A new MMP‐mediated prodomain cleavage mechanism to activate bone morphogenetic proteins from the extracellular matrix
2021; Wiley; Volume: 35; Issue: 3 Linguagem: Inglês
10.1096/fj.202001264r
ISSN1530-6860
AutoresAriane G. Furlan, Chara E. S. Spanou, Alan R. Godwin, Alexander P. Wohl, Laura‐Marie A. Zimmermann, Thomas Imhof, Manuel Koch, Clair Baldock, Gerhard Sengle,
Tópico(s)Signaling Pathways in Disease
ResumoThe FASEB JournalVolume 35, Issue 3 e21353 RESEARCH ARTICLEOpen Access A new MMP-mediated prodomain cleavage mechanism to activate bone morphogenetic proteins from the extracellular matrix Correction(s) for this article Erratum to "A new MMP-mediated prodomain cleavage mechanism to activate bone morphogenetic proteins from the extracellular matrix" Volume 36Issue 7The FASEB Journal First Published online: June 1, 2022 Ariane G. Furlan, Ariane G. Furlan Center for Biochemistry, Faculty of Medicine, University Hospital Cologne, University of Cologne, Cologne, GermanySearch for more papers by this authorChara E. S. Spanou, Chara E. S. Spanou Center for Biochemistry, Faculty of Medicine, University Hospital Cologne, University of Cologne, Cologne, Germany Department of Pediatrics and Adolescent Medicine, Faculty of Medicine, University Hospital Cologne, University of Cologne, Cologne, GermanySearch for more papers by this authorAlan R. F. Godwin, Alan R. F. Godwin Wellcome Trust Centre for Cell-Matrix Research, School of Biological Sciences, Faculty of Biology, Medicine and Health, Manchester Academic Health Science Centre, University of Manchester, Manchester, UK Division of Cell Matrix Biology and Regenerative Medicine, School of Biological Sciences, Faculty of Biology, Medicine and Health, Manchester Academic Health Science Centre, University of Manchester, Manchester, UKSearch for more papers by this authorAlexander P. Wohl, Alexander P. Wohl Center for Biochemistry, Faculty of Medicine, University Hospital Cologne, University of Cologne, Cologne, GermanySearch for more papers by this authorLaura-Marie A. Zimmermann, Laura-Marie A. Zimmermann Center for Biochemistry, Faculty of Medicine, University Hospital Cologne, University of Cologne, Cologne, Germany Department of Pediatrics and Adolescent Medicine, Faculty of Medicine, University Hospital Cologne, University of Cologne, Cologne, GermanySearch for more papers by this authorThomas Imhof, Thomas Imhof Institute for Dental Research and Oral Musculoskeletal Biology, Faculty of Medicine, University Hospital Cologne, University of Cologne, Cologne, GermanySearch for more papers by this authorManuel Koch, Manuel Koch Institute for Dental Research and Oral Musculoskeletal Biology, Faculty of Medicine, University Hospital Cologne, University of Cologne, Cologne, Germany Center for Molecular Medicine Cologne (CMMC), University of Cologne, Cologne, GermanySearch for more papers by this authorClair Baldock, Clair Baldock Wellcome Trust Centre for Cell-Matrix Research, School of Biological Sciences, Faculty of Biology, Medicine and Health, Manchester Academic Health Science Centre, University of Manchester, Manchester, UK Division of Cell Matrix Biology and Regenerative Medicine, School of Biological Sciences, Faculty of Biology, Medicine and Health, Manchester Academic Health Science Centre, University of Manchester, Manchester, UKSearch for more papers by this authorGerhard Sengle, Corresponding Author Gerhard Sengle [email protected] Center for Biochemistry, Faculty of Medicine, University Hospital Cologne, University of Cologne, Cologne, Germany Department of Pediatrics and Adolescent Medicine, Faculty of Medicine, University Hospital Cologne, University of Cologne, Cologne, Germany Center for Molecular Medicine Cologne (CMMC), University of Cologne, Cologne, Germany Cologne Center for Musculoskeletal Biomechanics (CCMB), Cologne, Germany Correspondence Gerhard Sengle, Center for Biochemistry, Faculty of Medicine, University Hospital Cologne, University of Cologne, Joseph-Stelzmann-Street 52, D-50931 Cologne, Germany. Email: [email protected]Search for more papers by this author Ariane G. Furlan, Ariane G. Furlan Center for Biochemistry, Faculty of Medicine, University Hospital Cologne, University of Cologne, Cologne, GermanySearch for more papers by this authorChara E. S. Spanou, Chara E. S. Spanou Center for Biochemistry, Faculty of Medicine, University Hospital Cologne, University of Cologne, Cologne, Germany Department of Pediatrics and Adolescent Medicine, Faculty of Medicine, University Hospital Cologne, University of Cologne, Cologne, GermanySearch for more papers by this authorAlan R. F. Godwin, Alan R. F. Godwin Wellcome Trust Centre for Cell-Matrix Research, School of Biological Sciences, Faculty of Biology, Medicine and Health, Manchester Academic Health Science Centre, University of Manchester, Manchester, UK Division of Cell Matrix Biology and Regenerative Medicine, School of Biological Sciences, Faculty of Biology, Medicine and Health, Manchester Academic Health Science Centre, University of Manchester, Manchester, UKSearch for more papers by this authorAlexander P. Wohl, Alexander P. Wohl Center for Biochemistry, Faculty of Medicine, University Hospital Cologne, University of Cologne, Cologne, GermanySearch for more papers by this authorLaura-Marie A. Zimmermann, Laura-Marie A. Zimmermann Center for Biochemistry, Faculty of Medicine, University Hospital Cologne, University of Cologne, Cologne, Germany Department of Pediatrics and Adolescent Medicine, Faculty of Medicine, University Hospital Cologne, University of Cologne, Cologne, GermanySearch for more papers by this authorThomas Imhof, Thomas Imhof Institute for Dental Research and Oral Musculoskeletal Biology, Faculty of Medicine, University Hospital Cologne, University of Cologne, Cologne, GermanySearch for more papers by this authorManuel Koch, Manuel Koch Institute for Dental Research and Oral Musculoskeletal Biology, Faculty of Medicine, University Hospital Cologne, University of Cologne, Cologne, Germany Center for Molecular Medicine Cologne (CMMC), University of Cologne, Cologne, GermanySearch for more papers by this authorClair Baldock, Clair Baldock Wellcome Trust Centre for Cell-Matrix Research, School of Biological Sciences, Faculty of Biology, Medicine and Health, Manchester Academic Health Science Centre, University of Manchester, Manchester, UK Division of Cell Matrix Biology and Regenerative Medicine, School of Biological Sciences, Faculty of Biology, Medicine and Health, Manchester Academic Health Science Centre, University of Manchester, Manchester, UKSearch for more papers by this authorGerhard Sengle, Corresponding Author Gerhard Sengle [email protected] Center for Biochemistry, Faculty of Medicine, University Hospital Cologne, University of Cologne, Cologne, Germany Department of Pediatrics and Adolescent Medicine, Faculty of Medicine, University Hospital Cologne, University of Cologne, Cologne, Germany Center for Molecular Medicine Cologne (CMMC), University of Cologne, Cologne, Germany Cologne Center for Musculoskeletal Biomechanics (CCMB), Cologne, Germany Correspondence Gerhard Sengle, Center for Biochemistry, Faculty of Medicine, University Hospital Cologne, University of Cologne, Joseph-Stelzmann-Street 52, D-50931 Cologne, Germany. Email: [email protected]Search for more papers by this author First published: 25 February 2021 https://doi.org/10.1096/fj.202001264RCitations: 4AboutSectionsPDF ToolsRequest permissionExport citationAdd to favoritesTrack citation ShareShare Give accessShare full text accessShare full-text accessPlease review our Terms and Conditions of Use and check box below to share full-text version of article.I have read and accept the Wiley Online Library Terms and Conditions of UseShareable LinkUse the link below to share a full-text version of this article with your friends and colleagues. Learn more.Copy URL Abstract Since their discovery as pluripotent cytokines extractable from bone matrix, it has been speculated how bone morphogenetic proteins (BMPs) become released and activated from the extracellular matrix (ECM). In contrast to TGF-βs, most investigated BMPs are secreted as bioactive prodomain (PD)–growth factor (GF) complexes (CPLXs). Recently, we demonstrated that PD-dependent targeting of BMP-7 CPLXs to the extracellular fibrillin microfibril (FMF) components fibrillin-1 and -2 represents a BMP sequestration mechanism by rendering the GF latent. Understanding how BMPs become activated from ECM scaffolds such as FMF is crucial to elucidate pathomechanisms characterized by aberrant BMP activation and ECM destruction. Here, we describe a new MMP-dependent BMP-7 activation mechanism from ECM-targeted pools via specific PD degradation. Using Edman sequencing and mutagenesis, we identified a new and conserved MMP-13 cleavage site within the BMP-7 PD. A degradation screen with different BMP family PDs and representative MMP family members suggested utilization of the identified site in a general MMP-driven BMP activation mechanism. Furthermore, sandwich ELISA and solid phase cleavage studies in combination with bioactivity assays, single particle TEM, and in silico molecular docking experiments provided evidence that PD cleavage by MMP-13 leads to BMP-7 CPLX disintegration and bioactive GF release. Abbreviations BMP bone morphogenetic protein CD circular dichroism spectroscopy CPLX complex DPP decapentaplegic EBNA Epstein-Barr virus nuclear antigen ECM extracellular matrix FMF fibrillin microfibrils FRET fluorescence resonance energy transfer GDF growth and differentiation factor GF growth factor HEK human embryonic kidney LAP latency-associated peptide LLC large latent complex LTBP latent TGF-β-binding protein Mab monoclonal antibody MMP matrix metalloproteinase PD prodomain TEM transmission electron microscopy TGF transforming growth factor TLL2 tolloid-like protein 2 TMB tetramethylbenzidine 1 INTRODUCTION Bone morphogenetic proteins (BMPs) belong to the TGF-β superfamily of growth factors (GFs) and play a central role in a multitude of cellular processes during embryogenesis and postnatal homeostasis by guiding cell differentiation, proliferation, survival, and apoptosis.1, 2 Originally, BMPs were discovered as pluripotent cytokines extractable from bone matrix that are capable to induce ectopic bone formation.3 Further studies confirmed that BMPs are stored in embryonic and perinatal connective tissues such as kidney, skin, and blood vessels.4-6 These findings implicate that ECM-bound BMPs serve an important function and that mechanisms for their utilization must exist to control their release at the appropriate time points when their action is required. For instance, the importance of ECM-bound BMPs is not restricted to fetal osteogenesis but also for regeneration of adult bones as illustrated by limb-specific Bmp2 null mice which presented with irreversible spontaneous fractures.7 Among extracellular matrix (ECM) networks collagen fibers were first considered as the primary scaffold responsible for BMP sequestration.8, 9 Thereby, it was shown that procollagen-2 is able to specifically bind BMP-2 and thereby influencing its bioactivity.10 In another prominent example it was found that collagen IV controls the bioavailability of the BMP homolog decapentaplegic (dpp) in drosophila by binding dpp directly or promoting the interaction with its receptor complex.11 Our previous work identified highly specific interactions between the prodomains (PDs) of BMPs and the extracellular glycoproteins fibrillin-1 and -2.6, 12 In addition, extracellular co-immunostaining of BMPs and fibrillin-1 suggested that fibrillin-1 microfibrils serve as storage platforms for these GFs.4-6 Fibrillins are 350 kDa glycoproteins with a conserved multidomain structure. In tissues, fibrillin-1 and -2 monomers are arranged into supramolecular, beads-on-a-string fibrillin microfibrils (FMF) with a diameter of 10-12 nm.13 The importance of FMF integrity becomes evident in congenital connective tissue disorders, caused by mutations in the fibrillin-1 and -2 coding genes (FBN1 and FBN2), the so-called fibrillinopathies.14 The fibrillinopathies represent disorders with similar, but also opposing clinical features affecting the musculoskeletal, cardiovascular, ocular, pulmonary, and dermal system.15 From these clinical features it can be concluded that fibrillins modulate GF-driven growth and differentiation processes in connective tissues. Also, analysis of fibrillin deficient mouse models corroborated this notion. An accelerated maturation of Fbn1-/- null osteoblasts was detected due to increased availability of non-ECM-targeted BMPs.16 Furthermore, genetic ablation of fibrillin-2 resulted in limb patterning and muscle maturation defects caused by dysregulated BMP signaling.17, 18 Meanwhile it is an established concept that FMF target and sequester TGF-β superfamily GFs and thereby regulate their bioavailability.19, 20 TGF-β depends on latent TGF-β-binding proteins (LTBPs) for sufficient secretion and targeting to the ECM in the form of a large latent TGF-β complex (LLC) formed by cysteine bridges of its PD to specialized 8-cysteine LTBP domains.21-23 Once secreted, the LLC is targeted to the ECM via LTBP binding to FMF and fibronectin.24-27 The mechanisms of TGF-β activation from FMF-bound pools, either by mechanical activation through integrin αv binding to the TGF-β PD (also known as latency-associated peptide: LAP),28-30 or upon metalloproteinase cleavage of LAP31 have been described. The concept that proteolytic cleavage of LAP leads to the activation of TGF-β-1 GF has been explored in in vitro experiments using MMP-2 and MMP-9.32 Also a combined activation mechanism has been described, where αvβ8 integrin binds to TGF-β-1 via LAP and enables MMP-14 cleavage of LAP and release of TGF-β-1 GF.33 Another TGF-β-1 GF activation model proposed that BMP-1 cleavage of LTBP-1 at N- and C-terminal sites releases truncated LLC from the ECM which is followed by a second MMP-2-mediated cleavage event of LAP.34 GDF-8 and GDF-11 are activated by BMP-1/Tolloid (TLD) metalloprotease-mediated cleavage of the PD.35, 36 Recently, tolloid-like protein 2 (TLL2) was also shown to cleave the PD of GDF-8 and thereby activate it from the latent state.37 Similarly, our previous in vitro results predicted a similar mechanism for BMP-10.12 Although much work has been undertaken to investigate the mechanisms of TGF-β activation, the regulatory pathways of other TGF-β superfamily members such as BMPs remain largely unknown. In contrast to TGF-β, most BMPs are secreted as bioactive PD-GF complexes (CPLXs) to the extracellular space.12, 38-41 Previously, we showed that the BMP-7 PD does not confer latency to the GF in solution, and that BMP type II receptors have free access to their binding sites on the GF. This suggestion arose from velocity sedimentation experiments in sucrose gradients showing that BMP type II receptor binding to the GF results in release of the free PD as a dimer from the CPLX.39 Recently, we could show that upon binding of the BMP-7 PD to the fibrillin-1N-terminal unique domain, a conformational change in the BMP-7 CPLX is induced which renders the GF inactive by locking the α2-helix of the PD in place, denying access to the BMP type II receptor site.42 However, little is known about how BMPs are released and activated once they are targeted to the ECM. Therefore, the aim of this study was to investigate new BMP activation mechanisms from FMF-targeted pools. 2 MATERIALS AND METHODS Ethics statement This study was carried out in strict accordance with German federal law on animal welfare, and the protocols were approved by the "Landesamt für Natur, Umwelt und Verbraucherschutz Nordrhein-Westfalen" (permit no. 84-02.04.2014.A397 for breeding and permit No. 84-02.05.40.14.115 for euthanasia). Antibodies Previously described monoclonal anti-BMP-7 PD antibodies mab2 and mab334 were kindly provided by Dr Lynn Sakai (Oregon Health and Science University). For western blots, mab33 was either used alone (1:1000 dilution), or in a mixture (1:1 molar ratio) together with mab2 (1:1000 dilution). The generation of polyclonal anti-fibrillin-1 antibody was previouly described.43 Polyclonal antibody against BMP-7 GF was purchased from PeproTech (#500-P198, Rocky Hill, NJ). Cell culture Primary murine skin fibroblasts were isolated from newborn mice.44 Primary dermal fibroblasts and HEK 293 cells were cultured in Dulbecco's Modified Eagle's medium (DMEM GlutaMAX, Invitrogen, Carlsbad, CA) supplemented with 10% of fetal bovine serum and penicillin/streptomycin. mRNA expression analysis via quantitative real-time PCR A total of 1 × 105 HEK 293 cells or primary murine skin fibroblasts cells were grown in 6-well plates prior to BMP GF stimulation at 100 ng/mL. After 24 hours of BMP stimulation, RNA extraction was performed by adding 1 mL of TRIzol Reagent (Thermo Fisher Scientific, Waltham, MA) according to the manufacturer's protocol. A subsequent sample purification step was included using the RNeasy kit (Qiagen, Venlo, The Netherlands), and residual DNA contamination was removed from each sample using the Turbo DNA-free kit (Ambion, Austin, TX). RNA samples were quantified by photospectrometry, and 1.0 μg of RNA per sample was reverse-transcribed using the Bio-Rad iScript cDNA synthesis kit (Bio-Rad, Hercules, CA). Quantitative PCR was performed using SensiFAST SYBR Hi-ROX Kit in 25 µL reaction volume (Meridian Bioscience, Cincinnati, OH). PCR was conducted with the StepOnePlus system (Applied Biosystems, Thermo Fisher Scientific). The standard annealing temperature of 60°C was chosen for the selected primer pairs (Mmp2F: CAAGTTCCCCGGCGATGTC, Mmp2R: TTCTGGTCAAGGTCACCTGTC; Mmp3F: ACATGGAGACTTTGTCCCTTTTG, Mmp3R: TTGGCTGAGTGGTAGAGTCCC; Mmp13F: TGTTTGCAGAGCACTACTTGAA, Mmp13R: CAGTCACCTCTAAGCCAAAGAAA). Analysis of data was performed using the 2−ΔΔCt method45 and quantitated relative to the murine Arbp or human GAPDH gene. Gene expression was normalized to BSA-treated control samples, which provided an arbitrary constant for comparative fold expression. Primer pairs for human MMP genes were purchased from Qiagen. Protein expression and purification BMP-7 CPLX was expressed and purified as described before.4 Briefly, the HEK 293 EBNA cell line stably transfected with N-terminally His6-tagged BMP-7 CPLX was kindly provided by Dr Lynn Sakai (Oregon Health and Science University). Cells were maintained in triple flasks, medium was collected, and affinity purified via nickel chelate affinity chromatography using the Ni-NTA resin (Cube Biotech, Germany). The highest purity fractions of BMP-7 CPLX were eluted with imidazole at a concentration of 50-250 mM. PDs of BMP-4, -5, -7, and -10 were expressed in E. coli and purified as previously described.6, 12 cDNAs encoding for BMP-7 PD mutant variants, PDs of human BMP-9 (K23-R319), human TGF-β-1 (L30-R279), and TGF-β-2 (L21-R330) were generated by gene synthesis (Genewiz, South Plainfield, NJ), cloned into the pET11a vector, overexpressed in E coli with a C-terminally placed His6-tag, and purified via Ni-NTA.6, 46 The murine proMMP-2, -7, -8, -9, and -13 (MMP2: NP_032636.1, aa A30-C662; MMP7: NP_034940.2 aa L21-L267, MMP8: NP_032637.3, aa F21-S465; MMP9: NP_038627.1, aa A20-P730; MMP13: NP_032633.1, L19-C472) were expressed and purified as described previously.47 MMP-3, MMP-12, GDF-8 PD, and BMP-7 GF were purchased from R&D Systems (Minneapolis, MN). Proteolytic cleavage assays MMPs were activated with 250 µM of amino-phenyl mercuric acetate (APMA) (Sigma-Aldrich, St. Louis, MO) for 2 hours at 37°C. For buffer exchange of solubilized BMP PDs to TC buffer (50 mM of Tris-HCl pH 7.5 and 1 mM of CaCl2), Amicon ultra 0.5 mL centrifugal filters (Merck Millipore, Burlington, MA) were used. A total of 10 nM of each activated MMP was incubated with 1 µM of BMP PD in 50 µL for 2 hours at 25°C. Fragments were analyzed by western blotting and silver staining. For Edman sequencing, 6 µg of BMP-7 PDs were incubated with 60 ng of MMP-2, MMP-3, or MMP-13. Fragments were separated by 10%-20% SDS-PAGE and transferred to a PVDF membrane. After staining with Ponceau S, the cleavage products were excised, and subjected to N-terminal Edman degradation performed by Proteome Factory AG (Berlin, Germany). MMP activity was assessed through incubation with a quenched Omni-MMP fluorogenic substrate (#BML-P126-0001, Enzo Life Sciences, Lörrach, Germany) in black 96-well plates (Thermo Fisher Scientific, Waltham, MA). Cleavage of the fluorescence resonance energy transfer (FRET)-based substrate (acceptor: MCA, donor: Dpa), led to fluorescence at 400 nm. For each assay, 2 µM of MMP substrate was incubated with 10 nM of the respective MMP in 100 µL of TC buffer for 30 minutes, at 25°C, followed by detection of fluorescence emission at 400 nm by an Infinite M1000 spectral photometer (Tecan, Switzerland). Circular dichroism spectroscopy BMP-7 PD variants were dialyzed into 5 mM of HClO4. CD spectra were recorded using a Jasco J-715 spectropolarimeter at 260-170 nm in a 0.1-mm path length quartz cell (Hellma, Germany) at 20°C. After subtraction of the buffer contribution, data were converted to Δϵ. ELISA and sandwich ELISA For ELISA, 100 ng/mL of BMP-7 CPLX was coated to microtiter plates (Nalge Nunc, Rochester, NY) in PBS overnight at 4°C. Wells were blocked with 5% nonfat dry milk/TBS for 1 hour at RT and washed three times with 0.025% TBS-tween, afterwards. Directly coated BMP-7 CPLX was incubated with MMPs at a molar ratio of 1:100 (MMP:BMP-7 CPLX) for 2 hours at RT in TC buffer. For sandwich ELISA detection, BMP-7 CPLX after MMP-13 cleavage was transferred to anti-BMP-7 GF antibody coated wells (10 µg/mL, PeproTech) and incubated for 1 hour. Wells were washed three times with 0.025% TBS-tween and incubated with detection antibody against BMP-7 PD (mab33 at 1:1000 dilution) in 2.5% nonfat dry milk/TBS for 2 hours at RT, followed by 1 hour incubation of HRP-conjugated anti-rabbit antibody in 2.5% nonfat dry milk/TBS at RT. Subsequently, wells were washed three times with TBS-tween, and incubated with 50 µL of 1-Step Ultra TMB-ELISA substrate solution for signal development (Thermo Fisher Scientific, Waltham, MA). OD was read at 450 nm using a Sunrise microplate reader (Tecan). MMP-13 cleavage assays on solid phase For the generation of an assembled ECM fiber network, 1 × 106 mouse skin fibroblasts were seeded on 6-well plates and cultivated for 4 days, followed by cell removal using deoxycholate.48 In brief, cell cultures were washed once with PBS and then, treated twice with 0.5% sodium deoxycholate in 10 mM of Tris-HCl buffer, pH 8.0, at 0°C for 10 minutes. The plates were then allowed to dry overnight at RT. Subsequently, wells were blocked in 5% BSA followed by incubation with BMP-7 CPLX. To assess co-localization between added BMP-7 CPLX and fibrillin-1 fibers by immunofluorescence, cells were grown on 24-well plates. To demonstrate a direct interaction between added BMP-7 CPLX and assembled ECM fibers by ELISA-style solid phase interaction assay, mouse fibroblasts were grown on 96-well plates. For this, 100 ng/mL of BMP-7 CPLX was titrated onto ECM-coated dishes following a 1:2 serial dilution in TBS buffer containing 1% of BSA at RT for 2 hours. BMP-7 CPLX immobilized to ECM fibers was submitted to MMP-13 cleavage (50 ng/mL) in 1 mL of TC buffer for 2 hours at 37°C. Afterwards, the supernatant was collected and subjected to TCA precipitation for western blot analysis or lyophilized to be subjected to BMP bioactivity assays. BMP bioactivity assay Supernatant from 6-well plates containing GF released from ECM-targeted BMP-7 CPLX after MMP-13 cleavage was collected and dialyzed in mini dialysis tubes with a molecular weight cut-off of 2 kDa into 100 mM acetic acid overnight at 4°C. After dialysis, samples were shock-frozen in liquid nitrogen and lyophilized overnight. Subsequently, samples were resuspended in 10 µL of 4 mM HCl and administered to BMP bioactivity assays. To measure BMP bioactivity murine C2C12 myoblasts were utilized as reporter cell line. For each measurement, 3.5 × 104 cells/well were seeded onto 96-well plates. Stimulation was performed in eight wells per concentration in triplicates. Two or three microliters of the obtained supernatant after MMP-13 cleavage of ECM-bound BMP-7 CPLX and 2 or 3 µL of the supernatant without MMP-13 incubation were added for C2C12 cell stimulation. After 5 hours, the total mRNA content of cells was harvested, reverse-transcribed, and subjected to qPCR to measure the mRNA levels of BMP response gene Id3 (inhibitor of differentiation 3).39 Id3 mRNA levels were normalized to the mRNA expression of ARBP ("acidic ribosomal binding protein") which served as housekeeping gene. A total of 10 ng/mL of BMP-7 GF (R&D systems) was added to the medium as positive control, and incubation of cells with 0.1% of BSA served as untreated negative control. Transmission electron microscopy (TEM) and single particle analysis BMP-7 CPLX alone, after dialysis into 1 M urea, or after incubation with MMP-13 for 2 hours was negatively stained as described previously.49 BMP-7 CPLX alone and cleaved with MMP-13 data were recorded at ×30 000 magnification using a FEI Tecnai 12 twin TEM at 120 kV using a Tietz TVIPs F214A CCD camera. Images were recorded with a 1-s exposure at defocus values of −0.5 to −1.6 μm at 1.5 Å/pixel (Figure 9C). BMP-7 CPLX after dialysis into 1 M urea data were collected on a FEI Tecnai G2 Polara TEM operating at 300 kV equipped with a Gatan K2 summit direct detector. Images were recorded at ×23 000 magnification with a 1-s exposure in integrating mode at defocus values of −0.5 to −1.6 μm at 1.67 Å/pixel. Single particle analysis was performed using Relion.50, 51 Particles were selected by a combination of manual and automated picking. The total number of particles selected for BMP-7 CPLX either alone, after MMP-13 incubation, or in the presence of 1 M urea was approximately 900, 6770, or 9600, respectively. Each data set was subjected to two-dimensional classification. Dynamic light scattering BMP-7 CPLX was dialyzed overnight into TC buffer and cleaved with MMP-13 as described above. DLS measurements of the cleaved and non-cleaved control BMP-7 CPLX were then taken using a Zetasizer Nano-S (Malvern, Herfordshire, UK) at a controlled temperature of 25°C. Molecular docking experiments The generation of the BMP-7 CPLX closed-ring shape model (Figure 6), based on the TGF-β-1 LAP crystal structure (Protein Data Bank code 3RJR) as a template, with MODELLER52 in UCSF Chimera53 was as described.42 In this model, a break in the peptide chain was introduced at residue Pro80 to allow the PD to be rotated into an open conformation without moving the N-terminal region. Next, using the Chimera software a peptide bond with a phi torsional angle of −60° was introduced at the exact same position in order to re-join the polypeptide chains without moving the N-terminal PD region. The BMP-9 CPLX structure (Protein Data Bank code 4YCI) and the BMP-7 CPLX EM map were used to guide rotation of the PD into an open conformation. To obtain a structural model of BMP-7 PD in the open V-shape conformation, BMP-7 PD was modeled on the proactivin CPLX structure (Protein Data Bank code 5HLZ) using Swiss-model. To gain structural insight into the MMP-13 cleavage mechanism, BMP-7 PD in the open conformation or the closed-ring BMP-7 CPLX, inputted as the "receptors" were docked to the crystal structure of the activated MMP-13 (Protein Data Bank code 4fu4) inputted as the "ligand" in the ClusPro2.0 server54 after deleting the structure of the co-crystallized peptide using UCSF chimera software. Subsequently, 30 molecular docking models were screened for each in silico experiment and models 21 and 16 were selected for closed and open BMP-7 PD conformations, respectively, due to excellent alignment of the MMP-13 catalytic site (His222, His226, and His232) to the prime region 83MLD85 of the cleavage site. To obtain a theoretical model of MMP-13 cleaving the BMP-7 CPLX closed-ring, model 21 was structurally aligned to each BMP-7 PD monomer of the closed BMP-7 CPLX model at the 83MLD85 site. Images were taken both in ribbon and surface representations. To pinpoint the exact molecular requirements for cleavage in the open BMP-7 PD conformation, model 16 was superimposed to the activated MMP-13 structure (Protein Data Bank code 4fu4) and the positioning of the co-crystallized peptide was compared to the 83MLD85 site of BMP-7 PD. Ions and metals were visualized using the 4fu4 template and images were taken in ribbon representation. To generate the open V-shape BMP-7 CPLX model, BMP-7 PD was assembled into a dimer using Swiss-model and the proactivin CPLX was structurally aligned to the BMP-7 PD dimer. Next, the monomers of the BMP-7 GF crystal structure (Protein Data Bank code 1LX5) were structurally aligned to the monomers of the proactivin GF after deleting the structure of the ActRII extracellular domain using UCSF chimera. To understand how MMP-13 is cleaving the open V-shape BMP-7 CPLX, model 16 was structurally aligned to one BMP-7 PD of the CPLX and images were taken both in ribbon and surface representations. Statistical analysis Data are expressed as mean ± SD. Statistical analyses were performed using GraphPad Prism software and the significance of differences between groups was determined by applying an unpaired two-tailed Student's test. Values of P ≤ .05 were considered significant. 3 RESULTS BMP-7 GF stimulates metalloproteinase activity leading to specific cleavage of BMP-7 PD Western blot analysis of HEK 293 cells transiently transfected with full length BMP-7 cDNA encoding for the entire BMP-7 PD-GF CPLX showed two bands for BMP-7 PD after 3 days of culture. After 1 day of transfection, only the expected size of full length BMP-7 PD at approximately 37 kDa was detected, while at day 3 an additional band of about 25 kDa could be observed (Figure 1A). This suggested that after secretion a specific proteolytic event
Referência(s)