Biologically Active Decorin Is a Monomer in Solution
2004; Elsevier BV; Volume: 279; Issue: 8 Linguagem: Inglês
10.1074/jbc.m310342200
ISSN1083-351X
AutoresSilvia Goldoni, Rick T. Owens, David J. McQuillan, Zachary Shriver, Ram Sasisekharan, David E. Birk, Shelly Campbell, Renato V. Iozzo,
Tópico(s)Cell Adhesion Molecules Research
ResumoIt has been reported that decorin and its protein core can have molecular masses nearly double the size of those previously published, suggesting a dimeric structure. In this study we tested whether biologically active decorin and its glycoprotein core would form dimers in solution. We used homo- and hetero-bifunctional chemical cross-linking reagents, BS3 and sulfo-SMPB, respectively, as well as glutaraldehyde and found no preferential dimer formation, whether chemical cross-linking was performed in the presence or absence of live cells. Under the same experimental conditions, we easily detected dimers of epidermal growth factor receptor and basic fibroblast growth factor, two glycoproteins known to dimerize. Only at very high cross-linker to decorin molar ratios (2000:1) were trimers and multimers observed, but performing the chemical cross-linking in the presence of a reducing agent abolished these. The elution of decorin protein core in Superose 6 gel chromatography gave masses compatible with monomeric proteins, both before and after denaturation with 2.5 m guanidine HCl. Matrix-assisted laser desorption ionization gave a mass of 44,077 Da for decorin protein core, without any evidence of dimers or oligomers. Extensive oligomerization of the decorin protein core was observed only after dialysis against water and freeze-drying. These oligomers were considered artifacts because they were independent of chemical cross-linking and were resistant to heat denaturation and disulfide-bond reduction. Oligomeric preparations showed markedly reduced biological activity in both phosphorylation and collagen fibrillogenesis assays. Thus, biologically active decorin is a monomer in solution and, as such, is a monovalent ligand for various extracellular matrix proteins, growth factors, and cell surface receptors. It has been reported that decorin and its protein core can have molecular masses nearly double the size of those previously published, suggesting a dimeric structure. In this study we tested whether biologically active decorin and its glycoprotein core would form dimers in solution. We used homo- and hetero-bifunctional chemical cross-linking reagents, BS3 and sulfo-SMPB, respectively, as well as glutaraldehyde and found no preferential dimer formation, whether chemical cross-linking was performed in the presence or absence of live cells. Under the same experimental conditions, we easily detected dimers of epidermal growth factor receptor and basic fibroblast growth factor, two glycoproteins known to dimerize. Only at very high cross-linker to decorin molar ratios (2000:1) were trimers and multimers observed, but performing the chemical cross-linking in the presence of a reducing agent abolished these. The elution of decorin protein core in Superose 6 gel chromatography gave masses compatible with monomeric proteins, both before and after denaturation with 2.5 m guanidine HCl. Matrix-assisted laser desorption ionization gave a mass of 44,077 Da for decorin protein core, without any evidence of dimers or oligomers. Extensive oligomerization of the decorin protein core was observed only after dialysis against water and freeze-drying. These oligomers were considered artifacts because they were independent of chemical cross-linking and were resistant to heat denaturation and disulfide-bond reduction. Oligomeric preparations showed markedly reduced biological activity in both phosphorylation and collagen fibrillogenesis assays. Thus, biologically active decorin is a monomer in solution and, as such, is a monovalent ligand for various extracellular matrix proteins, growth factors, and cell surface receptors. Decorin, the prototypic member of an enlarging family of small leucine-rich proteoglycans (1Iozzo R.V. Crit. Rev. Biochem. Mol. Biol. 1997; 32: 141-174Crossref PubMed Scopus (454) Google Scholar, 2Hocking A.M. Shinomura T. McQuillan D.J. Matrix Biol. 1998; 17: 1-19Crossref PubMed Scopus (416) Google Scholar, 3Iozzo R.V. J. Biol. Chem. 1999; 274: 18843-18846Abstract Full Text Full Text PDF PubMed Scopus (577) Google Scholar, 4Neame P.J. Kay C.J. Iozzo R.V. Proteoglycans: Structure, Biology and Molecular Interactions. Marcel Dekker, Inc., New York, NY2000: 201-236Google Scholar, 5Reed C.C. Iozzo R.V. Glycoconj. J. 2003; 19: 249-255Crossref Scopus (312) Google Scholar), has attracted considerable attention in the past decade primarily because of its ability to affect several key biological processes. Decorin is known to modulate growth factor activity (6Yamaguchi Y. Mann D.M. Ruoslahti E. Nature. 1990; 346: 281-284Crossref PubMed Scopus (1303) Google Scholar), collagen fibrillogenesis (7Toole B.P. Lowther D.A. Biochem. J. 1968; 109: 857-866Crossref PubMed Scopus (127) Google Scholar, 8Toole B.P. Nature. 1969; 222: 872-873Crossref PubMed Scopus (39) Google Scholar, 9Vogel K.G. Paulsson M. Heinegård D. Biochem. J. 1984; 223: 587-597Crossref PubMed Scopus (705) Google Scholar, 10Rada J.A. Cornuet P.K. Hassell J.R. Exp. Eye Res. 1993; 56: 635-648Crossref PubMed Scopus (291) Google Scholar, 11Danielson K.G. Baribault H. Holmes D.F. Graham H. Kadler K.E. Iozzo R.V. J. Cell Biol. 1997; 136: 729-743Crossref PubMed Scopus (1189) Google Scholar), receptor tyrosine kinase activity (12Iozzo R.V. Moscatello D. McQuillan D.J. Eichstetter I. J. Biol. Chem. 1999; 274: 4489-4492Abstract Full Text Full Text PDF PubMed Scopus (322) Google Scholar, 13Santra M. Reed C.C. Iozzo R.V. J. Biol. Chem. 2002; 277: 35671-35681Abstract Full Text Full Text PDF PubMed Scopus (197) Google Scholar), cancer growth (14Santra M. Mann D.M. Mercer E.W. Skorski T. Calabretta B. Iozzo R.V. J. Clin. Invest. 1997; 100: 149-157Crossref PubMed Scopus (184) Google Scholar, 15Santra M. Eichstetter I. Iozzo R.V. J. Biol. Chem. 2000; 275: 35153-35161Abstract Full Text Full Text PDF PubMed Scopus (183) Google Scholar, 16Iozzo R.V. Chakrani F. Perrotti D. McQuillan D.J. Skorski T. Calabretta B. Eichstetter I. Proc. Natl. Acad. Sci. U. S. A. 1999; 96: 3092-3097Crossref PubMed Scopus (122) Google Scholar, 17Reed C.C. Gauldie J. Iozzo R.V. Oncogene. 2002; 21: 3688-3695Crossref PubMed Scopus (126) Google Scholar, 18Troup S. Njue C. Kliewer E.V. Parisien M. Roskelley C. Chakravarti S. Roughley P.J. Murphy L.C. Watson P.H. Clin. Cancer Res. 2003; 9: 207-214PubMed Google Scholar), angiogenesis (19Schönherr E. O'Connell B.C. Schittny J. Robenek H. Fastermann D. Fisher L.W. Plenz G. Vischer P. Young M.F. Kresse H. Eur. J. Cell Biol. 1999; 78: 44-55Crossref PubMed Scopus (76) Google Scholar, 20de Lange Davies C. Melder R.J. Munn L.L. Mouta-Carreira C. Jain R.K. Boucher Y. Microvasc. Res. 2001; 62: 26-42Crossref PubMed Scopus (90) Google Scholar, 21Grant D.S. Yenisey C. Rose R.W. Tootell M. Santra M. Iozzo R.V. Oncogene. 2002; 21: 4765-4777Crossref PubMed Scopus (189) Google Scholar), tissue remodeling (22Funderburgh J.L. Hevelone N.D. Roth M.R. Funderburgh M.L. Rodrigues M.R. Nirankari V.S. Conrad G.W. Investig. Ophthalmol. Vis. Sci. 1998; 39: 1957-1964PubMed Google Scholar, 23Huttenlocher A. Werb Z. Tremble P. Huhtala P. Rosenberg L. Damsky C.H. Matrix Biol. 1996; 15: 239-250Crossref PubMed Scopus (38) Google Scholar, 24Xu G. Guimond M.-J. Chakraborty C. Lala P.K. Biol. Reprod. 2002; 67: 681-689Crossref PubMed Scopus (110) Google Scholar), bacterial infection (25Guo B. Norris S. Rosenberg L.C. Höök M. Infect. Immun. 1995; 63: 3467-3472Crossref PubMed Google Scholar), and cardiovascular (26Kinsella M.G. Fischer J.W. Mason D.P. Wight T.N. J. Biol. Chem. 2000; 275: 13924-13932Abstract Full Text Full Text PDF PubMed Scopus (73) Google Scholar) and periodontal (27Häkkinen L. Strassburger S. Kahari V.M. Scott P.G. Eichstetter I. Iozzo R.V. Larjava H. Lab. Invest. 2000; 80: 1869-1880Crossref PubMed Scopus (99) Google Scholar) diseases. This multiplicity of functions is due to the unusual structure of the decorin glycoprotein core which harbors 12 leucine-rich repeats flanked by cysteine-rich regions with four at the N terminus and two at the C terminus. Molecular modeling has predicted a horseshoe shape for the decorin core (28Weber I.T. Harrison R.W. Iozzo R.V. J. Biol. Chem. 1996; 271: 31767-31770Abstract Full Text Full Text PDF PubMed Scopus (307) Google Scholar), with the inner face made up of parallel β-strands thought to bind directly to various ligands such as collagen type I (29Svensson L. Heinegård D. Oldberg Å. J. Biol. Chem. 1995; 270: 20712-20716Abstract Full Text Full Text PDF PubMed Scopus (202) Google Scholar, 30Schönherr E. Hausser H. Beavan L. Kresse H. J. Biol. Chem. 1995; 270: 8877-8883Abstract Full Text Full Text PDF PubMed Scopus (184) Google Scholar, 31Kresse H. Liszio C. Schönherr E. Fisher L.W. J. Biol. Chem. 1997; 272: 18404-18410Abstract Full Text Full Text PDF PubMed Scopus (85) Google Scholar, 32Keene D.R. San Antonio J.D. Mayne R. McQuillan D.J. Sarris G. Santoro S.A. Iozzo R.V. J. Biol. Chem. 2000; 275: 21801-21804Abstract Full Text Full Text PDF PubMed Scopus (178) Google Scholar), transforming growth factor-β (33Schönherr E. Broszat M. Brandan E. Bruckner P. Kresse H. Arch. Biochem. Biophys. 1998; 35: 241-248Crossref Scopus (134) Google Scholar), or EGFR 1The abbreviations used are: EGFR, epidermal growth factor receptor; FGF2, basic fibroblast growth factor; sulfo-SMPB, sulfosuccinimidyl-4-(p-maleimidophenyl) butyrate; BS3, bis(sulfosuccinimidyl) suberate; MAPK, mitogen activated protein kinase; βME, β-mercaptoethanol; TBS, Tris-buffered saline; GdnHCl, guanidine hydrochloride; CHAPS, 3-[(3-cholamidopropyl)dimethylammonio]-1-propanesulfonate; BSA, bovine serum albumin; MALDI-MS, matrix-assisted laser desorption ionization mass spectrometry; PBS, phosphate-buffered saline. (13Santra M. Reed C.C. Iozzo R.V. J. Biol. Chem. 2002; 277: 35671-35681Abstract Full Text Full Text PDF PubMed Scopus (197) Google Scholar). This view implies that the biologically active form of decorin is a monomer in solution and thus a monovalent ligand for various extracellular matrix proteins and surface receptors. However, a recent paper (34Scott P.G. Grossmann J.G. Dodd C.M. Sheehan J.K. Bishop P.N. J. Biol. Chem. 2003; 278: 18353-18359Abstract Full Text Full Text PDF PubMed Scopus (52) Google Scholar) has challenged this body of evidence by reporting that the predominant form of decorin in solution is a dimer. According to this report, the inner concave face of decorin, a structural feature that has been proposed to provide sites for protein-protein interaction in the prototype ribonuclease inhibitor (35Kobe B. Deisenhofer J. Nature. 1995; 374: 183-186Crossref PubMed Scopus (578) Google Scholar) as well as other leucine-rich repeat proteins (36Kobe B. Kajava A.V. Curr. Opin. Struct. Biol. 2001; 11: 725-732Crossref PubMed Scopus (1298) Google Scholar), is indeed facilitating the dimerization. In the present study, we provide several lines of evidence that decorin proteoglycan and its glycoprotein core are primarily, if not exclusively, monomers in solution. First, we purified recombinant decorin under non-denaturing conditions from the secretions of cultured human fibrosarcoma or 293 embryonic kidney cells and verified its biological activity by specific interaction with the EGFR. Second, this decorin preparation was tested in affinity chemical cross-linking reactions using several cross-linking agents of diverse functionality and bridge length, in the presence of live cells and in cell-free systems. In all cases, there was no preferential dimer formation. In contrast, we detected significant dimerization of both EGFR and FGF2, two glycoproteins known to dimerize. Only at very high cross-linker to decorin molar ratios (2000:1) were trimers and multimers observed, but these could be abolished by a reducing agent. Third, the elution of decorin protein core in Superose 6 gel chromatography gave masses compatible with the monomeric proteins, both before and after denaturation with the chaotropic agent GdnHCl. Fourth, MALDI-MS gave a mass of 44,077 Da for decorin protein core, without any evidence for dimers or oligomers. Finally, extensive oligomerization of decorin protein core was observed after dialysis against water and freeze-drying, and these preparations lost activity in phosphorylation and collagen fibrillogenesis assays. We conclude that biologically active decorin and its glycoprotein core are monomers in solution. Materials—All the chemicals unless specified were obtained from Sigma. Dulbecco's modified Eagle's medium, fetal bovine serum, 100× antibiotic-antimycotic solution, and Dulbecco's phosphate-buffered saline (DPBS) were purchased from Mediatech (Herndon, VA). Recombinant human EGF and FGF2 were purchased from Promega (Madison, WI). Bis(sulfosuccinimidyl) suberate (BS3) and sulfosuccinimidyl-4-(pmaleimidophenyl)butyrate (sulfo-SMPB) were purchased from Pierce. Solutions of cross-linkers were freshly prepared before each experiment. Glutaraldehyde (25% solution) EM grade (Electron Microscopy Sciences, Washington, PA) was stored in aliquots at -80 °C. Nitrocellulose membrane was purchased from Bio-Rad. Antibodies include polyclonal rabbit antibodies against the N-terminal region of decorin (37Fisher L.W. Stubbs III, J.T. Young M.F. Acta Orthop. Scand. 1995; 66: 61-65Crossref Scopus (0) Google Scholar), against the C-terminal region of the EGF receptor (sc-03, Santa Cruz Biotechnology, Santa Cruz, CA), against FGF2 (sc-79, Santa Cruz Biotechnology), against phospho-p44/42 MAPK (Cell Signaling Technology, Beverly, MA), and a monoclonal antibody against phosphotyrosine (PY20, BD Transduction Laboratories, San Diego, CA). Horseradish peroxidase-conjugated donkey anti-rabbit and sheep anti-mouse antibodies were purchased from Amersham Biosciences. SuperSignal West Pico chemiluminescent substrate was purchased from Pierce. Expression and Purification of Recombinant Decorin and Biglycan and Gel Filtration Chromatography—Recombinant decorin was produced by using either a recombinant vaccinia virus expression system or a stably transfected 293-EBNA cell line. Generation of the recombinant vaccinia virus encoding the mature form of decorin (38Ramamurthy P. Hocking A.M. McQuillan D.J. J. Biol. Chem. 1996; 271: 19578-19584Abstract Full Text Full Text PDF PubMed Scopus (89) Google Scholar) and biglycan (39Hocking A.M. Strugnell R.A. Ramamurthy P. McQuillan D.J. J. Biol. Chem. 1996; 271: 19571-19577Abstract Full Text Full Text PDF PubMed Scopus (57) Google Scholar) has been described previously. This system was adapted for large-scale expression using a Celligen Plus bioreactor (New Brunswick Scientific, Edison, NJ) packed with Fibra-Cel disks. In brief, HT-1080 cells were seeded in the bioreactor and allowed to grow to saturation. At that time, recombinant virus was added in serum-free media, and the infection was allowed to proceed for 48 h. Serum-free media was then collected, and the recombinant polyhistidine decorin fusion protein was purified by nickel-chelating chromatography and elution with a gradient of 0–250 mm imidazole in 20 mm Tris-HCl, 500 mm NaCl, 0.2% CHAPS, pH 8.0. Fractions containing decorin were pooled, concentrated, and desalted into PBS containing 0.2% CHAPS. The decorin-expressing 293-EBNA cell line was created by transferring the vaccinia decorin construct into the pCEP4 (Invitrogen) expression vector. After transfection, stable expressing cells were selected with hygromycin. Cells were then grown to saturation in the Celligen Plus bioreactor, and protein production was achieved by switching to serum-free culture media. Conditioned media was collected every 48 h. After concentration of the conditioned media using a Pellicon 2 Tangential Flow system (Millipore, Bedford, MA), recombinant decorin was purified as described above. Both expression systems resulted in the production of protein core and proteoglycan forms of decorin. In some experiments, protein core was separated from proteoglycan after anion-exchange chromatography on Q-Sepharose and elution with a linear gradient of 0.15–2 M NaCl in PBS, 0.2% CHAPS. Protein core samples were analyzed by gel filtration chromatography before and after dialysis against water and freeze-drying. Dried proteins were resuspended in TBS (20 mm Tris-HCl, 150 mm NaCl, pH 7.0) and chromatographed on Superose 6 HR 10/30 (Amersham Biosciences) in TBS with or without 2.5 m GdnHCl. Biological Activity of Decorin and Cross-linking Experiments—To assess the biological activity of our decorin and decorin protein core preparations, confluent cultures of A431 squamous carcinoma cells (40Csordás G. Santra M. Reed C.C. Eichstetter I. McQuillan D.J. Gross D. Nugent M.A. Hajnóczky G. Iozzo R.V. J. Biol. Chem. 2000; 275: 32879-32887Abstract Full Text Full Text PDF PubMed Scopus (199) Google Scholar) were serum-starved for 16 h and incubated for various times at 37 °C with decorin or its protein core (0.5–1 μm) or EGF (16 nm)in400 μl total volume of DPBS (2.5 mm Ca2+), 20 mm HEPES, pH 7.5. The cell lysates were subjected to immunoblotting with anti-P-tyrosine, anti-EGFR, and anti-P-MAPK antibodies. Fibrillogenesis assays were done essentially as described previously (41Birk D.E. Silver F.H. Arch. Biochem. Biophys. 1984; 235: 178-185Crossref PubMed Scopus (82) Google Scholar). Briefly, at 4 °C, stock solutions (1–2 mg ml-1) of pepsin-extracted, bovine dermal type I collagen were neutralized with 10× PBS and brought to the desired concentration with 1× PBS (15 mm sodium phosphate, 0.15 m NaCl, pH 7.2). Decorin protein core was added to aliquots of the mixture at 25–50 μgml-1. The samples were transferred to the water-jacketed sample holder of a Beckman DU640 spectrophotometer, and assays were performed at 37 °C. Chemical cross-linking reactions were performed in the presence of A431 cells or in cell-free experimental conditions. In the former case, cells were incubated with decorin for 30 min, and then the cross-linker sulfo-SMPB (1 mm) was added for 20–30 min at 37 °C before being quenched with 90 mm (final concentration) glycine, 9 mm Tris-HCl, pH 8.5, for 5 min. After the cross-linking reaction, 250-μl aliquots were collected and precipitated with 1.5 ml of ethanol/potassium acetate at -20 °C for 18 h. Samples were subjected to a linear gradient SDS-PAGE (3–15% (w/v) gel) with the addition of reducing agent (0.2% βME). In the cell-free experiments, chemical cross-linking was carried out in a 40-μl volume containing either decorin or decorin protein core (100–200 nm) in DPBS (2.5 mm Ca2+) in the presence or absence of 0.2% β ME. 10-μl aliquots of each serially diluted cross-linker (sulfo-SMPB, BS3, or glutaraldehyde) were added to the mixture and incubated for 30 min at room temperature with gentle shaking. The samples were quenched as above, boiled for 5 min, and subjected to SDS-PAGE (3–15% linear gradient and 8.5% gel for decorin and its protein core, respectively) under reducing conditions. Western immunoblottings to detect decorin were carried out with specific antibodies (37Fisher L.W. Stubbs III, J.T. Young M.F. Acta Orthop. Scand. 1995; 66: 61-65Crossref Scopus (0) Google Scholar, 38Ramamurthy P. Hocking A.M. McQuillan D.J. J. Biol. Chem. 1996; 271: 19578-19584Abstract Full Text Full Text PDF PubMed Scopus (89) Google Scholar). For additional information, see the text and figure legends. MALDI-MS—Sinapinic acid (∼10 mg ml-1) in 30% (v/v) acetonitrile was used as a matrix solution. About 1 pmol of myoglobin, carbonic anhydrase, or decorin protein core was added to the matrix at 1:10 ratio. After crystallization, the surfaces were washed with water, dried under a stream of nitrogen, and placed into the mass spectrometer. MALDI-MS spectra of the protein preparations were acquired in the linear mode with a Voyager Elite Reflectron time-of-flight instrument (PerSeptive Biosystems) fitted with a 337-nm wavelength nitrogen laser essentially as described before (42Davis J.C. Venkataraman G. Shriver Z. Raj P.A. Sasisekharan R. Biochem. J. 1999; 341: 613-620Crossref PubMed Google Scholar). Decorin protein core preparations were analyzed under several different instrument conditions with several different protein concentrations and under conditions that have previously been shown to be amenable to measuring dimer formation for FGF2 (42Davis J.C. Venkataraman G. Shriver Z. Raj P.A. Sasisekharan R. Biochem. J. 1999; 341: 613-620Crossref PubMed Google Scholar). Biologically Active Decorin Does Not Form Dimers in Solution with Three Cross-linkers of Diverse Bridge Length and Functional Properties—Our recombinant decorin preparation, which was kept in solution at a concentration of ∼1–1.5 mg ml-1 (38Ramamurthy P. Hocking A.M. McQuillan D.J. J. Biol. Chem. 1996; 271: 19578-19584Abstract Full Text Full Text PDF PubMed Scopus (89) Google Scholar), contained about 85% decorin proteoglycan, ∼13% protein core, and only 50 μg ml-1 and in the presence of 5 mm Zn2+, existed predominantly as a hexamer with a z-average molecular weight of ∼6 × 105 (43Liu J. Laue T.M. Choi H.U. Tang L.H. Rosenberg L. J. Bi
Referência(s)