Analysis of Low Abundance Membrane-Associated Proteins from Rat

Aug 14, 2010 - 16 unique proteins, including rat mast cell chymase (RMCP-1) and peptidyl-prolyl cis-trans ... within the acinar cells of the exocrine ...
0 downloads 0 Views 4MB Size
Download PDF
Analysis of Low Abundance Membrane-Associated Proteins from Rat Pancreatic Zymogen Granules Heike Borta,†,‡ Miguel Aroso,†,§ Cornelia Rinn,§ Maria Gomez-Lazaro,§ Rui Vitorino,| Dagmar Zeuschner,⊥ Markus Grabenbauer,# Francisco Amado,| and Michael Schrader*,§ Department of Cell Biology and Cell Pathology, Philipps University of Marburg, Robert Koch Strasse 6, 35037 Marburg, Germany, Max-Planck-Institute of Molecular Biomedicine, Ro¨ntgenstrasse 20, 48149 Mu ¨ nster, Germany, Max-Planck-Institute for Molecular Physiology, Otto-Hahn-Strasse 11, 44227 Dortmund, Germany, Department of Chemistry, and Centre for Cell Biology and Department of Biology, University of Aveiro, Campus Universita´rio de Santiago, 3810-193 Aveiro, Portugal Received January 20, 2010

Zymogen granules (ZG) are specialized storage organelles in the exocrine pancreas that allow the sorting, packaging, and regulated apical secretion of digestive enzymes. As there is a critical need for further understanding of the key processes in regulated secretion to develop new therapeutic options in medicine, we applied a suborganellar proteomics approach to identify peripheral membraneassociated ZG proteins. We focused on the analysis of a “basic” group (pH range 6.2-11) with about 46 spots among which 44 were identified by tandem mass spectrometry. These spots corresponded to 16 unique proteins, including rat mast cell chymase (RMCP-1) and peptidyl-prolyl cis-trans isomerase B (PpiB; cyclophilin B), an ER-resident protein. To confirm that these proteins were specific to zymogen granules and not contaminants of the preparation, we conducted a series of validation experiments. Immunoblotting of ZG subfractions revealed that chymase and PpiB behaved like bona fide peripheral membrane proteins. Their expression in rat pancreas was regulated by feeding behavior. Ultrastructural and immunofluorescence studies confirmed their ZG localization. Furthermore, a chymase-YFP fusion protein was properly targeted to ZG in pancreatic AR42J cells. Interestingly, for both proteins, proteoglycan-binding properties have been reported. The importance of our findings for sorting and packaging during ZG formation is discussed. Keywords: organelle biogenesis • zymogen granules • exocrine pancreas • 2D gel electrophoresis • tandem mass spectrometry • membrane-associated proteins

Introduction Zymogen granules (ZGs) are specialized storage organelles within the acinar cells of the exocrine pancreas. Their main cargo is constituted by pancreatic digestive enzymes as inactive precursors which are released by regulated apical secretion, triggered by an external stimulus. ZG formation is initiated at the trans-Golgi network (TGN) where the regulated secretory ZG proteins coaggregate at the mildly acidic pH and high Ca2+ levels and condensing vacuoles/immature secretory granules are formed.1-4 Their maturation involves further concentration of the cargo proteins with selective removal of components not destined for regulated secretion, and a reduction in granule size.5-7 The mature ZGs are transported to the apical domain * To whom correspondence should be addressed. Michael Schrader, Centre for Cell Biology & Dept. of Biology, University of Aveiro, Campus Universita´rio de Santiago, 3810-193 Aveiro, Portugal. Tel.: + 351-234 370 200 ext. 22789. Fax: + 351-234 372 587. E-mail: [email protected]. † These authors contributed equally to this work. ‡ Philipps University of Marburg. § Centre for Cell Biology & Department of Biology, University of Aveiro. | Department of Chemistry, University of Aveiro. ⊥ Max-Planck-Institute of Molecular Biomedicine. # Max-Planck-Institute for Molecular Physiology. 10.1021/pr100052q

 2010 American Chemical Society

of the acinar cell, where they are stored until a neuronal or hormonal stimulus (e.g., acetylcholine and cholecystokinin) releases intracellular calcium stores, thus triggering the fusion of ZGs with the plasma membrane and with neighboring granules. Fusion results in exocytosis of digestive enzymes into the apical lumen and the pancreatic duct system.8,9 Under normal conditions, the digestive enzymes are finally activated by enterokinase via proteolytic cleavage of trypsinogen in the small intestine. Once in the intestinal tract, granule proteins are also supposed to fulfill regulatory and protective functions, e. g. in host defense.10,11 Although the ZG has long been a model for the understanding of secretory granule biogenesis and functions, the molecular mechanisms required for ZG formation at the TGN, for packaging and sorting of cargo proteins, as well as for granule fusion and exocytosis are still poorly defined.6,9,12 According to recent models, part of the molecular machinery required for digestive enzyme sorting, granule trafficking and exocytosis is supposed to be associated with the granule membrane (ZGM). In addition to basic research interests, ZG play important roles in pancreatic injury and disease.13 Thus, there is currently great interest in the identification and characterization of ZG and ZGM components Journal of Proteome Research 2010, 9, 4927–4939 4927 Published on Web 08/14/2010

research articles by conducting antibody screens, raft analyses and proteomic studies.14-19 In this study, we have focused on the identification and characterization of a “basic” group of peripheral granule membrane proteins from rat exocrine pancreas. A 2D-gel approach combined with tandem mass spectrometry led to the identification of membrane-associated proteins including classical ZG content proteins, lipid binding proteins as well as previously unknown low abundant proteins such as chymase (RMCP-1), a serine protease previously described in mast cell granules, and peptidyl-prolyl cis-trans isomerase B (PpiB), a known ER-resident enzyme. We performed a series of validation experiments and for the first time demonstrated that chymase and PpiB are bona fide peripheral membrane proteins of ZG. As we identified several peripheral ZGM proteins with proteoglycan-binding properties, our findings may help to unveil the possible role of proteoglycans in the sorting and packaging of zymogens.

Experimental Section Antibodies and cDNA. Antibodies were used as follows: rabbit polyclonal antibodies to carboxypeptidase A (Rockland Immunochemicals, Gilbertsville, PA), cyclophilin B (PpiB) (Abcam, Cambridge, U.K.), RNase A (Sigma-Aldrich, St. Louis, MO), PDI (kindly provided by H. D. So¨ling, MPI for Biophysiological Chemistry, Go¨ttingen, Germany), Calnexin (Stressgen, Ann Arbor, MI) and ZG16p,20 mouse monoclonal antibodies directed to GP2 (kindly provided by A. Lowe, Stanford University School of Medicine, Palo Alto, CA), Myc epitope 9E10 (Santa Cruz Biotechnology, USA), p115 (BD Transduction Laboratories), BiP (BD Diagnostics, NJ), R-amylase and R-tubulin DM1R (Sigma, St. Louis, MO). A sheep polyclonal antibody to RMCP1/chymase was kindly provided by H. R. Miller, University of Edinburgh, U.K.; goat polyclonal antibody to RMCP-1/chymase was kindly provided by L. B. Schwartz, Commonwealth University, Richmond, VA; goat anti-mouse-tryptase β-1 was obtained from R&D systems (Minneapolis, MN). A polyclonal antibody to recombinant carboxyl ester lipase from rat (CEL) was raised in chicken (Eurogentec, Belgium) and isolated from egg yolk as described.21 Species-specific IgG antibodies conjugated to HRP were obtained from BioRad (Richmond, CA), Molecular Probes Europe (Leiden, The Netherlands) and SigmaAldrich (St. Louis, MO). Species-specific anti-IgG antibodies conjugated to the fluorophores TRITC and Alexa 488 were obtained from Jackson ImmunoResearch (West Grove, PA), and Invitrogen (Carlsbad, CA). Concanavalin A conjugated with Tetramethylrhodamine (TRITC) was obtained from Molecular Probes Europe (Leiden, The Netherlands). The following primer sequences were used to amplify the coding sequence of rat chymase (NM_017145.1) from a rat pancreas cDNA library (Clontech, Heidelberg, Germany): 5′TTG GAT CCA TGC AGG CCC TAC TAT TCC-3′ (forward) and 5′-TTG AAT TCC TAG CTT GGA GAC TCT GAC-3′ (reverse). Using the restriction sites for BamHI and EcoRI at the ends of the PCR products (underlined), the cDNA was cloned in frame into the pcDNA3 vector (Invitrogen). Fusion of YFP to the C-terminus of chymase was achieved by insertion into pEYFPN1 vector (Clontech, Saint-Germain-en-Laye, France) using the restriction sites for EcoRI and BamHI after amplification with the primer sequences 5′-TTG AAT TCC CAT GCA GGC CCT ACT ATT CC-3′ (forward) and 5′-TTG GAT CCC CGC TTG GAG ACT CTG ACT CG-3′ (reverse). In frame insertion of all constructs was verified by sequencing (MWG). Plasmid YFP-ER was kindly 4928

Journal of Proteome Research • Vol. 9, No. 10, 2010

Borta et al. provided by R. Jacob (University of Marburg, Germany) and plasmid GFP-Sec61β was a kind gift from W. A. Prinz (National Institute of Diabetes and Digestive and Kidney Diseases, NIH, Bethesda, MD). Isolation of Zymogen Granules. ZGs were isolated from the pancreas of male Wistar rats (200-230 g) (Charles River, Sulzfeld, Germany) which were fasted overnight. Animals were handled according to the German law for the protection of animals, with the permit of the local authorities. Tissue homogenization was performed in the following buffer: 0.25 M sucrose, 5 mM 2-N-morpholino-ethanesulfonic acid (MES), pH 6.25, 0.1 mM MgSO4, 10 µM Foy 305 (Sanol Schwarz), 2.5 mM Trasylol (Bayer, Leverkusen, Germany), and 0.1 mM phenylmethylsulfonyl fluoride (PMSF).4 Pancreata were homogenized on ice using a brendle type homogenizer (Yellow line, OST20 Digital). The homogenate was centrifuged for 10 min, 4 °C at 500× g, and the resulting post nuclear supernatant was further centrifuged for 10 min, 4 °C at 2,000× g to sediment ZG. The brownish layer of mitochondria on top of the pellet was removed. The white zymogen granules were collected in homogenization buffer and centrifuged for 20 min, 4 °C at 2000× g. Tissue homogenization was repeated 2-3 times. Purified granules were resuspended in 50 mM Hepes, pH 8.0, carefully lysed by freezing and thawing and separated into zymogen granule content (ZGC) and membrane (ZGM) fractions by centrifugation at 100 000× g for 30 min in a swingingbucket rotor (Beckman SW50.1). The membranes were rinsed and resuspended in Hepes buffer and treated with 150 mM Na2CO3, pH 11.5 for 2 h on ice. Treated membranes (ZGMcarb) were recovered by centrifugation at 100 000× g for 30 min and thereby separated from peripheral membrane proteins (wash). Alternatively, purified ZG were resuspended in 50 mM Hepes, pH 8.0, 80 mM KCl, carefully lysed by freezing and thawing and centrifuged through a 0.3/1 M sucrose step gradient for 1 h, 4 °C at 200 000× g (Beckman 80 Ti rotor). ZGM were recovered at the interface and washed twice in 50 mM Hepes, pH 8.0 or in 100 mM NaHCO3, pH 8.1. After each washing step, ZGM were recovered by centrifugation for 1 h, 4 °C at 150 000× g (Beckman 80 Ti rotor). Finally, membranes were rinsed and resuspended in 50 mM Hepes, pH 8.0. A microsome-enriched fraction was obtained from a pancreas homogenate after subcellular fractionation and ultracentrifugation (100 000× g for 1 h, 4 °C; Beckman SW50.1). A postnuclear supernatant of rat tongue homogenate was obtained by centrifugation at 500× g, 4 °C for 10 min. Protein concentrations were determined using the Bradford assay. Assays were run with a recording spectrophotometer (Ultraspec 100 pro, Amersham Biosciences, Uppsala, Sweden). 1D and 2D-Gel Electrophoresis and Immunoblotting. Protein samples were separated by SDS-PAGE on 12.5% polyacrylamide mini gels, transferred to nitrocellulose (Schleicher and Schu ¨ ll, Dassel, Germany) via a semidry apparatus (Trans-Blot SD, Biorad) and analyzed by immunoblotting using horseradish peroxidase-conjugated secondary antibodies and enhanced chemiluminescence reagents (Amersham Bioscience, Arlington Heights, IL). For quantification, immunoblots were scanned and processed using “Gel Pro Analyzer” software. The first dimension of 2D-gel electrophoresis was performed in a horizontal apparatus (Ettan IPGphor, GE Healthcare, San Francisco, USA). Granule subfractions were precipitated with 20% TCA (ratio 1:1). Protein samples (300 µg) were solubilized for 30 min at 30 °C in rehydration buffer.22 The samples were then applied onto IPG strips (11 cm, pH 3-11) and isoelectric

Peripheral Membrane Proteins of Zymogen Granules focusing was conducted at 20 °C with 50 µA, for a minimum of 10 h at 50 V, 1 h at 500 V, 1 h at 1000 V and 110 min at 8000 V. The strips were afterward incubated for 15 min in equilibration buffer containing 6 M urea, 75 mM Tris-HCL pH 8.8, 34.5% Glycerol (87%), 2% SDS, 0.002% bromophenol blue and 1.5 mM DTT and then applied on top of a SDS-PAGE gel (14 cm by 14 cm, 15%). Proteins were separated according to molecular weight in a Hoefer 600 SE RUBY chamber (GE Healthcare). Silver staining of gels was performed according to.23 For tryptic digestion the SDS-PAGE gels were stained using colloidal Coomassie blue and silver staining. Tryptic Digestion and Mass Spectrometry. The protein spots were excised from the gels and washed three times with 25 mM ammonium bicarbonate/50% acetonitrile (ACN), one time with ACN and dried. Twenty-five µL of 10 µg/mL sequence grade modified porcine trypsin (Promega, Madison, WI) in 25 mM ammonium bicarbonate were added to the dried gel pieces and the samples were incubated overnight at 37 °C. Extraction of tryptic peptides was performed by adding 10% formic acid (FA)/50% ACN (3×) being lyophilized in a SpeedVac (Thermo Fisher Scientific, Asheville, NC). Tryptic peptides were resuspended in 13 µL of a 50% acetonitrile/0.1% formic acid solution. Aliquots of samples (0.5 µL) were spotted onto the MALDI sample target plate and mixed (1:1) with a matrix consisting of a saturated solution of R-cyano-4-hydroxycinnamic acid (5 mg/mL) prepared in 50% acetonitrile/0.1% formic acid. Peptide mass spectra were obtained on a MALDI-TOF/TOF mass spectrometer (4800 Proteomics Analyzer, Applied Biosystems, Europe) in the positive ion reflector mode. Spectra were obtained in the mass range between 800 and 4500 Da with ca. 1500 laser shots. For each sample spot, a data dependent acquisition method was created to select the six most intense peaks, excluding those from the matrix, trypsin autolysis, or acrylamide peaks, for subsequent MS/MS data acquisition. Trypsin autolysis peaks were used for internal calibration of the mass spectra, allowing a routine mass accuracy of better than 20 ppm. Spectra were processed and analyzed by the Global Protein Server Workstation (Applied Biosystems, Foster City, CA), which uses internal Mascot (Matrix Science Inc., Boston, MA) software for searching the peptide mass fingerprints and MS/MS data. Searches were performed against the NCBI nonredundant protein database. Cell Culture and Transfection Experiments. AR42J cells (ATCC no.CRL-1492) were cultured as described.19 Briefly, cells were maintained in Dulbecco’s modified Eagle’s medium (DMEM) containing 10% fetal calf serum and penicillin (100 U/mL)/streptomycin (100 µg/mL) (PAA Laboratories GmbH, Linz, Austria) at 37 °C in a 5% CO2-humidified incubator. To improve cell adherence, the culture dishes were coated with an extract of Engelbreth-Holm-Swarm tumor.24 To induce differentiation and zymogen granule formation, cells were incubated with 10 nM dexamethasone for 2-3 days. Cells were transfected with DNA constructs by electroporation.19 Briefly, cells grown to 90% confluency were harvested by trypsination, resuspended in 0.5 mL complete cell culture medium and transferred to a sterile 0.4 cm gap electroporation cuvette containing 10 µg of DNA. Electroporation was performed with an ECM 630 Electro Cell Manipulator (BTX Harvard Apparatus, Holliston, MA) at 250 V, 1500 µF and 125 Ω. After electroporation, cells were immediately resuspended in complete medium, plated on coverslips, and dexamethasone was added 24 h later. To knock down the expression of rat PpiB (Acc. No. NM_022536) by RNA interference, predesigned 21-nucleotide

research articles small interfering RNAs (siRNA) (sense strands: 5′-GCAAGUUCCAUCGUGUCAUtt-3′; 5′-GGAUGUGAUCAUUGUAGACtt-3′; 5′CGAUAAGAAGAAGGGACCUtt-3′) (Ambion, Austin, TX) were transfected into the cells using electroporation. As a control cells were transfected with siRNA duplexes targeting luciferase (Dharmacon, Lafayette, CO). Dexamethasone (10 nM) was added 24 h after transfection and cells were assayed for silencing and organelle morphology after 2-3 days. Immunofluorescence and Microscopy. Cryostat sections of rat pancreas and AR42J cells grown on glass coverslips were fixed with 4% paraformaldehyde in PBS, pH 7.4. The samples were permeabilized with 0.2% Triton X-100, blocked with 1% BSA and incubated with primary and secondary antibodies as described.25 Samples were examined using an Olympus BX-61 microscope (Olympus Optical Co. GmbH, Hamburg, Germany) equipped with the appropriate filter combinations and a 100× objective (Olympus Plan-Neofluar; numerical aperture, 1.35). Fluorescence images were acquired with an F-view II CCD camera (Soft Imaging System GmbH, Mu ¨ nster, Germany) driven by Soft imaging software. Confocal images were acquired on a Zeiss LSM 510 confocal microscope (Carl Zeiss, Oberkochen, Germany) using a Plan-Apochromat 63× or 100 × /1.4 oil objective. Images were processed and quantified using LSM510 software (Carl Zeiss MicroImaging, Inc.). Background noise was minimal when optimal gain/offset settings for the detectors were used. Digital images were optimized for contrast and brightness using Adobe Photoshop (Adobe Systems, San Jose, CA). For quantification of granule formation and morphology in AR42J cells, usually 3-5 slides per preparation were analyzed, and 3-5 independent experiments were performed. Immunoelectron Microscopy. Tissue from rat pancreas and rat tongue was fixed in 0.1 M cacodylate buffer, pH 7.35 containing 2% paraformaldehyde and 0.1% glutaraldehyde (Serva, Heidelberg, Germany). The samples were dehydrated in a graded series of alcohol, embedded in Lowicryl K4M (Polysciences Ltd., Eppenheim, Germany) and polymerized at -20 °C and UV light (360 nm) for 48-72 h. Thin sections (70 nm) were incubated with polyclonal antibodies directed to chymase (1:200-500) and visualized using a 10 nm Protein A-Gold solution (J. Slot, University of Utrecht, The Netherlands) at a dilution of 1:60 or 1:70, both in 0.5% BSA in PBS. Sections were stained with uranyl acetate/lead citrate and analyzed using a Zeiss EM 109 electron microscope (Carl Zeiss, Oberkochen, Germany). The labeling density (gold particles/ µm2) on zymogen granules and control regions was determined manually on images with the same magnification. For cryosectioning and PpiB-labeling, rat exocrine pancreas got initially fixed in 2% paraformaldehyde, 0.1% glutaraldehyde, 2% sucrose in 0.1 M phosphate buffer, pH 7.4. Samples were washed extensively in 0.1 M phosphate buffer, pH 7.4 and infiltrated in 2.3 M sucrose for cryoprotection. Ultrathin-cryosectioning and immunogold-labeling was performed as described.26 For HM20 embedding, fixed samples were dehydrated by the progressive lowering of temperature (PLT) method,27 polymerized at -35 °C with subsequent immunolabeling of ultrathin sections at room temperature. Samples were analyzed at 80 kV on a FEI-Tecnai 12 electron microscope (FEI, Eindhoven, Netherlands) or at 120 kV on a JEM-1400 (JEOL Germany, Eching, Germany) and selected areas were documented with imaging plates (Ditabis, Pforzheim, Germany) or via CCD camera (F-214, TVIPS, Gauting, Germany). Isolation of RNA, Reverse Transcription, and PCR. RT-PCR was performed to amplify parts of the coding sequence of rat Journal of Proteome Research • Vol. 9, No. 10, 2010 4929

research articles

Borta et al.

Figure 1. Separation of rat ZG content and ZGM wash subfractions by two-dimensional IEF/SDS-PAGE. The whole complement of ZGC (A) and the supernatant fraction (wash) (B) of carbonate-treated ZGM were subjected to 2D-PAGE followed by Coomassie staining. For IEF, 300 µg of protein were separated on 11 cm IPG strips (pH 3-11NL) and on 15% polyacrylamide gels in the 2nd dimension. Note the differences in the spot pattern of ZGC (A) and the wash fraction (B). The boxed areas in (B, C) highlight basic protein spots (chymase, PpiB, RNase A) which have been selected for further analysis and are verified by immunoblotting (C) using specific antibodies to rat mast cell chymase, PpiB, and RNase A.

chymase (NM_017145.1), tryptase β1 (NM_019322), amylase (NM_031502), PpiB (BC061971) and GAPDH (BC087069) from mRNA isolated from pancreas of fasted or Foy-treated rats, and from tongue tissue. Foy 305 (Sanol Schwarz, Monheim, Germany) is a low-molecular-weight serine protease inhibitor.28 Total RNA was isolated using the RNeasy Protect Mini Kit (Qiagen, Hilden, Germany), and reverse transcribed using oligo(dT) primer and M-MuLV reverse transcriptase (Stratagene Amsterdam, The Netherlands) at 42 °C. PCR was performed with 100 ng of template using the following primer pairs: 5′-GTT TCT TGT GAC CCG CCA ATT-3′ (forward chymase); 5′-TTA ATC CAG GGC ACA TAT GGG-3′ (reverse chymase); 5′-GAA TAA GGC TGA CCC CAA CA-3′ (forward tryptase β1); 5′-CTT GGG GAC ATA GCG GTA GA-3′ (reverse tryptase β1); 5′- GCC TTC TGG ATC TTG CAC TC-3′ (forward amylase); 5′-AGT GCT TGA CAA AGC CCA GT-3′ (reverse amylase); 5′-TCC GTT GTC TTC CTT TTG CT- 3′ (forward PpiB); 5′-GTT CTC CAC CTT CCG TAC CA-3′ (reverse PpiB); 5′ACG ACC CCT TCA TTG ACC-3′ (forward GAPDH); 5′-CCA GTG AGC TTC CCG TTC AGC-3′ (reverse GAPDH). GAPDH was used as a loading control. Samples were analyzed by 1% agarose gel electrophoresis.

Results Identification of Peripheral Membrane Proteins of Zymogen Granules. To understand the biogenesis and function of pancreatic zymogen granules as well as their role in disease a profound analysis of their membrane and content components is required. We have special interest in the analysis of membrane-associated ZG proteins, andstogether with our coworkersshave contributed to the further understanding of the composition and architecture of the ZGM.16,17,19,29-32 In this study, we have applied a suborganellar proteomics approach to identify peripheral membrane-associated ZG proteins (Figure 1, Supplementary Figure 1, Supporting Information). Zymogen granules were isolated from rat pancreas according to a standardized protocol.4,16,19 The purity of the isolated ZG fractions (g90%) was controlled by electron microscopy and immunoblotting as shown recently.19 Intact granules were gently lysed and further separated in a membrane (ZGM) and content fraction (ZGC). The isolated membranes (ZGM) were treated with carbonate to liberate membrane-associated proteins and separated in a pellet (ZGMcarb) and supernatant fraction (wash) (Figures 2, 3). The distribution of distinct 4930

Journal of Proteome Research • Vol. 9, No. 10, 2010

Figure 2. (A) Proteins identified from the “basic group” of a ZGM carbonate wash fraction. Functional annotation and organelle assignments were made using the UniprotKB database, additional annotation was incorporated from literature search. Extra information supporting the identification of the potential ZGM proteins is summarized in Supplementary Table 1 (Supporting Information). (B) Diagram illustrating the intracellular distribution of the identified proteins of the wash fraction. On the basis of published data, annotations in databases or predictions based on similarity to related proteins, the identified proteins are grouped in a pie chart according to their subcellular distribution and function.

granule marker proteins (e.g., amylase, GP2, ZG16p) among the granule subfractions was controlled by immunoblotting (Figure 3). Usually, a contamination with ER or mitochondrial marker proteins was not detected in the ZG subfractions (see Figure 3 and ref 19). To characterize the ZG peripheral membraneassociated proteins, 2D gel maps were generated for the supernatant fraction (wash) of Na2CO3-washed ZGM. For comparison typical 2D gel patterns of the wash and the content subfractions (ZGC) separated under equal conditions are shown

Peripheral Membrane Proteins of Zymogen Granules

research articles

Figure 3. Chymase and PpiB represent peripheral membrane proteins of rat ZG. (A) Lysed granules were separated into a content (ZGC) and membrane fraction (ZGM). In addition, isolated membranes were treated with Na2CO3 at pH 11.5 and separated into pellet (ZGMcarb) and supernatant (Wash) fractions. Equal amounts of protein (20 µg) were run on 12.5% acrylamide gels, blotted onto nitrocellulose membranes and incubated with antibodies to amylase (ZGC marker protein), GP2 (ZGM marker protein), ZG16p (peripheral ZGM marker protein), chymase, tryptase β1 (mast cell control), carboxyl ester lipase (CEL), RNase A, BiP, PDI and calnexin (ER control) and PpiB. (B) Densitometric quantification of immunoblots shown in (A). The distribution of the labeling density to ZGC and ZGM (% labeling density of total ZGC + ZGM) as well as the distribution of the labeling density to ZGMcarb and wash (% labeling density of total ZGM) is depicted (see also Supplementary Table 2, Supporting Information). Note that the overall distribution of chymase and PpiB resembles that of ZG16p, a peripheral ZGM marker protein. A lysate from rat tongue and a microsome-enriched fraction (Micros) served as controls for the detection of mast cell proteins and ER resident proteins, respectively. Antibodies to BiP, PDI, and PpiB have been incubated successively on the same immunoblot.

in Figure 1. Note that the representative gels exhibit unique spot patterns for the two different subfractions. In a representative gel for the wash fraction about 104 spots were reproducibly visualized, which represent an acidic and more basic group of ZG protein spots (Supplementary Figure 1, Supporting Information). The identification of the “acidic” group, which appears to contain several proteins from the cytosolic side of the ZG membrane is currently under investigation. In the present study, we focused on the analysis of the spots of the “basic” group (pH range 6.2-11), which appears to be composed mainly of luminal ZG proteins. Two previously identified novel peripheral ZGM proteins, ZG16p and syncollin, are as well proteins with a pI of 9.17 and 8.62, respectively.20,29,32,33

In this region, 46 spots were visualized among which 44 were identified by tandem mass spectrometry (Supplementary Figure 1, Supplementary Table 1, Supporting Information). These spots corresponded to 16 unique proteins (Figure 2A), which were categorized in 6 groups based on their known subcellular localization: ZGC (n ) 6; 38%) and ZGM (n ) 6; 38%) proteins, mast cell proteins (n ) 1; 6%), ER resident proteins (n ) 1; 6%), and proteins with other localizations (n ) 2; 12%) (Figure 2B). Based on their predicted biological functions we identified 12 enzymes including 9 digestive enzymes (usually attributed to the ZGC), 2 matrix (ZG16p, syncollin) and 3 slightly acidic glycoproteins (CEL, pancreatic lipase related protein 1 and 2). Based on literature and the Journal of Proteome Research • Vol. 9, No. 10, 2010 4931

research articles Protein Knowledgebase (UniProtKB), all of the identified proteins are supposed to be soluble or peripheral membrane proteins, and no transmembrane proteins or membraneanchored proteins have been identified in the wash fraction. Furthermore, except for two cytosolic proteins (vinculin (fragment), an actin-binding protein, and ubiquitin carrier protein involved in ubiquitination/quality control) all identified proteins are supposed to be components of the secretory pathway. This further confirms the applicability of the carbonate and gel-based proteomic approach. In addition, some classical content proteins such as amylase, elastase, colipase and triacylglycerol lipase were identified in the wash subfraction (Figure 2, Supplementary Table 1, Supporting Information). This can be due to a cross-contamination of the subfractions or due to the fact that the interactions established between the proteins from the different subfractions are not completely disrupted in the separation procedure. The identification of some abundant theoretical acidic proteins within the “basic” group (e.g., CEL, lipase related protein 1) is likely to be a result of protein degradation and/or deglycosylation. The corresponding spots are generally weakly stained, but the intact proteins with the theoretical MW have also been identified (Figure 2, Supplementary Table 1, Supplementary Figure 1, Supporting Information). To validate our suborganellar proteomic data, we selected two proteins for further characterization. Chymase (RMCP-1) and peptidyl-prolyl cis-trans isomerase B (PpiB; cyclophilin B) are likely to be previously unknown genuine membraneassociated ZG proteins. Chymase was initially identified as a proteoglycan-associated serine protease in granules of mast cells.34-36 In support of our findings, chymase was also identified in ZGM fractions in a recent proteomic study combining two-dimensional LC and tandem mass spectrometry.37 PpiB has been reported to be an ER-resident enzyme with proteoglycan-binding properties.38,39 Furthermore, we selected RNase A originally known as a typical content protein of ZG, but prominently present in our ZGM fraction. Based on the spot size, chymase and PpiB appear to be low abundant components of ZGM (Supplementary Figure 1, Supporting Information). To confirm their presence in our 2D gels and thus, their correct identification, immunoblotting was performed with the 2D-gels of the wash fraction. As shown in Figure 1, the corresponding chymase and PpiB spots identified by tandem mass spectrometry were recognized by specific antibodies to chymase and PpiB after immunoblotting. To rule out that chymase and PpiB represent contaminating proteins,40 we performed a series of validation experiments (see below). Chymase and PpiB are Novel Peripheral ZGM Proteins. To monitor the distribution of chymase and PpiB among granule subfractions, equal amounts of protein from the four granule subfractions (ZGC, ZGM, ZGMcarb and wash) were separated on 12.5% acrylamide gels and immunoblotted using antibodies directed to chymase and PpiB (Figure 3 and Supplementary Table 2). Furthermore, antibodies to the granule marker proteins amylase, CEL, RNase A, GP2, and ZG16p were used to assess proper granule subfractionation. As shown in Figure 3, immunoreactivity for amylase was mainly found in the ZGC fraction, whereas GP2, a major GPI-anchored glycoprotein of ZG, was predominantly present in the ZGM fraction. The secretory lectin ZG16p, a peripheral ZGM protein,20 was concentrated on isolated ZGM, and a major portion was liberated by carbonate treatment (Figure 3 and Supplementary Table 2). Interestingly, chymase and PpiB showed a similar 4932

Journal of Proteome Research • Vol. 9, No. 10, 2010

Borta et al. distribution pattern than ZG16p. Both enzymes were associated with isolated ZGM, but barely detectable in the ZGC fraction. Upon carbonate-treatment the majority of chymase and PpiB was removed from the ZGM and found in the wash fraction indicating that both proteins are peripheral components of the ZGM. As chymase is a chymotrypsin-like protein, we also tested for cross-reactivity of the anti-chymase antibody with chymotrypsin, but no binding was observed (not shown). Furthermore, we investigated the distribution of CEL and RNase A within the granule subfractions. For this purpose, a CEL-specific antibody was generated in chicken and isolated from chicken egg. This antibody has been successfully used in other studies.41 Interestingly, the labeling density for CEL was highest in the ZGM fraction (Figure 3). However, only a small portion can be removed from the ZGM by carbonate treatment indicating that the enzyme is more tightly associated with the ZGM. This might be due to its affinity to lipid surfaces.42-45 In the case of RNase A, most of the labeling density is found in the ZGC fraction, but a surprisingly high amount (about 40% of the labeling density, Figure 3) is associated with the ZGM and can be liberated by carbonate treatment. Note that the ZGM associated labeling density for RNase A is much higher than that of membrane-associated amylase (13%, Figure 3). To provide a more quantitative approach, we have analyzed dilution series of content and membrane fractions for PpiB, chymase, RNase A, CEL and amylase by immunoblotting. Intensities of the corresponding signals were determined by densitometry, and the ratios of the proteins in the ZGC to ZGM fractions were calculated using these titration curves (Supplementary Figure 2, Supporting Information). A ZGC to ZGM concentration ratio of 1:35 was calculated for PpiB, and 1:46 for chymase indicating an enrichment in the membrane fraction. An enrichment in the ZGM fraction was also calculated for CEL (1:23) and, more moderate, for RNase A (1:10), whereas amylase was highly enriched in the content fraction (about 80: 1). Taking into account the distribution of total ZG proteins (approximately 95% content proteins and 5% membranebound proteins), this implicates that about 70% of total chymase and 43% of total PpiB are associated with the ZGM fraction. Furthermore, 54% of total CEL and 33% of total RNase A were found in the ZGM fraction. In contrast, only 0.06% of total amylase is recovered in the membrane fraction, and 99.94% in the content fraction. The values for the prominently membrane-associated proteins are presumably even higher, as the proteins are partially liberated from the membranes during the isolation procedure. To exclude that not all of the content is solubilized and that aggregated zymogens might sediment with the membranes, we have isolated ZGs in a higher salt buffer and recovered ZGM without pelleting in a 0.3 M/1 M sucrose gradient. The ZGM were as well more extensively washed either in 50 mM Hepes, pH 8.0 or in 100 mM NaHCO3, pH 8.1 (Supplementary Figure 3, Supporting Information). Whereas amylase was more efficiently removed from the ZGM fractions after extensive washing, the distribution of CEL, chymase and PpiB was not grossly altered by the different isolation procedures. The minor amount of amylase found in the ZGM fractions is as well an indicator for complete ZG lysis and solubilization of the granule content (at least of amylase-containing complexes). Furthermore, membrane-association of some content proteins was also observed by others after more stringent washing/purification conditions and gradient centrifugation.14,18

Peripheral Membrane Proteins of Zymogen Granules

Figure 4. Expression of chymase and PpiB in rat pancreas is regulated by feeding behavior. Total RNA was isolated from pancreas of fasted (1) or Foy 305-treated rats (2), and from tongue tissue (t). Semiquantitative RT-PCR was performed to amplify parts of the coding regions of chymase, tryptase β1, PpiB, amylase, and GAPDH (loading control). Tongue tissue (t) was used as a positive control for the detection of mRNA coding for mast cell proteins. Samples were analyzed by 1% agarose gel electrophoresis. Note the increase of the mRNA levels for amylase, chymase, and PpiB after stimulation of acinar protein synthesis by application of the synthetic trypsin inhibitor Foy 305.

To exclude a contamination of the ZG subfractions with granules from mast cells (e.g., from pancreatic tissue), an antibody to tryptase β1, a prominent mast cell marker, was applied. A lysate of rat tongue, which is rich in mast cells containing RMCP-1, was used as a positive control. In contrast to chymase, tryptase β1 was absent from the ZG subfractions. However, both tryptase β1 and chymase were detected in lysates of rat tongue. To check for contamination of the ZG subfractions with ER or ER-resident proteins, specific antibodies directed to the ER marker proteins BiP (GRP78, 78 kDa glucose regulated protein), PDI (protein disulfide isomerase) and calnexin (an abundant ER transmembrane protein) were applied. PpiB, BiP, PDI and calnexin were all detected in a microsome-enriched fraction which was used as a positive control. In contrast to PpiB, which was prominently labeled in the ZGM and wash fractions, BiP and calnexin were below the detection level in all ZG subfractions. Interestingly, a small amount of PDI was detected in the ZGC fraction, but PDI was not detectable in all other ZG subfractions. This is consistent with morphological studies where small amounts of PDI have

research articles been detected in the ZG content. The presence of PDI is supposed to be due to occasional escape from the ER.46,47 PpiB was as well identified in ZGM fractions in a recent proteomics report using 2D-GEP and LC-MS/MS.14 As PpiB is selectively enriched in the ZG or ZGM fraction, but calnexin, BiP and PDI are not (or are barely detectable), a contamination with ER appears unlikely. In addition, other major microsomal proteins were not identified in our 2D gel analyses. These findings support the notion that PpiB is a bona fide ZG protein and not the result of a contamination (see also Figure 6). Expression of Chymase and PpiB in Rat Pancreas Is Regulated by Feeding Behavior. To investigate whether chymase and PpiB expression levels are modulated by feeding behavior, their mRNA expression levels were analyzed by RTPCR in the pancreas of fasted and Foy 305-treated rats (Figure 4). The administration of Foy 305, a low-molecular-weight serine protease inhibitor, inhibits trypsin activity in the small intestine, thus inducing the expression of digestive enzymes (e.g., amylase) (Figure 4), whereas fasting has the opposite effect.28 Interestingly, in addition to amylase expression, chymase and PpiB mRNA levels were increased in Foy-treated rats demonstrating their dependence on feeding behavior (Figure 4). GAPDH served as a loading control, and mRNA isolated from rat tongue served as a positive control for chymase expression. Tryptase β1 expression was monitored to rule out mast cell contamination. Chymase and PpiB Localize to Zymogen Granules in Rat Pancreas. To verify the localization of endogenous chymase to ZG in rat pancreatic tissue, immunohistological and ultrastructural studies were performed (Figures 5 and 6). For immunohistology, frozen sections of rat pancreas were incubated with specific antibodies directed to chymase and to carboxypeptidase A, a prominent granule marker protein (Figure 5A, B). Chymase was observed to colocalize with carboxypeptidase A over the granule region/zymogen granules surrounding the acinar lumens (Figure 5C). As a negative control, sections were incubated with a preabsorbed antibody and with anti- carboxypeptidase A (Figure 5D-F). Next, we performed immunoelectron microscopy on K4M-embedded tissues of rat tongue and rat pancreas (Figure 6A, B). Ultrathin sections were stained with antibodies directed to chymase and afterward with protein A-gold. Rat tongue, which is rich in mast cells and RMCP-1/chymase-containing granules, served as a

Figure 5. Immunofluorescence microscopy of pancreatic sections from rat pancreas. Cryosections of rat pancreas were incubated with antibodies specific for chymase (A) and carboxypeptidase (CBP) (B). (C) Overlay (Merge) of (A, B). (D, E) Negative control using preadsorbed chymase and anticarboxypeptidase antibodies. (F) Overlay (Merge) of (D, E). Asterisks in (F) mark the nuclei of acinar cells. Bars, 10 µm. Journal of Proteome Research • Vol. 9, No. 10, 2010 4933

research articles

Figure 6. Chymase and PpiB localize to zymogen granules in acinar cells of the exocrine pancreas. Tissue from rat tongue (A) and pancreas (B) was processed for immunogold electron microscopy, incubated with polyclonal antichymase antibody and visualized using 10 nm protein A-gold. Ultrathin cryosection of rat pancreas (C) and section of HM20 embedded pancreas (D), labeled for PpiB, and detected with 15 nm protein A-gold. Note the dual labeling on the ER cisternae (arrows) and on zymogen granules. Arrowheads in (B, C, D) indicate gold particles on zymogen granules (ZG). Gr, granule of a mast cell; N, nucleus. Bars, 1 µm.

positive control, and a prominent, uniform labeling with gold particles was observed on cytoplasmic granules (Figure 6A). Specific gold-labeling of ZGs in pancreatic acinar cells was also observed. The staining was less prominent, but specific for ZGs (Figure 6B) (labeling density on ZG: 1.16 ( 0.1 gold particles/ µm2, labeling density on control areas outside ZG: 0.18 ( 0.8 gold particles/µm2). This might be due to the low abundance of chymase in ZGs and the regulation of its expression level. Furthermore, the accessibility of the protein might be reduced in highly condensed ZGs. The localization of PpiB was determined on ultrathin cryosections and on HM20 embedded samples of rat pancreas, which were stained with antibodies directed to PpiB and afterward with protein A-gold (Figure 6C, D). With both 4934

Journal of Proteome Research • Vol. 9, No. 10, 2010

Borta et al. methods specific labeling (with virtually no unspecific background staining) was found on the ER cisternae and on zymogen granules, confirming the dual localization of PpiB in the exocrine pancreas. The ZG labeling density may be lower than expected due to dense cargo packing and, thus, steric hindrance for antibodies. Chymase is Sorted to Secretory Granules in Pancreatic AR42J Cells. To examine if exogenously expressed chymase is sorted to zymogen granules, we used pancreatic AR42J cells as a model. Treatment of AR42J cells with glucocorticoids (dexamethasone) initiates their differentiation into acinar-like cells and the de novo formation of electron-opaque secretory granules, which contain the major pancreatic zymogens.48 Granule formation was usually induced 24 h after plating by the addition of 10 nM dexamethasone, and cells were processed after 2-3 days. RT-PCR with mRNA from stimulated and control AR42J cells revealed that chymase was not endogenously expressed under our experimental conditions (Figure 7L). Furthermore, it is known that AR42J cells do not express all ZG proteins found in rat pancreas.49 Next AR42J cells were transfected with a generated chymase-YFP fusion protein (Figure 7A-C) or a YFP containing an ER-targeting signal (YFPER) (Figure 7D-I) and stimulated for granule formation. After 2-3 days, cells were processed for indirect immunofluorescence using antibodies to carboxypeptidase A (Figure 7B, E). Dexamethasone-treatment resulted in the formation of numerous characteristic granules positive for carboxypeptidase A. Confocal microscopy revealed that many of these granules showed a colocalization with chymase-YFP (Figure 7A, C), whereas YFP-ER localized mainly to the Golgi complex in AR42J cells (Figure 7G-I), and not to secretory granules (Figure 7D, F). Similar observations were made with an ER-mRubyconstruct50 (not shown). Furthermore, VSVG-GFP and YFP-GPI fusion proteins were not observed to be sorted to secretory granules in AR42J cells.19 Quantification by pixel-by-pixel analysis from confocal images revealed a colocalization coefficient of 77 ( 18% (n ) 6, 15 z-planes from each cell) for chymase-YFP and carboxypeptidase A (Figure 7K). These findings demonstrate that rat mast cell chymase enters the secretory pathway and is properly sorted to zymogen granules in AR42J cells, an acinar model system. Partial Localization of PpiB to Secretory Granules in Pancreatic AR42J Cells. To characterize the subcellular localization of PpiB in pancreatic AR42J cells, we conducted immunofluorescence microscopy using antibodies directed to PpiB and chymotrypsin (a granule marker enzyme). Furthermore, a GFP-Sec61β fusion protein, a marker for the ER, was expressed (Figure 8A-F). In cells stimulated for granule formation, PpiB showed an ER and granule-like distribution pattern (Figure 8A, D). A pixel-by-pixel analysis of confocal serial images revealed a colocalization coefficient of 72 ( 20% (n ) 12, 15 z-planes from each cell) for PpiB with the ER marker GFP-Sec61β (Figure 8H, I). A similar colocalization coefficient (86 ( 12%) was obtained for PpiB with the ER marker concanavalin A (not shown). However, a partial colocalization for PpiB with the granule marker chymotrypsin was also observed (colocalization coefficient of 24 ( 19%) in cells with pronounced granule formation (Figure 8G, I). Our findings show that PpiB is mainly an ER-resident protein in AR42J cells, but is partially directed to ZG in this acinar model system. Next we examined whether PpiB protein levels are upregulated after stimulation of AR42J cells with dexamethasone. Cell lysates of controls and stimulated AR42J cells were prepared

research articles

Peripheral Membrane Proteins of Zymogen Granules

Figure 7. Chymase is targeted to zymogen granules in pancreatic acinar AR42J cells. AR42J cells were transfected with a chymase-YFP construct (A), a p-EYFP-ER construct (D, G) or a p-EYFP-N1 vector (J), stimulated for granule formation and processed for indirect immunofluorescence 72 h after transfection using antibodies directed to carboxypeptidase A (CBP) (B, E) and the Golgi marker p115 (H). Overlays (Merge) of the confocal images (A-B; D-E; G-H) are shown in (C, F, I). Arrows point to some granules showing colocalization of chymase-YFP and CBP. EYFP-ER localizes mainly to the Golgi complex and is not targeted to granules in AR42J cells. Cytosolic YFP (J) is not targeted to granules either. (K) Quantitative correlation analysis. Image analysis was carried out with the Zeiss LSM 510 4.0 software (Carl Zeiss MicroImaging, Inc.). The correlation plot describes the pixel colocalization depending on their intensity in the Alexa 488 (for chymase-YFP) and TRITC (for CBP) channels with region 3 displaying colocalizing pixels, whereas regions 1 and 2 contain the noncolocalizing pixels for each label, respectively. The signals below the background (indicated by the axes) in each picture were not included in the quantification of the colocalization coefficients. An average value of 77% of colocalization has been obtained measured from 6 cells (15 z-planes from each cell) with a standard deviation of 18%. (L) RT-PCR of mRNA isolated from AR42J cells (treated with 10 nM dexamethasone) (a) and from rat tongue (t) (positive control for chymase expression). GAPDH was used as a loading control. Note that chymase is not endogenously expressed in pancreatic acinar AR42J cells. Bars, 10 µm.

after 3 days of stimulation. Equal amounts of protein were separated by 12.5% SDS-PAGE and immunoblotted using antibodies to amylase, PpiB, and R-tubulin (loading control) (Figure 8J) Whereas the protein levels of amylase were increased after stimulation with dexamethasone, the protein levels of PpiB remained almost unchanged in controls and stimulated cells. Similar results were obtained for PpiB mRNA levels (not shown). This is in contrast to the results obtained for the expression of PpiB in pancreatic tissue, and reflects differences in the model systems used. Furthermore, silencing of PpiB in AR42J cells did not result in any obvious alterations regarding granule morphology, number or distribution when compared to controls (Supplementary Figure 4, Supporting Information). The localization of some granule marker proteins tested

(carboxypeptidase A, amylase, chymotrypsin) was not altered (e.g., there was no accumulation in the ER or Golgi complex). Thus, PpiB does not appear to be an essential protein for granule formation and sorting, at least in AR42J cells.

Discussion The ZG membrane has been the focus of recent proteomics studies,14,18,37 as their components are supposed to contributesat least partiallysto the sorting and packaging of digestive enzymes into ZG, their apical transport, membrane fusion and regulated, Ca2+-dependent secretion as well as to pancreatic and intestinal disorders. Here we report on the identification of peripheral, membrane-associated proteins of the ZGM by a Journal of Proteome Research • Vol. 9, No. 10, 2010 4935

research articles

Borta et al.

Figure 8. PpiB partially localizes to zymogen granules in pancreatic acinar AR42J cells. Dexamethasone-treated AR42J cells were processed for immunofluorescence and stained with antibodies to PpiB (A, D) and chymotrypsinogen (Chym) (B). For ER colocalization studies AR42J cells were transfected with GFP-Sec61β (E). Overlays (Merge) of the confocal images are shown on the right (C, F). Arrowheads highlight some regions of colocalization. (G, H, I) Pixel-by-pixel analysis of colocalization (correlation plot) for PpiB and chymotrypsinogen (G, I) and for PpiB and GFP-Sec61β (H, I). (I) Quantitative evaluation of the colocalization coefficient for PpiB (p < 0.001, n ) 12 cells with 15 z-planes each). (J) Immunoblots of cell lysates prepared from control (- dexamethasone) or stimulated (+ dexamethasone) AR42J cells using anti-PpiB, anti-amylase (Amyl) and anti-tubulin (Tub) antibodies. Equal amounts (20 µg) of protein were loaded onto the gels. N, nucleus. Bars, 10 µm.

suborganellar proteomic approach based on carbonate-treatment of ZGM and 2D-PAGE of the corresponding supernatant (wash) fraction. In this study, we focused on the identification and characterization of a “basic” group of membrane-associated proteins (pI range 6.2-11). As a verification of our approach, two previously reported peripheral (basic) ZG proteins, the secretory lectin ZG16p and syncollin, were identified in this study. ZG16p is supposed to interact with the ZGM via its lectin domain, whereas syncollin has been shown to interact with cholesterol and the GPI-anchored membrane glycoprotein GP2.29,30,32 Furthermore, no transmembrane proteins or classical membrane-anchored proteins were detected. On the basis of bioinformatic information and published results, almost all 4936

Journal of Proteome Research • Vol. 9, No. 10, 2010

of the identified proteins are able to enter the secretory pathway, thus confirming their interaction with the luminal side of the ZGM. The two identified cytosolic proteins vinculin (fragment), an actin-binding protein, and ubiquitin carrier protein involved in ubiquitination/quality control may be associated with the cytosolic side of the ZG. Two novel proteins, rat mast cell chymase and PpiB, an ER-resident protein, appeared to be of low abundance in the 2D-PAGE, and were thus selected for further validation of our approach. It is important to note that while increased instrument sensitivity allowed the identification of many more low abundant ZGM proteins, it also uncovered more contaminating proteins.40 PpiB, for example, was listed as a contaminant in a recent

Peripheral Membrane Proteins of Zymogen Granules 14

proteomics study of ZGM. The confirmation of identified proteins on ZGM is thus a major and important task.40 By applying RT-PCR and immunoblotting for the detection of ER and mast cell marker proteins, we were able to rule out that the presence of chymase and PpiB in our granule subfractions was the result of a contamination with ER or mast cell granules/ proteins. The distribution of chymase and PpiB in the granule subfractions resembled that of ZG16p, a typical peripheral ZGM marker, further confirming that chymase and PpiB are genuine peripheral ZGM proteins. In addition, their granule localization was for the first time confirmed by immunocytochemistry and morphological studies with pancreatic tissue and/or AR42J cells, an acinar cell model. In contrast to the biochemical data, membrane-association of PpiB and chymase appears to be less pronounced based on the ultrastructural studies. As mentioned before, this might be explained by restrictions in antibody binding. It is as well possible that PpiB and chymase interact with membrane-associated proteoglycans which extend into the content (see below). Furthermore, we cannot rigorously exclude that the biochemical experiments exaggerate the amount of membrane-association, for example, due to differential solubility of the ZG proteins in the lysis buffer. Importantly, an exogenously expressed chymase-YFP fusion protein was targeted to ZG in pancreatic AR42J cells, whereas other fusion proteins (e.g., YFP-ER, Ruby-ER, VSVG-GFP, YFPGPI) entering the secretory pathway in AR42J cells were not observed to be targeted to ZG. Furthermore, physiological studies revealed that the expression of both chymase and PpiB in the rat pancreas was dependent on feeding behavior. The chymase and PpiB mRNA levels were increased after inhibition of trypsin activity by Foy 305, a serine protease inhibitor, which after oral administration results in increased expression of digestive enzymes. In contrast, chymase and PpiB mRNA levels were low under fasting conditions, where digestive enzyme synthesis is decreased. It should be noted that chymase was also suggested to be a potential peripheral ZGM protein based on a recent iTRAQ quantitative proteomic analysis.37 Taken together, these results clearly demonstrate that both chymase and PpiB are genuine peripheral ZGM proteins and no contaminants. Chymase has been identified as a major, highly basic chymotrypsin-like serine protease of mast cells, where it is stored in regulated secretory granules. Thus, its presence in ZGs of the exocrine pancreas is not that unusual at all. With respect to function, chymase is likely to act as a serine protease after ZG secretion. RMCP-1 of mast cells is involved in inflammatory processes and tissue-remodelling.51 Its capability to cleave proteins of the extracellular matrix (e.g., fibronectin) might be a reason for its low and regulated abundance in ZGs. PpiB (Cyclophilin B) is a highly basic cyclosporin A-binding protein with peptidyl-prolyl cis-trans isomerase activity. It belongs to a group of enzymes (cyclophilins) that bind to the immunosuppressor cyclosporin A52 and catalyze protein folding of prolin-containing polypeptide chains.53-55 PpiB is sorted to the secretory pathway via an ER targeting signal56,57 and has been reported to be retained in specialized regions of the ER,58 but also as a secretory protein without ER-retention signal.56,59 On the basis of peptide coverage, the identified ZG form of PpiB appears to possess the C-terminal ER-retention signal (data not shown). The first Ppi (PpiG) within regulated secretory granules was identified in blood cells of the horseshoe crab Limulus polyphemus,60 where it is supposed to play a role in protein storage. PpiG shows highest homology to PpiB and is

research articles missing an ER-retention signal. PpiB was also identified in other exocrine secretions such as human saliva in a recent proteomic approach.61 Thus, besides its intracellular localization, PpiB can also be found in the extracellular space, where it might contribute to intercellular communication.62,63 PpiB in ZG might act as a chaperone in protein folding, and can assist in protein sorting and/or regulation of an active/inactive conformation of digestive enzymes. However, a function in the regulation of membrane channels, Ca2+-dependent processes or in immune defense after secretion are alternative or additional modes of action.58,64-66 Interestingly, several known digestive enzymes, among them many lipid-interacting enzymes, were also identified in the wash subfraction. The presence of these classical content (ZGC) proteins in a ZGM fraction is usually interpreted as a contamination. However, it should be noted that these enzymes remain partially attached to the ZGM even under more stringent purification conditions (e.g., KBr wash before carbonate treatment) or after further purification of crude ZGM by gradient centrifugation.14,18 Interestingly, many of the membraneassociated enzymes identified in this study in a membrane wash fraction have been predicted to be potential luminal ZGM proteins in a recent topology analysis of purified ZGM.37 Thus, it is very likely that the identified subset of peripheral digestive enzymes exhibits a more specific interaction with the ZGM than previously expected. Enzymes with lipid-binding properties might interact with membrane lipids to specifically associate with the ZGM. The presence of CEL at the ZGM (1:23; about 54% of total CEL) might be mediated by its capability to bind to lipid surfaces. In support of this notion, sphingolipid-binding domains of CEL are supposed to mediate its interaction with lipid rafts.67 Furthermore, CEL has been found to be associated with lipid microdomains, which have been identified in ZGM.15,17 Their importance for granule biogenesis has been demonstrated in recent studies.17,68 We also detected a prominent amount of RNase A at the ZGM (1:10; about 33% of total RNase A compared to 0.06% of total amylase). Interestingly, this interaction might be mediated by the interaction of RNase A with proteoglycans, for example, of a predicted submembranous granule matrix.16,69 RNase A hasbeenshowntointeractwithheparinandchondroitinsulfate,70,71 and on the basis of these properties, a protein A-gold-RNase A technique has been developed to detect proteoglycans in cellular compartments and tissues.72 Protein A-gold-RNase A also labeled putative proteoglycans in ZG at the membrane and in the content.73 Furthermore, an interaction with proteoglycans has been described for chymase, PpiB, and CEL.35,74-76 This interaction is supposed to influence enzyme function.35,71 The negatively charged proteoglycans in secretory granules of hematopoietic cells and mast cells are engaged in the binding of small positively charged molecules, such as histamine77-79 and proteases,80,81 and have therefore been considered to promote the efficient packaging and concentration of secretory products.82 Similarly, proteoglycans within ZGs are supposed to interact electrostatically and through specific protein-protein and carbohydrate-protein binding domains with the secretory proteins of the granule content.12,16,69 A potential role of serglycin in granule formation has, however, recently been questioned.83 It will be a great challenge for future studies to also identify and characterize high molecular mass components such as proteoglycans and glycoproteins in ZG fractions to unveil the complex architecture and putative interactions at the granule membrane and in the granule content. Journal of Proteome Research • Vol. 9, No. 10, 2010 4937

research articles Abbreviations: CBP, carboxypeptidase; CEL, carboxyl ester lipase; IEF, isoelectric focusing; PpiB, peptidyl-prolyl cis-trans isomerase B; RMCP-1, rat mast cell protease 1/chymase; ZG, zymogen granule; ZGC, zymogen granule content; ZGM, zymogen granule membrane.

Acknowledgment. We thank all those colleagues who provided antibodies and cDNA (see Materials and Methods). We thank V. Kramer, W. Ackermann and B. Agricola (Marburg, Germany) for excellent technical assistance and J. Nyalwidhe, D. Delacour and J. Adamkievicz for support with protein analyses in Marburg. This work was supported by the German Research Foundation (DFG; SCHR 518/5-1, 2), the J. Manchot foundation (Du ¨sseldorf, Germany), the Portuguese Foundation for Science and Technology (FCT) [REDE/1504/REM/2005, REEQ/1023/BIO/2005; SFRH/BD/38629/2007 (to C.R.), ESF/ JCCM 2007-2013, SFRH/BPD/37725/2007 (to M.G.L.), SFRH/ BD/48722/2008 (to M. A.)], and the University of Aveiro. Supporting Information Available: Supplementary tables and figures. This material is available free of charge via the Internet at http://pubs.acs.org. References (1) Leblond, F. A.; Viau, G.; Laine, J.; Lebel, D. Reconstitution in vitro of the pH-dependent aggregation of pancreatic zymogens en route to the secretory granule: implication of GP-2. Biochem. J. 1993, 291 (Pt 1), 289–96. (2) Freedman, S. D.; Scheele, G. A. Regulated secretory proteins in the exocrine pancreas aggregate under conditions that mimic the trans-Golgi network. Biochem. Biophys. Res. Commun. 1993, 197 (2), 992–9. (3) Colomer, V.; Kicska, G. A.; Rindler, M. J. Secretory granule content proteins and the luminal domains of granule membrane proteins aggregate in vitro at mildly acidic pH. J. Biol. Chem. 1996, 271 (1), 48–55. (4) Dartsch, H.; Kleene, R.; Kern, H. F. In vitro condensation-sorting of enzyme proteins isolated from rat pancreatic acinar cells. Eur. J. Cell Biol. 1998, 75, 211–22. (5) Arvan, P.; Castle, D. Sorting and storage during secretory granule biogenesis: looking backward and looking forward. Biochem. J. 1998, 332 (Pt 3), 593–610. (6) Borgonovo, B.; Ouwendijk, J.; Solimena, M. Biogenesis of secretory granules. Curr. Opin. Cell Biol. 2006, 18 (4), 365–70. (7) Tooze, S. A.; Martens, G. J. M.; Huttner, W. B. Secretory granule biogenesis: rafting to the SNARE. Trends Cell Biol. 2001, 11 (3), 116–22. (8) Wasle, B.; Edwardson, J. M. The regulation of exocytosis in the pancreatic acinar cell. Cell Signal. 2002, 14 (3), 191–7. (9) Williams, J. A. Regulation of pancreatic acinar cell function. Curr. Opin. Gastroenterol. 2006, 22 (5), 498–504. (10) Yu, S.; Lowe, A. W. The pancreatic zymogen granule membrane protein, GP2, binds Escherichia coli Type 1 fimbriae. BMC Gastroenterol. 2009, 9, 58. (11) Bach, J. P.; Borta, H.; Ackermann, W.; Faust, F.; Borchers, O.; Schrader, M. The secretory granule protein syncollin localizes to HL-60 cells and neutrophils. J. Histochem. Cytochem. 2006, 54 (8), 877–88. (12) Schrader, M. Membrane targeting in secretion. Subcell. Biochem. 2004, 37, 391–421. (13) Gaisano, H. Y.; Gorelick, F. S. New insights into the mechanisms of pancreatitis. Gastroenterology 2009, 136 (7), 2040–4. (14) Rindler, M. J.; Xu, C. F.; Gumper, I.; Smith, N. N.; Neubert, T. A. Proteomic Analysis of Pancreatic Zymogen Granules: Identification of New Granule Proteins. J. Proteome Res. 2007, 6, 2978–92. (15) Berkane, A. A.; Nguyen, H. T.; Tranchida, F.; Waheed, A. A.; Deyris, V.; Tchiakpe, L.; Fasano, C.; Nicoletti, C.; Desseaux, V.; Ajandouz el, H.; Comeau, D.; Comeau, L.; Hiol, A. Proteomic of lipid rafts in the exocrine pancreas from diet-induced obese rats. Biochem. Biophys. Res. Commun. 2007, 355 (3), 813–9. (16) Schmidt, K.; Dartsch, H.; Linder, D.; Kern, H.; Kleene, R. A submembranous matrix of proteoglycans on zymogen granule membranes is involved in granule formation in rat pancreatic acinar cells. J. Cell Sci. 2000, 113 (Pt 12), 2233–42.

4938

Journal of Proteome Research • Vol. 9, No. 10, 2010

Borta et al. (17) Schmidt, K.; Schrader, M.; Kern, H. F.; Kleene, R. Regulated apical secretion of zymogens in rat pancreas: Involvement of the GPIanchored glycoprotein GP-2, the lectin ZG16p and cholesterolglycosphingolipid enriched microdomains. J. Biol. Chem. 2001, 276, 14315–23. (18) Chen, X.; Walker, A. K.; Strahler, J. R.; Simon, E. S.; TomanicekVolk, S. L.; Nelson, B. B.; Hurley, M. C.; Ernst, S. A.; Williams, J. A.; Andrews, P. C. Organellar proteomics: analysis of pancreatic zymogen granule membranes. Mol. Cell. Proteomics 2006, 5 (2), 306–12. (19) Faust, F.; Gomez-Lazaro, M.; Borta, H.; Agricola, B.; Schrader, M. Rab8 is involved in zymogen granule formation in pancreatic acinar AR42J cells. Traffic 2008, 9 (6), 964–79. (20) Cronshagen, U.; Voland, P.; Kern, H. F. cDNA cloning and characterization of a novel 16 kDa protein located in zymogen granules of rat pancreas and goblet cells of the gut. Eur. J. Cell Biol. 1994, 65 (2), 366–77. (21) Schrader, M.; Temm-Grove, C. J.; Lessard, J. L.; Jockusch, B. M. Chicken antibodies to rabbit muscle actin with a restricted repertoire of F-actin recognition. Eur. J. Cell Biol. 1994, 63 (2), 326– 35. (22) Rabilloud, T. Use of thiourea to increase the solubility of membrane proteins in two-dimensional electrophoresis. Electrophoresis 1998, 19 (5), 758–60. (23) Yan, J. X.; Wait, R.; Berkelman, T.; Harry, R. A.; Westbrook, J. A.; Wheeler, C. H.; Dunn, M. J. A modified silver staining protocol for visualization of proteins compatible with matrix-assisted laser desorption/ionization and electrospray ionization-mass spectrometry. Electrophoresis 2000, 21 (17), 3666–72. (24) Kleinman, H. K.; McGoodwin, E. B.; Rennard, S. I.; Martin, G. R. Preparation of collagen substrates for cell attachment: effect of collagen concentration and phosphate buffer. Anal. Biochem. 1979, 94 (2), 308–12. (25) Schrader, M.; Reuber, B. E.; Morrell, J. C.; Jimenez-Sanchez, G.; Obie, C.; Stroh, T. A.; Valle, D.; Schroer, T. A.; Gould, S. J. Expression of PEX11beta mediates peroxisome proliferation in the absence of extracellular stimuli. J. Biol. Chem. 1998, 273 (45), 29607–14. (26) Slot, J. W.; Geuze, H. J. Cryosectioning and immunolabeling. Nat. Protoc. 2007, 2 (10), 2480–91. (27) Schwarz, H.; Humbel, B. M. Correlative light and electron microscopy using immunolabeled resin sections. Methods Mol. Biol. 2007, 369, 229–56. (28) Yonezawa, H. Secretory responses of hypertrophied rat pancreas induced by repeated oral administrations of a synthetic protease inhibitor. Jpn. J. Physiol. 1983, 33 (2), 183–95. (29) Hodel, A.; An, S. J.; Hansen, N. J.; Lawrence, J.; Wasle, B.; Schrader, M.; Edwardson, J. M. Cholesterol-dependent interaction of syncollin with the membrane of the pancreatic zymogen granule. Biochem. J. 2001, 356 (Pt 3), 843–50. (30) Kalus, I.; Hodel, A.; Koch, A.; Kleene, R.; Michael Edwardson, J.; Schrader, M. Interaction of syncollin with GP-2, the major membrane protein of pancreatic zymogen granules, and association with lipid microdomains. Biochem. J. 2002, 362 (Pt 2), 433– 42. (31) Kleene, R.; Kastner, B.; Rosser, R.; Kern, H. Complex formation among rat pancreatic secretory proteins under mild alkaline pH conditions. Digestion 1999, 60 (4), 305–13. (32) Kleene, R.; Dartsch, H.; Kern, H. F. The secretory lectin ZG16p mediates sorting of enzyme proteins to the zymogen granule membrane in pancreatic acinar cells. Eur. J. Cell Biol. 1999, 78, 79–90. (33) Edwardson, J. M.; An, S.; Jahn, R. The secretory granule protein syncollin binds to syntaxin in a Ca2(+)- sensitive manner. Cell 1997, 90 (2), 325–33. (34) Pejler, G.; Maccarana, M. Interaction of heparin with rat mast cell protease 1. J. Biol. Chem. 1994, 269 (20), 14451–6. (35) Pejler, G.; Sadler, J. E. Mechanism by which heparin proteoglycan modulates mast cell chymase activity. Biochemistry 1999, 38 (37), 12187–95. (36) Pejler, G.; Berg, L. Regulation of rat mast cell protease 1 activity. Protease inhibition is prevented by heparin proteoglycan. Eur. J. Biochem. 1995, 233 (1), 192–9. (37) Chen, X.; Ulintz, P. J.; Simon, E. S.; Williams, J. A.; Andrews, P. C. Global topology analysis of pancreatic zymogen granule membrane proteins. Mol. Cell. Proteomics 2008, 7 (12), 2323–36. (38) Iwai, N.; Inagami, T. Molecular cloning of a complementary DNA to rat cyclophilin-like protein mRNA. Kidney Int. 1990, 37 (6), 1460– 5. (39) Hanoulle, X.; Melchior, A.; Sibille, N.; Parent, B.; Denys, A.; Wieruszeski, J. M.; Horvath, D.; Allain, F.; Lippens, G.; Landrieu,

research articles

Peripheral Membrane Proteins of Zymogen Granules

(40) (41)

(42) (43)

(44) (45) (46)

(47)

(48)

(49) (50)

(51)

(52) (53) (54) (55) (56)

(57) (58)

(59) (60) (61)

(62)

I. Structural and functional characterization of the interaction between cyclophilin B and a heparin-derived oligosaccharide. J. Biol. Chem. 2007, 282 (47), 34148–58. Butler, G. S.; Overall, C. M. Proteomic identification of multitasking proteins in unexpected locations complicates drug targeting. Nat. Rev. Drug Discovery 2009, 8 (12), 935–48. Ackerman, S. J.; Liu, L.; Kwatia, M. A.; Savage, M. P.; Leonidas, D. D.; Swaminathan, G. J.; Acharya, K. R. Charcot-Leyden crystal protein (galectin-10) is not a dual function galectin with lysophospholipase activity but binds a lysophospholipase inhibitor in a novel structural fashion. J. Biol. Chem. 2002, 277 (17), 14859–68. Lombardo, D. Bile salt-dependent lipase: its pathophysiological implications. Biochim. Biophys. Acta 2001, 1533 (1), 1–28. Lombardo, D.; Fauvel, J.; Guy, O. Studies on the substrate specificity of a carboxyl ester hydrolase from human pancreatic juice. I. Action on carboxyl esters, glycerides and phospholipids. Biochim. Biophys. Acta 1980, 611 (1), 136–46. Bruneau, N.; de la Porte, P. L.; Sbarra, V.; Lombardo, D. Association of bile-salt-dependent lipase with membranes of human pancreatic microsomes. Eur. J. Biochem. 1995, 233 (1), 209–18. Bruneau, N.; Lombardo, D. Chaperone function of a Grp 94-related protein for folding and transport of the pancreatic bile saltdependent lipase. J. Biol. Chem. 1995, 270 (22), 13524–33. Yoshimori, T.; Semba, T.; Takemoto, H.; Akagi, S.; Yamamoto, A.; Tashiro, Y. Protein disulfide-isomerase in rat exocrine pancreatic cells is exported from the endoplasmic reticulum despite possessing the retention signal. J. Biol. Chem. 1990, 265 (26), 15984–90. Bruneau, N.; Lombardo, D.; Levy, E.; Bendayan, M. Roles of molecular chaperones in pancreatic secretion and their involvement in intestinal absorption. Microsc. Res. Tech. 2000, 49 (4), 329– 45. Logsdon, C. D.; Moessner, J.; Williams, J. A.; Goldfine, I. D. Glucocorticoids increase amylase mRNA levels, secretory organelles, and secretion in pancreatic acinar AR42J cells. J. Cell Biol. 1985, 100 (4), 1200–8. Yu, S.; Hao, Y.; Lowe, A. W. Effects of GP2 expression on secretion and endocytosis in pancreatic AR4-2J cells. Biochem. Biophys. Res. Commun. 2004, 322 (1), 320–5. Kredel, S.; Oswald, F.; Nienhaus, K.; Deuschle, K.; Rocker, C.; Wolff, M.; Heilker, R.; Nienhaus, G. U.; Wiedenmann, J. mRuby, a bright monomeric red fluorescent protein for labeling of subcellular structures. PLoS One 2009, 4 (2), e4391. Leskinen, M. J.; Lindstedt, K. A.; Wang, Y.; Kovanen, P. T. Mast cell chymase induces smooth muscle cell apoptosis by a mechanism involving fibronectin degradation and disruption of focal adhesions. Arterioscler. Thromb. Vasc. Biol. 2003, 23 (2), 238–43. Galat, A. Peptidylproline cis-trans-isomerases: immunophilins. Eur. J. Biochem. 1993, 216 (3), 689–707. Wang, P.; Heitman, J. The cyclophilins. Genome Biol. 2005, 6 (7), 226. Fischer, G.; Wittmann-Liebold, B.; Lang, K.; Kiefhaber, T.; Schmid, F. X. Cyclophilin and peptidyl-prolyl cis-trans isomerase are probably identical proteins. Nature 1989, 337 (6206), 476–8. Takahashi, N.; Hayano, T.; Suzuki, M. Peptidyl-prolyl cis-trans isomerase is the cyclosporin A-binding protein cyclophilin. Nature 1989, 337 (6206), 473–5. Price, E. R.; Jin, M.; Lim, D.; Pati, S.; Walsh, C. T.; McKeon, F. D. Cyclophilin B trafficking through the secretory pathway is altered by binding of cyclosporin A. Proc. Natl. Acad. Sci. U.S.A. 1994, 91 (9), 3931–5. Bose, S.; Freedman, R. B. Peptidyl prolyl cis-trans-isomerase activity associated with the lumen of the endoplasmic reticulum. Biochem. J. 1994, 300 (Pt 3), 865–70. Arber, S.; Krause, K. H.; Caroni, P. s-cyclophilin is retained intracellularly via a unique COOH-terminal sequence and colocalizes with the calcium storage protein calreticulin. J. Cell Biol. 1992, 116 (1), 113–25. Mariller, C.; Allain, F.; Kouach, M.; Spik, G. Evidence that human milk isolated cyclophilin B corresponds to a truncated form. Biochim. Biophys. Acta 1996, 1293 (1), 31–8. Takaki, Y.; Muta, T.; Iwanaga, S. A peptidyl-prolyl cis/transisomerase (cyclophilin G) in regulated secretory granules. J. Biol. Chem. 1997, 272 (45), 28615–21. Walz, A.; Stuhler, K.; Wattenberg, A.; Hawranke, E.; Meyer, H. E.; Schmalz, G.; Bluggel, M.; Ruhl, S. Proteome analysis of glandular parotid and submandibular-sublingual saliva in comparison to whole human saliva by two-dimensional gel electrophoresis. Proteomics 2006, 6 (5), 1631–9. Daum, S.; Schumann, M.; Mathea, S.; Aumuller, T.; Balsley, M. A.; Constant, S. L.; de Lacroix, B. F.; Kruska, F.; Braun, M.; Schiene-

(63) (64)

(65) (66) (67)

(68)

(69) (70) (71) (72)

(73)

(74)

(75) (76)

(77) (78)

(79) (80) (81)

(82)

(83)

Fischer, C. Isoform-specific inhibition of cyclophilins. Biochemistry 2009, 48 (26), 6268–77. Bukrinsky, M. I. Cyclophilins: unexpected messengers in intercellular communications. Trends Immunol. 2002, 23 (7), 323–5. Montague, J. W.; Hughes, F. M., Jr.; Cidlowski, J. A. Native recombinant cyclophilins A, B, and C degrade DNA independently of peptidylprolyl cis-trans-isomerase activity. Potential roles of cyclophilins in apoptosis. J. Biol. Chem. 1997, 272 (10), 6677–84. Bram, R. J.; Crabtree, G. R. Calcium signalling in T cells stimulated by a cyclophilin B-binding protein. Nature 1994, 371 (6495), 355– 8. Galat, A. Peptidylprolyl cis/trans isomerases (immunophilins): biological diversity--targets--functions. Curr. Top. Med. Chem. 2003, 3 (12), 1315–47. Aubert-Jousset, E.; Garmy, N.; Sbarra, V.; Fantini, J.; Sadoulet, M. O.; Lombardo, D. The combinatorial extension method reveals a sphingolipid binding domain on pancreatic bile salt-dependent lipase: role in secretion. Structure 2004, 12 (8), 1437–47. Gondre-Lewis, M. C.; Petrache, H. I.; Wassif, C. A.; Harries, D.; Parsegian, A.; Porter, F. D.; Loh, Y. P. Abnormal sterols in cholesterol-deficiency diseases cause secretory granule malformation and decreased membrane curvature. J. Cell Sci. 2006, 119 (Pt 9), 1876–85. Scheele, G. A.; Fukuoka, S.; Freedman, S. D. Role of the GP2/THP family of GPI-anchored proteins in membrane trafficking during regulated exocrine secretion. Pancreas 1994, 9 (2), 139–49. Dvorak, A. M.; Morgan, E. S. Ribonuclease-gold labels chondroitin sulphate in guinea pig basophil granules. Histochem. J. 1998, 30 (8), 603–8. Dvorak, A. M.; Morgan, E. S. Ribonuclease-gold labels heparin in human mast cell granules. New use for an ultrastructural enzyme affinity technique. J. Histochem. Cytochem. 1998, 46 (6), 695–706. Dvorak, A. M.; Morgan, E. S. Ribonuclease-gold ultrastructural localization of heparin in isolated human lung mast cells stimulated to undergo anaphylactic degranulation and recovery in vitro. Clin. Exp. Allergy 1999, 29 (8), 1118–28. Dvorak, A. M.; Morgan, E. S. Ribonuclease-gold labels proteoglycan-containing cytoplasmic granules and ribonucleic acid-containing organelles--a survey. Histol. Histopathol. 1999, 14 (2), 597– 626. Vanpouille, C.; Denys, A.; Carpentier, M.; Pakula, R.; Mazurier, J.; Allain, F. Octasaccharide is the minimal length unit required for efficient binding of cyclophilin B to heparin and cell surface heparan sulphate. Biochem. J. 2004, 382 (Pt 2), 733–40. Rebai, O.; Le Petit-Thevenin, J.; Bruneau, N.; Lombardo, D.; Verine, A. In vitro angiogenic effects of pancreatic bile salt-dependent lipase. Arterioscler. Thromb. Vasc. Biol. 2005, 25 (2), 359–64. Allain, F.; Vanpouille, C.; Carpentier, M.; Slomianny, M. C.; Durieux, S.; Spik, G. Interaction with glycosaminoglycans is required for cyclophilin B to trigger integrin-mediated adhesion of peripheral blood T lymphocytes to extracellular matrix. Proc. Natl. Acad. Sci. U.S.A. 2002, 99 (5), 2714–9. Castle, D.; Castle, A. Intracellular transport and secretion of salivary proteins. Crit. Rev. Oral Biol. Med. 1998, 9 (1), 4–22. Brion, C.; Miller, S. G.; Moore, H. P. Regulated and constitutive secretion. Differential effects of protein synthesis arrest on transport of glycosaminoglycan chains to the two secretory pathways. J. Biol. Chem. 1992, 267 (3), 1477–83. Grimes, M.; Kelly, R. B. Intermediates in the constitutive and regulated secretory pathways released in vitro from semi-intact cells. J. Cell Biol. 1992, 117 (3), 539–49. Huang, C.; Sali, A.; Stevens, R. L. Regulation and function of mast cell proteases in inflammation. J. Clin. Immunol. 1998, 18 (3), 169– 83. Lutzelschwab, C.; Pejler, G.; Aveskogh, M.; Hellman, L. Secretory granule proteases in rat mast cells. Cloning of 10 different serine proteases and a carboxypeptidase A from various rat mast cell populations. J. Exp. Med. 1997, 185 (1), 13–29. Matsumoto, R.; Sali, A.; Ghildyal, N.; Karplus, M.; Stevens, R. L. Packaging of proteases and proteoglycans in the granules of mast cells and other hematopoietic cells. A cluster of histidines on mouse mast cell protease 7 regulates its binding to heparin serglycin proteoglycans. J. Biol. Chem. 1995, 270 (33), 19524–31. Niemann, C. U.; Cowland, J. B.; Ralfkiaer, E.; Abrink, M.; Pejler, G.; Borregaard, N. Serglycin proteoglycan is not implicated in localizing exocrine pancreas enzymes to zymogen granules. Eur. J. Cell Biol. 2009, 88 (8), 473–9.

PR100052Q

Journal of Proteome Research • Vol. 9, No. 10, 2010 4939