TY - JOUR
AU - Rogowsky, Peter M.
AB - Abstract Three highly homologous Esr genes are expressed specifically in the embryo surrounding region at the micropylar end of the maize endosperm. The proteins belong to a family of small hydrophilic proteins that share a conserved motive with Clv3, the ligand of the receptor‐like kinase Clv1. In this study, co‐localization of Esr proteins with their mRNAs in the embryo surrounding region was shown with polyclonal antibodies recognizing all three Esr proteins. On a subcellular level the secretion of Esr proteins and their association with the cell wall was shown independently by cell fractionation, immunohistochemistry and transient expression of Gfp fusion proteins. Furthermore, a possible interaction of Esr proteins with a 35 kDa protein present in the lower half of maize kernels was suggested by in vitro affinity chromatography. Therefore Esr proteins share two characteristics with ligands of receptor‐like kinases: they are released in the extracellular space and they have the capacity to form protein–protein interactions. Key words: Endosperm, immunohistochemisty, ligand, protein–protein interaction, Zea mays. Received 21 August 2001; Accepted 14 March 2002 Introduction Maize endosperm is a highly differentiated organism with a complex development. The first two weeks of its existence are characterized by pattern formation and lead to the differentiation of four domains (Olsen et al., 1999) with very distinct cytological and functional characteristics (Randolph, 1936; Schel and Kieft, 1986). Most of its volume is taken up by the starchy endosperm, where starch and storage proteins are accumulated. It is surrounded by a single cell layer called the aleurone layer, where anthocyanin pigments and numerous hydrolases are deposited; by contrast to the starchy endosperm, the aleurone layer stays alive during seed dormancy and is essential for the mobilization of the reserve substances during germination (Kyle and Styles, 1977). At the micropylar end of the endosperm, the aleurone layer is replaced by another specialized domain, the basal endosperm transfer layer (BETL). The cells of this domain are in close contact with the vascular bundles of the mother plant and play a role in the nutrient transfer from source organs into the developing endosperm which is a nutritional sink (Davis et al., 1990). The fourth domain, the embryo surrounding region (ESR), is characterized by small and densely cytoplasmic cells with a high content of endoplasmatic reticulum and Golgi vesicles. Its function remains unknown. It has been hypothesized that it could play a role in different types of exchanges between endosperm and embryo (Opsahl‐Ferstad et al., 1997; van Lammeren, 1986) and, more generally, that the base of the endosperm could be devoted to exchanges between maternal tissues, endosperm and embryo. Little is known about the genes expressed in the two basal domains of maize endosperm (Matthys‐Rochon et al., 1997). Marker genes for the BETL include the genes Betl1 (Hueros et al., 1995), Betl2, Betl3, and Betl4 (Hueros et al., 1999), Mrp1 coding for a transcription factor (G Hueros, personal communication) and Incw2 coding for a cell wall bound invertase (Cheng et al., 1996). The ESR is defined by the expression of the highly homologous genes Esr1, Esr2, and Esr3 (Opsahl‐Ferstad et al., 1997), and the non‐related genes Ae1 and Ae3 (Magnard et al., 2000). With the exception of Mrp1 and Incw2, all these genes code for small hydrophilic proteins of unknown function without substantial homologies to the databases. The localization of these proteins in exchange areas is suggestive for three potential functions. First of all, these proteins might play a role as signal proteins. This hypothesis is reinforced by the presence of putative signal peptides at the N‐terminus, a prerequisite for signal transduction between symplastically isolated tissues. There are several examples of small proteins acting as signals. Extracellular glycoproteins control the early embryo development in Citrus (Gavish et al., 1991), Clavata3 plays a role in the differentiation of vegetative and floral meristems (Fletcher et al., 1999), phytosulphokine (Matsubayashi and Sakagami, 1996) controls cellular proliferation, and systemin intervenes in defence reactions (for a review see Bisseling, 1999). An alternative hypothesis proposes that these small proteins have a nutritional function, as do some of the vegetative storage proteins, which act as a nitrogen source in vegetative tissues (Mason and Mullet, 1990). A last hypothesis claims that these small proteins help to avoid transfer of pathogens from the mother plant into the developing seed: this might be the case for the BAP proteins, a family of antifungal proteins related to BETL2 which present partial homologies to plant defensins (Serna et al., 2001). In this study, the main interest is the Esr genes, which represent an entry point in the understanding of the function of the ESR. Genes Esr1, Esr2 and Esr3 were isolated by differential display and present homologies between 80% and 90% at the nucleotide level. All three Esr genes show very similar expression patterns (Bonello et al., 2000) being confined to the ESR between 4 d and 28 d after pollination (DAP). They code for small hydrophilic proteins with the exception of an N‐terminal signal peptide and show a partial similarity to Clavata3 (Clv3) from Arabidopsis thaliana. Clv3 is a putative ligand of the LRR type receptor‐like kinase Clavata1 and plays a role during meristem formation. The similarity between Esr and Clv3 is limited to a conserved region of 15 aa (Fig. 3) and is likely to be significant for two reasons: Firstly, the conserved region is probably a functional domain of Clv3 because the only two existing point mutations are due to substitutions in this region (Fletcher et al., 1999). Secondly, this region of 15 aa is shared by a large family of over 40 deduced proteins called Cle (Clv3/Esr‐related) that are all small hydrophilic proteins with a signal peptide (Cock and McCormick, 2001). To substantiate the hypothesis that Esr proteins are ligands of receptor kinases or signalling molecules in a larger sense, the isolation of interacting proteins and determination of the subcellular localization was started. The existence of interacting proteins in vitro is reported here and evidence is presented for the secretion of Esr proteins. Material and methods Plant material Maize inbred lines A188 (Gerdes and Tracy, 1993) and F485 (gift of M Beckert, INRA Clermont‐Ferrand) were grown in a growth chamber with a 16 h illumination period (100 W m–2) at 24/19 °C (day/night) and 80% relative humidity. Plants were hand‐pollinated and samples taken at defined developmental stages. Overexpression of Esr protein DNA fragments coding for the mature proteins Esr1, Esr2 and Esr3 without their putative signal peptides were amplified with Pfu polymerase (Stratagene) with primers E1N3 (5′‐CCCG GATCCAATGTCGTGGCAGACTGATGA‐3′), and E1C5 (5′‐CC CAAGCTTAAG‐TTATTGCTTCATGTTTAC‐3′) and cloned in frame into the bacterial expression vector pQE30 (Qiagen) downstream of an N‐terminal 6‐His tag. The overexpression in E. coli strain SG14 (Gottesman et al., 1981) was induced by IPTG and the proteins purified under denaturing conditions on Ni‐agarose columns according to the instructions of the supplier (Qiagen). Antibody production Partially purified Esr proteins were separated on 15% preparative gels (acrylamide/bisacrylamide 29/1) under denaturing conditions according to Amero et al. (1994). The bands corresponding to Esr proteins were ground in liquid nitrogen and resuspended in PBS (50 mM NaH2PO4 pH 7.25; 150 mM NaCl). Rabbit antiserum was raised by repeated subcutaneous injections of 100 µg of protein (Eurogentec). In parallel, antibodies were raised against the synthetic Esr2 peptide KAPYEDDRSR coupled to a Keyhole Limpet Hemocyanin carrier (Coval‐Ab, Lyon). Protein extraction, electrophoretic analysis and Western blotting Maize endosperms were ground in liquid nitrogen and 200 µg of the powder was resuspended in 200 µl SDG buffer (62.5 mM Tris–HCl, 2.5% w/v SDS, 2% m/v DTT, 10% m/v glycerol, pH 6.8). After additional grinding the extract was heated 10 min at 100 °C and cellular debris was eliminated by centrifugation for 15 min at 18 000 g. When indicated, 6 M guanidinium was used as extraction buffer instead of SDG buffer and the extracts purified on Bio Gel P‐6DG (BioRad) prior to electrophoresis. For extraction with cell wall lysis, 200 µg of ground kernels were incubated 2 h in a digestion buffer (200 mM KH2PO4, 1 M KNO3, 100 mM CaCl2.2H2O, 10 mM MgSO4, 1 mM KI, 0.1 mM CuSO4, 1% w/v cellulase, 2% w/v macerozyme, 0.6% w/v PVPP, 1 mM DTT, 1 mM PMSF, 10 µg ml–1 leupeptin, 10 µg ml–1 aprotinin, and 20 µg ml–1 chymostatin). After precipitation by acetone the proteins were resuspended in 150 µl SDG buffer prior to electrophoresis. SDS–PAGE, electrotransfer to nitrocellulose membranes and detection of antigen with antibodies has been described previously (Gaude et al., 1993). Microsome preparation Maize kernels were homogenized with an ultrathurax device in a lysis buffer preserving membrane structure (500 mM sucrose, 50 mM HEPES pH 7.4, 5 mM EDTA, 5 mM EGTA, 1 mM DTT, and 0.6% w/v PVPP) in the presence of phospholipase and protease inhibitors (10 mM NaF, 1 mM PMSF, 10 µg ml–1 leupeptin; 20 µg ml–1 chymostatin). Cellular debris was eliminated by centrifugation at 10 000 g for 10 min. The supernatant was centrifuged for 1 h at 100 000 g at 4 °C in a TFT 50.13 rotor in a Centrikon ultracentrifuge (Kontron). Soluble proteins were present in the supernatant. The pellet corresponding to the microsomal fraction was resuspended in 200 µl TBS. Affinity chromatography A 4 ml cleared lysate of bacteria overexpressing Esr2 was prepared under native conditions according to the protocol Qia Expressionist (Qiagen). The cleared lysate was mixed with a protein extract obtained from 200 mg of 9 DAP maize endosperm in 200 µl of lysis buffer (50 mM NaH2PO4, pH 8; 300 mM NaCl; 10 mM imidazole) and 1 ml of Ni‐NTA resin during 1 h at room temperature. The mixture was loaded into a column and washed with wash buffer (50 mM NaH2PO4, pH 8.0, 300 mM NaCl, 20 mM imidazole). After addition of elution buffer (50 mM NaH2PO4, pH 8.0, 300 mM NaCl, 250 mM imidazole) fractions were collected for subsequent SDS–PAGE and silver staining according to Morrissey (1981). IgG purification IgG were purified from serum by precipitation with saturated ammonium sulphate. The serum was centrifuged 10 min at 10 000 g in order to eliminate lipids. One volume of ammonium sulphate was progressively added to 1 vol of serum at 4 °C. After a 30 min incubation on ice, the suspension was centrifuged 10 min at 10 000 g. The pellet was washed twice with 50% ammonium sulphate and resuspended in two volumes of PBS. Immunocytochemistry for light microscopy Maize kernels were sliced into equal parts along their longitudinal axis and the central slice, fixed in 4% PFA (4% paraformaldehyde, 0.1 M phosphate buffer pH 7.0) at 4 °C for 24 h. Dehydration and paraffin embedding in Paraplast Plus (Sherwood Medical) were carried out according to Jackson (1991). Microtome sections of 10 µm were placed on slides coated with 2% Tespa (3‐aminopropyl‐triethoxysilan), dewaxed with HistoClear‐II (Prolabo) and rehydrated through an ethanol series. After 1 h blocking with 2% BSA in TBS (100 mM Tris–HCl pH7.5; 150 mM NaCl), the sections were incubated 2 h with a purified IgG fraction diluted 1:30 in TBS. After three washes with TBS, the sections were incubated with goat anti‐rabbit IgG antibody conjugated with alkaline phosphatase (1:40 in blocking buffer). Alkaline phosphatase activity was revealed with NBT and BCIP according to the manufacturer (BioRad). Immunocytochemistry for electron microscopy Maize kernels to be processed for transmission electron microscopy were sliced as described above and fixed overnight at 4 °C in 4% formaldehyde in PBS. After a brief wash with PBS, the samples were dehydrated through a methanol series at 4 °C and embedded in lowicryl K4M resin at the same temperature under UV irradiation. Lowicryl ultrathin sections were placed on colodion and carbon‐coated nickel grids, then used for immunogold labelling. Lowicryl ultrathin sections were floated successively for some min on drops of distilled water, PBS and 5% BSA in PBS. Subsequently, the grids were incubated for 1 h with purified IgG fraction diluted 1:20 in PBS with 1% BSA, rinsed with PBS and incubated with a goat anti‐rabbit IgG conjugated to colloidal gold particles of 10 nm diameter (BioCell, Cardiff UK) diluted 1:20 in PBS. After a final wash in the same buffer, ultrathin sections were stained with 5% aqueous uranyl acetate for 20 min and citrate for 20 s. Gfp constructs A novel fusion vector containing in order a 35S promoter, a ribosomal binding site, an ATG start codon, a multiple cloning site, a Gfp gene without cell sorting signals, and a nos terminator was constructed. The ORF of Esr2 missing the ATG start codon and four bases upstream of the stop codon was amplified with primers Bam‐ 5‐Esr2 (5′‐CGCGGATCCGCATCAAGAATGGGGATGGTGG CT‐3′) and Xho‐3‐Esr2 (5′‐CCGCTCGAGGCCCTATCTAAA AATGGTGGAGGACC‐3′) and cloned in the fusion vector. Similarly a control plasmid for nuclear Gfp localization was constructed by amplification of the putative nuclear localization signal of the maize HsfB gene with primers NLS2‐GS (5′‐GATCCAAGAAAAGGAGACTAGCC‐3′) and NLS‐GA (5′‐TCGAGGCTAGTCTCCTTTTCTTG‐3′). The control plasmids for Gfp localization in the ER or the cell wall were provided by J Haseloff (Cambridge). Transient expression in onion epidermis cells White onions were bought on a local market, peeled and surface‐sterilized with Pursept®‐A (Polylabo). Pieces of the inner epidermis of onion peels were placed on Murashige and Skoog medium (4.32 g l–1 MS salts, 30 g l–1 sucrose, 4.9 g l–1 phytagel) adjusted to pH 5.7 or pH 7. The inner side was exposed to bombardment with DNA coated gold particles (Vain et al., 1993) propelled by a particle inflow gun (Finer et al., 1992). After a 24 h incubation at 24 °C (16/8 h light/dark) fluorescence was observed in a LSM510 confocal mircroscope (Zeiss). Results Production of polyclonal antibodies Antibodies were needed to detect Esr in protein complexes or to localize them in subcellular compartments. To raise antisera against proteins Esr1, Esr2 and Esr3, the respective coding sequences lacking the putative signal peptides were cloned in a bacterial expression vector downstream of an N‐terminal 6‐His tag. The three proteins were purified by Ni2+‐affinity chromatography (Fig. 1) and protein bands cut from preparative gels were used to immunize rabbits. The specificity of the antisera obtained was quite satisfactory. In whole protein extracts from bacteria overexpressing individual Esr proteins, the antisera revealed single bands of approximately 13 kDa which were in good agreement with the theoretical values (Fig. 2A). In control reactions with the respective preimmune sera, no major bands were seen (Fig. 2A). As expected, due to the extensive homology between Esr proteins, cross‐reactions were observed (data not shown). Esr1 antiserum recognized Esr1 and Esr2 antigen with the same affinity. Esr2 antiserum recognized all three Esr antigens showing a higher affinity for Esr2. Esr3 antiserum revealed Esr1 and Esr3 antigens, showing a higher affinity for Esr3. The titre of the antisera was determined by dot blot of serial dilutions of purified Esr proteins (Fig. 2B). It was defined as the strongest protein dilution necessary to obtain a signal, with a dilution 1:1000 of antisera. The best titre was obtained with Esr1 antiserum, which detected a 1:2000 dilution of Esr1 protein. However, in the following studies, Esr2 antiserum, detecting a 1:1000 dilution of Esr2 protein, was privileged because it recognized all three Esr proteins. Immunological characterization of Esr proteins In protein extracts of maize kernel harvested at 9 DAP, the Esr2 antiserum revealed a band of 50 kDa, which was very different from the theoretical value of 14 kDa for mature Esr proteins (Fig. 4, lanes 1 and 2). The signal was probably related to Esr proteins because it only appeared in extracts from the basal third of the kernel and was absent in extracts of the upper part (Fig. 4, lanes 3 and 4). Since there were no canonical sites for post‐translational modifications in the Esr sequences, this high molecular weight form was probably not due to glycosylation or other classical modifications. A possible explanation of the 50 kDa band was the formation of Esr multimers that were not dissociated by the SDS containing extraction buffer. To test this hypothesis endosperm proteins were extracted in the presence of chaotropic reagents such as guanidinium (Fig. 4, lanes 5 and 6) or urea (data not shown). In both cases the 50 kDa band remained clearly visible while an additional band of 14 kDa appeared. The intensity of the 50 kDa band was hardly altered and protein–protein interactions would have to be unusually stable if the 50 kDa form corresponded to Esr multimers. It remained unclear whether the smaller band corresponded to a partial dissociation of a 50 kDa complex or originated from material eliminated with cell debris in the extractions for lanes 1 to 4. The second explanation was corroborated by an extraction after cell wall digestion with cellulase and macerozyme. The larger band remained clearly visible, while the smaller one appeared even without chaotropic reagents (Fig. 4, lanes 7 and 8). These data suggested a possible association of Esr proteins with the cell wall. This association was not disrupted by SDS and the 14 kDa band remained associated with cellular debris, whereas more drastic methods allowed a partial extraction of the 14 kDa band. In order to determine the subcellular localization of both forms in endosperm cells, a cellular fractionation was carried out. The 14 kDa form was detected by the Esr2 antiserum as a major band in the microsomal fraction (Fig. 4, lanes 9 and 10) likely reflecting the synthesis of Esr in the endoplasmic reticulum and its transport via the Golgi apparatus to its final destination in the cell wall. A minor band in the soluble fraction (Fig. 4, lanes 11 and 12) was probably caused by a contamination with ruptured vesicles. The absence of a 14 kDa band in the lanes with cellular debris (Fig. 4, lanes 13 and 14) was expected because only SDS had been used for the extraction and only solubilized proteins migrated in the gel. The 50 kDa form was detected exclusively in the soluble fraction. Neither form was observed after fractionation of the upper part of the kernel (data not shown). Identical results to the ones presented in Fig. 4 (lanes 1 to 14) were obtained with an independent antibody raised against a synthetic oligopeptide of Esr2 as well as with antibodies cleaned up by adsorption to purified Esr protein fixed to a nitrocellulose membrane. Therefore the 50 kDa complex must contain Esr or a protein structurally similar to the oligopeptide KAPYEDDRSR. Interactions of Esr proteins The Esr2 antibodies provided the tool necessary to test the hypothesis that Esr proteins interacted with other proteins. Bacterial extracts overexpressing 6‐His‐tagged mature Esr2 protein were incubated with a whole protein extract made from the base of 9 DAP endosperm, and the mix purified on a Ni‐NTA column. After elution, two major proteins were observed in overlapping fractions: a protein of 13 kDa corresponding to the overexpressed Esr2 protein, and a protein of 35 kDa (Fig. 5A). Control experiments omitting either one of the two extracts showed that the 35 kDa protein was not present in the bacterial extract containing the Esr2 protein (Fig. 5B) and that it did not interact directly with the Ni‐NTA resin (Fig. 5C). Therefore, the 35 kDa protein was present in the endosperm extract and interacted with the 6‐His tagged Esr2 protein in vitro. Further experiments will be needed to exclude an involvement of the His tag or mediating bacterial proteins prior to in vivo studies. Tissue localization of Esr proteins To determine whether Esr proteins co‐localized with their mRNA, in situ immunolocalization at the light microscopical level was carried out with a purified anti‐Esr2 IgG fraction. As expected, a signal was observed in the embryo surrounding region on sagittal sections of 9 DAP maize kernels. Surprizingly an additional, although much weaker signal was seen in the basal endosperm transfer layer (Fig. 6A). Control experiments with pre‐immune serum did not reveal any signal (Fig. 6B). By contrast to the BAP proteins possibly involved in the defence of the maize endosperm against fungal pathogens (Serna et al., 2001), no signal was detected in the pedicel or other maternal tissues. Subcellular localization of Esr proteins Since the purified IgG fractions allowed the immunochemical detection of Esr proteins, the hypothesis that Esr proteins were secreted from the cell, as one would expect it for a potential ligand of a receptor kinase, could now be tested. In the electron microscope the ESR cells were easily identified at 9 DAP because of their cytological characteristics: dense cytoplasm with abundant endoplasmic reticulum and dictyosomes, numerous vesicles of different sizes, and a complete lack of large vacuoles. Two types of signals were detected in these cells: the first one in the secretory machinery (endoplasmic reticulum, Golgi apparatus, and secretion vesicles) and the cell wall (Fig. 7A, B, C), and a second one in the nucleus, mainly in the interchromatic region (Fig. 7D). No signal was observed in endosperm cells outside the ESR (Fig. 7E) or with the pre‐immune serum (Fig. 7F). One of many possible explanations was that the two signals corresponded to the bands of 14 kDa and 50 kDa, respectively. To gain further insight into the subcellular localization of Esr proteins, the entire ORF of Esr2 was fused to a Gfp gene lacking any known cell sorting signals. After transient expression of the construct in onion epidermis cells, fluorescence was observed in a fine network lining the central vacuole and the nucleus (Fig. 8A, B). The signal was very similar to the ones observed with control constructs targeted to the endoplasmatic reticulum (Fig. 8C, D) or the cell wall. No fluorescence was observed in the nucleus in contrast to control constructs targeted to the nucleus (Fig. 8E, F) or the cytoplasm (Fig. 8G, H). In the latter case the signal in the nucleus was caused by diffusion of the Gfp, which was small enough to pass through nuclear pores. Since the fluorescence of the Gfp was quenched by the low pH in the cell wall and periplasmic space, further experiments were conducted at pH 7. In addition, the cells were plasmolysed to distinguish signal in or on the cytoplasm from signal in the cell wall. Cells expressing the Esr2‐Gfp fusion (Fig. 8I, J) behaved in a similar way to the control cells where the Gfp was targeted to the cell wall (Fig. 8K, L). A faint, spotty signal was seen along the cell wall in addition to a stronger signal lining the cytoplasm. Taken together these results exclude a targeting of Esr2 to the nucleus or the cytoplasm and strongly suggest a secretion via the endoplasmatic reticulum into the cell wall. Discussion The results presented here show that Esr proteins are secreted from the cells of the embryo surrounding region and likely remain associated with the cell wall. They move very little and seem to have the capacity to bind other proteins. These results are compatible with a function as a ligand of a receptor‐like kinase, as suggested by a partial sequence homology to the Clv3, the ligand of the LRR receptor‐like kinase Clv1 of Arabidopsis thaliana (Trotochaud et al., 2000). However, other hypotheses such as a function in the nutrition of the embryo or a role in defence against pathogens remain valid. Esr proteins are located in the embryo surrounding region Immunocytological studies with polyclonal anti‐Esr2 antibodies showed a major signal in the ESR. The spatial localization of Esr proteins corresponded precisely to the pattern of Esr gene expression and therefore a great majority of the proteins are transported neither to other domains of the endosperm nor to the embryo nor to maternal tissues. In addition, a rather weak protein signal was observed in the BETL, in the absence of a tangible mRNA signal. It may well be due to a cross reaction of the antibody rather than protein transport. These results are different from the ones obtained with basal layer antifungal proteins (BAP) of maize, which are transcribed in the BETL, but act in the pedicel (Serna et al., 2001). Since movement is a general characteristic of many defence compounds and since the deduced amino acid sequence of Esr proteins lacks any similarities with defence proteins, the data do not strengthen the hypothesis that Esr proteins play a role in plant defence. Esr proteins are secreted from the cell Analysis of the deduced amino acid sequences with PSORT (http://psort.nibb.ac.jp) indicated almost equal probabilities for localization of Esr1 and Esr2 in the outer mitochondrial membrane (0.85) or an export outside of the cell (0.82). However, no targeting to the mitochondria was predicted for Esr3 and the use of other programs such as Predotar (http://www.inra.fr/Internet/Produits/Predotar/) did not corroborate mitochondrial targeting. Three different experimental approaches clearly established export of Esr proteins: (1) In cell fractionation experiments a 14 kDa band, corresponding well to the deduced amino acid sequence of Esr proteins, was found in the microsomal cell fraction and released from the cell wall by enzymatic digestion. (2) Immunocytological studies showed a signal in the cell wall and in all components of the secretory machinery (the endoplasmatic reticulum, Golgi apparatus, secretory vesicles). (3) Transient expression of an Esr2‐Gfp fusion protein resulted in fluorescent patterns indistiguishable from those of a control construct targeted to the cell wall. Taken together with the presence of putative signal peptides at the N‐terminus of Esr sequences, these results are a clear indication for the secretion of Esr proteins and their association with the cell wall. The biochemical interactions with the cell wall seem to be rather weak because Esr proteins can be detected in a soluble fraction. On the other hand they seem to be mechanically ‘trapped’ because their extraction requires either an enzymatic digestion or a strong mechanical disruption of the cell wall. The motif GPPP may play a role in this interaction, although it is shorter than the ones found in structural cell wall proteins such as extensins, proline‐rich or glycine‐rich proteins (Showalter, 1993). The secretion of Esr proteins does not seem to be polar. The proteins are present in all cells of the ESR and no accumulation at the interface either with the embryo or other parts of the endosperm was observed. Esr antisera detect a 50 kDa band Esr antisera detect two major bands in maize endosperm, a 14 kDa band of the size expected for Esr proteins and a higher molecular mass band of 50 kDa. Cellular fractionation showed that the 50 kDa band was present exclusively in the soluble fraction where it is a lot more abundant than the 14 kDa one. The corresponding protein must share some common features with Esr because it reacted with two different types of Esr antisera raised against overexpressed Esr proteins and an oligopeptide, respectively. It also seems to be specific for the embryo surrounding region because it was detected only in protein extracts of the basal third of the endosperm. Furthermore, electron microscope studies with Esr antibodies showed signal only in the embryo surrounding region. The dual nature of this signal in the cell wall and secretion pathway and in the nucleus is difficult to reconcile with a single protein. In fact proteins observed in Golgi apparatus and secretion vesicles are secreted or directed to the lysosomes (Bradshaw, 1989). Nuclear proteins are orientated to the nucleus just after their synthesis in endoplasmatic reticulum (Nigg et al., 1991). They possess a short sequence of four to eight amino acids, rich in lysine and arginine, and usually containing proline (Dingwall and Laskey, 1991), which is not the case for Esr proteins (Opsahl‐Ferstad et al., 1997). Since the 14 kDa Esr proteins are likely secreted, the 50 kDa protein is a prime candidate to explain the nuclear signal. It is difficult to imagine that the 50 kDa form corresponds to a simple aggregation of Esr proteins. It resists denaturing conditions or treatments with chaotropic reagents and would have to be held together by covalent links other than disulphide bonds. Therefore, the idea that the 50 kDa protein does not belong to the Esr family and simply shares a highly antigenic domain with Esr proteins is preferred. Esr protein binds a 35 kDa protein in vitro Affinity chromatography with overexpressed Esr2 protein provided preliminary evidence for binding of Esr to a 35 kDa protein present in protein extracts of the lower half of the maize kernel. Attempts to identify the interacting protein by MALDI‐TOF analysis failed (data not shown). This failure was partly due to the fact that the gel band was certainly contaminated with other proteins and partly due to the fact that the identification of maize proteins is hampered by the use of databases established with other species. 2D gels and subsequent sequencing will hopefully yield the expected results. Nevertheless, the possible interaction with a maize protein in the absence of any visible pathogen attack favours the idea that Esr might be a ligand. A way to localize potential receptors would be the use of labelled Esr proteins on kernel sections. Preliminary experiments using biotinylation had to be abandoned due to a strong endogenous activity. Alternative methods based on fusions to alkaline phosphatase are underway. Acknowledgements We thank Pilar Sánchez Testillano and María José Coronado Albí for their invaluable help in electron microscopy studies, Richard Blanc and Alexis Lacroix for plant culture, and Hervé Leyral and Anne‐Marie Thierry for excellent technical assistance. Jim Haseloff is gratefully acknowledged for his Gfp constructs and Thierry Gaude for his advice all along this work. View largeDownload slide Fig. 1. Purification of Esr proteins. SDS–PAGE of total bacterial extracts without (lanes 2, 5 and 8) or with (lanes 3, 6 and 9) induction of Esr expression. Esr proteins purified by affinity chromatography were loaded in lanes 4, 7 and 10. The sizes of the size standard (st) in lane 1 are indicated in kDa. View largeDownload slide Fig. 1. Purification of Esr proteins. SDS–PAGE of total bacterial extracts without (lanes 2, 5 and 8) or with (lanes 3, 6 and 9) induction of Esr expression. Esr proteins purified by affinity chromatography were loaded in lanes 4, 7 and 10. The sizes of the size standard (st) in lane 1 are indicated in kDa. View largeDownload slide Fig. 2. Specificity and titre of Esr antisera. (A) Western blots of total bacterial extracts after induction of Esr expression were revealed with the respective Esr antisera (+) or with the corresponding pre‐immun sera (–). (B) Serial dilutions of purified Esr proteins (10 µg in the undiluted sample) were detected with the respective Esr antisera. View largeDownload slide Fig. 2. Specificity and titre of Esr antisera. (A) Western blots of total bacterial extracts after induction of Esr expression were revealed with the respective Esr antisera (+) or with the corresponding pre‐immun sera (–). (B) Serial dilutions of purified Esr proteins (10 µg in the undiluted sample) were detected with the respective Esr antisera. View largeDownload slide Fig. 3. Sequence similarities between Esr proteins, Clv3 and Cle proteins. The deduced amino acid sequences of Esr1, Esr2, Esr3, and Clv3 were aligned with the consensus sequence of Cle proteins (Cock and McCormick, 2001) by the CLUSTAL algorithm. Amino acid residues differing from the consensus are shaded in black. The conserved region is shaded in grey. Signal peptides predicted by SignalP (Nielsen et al., 1997) are boxed. View largeDownload slide Fig. 3. Sequence similarities between Esr proteins, Clv3 and Cle proteins. The deduced amino acid sequences of Esr1, Esr2, Esr3, and Clv3 were aligned with the consensus sequence of Cle proteins (Cock and McCormick, 2001) by the CLUSTAL algorithm. Amino acid residues differing from the consensus are shaded in black. The conserved region is shaded in grey. Signal peptides predicted by SignalP (Nielsen et al., 1997) are boxed. View largeDownload slide Fig. 4. Immunochemical analysis of Esr proteins. Endosperm proteins were separated by SDS–PAGE. After electroblotting, the nitrocellulose membranes were immuno‐reacted with Esr2 antiserum (+) or pre‐immune serum (–). Proteins were extracted from the basal part (lanes 1, 2, 5–14) or the upper part (lanes 3, 4) of maize endosperm, in SDG buffer (lanes 1–4, 9–14), guanidinium buffer (lanes 5, 6) or after cell wall digestion (lanes 7, 8). Molecular masses of a size standard are indicated in kDa. Arrowheads indicate bands revealed by the Esr2 antiserum. View largeDownload slide Fig. 4. Immunochemical analysis of Esr proteins. Endosperm proteins were separated by SDS–PAGE. After electroblotting, the nitrocellulose membranes were immuno‐reacted with Esr2 antiserum (+) or pre‐immune serum (–). Proteins were extracted from the basal part (lanes 1, 2, 5–14) or the upper part (lanes 3, 4) of maize endosperm, in SDG buffer (lanes 1–4, 9–14), guanidinium buffer (lanes 5, 6) or after cell wall digestion (lanes 7, 8). Molecular masses of a size standard are indicated in kDa. Arrowheads indicate bands revealed by the Esr2 antiserum. View largeDownload slide Fig. 5. Chromatography of proteins interacting with Esr2. (A) Extracts of bacteria overexpressing 6 His‐tagged mature Esr2 protein were incubated with a protein extract from the base of 9 DAP maize kernels, loaded on a Ni‐NTA column and eluted with an imidazole gradient. The eight fractions with the highest protein content (F6–F13) were separated by SDS–PAGE and visualized by silver staining. Molecular masses of the size standard are indicated in kDa. Bands marked with arrowheads are discussed in the text. (B) Control without maize kernel extract. (C) Control without bacterial extract. View largeDownload slide Fig. 5. Chromatography of proteins interacting with Esr2. (A) Extracts of bacteria overexpressing 6 His‐tagged mature Esr2 protein were incubated with a protein extract from the base of 9 DAP maize kernels, loaded on a Ni‐NTA column and eluted with an imidazole gradient. The eight fractions with the highest protein content (F6–F13) were separated by SDS–PAGE and visualized by silver staining. Molecular masses of the size standard are indicated in kDa. Bands marked with arrowheads are discussed in the text. (B) Control without maize kernel extract. (C) Control without bacterial extract. View largeDownload slide Fig. 6. Immunolocalization of Esr proteins. Sagittal sections of maize kernels at 9 DAP were used to immunolocalize Esr proteins with Esr2 antiserum (A) or preimmune serum (B). Arrowheads indicate signal in the ESR and arrows point at signal in the BETL: e, embryo; en, endosperm; p, pericarp. The scale bar represents 1 mm. View largeDownload slide Fig. 6. Immunolocalization of Esr proteins. Sagittal sections of maize kernels at 9 DAP were used to immunolocalize Esr proteins with Esr2 antiserum (A) or preimmune serum (B). Arrowheads indicate signal in the ESR and arrows point at signal in the BETL: e, embryo; en, endosperm; p, pericarp. The scale bar represents 1 mm. View largeDownload slide Fig. 7. Immunogold labelling of maize kernels. Ultrathin lowicryl sections of 9 DAP maize kernels were exposed to purified IgG fraction of Esr2 antiserum (A–E) or preimmune serum (F), reacted with goat anti‐rabbit IgG antibodies conjugated to colloidal gold particles and observed in the electron microscope. (A–D, F) Cells of the embryo surrounding region. (E) Cells of the starchy endosperm. Arrows point at some of the gold particles corresponding to signal. The dotted line in (D) indicates the boundary between the nucleus (top) and the cytoplasm (bottom): cc, condensed chromatin masses; cw, cell wall; cy, cytoplasm; er, endoplasmatic reticulum; G, Golgi apparatus; ic, interchromatin region; n, nuucleus; nl, nucleolus; va, vacuole; ve, secretion vesicle. View largeDownload slide Fig. 7. Immunogold labelling of maize kernels. Ultrathin lowicryl sections of 9 DAP maize kernels were exposed to purified IgG fraction of Esr2 antiserum (A–E) or preimmune serum (F), reacted with goat anti‐rabbit IgG antibodies conjugated to colloidal gold particles and observed in the electron microscope. (A–D, F) Cells of the embryo surrounding region. (E) Cells of the starchy endosperm. Arrows point at some of the gold particles corresponding to signal. The dotted line in (D) indicates the boundary between the nucleus (top) and the cytoplasm (bottom): cc, condensed chromatin masses; cw, cell wall; cy, cytoplasm; er, endoplasmatic reticulum; G, Golgi apparatus; ic, interchromatin region; n, nuucleus; nl, nucleolus; va, vacuole; ve, secretion vesicle. View largeDownload slide Fig. 8. Subcellular localization of the Esr2‐Gfp fusion protein. After transformation onion epidermis cells were incubated for 24 h at pH 5.7 (A–H) or pH 7 (I–L). Cells transiently expressing different Gfp fusion proteins were photographed twice: with a filter 505–550 nm (A, C, E, G, I, K) or without filter (B, D, F, H, J, L). (I–L) Cells were plasmolysed 30 min prior to observation. (A, B, I, J) Esr2‐Gfp fusion protein. (C, D) Gfp targeted to the endoplasmatic reticulum. (E, F) Gfp targeted to the nucleus. (G, H) Gfp remaining in the cytoplasm. (K, L) Gfp targeted to the cell wall. View largeDownload slide Fig. 8. Subcellular localization of the Esr2‐Gfp fusion protein. After transformation onion epidermis cells were incubated for 24 h at pH 5.7 (A–H) or pH 7 (I–L). Cells transiently expressing different Gfp fusion proteins were photographed twice: with a filter 505–550 nm (A, C, E, G, I, K) or without filter (B, D, F, H, J, L). (I–L) Cells were plasmolysed 30 min prior to observation. (A, B, I, J) Esr2‐Gfp fusion protein. (C, D) Gfp targeted to the endoplasmatic reticulum. (E, F) Gfp targeted to the nucleus. (G, H) Gfp remaining in the cytoplasm. (K, L) Gfp targeted to the cell wall. References AmeroSA, James TC, Elgin SC. 1994. Production of antibodies using proteins in gel bands. Methods in Molecular Biology  32, 401–406. Google Scholar BisselingT. 1999. The role of plant peptides in intercellular signalling. Current Opinion in Plant Biology  2, 365–368. Google Scholar BonelloJ‐F, Opsahl‐Ferstad H‐G, Perez P, Dumas C, Rogowsky PM. 2000. Esr genes show different levels of expression in the same region of maize endosperm. Gene   246, 219–227. Google Scholar BradshawRA. 1989. Protein translocation and turnover in eukaryotic cells. Trends in Biochemical Sciences   14, 276–279. Google Scholar ChengWH, Taliercio EW, Chourey PS. 1996. The Miniature1 seed locus of maize encodes a cell wall invertase required for normal development of endosperm and maternal cells in the pedicel. The Plant Cell  8, 971–983. Google Scholar CockJM, McCormick S. 2001. A large family of genes that share homology with CLAVATA3. Plant Physiology  (in press). Google Scholar DavisRW, Smith JD, Cobb BG. 1990. A light and electron microscope investigation of the transfer cell region of maize caryopses. Canadian Journal of Botany  68, 471–479. Google Scholar DingwallC, Laskey RA. 1991. Nuclear targeting sequences—a consensus? Trends in Biochemical Sciences   16, 478–481. Google Scholar FinerJJ, Vain P, Jones MW, Mcmullen MD. 1992. Development of the Particle Inflow Gun for DNA Delivery to Plant Cell s. Plant Cell  Reporter  11, 323–328. Google Scholar FletcherJC, Brand U, Running MP, Simon R, Meyerowitz EM. 1999. Signaling of cell fate decisions by CLAVATA3 in Arabidopsis shoot meristems. Science  283, 1911–1914. Google Scholar GaudeT, Friry A, Heizmann P, Mariac C, Rougier M, Fobis I, Dumas C. 1993. Expression of a self‐incompatibility gene in a self‐compatible line of Brassica oleracea. The Plant Cell  5, 75–86. Google Scholar GavishH, Vardi A, Fluhr R. 1991. Extracellular proteins and early embryo development in Citrus nucellar cell cultures. Physiologia Plantarum  82, 606–616. Google Scholar GerdesJT, Tracy WF. 1993. Pedigree diversity within the Lancaster surecrop heterotic group of maize. Crop Science  33, 334–337. Google Scholar GottesmanS, Halpern E, Trisler P. 1981. Role of sulA and sulB in filamentation by lon mutants of Escherichia coli K‐12. Journal of Bacteriology  148, 265–273. Google Scholar HuerosG, Royo J, Maitz M, Salamini F, Thompson RD. 1999. Evidence for factors regulating transfer cell‐specific expression in maize endosperm. Plant Molecular Biology  41, 403–414. Google Scholar HuerosG, Varotto S, Salamini F, Thompson RD. 1995. Molecular characterization of BET1, a gene expressed in the endosperm transfer cells of maize. The Plant Cell  7, 747–757. Google Scholar JacksonDP. 1991. in situ hybridization in plants. In: Bowles DJ, Gurr SJ, McPhereson M, eds. Molecular plant pathology: a practical approach. Oxford University Press, 163–174. Google Scholar KyleDJ, Styles ED. 1977. Development of aleurone and sub‐aleurone layers in maize. Planta  137, 185–193. Google Scholar MagnardJ‐L, Le Deunff E, Domenech J, Rogowsky PM, Testillano PS, Rougier M, Risueno MC, Vergne P, Dumas C. 2000. Endosperm‐specific features at the onset of microspore embryogenesis in maize. Plant Molecular Biology  44, 559–574. Google Scholar MasonHS, Mullet JE. 1990. Expression of two soybean vegetative storage protein genes during development and in response to water deficit, wounding, and jasmonic acid. The Plant Cell  2, 569–579. Google Scholar MatsubayashiY, Sakagami Y. 1996. Phytosulphokine, sulphated peptides that induce the proliferation of single mesophyll cells of Asparagus officinalis L. Proceedings of the National Academy of Sciences , USA   93, 7623–7627. Google Scholar Matthys‐RochonE, Piola F, Bonello J‐F, Dumas C. 1997. Genes involved in early embryo and endosperm development in flowering plants: a minireview. Bulletin of the Polish Academy of Science  45, 157–165. Google Scholar MorrisseyJH. 1981. Silver stain for proteins in polyacrylamide gels: a modified procedure with enhanced uniform sensitivity. Annals of Biochemisty   117, 307–310. Google Scholar NielsenH, Engelbrecht J, Brunak S, von Heijne G. 1997. Identification of prokaryotic and eukaryotic signal peptides and prediction of their cleavage sites. Protein Engineering   10, 1–6. Google Scholar NiggEA, Baeuerle PA, Luhrmann R. 1991. Nuclear import–export: in search of signals and mechanisms. Cell  66, 15–22. Google Scholar OlsenO‐A, Linnestad C, Nichols SE. 1999. Developmental biology of the cereal endosperm. Trends in Plant Science  4, 253–257. Google Scholar Opsahl‐FerstadHG, Le Deunff E, Dumas C, Rogowsky PM. 1997. ZmEsr, a novel endosperm‐specific gene expressed in a restricted region around the maize embryo. The Plant Journal  12, 235–246. Google Scholar RandolphLF. 1936. Developmental morphology of the caryopsis in maize. Journal of Agricultural Research   53, 882–916. Google Scholar SchelJHN, Kieft H. 1986. An ultrastructural study of embryo and endosperm development during in vitro culture of maize ovaries (Zea mays). Canadian Journal of Botany  64, 2227–2238. Google Scholar SernaA, Maitz M, O’Connell T, Santandrea G, Thevissen K, Tienens K, Hueros G, Faleri C, Cai G, Lottspeich F, Thompson RD. 2001. Maize endosperm secretes a novel antifungal protein into adjacent maternal tissue. The Plant Journal  25, 687–698. Google Scholar ShowalterAM. 1993. Structure and function of plant cell wall proteins. The Plant Cell  5, 9–23. Google Scholar TrotochaudAE, Jeong S, Clark SE. 2000. CLAVATA3, a multimeric ligand for the CLAVATA1 receptor‐kinase. Science  289, 613–617. Google Scholar VainP, Keen N, Murillo J, Rathus C, Nemes C, Finer JJ. 1993. Development of the particle inflow gun. Plant Cell , Tissue and Organ Culture   33, 237–246. Google Scholar van LammerenAAM. 1986. Developmental morphology and cytology of the young maize embryo (Zea mays L.). Acta Botanica Neerlandica  35, 169–188. Google Scholar
TI - Esr proteins are secreted by the cells of the embryo surrounding region
JF - Journal of Experimental Botany
DO - 10.1093/jxb/erf010
DA - 2002-07-01
UR - https://www.deepdyve.com/lp/oxford-university-press/esr-proteins-are-secreted-by-the-cells-of-the-embryo-surrounding-01aucpawAS
SP - 1559
EP - 1568
VL - 53
IS - 374
DP - DeepDyve
ER -