Self-Assembly of Elastin-Based Peptides into the ECM: the Importance

432

Biomacromolecules 2011, 12, 432–440

Self-Assembly of Elastin-Based Peptides into the ECM: the Importance of Integrins and the Elastin Binding Protein in Elastic Fiber Assembly Dhaval Patel, Rohan Menon, and Lakeshia J. Taite* School of Chemical and Biomolecular Engineering, Georgia Institute of Technology, 311 Ferst Drive NW, Atlanta, Georgia 30332-0100, United States Received October 12, 2010; Revised Manuscript Received December 8, 2010

The formation of a suitable extracellular matrix (ECM) that promotes cell adhesion, organization, and proliferation is essential within biomaterial scaffolds for tissue engineering applications. In this work, short elastin mimetic peptide sequences, EM-19 and EM-23, were engineered to mimic the active motifs of human elastin in hopes that they can stimulate ECM development in synthetic polymer scaffolds. Each peptide was incubated with human aortic smooth muscle cells (SMCs) and elastin and desmosine production were quantified after 48 h. EM-19 inhibited elastin production through competitive binding phenomena with the elastin binding protein (EBP), whereas EM-23, which contains an RGDS domain, induces recovery of elastin production at higher concentrations, alluding to a higher binding affinity for the integrins than for the EBP and the involvement of integrins in elastin production. Colocalization of each peptide with the elastin matrix was confirmed using immunofluorescent techniques. Our data suggest that with appropriate cell-binding motifs, we can simulate the cross-linking of tropoelastin into the developing elastin matrix using short peptide sequences. The potential for increased cell adhesion and the incorporation of elastin chains into tissue engineering scaffolds make these peptides attractive bioactive moieties that can easily be incorporated into synthetic biomaterials to induce ECM formation.

Introduction The sustainable growth and maintenance of normal tissues require specific cellular interactions in a microenvironment that incorporates the appropriate growth factors and adhesion molecules and allows for the deposition of a complex extracellular matrix (ECM).1 The ECM is comprised of several cellsecreted proteins that serve to regulate the aforementioned interactions for maintaining sustainable cell growth.2 Elastin, the ECM component prominent in connective tissues such as the lungs, arteries, and skin, is a highly cross-linked, insoluble biological polymer that is essential to proper tissue function.3-6 The largely hydrophobic 70 kDa protein makes up about 50% of the ECM dry weight of an artery7-9 and is a key factor in promoting smooth muscle cell (SMC) proliferation and migration throughout the medial layer of the arterial wall.10-12 The synthesis of elastin is a multistep process in which soluble tropoelastin is synthesized intracellularly and cross-linked into elastic fibers.13,14 Though there is still some debate over the exact mechanisms of tropoelastin transport,15,16 it is clear that the transmembrane receptor known as the elastin binding protein (EBP) plays a crucial role in chaperoning the protein to extracellular microfibrils for elastic fiber assembly before being recycled to bind additional intracellular tropoelastin (Figure 1).17-19 Moreover, recent work has shown that the C-terminal GRKRK motif of tropoelastin also interacts with the Rvβ3 integrin in cultures of fibroblasts.20 Though the exact role of this interaction in the development of elastic fibers is yet unknown, it is likely that cell adhesion to elastin fibers is mediated through integrin binding. The EBP forms a complex with intrinsic membrane proteins sialidase and carboxypeptidase A to form a transmembrane * To whom correspondence should be addressed. Tel.: +1 404-894-8795. Fax: 404-894-2866. E-mail: [email protected].

Figure 1. Elastin production involves a sequence of events involving the secretion of tropoelastin, which binds the elastin binding protein (EBP) on cell surfaces. Tropoelastin aggregates are then transported to adjacent microfibrils involved in elastin fiber assembly, and the EBP is recycled. The Rvβ3 integrin also binds the C-terminal region of tropoelastin, playing a yet unknown role in the synthesis of mature elastin.

elastin receptor20-22 and has high affinity for elastin’s hexapeptide repeat sequence, Val-Gly-Val-Ala-Pro-Gly (VGVAPG).23,24 The interaction of VGVAPG and the EBP has been shown to influence chemotaxis16 and vasorelaxation25 and inhibit platelet aggregation.26 Intracellularly, the EBP binds tropoelastin and chaperones the elastin precursor to the extracellular space; tropoelastin is only released from the EBP once galactosugars on microfibrils in the extracellular space bind to the EBP,27,28 and subsequently, lysine residues on tropoelastin undergo spontaneous oxidative deamination catalyzed by the enzyme lysyl oxidase (LOX) to yield either desmosine or isodemsosine cross-link sites, forming mature elastin.29,30 Along with providing flexibility and contractility, developed elastic fibers also guide further elastin deposition in vivo.17,31,32

10.1021/bm101214f  2011 American Chemical Society Published on Web 12/30/2010

Self-Assembly of Elastin-Based Peptides into the ECM

It is only after cross-linking that elastin achieves high elasticity, stability, and tensile strength.33-35,6 Unfortunately, the overwhelmingly hydrophobic nature of the protein, coupled with its insolubility, limits the use of elastin as a biomaterial.17,36,37 As an alternative approach, several researchers have investigated engineering artificial elastin mimetic proteins as potential substitutes for engineering materials for vascular prostheses.38-43 These materials are designed to exploit the active domains of elastin assembly, however, engineering protein expression and purification can be an arduous, costly, and time-consuming process. Herein, we have engineered short synthetic peptides that employ the cell adhesion and cross-linking motifs that are found in natural elastin to induce ECM deposition and crosslinking within a synthetic construct for the engineering of vascular tissues. The peptides include the hydrophobic VGVAPG domain and a hydrophilic domain consisting of lysine (Lys, K) and alanine (Ala, A) residues, which facilitates cross-linking between adjacent peptides and with secreted tropoelastin from SMCs seeded within the scaffold. Our study also includes the fibronectin-derived cell adhesion sequence Arg-Gly-Asp-Ser (RGDS), which binds with high affinity to cell surface integrins,44,45 in order to further enhance adhesion and allow investigation of the role that integrin binding might play in elastin production and organization. The binding of RGDS to integrins has previously been shown to trigger chemotactic responses, including cell adhesion and migration, for SMCs in vitro,46,47 but the role of integrins in elastin binding and deposition is not known. Cross-linking of the peptide within a synthetic construct will create a network of elastin fibers physically linked to the synthetic polymer scaffold, engineering the appropriate biochemical and biomechanical cues for vascular cell growth. Our chosen peptides show an intriguing ability to guide elastin deposition and desmosine formation in human SMCs. There is evidence to support competition between our peptides bearing the VGVAPG sequence with secreted tropoelastin, but this is tempered by the addition of RGDS into the peptide. Peptides bearing a scrambled version of VGVAPG and an unmodified RGDS were able to increase elastin content in culture, suggesting that integrins play a large role in tropoelastin production and the cross-linking of elastin fibers, as recently suggested by Weiss and colleagues.20 Our results are reinforced in cultures maintained in R-lactose, which is known to disrupt the EBP and blocks elastin production. Visualization of fluorescently tagged peptides within the elastin matrix by immunostaining with a fluorescently tagged antibody specific to the VGVAPG sequence also allowed us to view elastin within the ECM while simultaneously identifying any of the synthetic peptide that has been cross-linked into the ECM. These data provide further insight into the mechanisms required for elastin production and organization in vascular cell cultures and present an opportunity to incorporate cell-secreted elastin into synthetic biomaterials for tissue engineering applications.

Materials and Methods Peptide Synthesis. All materials and reagents for peptide synthesis were purchased from Advanced Automated Peptide Protein Technologies (AAPPTec; Louisville, KY). Six different peptides were synthesized using standard O-Benzotriazole-N,N,N′,N′-tetramethyl-uroniumhexafluoro-phosphate (HBTU) coupling and fluorenylmethyloxycarbonyl (Fmoc) protection chemistry on an Apex 396 peptide synthesizer (AAPPTec, Louisville, KY). A 23 amino acid sequence (EM-23; AAKAAKVGVAPGRGDSAAKAAKK), containing two cell adhesion motifs (VGVAPG and RGDS), as well as Lys rich cross-linking

Biomacromolecules, Vol. 12, No. 2, 2011

433

domains, was synthesized along with a 23 amino acid control peptide in which the VAPG sequence is scrambled and the RGDS sequence is left intact (EM-23S; AAKAAKVGVPAGRGDSAAKAAKK). A 19 amino acid sequence (EM-19; AAKAAKVGVAPGAAKAAKK) and a 19 amino acid peptide in which the VAPG sequence is scrambled (EM-19S; AAKAAKVGVPAGAAKAAKK) were synthesized, omitting the RGDS motif completely, to observe the importance of integrin binding. Both 23 and 19 amino acid sequences containing Gly residues in place of Lys were used to observe the importance of the hydrophilic cross-linking domain (EM-23G; AAGAAGVGVAPGRGDSAAGAAGG; EM-19G, AAGAAGVGVAPGAGAAGG). Peptides were dialyzed using cellulose ester membrane tubing (MWCO 1000; Spectrum laboratories, Rancho Dominguez, CA). The dialyzed peptides were then frozen and lyophilized overnight. Peptides were characterized using mass spectrometry and gel permeation chromatography (GPC). RGDS was purchased (American Peptide Company, Sunnyvale, CA) and used as received. Circular dichroism (CD) measurements were performed by Alliance Protein Laboratories (Camarillo, CA) to confirm secondary peptide structures and ensure that any peptide incorporated into elastic fibers will not cause significant changes in the overall structural conformation of the fibers. Peptides were dissolved in water at 0.2 mg/mL. CD measurements were carried out on a Jasco J-715 spectropolarimeter at controlled temperature using a 0.1 cm cell. The temperature was controlled by a PTC-348WI temperature controller and Peltier cell holder. Solvent spectrum was subtracted from the sample spectrum and the spectrum was converted to the mean residue ellipticity (CD intensity per amino acid). Acrylate Poly(ethylene glycol) Synthesis. Polyethylene glycol diacrylate (PEG-DA) was synthesized by dissolving 24 g dry PEG (MW: 10000; Fluka, Milwaukee, WI) in 20 mL of anhydrous dichloromethane (DCM) with an equimolar amount of triethylamine and 0.869 g acryloyl chloride (Lancaster Synthesis, Windham, NH) added dropwise. The mixture was stirred under argon for 24 h, washed with 2 M K2CO3, and separated into aqueous and DCM phases to remove HCl. The DCM phase was dried with anhydrous MgSO4 (VWR, West Chester, PA), and the PEG-DA was then precipitated in diethyl ether, filtered, and dried under vacuum at room temperature overnight. The resulting polymer was dialyzed overnight against DI water with 5 kDa MWCO cellulose ester tubing (Spectrum Laboratories, Rancho Dominguez, CA) to remove any residual salts and impurities, dissolved in chloroform-d, and characterized via proton NMR to determine the extent of acrylation. Peptide sequences were conjugated to PEG monoacrylate by reaction with acryloyl- PEG-succinimidyl valerate (PEG-SVA, MW 3400; Laysan Bio Inc., Arab, AL) in 50 mM sodium bicarbonate (pH 8.5) at a 1:1 molar ratio for 2 h. The mixture then was dialyzed (3500 MWCO dialysis cassettes Pierce Biotechnology, Rockford, IL) against deionized water for 2 h, lyophilized, and stored at -80 °C. Ultraviolet-visible (UV-vis) spectroscopy and gel permeation chromatography with UV and evaporative light scattering detectors (Polymer Laboratories, Amherst, MA) were used to determine the coupling efficiency. Cell Culture and Maintenance. Human aortic smooth muscle cells (SMCs, passages 3-5; Invitrogen Inc., Carlsbad, CA) were used in this work. SMCs were maintained in Medium 231 (Invitrogen Inc., Carlsbad, CA) supplemented with smooth muscle cell growth supplement (SMGS, Invitrogen Inc., Carlsbad, CA), 2 mM L-glutamine, 1 unit/mL penicillin, and 100 mg/L streptomycin (GPS; Cellgro, Manassas, VA) at 37 °C in a 5% CO2 environment. Cell-Adhesive PEG-Peptide Hydrogels. The extent of SMC adhesion on surfaces modified with RGDS, EM-23, EM-23S, EM-19 or EM-19S was determined by conjugating the peptides to PEG-DA hydrogels. Briefly, PEG-DA was dissolved in 10 mM HEPES buffer (pH 7.4) at 0.1 g/mL and sterile-filtered using a 0.2 µm pore size nylon filters (Pall Life Sciences, Port Washington, NY). To this solution, 10 µL/mL of the photoinitiator 2,2 dimethyl-2-phenyl-acetophenone (DMAP, Alfa Aesar, Ward Hill, MA) in N-vinylpyrrolidone (300 mg/

434


mL) was added. This precursor solution was injected between two glass slides with a 1 mm spacer (CBS Scientific Company) in between and exposed to UV light (365 nm, 10 mW/cm2) to form a thin, flat hydrogel. The top glass slide was then lifted and phosphate buffered saline (PBS; pH 7.4) was used to rinse the hydrogel. A 1 mL solution of 1 µmol/ mL PEG-peptide containing 10 µL/mL DMAP was used to coat the surface and the top glass slide was carefully replaced to distribute the PEG-peptide solution evenly on the surface. The mold was then exposed to UV light and the monoacrylated PEG peptide-free acrylates were allowed to cross-link to the surface of PEG-DA. Previous studies have shown that limitations imposed by the number of acrylate groups available on a PEG-DA surface allows for the immobilization of 6.25 µmol peptide/cm2.48 Residual peptide solution and DMAP were rinsed away and 14.3 mm diameter gel samples were cut and carefully placed in a 24-well plate. The wells were incubated with SMC media for 2 h and rinsed three times with PBS, and 30000 SMCs/cm2 were then seeded onto the hydrogel surfaces. To observe any nonspecific cell adhesion, gels were swollen in both serum-free media and media containing FBS. Cell adhesion was assessed after 48 h by detaching the cells through incubation with trypsin containing 0.25% ethylenediaminetetraacetic acid (EDTA; Cellgro, Manassas, VA) at 37 °C for 4 min. The dissociated cells were counted using a Beckman Z1 particle counter (Beckman Coulter, Brea, CA). Peptide Influence on Elastin and Desmosine Formation. Elastin production and desmosine content were characterized in vitro after a 48 h incubation of each peptide with SMCs. The 48 h incubation period was chosen after initial experiments confirmed that the protein concentration following incubation with elastin mimetic peptides was well within the range of the Fastin assay. A stock solution of 5 µmol/ mL of each peptide was made by dissolving in 10 mM HEPES buffered saline (HBS) at pH 7.4. The solution was then sterile-filtered using 0.2 µm pore size nylon filters (Pall Life Sciences, Port Washington, NY). A 24-well plate was seeded with 30000 SMCs/cm2. Peptides were added to the cell suspensions during the seeding procedure at concentrations of 0.01, 0.1, or 1 µmol/mL. After 48 h, the media was removed, the wells were thoroughly washed with PBS, and the well plates were frozen and lyophilized overnight. The contents of each well were hydrolyzed by adding 0.1 N NaOH and incubating for 1 h at 100 °C. The resulting lysate was centrifuged at 8000 × g for 10 min and the supernatant containing the cell debris and soluble matrix components were removed. The remaining insoluble elastin was suspended in 0.25 N oxalic acid and further hydrolyzed for 48 h at 100 °C. Following the oxalic acid treatment, the solubilized elastin was purified using Amicon Ultra-0.5 Centrifugal Filters with Ultracel-10 membranes (MWCO 10000; Millipore, Billerica, MA). A Fastin assay (Accurate Chemical, Westbury, NY) and an enzyme-linked immunosorbent assay (ELISA) were used to quantify elastin and desmosine content, respectively. Peptide Competition with r-Lactose and RGDS. In separate experiments, 30000 SMCs/cm2 were incubated with elastin mimetic peptides as well as 10 mM R-lactose or 1 mM RGDS solution (sterilefiltered using 0.2 µm pore size nylon filters; Pall Life Sciences, Port Washington, NY). The EBP also has strong binding interactions with galactosugars, such as R-lactose,28 which triggers the EBP to undergo a conformational change and shed from the cell surface, resulting in decreased elastin production.49 Our elastin mimetic peptides were again added to cell suspensions during the seeding procedure at concentrations of 0.01, 0.1, or 1 µmol/mL, this time also in the presence of either R-lactose or RGDS. After a 48 h incubation period, cell cultures were washed and the elastin and desmosine were isolated as previously described. Fastin Assay for Assessment of Elastin Production. A Fastin assay was used to quantify elastin deposition after sample hydrolysis. Briefly, samples and R-elastin standards in oxalic acid were treated with an equal volume of elastin precipitation reagent containing trichloroacetic acid and hydrochloric acid (HCl) for 10 min. Following precipitation of elastin, the mixture was centrifuged at 10000 × g for 10 min and

Patel et al. the supernatant was discarded. The recovered sample was incubated with 1.0 mL of Fastin dye reagent, containing 5,10,15,20-tetraphenyl21,23-porphine tetra-sulfonate (TPPS), for 90 min. TPPS has an affinity for the basic amino acid side chains of elastin. The elastin-dye complex was then separated via centrifugation at 10000 × g for 10 min. The supernatant was again discarded, and the dye-elastin complex was dissociated by adding 250 µL of dye dissociation reagent containing guanidine HCl and propan-1-ol. The final dye extract was placed in a 96-well plate and absorbance was read using a Beckman DTX 880 Multimode Plate Reader (Beckman Coulter, Brea, CA) at 513 nm. Concentrations of elastin were calculated by comparison with R-elastin standards. Desmosine ELISA. A competitive ELISA was used to quantify desmosine deposition. Briefly, an ovalbumin-desmosine conjugate in 50 mM sodium bicarbonate buffer (pH 8.5) was adsorbed to the wells of a 96-well plate for 1 h at room temperature. After 1 h, the wells were washed three times with washing buffer composed of 10 mM PBS with 0.05% Tween 20. The wells were blocked by adding 1% bovine serum albumin (BSA; Sigma, St. Louis, MO) in 10 mM PBS for 30 min at room temperature and then washed three times with washing buffer. A primary desmosine antibody (Elastin Products Company, EPC, Owensville, MO) in 1% BSA solution (1:1000 dilution) along with either samples or standards were added and allowed to interact for 1 h at room temperature. After primary antibody interaction, the wells were washed three times with washing buffer. A secondary antirabbit IgG conjugated with horseradish peroxidase (1:3500 dilution; Promega, Madison, WI) was then added to each well and incubated for 1 h at room temperature. The wells were again washed three times and incubated with o-phenylenediamine (OPD; Sigma, St. Louis, MO) for 30 min at room temperature in the dark. The final reaction was quenched by adding 3 M HCl to each well and absorbance was read using a Beckman DTX 880 Multimode Plate Reader at 492 nm. Immunofluorescence. EM-23, EM-23S, EM-19, and EM-19S were each conjugated with fluorescein isothiocyanate (FITC). A total of 10 mg of each peptide was dissolved in 5 mL of deionized water, and FITC (1:1 amine mole ratio) was dissolved in 4 mL of 0.1 M sodium bicarbonate buffer (pH 8.5). The dye solution was added to the peptide solution dropwise and allowed to react for 2 h. After 2 h, any unreacted dye and peptide were dialyzed using 2000 MWCO dialysis cassettes (Pierce Biotechnology, Rockford, IL) for 2 h. The dialyzed product was frozen and lyophilized overnight. The conjugated peptide-dye moiety was dissolved in 10 mM HEPES buffer (pH 7.4) such that the final concentration in one 24-well plate would be 1 µmol peptide/mL. The solution was sterilized using 0.2 µm pore size nylon filters (Pall Life Sciences, Port Washington, NY). The sterile solution was added to each well plate containing 80000 SMCs/cm2 and incubated for 24 h. Immunostaining techniques were then utilized to visualize the elastin fibers. Briefly, each well was washed with 10 mM PBS followed by fixing the cells with 100% methanol for 5 min. Wells were again washed

Figure 2. CD spectra of elastin mimetic peptides at room temperature and 37 °C. Both peptides show a significant amount of disorder, indicated by the large negative peak at approximately 200 nm, which is a characteristic of the random coil structure of human elastins.



435

using a Leica DMI 4000B fluorescent microscope (Leica Microsystems, Inc.; Bannockburn, IL). Statistical Analysis. All experiments were performed minimally in triplicate. Error bars reflect standard deviations, and p-values were assessed using two-tailed Student’s t tests; p-values less than 0.05 were considered significant.

Results and Discussion

Figure 3. Both PEG-EM19 and PEG-EM23 promote significant SMC adhesion, and PEG-EM23 encouraged more cell adhesion than PEGRGDS. PEG-EM19S did not facilitate cell adhesion, while adhesion to PEG-EM-23S is comparable to that in the presence of PEG-RGDS (n ) 4, *p < 0.03 compared to PEG-DA).

with 10 mM PBS and a primary elastin antibody at a 1:100 dilution (BA-4, Santa Cruz Biotechnology, Santa Cruz, CA) in 10 mM PBS was incubated with cell cultures for 2 h at room temperature. The wells were washed with 10 mM PBS and incubated with secondary goat antimouse IgG conjugated with Texas Red (SCBT, Santa Cruz, CA) at a 1:400 dilution for 1 h at room temperature. Elastin fibers, along with peptide-dye moieties, were visualized simultaneously in each well

Peptide Characterization. Peptides structures were confirmed by both GPC and mass spectroscopy (see Supporting Information). Peptides were characterized by CD at room temperature and 37 °C, and each spectrum contains a large negative band at 197 nm, which is representative of a disordered structure, such as the random coil of native human elastin, but do not contain additional secondary structures observed in soluble human elastins (Figure 2). The CD spectrum of soluble human κ-elastin contains a large negative band at 200 nm and includes a shoulder at 222 nm suggested to be indicative of a substantial number of irregular β structures,50 which are typical features of noncoacervated solutions of soluble elastins.51-56 The addition of RGDS into the peptide decreases the intensity of the negative peak, and at physiological temperature, there are small but significant changes in each spectrum, indicating changes in secondary structure with temperature. These results

Figure 4. Elastin deposition and desmosine deposition from EM-19 and EM-23 and their respective scrambled sequences. (A) EM-19 and EM-19S stimulate decreases in elastin production with increasing peptide concentration. Though only EM-19 can bind to the EBP, both peptides are capable of being cross-linked into the elastin matrix, thus, the total mass of tropoelastin being cross-linked in decreased. (B) Both EM-19 peptides cause only small fluctuation in desmosine production, indicating that LOX activity is not significantly changed. (C) The addition of EM-23 in SMC cultures leads to reduced levels of elastin at lower concentrations, followed by recovery at the highest concentration indicating an added effect of the RGDS domain (the data taken at 0.01 µmol and 1 µmol are not statistically different, while both are significantly different from data taken at 0.1 µmol; p < 0.02). (D) EM-23 and EM-23S also stimulate only small changes in desmosine production compared to controls (n ) 7; *p < 0.05 compared to TCPS; **p < 0.005 compared to scrambled sequences).

436


Patel et al.

Figure 5. Elastin and desmosine production encouraged by EM-19 and EM-23 in the presence of 10 mM R-lactose. (A) EM-19 undergoes competition with R-lactose but elastin levels show signs of recovery at high concentrations due to competition of the VGVAPG sequence with lactose. EM-19S does not compete to bind the EBP and no functional recovery is possible. (B) EM-19 and EM-19S stimulate significant increases in desmosine production. (C) Elastin levels in the presence of EM-23 and EM-23S remain drastically decreased, suggesting the peptide binds to integrins and cannot stimulate EBP activity. (D) EM-23 and EM-23S are able to promote desmosine production, yet not at the levels observed for the EM-19 peptides (n ) 5; *p < 0.05 compared to TCPS).

indicate that the inclusion of our peptide into the elastic matrix should not cause effects in the conformation of the protein structure. Smooth Muscle Cell Adhesion on Bioactive PEG Hydrogels. Elastin mimetic peptides were conjugated to acrylatePEG-SVA for incorporation into PEG-DA hydrogels; the final yield for peptide-conjugated moieties was approximately 70%. PEG hydrogels are well-characterized materials with desirable properties for use in tissue engineering, including their ease of functionalization. The hydrophilicity of PEG inhibits cell adhesion without the incorporation of cell adhesive peptide sequences, and thus, we chose these hydrogels as our base material to study the ability of our peptides to facilitate cell adhesion. Derivatization of the monoacrylate PEG was monitored by UV-vis spectroscopy before and after the coupling reaction by measuring the increase in absorptivity at 260 nm as the NHS group is cleaved and by GPC. PEG-peptides were then photopolymerized onto the surfaces of thin, flat PEGhydrogels, which were then swollen in SMC culture media. Cell adhesion was determined after 48 h of culture. There were no differences in cell adhesion when hydrogels were swollen in serum-free media compared to those swollen in media containing FBS, suggesting that there is no nonspecific protein adsorption to our peptides. Each peptide was able to significantly increase cell adhesion as compared to the PEG-DA control (p < 0.03), and though PEG-EM19 was able to induce slightly less cell adhesion than PEG-RGDS (20% decrease in cell number), PEG-EM23 was able to stimulate 31% more cell adhesion than

RGDS and 45% more than EM-19 when conjugated to PEG hydrogel surfaces (Figure 3). Results from this study indicate the RGDS motif on EM-23 plays a significant role in cell adhesion compared to the VGVAPG moiety. Human tropoelastin has previously been classified as cell-adhesive through binding both the EBP and cell surface integrins,23,24,20 and our results indicate that though the presence of VGVAPG alone can stimulate cell adhesion, the presence of this domain in conjunction with RGDS leads to a more significant cell attachment on synthetic substrates. Elastin and Desmosine Production in Smooth Muscle Cell Cultures. Elastin and desmosine content were quantified after incubating SMCs with either EM-19, EM-23, or scrambled control sequences EM-19S and EM-23S (Figure 4). As expected, EM-19 peptide was able to compete with tropoelastin and bind the EBP, decreasing the total mass of elastin produced. EM19S also unexpectedly decreased elastin production to similar levels through unrestricted LOX activity, without binding the EBP. EM-19 and EM-19S both contain the lysine-rich motifs found in tropoelastin that are involved in forming desmosine cross-links, and both are being actively incorporated into the elastic matrix instead of tropoelastin, resulting in the decreased mass of elastin over all peptide concentrations. For both peptides, the overall desmosine production did not change significantly from the control levels (cells cultured on tissue culture polystyrene, TCPS; Figure 4B), suggesting that though the LOX activity and desmosine formation are fairly constant over our conditions, the overall decrease in elastin mass is due



437

Figure 6. Elastin and desmosine production stimulated by EM-19 and EM-23 in the presence of 1 µmol RGDS. (A) The addition of RGDS alone decreases elastin production in SMC culture, and EM-19 promotes further decreases in elastin levels, suggesting that RGDS plays an important role in elastin production. (B) Desmosine production decreases with increasing peptide concentration, a trend similar to that observed in the presence of lactose. (C) EM-23 stimulates a similar trend in elastin content as cultures that did not contain RGDS. The data suggests that there is a synergistic interplay between RGDS and VGVAPG on EM-23 that is influencing elastin production. (D) Desmosine production from EM-23 only increases marginally at all peptide concentrations (n ) 5; *p < 0.05 compared to TCPS).

to the incorporation of both peptides into the matrix through enzymatic cross-linking of the peptide lysine groups, rather than the much larger tropoelastin. Figure 4A suggests that EM-19S is also able to participate in cross-linking without having any affinity to bind to the surfaces of SMCs, alluding to the high activity of the cross-linking lysine motifs. Further, Figure 4B reveals very slight changes in the amount of desmosine content for both sequences; however, desmosine is only 0.9% of elastin57 and any increase in desmosine correlates to a higher potential for incorporation of the peptide into elastic fibers. EM-23 stimulates a similar trend, though not to the extent that EM-19 and EM-19S influenced the mass of elastin (Figure 4C). This data imply that though there is significant competition to bind to the EBP with peptides containing the VAPG sequence, the presence of the RGD motif leads to preferential binding of integrins, which do not inhibit tropoelastin transport or lead to the same level of decreased elastin mass. The small fluctuations observed in the desmosine content across all conditions indicate that the changes in elastin levels detected are only due to differences in the amount of tropoelastin that is integrated into the elastin matrix and not any inhibition of the overall amount of cross-linking occurring in the system (Figure 4D). The lysine moieties of the peptides are actively cross-linked instead of tropoelastin, thus, similar amounts of desmosine are being produced. Bioactive Peptide Competition with r-Lactose. To further confirm the competition of EM-19 with secreted tropoelastin to bind the EBP, we repeated the previously described experi-

ments with the addition of 10 mM R-lactose to each well of the cell cultures. Lactose and VGVAGP have a similar binding affinity to the EBP and lactose is known to decrease the affinity of the EBP for immobilized elastin and blocks elastin production.20,49,28 Figure 5A shows the competitive binding phenomenon between EM-19 and R-lactose for the EBP. The binding of EM-19 to the EBP blocks the adhesion of cellsecreted tropoelastin to the EBP; however, binding of R-lactose changes the structural conformation of the EBP and sheds it from the cell surface, resulting in the observed decreased elastin production.23 Tropoelastin is synthesized and then transported to the cell surface where it binds to the EBP and is then transferred to adjacent microfibrils involved in elastin fiber assembly.17 Lactose also binds the EBP and the receptor is subsequently shed from the cell surface, leading to a drastic decrease in elastin production in the control culture, because the EBP cannot be recycled (Figure 5A, TCPS). The addition of increasing amounts of EM-19 restricts the number of EBP receptors that are readily bound by lactose, and elastin production is able to occur as the peptide is cross-linked into the forming elastic fibers; the EBP is recycled to the cell surface and able to subsequently bind tropoelastin. As larger amounts of the EBP are available for recycling to the cell surface at increasing concentrations, more tropoelastin is chaperoned to the elastic fiber assembly. This correlates to the increased elastin mass (containing more of the large protein and less small peptide) and a decrease in desmosine content to the control level (Figure 5B); in effect, more mass is added with less cross-links

438


Patel et al.

Figure 7. Immunofluorescent images of peptide activity in vitro after 24 h. Arrows indicate sites of peptide localization with the elastin matrix. EM-19 colocalizes with the elastin matrix (A) as does EM-19S (B). Both EM-23 (C) and EM-23S (D) are able to bind the cell surface and greater colocalization is observed compared to the EM-19 peptides.

formed. Though the data is normalized to the control, it should be noted that the elastin content recovers to 5.48 ( 1.81 µg, which is only slightly less than the 7.23 ( 1.06 µg of total elastin produced at the 1 µmole EM-19 concentration in the absence of lactose. The scrambled sequence is not able to compete for the receptor or promote recovery of elastin production. Of note, the amount of desmosine observed at the lower concentrations of both EM-19 sequences was increased (Figure 5B), decreasing to control levels at the highest peptide concentration. These results indicate that, when EM-19 is present in low concentrations and is able to bind the EBP competitively, cross-linking is greatly enhanced through the incorporation of our peptide to the limited amounts of tropoelastin produced. On the other hand, elastin production decreased throughout all concentrations of both EM-23 and EM-23S in the presence of R-lactose (Figure 5C), and desmosine levels do not exceed the control levels (Figure 5D). Here we demonstrate that EM-23 preferentially binds to integrins instead of the EBP in a competitive environment, allowing R-lactose to bind the EBP and leading to a decrease in elastin production with no recovery at increasing peptide concentrations. As cross-linking controls, glycine derivatized EM-19 and EM-23 were used in identical experiments. These peptides did not contain the lysine moieties necessary for cross-linking by LOX, and when incubated with SMCs in the presence of lactose, elastin production was inhibited and unable to recover. Levels of detected desmosine cross-links

also remained constant levels across the control and three different concentrations of EM-19G and EM-23G. Peptide Competition with RGDS. To further validate a synergistic role of integrin activity with the EBP in elastin and desmosine production, soluble RGDS (1 µmol/mL) was added to cultures containing EM-19 and EM-23, and elastin and desmosine content was assessed after 48 h. Incubating SMCs with RGDS alone decreases elastin production by approximately 30% when compared to elastin production without RGDS (Figure 6). This suggests that by occupying integrins on the cell surface, the mechanism of elastin production is somehow inhibited. Recently, Bax et al. have reported a C-terminal motif on tropoelastin that binds Rvβ3 and mediates fibroblast adhesion to tropoelastin,20 suggesting that Rvβ3 actively participates in elastic fiber deposition and assembly. Our results reinforce this theory, as the presence of RGDS and EM-19 also causes a decrease in elastin production (Figure 6A), but remarkably, the levels of elastin detected were higher than when SMCs are incubated only with EM-19 (as shown in Figure 4A). EM-23 is able to efficiently compete with RGDS and maintain similar levels of elastin and desmosine as those previously observed in cultures without RGDS (Figure 6C,D). Immunofluorescence. For visual characterization of peptide incorporation into the elastin matrix, 1 µmol of each peptide was tagged with FITC and using immunostaining techniques, we were able to visualize peptide activity and elastin fibers in


vitro. Visualization of our peptides correlate well with the quantified elastin and desmosine data, showing FITC-labeled peptides colocalized with immunostained elastin fibers (Figure 7). Localized cross-linking of EM-19 (Figure 7A) can be seen around the periphery of the cell membrane, as can EM-19S (Figure 7B). On the other hand, both EM-23 and EM-23S show similar localization on the cell periphery to a much greater extent (Figure 7C and D, respectively). The EM-23 sequence has the ability to bind to either cell surface integrins or the EBP whereas EM-23S can bind only through integrins. Cultures containing EM-23 show an abundance of peptide localized throughout the image, resulting in an overwhelming colocalized (yellow) fluorescence (Figure 7C), and though quite a bit of localization of the fluorescent peptide with the immunostained elastin can be observed in cultures containing EM-23S (Figure 7D), there is less colocalized fluorescence than the high levels observed in cultures containing EM-23 (Figure 7C).

Conclusion Through rational design and a thorough in vitro characterization, we have shown the potential of short elastin mimetic peptides as a means of supporting vascular cell adhesion and catalyzing elastic fiber assembly that incorporates our peptide into the developing ECM. These peptides include essential cellbinding motifs, moieties that allow enzymatic cross-linking with ECM proteins, and the characteristic disordered structure of human elastin. We have shown that there is a synergistic interplay between the EBP and integrins in elastic fiber production and organization that can be exploited to engineer elastin that will be physically tethered to bioactive hydrogels through LOX-catalyzed cross-linking. The incorporation of these peptides into synthetic matrices for vascular tissue engineering will promote the deposition of an ECM template for tissue development based on cell-secreted elastin and can serve as a means to further investigate the mechanisms of elastic fiber organization. Acknowledgment. The authors would like to thank Dr. Barry Starcher of the University of Texas Health Science Center at Tyler for assistance with the desmosine ELISA and Dr. Tsutomu Arakawa of Alliance Protein Laboratories for CD analysis. Funding for this project has been provided by the Georgia Institute of Technology. Supporting Information Available. Additional peptide characterization (mass spectra) of EM-19 and EM-23. This material is available free of charge via the Internet at http:// pubs.acs.org.

References and Notes (1) Tibbitt, M. W.; Anseth, K. S. Biotechnol. Bioeng. 2009, 103, 655– 663. (2) Pratt, A. B.; Weber, F. E.; Schmoekel, H. G.; Muller, R.; Hubbell, J. A. Biotechnol. Bioeng. 2004, 86, 27–36. (3) Faury, G. Pathol. Biol. 2001, 49, 310–325. (4) Cox, B. A.; Starcher, B. C.; Urry, D. W. J. Biol. Chem. 1974, 249, 997–998. (5) Wu, W. J.; Vrhovski, B.; Weiss, A. S. J. Biol. Chem. 1999, 274, 21719–21724. (6) Powell, J. T.; Vine, N.; Crossman, M. Atherosclerosis 1992, 97, 201– 208. (7) Chrzanowski, P.; Keller, S.; Cerreta, J.; Mandl, I.; Turino, G. M. Am. J. Med. 1980, 69, 351–359. (8) Ayad, S.; Boot-Handford, R.; Humphries, M. J.; Kadler, K. E.; Shuttleworth, C. A. The extracellular matrix factsbook, 2nd ed.; Academic Press: San Diego, CA, 1998.


439

(9) Rosenbloom, J.; Abrams, W. R.; Mecham, R. FASEB J. 1993, 7, 1208– 1218. (10) Mochizuki, S.; Brassart, B.; Hinek, A. J. Biol. Chem. 2002, 277, 44854–44863. (11) Li, D. Y.; Brooke, B.; Davis, E. C.; Mecham, R. P.; Sorensen, L. K.; Boak, B. B.; Eichwald, E.; Keating, M. T. Nature 1998, 393, 276– 280. (12) Rodgers, U. R.; Weiss, A. S. Pathol. Biol. 2005, 53, 390–398. (13) Wagenseil, J. E.; Mechamp, R. P. Birth Defects Res. C 2007, 81, 229– 240. (14) Kielty, C. M.; Sherratt, M. J.; Shuttleworth, C. A. J. Cell Sci. 2002, 115, 2817–2828. (15) Bressan, G. M.; Prockop, D. J. Biochemistry 1977, 16, 1406–1412. (16) Grosso, L. E.; Mecham, R. P. Biochem. Biophys. Res. Commun. 1988, 153, 545–551. (17) Daamen, W. F.; Veerkamp, J. H.; van Hest, J. C. M.; van Kuppevelt, T. H. Biomaterials 2007, 28, 4378–4398. (18) Wise, S. G.; Weiss, A. S. Int. J. Biochem. Cell Biol. 2009, 41, 494– 497. (19) Hinek, A.; Pshezhetsky, A. V.; von Itzstein, M.; Starcher, B. J. Biol. Chem. 2006, 281, 3698–3710. (20) Bax, D. V.; Rodgers, U. R.; Bilek, M. M. M.; Weiss, A. S. J. Biol. Chem. 2009, 284, 28616–23. (21) Seyrantepe, V.; Hinek, A.; Peng, J.; Fedjaev, M.; Ernest, S.; Kadota, Y.; Canuel, M.; Itoh, K.; Morales, C. R.; Lavoie, J.; Tremblay, J.; Pshezhetsky, A. V. Circulation 2008, 117, 1973–1981. (22) Hinek, A. J. Biol. Chem. 1996, 377, 471–480. (23) Mecham, R. P.; Hinek, A.; Entwistle, R.; Wrenn, D. S.; Griffin, G. L.; Senior, R. M. Biochemistry 1989, 28, 3716–3722. (24) Robinet, A.; Fahem, A.; Cauchard, J. H.; Huet, E.; Vincent, L.; Lorimier, S.; Antonicelli, F.; Soria, C.; Crepin, M.; Hornebeck, W.; Bellon, G. J. Cell Sci. 2005, 118, 343–356. (25) Faury, G.; Garnier, S.; Weiss, A. S.; Wallach, J.; Fulop, T.; Jacob, M. P.; Mecham, R. P.; Robert, L.; Verdetti, J. Circ. Res. 1998, 82, 328–336. (26) Floquet, N.; Hery-Huynh, S.; Dauchez, M.; Derreumaux, P.; Tamburro, A. M.; Alix, A. J. P. Biopolymers 2004, 76, 266–280. (27) Vrhovski, B.; Weiss, A. S. Eur. J. Biochem. 1998, 258, 1–18. (28) Hinek, A.; Wrenn, D. S.; Mecham, R. P.; Barondes, S. H. Science 1988, 239, 1539–1541. (29) Maki, J. M.; Sormunen, R.; Lippo, S.; Kaarteenaho-Wiik, R.; Soininen, R.; Myllyharju, J. Am. J. Pathol. 2005, 167, 927–936. (30) Sibon, I.; Sommer, P.; Lamaziere, J. M. D.; Bonnet, J. Heart 2005, 91, e33. (31) Kozel, B. A.; Ciliberto, C. H.; Mecham, R. P. Matrix Biol. 2004, 23, 23–34. (32) Li, D. Y.; Faury, G.; Taylor, D. G.; Davis, E. C.; Boyle, W. A.; Mecham, R. P.; Stenzel, P.; Boak, B.; Keating, M. T. J. Clin. InVest. 1998, 102, 1783–1787. (33) Urry, D. W.; Parker, T. M. J. Muscle Res. Cell Motil. 2002, 23, 543– 559. (34) Buttafoco, L.; Kolkman, N. G.; Engbers-Buijtenhuijs, P.; Poot, A. A.; Dijkstra, P. J.; Vermes, I.; Feijen, J. Biomaterials 2006, 27, 724–734. (35) Ratcliffe, A. Matrix Biol. 2000, 19, 353–357. (36) Kielty, C. M.; Stephan, S.; Sherratt, M. J.; Williamson, M.; Shuttleworth, C. A. Philos. Trans. R. Soc. B 2007, 362, 1293–1312. (37) Nimni, M. E.; Myers, D.; Ertl, D.; Han, B. J. Biomed. Mater. Res. 1997, 35, 531–537. (38) McHale, M. K.; Setton, L. A.; Chilkoti, A. Tissue Eng. 2005, 11, 1768– 1779. (39) Lee, J. Recent Res. DeV. Polym. Sci. 2002, 6, 27–43. (40) Mithieux, S. M.; Rasko, J. E. J.; Weiss, A. S. Biomaterials 2004, 25, 4921–4927. (41) Berglund, J. D.; Nerem, R. M.; Sambanis, A. Tissue Eng. 2004, 10, 1526–1535. (42) Kaufmann, D.; Fiedler, A.; Junger, A.; Auernheimer, J.; Kessler, H.; Weberskirch, R. Macromol. Biosci. 2008, 8, 577–588. (43) Lee, J.; Macosko, C. W.; Urry, D. W. Biomacromolecules 2001, 2, 170–179. (44) Ruoslahti, E.; Pierschbacher, M. D. Science 1987, 238, 491–497. (45) Hern, D. L.; Hubbell, J. A. J. Biomed. Mater. Res. 1998, 39, 266– 276. (46) Liaw, L.; Almeida, M.; Hart, C. E.; Schwartz, S. M.; Giachelli, C. M. Circ. Res. 1994, 74, 214–224. (47) Liaw, L.; Skinner, M. P.; Raines, E. W.; Ross, R.; Cheresh, D. A.; Schwartz, S. M.; Giachelli, C. M. J. Clin. InVest. 1995, 95, 713–724. (48) Hahn, M. S.; Taite, L. J.; Moon, J. J.; Rowland, M. C.; Ruffino, K. A.; West, J. L. Biomaterials 2006, 27, 2519–2524.

440


(49) Wachi Hiroshi, S. H.; Murata, H.; Nakazawa, J.; Mecham, R. P.; Seyakama, Y. J. Atheroscler. Thromb. 2004, 11, 159–166. (50) Debelle, L.; Alix, A. J.; Wei, S. M.; Jacob, M. P.; Huvenne, J. P.; Berjot, M.; Legrand, P. Eur. J. Biochem. 1998, 258, 533–539. (51) Mammi, M.; Gotte, L.; Pezzin, G. Nature 1968, 220, 371–373. (52) Urry, D. W.; Starcher, B.; Partridge, S. M. Nature 1969, 222, 795– 796. (53) Tamburro, A. M.; Guantieri, V.; Dagagordini, D.; Abatangelo, G. Biochim. Biophys. Acta 1977, 492, 370–376.

Patel et al. (54) Tamburro, A. M.; Guantieri, V.; Gordini, D. D. Int. J. Biol. Macromol. 1982, 4, 111–115. (55) Urry, D. W. Ultrastruct. Pathol. 1983, 4, 227–251. (56) Urry, D. W. J. Protein Chem. 1988, 7, 1–34. (57) Yamamoto, Y.; Sakata, N.; Meng, J.; Sakamoto, M.; Noma, A.; Maeda, I.; Okamoto, K.; Takebayashi, S. Nephrol., Dial., Transplant. 2002, 17, 630–636.

BM101214F

Self-Assembly of Elastin-Based Peptides into the ECM: the Importance

Recommend Documents