Article pubs.acs.org/molecularpharmaceutics
Secreted Matrix Metalloproteinase‑9 of Proliferating Smooth Muscle Cells as a Trigger for Drug Release from Stent Surface Polymers in Coronary Arteries Daniel G. Gliesche,† Janine Hussner,† Dominik Witzigmann,‡ Fabiola Porta,‡ Timo Glatter,§ Alexander Schmidt,§ Jörg Huwyler,‡ and Henriette E. Meyer zu Schwabedissen*,† †
Biopharmacy, Department of Pharmaceutical Sciences, University of Basel, 4056 Basel, Switzerland Pharmaceutical Technology, Department of Pharmaceutical Sciences, University of Basel, Basel 4056, Switzerland § Proteomics Core Facility, Biozentrum, University of Basel, Basel 4056, Switzerland ‡
S Supporting Information *
ABSTRACT: Cardiovascular diseases are the leading causes of death in industrialized countries. Atherosclerotic coronary arteries are commonly treated with percutaneous transluminal coronary intervention followed by stent deployment. This treatment has significantly improved the clinical outcome. However, triggered vascular smooth muscle cell (SMC) proliferation leads to in-stent restenosis in bare metal stents. In addition, stent thrombosis is a severe side effect of drug eluting stents due to inhibition of endothelialization. The aim of this study was to develop and test a stent surface polymer, where cytotoxic drugs are covalently conjugated to the surface and released by proteases selectively secreted by proliferating smooth muscle cells. Resting and proliferating human coronary artery smooth muscle cells (HCASMC) and endothelial cells (HCAEC) were screened to identify an enzyme exclusively released by proliferating HCASMC. Expression analyses and enzyme activity assays verified selective and exclusive activity of the matrix metalloproteinase-9 (MMP-9) in proliferating HCASMC. The principle of drug release exclusively triggered by proliferating HCASMC was tested using the biodegradable stent surface polymer poly-L-lactic acid (PLLA) and the MMP-9 cleavable peptide linkers named SRL and AVR. The specific peptide cleavage by MMP-9 was verified by attachment of the model compound fluorescein. Fluorescein release was observed in the presence of MMP-9 secreting HCASMC but not of proliferating HCAEC. Our findings suggest that cytotoxic drug conjugated polymers can be designed to selectively release the attached compound triggered by MMP-9 secreting smooth muscle cells. This novel concept may be beneficial for stent endothelialization thereby reducing the risk of restenosis and thrombosis. KEYWORDS: atherosclerosis, drug eluting stent, bioresorbable scaffold, MMP-9, stenosis, HCASMC, coronary arteries, LC−MS, triggered release
■
INTRODUCTION
cells, which is triggered by vascular overstretching or injuries leading to in-stent restenosis.7,8 This observation resulted in the development of drug eluting stents (DES) coated with cytotoxic compounds such as paclitaxel (TAXUS, Boston Scientific) or sirolimus (CYPHER, Cordis Corporation).9 Both have been the first DES in clinical use. The inhibition of smooth muscle cell proliferation has significantly improved the clinical outcome after stent implantation.10−13 However, the unspecific cytotoxicity of the released drugs leads to abrogated neointima formation and inhibition of endothelialization of the stent struts. This has
Fatal vascular events such as myocardial infarction or stroke are the most severe consequences of atherosclerosis.1 One of the treatments of vascular stenosis is percutaneous transluminal coronary angioplasty. This includes vessel dilatation followed by stent deployment. The first devices used in coronary arteries were bare metal stents (BMS). The development of BMS was based on the assumption that a metal structure would mechanically prevent the major adverse events of acute elastic recoil and restenosis after balloon angioplasty.2 Even if BMS have significantly improved the outcome of patients after stent implantation,3,4 it has been shown that BMS are associated with the adverse event of neointima formation. This process is characterized by cell migration, cell division, and matrix secretion.5,6 The pathophysiological basis of neointima formation is the enhanced proliferation of smooth muscle © XXXX American Chemical Society
Received: January 13, 2016 Revised: May 23, 2016 Accepted: May 30, 2016
A
DOI: 10.1021/acs.molpharmaceut.6b00033 Mol. Pharmaceutics XXXX, XXX, XXX−XXX
Article
Molecular Pharmaceutics
using the method described by Petersen et al. with slight modifications.26 PLLA films were formed by pouring a 40 mL of chloroform solution containing 1 g of PLLA (Resomer L210, MW 280 000 g/mol, Boehringer Ingelheim, Ingelheim, Germany) into a glass Petri dish (⌀ = 9 cm). The Petri dish was incubated at room temperature (rt) under a ventilated hood until the solvent was totally evaporated. Afterwards, PLLA uniform films were washed with distilled water for 5 days and dried in a vacuum chamber (3 d, rt, 50 mbar). The polymer films were cut into circular 6 mm pieces. For integration of amino groups, ten PLLA pieces were incubated with 3 mL of hexamethylenediamine/2-propanol (10% w/w) stirring at 37 °C for 1 h, then washed with distilled water three times for 30 min at rt, and dried in a vacuum chamber (rt, 50 mbar) overnight. Finally, the amino group transfer was analyzed using the NHS-fluorescein assay (for details see Supporting Information). Attachment of Peptides to PLLA Polymer Films. For membrane functionalization, a total amount of 1.171 nmol/ mm2 peptide was used. To link the carboxyl-terminal end of commercially synthesized and amino-terminal fluorenylmethoxycarbonyl (Fmoc) protected peptides (Biomatik, Wilmington, Delaware, USA) to the transferred amino groups, ten activated polymers were placed in 10 mL of dimethylformamide (DMF) containing 660 mmol of the peptide, 0.857 mL of N,Ndiisopropylethylamine (DIPEA), and 1.13 g of 2-(6-chloro-1Hbenzotriazol-1-yl)-1,1,3,3-tetramethyluronium-hexafluorophosphate (HCTU) stirring at rt overnight. The polymers were washed three times with DMF and 2-propanol. Subsequently, free amino residues of the PLLA polymers were acetylated using 10 mL of DMF containing 160 μL of acetic anhydride and 275 μL of DIPEA at rt for 1.5 h. Then polymers were again washed with three series of 10 mL of DMF and 2-propanol. Samples were stored at −20 °C in DMF. Modification of Peptides Linked to PLLA Polymer Films. For cleavage of the Fmoc protection group, samples were treated with 10 mL of DMF containing 16% piperidine at rt for 1.5 h, stirring in the dark, and then washed three times with 10 mL of DMF or 2-propanol. Subsequently, polymers were incubated with NHS-fluorescein (Pierce, Fisher Scientific AG, Reinach, Switzerland) dissolved in 20 mM HEPES buffer pH 8.0 (100 μg/mL) at rt for 1.5 h in the dark (see also manufacturer’s protocol). Finally, modified polymer films were washed three times for 10 min with 20 mM HEPES and then with PBS. Quantification of Fluorescein Release from PLLACoupled Peptides. For the assessment of fluorescein release from PLLA-coupled peptides, each polymer film was placed in a 1.5 mL reaction tube containing 500 μL of PBS supplemented with 10 μg of secretome, 5 ng of activated enzyme (see below), or solvent as control and incubated at 37 °C for 24 h. Fluorescence of released fluorescein (ex λ = 460 nm, em λ = 518 nm) was detected using the NanoQuant Plate and microplate reader Infinite M200 PRO (Tecan, Männedorf, Switzerland) at the beginning (0 h background) and after 24 h incubation. Purified and APMA activated MMP-2 or MMP-9, both obtained from AnaSpec (ANAWA, Wangen, Switzerland), served as control for fluorescein release, and the supernatant was not treated. After subtraction of background, fluorescence data was normalized to that determined after incubation with the solvent control. Qualitative analysis of the PLLA surfaces and their fluorescence was performed using laser scanning
been shown to be associated with a different adverse event, namely, the late and the very late stent thrombosis.14,15 A significantly delayed or partial healing in DES is multifactorial and characterized by fibrin disposition and incomplete reendothelialization.16 From a pharmacological perspective there are different approaches to enhance specificity of cytotoxic compounds. A first approach would be the identification of drug targets, which are specifically expressed in proliferating smooth muscle cells (SMCs). This potential target modulating cell proliferation could be used to decrease the tendency of neointima formation. Second, cellular differences in pharmacokinetics could be used to enhance accumulation in SMCs, thereby increasing the specificity of a drug with an intracellular drug target. In this regard, we recently reported a proof-of-concept study, showing that adenoviral driven overexpression of an uptake transporter changes pharmacokinetics in the microcompartment of the vascular wall, thereby enhancing pharmacological efficacy of substrate drugs in SMCs.17 A third option would be a release of cytotoxic compounds triggered by proliferation of SMCs. This concept is inspired by the so-called prodrug concept18 with bioactivation in close vicinity of the targeted organ or cell. One example of this concept is the small molecule acyclovir, which exerts its drug effect after phosphorylation mediated by the viral thymidine kinase. This enzyme is only present in herpes simplex virus infected cells.19 In the perspective to implant a drug delivery device in the close vicinity of the targeted cells, namely, the coronary artery smooth muscle cells, bioactivation not only would mean chemical modification of a small molecule by metabolism,20 but also would include the release from the stent surface.21 Based on the current pathophysiological understanding of late and very late stent thrombosis,14,15 cytotoxic compounds should be released during SMC proliferation to prevent neointima formation.13 The impact of a cytotoxic compound on endothelial cell proliferation and migration would be limited by reducing the exposure to the area and period of enhanced SMC proliferation. The increased proliferation of vascular SMCs with neointima formation and the re-endothelialization of the injured vessel area are two processes that proceed at the same time. However, postmortem data revealed that up to 18 months after stent implantation the vascular lesions do primarily consist of confluent SMCs lacking full endothelial coverage, thereby suggesting that SMC proliferation22,23 is an early reaction followed by the process of endothelialization. This is supported by findings in large animal studies.24,25 To provide basis for the latter concept, we focused on the identification of an extracellular proteolytic enzyme specifically secreted by proliferating SMCs and the development of a specific drug release concept for this enzyme. Therefore, we isolated all secreted proteins from a cell type that is now designated as secretome. We characterized the functionality of the identified matrix metalloproteinase for proteolytic drug release from a stent surface polymer modified with two different peptides. Both peptide linkers used in the herein reported study were previously reported as MMP-9 specific substrates. Specificity of the drug release from the polymer triggered by the enzyme or by proliferating SMCs was confirmed.
■
EXPERIMENTAL SECTION Manufacturing of PLLA Polymer Films with Amino Residues. Poly-L-lactic acid (PLLA) polymers were produced B
DOI: 10.1021/acs.molpharmaceut.6b00033 Mol. Pharmaceutics XXXX, XXX, XXX−XXX
Article
Molecular Pharmaceutics microscopy. Detailed description of fluorescence microscopy is provided in the Supporting Information. Cell Culture. Primary human coronary artery smooth muscle cells (HCASMC) and endothelial cells (HCAEC) (PromoCell GmbH, Heidelberg, Germany) were cultured at 37 °C in a humidified atmosphere with 5% CO 2 using commercially available optimized culture media, i.e., Smooth Muscle Cell Growth Medium 2 and Endothelial Cell Growth Medium MV (PromoCell). HCASMC and HCAEC were seeded at a density of 1.5 × 104 cells/cm2 24 h prior to the start of the respective experiment in the indicated culture medium. Cell Proliferation Assay. The method is described in the Supporting Information. Enrichment of Secreted Proteins from Proliferating Cells. HCASMCs and HCAECs were grown in growth media (i.e., Endothelial Cell Growth Medium MV and Smooth Muscle Cell Growth Medium 2, PromoCell) and seeded on three 10 cm2 dishes each individual. After cultivation for 24 h in the respective basal medium (i.e., Endothelial Cell Basal Medium MV and Smooth Muscle Cell Basal Medium 2, PromoCell GmbH, Heidelberg, Germany) supplemented with FCS (0.1%), they were restimulated with growth medium for 24 h. Cells were washed twice with PBS, and medium was changed to 4 mL of serum-free medium. After 17 h incubation the cell supernatant of three dishes was pooled to enrich the secreted proteins. In detail, proteins of the supernatant were precipitated by addition of 10% trichloroacetic acid to a total volume of 12 mL of culture supernatant. After incubation on ice for 15 min, the samples were centrifuged at 4 °C (15000g for 15 min). Protein pellets were washed twice with ice-cold acetone and dried at rt for 5 min in a Concentrator plus (Vaudaux-Eppendorf, Basel, Switzerland) before the samples were diluted in 100 μL of 6 M urea. Protein concentration was determined using a Bradford assay as described by the manufacturer (Pierce, Fisher Scientific AG). Mass Spectrometric Analysis. For identification of secreted proteins by quantitative liquid chromatography− mass spectrometry (LC−MS), samples were prepared and analyzed as previously described by Glatter et al.27 For detailed information on protein preparation and LC−MS analysis see Supporting Information. Data were quantitatively analyzed using the method NSAF (normalized abundance factor).28 In detail, the peptide scores of every protein, presented in Table S1, correspond to the number of peptides identified per protein normalized to all detected peptides in one sample. The spectrum report and the results of the database search are provided in Table S2. Western Blot Analysis. Samples were separated by 10% SDS−PAGE and blotted onto a nitrocellulose membrane. After visualization of protein transfer by Ponceau S (Sigma-Aldrich, Ahrlesheim, Switzerland), the membranes were incubated with one of the following primary antibodies overnight at 4 °C, namely, anti-MMP-1 (ab38929), anti-MMP-2 (ab79781), antiMMP-9 (ab38898), or anti-MMP-10 (ab38930) (Abcam, Cambridge, U.K.), and then incubated (rt, 1 h) with the respective peroxidase-labeled secondary antibody (Bio-Rad Laboratories, Cressier, Switzerland). Protein expression was quantified using the ChemiDoc MP imaging system (Bio-Rad). Due to the fact that MMP-9 appears as monomer, dimer, or multimers and as glycosylated protein, we selected the 82 kDa (monomer) for protein quantification.29 Levels of extracellular proteins detected in the supernatant were normalized to that of Ponceau S as there is a lack of a known extracellular protein
which could be used as loading control and therefore for normalization. Coomassie Blue Staining. For visualization of proteins enriched from cell culture supernatant, samples were separated by 10% SDS−PAGE, fixed in 40% ethanol with 10% acetic acid, and stained using the commercially available Roti-Blue Colloidal Coomassie solution (Roth AG, Arlesheim, Switzerland). After 5 h the gels were washed several times with 25% methanol. Staining was digitalized using the ChemiDoc MP imaging system (Bio-Rad). In Gel Zymography. Zymography was performed to visualize gelatinolytic activity in the secretome of HCAEC and HCASMC after separation by nonreducing SDS−PAGE. The method was performed as described previously.30,31 Ten micrograms of secretome were separated by 7.5% SDS−PAGE copolymerized with 0.7 mg/mL gelatin (300 Bloom, SigmaAldrich). After replacing SDS by Triton X-100 in a washing step, the gel with refolded enzymes was incubated at 37 °C for 36 h for degradation of the copolymerized gelatin. The proteolytic zones were visualized by Coomassie blue staining; the resulting bands were quantified using the ChemiDoc MP imaging system (Bio-Rad). Quantification of MMP-2 and MMP-9 Enzyme Activity. Proteolytic activity of MMPs was quantified using 10 μg of secretome and the fluorometric SensoLyte Plus 520 MMP-2 or MMP-9 Assay Kit from AnaSpec (ANAWA), following the manufacturers’ protocol. Turnover of a fluorescence resonance energy transfer peptide substrate was measured at the beginning and after 24 h using the microplate reader Infinite M200 PRO (Tecan) (ex λ = 490 nm, em λ = 520 nm). Inactive proenzymes in the secretome samples were activated with 1 mM 4-aminophenylmercuric acetate (APMA) for 1 h (MMP-2) or 2 h (MMP-9) at 37 °C to determine total enzyme. Purified proenzymes obtained from the manufacturer were activated with APMA and served as standard. For comparison of active MMPs in the supernatant of proliferating and resting cells, the enriched secretome from cultured cells was not treated with APMA. Quantitative Real-Time PCR. For quantification of mRNA expression, total RNA was isolated followed by reverse transcription. The cDNA of HCASMC and HCAEC was used for quantitative analysis of MMP-2 and MMP-9 gene expression by specific SYBR green primer pairs. The analysis was realized using a ViiA 7 Real-Time PCR system (Fisher Scientific AG). 18S rRNA gene expression served as control. Gene expression was normalized to proliferating HCASMC. See Supporting Information for details. Statistical Analysis. Data are presented as means ± SD from at least three independent experiments performed in technical duplicates or triplicates (GraphPad PRISM 6.0, La Jolla, CA, USA). Single comparison analysis was performed using Student’s t test or if indicated one-way ANOVA for multiple comparisons. The p-values
