Combined Effect of Osteopontin and BMP-2 Derived Peptides Grafted

Feb 28, 2012 - ... and BMP-2 Derived Peptides Grafted to an Adhesive Hydrogel on Osteogenic and Vasculogenic Differentiation of Marrow Stromal Cells...
1 downloads 0 Views 4MB Size
Download PDF
Article pubs.acs.org/Langmuir

Combined Effect of Osteopontin and BMP-2 Derived Peptides Grafted to an Adhesive Hydrogel on Osteogenic and Vasculogenic Differentiation of Marrow Stromal Cells Xuezhong He,† Xiaoming Yang,‡ and Esmaiel Jabbari*,† †

Biomimetic Materials and Tissue Engineering Laboratories, Department of Chemical Engineering, University of South Carolina, Columbia, South Carolina 29208, United States ‡ Dorn Research Institute, Columbia, South Carolina 29209, United States ABSTRACT: The objective of this work was to investigate the combined effect of grafting the peptide corresponding to amino acid residues 162−168 of osteopontin (OPD peptide) and the peptide corresponding to amino acid residues 73−92 of bone morphogenetic protein-2 (BMP peptide) to an RGDconjugated inert hydrogel on osteogenic and vasculogenic differentiation of bone marrow stromal (BMS) cells. RGD-conjugated three-dimensional (3D) porous hydrogel scaffolds with well-defined cylindrical pore geometry were produced from sacrificial wax molds fabricated by fused deposition modeling rapid prototyping system. Propargyl acrylate and 4-pentenal were conjugated to the hydrogel for orthogonal grafting of BMP and OPD peptides by click reaction and oxime ligation, respectively. The OPD peptide was grafted by the reaction between aminooxy moiety of aminooxy-mPEG-OPD (mPEG = mini-poly(ethylene glycol)) and the aldehyde moiety in the hydrogel. The BMP peptide was grafted by the reaction between the azide moiety of Az-mPEGBMP and the propargyl moiety in the hydrogel. The hydrogels seeded with BMS cells were characterized by biochemical, immunocytochemical, and mRNA analyses. Groups included RGD control hydrogel (RGD), RGD and BMP peptides without OPD (RGD+BMP), RGD and BMP peptides with mutant OPD (RGD+BMP+mOPD), and RGD and BMP peptides with OPD (RGD+BMP+OPD) grafted hydrogels. The extent of mineralization of RGD, RGD+BMP, RGD+BMP+mOPD, and RGD +BMP+OPD groups after 28 days was 650 ± 70, 990 ± 30, 850 ± 30, and 1150 ± 40 mg/(mg of DNA), respectively, indicating that the BMP and OPD peptides enhanced osteogenic differentiation of the BMS cells. The BMS cells seeded on RGD+BMP +OPD grafted hydrogels stained positive for vasculogenic markers α-SMA, PECAM-1, and VE-cadherin while the groups without OPD peptide (RGD+BMP and RGD+BMP+mOPD) stained only for α-SMA but not PECAM-1 or VE-cadherin. These results were consistent with the significantly higher PECAM-1 mRNA expression for RGD+BMP+OPD group after 21 and 28 days, compared to the groups without OPD. These findings suggest that the RGD+BMP+OPD peptides provide a favorable microenvironment for concurrent osteogenic and vasculogenic differentiation of progenitor marrow-derived cells.

1. INTRODUCTION Reconstruction of skeletal defects that require bone graft procedures remains a significant problem.1−3 Autologous bone graft is the standard for treatment because it provides an osteoconductive matrix for adhesion of progenitor cells and inductive factors for their differentiation, all maintained by a vascular network.4 However, there is a limited supply of autograft tissue.2 Novel functional materials that can support adhesion of progenitor cells and induce their differentiation to multiple lineages hold great promise for treating skeletal defects. Natural, synthetic, and hybrid materials have been investigated as a matrix for seeding osteoprogenitor cells.5 Natural materials provide a conductive matrix for cell migration and adhesion but lack sufficient mechanical strength to prevent soft tissue compression.6 Synthetic materials provide enormous flexibility in the design of materials with defined physical and mechanical properties, but they lack bioactive recognition ligands to support cell−matrix interactions required for adhesion, proliferation, differentiation, and maturation of the © 2012 American Chemical Society

seeded cells. This limitation can be overcome by chemically conjugating bioactive ligands or partial sequences of the extracellular matrix (ECM) proteins to the synthetic matrix.7 The focal adhesion RGD sequence that interacts with integrinbinding receptors on the cell surface is the most widely used peptide.7 Conjugation of RGD to many polymer matrices including poly(ε-caprolactone) (PCL), 8 poly( L -lactide) (PLA),9 poly(lactide-co-glycolide) (PLGA),10 poly(ethylene glycol) (PEG),11 and their copolymers12 has demonstrated that the peptide facilitates adhesion of bone marrow stromal (BMS) cells to the substrate and is a mild promoter of osteogenic differentiation.11,13−15 In a previous work, we demonstrated that the covalent attachment of integrin-binding RGD peptide and the osteogenic BMP peptide, corresponding to amino acid residues 73−92 of the knuckle epitope of recombinant human bone morphogenetic protein (rhBMP-2), Received: December 19, 2011 Revised: February 27, 2012 Published: February 28, 2012 5387

dx.doi.org/10.1021/la205005h | Langmuir 2012, 28, 5387−5397

Langmuir

Article

evaluated by biochemical (DNA content, ALPase activity, calcium content, and Alizarin red staining), mRNA (osteopontin, osteocalcin, osteonectin, and PECAM-1), and immunocytochemical (α-SMA, VE-cadherin, PECAM-1) analyses.

enhanced osteogenic differentiation and mineralization of the seeded BMS cells.16 The RGD peptide provided sites for spreading and adhesion of BMS cells to the matrix, which in turn increased the interaction of BMP peptide with type I and II cell surface receptors, leading to a higher extent of osteogenic differentiation and mineralization. Several studies have shown that inadequate vascularity limits the supply of oxygen and nutrients to the engineered matrix, which often leads to poor differentiation and mineralization of the seeded cells.17−20 For example, osteoblasts seeded in a cancellous bone matrix did not survive 1 week after implantation in a calvarial defect.21 It has been reported that the peptide SVVYGLR (OPD peptide), corresponding to amino acid residues 162−168 of osteopontin, induces vasculogenic differentiation of the bone marrow cells.22,23 The OPD peptide interacts with α9β1 integrin receptors on the surface of BMS cells, analogous to the interaction of this receptor with vascular cell adhesion protein-1 (VCAM-1). The peptide is not active until it is exposed upon thrombin cleavage of the protein at the sites of inflammation and remodeling.24 The OPD peptide induces tube formation by progenitor endothelial cells in 3D collagen gels with as much potency as VEGF.22,24 In an effort to induce vascularization in synthetic matrices for bone regeneration, in this work we used orthogonal conjugation chemistries to graft BMP and OPD peptides to a cell-adhesive hydrogel matrix. The objective of this work was to investigate the effect of grafting BMP and OPD peptides to a cell-adhesive RGD-conjugated 3D porous hydrogel scaffold on concurrent differentiation of the seeded BMS cells to osteogenic and vasculogenic lineages. Many different peptide combinations need to be tested when three adhesive, osteoinductive, and vasculogenic peptides are conjugated/grafted to the hydrogel matrix. Since the effect of grafted OPD alone on osteogenic and vasculogenic differentiation of BMS cells has been previously investigated,22,23 we focused on the combined effect of OPD and BMP peptides on differentiation of BMS cells. In that regard, the experimental groups included control hydrogel (RGD only) and hydrogel without OPD (RGD+BMP), with mutant OPD (RGD+BMP +mOPD), and with OPD peptide (RGD+BMP+OPD). Pore geometry and pore size distribution affect vasculogenic and osteogenic differentiation of progenitor cells.25−27 Therefore, it is important to investigate the effect of bioactive peptides on BMS cell differentiation in a three-dimensional (3D) environment with well-defined pore geometry and size to minimize geometric and size distribution effects. To reach the objective, we used solid free-form fabrication to produce porous wax molds with uniform pore size. The wax mold was used as a sacrificial negative mold to produce an RGD-conjugated 3D hydrogel scaffold with cylindrical pore geometry by infusing the hydrogel precursor solution in the mold, followed by crosslinking and dissolving the wax. The hydrogel precursor solution is based on poly(lactide-co-ethylene oxide fumarate) (PLEOF) macromer, developed in our laboratory.16,28−30 The RGD peptide was functionalized with an acrylamide group for bulk conjugation to the hydrogel matrix. For grafting reactions, the hydrogel was functionalized with aldehyde and propargyl groups by adding 4-pentenal and propargyl acrylate, respectively, to the hydrogel precursor solution. The aminooxyfunctionalized OPD and azide-functionalized BMP peptides were grafted to the pore surface area of the hydrogel by aminooxy/aldehyde and azide/propargyl reactions, respectively. The peptide grafted hydrogels were seeded with BMS cells and

2. EXPERIMENTAL SECTION 2.1. Materials. The protected amino acids and Rink Amide NovaGel resin were purchased from EMD Biosciences (San Diego, CA, USA). N,N-Dimethylformamide (DMF), methylene chloride (DCM), acetonitrile (MeCN), N,N-diisopropylethylamine (DIEA), N,N′-diisopropylcarbodiimide (DIC), diethylene glycol (DEG), triisopropylsilane (TIPS), (N,N-dimethylamino)pyridine (DMAP), hydroxybenzotriazole (HOBt), and trifluoroacetic acid (TFA) were received from Acros (Pittsburgh, PA, USA). Diethyl ether and hexane were obtained from VWR (Bristol, CT, USA). Piperidine, deuterated chloroform (CDCl3), fumaryl chloride (FuCl), triethylamine (TEA), poly(ethylene glycol) (PEG, 3.4 kDa), ammonium persulfate (APS), N,N-methylenebis(acrylamide) (BISAM), 4-pentenal, 4-carboxybenzenesulfonazide, acrylic acid, N,N,N,N-tetramethylethylenediamine (TMEDA), ethylenediaminetetraacetic acid disodium salt (EDTA), penicillin, streptomycin, Alizarin red S, and paraformaldehyde were purchased from Sigma-Aldrich (St. Louis, MO, USA). Dulbecco’s phosphate-buffered saline (PBS) and modified Eagle’s medium (DMEM; 4.5 g/L glucose with L-glutamine and without sodium pyruvate) were obtained from Cellgro (Herndon, VA, USA). Fetal bovine serum (FBS, screened for compatibility with rat bone marrow stromal cells) was obtained from Atlas Biologicals (Fort Collins, CO, USA). Trypsin and Quant-it PicoGreen dsDNA reagent kit were obtained from Invitrogen (Carlsbad, CA, USA). QuantiChrom calcium and alkaline phosphatase (ALPase) assay kits were purchased from Bioassay Systems (Hayward, CA, USA). All other solvents and chemicals were used as received. 2.2. Instrumentation. The chemical structure of the PLEOF macromer was characterized by a Varian Mercury-300 1H NMR (Varian, Palo Alto, CA, USA). The polymer sample was dissolved in CDCl 3 at a concentration of 50 mg/mL, and 1% (v/v) tetramethylsilane (TMS) was used as the internal standard. The molecular weight distribution of the synthesized polymers was measured by gel permeation chromatography (GPC, Waters, Milford, MA, USA) as previously described.28 Monodisperse polystyrene standards (Waters, 0.58−66.35 kDa and polydispersities of 14 mM) did not significantly increase grafting density, indicating that the reaction was limited by the aldehyde density on the hydrogel surface. The grafting density increased from 3.8 ± 0.4 to 5.4 ± 0.7 pmol/cm2 when the BMP peptide concentration was increased from 2 to 10 mM. Higher BMP peptide concentrations (>5 mM) did not significantly increase the grafting density. 3.2. Biochemical Analysis of the BMS Seeded Hydrogels. DNA content of the BMS cells seeded in the hydrogels is shown in Figure 5a for the four groups. There was no statistically significant difference in cell count between the four groups for all time points which is consistent with the fact that RGD modification of hydrogels promotes BMS cell adhesion.13 We have previously shown that BMS cells spread and exhibit focal-point adhesion on scaffolds conjugated with RGD peptide.30 Hamada et al. has shown that the OPD peptide does not have a significant effect on cell proliferation.42 ALPase activity of the cell-seeded hydrogels for the four groups is shown in Figure 5b. In Figure 5b,c, one star indicates there was a statistically significant difference between the three BMP-containing hydrogels and the RGD group (control) for a given time point. Two stars indicate a significant difference between the RGD+BMP+OPD group and the other BMP

Figure 3. HPLC chromatographs of the functionalized aminooxymPEG-OPD (a) and Az-mPEG-BMP (b) peptides, respectively; mass spectra of the aminooxy-mPEG-OPD (c) and Az-mPEG-BMP (d) peptides, respectively. 5391

dx.doi.org/10.1021/la205005h | Langmuir 2012, 28, 5387−5397

Langmuir

Article

Figure 5. (a) DNA content, (b) ALPase activity, and (c) calcium content of BMS cells seeded in peptide grafted hydrogels as a function of incubation time in osteogenic media supplemented with vasculogenic factors. Groups include RGD control (light blue), RGD and BMP grafted peptide (RGD+BMP, green), RGD, BMP, and mutant OPD grafted (RGD+BMP+mOPD, blue), and RGD, BMP, and OPD grafted (RGD+BMP +OPD, red) hydrogels. In the panels, one star indicates a statistically significant difference (p < 0.05) between the three BMP groups (RGD+BMP, RGD+BMP+mODP, and RGD+BMP+ODP) and the control group (RGD) at the same time point. Two stars indicate a significant difference between the RGD+BMP+OPD group and other BMP groups. Error bars correspond to means ± SD (n = 3).

groups. ALPase activity of all groups peaked after 14 days and returned to the baseline level at day 21. At day 14, ALPase activity of RGD, RGD+BMP, RGD+BMP+mOPD, and RGD +BMP+OPD groups was 1.3 ± 0.1, 1.8 ± 0.2, 2.3 ± 0.2, and 2.5 ± 0.2 IU/(mg of DNA), respectively. The ALPase activity of BMP peptide grafted groups was significantly higher than the group that was not grafted with BMP peptide (RGD group). At day 14, ALPase activity of the RGD+BMP+OPD group was significantly higher than BMP only groups, which indicated that osteogenic differentiation of the BMS cells was modulated by OPD peptide (indicated by two stars). Calcium content of the cell seeded hydrogels is shown in Figure 5c. For days 4, 7, and 14, there was no significant difference in calcium content among the four groups. For days 21 and 28, there was a significant difference between the calcium content of the BMP peptide groups and the control, as indicated by one star. For those time points, the calcium content of RGD+BMP+OPD peptide group was significantly higher than the other BMP peptide groups (RGD+BMP and RGD+BMP+mOPD). For example, at day 28, the calcium content of RGD, RGD+BMP, RGD+BMP+mOPD, and RGD +BMP+OPD groups was 650 ± 70, 990 ± 30, 850 ± 30, and 1150 ± 40 mg/(mg of DNA), respectively. The biochemical results show that the BMP and OPD peptides, grafted to an adhesive hydrogel, enhanced osteogenic differentiation, and mineralization of BMS cells. Figure 6 compares Alizarin red staining of the cell seeded hydrogels after 28 days incubation in differentiation media. The BMP peptide groups (RGD+BMP, RGD+BMP+mOPD, and RGD+BMP+OPD) exhibited intense Alizarin red staining. The alizarin red staining of the RGD control group was significantly lower than the BMP groups, consistent with the calcium content of the four groups shown in Figure 5c. 3.3. mRNA Analysis. The expression of osteogenic markers OP, OC, and ON and vasculogenic marker PECAM-1 as a function of time for the cell seeded hydrogels is shown in Figure 7a−d, respectively. In the figure, one star indicates a statistically significant difference between the BMP peptide groups and the RGD control group at the same time point while two stars indicate a significant difference between the RGD+BMP+OPD group and the other BMP groups. Overall,

Figure 6. Alizarin red S staining of mineral deposits for the BMS cells seeded in RGD-conjugated hydrogels grafted with (a) RGD control, (b) RGD+BMP peptide, (c) RGD+BMP+mOPD peptide, and (d) RGD+BMP+OPD peptide and incubated for 28 days in osteogenic media supplemented with vasculogenic factors. The squares in the images delineate the gel−pore interface of the hydrogel with welldefined pore geometry. The scale bars in images a−d are 100 μm.

OP and OC expressions increased with time while the expression of ON did not change appreciably with incubation time for all four groups. At day 14, the OP expression level of the BMP peptide groups was significantly higher than that of the RGD control group (as marked by a star in Figure 7a). After 28 days, the OP expression level of RGD, RGD+BMP, RGD+BMP+mOPD, and RGD+BMP+OPD groups showed 12.7 ± 1.3-, 17.2 ± 1.7-, 16.0 ± 1.0-, and 26.7 ± 1.0-fold increases, respectively, compared to the initial time zero. Interestingly, the expression level of the RGD+BMP+OPD group after 28 days was significantly higher than the RGD+BMP or RGD+BMP +mOPD groups. For OC, at days 14, 21, and 28, the OC expression level of the BMP peptide groups was significantly higher than the RGD control group. For example at day 28, the OC expression of RGD, RGD+BMP, RGD+BMP+mOPD, and 5392

dx.doi.org/10.1021/la205005h | Langmuir 2012, 28, 5387−5397

Langmuir

Article

Figure 8 show the nuclei, α-SMA, VE-cadherin, and the composite images, respectively, and rows 1−4 show the staining for RGD, RGD+BMP, RGD+BMP+mOPD, and RGD+BMP+OPD groups after 28 days of incubation. The RGD and RGD+BMP groups showed a weak staining for αSMA (images 1b and 2b) and practically no staining for VEcadherin (images 1c and 2c). The RGD+BMP+mOPD group showed moderate and weak staining for α-SMA (image 3b) and VE-cadherin (image 3c), respectively. On the other hand, the RGD+BMP+OPD group had strong staining for both α-SMA and VE-cadherin, as shown in images 4b and 4c. Furthermore, the α-SMA staining in the RGD+BMP+OPD group was closely associated with VE-cadherin, as shown in image 4d. The PECAM-1 staining of the groups in Figure 9 was similar to that of VE-cadherin in Figure 8 with slight differences. The RGD group showed no staining for PECAM-1 while RGD+BMP and RGD+BMP+mOPD groups showed a weak staining for PECAM-1 (images 1c, 2c, and 3c). Meanwhile, the RGD +BMP+OPD group stained strongly for both α-SMA and PECAM-1, as shown in images 4b−4d.

4. DISCUSSION The interactions of the cell with biomolecules in the ECM facilitate adhesion, proliferation, differentiation, matrix formation, and remodeling. In a previous study, we demonstrated that the covalent attachment of integrin-binding RGD peptide and the osteogenic BMP peptide enhanced osteogenic differentiation and mineralization by the seeded BMS cells.16 To test the effect of BMP and OPD peptides in a 3D geometry, PLEOF hydrogel scaffolds with well-defined cylindrical pores were fabricated by hydrogel infusion in a sacrificial wax mold followed by cross-linking and wax removal. The RGD peptide was bulk-conjugated by cross-reaction between the acrylamide group of Ac-GRGD and the fumarate groups of PLEOF. The surface density of RGD on the pore surface area of the hydrogel was 1.6 ± 0.4 pmol/cm2. Propargyl acrylate and 4-pentenal were bulk-conjugated to the hydrogel for click reaction and oxime ligation, respectively. The double bond reactivity in radical copolymerization depends on electron-donating or electron-withdrawing ability of the monomers (4-pentenal, propargyl acrylate, and Ac-GRGD) and PLEOF macromer as well as the resonance structures, polarity, and size of the substituent group on the double bond and temperature. The double bond in 4-pentenal, unlike other comonomers, is electron-rich which may affect the amount of aldehyde incorporated in the hydrogel. To determine the amount of incorporated aldehyde, the hydrogel was dried and the uncross-linked monomers and macromers were extracted with DCM. 4-Pentenal has very good solubility in DCM. Next, the extract was dried and re-dissolved in CDCl3 with 1% (v/v) TMS, and the amount of 4-pentenal was measured by 1H NMR. Analysis of the NMR chemical shift with peak position at 9.78 ppm, due to the aldehyde group of 4-pentenal, showed that >89% 4-pentenal was incorporated in the hydrogel. To graft BMP and OPD peptides, orthogonal conjugation chemistries were used.28,29 First the OPD peptide was grafted to the pore surface area of the hydrogel by oxime ligation reaction between the aminooxy moiety on the peptide and the aldehyde moiety in the hydrogel to form a stable oxime.32,45 The advantages of oxime ligation are high chemoselectivity and reaction efficiency without the need for a protecting strategy or preactivation step.32,45 Next, the BMP peptide was grafted to the hydrogel surface by click reaction between the azide moiety

Figure 7. mRNA expression level (as fold difference) of osteogenic markers OP (a), OC (b), and ON (c), and vasculogenic marker PECAM-1 of the cell seeded hydrogels as a function of incubation time in osteogenic media supplemented with vasculogenic factors. The expression level of the ARBP control gene was used as the reference, and the fold difference in expression was normalized to that at time zero. Groups include RGD control (light blue), RGD and BMP grafted peptide (RGD+BMP, green), RGD, BMP, and mutant OPD grafted (RGD+BMP+mOPD, blue), and RGD, BMP, and OPD grafted (RGD+BMP+OPD, red) hydrogels. In the panels, one star indicates a statistically significant difference (p < 0.05) between the three BMP groups (RGD+BMP, RGD+BMP+mODP, and RGD +BMP+ODP) and the control group (RGD) at the same time point. Two stars indicate a significant difference between the RGD+BMP +OPD group and other BMP groups. Error bars correspond to means ± SD (n = 3).

RGD+BMP+OPD groups showed 128 ± 16-, 390 ± 46-, 358 ± 38-, and 468 ± 78-fold increases, respectively. However, for all time points, there was not a significant difference in the OC expression among BMP groups. The OP, OC, and ON expressions are consistent with previously reported expression levels for BMS cells.43 For example, BMS cells seeded in porous collagen−glycosaminoglycan scaffolds had significantly higher expression of OP and OC after 14 and 21 days in osteogenic media.43 There was no change in ON expression of MC3T3 murine osteoblast cells seeded on tissue culture plates with incubation time.44 The expression of vasculogenic marker PECAM-1 increased with incubation time for all groups. Similar to OC, at days 14, 21, and 28, the PECAM-1 expression level of the BMP peptide groups was significantly higher than the RGD control group. At days 21 and 28, the PECAM-1 expression level of the RGD+BMP+OPD group was significantly higher than the other BMP groups. For example, at day 28, the PECAM-1 expression of RGD, RGD+BMP, RGD+BMP +mOPD, and RGD+BMP+OPD groups showed 29 ± 7-, 92 ± 21-, 99 ± 29-, and 244 ± 17-fold increases, respectively, compared to the initial time. 3.4. Immunocytochemical Analysis. Figures 8 and 9 show the immunostained images of the cell seeded hydrogels against α-SMA, VE-cadherin, and PECAM-1. Columns a−d in 5393

dx.doi.org/10.1021/la205005h | Langmuir 2012, 28, 5387−5397

Langmuir

Article

Figure 8. Effect of BMP and OPD grafting of RGD-conjugated hydrogels on the expression of vasculogenic markers for the cell seeded hydrogels after 28 days of incubation. Fluorescent images are for nuclei (blue), α-SMA (green), VE-cadherin (red), and the composite of the three (far right). Groups include RGD control, RGD+BMP, RGD+BMP+mOPD, and RGD+BMP+OPD grafted hydrogels.

on the peptide and propargyl moiety in the hydrogel.46 The click reaction does not cross-react with the aminooxy moiety in the hydrogel, and other functional groups in the BMP peptide (amino and carboxyl groups) do not interfere with the click reaction.46 Subsequent to grafting, the OPD density on the hydrogel surface was measured using a lysine-terminated OPD peptide and FITC−dextran. The grafting densities were measured for each peptide individually. The MW of OPD and BMP peptides are 1067 and 2444 Da, respectively, corresponding to equivalent sizes of 1.5 and 2.0 nm, respectively. Therefore, the areas occupied by OPD and BMP peptides are 1.8 and 3.2 nm2, respectively (2.5 nm2 average area per peptide). The amount of 4-pentenal (for OPD grafting) and propargyl acrylate (for BMP grafting) was 1.5 mg/(1170 mg of the gel) (see Experimental Section), corresponding to 0.26% OPD or BMP peptide on the hydrogel surface. This means that the average distance between two grafted peptides was approximately 1 μm, much larger than the average size of the peptides (1.5−2.0 nm). Since the “click” and “oxime” reactions do not cross-react and separation distance between the surface reactive moieties was 1000× higher than the size of the reactive moieties, the effect of first peptide grafting on the grafting efficiency of the second peptide was minimal. A factor that can complicate the measurements is diffusion of FITC−dextran in the hydrogel, leading to a higher than expected value for OPD surface density. We and others have shown that FITC−dextran (20 kDa MW and stokes’ radius ∼ 3.3 nm) has slow diffusion in PEG-based hydrogels.11,16 The OPD peptide concentrations > 14 mM did not significantly increase grafting density (see Figure 4a), indicating that the

grafting reaction was limited by the density of the aldehyde moiety on the hydrogel surface. At the same solution concentration, the OPD peptide density was approximately twice that of BMP peptide, which was attributed to the smaller size and higher solubility, thus less steric hindrance, of aminooxy-mPEG-OPD peptide (MW = 1068 Da) compared to azide-mPEG-BMP (MW = 2445 Da). The biological response of the grafted peptides depends on their surface density. We previously demonstrated that osteogenic differentiation of BMS cells was significantly enhanced with grafted BMP peptide density of ∼5 pmol/cm2.16 Hamada and collaborators reported that a collagen scaffold loaded with 100 ng/mL OPD peptide induced vasculogenic differentiation of cells derived from the bone marrow,23 comparable to the effective concentration of the BMP peptide.16 Therefore, OPD grafting density (13.8 pmol/cm2) comparable to that of BMP peptide (5.4 pmol/cm2) was used in this study. The results demonstrate that the above OPD and BMP grafting densities significantly enhance osteogenic and vasculogenic differentiation of the BMS cells. The effect of BMP and OPD grafting density on the extent of BMS cell differentiation is under investigation and will be reported in the future. Another issue is the stability of the grafted OPD peptide toward enzymatic degradation. Chan and collaborators reported that soluble OPD peptide was unstable in 100% human serum.47 They also reported that the OPD peptide was stabilized by grafting to a disulfide-rich cyclic peptide scaffold.47 To test the stability of the OPD peptide grafted to the hydrogel, the fluorescent dye dansyl chloride (Sigma-Aldrich) was conjugated to the free amine group of the peptide prior to 5394

dx.doi.org/10.1021/la205005h | Langmuir 2012, 28, 5387−5397

Langmuir

Article

Figure 9. Effect BMP and OPD grafting of RGD-conjugated hydrogels on the expression of vasculogenic markers for the cell seeded hydrogels after 28 days of incubation. Fluorescent images are for nuclei (blue), α-SMA (green), PECAM-1 (red), and the composite of the three (far right). Groups include RGD control, RGD+BMP, RGD+BMP+mOPD, and RGD+BMP+OPD grafted hydrogels.

grafting. After grafting, the gel was incubated in primary media supplemented with vasculogenic and osteogenic factors. At each time point after washing the gel thoroughly with PBS, the gel fluorescence was measured with a plate reader (Synergy HT, 557 nm wavelength) and compared with the intensity at zero time. If the grafted OPD degraded at any of the peptide’s amide bonds, the dansyl chromophore would be detached from the hydrogel, resulting in a reduction in fluorescent intensity. On the basis of fluorescent measurements, 58 and 47% of the OPD peptide was retained by the gel after 4 and 7 days of incubation, respectively. Therefore, a significant fraction of the OPD peptide was still attached to the gel after 7 days of incubation. It should be noted that the oxime linkage is susceptible to hydrolysis under strong basic (pH > 9) or acidic (pH < 1) conditions. However, Singh and collaborators reported that the linkage is stable in the pH 4−7 range. They reported