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

J. Biol. Chem., Vol. 275, Issue 49, 38190-38196, December 8, 2000

This Article Abstract Full Text (PDF) All Versions of this Article: 275/49/38190    most recent M006093200v1 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 Egas, C. Articles by Faro, C. Search for Related Content PubMed PubMed Citation Articles by Egas, C. Articles by Faro, C. Social Bookmarking                      What's this?

The Saposin-like Domain of the Plant Aspartic Proteinase Precursor Is a Potent Inducer of Vesicle Leakage^*

Conceição Egas^§, Nuno Lavoura, Rosa Resende, Rui M. M. Brito^¶, Euclides Pires, Maria C. Pedroso de Lima, and Carlos Faro

From the  Centro de Neurociências de Coimbra, Universidade de Coimbra, 3004-517 Coimbra, Portugal, the  Departamento de Bioquimica, Faculdade de Ciências e Tecnologia, Universidade de Coimbra, 3001-401 Coimbra, Portugal, and the ^¶ Departamento de Química, Faculdade de Ciências e Tecnologia, Universidade de Coimbra, 3004-535 Coimbra, Portugal

Received for publication, July 11, 2000, and in revised form, September 11, 2000

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

A unique feature of plant aspartic proteinase precursors is the presence of an internal domain, known as plant-specific insert, whose function is not completely understood. The three-dimensional structure of the plant-specific insert resembles that of saposin-like proteins, a group of lipid-binding proteins involved in a variety of physiological processes. Here we show that recombinant plant-specific insert is able to interact with phospholipid vesicles and to induce leakage of their contents in a pH- and lipid-dependent manner. The leakage activity is higher at pH 4.5 and requires the presence of acidic phospholipids such as phosphatidylserine. To determine whether the same effect could be observed when the plant-specific insert is part of the precursor form, procardosin A and a mutant form lacking this specific domain were produced and characterized. Procardosin A displays a similar activity profile, whereas the mutant without the plant-specific insert shows only residual activity. These findings indicate that the plant-specific insert domain of plant aspartic proteinases mediates an interaction of their precursors with phospholipid membranes and induces membrane permeabilization. It is therefore possible that the plant-specific insert, alone or in conjunction with the proteolytic activity of plant aspartic proteinases, may function either as a defensive weapon against pathogens or in late autolysis of plant cells.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Plant aspartic proteinases have been purified and characterized from several species. They share high sequence and structure homology, as well as general biochemical properties, with the animal and microbial aspartic proteinases (1). A unique feature present in most aspartic proteinases of plant origin is an extra protein domain of about 100 amino acids, generally named plant-specific insert (PSI)¹, that has no homology to any other aspartic protease sequence (2). The PSI is present in the precursors of plant aspartic proteinases but is absent from the mature form of the enzymes (2-6). The function of this plant-specific insert is still unclear. An important role in the vacuolar targeting of aspartic proteinase precursors has been proposed, either by its direct interaction with the membrane, where it would act as a targeting signal (4), or by the interaction of a putative membrane binding region, formed by a cluster of the mature protein and the PSI in prophytepsin, with the membrane itself or with a receptor in the Golgi apparatus (7). It has been suggested (8) that the PSI would interact with the membrane probably in the endoplasmic reticulum as a prerequisite for signaling plant aspartic proteases into the vacuole, because previous work identified a putative vacuolar targeting sequence on the C terminal end of the aspartic protease of cardoon (5). On the other hand, the PSI could be implicated in the correct folding of the enzyme, as suggested by the lack of activation of a mutant of recombinant procyprosin bearing a deletion of the PSI sequence (2).

The plant-specific insert bears high homology with a family of proteins, the saposin-like proteins (SAPLIPs). This family includes proteins such as saposins A, B, C, and D, NK-lysin, granulysin, surfactant protein B, amoebapores, and domains of acid sphingomyelinase and acyloxyacyl hydrolase (9-12). The PSI is not a true saposin but a swaposin, corresponding to a permutation of amino- and carboxyl-terminal saposin fragments, where the C-terminal portion of one saposin is linked by a sequence of about 20 amino acids to the N-terminal portion of the other saposin (10). The sequence homology of this family of proteins resides in three conserved disulfide bridges, a set of hydrophobic residues, and a consensus glycosylation site. The structure of NK-lysin was elucidated by NMR, revealing five -helices folded into a single globular domain. The hydrophobic conserved residues are located in the interior of the globular domain, whereas solvent-exposed side chains are those belonging to less conserved amino acid residues in the SAPLIP family (9). Recently, the crystal structure of prophytepsin, the precursor of the aspartic protease of barley, was reported, revealing that the PSI forms an independent globular domain with five -helices, quite similar to the structure described for NK-lysin (7). Because of the homology within the saposin-like proteins, the proteins of this family most probably share the five -helical globular structure. The proteins belonging to the SAPLIP group are implicated in different physiological functions. Saposins are involved in the lysosomal degradation of sphingolipids (11); NK-lysin has antibacterial activity and the capability to lyse tumor cells but not red blood cells (13); granulysin has antimicrobial activity (12); and surfactant protein B has an important function in the reduction of the surface tension of pulmonary surfactant (14). On the other hand, amoebapores, the pore-forming peptides of Entamoeba histolytica, induce lysis of bacteria and eukaryotic cells (15). The common theme among these proteins appears to be the membrane interaction required for their specific functions.

We have been studying two aspartic proteinases from the flowers of the cardoon Cynara cardunculus L., cardosin A and B (16). Cardosin A, the most abundant, is produced as a single chain zymogen, procardosin A, from which the PSI and the N-terminal propeptide segment are removed to render an active two-chain enzyme (5). The enzyme is mainly detected in pistils, where it accumulates in protein storage vacuoles of the epidermic papillae of the stigma, the pollen-receptive surface, and in a lesser degree in the vacuolar protein masses of epidermic cells of the style (5). Recent studies revealed that expression of cardosin A is developmentally regulated. High levels of mRNA have been identified in the first stages of floral development; then the precursor is converted into the mature form as the flower matures and is accumulated until the later stages of flower senescence, where no mRNA is found (8). The function of cardosin has not been completely clarified, but the data collected so far suggest an involvement in pollen-pistil interaction (17), most probably in adhesion-mediated proteolytic mechanisms associated with pollen tube growth (8), and a possible role in defense mechanisms against pathogens or invasion (17).

To elucidate the function of the plant-specific insert of procardosin A, taking into account the homology between this protein domain and the SAPLIP members, we cloned and expressed recombinant PSI and studied its interaction with membrane vesicles. The membrane interaction of recombinant procardosin A and of a mutated procardosin A lacking the PSI was also determined. To our knowledge this is the first report on the ability of the plant-specific insert, a SAPLIP of plant origin, to interact with membranes and induce vesicle leakage.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Construction of the Expression Vector for Recombinant PSI-- The sequence coding for the plant-specific insert of procardosin A was amplified by PCR using as a template a full-length precursor cDNA clone of cardosin A, obtained by the method described in Faro et al. (8). The primers 5'-TGAAGTGGATAAGTGTTCAC-3' and 5'-GGATCCGTCATGAACCAGCAATGC-3' (forward and reverse, respectively) were used for the amplification. The reverse primer introduced a BamHI restriction site, allowing for directed cloning into the expression vector pGEX-1T. The PCR product was cloned in pCR 2.1 vector (Invitrogen, Groningen, Netherlands). This vector was digested with BamHI/EcoRI, and the insert was subcloned into the pGEX-1T vector (Amersham Pharmacia Biotech). Positive clones were selected by restriction analyses, and the correct orientation was confirmed by sequencing on a Vistra DNA Automatic Sequencer 725 (Amersham Pharmacia Biotech).

Construction of the Expression Vectors for Procardosin A and Mutated Procardosin A-- The procardosin A (pCA) cDNA was cloned as described previously (8). The construction of the mutant cDNA of procardosin A without the PSI sequence (pCA#PSI) was as follows. Using a cDNA encoding the full-length cardosin A as template, a first PCR reaction was performed with Pfu polymerase. The product thus obtained (PCR1) had the prepro sequence and the 31-kDa chain coding regions of cardosin A cDNA followed by 4 glycines in the corresponding C terminus. A second PCR reaction with Pfu polymerase enzyme using the same full-length cDNA of cardosin A as template yielded one fragment (PCR2) that corresponded to the 15-kDa chain of cardosin A with the same 4 glycine sequence in its N terminus. The two PCR fragments (PCR1 and PCR2) were fused in a blunt-ended ligation reaction using T4 DNA ligase (Life Technologies, Inc.). The pCA#PSI cDNA was then obtained through a PCR reaction with the enzyme Taq polymerase using the ligation product as template and the same primers used to amplify procardosin A with NheI sites in the 5' ends. The amplified product was purified, cloned in pCR 2.1 vector, and later subcloned into the NheI site of the pET11-b vector (Novagene, Madison, WI) for expression. The correct cloning direction of positive clones was confirmed by sequencing.

Expression and Purification of Recombinant PSI-- The recombinant plasmid (pGEX-PSI) coding for the plant-specific insert was transformed into Escherichia coli strain BL21, and the recombinant PSI was expressed as a fusion protein with glutathione S-transferase. Bacterial cultures were grown at 37 °C to an A_{600 nm} of 1.0. Protein expression was induced by addition of isopropyl-1-thio--D-galactopyranoside at a final concentration of 0.1 mM. After 2 h at 30 °C, the cells were harvested by centrifugation (4 °C, 8000 × g, 10 min) and washed with 10 mM Na₂HPO₄, 1.8 mM KH₂PO₄, 140 mM NaCl, 2.7 mM KCl, pH 7.3 (phosphate-buffered saline). The recovery of the fusion protein was done according to the method of Frangioni and Neel (18). Briefly, the cells were treated with lysozyme (100 µg/ml) in 10 mM Tris pH 8.0, 150 mM NaCl, 1 mM EDTA for 15 min on ice. Dithiothreitol was added to a final concentration of 5 mM. The protein was then solubilized by the addition of the detergent N-laurylsarcosine (Sigma Aldrich Quimica) to a final concentration of 0.25%, and the resulting mixture was sonicated for 1 min in a water bath (Sonorex RK100H). Clarification of the lysate was obtained by centrifugation at 15,000 × g for 15 min at 4 °C. Triton X-100 was added to the supernatant at the same molar ratio as N-laurylsarcosine, and the protein solution was incubated overnight at 4 °C with the affinity resin glutathione-Sepharose (Amersham Pharmacia Biotech), previously equilibrated in phosphate-buffered saline. The hydrolysis of PSI from the glutathione S-transferase was performed with the addition of thrombin (Amersham Pharmacia Biotech) for 24 h at 20 °C with the fusion protein still bound to the resin. The resulting fraction of PSI was further purified by ion-exchange chromatography in a Hitrap Q column (Amersham Pharmacia Biotech) equilibrated in 25 mM Tris pH 8.0 and eluted with a sodium chloride gradient. Recombinant PSI purity was assessed by SDS-PAGE (Phast system, Amersham Pharmacia Biotech). Glutathione S-transferase was produced by the above procedure, using the vector pGEX-1T without any insertion, and used as a control protein.

Expression and Purification of Recombinant Procardosin A and Mutated Procardosin A-- Both recombinant proteins (pCA and pCA#PSI) were synthesized as "inclusion bodies" in E. coli strain BL21 (DE3). The cells were grown to an A_{600 nm} of 0.5-0.7 at 30 °C. At this time, the expression of the proteins was induced by the addition of isopropyl-1-thio--D-galactopyranoside to a final concentration of 0.5 mM, and the incubation at 30 °C was continued for an additional 2-3 h. After this period the cells were pelleted by centrifugation for 25 min at 3000 × g and resuspended in 50 mM Tris-HCl, pH 7.2, 0.15 M NaCl (TN buffer). Lysozyme (4 mg/ml) was added, and the mixture was stirred at room temperature for 15 min. After freezing at 80 °C overnight and thawing, 2 mM MgCl₂ and 2 units/ml DNase were added, and the homogenized solution was kept for 1 h at 4 °C with agitation. The solution was then diluted with TN buffer containing 1% Triton X-100 and further stirred at room temperature for 30 min, and the "inclusion bodies" were recovered by centrifugation at 5000 × g for 15 min. The final pellet was dissolved in 8 M urea, 50 mM Tris, 1 mM glycine, 1 mM EDTA, pH 8.0, and stirred overnight at 4 °C to solubilize the recombinant material. The soluble material was dialyzed for 24 h against 25 mM Tris pH 8.0 at 4 °C with continuous agitation. Precipitates, which formed, were removed by centrifugation. The clear supernatant was then applied to a Sepharose CL-6B (Amersham Pharmacia Biotech) column equilibrated in 25 mM Tris pH 8.0 at a flow rate of 20.4 ml/h. The fractions with the recombinant proteins were detected by SDS-PAGE and further fractionated on a fast protein liquid chromatography anion-exchange Hitrap Q column (Amersham Pharmacia Biotech). The elution was developed with a gradient of 0-0.5 M NaCl in 25 mM Tris-HCl, pH 8.0. Purity of the recombinant proteins was assessed by SDS-PAGE (Phast system, Amersham Pharmacia Biotech).

Biochemical Characterization-- The identity of the recombinant plant-specific insert was confirmed by N-terminal sequencing of an electroblotted sample of purified protein (Applied Biosystems 473A protein sequencer, Perkin Elmer, Applied Biosystems). The purity of the expressed PSI was also assessed by reverse-phase HPLC in a C4 column (Vydac), equilibrated in 0.1% trifluoroacetic acid, at 1 ml/min. Elution of the bound PSI was accomplished with a linear gradient of 80% acetonitrile in 0.1% trifluoroacetic acid over 35 min. The recombinant PSI was additionally subjected to size exclusion chromatography on Superdex 75 (Amersham Pharmacia Biotech), equilibrated in 20 mM phosphate, 150 mM NaCl, pH 7.0, at 24 ml/h. Apparent molecular mass was calculated by interpolation on an elution volume versus log (molecular mass) calibration curve for four protein standards: bovine serum albumin (66 kDa), carbonic anhydrase (29 kDa), cytochrome c (12.4 kDa), and aprotinin (6.5 kDa). Protein was quantified by the bicinchoninic acid method (Pierce), using lysozyme as the standard protein.

Circular Dichroism (CD) Measurement-- Far-UV CD spectra were recorded on a Jasco J-715 spectropolarimeter equipped with a Peltier temperature control unit, using 1-mm path length cells. Samples were extensively dialyzed against 10 mM sodium phosphate buffer at pH 7.0, and CD spectra of the dialysis buffer were subtracted from the protein spectra. Spectropolarimeter calibration and cell path length were verified using d-10-camphorsulfonic acid. To quantitate the secondary structure content of PSI we used the program CONTIN (19), running on an SGi Octane.

Preparation of Phospholipid Vesicles-- L--Phosphatidylserine (PS), L--phosphatidic acid (PA), L--phosphatidylcholine (PC), and L--phosphatidylethanolamine (PE) were purchased from Avanti Polar Lipids. Large unilamellar vesicles (LUV) were prepared by filter exclusion of multilamellar liposomes using a high pressure apparatus (Lipex Biomembranes, Vancouver, British Columbia, Canada). Dried lipid films of the desired compositions were hydrated by extensive vortexing in 10 mM Hepes, 140 mM NaCl, pH 7.4 for tryptophan fluorescence measurements, or in 80 mM calcein (Sigma), 10 mM Hepes, pH 7.4, for the leakage assays. The resulting multilamellar liposomes were extruded six times through two stacked polycarbonate filters (pore size 0.1 µm; Nucleopore) under a stream of N₂. Free calcein was separated from the dye-containing LUV by size-exclusion chromatography on a Sephadex G-75 column. Phospholipid concentration was measured by the method of Fiske and Subbarow (20). Lipid composition of the vesicles was as indicated in the text and figure legends.

Leakage Assay-- The leakage of liposome contents was monitored by the release of encapsulated calcein at a self-quenching concentration. The leakage experiments were carried out at 37 °C, and fluorescence measurements were performed in a Fluorolog spectrofluorometer (Spex Industries, Edison, NY) equipped with a constant-temperature cell holder and stirrer. The experiments were carried out in 10 mM Hepes, 140 mM NaCl, pH 7.4 or 20 mM Mes, 140 mM NaCl, pH 4.5-6.0. After protein addition, the leakage of calcein into the external medium was monitored by the increase in fluorescence caused by the relief of self-quenching upon calcein dilution (excitation 494 nm, slit width 0.5; emission 517 nm, slit width 1.5). Baseline fluorescence (0%) was determined before protein addition, and complete leakage (100%) was established by lysing the liposomes with 40 mM C₁₂E₈ (octaethylene glycol mono-n-dodecyl ether, Sigma).

Tryptophan Fluorescence Measurements-- The intrinsic fluorescence of the sole tryptophan residue of the plant-specific insert was measured at 37 °C in a Spex Fluorolog spectrofluorometer equipped with a constant-temperature cell holder. The emission spectra were obtained by exciting PSI samples (0.5 µM) at 280 nm (0.7 nm slit width) and scanning emission from 290 to 420 nm (1.2 nm slit width). The fluorescence spectra of PSI were recorded in the presence and absence of LUV (100 µM) composed of PA/PE/PS (1:1:1) and PC/PE (1:1). The spectra were corrected for the intrinsic fluorescence of buffer and phospholipids (scattering).

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Recombinant Plant-specific Insert-- The sequence corresponding to the plant-specific insert of procardosin A was cloned into pGEX-1T and expressed as a fusion protein with glutathione S-transferase. Although the levels of expression of the fusion protein in the culture were high, the purification of the protein upon solubilization with Triton X-100 yielded low protein amounts. The presence of six cysteine residues and a set of hydrophobic amino acids in PSI may account for the increased insolubility of the fusion protein. Different extraction protocols were used to improve PSI recovery, including methods of solubilization of inclusion bodies, without significant increases in protein yield. Limited success was obtained when the detergent N-laurylsarcosine, which is known to reduce protein-protein interactions, was included in the purification procedure. The soluble fusion protein thus obtained (glutathione S-transferase-PSI) was hydrolyzed with thrombin, allowing the independent recovery of PSI. However, this protein fraction showed a two-protein molecular weight band profile in SDS-PAGE. The addition of a mixture of protease inhibitors during extraction and purification did not alter the two-protein band profile; thus an additional purification step was required. The two proteins were efficiently separated by anion-exchange chromatography (Fig. 1). The final yield of PSI expression and purification varied between 150-200 µg/liter of culture. The identity of the purified PSI was confirmed by N-terminal protein sequencing. The recombinant PSI was judged to be homogeneous by SDS-PAGE, presenting an apparent molecular mass of about 12,400 Da, a value that is in close agreement with that calculated from the amino acid sequence. The homogeneity of the protein sample was further confirmed by reverse-phase HPLC, where a single protein peak was detected (Fig. 2A) and by size-exclusion chromatography, where recombinant PSI was eluted in a volume corresponding to 19,000 Da (Fig. 2B).

View larger version (42K): [in this window] [in a new window]   Fig. 1.   SDS-PAGE of purified plant-specific insert. Protein was analyzed by SDS-PAGE in Homogeneous 20 gels on a Phast system. Gels were silver stained. Lane 1, culture sample 2 h after isopropyl-1-thio--D-galactopyranoside induction; lane 2, fraction from glutathione S-transferase; lane 3, purified recombinant plant-specific insert.

View larger version (16K): [in this window] [in a new window]   Fig. 2.   Chromatographic elution profile of recombinant plant-specific insert. Purified PSI was applied to reverse-phase HPLC on a C4 column, equilibrated in 0.1% trifluoroacetic acid at a flow rate of 1 ml/min, and eluted with a linear gradient of 80% acetonitrile in equilibration buffer, over 35 min (A). PSI was also analyzed by size-exclusion chromatography on a Superdex 75 column, equilibrated in 20 mM phosphate buffer, 150 mM NaCl, pH 7.4 at 24 ml/h (B).

To evaluate the secondary structure content of the plant-specific insert, we recorded its far-UV CD spectrum (Fig. 3). A qualitative interpretation of the spectrum, which shows double minima around 207 and 222 nm, indicates the presence of a large percentage of -helix in the protein. This clearly agrees with what is observed for the plant-specific insert in the crystal structure of the zymogen prophytepsin from barley (7). In this structure, the plant-specific domain of prophytepsin also shows a high percentage of -helix.

View larger version (17K): [in this window] [in a new window]   Fig. 3.   Far-UV circular dichroism of plant-specific insert. Far-UV circular dichroism spectra of PSI were recorded at a protein concentration of 11.5 µM in 10 mM sodium phosphate at pH 7.0. MRE, mean residue ellipticity.

Recombinant PSI Induces Vesicle Leakage-- To gain a better insight into the properties and functions of the plant-specific insert, we investigated the interaction of the protein with phospholipid vesicles. This interaction was first determined by the pH-dependent release of the fluorescent probe calcein, trapped inside the vesicles. The influence of the phospholipid composition of the vesicles on leakage was also studied. For all the phospholipid compositions tested the leakage induced by PSI was found to be pH-dependent, with calcein release increasing with decreasing pH (Fig. 4). At physiological pH no calcein release was observed, and at pH 6.0 only residual leakage was detected from vesicles consisting of either PA/PE or PA/PE/PS. As the pH turned into more acidic values, both the initial rates and extents of leakage increased. This effect was more notorious in PA/PE/PS vesicles, where at pH 4.5 most of the leakage was achieved within the first minute (Fig. 4A). At pH 5.0, although the extent of calcein release was smaller than at pH 4.5, again more than 50% of the leakage observed after 10 min of incubation was detected at the first minute after protein addition. However, despite the rapid and extensive calcein release at pH 4.5, complete (100%) leakage was not observed. In PA/PE vesicles a pH-dependent calcein release was also found, although overall values of leakage were smaller. Once more, 50% of the leakage extent observed was detected at the first minute of the experiment. On the other hand, when vesicles of PC/PE were tested, only marginal levels of probe release were measured, either at acidic or neutral pH (data not shown). Glutathione S-transferase was used as a control protein, and its leakage activity was tested at the same concentration as PSI. However, glutathione S-transferase showed no ability to induce calcein release. Altogether these results indicate that the PSI destabilized phospholipid membranes, causing the release of vesicle aqueous contents. This effect was dependent on phospholipid composition, with higher leakage activity in the presence of acidic phospholipids. The pH was also a determinant factor on PSI leakage activity, with increased calcein release in acidic environments, where a rapid initial rate of leakage was found. These data suggest that under acidic conditions, recombinant PSI is a potent inducer of vesicle leakage in the presence of acidic phospholipids.

View larger version (16K): [in this window] [in a new window]   Fig. 4.   PSI-induced leakage of vesicle contents as a function of pH. LUV (15 µM) with entrapped calcein composed of PA/PE/PS (1:1:1) (A) and PA/PE (1:1) (B) were incubated with plant-specific insert (75 nM) in a final volume of 2 ml at 37 °C in a stirred cuvette. The release of calcein was monitored for 10 min in 20 mM Mes, 140 mM NaCl at pH 4.5 (a), 5.0 (b), 5.5 (c), and 6.0 (d). Curves are representative of three to four independent time-course measurements of calcein fluorescence.

The leakage activity of PSI was further studied by evaluating the effect of protein concentration on calcein release in PA/PE vesicles at pH 4.5, while maintaining the lipid concentration constant. Over the protein concentration range tested (10-200 nM) a rapid initial release of calcein was observed. As presented in Fig. 5, the initial rate of calcein release showed a nonlinear dependence on protein concentration, suggesting that leakage was induced by the interaction of PSI multimers with the membranes and/or that there was a limited number of available sites at the membrane surface for protein binding. The latter hypothesis was supported by the results obtained in a set of experiments at pH 4.5, where after protein-induced leakage had reached a plateau (corresponding to approximately 300 s in a curve such as curve a in Fig. 4A), addition of a fresh aliquot of PSI caused only a minor increase in calcein release, suggesting that most of the membrane binding sites were already occupied by the molecules of the first aliquot of PSI added (data not shown).

View larger version (9K): [in this window] [in a new window]   Fig. 5.   Effect of plant-specific insert concentration on the initial rate of calcein release. Recombinant PSI was incubated at different concentrations with PA/PE (1:1) (15 µM) containing calcein for 10 min at pH 4.5. The initial rate of release was determined from the tangent at t = 0 to curves such as those presented in Fig. 4 and plotted as a function of PSI concentration. The data points represent the mean of at least three independent experiments.

To further investigate the interaction of PSI with membrane vesicles, the intrinsic fluorescence of the single tryptophan residue of PSI was measured in the absence and presence of LUV composed of PA/PE/PS or PC/PE at pH 4.5 and 7.4. These particular conditions were chosen because they corresponded to the phospholipid compositions and pH at which PSI induced maximal and minimal leakage, respectively. At pH 7.4 the fluorescence emission maximum of PSI in the absence of vesicles was detected at 345 nm, consistent with the tryptophan being exposed to the solvent (Table I). In the presence of both PA/PE/PS or PC/PE vesicles, similar fluorescence emission values were observed. At pH 4.5 there was a slight shift of the emission maximum to 342 nm for PSI in buffer and to 341 nm in the presence of PC/PE LUV. However, a large blue shift in the emission maximum (332 nm) was obtained in the presence of PA/PE/PS vesicles, consistent with a transfer of the tryptophan to a more hydrophobic environment, most probably reflecting its insertion into the membrane. Additional data supporting the binding of PSI to membranes at pH 4.5 was obtained upon the observation of only a slight increase in calcein release (data not shown) when a fresh aliquot of vesicles was added to the incubation medium of an ongoing leakage experiment with PA/PE/PS (such as curve a in Fig. 4A), suggesting that most of the protein was bound to the first set of vesicles and therefore became unavailable for further leakage.

Recombinant Procardosin A and Procardosin A Lacking PSI Domain Induce Vesicle Leakage-- The results obtained for recombinant PSI motivated the study of the leakage activity of procardosin A and of a mutant construct lacking the PSI domain, under similar conditions as those for individual PSI. Procardosin A and mutant procardosin A were cloned and expressed as inclusion bodies in E. coli. The purified proteins presented apparent molecular masses of 54,200 Da and 42,200 Da on SDS-PAGE, respectively (data not shown), which are in close agreement with the values derived from the amino acid sequence. The activity of recombinant procardosins toward membrane vesicles was studied by measuring the leakage of encapsulated calcein as a function of phospholipid composition and pH. No calcein release was observed with any of the proteins when PC/PE (1:1) vesicles were used (data not shown), reinforcing the importance of the lipid composition, namely the presence of acidic phospholipids, in membrane destabilization. When LUV of PA/PS/PE (1:1:1) were tested, procardosin A showed a similar profile to that of PSI, although with lower leakage extents. Like PSI, at pH 4.5 and 5.0, most of the leakage was achieved within the first minute after protein addition, whereas only residual leakage was detected at pH 6.0, and no calcein release at all was detected at pH 7.4 (Fig. 6A). The leakage activity at pH 4.5 was not altered by the presence of protease inhibitors (1 mM phenylmethylsulfonyl fluoride, 1 mM EDTA, 10 µM E-64, and 1 µM pepstatin), precluding the possibility of calcein leakage being induced by hydrolyzed molecules of PSI. When the leakage activity of mutated procardosin A lacking the PSI domain was measured under similar conditions as to procardosin A, only residual leakage was measured (Fig. 6B), most probably derived from the interaction of a hydrophobic cluster with the membrane vesicles. The data suggest that the binding of procardosin A to membranes is accomplished via PSI, which is also capable of leakage activity while being a part of the precursor protein.

View larger version (15K): [in this window] [in a new window]   Fig. 6.   Leakage of vesicles induced by procardosin A and procardosin A lacking the PSI domain. Recombinant procardosin A (A) and procardosin A lacking the PSI domain (B) (75 nM) were incubated with PA/PE/PS (1:1:1) (15 µM) containing calcein in a final volume of 2 ml. The time course of calcein release as a function of pH was measured at 37 °C in 20 mM Mes, 140 mM NaCl at pH 4.5 (a), 5.0 (b), 5.5 (c), and 6.0 (d). Curves are representative of three to four independent time-course measurements of calcein fluorescence.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

In the present report we studied the interaction of the plant aspartic proteinase-specific insert with membrane vesicles. The PSI was found to interact with phospholipid membranes, promoting the release of the aqueous contents of phospholipid vesicles in a pH-dependent manner, with higher leakage activity at acidic pH. A similar dependence has also been found for saposins C and D regarding their ability to bind and induce vesicle leakage. These observations were related to changes in protein hydrophobicity caused by acidification (21, 22). Several proteins, such as carboxypeptidase E, colicin, and clathrin (21), and peptides, like the N-terminal peptide of HA2 influenza virus hemagglutinin (23) or the GALA peptide (24), are also known to undergo conformational changes at acidic pH, which result in an increased hydrophobicity, thus favoring membrane interaction. In the case of PSI, a change in hydrophobicity could also explain the different activity observed at different pH values, where at acidic pH the conformation adopted would be more favorable for tighter association, leading to a higher degree of embedding into the membrane, thus causing perturbation of the lipid bilayer and the concomitant release of vesicle contents. Taking into account the acidic character of PSI and its negative charge at neutral pH, the protonation of glutamic and aspartic residues at acidic pH might account for an increased hydrophobicity under these conditions. The protonation of the negatively charged residues is also known to stabilize -helices, a preferred structural motif for membrane interaction (23, 25). Regarding PSI features and membrane interaction, it is interesting to note that the absence of protein glycosylation, an exception to the SAPLIP members, did not prevent leakage activity, a result which is consistent with the data recently obtained for saposin D that showed that deglycosylated protein was still able to interact with membrane vesicles (22).

The effect of PSI on the leakage of vesicles was found to vary with lipid composition, indicating a clear preference for acidic phospholipids. Several proteins of the SAPLIP family, namely saposins C and D and surfactant protein B, show a higher leakage activity in the presence of negatively charged phospholipids (21, 22, 26). The preference for interaction with acidic phospholipids is also a feature of several cationic peptides that exhibit antimicrobial activity, such as maiganin 2 (27), defensin A (28), and several other peptide toxins (25). For these peptides an electrostatic interaction between anionic phospholipids in the bacterial membranes and positively charged domains in the protein has been suggested (29-31). A similar mechanism has been proposed for the interaction of NK-lysin with asolectin membranes (13), at least as a first stage of membrane interaction, where a differential charge distribution is responsible for an equatorial belt of positively charged residues, which would interact with the phospholipid head groups. In the case of PSI, experiments carried out with the protein labeled with the fluorescent probe 12-(N-methyl-N-(7-nitrobenz-2-oxa-1,3-diazol-4-yl)), which binds amine groups and whose fluorescence increases in hydrophobic environments, showed no significant fluorescence increase upon interaction with membranes, suggesting that a localized distribution of positive charges may be involved as a first approach to membrane interaction.

Procardosin A was also found to induce vesicle leakage of aqueous contents in a pH-dependent manner. Instead, procardosin A lacking the PSI domain only presented a marginal leakage activity. Although the extent of leakage induced by procardosin A was detected at lower levels than for PSI, initial rates of leakage were quite similar. Altogether these results indicate that the association of procardosin A to membranes is mediated by PSI. The association of procardosin A with microsomal membranes was already observed by Faro et al. (8) and predicted for other plant aspartic proteases (4, 7). The similarity of leakage results obtained for procardosin A and PSI might indicate that PSI is interacting with the membranes through the same structural motifs in both proteins. In the case of procardosin A the slightly reduced activity found might be caused by conformational restrictions induced by the covalent binding of PSI to the rest of the precursor. A recent study identified the structural domains in NK-lysin and granulysin that are responsible for membrane interaction and microbial activity as the motif helix-loop-helix, namely helix 2-loop-helix 3 (32). In the case of PSI, which is a swaposin instead of a saposin, this structural motif corresponds to the N and C termini and in procardosin A is involved in interactions with the rest of the precursor, strongly suggesting that the interaction of PSI with the membrane is accomplished via other structural motifs.

The calcein release induced by procardosin A and PSI in acidic environments was observed with a protein concentration of 75 nM, where most of the leakage was obtained in the first minute of reaction. In leakage assays with other saposin-like proteins and peptides, higher protein concentrations were required to induce lower levels of vesicle leakage: 224 nM for NK-lysin (13), 100 nM for saposin C (21), 800 nM for the N-terminal peptide of HA2 influenza virus hemagglutinin (23), and 50 nM for -hemolysin (33). In comparison to these proteins the plant-specific insert exhibits higher leakage activity in membranes at acidic pH. Such a potent action suggests that the PSI domain is likely to interfere with cell permeability. Saposin-like proteins like NK-lysin, granulysin, and amoebapores are known for their antimicrobial activities (12, 15, 33), and granulysin has also been implicated in apoptotic mechanisms (34). On the other hand, plant aspartic proteinases such as phytepsin (35), nucellin (36), and cardosin B (37) have been found in tissues undergoing programmed cell death. It is possible therefore that the vesicle leakage of PSI may function as part of a defensive mechanism against pathogens and as an effector of cell death.

In any case, the results herein described strongly support the idea that plant aspartic proteinases are bifunctional molecules containing at a certain stage a membrane-destabilizing domain in addition to their proteolytic domain. The PSI domain not only folds as an independent domain, which is able to interact with phospholipid membranes, but also contains a potent vesicle leakage activity whose toxicity has still to be thoroughly investigated in cell culture systems using molecular genetic approaches. In our view, the data presented offer a new approach to the investigation of the functional aspects of the PSI domain with regard to its ability to interact with membranes and induce vesicle leakage, which will certainly help elucidate the function of this domain in the precursors of plant aspartic proteases and in plant cell mechanisms.

	ACKNOWLEDGEMENTS

We thank Dr. Paula Veríssimo for the N-terminal sequencing of the recombinant plant-specific insert and Dr. Teresa Pinheiro and Dr. Alison Roger for the use of their Jasco spectropolarimeter at the University of Warwick, UK.

	FOOTNOTES

^* 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.

^§ Recipient of a fellowship from the PRAXIS XXI program Fundagão para a Ciência e Tecnologia-Ministerio da Ciência e Tecnologia). To whom correspondence should be addressed: Departamento de Biologia Molecular e Biotecnologia, Centro de Neurociências de Coimbra, Instituto Biomédico de Investigacão da Luz e da Imagem, Azinhaga de Santa Comba, 3000 Coimbra, Portugal. Tel.: 351-239-480210; Fax: 351-239-480217; E-mail: cve@imagem.ibili.uc.pt.

Published, JBC Papers in Press, September 11, 2000, DOI 10.1074/jbc.M006093200

	ABBREVIATIONS

The abbreviations used are: PSI, plant-specific insert; SAPLIP, saposin-like protein; PCR, polymerase chain reaction; SDS-PAGE, SDS-polyacrylamide gel electrophoresis; PS, L--phosphatidylserine; PA, L--phosphatidic acid; PC, L--phosphatidylcholine; PE, L--phosphatidylethanolamine; LUV, large unilamellar vesicles; 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:

				S. Senthilkumar, D. Ramasamy, and S. Subramanian Isolation and Partial Characterisation of Milk-clotting Aspartic Protease from Streblus asper Food Science and Technology International, April 1, 2006; 12(2): 103 - 109. [Abstract] [PDF]

				P. Castanheira, B. Samyn, K. Sergeant, J. C. Clemente, B. M. Dunn, E. Pires, J. Van Beeumen, and C. Faro Activation, Proteolytic Processing, and Peptide Specificity of Recombinant Cardosin A J. Biol. Chem., April 1, 2005; 280(13): 13047 - 13054. [Abstract] [Full Text] [PDF]

				C. L. C. Esteves, J. A. Lucey, T. Wang, and E. M. V. Pires Effect of pH on the Gelation Properties of Skim Milk Gels Made From Plant Coagulants and Chymosin J Dairy Sci, August 1, 2003; 86(8): 2558 - 2567. [Abstract] [Full Text] [PDF]

				H. X. You, X. Qi, G. A. Grabowski, and L. Yu Phospholipid Membrane Interactions of Saposin C: In Situ Atomic Force Microscopic Study Biophys. J., March 1, 2003; 84(3): 2043 - 2057. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) All Versions of this Article: 275/49/38190    most recent M006093200v1 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 Egas, C. Articles by Faro, C. Search for Related Content PubMed PubMed Citation Articles by Egas, C. Articles by Faro, C. Social Bookmarking                      What's this?