worden uitgescheiden door Brassica microsporen embryo s om zodoende verschillen te identificeren in media van embryo culturen met een hoge of lage opbrengst. De scriptie wordt afgesloten met een nieuw ontwikkelde aanpak om door middel van peptiden inzicht te verkrijgen in de CLE signaaloverdracht in scheut en wortel meristemen. Hoofdstuk 1 is een introductie in de signaaloverdracht door de CLE familie in Arabidopsis. Dit hoofdstuk combineert alle gepubliceerde data gerelateerd aan de CLE genfamilie. De Arabidopsis CLE familie was gebruikt als startpunt om het rijst genoom door te lichten op zoek naar rijst CLE genen. Dit onderzoek resulteerde in de identificatie van 33 mogelijke rijst CLE genen. Alle gevonden rijst CLE s werden vergeleken met de al bekende Arabidopsis CLE s om orthologen te vinden gebaseerd op de aminozuur sequentie van het gehele eiwit of alleen het CLE motief. De vergelijking gebaseerd op het gehele eiwit resulteerde niet in de identificatie van CLE orthologen tussen rijst en Arabidopsis CLE families. De vergelijking van de 14 aminozuren CLE domeinen resulteerde in de identificatie van enkele mogelijke 131

132 Samenvatting orthologen, maar de meeste overeenkomstige CLE domeinen werden gevonden binnen de rijst of Arabidopsis CLE familie en niet tussen deze twee families. Het CLE domein is waarschijnlijk te kort om orthologen te identificeren tussen een mono-cotelydon zoals rijst en een di-cotelydon zoals Arabidopsis maar kan wel als startpunt worden gebruikt voor verder onderzoek. Dit hoofdstuk wordt afgesloten met een overzicht van deze scriptie. In hoofdstuk 2 wordt een overzicht gegeven van de biochemische aspecten van haploïde embryogenese vanuit microsporen met in het bijzonder de identificatie van extracellulaire signaalmoleculen betrokken bij de ontwikkeling van microsporen embryo s van Brassica napus. Verschillen in eiwit profielen werden onderzocht van media met een hoge embryo opbrengst versus media met een lage embryo opbrengst. Verscheidene eiwitten werden geïsoleerd en geïdentificeerd door middel van het het sequencen van de eiwitten waaronder enige onbekende eiwitten zonder een duidelijke homoloog in het Arabidopsis genoom. Hoofdstuk 3 beschrijft DD3-12, een van de genen die werden geïsoleerd in een onderzoek om markers te identificeren voor de ontwikkeling van embryo s in Brassica napus. DD3-12 codeert voor BnCLE19, een klein 74 aminozuur gesecreteerd eiwit dat behoort tot de CLE familie van eiwitten. Met gebruik van northerns, een pbncle19::gus construct en RT-PCR analyse werd het expressie patroon van CLE19 bepaald in Brassica and Arabidopsis. CLE19 komt tot expressie in de L1 en L2 lagen in de periferie van het scheut- en bloeimeristeem wat interessant is vergeleken met CLV3, welke ook tot expressie komt in de top van beide meristemen, maar dan in de eerste drie lagen in de top van het meristeem. Ook in embryo s komt CLE19 tot expressie zoals te zien is vanaf het globulair embryo stadium, in de epidermiscellen die de ontwikkelende cotelydons bedekken. BnCLE19 en CLE19, de ortholoog in Arabidopsis, werden tot overexpressie gebracht in Arabidopsis wat onder andere resulteerde in de terminatie van het wortel meristeem. Deze terminatie van het wortelmeristeem is ook zichtbaar bij het tot overexpressie brengen van andere leden van de CLE familie, namelijk CLV3 en CLE40. Het enige gemeenschappelijke domein in de CLE eiwitten, behalve het secretiesignaal, is een 14 aminozuur CLE domein aan of vlakbij de C-terminus. Om te testen of dit domein het functionele deel van het CLE eiwit is en of dit domein ook onafhankelijk functioneel kan zijn werd een in-vitro systeem ontwikkeld gebruikmakend van 14 aminozuur CLE peptiden. 132

133 Samenvatting Hoofdstuk 4 beschrijft het onderzoek gericht op de analyse van het functionele domein van de CLE eiwitten. We laten zien dat de CLE peptiden een overexpressie fenotype kunnen veroorzaken in de wortels van Arabidopsis, namelijk de terminatie van het wortelmeristeem zoals eerder beschreven bij de overexpressie van CLE19 in hoofdstuk 3. Al na 4 dagen van groei, op CLE peptide bevattend media, werden de eerste verschijnselen van een algemene miscommunicatie tussen cellen en cellagen zichtbaar. Vroeg in de wortelontwikkeling veroorzaken de CLE peptiden een miscommunicatie tussen de pericycle, endodermis en cortex cellagen en een vervroegde differentiatie van de cortical dochtercellen. In de zoektocht naar receptoren die betrokken zijn bij de perceptie van de CLE liganden werden de clv mutanten met verschillende CLE peptiden getest. Alleen de clv2 mutant, wanneer die geïncubeerd werd meteen van de CLE peptiden, resulteerde in een normale wortelontwikkeling zonder defecten in het meristematische weefsel. Dit resultaat onthulde een signaaltransductie in de wortels, lijkend op de CLV signaaloverdracht in de SAM, met CLV2 als een van de receptoren die betrokken is bij de perceptie van de CLE peptiden tijdens wortelontwikkeling. In Hoofdstuk 5 wordt de functionele analyse beschreven van het CLV3 peptide in de ontwikkeling van het scheut apicaal meristeem (SAM) van de clv3 mutant wat resulteerde in een reductie van het vergrootte SAM tot wild type (WT) proporties. Verschillende CLE peptiden werden gebruikt om de redundantie te bepalen tussen verschillende CLE genen in de ontwikkeling van de SAM. De functionele analyse van de verschillende CLE peptiden in clv3 resulteerde in een variabel niveau van complementatie. De CLE40p en CLV3p peptiden veroorzaakten een bijna complete complementatie van het SAM tot WT terwijl CLE19p en CLE5p resulteerde in een gedeeltelijke complementatie. Dit in tegenstelling tot CLE22p welke niet in staat was om CLV3 te vervangen. Dit resulteerde in een sterk vergroot meristeem, zoals te zien is in de clv3 mutant. Deze resultaten tonen duidelijk dat niet alle CLE liganden uitwisselbaar zijn en dat een zekere mate van sequentie specificiteit vereist is wat werd geverifieerd met gemuteerde en verschillende gedeleteerde peptiden. Deze in-vitro peptide aanpak werd gecombineerd met een deletieanalyse van het CLV3 gen om te bewijzen dat het CLE domein, naast het secretiesignaal, essentieel en voldoende is om de clv3 mutant te complementeren. 133

134 Samenvatting Deze doctoraalscriptie eindigt met enige conclusies en een visie op toekomstig onderzoek beschreven in Hoofdstuk 6. Dit hoofdstuk geeft een beschrijving van toekomstig onderzoek in het veld van de CLE signaaltransductie. Hierbij wordt ingegaan op welke manieren inzicht kan worden verkregen in de ligand-receptor signaaloverdracht inclusief mogelijke valkuilen en nadelen behorend bij de identificatie van receptoren die betrokken zijn bij de perceptie van CLE liganden. 134

135 Dankwoord Dankwoord Als laatste, maar waarschijnlijk als eerste gelezen, begin ik aan het meest belangrijke deel van dit boekje namelijk het bedankt voor alles hoofdstuk. Na een energieke start van mijn promotieproject kon ik na 2 jaar werken ongeveer de achterkant van een bierviltje vullen met resultaten. Het schijnt dat elke AIO het meemaakt en dat het goed voor je is maar d r waren tijden dat ik wou zeggen ja, het zijn allemaal afbraakproducten, nou goed. Gelukkig kwam toen het geweldige idee om dat ene peptide, dat toch maar in de vriezer lag, maar eens te testen op zijn invloed op de wortelonwikkeling. En zo slecht als het eerst ging, zo snel en goed was het vervolg met mooie resultaten in overvloed. Natuurlijk kon ook ik het niet zonder de steun van mijn collega s die toch altijd voor me klaarstonden en, wanneer nodig, me weer op terug op mijn plaats zette als ik het weer eens te bont maakte. Om nu een boekwerk aan namen op te noemen is ook weer zoiets dus hier dan maar de bijnamen van collega s die in de loop der jaren voorbij kwamen en die op verschillende wijze een bijdrage hadden (hierbij ook maar meteen een oproep om die namen in het telefoonboek te zetten, dan hoef ik niet heel de tijd te vragen hoe die persoon ook weer echt heet); paps, Herr doctor, Stefanus, Lonnie, chickie (dames herken uzelf), kabouter, tante Jannie (een van de weinige waar ik wel naar luisterde), Janneman, de alpenzusjes, Floor, Snoekepoek, Poppedop (chicken tonight zal nooit meer hetzelfde zijn), Miep, Sjonnie, Ieroen, Jut en Jul (mijn lunches zijn nog nooit zo luidruchtig geweest), Suzie, Fiers (dat hoofdstuk Fiers & Fiers is er toch gekomen!) Eddy the eagle, Brammie, zonder schroevendraaier, stille Willie, Eggie (die grasmaaier komt er nog wel eens), Klaasje, Peeters, Chef, de Co en de Cu, allemaal bedankt!! Ook heb ik nog een tijdje op en neer gependeld naar Amsterdam waarbij ik Roel en Ka Wan, van de vakgroep neurobiology aan de VU, zeer erkentlijk ben voor al de hulp bij de analyse van mijn eiwit monsters op de HPLC en MALDI-TOF. Gerco wil ik ook bedanken voor alle ruimte die hij bood en creërde om mijn onderzoek uit te voeren, zelfs zonder dat ik gebruik moest maken van een of andere transcriptiefactor. 135

136 Dankwoord Mijn promotor Willem Stiekema wil ik bedanken voor zijn positieve steun in de totstandkoming van het boekje. Altijd was je positief ingesteld en bezig om te zien of we toch nog een extra hoofdstukje konden extraheren uit overgebleven data. Chun-Ming was my supervisor from the beginning and had the tough job of transforming someone who was just handy with a pipette into a solid scientist. I m very gratefull to you for teaching me all the ins and outs of doing good science. Lots of time we disagreed and even more times we argued over on how to proceed, but I think that your calm and scientific approach combined with my straight through the door approach was an ideal mix which resulted in some very nice results. Mijn twee nymphen verdienen natuurlijk een apart stukje, lelijk zwijn (ook wel.com) en Kimmie baby, betere collega s en vrienden kun je niet hebben. Talloze wilde feesten en uitstapjes hebben we gedaan die steevast eindigden met mijn kop in een emmer en spontaan geheugenverlies. Ik ben blij dat jullie naast me staan! Ons pa en ma ben ik ook heel wat verschuldigt, altijd hebben ze me gestimuleerd om door te leren. De vragen hoe het nou ging met mijn onderzoek en of mijn verslag (= artikel) al af was waren en zijn legio. Ons thuis is en blijft toch altijd gelegen aan de Zwartakkers in Bladel. Als laatste wil ik mijn twee vrouwen bedanken voor alle steun tijdens mijn promotie, al werd er wel eens getwijfeld of ik ook ooit wel eens werkte naast het organiseren van labuitjes, borrels, BBQs of het maken van toneelstukjes of film voor een of ander afscheid of feest. Ik hoop dat het wel duidelijk wordt als je dit boekje doorleest. De vraag van Rineke of ik het allemaal zelf heb verzonnen laat ik maar in het midden Houdoe en bedankt! Martijn 136

137 List of publications List of publications Patent: Liu, C.M., Cordewener, J.H.G., Fiers, M., Joosen, R., A.H.M. van der Geest (2000). A group of polypeptides and their coding polynucleiotide sequences that can modulate plant growth, development and defense response. PCT/NL01/ Book chapters: Custers, J.B.M., J.H.G. Cordewener, M.A. Fiers, B.T.H. Maassen, M.M. van Lookeren Campagne and C.M. Liu (2000). Androgenesis in Brassica: A model system to study the initiation of plant embryogenesis. In Current Trends in the Embryology of Angiosperm, S.S. Bhojwani and W.Y. Soh (Eds), pp Boutilier, K., Fiers, M., Liu, C.M., and Geest A.H.M. (2004). Biochemical and Molecular aspects of Haploid Embryogenesis. In Haploids in Crop Improvement II. D. Palmer, W. Keller and K. Kasha eds. Springer-Verlag, Heidelberg. pp Publications: Veldhuisen G., Saloheimo M., Fiers M.A., Punt P.J., Contreras R., Penttila M., van den Hondel C.A. (1997). Isolation and analysis of functional homologues of the secretion-related SAR1 gene of Saccharomyces cerevisiae from Aspergillus niger and Trichoderma reesei. Mol. Gen. Genet. 256,

138 List of publications Fiers, M., Hause, G., Boutilier, K., Casamitjana-Martinez, E., Weijers, D., Offringa, R., van der Geest, L., van Lookeren Campagne, M., and Liu, C.M. (2004). Mis-expression of the CLV3/ESR-like gene CLE19 in Arabidopsis leads to a consumption of root meristem. Gene 327, Fiers, M., Golemiec, E., Xu, J., van der Geest, L., Heidstra, R., Stiekema, W.J., and Liu, C.M. (2005). The 14-amino acid CLV3, CLE19 and CLE40 peptides trigger consumption of the root meristem in Arabidopsis through a CLAVATA2-dependent pathway. Plant Cell 17, Fiers, M., Golemiec, E., van der Schors, R., van der Geest, L., Li, K.W., Stiekema, W.J., and Liu, C.M. (2006). The CLV3/ESR motif of CLV3 is functionally independent from the non-conserved flanking sequences. Plant Physiology 141, Fiers, M., Ku, K.L., and Liu, C.M. (2007). CLE peptide ligands and their roles in establishing meristems. Curr. Opin. in Plant Biol

139 Curriculum Vitae Curriculum Vitae Martijn Adrianus Fiers werd geboren op 9 maart 1972 in het brabantse Bladel. In 1989 slaagde hij voor de HAVO waarna hij begon aan zijn studie in Eindhoven aan het Hoger Laboratorium Onderwijs (HLO) in de richting algemene microbiologie. De eerste voorzichtige stapjes op het gebied van de biotechnologie werden genomen tijdens zijn stage en afstuderen in het vierde jaar bij MBL-TNO bij de afdeling Moleculaire Genetica en Gentechnologie (MGG, onderdeel van het latere TNO voeding). Hier werkte hij aan het kloneren en karakteriseren van genen die betrokken zijn bij de secretie in de schimmel Aspergillus niger in de groep van Dr. Cees van den Hondel. In 1994 werd zijn HLO studie afgerond met het verkrijgen van de Ingenieurs titel en werd hij opgeroepen voor militaire dienst. Tot eind 1994 was hij in dienst als motorordonnans bij de verbindingstroepen in de rang van korporaal. In 1995 kwam hij weer te werken bij de afdeling MGG van TNO voeding als research analist waar hij werkzaam was in een industrieel project betreffende het tot expressie brengen en opschalen van de productie van heterologe enzymen in schimmels. In 1996 werd hij geïntroduceerd in de plantenbiotechnologie toen hij aangenomen werd als research analist bij de afdeling Ontwikkelingsbiologie van het CPRO-DLO (later opgegaan in de Business Unit Bioscience welke een onderdeel is van het latere Plant Research International). Hier was hij betrokken bij een project met als doel het isoleren en karakteriseren van genen die betrokken zijn bij de microsporen embryogenese van Brassica napus (koolzaad). Een van de geïsoleerde genen uit dit onderzoek was CLE19, welke de basis vormde voor het vervolgonderzoek naar peptideliganden in planten. De resultaten van dat onderzoek zijn beschreven in dit proefschrift. Per januari 2007 heeft hij een aanstelling als junior onderzoeker in de groep van Prof. Dr. Gerco Angenent in de cluster Plant Developmental Systems, BU Bioscience onderdeel van PRI waar hij werkt aan ligand-receptor signaaltransductie in de stamcellen van planten. 139

140 Curriculum Vitae Cover design: Brechje Raaijmakers 140

141 Appendix Colour figures Appendix Colour figures 141

142 Appendix Colour figures 142

143 Appendix Colour figures Chapter 1, Figure 4. Similarity matrix of a comparison between rice and Arabidopsis CLE proteins. The bar depicted on the right represents the similarity (%) as calculated by ClustalW. I

144 Appendix Colour figures Chapter 1, Figure 5. Similarity matrix of a comparison between rice and Arabidopsis CLE domains. The bar depicted on the right represents the similarity (%) as calculated by ClustalW. II

145 Appendix Colour figures Chapter 2, Figure 3. Expression analysis of BnCLE19. Northern blot analysis of BnCLE19 expression in B. napus. Ten µg of total RNA isolated from various organs and tissues was separated, blotted and probed with the BnCLE19 cdna. BnCLE19 is expressed in developing embryos (globular to heart stage embryos from microspore culture), flower buds and pistils. Petals, anthers and pistils were obtained from 5 mm flower buds containing tri-nucleate pollen. (B-E) Expression of BnCLE19 in zygotic embryos, as shown by the confocal microscopic observation of F1 embryos carrying both pbncle19:gal4-vp16) and puas::gfp-gus constructs. In tri-angular stage embryos, GFP expression was observed in the epidermal cell layer that covers the cotyledon primordia and the shoot apical meristem (B). In the heart- to torpedo-shaped embryos, the GFP signal was only observed in the epidermal cells covering the cotyledon primordia (C-E). The scale bars in (B) represents 10 µm for (B and C), and in (C) represents 25 µm for (D and E). III

146 Appendix Colour figures Chapter 2, Figure 4. Histological analysis of post-embryonic GUS expression in pbncle19-gus transgenic Arabidopsis. The photographs correspond to whole-mount materials cleared with Hoyer s solution (A-C, F and H) and paraffin sections (D, E and G). (A-D) In roots, pbncle19::gus is expressed in the root hair region and the differentiation zone above (B-D), but not in the root meristem (D), nor in the newly formed lateral root (B), nor in older roots with secondary thickenings (A). The scale bar in (D) represents 100 µm for A-D. (E) Transverse section of a root in the root hair region, showing GUS expression in 2 to 3 pericycle cells (arrowheads) facing the protoxylem poles. The tissue deformation was caused by the acetone pre-fixation used in the GUS assay. The scale bar represents 100 µm. (F) pbncle19::gus expression was seen in the periphery of meristems in the axillary bud, where the cauline leaves will form. The scale bar represents 25 µm. (G and H) During flower development, GUS expression was seen in the sepal primordia in stage 2-5 flower buds, but not in the main inflorescence meristem (marked with an asterisk). The scale bars represent 40 µm. (I) In a stage 10 flower bud, GUS expression was seen at the top of the pistil, where the stigma hairs will form. The scale bar represents 150 µm. IV

147 Appendix Colour figures Chapter 4, Cover Plant Cell 17, (2005) V

148 Appendix Colour figures Chapter 4, Figure 7. Effects of CLE peptides on the cell identity of roots. The ground tissue initial cells are marked by arrows, and the ground tissue daughter cells, including cells at this position, are labelled with arrowheads. Abnormal expression of GFP or YFP in other cell layers is marked by asterisks. VI

149 Appendix Colour figures (A) to (D) Confocal analysis after 4-day treatment of the roots of J0571. Note that treatments with CLV3p (C) or CLE19p (D) lead to a delayed separation of cortex and endodermis and expression of GFP in pericycle cells. (A) No peptide; 10 µm of (B) CLVm; (C) CLV3p and (D) CLE19p. (E) to (H) Confocal analyses of P SCR :GFP seedlings after 4-day treatments. (E) No peptide; (F) 10 µm CLV3s and (G) 10 µm CLV3p. The insert shows the delayed separation of the cortex and endodermis cells (with GFP expression). (H) 10 µm CLE19p. (I) to (M) P CO2 :YFP-H2B marker line treated with different peptides. Note the YFP expression in the ground tissue daughter cells and the pericycle cells, but not in ground tissue initial cells. (I) No peptide, 4 days after germination. The insert shows the absence of YFP expression in the QC, ground tissue initial and ground daughter cells. (J) 10 µm CLE19p, 4 days after treatment; (K) 10 µm CLE40p, 5 days after treatment; (L) and (M) 10 µm CLE19p, 5 days after treatment. The scale bars represents 30 µm. Chapter 4, Figure 8. Confocal observation of J0571 roots, after an 8-day treatment with CLE19p. (A) No peptide. (B) CLE19p. Note that at this stage the number of ground tissue daughter cells (marked with arrowheads) was not increased further, and no GFP expression was detectable in the pericycle layer. Ground tissue daughter cells are marked with arrowheads. The scale bars represents 30 µm. VII

150 Appendix Colour figures Chapter 5, Cover Plant Physiology 141, (2006) VIII