HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 277, Issue 10, 8708-8715, March 8, 2002

This Article Abstract Full Text (PDF) All Versions of this Article: 277/10/8708    most recent M109346200v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Watanabe, E. Articles by Hara-Nishimura, I. Search for Related Content PubMed PubMed Citation Articles by Watanabe, E. Articles by Hara-Nishimura, I. Social Bookmarking                      What's this?

Calcium-mediated Association of a Putative Vacuolar Sorting Receptor PV72 with a Propeptide of 2S Albumin^*

Etsuko Watanabe^§, Tomoo Shimada^¶, Miwa Kuroyanagi^¶, Mikio Nishimura^§, and Ikuko Hara-Nishimura^¶

From the  Department of Cell Biology, National Institute for Basic Biology, Okazaki 444-8585, Japan, the ^§ Department of Molecular Biomechanics, School of Life Science, The Graduate University for Advanced Studies, Okazaki 444-8585, Japan, and the ^¶ Department of Botany, Graduate School of Science, Kyoto University, Sakyo-ku, Kyoto 606-8502, Japan

Received for publication, September 27, 2001, and in revised form, December 5, 2001

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

PV72, a type I membrane protein with three epidermal-growth factor (EGF)-like motifs, was found to be localized on the membranes of the precursor-accumulating (PAC) vesicles that accumulated precursors of various seed storage proteins. To clarify the function of PV72 as a sorting receptor, we expressed four modified PV72s and analyzed their ability to bind the internal propeptide (the 2S-I peptide) of pro2S albumin by affinity chromatography and surface plasmon resonance. The recombinant PV72 specifically bound to the 2S-I peptide with a K_D value of 0.2 µM, which was low enough for it to function as a receptor. The EGF-like motifs modulated the Ca²⁺-dependent conformational change of PV72 to form a functional pocket for the ligand binding. The binding of Ca²⁺ stabilizes the receptor-ligand complex even at pH 4.0. The association and dissociation of PV72 with the ligand is modulated by the Ca²⁺ concentration (EC₅₀ value = 40 µM) rather than the environmental pH. Overall results suggest that Ca²⁺ regulates the vacuolar sorting mechanism in higher plants.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Most proteins that are synthesized on rough endoplasmic reticulum are delivered to various cellular destinations, including vacuoles and lysosomes. Such sorting involves recognition of targeting signals of proteins by receptors. In mammalian systems, mannose 6-phosphate residues in glycosyl side chains of glycoproteins are known to function as a targeting signal to the lysosomes, and mannose 6-phosphate receptors have been identified as a lysosome sorting receptor (1). In yeast systems, a short stretch sequence of amino acids, QRPL, found in the carboxypeptidase Y, is known to function as a targeting signal to the vacuoles, and Vps10p has been identified as a vacuolar sorting receptor for vacuolar hydrolases (2).

Higher plants have two types of vacuoles: one type, protein storage vacuoles, develop mainly in storage organs, such as seeds, and the other type, lytic vacuoles, which contain various lytic enzymes, develop in the vegetative organs. Both types of vacuoles, however, are found in the same cells of barley roots (3) and of maturing pea seeds (4). In these cells, vacuolar proteins synthesized on the rough endoplasmic reticulum are sorted and delivered to their respective vacuoles. Thus, different targeting machinery for each type of vacuole must be involved in protein transport in these cells.

For the lytic vacuoles, BP-80 was the first putative vacuolar sorting receptor isolated from pea (5). It binds in vitro to the vacuolar-targeting determinants (6, 7). Recently, Humair et al. (8) demonstrated the in vivo binding of BP-80 to the propeptide sequence of barley aleurain in a yeast mutant strain defective for its own vacuolar receptor, Vps10p. An Arabidopsis homolog, AtELP, was also found to interact with the propeptide of an aleurain homolog, AtALEU (9). BP-80 (10) and AtELP (11) are a type I integral membrane protein with epidermal growth factor (EGF)¹-like motifs. They have been shown to be rich in clathrin-coated vesicles (CCVs) and prevacuolar compartments (12, 13). This implied that the cysteine proteinases might be delivered from the Golgi complex to lytic vacuoles via the CCVs in a receptor-dependent manner (9, 14). In contrast, the CCVs isolated from maturing pea seeds were reported to contain no storage proteins (12). Therefore, the transport machinery for storage proteins should be different from that of the lytic enzymes. However, the molecular mechanism responsible for the transport of storage proteins are scarcely elucidated.

In maturing seeds of plants, seed storage protein precursors are synthesized on the rough endoplasmic reticulum and then transported to protein storage vacuoles, where the precursor proteins are converted into the respective mature forms by the action of vacuolar processing enzyme (15-20). Multiple transport pathways have been shown for storage proteins. Hohl et al. (21) and Hinz et al. (12) demonstrated immunocytochemically that dense vesicles with a diameter of about 100 nm associated with Golgi complex contain storage proteins in maturing pea cotyledons. We found the other unique vesicles responsible for delivery of precursors of seed storage proteins and a membrane protein into the vacuoles (22-24) and designated them PAC (precursor-accumulating) vesicles (25). The PAC vesicles are derived from the endoplasmic reticulum and mediate a transport for storage proteins directly to protein storage vacuoles. We have found an integral membrane protein, PV72, with EGF-like motifs in the PAC vesicle fraction prepared from maturing pumpkin seeds (26). We have also shown that PV72 exhibits an affinity for peptides derived from pumpkin 2S albumin (26).

The question is how the association and dissociation of a receptor and the respective ligand is regulated. In mammalian systems, the affinity of mannose 6-phosphate receptor for their glycoprotein ligands was reported to be reduced by the acidic pH of the lysosomes (1). From the analogy to the system, the binding of BP-80 and AtELP to the ligand was described to be regulated by the pH (5, 9). In this study, however, we found that PV72 still has an ability to bind to the ligand at pH 4.0 in the presence of Ca²⁺ and that the EGF-like motifs modulate a Ca²⁺-dependent conformational change of PV72. We describe here a unique mechanism by which Ca²⁺ acts as a modulator for the association and dissociation of PV72 with its ligand.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Plant Materials-- Pumpkin (Cucurbita sp cv Kurokawa Amakuri Nankin) seeds were purchased from Aisan Shubyo Seed Co. (Nagoya, Japan). The seeds were planted in the field. The cotyledons of maturing seeds, freshly harvested 22-28 days after anthesis, were used for the experiments.

Isolation of PAC Vesicles from Maturing Pumpkin Seeds-- PAC vesicles were isolated from pumpkin cotyledons at the middle stage of seed maturation as described previously (25). The cotyledons were homogenized in buffer A (20 mM sodium pyrophosphate, pH 7.5, 1 mM EDTA and 0.3 M mannitol) with an ice-chilled mortar and pestle and filtered through three layers of cheesecloth. The filtrate was centrifuged at 3,000 g for 15 min and the supernatant was centrifuged again at 8,000 g for 20 min at 4 °C. The pellet was suspended in buffer B (10 mM HEPES-KOH, pH 7.2, 1 mM EDTA and 0.3 M mannitol) and layered on 28% Percoll (Amersham Biosciences, Tokyo, Japan) solution. After centrifugation at 40,000 × g for 35 min, the vesicle fraction was pooled and washed once in buffer B. PAC vesicles were collected by the centrifugation at 10,000 × g for 20 min and resuspended in buffer B. The isolated vesicles were subjected to immunoblot analysis.

Ultrastructural Analysis and Immunogold Labeling-- Maturing pumpkin seeds were vacuum-infiltrated for 1 h with a fixative that consisted of 4% paraformaldehyde, 1% glutaraldehyde, and 0.06 M sucrose in 0.05 M cacodylate buffer, pH 7.4. The tissues were then cut into slices of less than 1 mm in thickness with a razor blade and treated for another 2 h with freshly prepared fixative.

Immunoblot Analysis-- Immunoblot analysis was performed essentially as described previously (27), except that in the present study we used specific antibodies against PV72 (diluted 5,000-fold) (26) and horseradish peroxidase-conjugated donkey antibodies against rabbit IgG (diluted 5,000-fold; Amersham Biosciences, Inc.). PV72 was immunologically detected with an enhanced chemiluminescence kit (an ECL system, Amersham Biosciences, Inc.).

Construction of Recombinant Baculoviruses-- Four modified PV72s were expressed in insect cells of Spodoptera frugiperda (Sf21) with a baculovirus expression system (Invitrogen, San Diego, CA). The system includes a transfer vector pBlueBac 4.5 and an expression vector Bac-N-Blue DNA composed of engineered baculoviral Autographa california multiple polyhedrosis virus. The KpnI-SacI fragment of pPV72 was produced from the amplified DNA with PV72 cDNA and a unique primer of 5'-ATT TGT TTA ACT GAA GAC GTG CAC CAC CAC CAC CAC CAC GAT GAG CTT TGA GGT ACC GAA TTC-3' and was ligated with the KpnI-SacI-digested pBlueBac 4.5 to produce the pBlueBac-rPV72. The pBlueBac-rPV72 encoded a fusion protein composed of both the signal sequence and the lumen domain of PV72 followed by a polyhistidine tag and an HDEL sequence. Constructs for three other modified PV72s were produced by the same procedure described above, except for using each primer: 5'-GAT GGA GTC CAC ACG TGT GAA CAC CAC CAC CAC CAC CAC GAT GAG CTT TGA GGT ACC GAA TTC-3' for PV723, 5'-YAC ACT CAT TGT GAA GCT CAC CAC CAC CAC CAC CAC GAT GAG CTT TGA GGT ACC GAA TTC-3' for PV722,3, and 5'-ATT TGT TTA ACT GAA CAC CAC CAC CAC CAC CAC GAT GAG CTT TGA GGT ACC GAA TTC-3' for PV721,2,3. We cotransfected Sf21 cells with Bac-N-Blue DNA and each of the produced plasmids to generate recombinant baculoviruses. The viruses were purified from the supernatant of the transfected cells by a plaque assay to generate a high titer recombinant viral stock (28).

Expression and Purification of the Modified PV72s-- For large scale expression of these modified PV72s, the optimal expression time was determined by monitoring the cellular extract using SDS-PAGE and an immunoblot with anti-PV72 antibodies. Three days after infection, the cells were collected by centrifugation at 500 g for 10 min, washed with PBS and gently suspended in 20 mM HEPES-NaOH, pH 7.0, 150 mM NaCl, 1% CHAPS, and 1 mM CaCl₂. The cells were lysed by three bursts of sonication for 1 min at 10-min intervals on ice and were centrifuged at 750,000 × g for 30 min. The supernatant was loaded on a Hi-Trap chelating column (Amersham Biosciences, Inc.) and was eluted with a gradient of 20-1,000 mM imidazol in the above buffer. The modified PV72 fractions were dialyzed against the HEPES buffer (20 mM HEPES-NaOH, pH 7.0, 150 mM NaCl, and 0.4% CHAPS) plus 1 mM CaCl₂ were concentrated using Centricon 30 (Amicon Inc., Beverly, MA) and then were loaded on a Superdex-200 column (Amersham Biosciences, Inc.) equilibrated with the HEPES buffer plus 1 mM CaCl₂. The purified modified PV72s were concentrated by Centricon 30 and subjected to a protein assay (Nippon Bio-Rad Laboratories, Tokyo) and a binding assay as described below.

Ligand Binding Assay by Affinity Column Chromatography-- Five peptides (SRDVLQMRGIENPWRREG (2S-I), SRDVLQMRGIENPWGGGG (2S-I/3G), SRDVLQMRGIENGWRREG (2S-I/P75G), SRDVLQMRGIGNPWRREG (2S-I/E73G), SRDVLQMRGIENPWRRGG (2S-I/E79G)) were chemically synthesized with a peptide synthesizer (model 431A; Applied Biosystems Inc., Tokyo, Japan) and were used for ligands of affinity columns. Each peptide (10 mg) was immobilized to N-hydroxysuccinimide-activated Sepharose HP (Amersham Biosciences, Inc.) to prepare affinity columns. The modified PV72s were applied to each column equilibrated with the HEPES buffer plus 1 mM CaCl₂ and then eluted with the HEPES buffer plus 2.5 mM EDTA (20 mM HEPES-NaOH, pH 7.0, 150 mM NaCl, 0.4% CHAPS, and 2.5 mM EDTA) with an automated chromatography system (ÄKTA, Amersham Biosciences, Inc.). Each fraction was subjected to SDS-PAGE and subsequently to an immunoblot analysis with anti-PV72 antibodies.

Surface Plasmon Resonance and Kinetic Assays-- We immobilized either 2S-I peptide or 2S-I/3G peptide on a sensor chip for a BIACORE system (BIACORE, Tokyo, Japan) in 10 mM HEPES, pH 7.4, 150 mM NaCl, 1 mM EDTA, and 0.005% P-20 (HBS, BIACORE). Carboxymethylated dextran on a sensor chip (CM5) was activated with the mixture (70 µl) of 0.05 M N-hydroxysuccinimide and 0.05 M N-ethyl-N-(3-diethylaminopropyl)carbodiimide and then coupled with either 2S-I peptide or 2S-I/3G peptide at 25 °C and a flow rate of 5 µl/min for the solutions used on a BIACORE system. A control flow cell was prepared with no peptide. The amount of the coupled peptide on the sensor chip was found to be 1200-1500 resonance units.

The modified PV72s were injected onto the sensor chip for 300 s, and then the HEPES buffer plus 1 mM CaCl₂ was eluted for 200-300 s at 25 °C and a flow rate of 30 µl/min. The sensor chip surface was regenerated with 30 µl of 20 mM HCl to remove residual PV72 from the immobilized peptides. Equal volumes of each protein dilution were also injected over a control flow cell to serve as blank sensorgrams for subtraction of bulk refractive index background and nonspecific binding of analyte. The sensorgrams shown in this study are made by subtracting the sensorgram made with the control flow cell.

Kinetic analysis was performed according to the manufacturer's protocol. The association, dissociation, and regeneration phases were followed in real time as the changes in the relative diffraction. The association phase (0-180 s) was analyzed by nonlinear least squares curve fitting to yield the association rate constants (k_a) as mean values. The dissociation phase (180-300 s) was also analyzed by nonlinear least squares curve fitting to yield the dissociation rate constants (k_d). To avoid mass transport, we worked at a low immobilization level, a high flow rate (30 µl/min), and using suitable concentrations of analyte. Kinetic constants (the association rate constant (k_a), the dissociation rate constant (k_d), and the dissociation constant (K_D = k_d/k_a)) were calculated from the sensorgrams using BIA evaluation software version 2.1 (BIACORE). These kinetic parameters were determined from three independent experiments.

Ca²⁺-dependent Binding to the 2S-I Peptide-- Ca²⁺-dependent binding was analyzed by two methods: surface plasmon resonance analysis and affinity chromatography. Both rPV72 and rPV721,2,3 were dialyzed against the HEPES buffer plus 2.5 mM EGTA, followed by dialysis against the HEPES buffer to remove Ca²⁺ and EGTA. The dialyzed PV72s were injected onto the 2S-I-immobilized sensor chip equilibrated with the HEPES buffer plus 0, 0.02, 0.05 0.1, and 1 mM CaCl₂ or 1 mM MgCl₂ to obtain each sensorgram.

Alternatively rPV72 was applied to the 2S-I affinity column with the HEPES buffer plus 1 mM CaCl₂ and then washed with the HEPES buffer containing decreasing concentration of CaCl₂ (500, 100, 50, 20, 0 µM). Finally, the column was eluted with the HEPES buffer plus 1 mM EGTA.

To compare the Ca²⁺ sensitivity of rPV72 with that of rPV721,2,3, both modified proteins were applied to the 2S-I column with either the HEPES buffer plus 1 mM CaCl₂ or the MES buffer (20 mM MES-NaOH, pH 5.5, 150 mM NaCl, 0.4% CHAPS) plus 1 mM CaCl₂. We used the HEPES buffer plus 50 µM CaCl₂ or the MES buffer plus 50 µM CaCl₂ as the washing solution for the columns and the HEPES buffer plus 2.5 mM EGTA or the MES buffer plus 2.5 mM EGTA as the elution buffers. Each fraction was subjected to SDS-PAGE and subsequently to an immunoblot analysis.

pH-dependent Binding-- To investigate the pH effect on the interaction of modified PV72s with either 2S-I peptide, 2S-I/E73G peptide, 2S-I/P75G peptide, and 2S-I/E79G peptide, the proteins were subjected to each affinity column equilibrated with the HEPES buffer plus 1 mM CaCl₂ or the sodium acetate buffer (20 mM sodium acetate, pH 4.0, 150 mM NaCl, 0.4% CHAPS) plus 1 mM CaCl₂. The column was washed with the respective buffer and eluted with the respective buffer plus 2.5 mM EGTA. Each fraction was subjected to SDS-PAGE and subsequently to an immunoblot analysis.

Spectroscopic Measurements-- Fluorescence emission spectra were recorded from 300 to 400 nm by a fluorescence spectrophotometer (Hitachi, F-4500, Tokyo, Japan) with an excitation wavelength at 280 nm in mixtures containing 1 µg/ml modified PV72s, in the HEPES buffer plus 1 mM CaCl₂ or the HEPES buffer plus 1 mM EDTA, as described previously (29).

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

PV72 Is the Fourth Abundant Protein of PAC Vesicles That Accumulate Storage Protein Precursors-- Previously, we found unique vesicles, PAC vesicles, which mediate the transport of the storage protein precursors to protein storage vacuoles in maturing pumpkin seeds (25). Electron microscopy of the maturing seeds revealed numerous electron-dense PAC vesicles within the cells, as indicated by arrowheads in Fig. 1. Isolation of the PAC vesicles showed that they accumulated proprotein precursors of seed storage proteins, but not their mature forms at all (Fig. 2, lane 1). The precursors included pro2S albumin, proglobulin, and PV100, which is a single precursor of multifunctional proteins including trypsin inhibitors, cytotoxic peptides, and 7S globulin (30). PV72 was detectable as a single band with a molecular mass of 72 kDa on an immunoblot of the PAC vesicles with anti-PV72 antibodies (Fig. 2, lane 2). The PV72 content was enough high to be visible on the SDS gel with Coomassie Blue staining (Fig. 2, lane 1). The result indicates that the pure PAC vesicles contain PV72 as the fourth abundant protein of the vesicles

View larger version (123K): [in this window] [in a new window]   Fig. 1.   Electron microscopy of pumpkin cotyledons at the middle stage of seed maturation. Numerous PAC vesicles (arrowheads) are visible in the cells. PSV, protein storage vacuole; ER, endoplasmic reticulum; LB, lipid body. Bar = 500 nm.

View larger version (24K): [in this window] [in a new window]   Fig. 2.   Localization of PV72 in PAC vesicles that accumulate a proprotein precursor of 2S albumin, a seed storage protein. Isolated PAC vesicles were subjected to SDS-PAGE and subsequent staining with Coomassie Blue (lane 1) or immunoblot analysis with anti-PV72 antibodies (lane 2). p2S, pro2S albumin; pG, proglobulin; PV100, a precursor of a proteinase inhibitor, cytotoxic proteins, and 7S globulin (30). Lane M contains molecular mass markers. The molecular mass of each marker protein is given on the left in kilodaltons.

PV72 Lacking the EGF-like Motifs Still Specifically Binds to the Internal Propeptide of 2S Albumin-- To clarify the ligand-binding mechanism of PV72, we expressed modified rPV72s with a His tag in insect Sf21 cells employing a baculovirus expression system. PV72 is a type I integral membrane protein with EGF-like motifs (26). Fig. 3A shows each construct of the modified proteins; rPV72 corresponds to the lumenal domain of PV72, rPV723 corresponds to the lumenal domain without the third EGF-like motif, rPV722,3 corresponds to the lumenal domain without the second and third EGF-like motifs, and rPV721,2,3 corresponds to the lumenal domain with no EGF-like motifs. These expressed proteins were purified with a chelating column and a gel filtration column. Each final preparation was highly pure as judged from SDS-PAGE with Coomassie Blue staining (Fig. 3B). All of their N-terminal amino acid sequences were determined to be RFVVEKNSLK, which corresponds to the N-terminal sequence of authentic pumpkin PV72 as reported by Shimada et al. (26). The results indicate that a signal peptide of the expressed proteins is correctly processed on the rough endoplasmic reticulum.

View larger version (32K): [in this window] [in a new window]   Fig. 3.   Constructs and the expressed proteins of four modified PV72s in insect cells. A, PV72 is a type I integral membrane protein with three EGF-like motifs at the C terminus of the lumenal domain. The N-terminal domain is indicated by a gray box, and each EGF-like motif is indicated by an open box (boxes 1, 2, and 3, respectively). The transmembrane domain is indicated by a closed box, and the cytoplasmic tail is indicated by an open box. rPV72 was composed of the lumen domain followed by a His tag and the HDEL sequence. rPV723, rPV722,3, and rPV721,2,3 were composed of rPV72 without the third EGF-like motif, without the second and third EGF-like motifs, and without all three of the EGF-like motifs, respectively. B, the four modified PV72s that were expressed in insect Sf21 cells were purified with a chelating column and a gel filtration column. Each purified protein was subjected to SDS-PAGE with Coomassie Blue staining: rPV72 (lane 1), rPV723 (lane 2), rPV722,3 (lane 3), and rPV721,2,3 (lane 4). The molecular mass of each marker protein (lane M) is given on the left in kilodaltons.

To investigate the binding ability of rPV72 to the ligand peptides, we performed a binding assay on the affinity column (2S-I column) that was conjugated with the 2S-I peptide, the internal propeptide of pumpkin 2S albumin. As shown in Fig. 4A (upper), rPV72 bound to the 2S-I column and then eluted with EDTA (discussed below). Previously we reported that the isolated PV72 from the maturing seeds of pumpkin binds to the 2S-I peptide but not to the mutant peptide 2S-I/3G with GGG instead of RRE of the internal propeptide. To clarify the specificity of the binding of rPV72, we used another affinity column (2S-I/3G column) that was conjugated with the 2S-I/3G peptide. As shown in Fig. 4A (lower), rPV72 did not bind to the 2S-I/3G column. The results indicate that the characteristics of rPV72 with respect to ligand binding were the same as those of the authentic PV72 has.

View larger version (30K): [in this window] [in a new window]   Fig. 4.   Modified PV72s showed a specific binding to the 2S-I peptide derived from the internal propeptide of a storage protein, 2S albumin. A, rPV72 was subjected to an affinity column with either the 2S-I peptide or the mutant peptide with a replacement of RRE by GGG (2S-I/3G). Each fraction was subjected to SDS-PAGE and then to an immunoblot with anti-PV72 antibodies. FT, flow-through fraction; W, washing fraction with the HEPES buffer, pH 7.0, with 1 mM CaCl2; EDTA, eluted fraction by 2.5 mM EDTA. B, four modified PV72s were injected onto the sensor chip coupled with either the 2S-I peptide or the 2S-I/3G peptide to obtain the sensorgrams by surface plasmon resonance. The protein concentrations used were 0.7 µM for both rPV72 and rPV723, 1.6 µM for rPV722,3, and 1.4 µM for rPV721,2,3.

To identify the ligand-binding region of PV72, we performed a surface plasmon resonance analysis for four modified PV72s with either the 2S-I sensor chip or 2S-I/3G sensor chip. Each modified PV72 of the same concentration was injected onto the sensor chips. rPV72 bound to the 2S-I peptide, but not to the 2S-I/3G peptide (Fig. 4B), as expected from the result in Fig. 4A. All of the deleted proteins, rPV723, rPV722,3, and rPV721,2,3, also bound to the 2S-I peptide, but not to the 2S-I/3G peptide (Fig. 4B). This demonstrates that the deleted proteins also specifically recognize the RRE sequence of the 2S-I peptide as rPV72 does. These results indicated the N-terminal region of PV72 corresponding to rPV721,2,3 includes a ligand-binding site.

The EGF-like Motifs Modulate the Association and Dissociation of PV72 with the Ligand-- We determined the kinetic parameters for the binding of each modified rPV72 to the 2S-I peptide by surface plasmon resonance. Each modified rPV72 was injected onto the 2S-I sensor chip to start the association reaction. Fig. 5A shows the association and dissociation curves obtained from the respective experiment with four different concentrations (0.07-2 µM) of each protein. The sensorgrams of rPV723, rPV722,3, and rPV721,2,3 exhibited more rapid association followed by more rapid dissociation after the injection was completed than did the sensorgram of rPV72.

View larger version (33K): [in this window] [in a new window]   Fig. 5.   Kinetics for the association and dissociation of the modified PV72s and the 2S-I peptide. A, the surface plasmon resonance profiles for the association and dissociation curves of the modified PV72s and the 2S-I peptide. The coupling efficiency of the 2S-I peptide to the sensor surface was 1,200 resonance units. Each of the modified PV72s was injected onto the 2S-I sensor chip at different concentrations from 0.07 µM to 1.5 µM. B, the kinetic constants, an association rate constant (ka), a dissociation rate constant (kd), and a dissociation constant (KD = kd/ka), were calculated from the above sensorgrams using BIA evaluation software version 2.1. These kinetic parameters were determined from three independent experiments.

The kinetic constants of association and dissociation were calculated from the slopes of the curves, as shown in Fig. 5B. In contrast to small differences of the k_a values among the modified rPV72, large differences of the k_d values were observed. The k_d value of each of rPV723, rPV722,3, and rPV721,2,3 was 16-23-fold higher than the k_d value of rPV72. The apparent equilibrium dissociation constant was determined from the ratio of these two kinetic constants (k_d/k_a). rPV72 has a high enough affinity for the 2S-I peptide (K_D = 0.2 µM) to function as a receptor. The K_D values of each of rPV723, rPV722,3, and rPV721,2,3 were 10-, 19-, and 21-fold higher than the K_d value of rPV72, respectively. Therefore, the affinity of rPV72 for the ligand peptide is much higher than the affinities of the rPV72s lacking the EGF-like motifs. It seems likely that the EGF-like motifs play a role in stabilizing the receptor-ligand complex.

The EGF-like Motifs Modulate a Ca²⁺-dependent Conformational Change of PV72-- The question is how the EGF-like motifs regulate the stability of the ligand binding of PV72. Previously, we found that the third EGF-like motif has a consensus sequence for Ca²⁺ binding, while the first and second motifs do not have such a sequence (26). Fig. 4A shows that rPV72 was eluted from the 2S-I column by the addition of chelating agents. This implied that Ca²⁺ binding to the third EGF-like motif might be important for the ligand binding. To clarify the requirement of Ca²⁺, we performed an analysis of surface plasmon resonance with rPV72 in the presence of either Ca²⁺ or Mg²⁺. Fig. 6A (left) shows that the interaction between rPV72 and the ligand was observed in the presence of Ca²⁺, but not in the presence of Mg²⁺ instead of Ca²⁺. This result indicates that Ca²⁺ is required for PV72 to interact with the 2S-I peptide.

View larger version (23K): [in this window] [in a new window]   Fig. 6.   Ca2+-dependent ligand-binding of PV72 and Ca2+-induced conformational change in PV72. A, divalent cation selectivity for the ligand binding. Either rPV72 or rPV721,2,3 was injected onto the 2S-I sensor chip in the presence of 1 mM CaCl2 or 1 mM MgCl2. B, fluorescence emission spectra showing the Ca2+-induced structural change. The fluorescence emission of Trp and Tyr residues in rPV72 and rPV721,2,3 was measured from 300 to 400 nm after excitation at 280 nm. rPV72 (1 µg/ml) was suspended in the HEPES buffer plus 1 mM CaCl2 (open circles) and the HEPES buffer plus 1 mM EDTA (closed circles). rPV721,2,3 (1 µg/ml) was in the buffer plus 1 mM CaCl2 (open squares) and the HEPES buffer plus 1 mM EDTA (closed squares).

Unexpectedly, however, rPV721,2,3, which lacks the EGF-like motifs, also showed a Ca²⁺-dependent interaction with the 2S-I peptide, but not a Mg²⁺-dependent interaction (Fig. 6A, right), as rPV72 did. This result indicates that the N-terminal region corresponding to rPV721,2,3 has another Ca²⁺-binding site (s), although no consensus sequence for Ca²⁺ binding was found in the region. It should be noted that the affinity of PV72 (K_D = 0.2 µM) was 20-fold stronger than that of rPV721,2,3 (K_D = 4.2 µM). Thus, the Ca²⁺-binding to the EGF-like motif must be required for the high affinity of PV72 for the ligand.

The next question raised is whether the Ca²⁺ binding causes a conformational change of PV72 that results in the higher affinity. To answer the question, we measured Ca²⁺-dependent changes in the fluorescence emission spectra of Tyr or Trp residues in both rPV72 and rPV721,2,3. The rPV72 polypeptide includes 23 Tyr residues and 11 Trp residues. The fluorescence of these residues was monitored in the presence or absence of Ca²⁺. Fig. 6B show a remarkable Ca²⁺-dependent change in the fluorescence emission spectra of rPV72, but only a little change in the spectra of rPV721,2,3. The result suggests that the EGF-like motifs induced a much larger conformational change of PV72 in a Ca²⁺-dependent manner than the N-terminal region corresponding to rPV721,2,3 did.

The Ligand Binding of PV72 Is Regulated by the Ca²⁺ Concentration Rather than pH-- The next issue to be determined was a critical concentration of Ca²⁺ for association and dissociation of PV72. We performed a binding assay under the conditions of various Ca²⁺ concentrations by surface plasmon resonance. The EC₅₀ value of the Ca²⁺-dependent interaction was determined to be 40 µM (Fig. 7A). We also performed another binding assay by affinity chromatography with the 2S-I column. rPV72 that had bound to the 2S-I column was exposed to decreasing concentrations of CaCl₂ from 500, 100, 50, 20, and 0 µM. The rPV72 was eluted from the column under CaCl₂ concentrations lower than 50 µM (Fig. 7B). The CaCl₂ concentration was consistent with the EC₅₀ value determined by the surface plasmon resonance. The results suggested that the interaction of PV72 to the ligand might be regulated by the Ca²⁺ concentration in the respective compartment of the maturing seed cells.

View larger version (26K): [in this window] [in a new window]   Fig. 7.   Regulation of the ligand binding of PV72 by the Ca2+ concentration. A, rPV72 was injected onto the 2S-I sensor chip in the presence of various concentrations of CaCl2. Relative responses were plotted against the concentration of Ca2+. EC50 value for Ca2+ calculated was 40 µM. B, rPV72 was subjected to the 2S-I affinity column and eluted by the HEPES buffer containing CaCl2 of various concentrations from 500 to 0 µM and finally by 1 mM EGTA. Each fraction was subjected to SDS-PAGE and then to an immunoblot with anti-PV72 antibodies. FT, flow-through fraction.

In general, binding of receptors to their ligands are known to be modulated by the environmental pH. This raised a question whether the interaction of PV72 with the ligand is also regulated by pH. To answer this question, we performed an assay with the rPV72 and rPV721,2,3 that had bound to the 2S-I column in the presence of 1 mM CaCl₂. Fig. 8A (upper) shows that the rPV72 that bound at pH 7.0 was not eluted by decreasing the CaCl₂ concentration to 50 µM nor by decreasing the pH to 5.5, but was eluted with an EGTA solution. In contrast, the rPV721,2,3 bound to the column was easily eluted with the neutral buffer (pH 7.0) containing 50 µM CaCl₂ (Fig. 8A, lower). Alternatively, we performed another assay with the rPV72 bound to the 2S-I column at pH 5.5 in the presence of 1 mM CaCl₂. The result was similar as shown in Fig. 8A. The bound rPV72 was not eluted at pH 5.5 in the presence of 50 µM CaCl₂ (Fig. 8B, upper), while the bound rPV721,2,3 was easily eluted with the acidic buffer (pH 5.5) containing 50 µM CaCl₂ (Fig. 8B, lower).

View larger version (38K): [in this window] [in a new window]   Fig. 8.   Modulation of Ca2+-dependent ligand binding of rPV72 by the EGF-like motifs. A, either rPV72 or rPV721,2,3 bound to the 2S-I column in the HEPES buffer, pH 7.0, with 1 mM CaCl2. The elution pattern of each protein was examined with the sequential solutions; the HEPES buffer, pH 7.0, with 50 µM CaCl2, the MES buffer, pH 5.5, with 50 µM CaCl2, and the MES buffer with EGTA. B, either rPV72 or rPV721,2,3 bound to the column in the MES buffer with 1 mM CaCl2. The elution pattern was examined with sequential solutions; the MES buffer with 50 µM CaCl2, the HEPES buffer with 50 µM CaCl2, and the HEPES buffer with EGTA. Each fraction was subjected to SDS-PAGE and then to an immunoblot with anti-PV72 antibodies. FT, flow-through fraction.

The results indicated that the pH change did not affect the interaction between rPV72 and the ligand in the presence of 50 µM CaCl₂. When a more acidic buffer (pH 4.0) was used instead of the pH 5.5 buffer, an elution profile similar to that in Fig. 8A was obtained (data not shown). The presence of 50 µM CaCl₂ made the complex of rPV72 and the ligand stable under acidic conditions. The overall results suggested that the EGF-like motifs might be involved in the stability of the complex in the presence of 50 µM CaCl₂. Thus, it appears that the association and dissociation of PV72 with the ligand was modulated by the Ca²⁺ concentration rather than by the pH.

To clarify the effect of the 2S-I peptide sequence on the binding, we prepared affinity columns conjugated with three mutant peptides: 2S-I/P75G with Gly instead of Pro-75, 2S-I/E73G with Gly instead of Glu-73, and 2S-I/E79G with Gly instead of Glu-79. At neutral pH (pH 7.0), both rPV72 and rPV72D1,2,3 bound to all the columns with the mutant peptides in the presence of 1 mM CaCl₂ and then eluted with the EDTA solution (Fig. 9, right). Even at pH 4.0, both rPV72 and rPV72D1,2,3 bound to the 2S-I column and the 2S-I/P75G column, but they did not bind to either the 2S-I/E73G column or 2S-I/E79G column (Fig. 9, left). The results indicate that the PV72 has an ability to bind to the 2S-I peptide not only at pH 7.0 but also at pH 4.0, in the presence of Ca²⁺. When Glu-73 or Glu-79 of the 2S-I peptide was substituted by Gly, the affinity of either rPV72 or rPV72D1,2,3 for the mutant peptide was reduced at pH 4.0, suggesting that both Glu-73 and Glu-79 are necessary for the binding of PV72 at acidic pH. PV72 might interact with pro2S albumin through the two Glu residues in Ca²⁺-dependent manner.

View larger version (30K): [in this window] [in a new window]   Fig. 9.   Effect of pH on the binding of rPV72s to the 2S-I peptide and the mutant peptides. The affinity column conjugated with each of the 2S-I peptide, 2S-I/P75G, 2S-I/E73G, or 2S-I/E79G was used, as shown on the left. Either rPV72 or rPV721,2,3 was applied to the column under neutral conditions (pH 7.0, right panel) or under acidic conditions (pH 4.0, left panel). The column was washed with the respective buffer and finally with the buffer containing 2.5 mM EGTA. Each fraction was subjected to SDS-PAGE and then to an immunoblot with anti-PV72 antibodies. FT, flow-through fraction; W, washing fraction.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Ca²⁺-mediated Association and Dissociation of PV72 and the Ligand-- Receptor-mediated protein sorting involves an association and dissociation of the receptor and the respective ligand by a modulator. In general, the environmental pH is known to regulate it as a modulator. Lysosomal proteins synthesized on the rough endoplasmic reticulum are reported to be recognized by a receptor and then delivered to the respective acidic compartment, where the ligands are dissociated from the receptor. The dissociation for both mannose 6-phosphate receptor has been shown to occur in the acidic organelles in mammals (1). Similarly in plants, the dissociation for BP-80 was known to occur at pH 4.0 (5).

We found, however, that PV72 binds to the 2S-I peptide even at pH 4.0 in the presence of Ca²⁺. It does not appear that the acidic pH is responsible for the dissociation of PV72 from the 2S-I peptide. Our results demonstrated that Ca²⁺ functions as a modulator for the association and dissociation of PV72 with the internal propeptide of 2S albumin. The Ca²⁺ concentration in the subcellular compartments was reported to range from 1 to 1,500 µM (31). The EC₅₀ (Ca²⁺) value of 40 µM for the ligand binding of rPV72 is reasonable for the regulation of the association and dissociation of the receptor with the ligand within the cells.

PV72 has a consensus sequence for Ca²⁺-binding in the third EGF-like motif (26). The motif might function as a Ca²⁺-binding EGF (cbEGF) domain. Binding of Ca²⁺ to the cbEGF domain might cause a conformational change in the PV72 molecule to make the receptor-ligand complex stable. On the other hand, reduction of the environmental Ca²⁺ concentration might cause the dissociation of the ligand from the receptor.

PV72 has another Ca²⁺-binding site in the N-terminal region corresponding to rPV721,2,3, which lacks a consensus sequence for Ca²⁺ binding. However, binding of Ca²⁺ to the region causes only a small conformational change in the PV72. The N-terminal region might play a role in the formation of the ligand-binding pocket with the assistance of Ca²⁺ (discussed below). The binding of Ca²⁺ to both the N-terminal region and the cbEGF domain of PV72 could induce the formation of a functional pocket for the ligand binding and stabilize the receptor-ligand complex.

The low density lipoprotein receptor has three EGF motifs, one of which is cbEGF (32). The receptor also has a ligand-binding region with another Ca²⁺-binding site (33), and the receptor-ligand complex is stabilized by Ca²⁺ in the receptor molecules (33, 34). Despite the very low identity of the sequence between the low density lipoprotein receptor and PV72, the mechanisms of Ca²⁺-mediated association and dissociation of the receptor and ligand are similar to each other. The Ca²⁺-mediated regulation through the cbEGF domain is reported to be crucial in mammals. Mutations in a cbEGF domain of the low density lipoprotein receptor have been shown to cause familial hypercholesterolemia (33) and a mutation in the cbEGF of fibrillin causes Marfan syndrome (35).

The Protease-associated Domain of PV72 Might Be Involved in the Ligand Binding-- The lumen domain of BP-80 consists of three domains: an N-terminal domain homologous to ReMembR-H2 (RMR) protein, a central domain, and a C-terminal EGF repeat domain (7). It has been shown that the former two domains together determine the NPIR-specific ligand binding site and an EGF repeat domain of BP-80 alters the conformation of the other two domains to enhance ligand binding (7). These results are consistent with our findings that PV72 forms the ligand-binding pocket in the N-terminal region corresponding to rPV721,2,3 and that the EGF-like motifs modulate a Ca²⁺-dependent conformational change of PV72. It seems likely that the EGF-like motifs play a role in stabilizing the receptor-ligand complex. Both the N-terminal region corresponding to rPV721,2,3 and the RMR homology domain of BP-80 contain a protease-associated domain, which is speculated to be involved in substrate determination for peptidases or to form protein-protein interactions (36, 37). It is possible that each protease-associated domain of PV72 and BP-80 mediates the interaction with the ligand.

We found that mutation of either Glu-73 or Glu-79 reduced the affinity of PV72 for the ligand peptide under acidic conditions. Two Glu residues of the 2S-I peptide might be important for stability of the ligand-receptor interaction under acidic conditions. PV72 might interact with the ligand through the two Glu residues, one of which is included in the RRE sequence in the 2S-I peptide. Previously we reported that the RRE sequence was essential for the interaction with PV72 (26). The essential sequence is different from the targeting determinant, NPIR, of the lytic enzymes.

PV72 Might Be Recycled between the PAC Vesicles and Golgi Complex-- When a fusion protein of green fluorescent protein with a transmembrane domain and a cytoplasmic tail of PV72 was expressed in tobacco BY2 cells, a fluorescent Golgi complex was observed. The localization of PV72 in the Golgi complex is supported by the findings that PV72 has a complex glycan.² The peripheral region of the PAC vesicles was labeled with gold particles for complex glycans in the immunoelectron micrograph (25). It is possible that PV72 is recycled between the Golgi complex and the PAC vesicles in maturing seeds. It was suggested that the cytosolic tail of BP-80 is responsible for the retrieve from the prevacuolar compartments to the Golgi complex in the protoplasts of the transgenic tobacco (14).

We clearly demonstrated that the interaction of PV72 with the 2S-I peptide is modulated by the Ca²⁺ concentration. The Ca²⁺ concentration in the Golgi complex was determined to be 300 µM (38), which is high enough for PV72 to bind the ligand. On the other hand, the Ca²⁺ concentration in the endoplasmic reticulum ranges from 1 to 1,500 µM depending on the region of the endoplasmic reticulum (31). Thus, the endoplasmic reticulum-derived PAC vesicles possibly have a Ca²⁺ concentration lower than 50 µM, which could dissociate the ligand from the receptor-ligand complex.

Previously we reported that most of pro2S albumin synthesized on the endoplasmic reticulum are directly incorporated into the PAC vesicles, in a Golgi-independent manner (24, 25). It is possible that PV72 might trap the escaped pro2S albumin that leaves the rough endoplasmic reticulum for the Golgi complex and recruit them from the Golgi complex to the PAC vesicles. On the contrary, BP-80 has been shown to be rich in CCVs, but not in dense vesicles responsible for transport seed globulins in maturing pea seeds (12). BP-80 might transport an NPIR-containing proteinase from Golgi complex to prevacuolar compartments via CCVs. Thus, the possibility cannot be excluded that PV72 mediates the transport of an NPIR-containing proteinase from Golgi complex to the PAC vesicles or from the PAC vesicles to lytic compartments. To clarify the intracellular pathway regulated by the vacuolar sorting receptors, further analysis of subcellular localization of the receptors in Golgi complex or prevacuolar compartments in addition to the PAC vesicles is required. Further investigation of the organ-specific and temporal expression of the receptors will also provide us an insight into the physiological function of them.

	ACKNOWLEDGEMENTS

We thank to Prof. J. C. Rogers (Washington State University) and Prof. D. G. Robinson (University of Heidelberg) for critical reading of the manuscript. We are grateful to Prof. S. Yoshida (Nagoya University) and C. Nanba (National Institute for Basic Biology) for their efforts on growing pumpkin plants and to Y. Makino (National Institute for Basic Biology) for assistance on peptide synthesis.

	FOOTNOTES

^* This work was supported by Grants-in-aid for the "Human Frontier Science Program" (RG0018/2000-M 103), for "Research for the Future Program from the Japan Society for the Promotion of Science" (JSPS-RFTF96L60407), and for Scientific Research from Ministry of Education, Culture, Sports, Science, and Technology of Japan (10182102, 12138205, and 12304049).The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

To whom correspondence should be addressed: Dept. of Botany, Graduate School of Science, Kyoto University, Kyoto 606-8502, Japan. Tel.:/Fax: 81-75-753-4142; E-mail: ihnishi@gr.bot.kyoto-u.ac.jp.

Published, JBC Papers in Press, December 17, 2001, DOI 10.1074/jbc.M109346200

² E. Watanabe, T. Shimada, M. Nishimura, and I. Hara-Nishimura, unpublished data.

	ABBREVIATIONS

The abbreviations used are: EGF, epidermal growth factor; 2S-I, an internal propeptide of pumpkin 2S albumin; CCVs, clathrin-coated vesicles; PAC vesicles, precursor-accumulating vesicles; CHAPS, 3-[(3-cholamidopropyl)dimethylammonio]-1-propanesulfonic acid; MES, 4-morpholineethanesulfonic acid.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				X. Li, S. Chanroj, Z. Wu, S. M. Romanowsky, J. F. Harper, and H. Sze A Distinct Endosomal Ca2+/Mn2+ Pump Affects Root Growth through the Secretory Process Plant Physiology, August 1, 2008; 147(4): 1675 - 1689. [Abstract] [Full Text] [PDF]

				E. L. Avila, M. Brown, S. Pan, R. Desikan, S. J. Neill, T. Girke, M. Surpin, and N. V. Raikhel Expression analysis of Arabidopsis vacuolar sorting receptor 3 reveals a putative function in guard cells J. Exp. Bot., April 1, 2008; 59(6): 1149 - 1161. [Abstract] [Full Text] [PDF]

				M. Yamazaki, T. Shimada, H. Takahashi, K. Tamura, M. Kondo, M. Nishimura, and I. Hara-Nishimura Arabidopsis VPS35, a Retromer Component, is Required for Vacuolar Protein Sorting and Involved in Plant Growth and Leaf Senescence Plant Cell Physiol., February 1, 2008; 49(2): 142 - 156. [Abstract] [Full Text] [PDF]

K. Fuji, T. Shimada, H. Takahashi, K. Tamura, Y. Koumoto, S. Utsumi, K. Nishizawa, N. Maruyama, and I. Hara-Nishimura Arabidopsi