- Email: [email protected]
Heparin-functionalized chitosan scaffolds for bone tissue engineering
Carbohydrate Research 346 (2011) 606–613
Contents lists available at ScienceDirect
Carbohydrate Research journal homepage: www.elsevier.com/locate/carres
Heparin-functionalized chitosan scaffolds for bone tissue engineering Menemsße Gümüsßdereliog˘lu ⇑, Sezin Aday Hacettepe University, Chemical Engineering and Bioengineering Departments, 06800 Beytepe, Ankara, Turkey
a r t i c l e
i n f o
Article history: Received 16 September 2010 Received in revised form 6 December 2010 Accepted 8 December 2010 Available online 14 December 2010 Keywords: Heparin Chitosan Osteogenic activity Electrostatic interactions Covalent immobilization
a b s t r a c t The aim of this study is to investigate the effects of heparin-functionalized chitosan scaffolds on the activity of preosteoblasts. The chitosan scaffolds having the pore size of 100 lm were prepared by a freeze-drying method. Two different methods for immobilization of heparin to chitosan scaffolds were successfully performed. In the first method, functionalization of the scaffolds was achieved by means of electrostatic interactions between negatively charged heparin and positively charged chitosan. The covalent immobilization of heparin to chitosan scaffolds by 1-(3-dimethylaminopropyl)-3-ethylcarbodiimide (EDAC) and N-hydroxysuccinimide (NHS) was used as a second immobilization method. Morphology, proliferation, and differentiation of MC3T3-E1 preosteoblasts on heparin-functionalized chitosan scaffolds were investigated in vitro. The results indicate that covalently bound heparin containing chitosan scaffolds (CHC) stimulate osteoblast proliferation compared to other scaffolds, that is, unmodified chitosan scaffolds (CH), electrostatically bound heparin containing chitosan scaffolds (EHC), and CH+free heparin (CHF). SEM images also proved the stimulative effect of covalently bound heparin on the proliferation of preosteoblasts. Alkaline phosphatase (ALP) and osteocalcin (OCN) levels of cells proliferated on CHC and EHC were also higher than those for CH and CHF. In vitro studies have demonstrated that chitosan scaffolds increase viability and differentiation of MC3T3-E1 cells especially in the presence of immobilized heparin. Ó 2010 Elsevier Ltd. All rights reserved.
1. Introduction Although heparin has been used as an anticoagulant for a great many years, much more attention has recently been given to its unique role in other biological processes. It is known that heparin can interact with a great number of proteins related to adhesion (e.g., fibronectin and vitronectin), proliferation (e.g., fibroblast growth factor, FGF)1–3 or osteogenic differentiation (e.g., bone morphogenic proteins, BMPs).4–7 It induces oligomerization of FGF molecules, which is an important step for receptor activation8 and helps BMPs to remain in the extracellular space for longer periods while preventing ligand inactivation as well.7 It is also suggested that heparin enhances BMP activities by making complexes.6 Because of its special interactions with proteins, heparin can be used as a delivery system for growth factors.1 In the literature, there are conflicting results related to the effect of heparin on cellular proliferation. Some studies have indicated that heparin increases endothelial cell proliferation in the presence of heparin-binding growth factors,9,10 which others claimed that it decreases cell proliferation or it has no effect at all.11,12 In addition, it
⇑ Corresponding author. Tel.: +90 312 2977447; fax: +90 312 2992124. E-mail address: [email protected] (M. Gümüsßdereliog˘lu). Present address: Biomaterials and Stem Cell-Based Therapeutics Group, BiocantCenter of Innovation and Biotechnology, Nucleo 4, Lote 3, 3060-197 Cantanhede, Portugal.
0008-6215/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.carres.2010.12.007
has been shown that glycosaminoglycans (GAGs) inhibit the proliferation of smooth muscle cells.13,14 However, it is known that chitosan membranes can stimulate endothelial regeneration by stimulating cell proliferation.14 Since incomplete endothelization and smooth muscle cell hyperplasia are the major problems for small vascular grafts,15,16 heparin–chitosan complexes can be used in the bone regeneration process that requires neo-vascularization. In addition, it has been indicated that heparin-modified polyethylene glycol (PEG) gels promote human mesenchymal stem cell (hMSC) adhesion, spreading, and proliferation.1 In our previous study, it was shown that heparin-immobilized non-woven polyester fabric (NWPF) discs supported proliferation of L929 fibroblasts in lowserum media.17 Chitosan is a hydrophilic, non-toxic, biodegradable, antibacterial, and biocompatible polymer that can be widely used in the biomedical area to develop tissue substitutes, delivery vehicles, and membranes for hemodialysis.18–20 In the literature, water-soluble chitosan/heparin ointments were studied as wound healing accelerators.21 Jeon et al.5 prepared heparin-conjugated nanospheres to enhance long-term delivery of BMP-2. Zhu et al.22,23 used expanded polytetrafluoroethylene (ePTFE)/chitosan and polylactic acid (PLA)/chitosan surfaces in order to immobilize heparin using ionic interactions between chitosan and heparin. Taking the advantage of the natural binding mechanism of heparin to design bioactive materials is promising due to the fact that this method prevents growth factor damage or degradation.1 Additionally, since different growth factors can bind to heparin, it is possible
M. Gümüsßdereliog˘lu, S. Aday / Carbohydrate Research 346 (2011) 606–613
to obtain dual or multiple growth factor-releasing systems by using heparin.24 In the present study, the aim was to develop heparin-immobilized chitosan scaffolds for bone tissue engineering applications. Our results are important in order to understand the effects of different forms of heparin, that is, electrostatically or covalently immobilized and the free form, on cell proliferation and differentiation. In addition, since no additional growth factor was used in our study, we were able to mimic the natural environment of the cells more precisely by using just the interactions between heparin and different growth factors released during proliferation and differentiation of MC3T3-E1 cells.
2. Results and discussion 2.1. Preparation and characterization of chitosan scaffolds Scaffolds, which are widely used in tissue engineering, are required to improve cell proliferation and differentiation, to prompt cells to obtain 3D tissues and to reinforce the regenerating tissue.25,26 It is known that scaffolds having a pore diameter greater than 100 lm can enhance osteoblast growth.27 In this study, porous chitosan scaffolds that have >85% DD with 2% (w/v; weight of chitosan in volume of acetic acid) composition were prepared by a freeze-drying method. SEM examination of these scaffolds indicated a highly interconnected morphological structure with a pore size of 100 lm. Since osteoblasts are 10–30 lm long, this pore size allows cells to migrate into the chitosan ‘sponges’. The porosity of the scaffolds calculated as 82 ± 1.7%. Further details of these scaffolds can be found in our previous studies.28,29 2.2. Heparin immobilization The scaffolds described above were modified with heparin to take the advantage of the unique properties of this molecule. Heparin is a member of the glycosaminoglycan family of polysaccharides (GAGs) that are responsible for regulating a great number of functions such as localization of mitogenic factors,30 bioactivity of growth factors,10 and cellular proliferation.14,31 Additionally, it is known that extracellular matrices (ECMs) are composed of different GAGs such as chondroitin sulfate, heparan sulfate, and hyaluronate. Different structural characteristics of GAGs (e.g., arrangement and degree of sulfation) or their size and location can effect the interactions of these molecules with other molecules, and, additionally, their activity.31,32 When all these important points are taken into consideration, it can be concluded that it is important to use immobilized heparin to mimic the cellular environment more closely. In our study, to find the best method for heparin immobilization, different buffers, stabilization, and immobilization methods were compared. Modified scaffolds were evaluated using the Toluidine Blue test which depends on the metachromatic interactions between heparin and the dye. In this test, higher concentrations of heparin give lower absorbances because of the extraction of heparin–dye complexes into the organic phase (n-hexane). The best test results were obtained when the scaffolds were stabilized by ethanol and covalently modified with heparin in MES buffer containing EDAC and NHS. These scaffolds did not lose their shapes and swelling characteristics in neutral aqueous solutions in contrast to scaffolds stabilized with NaOH. After these preliminary tests, ethanol-stabilized and heparin-functionalized scaffolds were further evaluated. Two different immobilization methods, that is, electrostatic interaction and covalent immobilization, were chosen to determine the effect of different modification methods on heparin release and cellular activity. Although heparin solutions
607
having concentrations of more than 1% (w/v) were used in the literature,3,33 we chose the initial heparin concentration for immobilization as 0.5% (w/v) according to the results of our preliminary studies. The mechanisms of immobilization methods are given in Figure 1. In the covalent method, heparin was immobilized onto scaffolds through the interaction between the carboxyl groups on heparin activated by NHS/EDAC and the amino groups of chitosan (Fig. 1A). Chitosan’s cationic nature and high charge density make it a potential polymer for biomaterial applications. Due to its charge density, it is possible to obtain insoluble ionic complexes of chitosan with anionic polysaccharides.34 As it can be seen in Figure 1B, the electrostatic immobilization of heparin was achieved using negative groups on heparin and positive groups on chitosan. The amount of heparin immobilized on the scaffolds was determined by the Toluidine Blue test, and the results are given as milligram heparin per gram of chitosan scaffold. These values were 3.8 ± 0.8 mg and 4.7 ± 0.4 mg for electrostatically (EHC) and covalently modified scaffolds (CHC), respectively. Heparin modification was further proved by elemental analysis. It is known that chitosan does not contain sulfur in its structure, while heparin is a highly sulfated polysaccharide. Elemental analysis results indicated that after modification of scaffolds, the percentage of sulfur in the complex was increased about 7–10-fold (1.478% and 2.044% for EHC and CHC, respectively) compared to unmodified CH scaffolds (0.210% from contamination). In addition, the nitrogen percentage of the scaffolds modified covalently with heparin (7.41%) was higher than that for unmodified scaffolds (6.60%). In general, it is known that covalent bonding has several advantages compared to other functionalization techniques (e.g., control of molecular orientation, minimization of nonspecific interactions, and prevention of the dissolution, desorption and degradation of the molecule).35 In our study, it was shown that covalent bonding supported the long-term stability of heparin, and even after 20 days, the amount of heparin released from the scaffolds was not significant enough to be determined using the Toluidine Blue test. However, it was reported that heparin can be easily released from the surface when it is immobilized by physical or ionic bonding.20 Since the charge density of chitosan is pH dependent, it is possible to use chitosan as a delivery vehicle for polyanions by taking advantage of the dissociation of the chitosan–polyanion complexes under physiological conditions.22,23 Similarly, in our study, when electrostatic interactions were used, 45% of immobilized heparin was released in the first three days, and during 20 days 71% of heparin was released (Fig. 2). The integration of heparin into the structure affected the mechanical properties of the scaffolds in a positive way. The compression modulus of scaffolds increased significantly, and the values of 1.993 ± 0.035 N/mm2 and 2.853 ± 0.429 N/mm2 were obtained for the CH and CHC scaffolds, respectively. 2.3. Cell culture studies In tissue engineering approaches three major steps, namely cell attachment, proliferation, and differentiations are crucial for regeneration of tissues. The MTT assay, which is a colorimetric test based on the reduction of a yellow tetrazolium salt into a purple formazan product, is widely used to determine cellular toxicity, proliferation or viability. Figure 3 shows the MTT assay results of the different scaffolds examined. As can be seen from this figure, CHC scaffolds showed relatively higher absorbances during the culture period. There was no significant difference among the MTT values for CH and EHC scaffolds and free heparin up to the 12th day of the culture. The early stimulative effect of heparin-immobilized scaffolds can be explained by initial attachment of the cells. It is known that serum
608
M. Gümüsßdereliog˘lu, S. Aday / Carbohydrate Research 346 (2011) 606–613
Figure 1. Schematic diagram showing the immobilization of heparin to chitosan scaffolds: (A) covalent immobilization, (B) electrostatic immobilization.
Figure 2. Heparin release graph for the scaffolds prepared using electrostatic interactions (in PBS, pH 7.4, 37 °C).
contains some attachment factors such as fibronectin and vitronectin that have important roles in cellular attachment onto surfaces. Since heparin can interact with these proteins,1 it can be expected that heparin-immobilized scaffolds will adsorb higher amounts of these proteins compared to unmodified scaffolds. Although this stimulative effect was maintained during the culture period for CHC scaffolds, EHC scaffolds did not induce the cell proliferation during the culture period. The low cell numbers on the EHC scaffolds can be related to desorption of electrostatically bound heparin from the chitosan scaffolds. Heparin release studies indicated that more than 50% of the immobilized heparin was released from EHC during the first 10 days of culture. However, no significant release profile was obtained for covalently immobilized heparin. Chupa et al.14 claimed that the desorption of GAGs may prevent the adsorption of serum components such as fibronectin and collagen. In addition, depending on the concentrations or ionic environment, the binding interactions with heparin can enhance or inhibit growth factor activity.14 High
Figure 3. MTT results for different chitosan scaffolds (statistically significant differences, n = 3, ⁄: p <0.05, ⁄⁄: p <0.001. Unmodified chitosan scaffold is control group).
M. Gümüsßdereliog˘lu, S. Aday / Carbohydrate Research 346 (2011) 606–613
levels of soluble GAGs present in the culture may bind and sequester growth factors that are either supplied by the medium or secreted endogenously. Although it is claimed that cells on chitosan–GAG membranes (2D structures) exhibit rounded shapes,14,31 SEM characterization of the scaffolds at the 4th and 11th days of culture indicated that cells on chitosan–heparin scaffolds attached well and showed a spread-out morphology (Fig. 4). These results were consistent with the ones obtained from 3D GAG–chitosan matrice studies in the literature.31 Similarly, Zhu et al.22 also showed that the attachment of fibroblastic cells increases significantly on chitosan and chitosan/ heparin surfaces. The cell numbers on the electrostatically modified scaffolds were lower as expected from the MTT results (Fig. 4A and D). The distribution of the cells in the 3D structure of the chitosan scaffolds was examined by CLSM on the 8th day of culture (Fig. 5). Alexa Fluor 488 phalloidin was used to stain F-actin (green), and propidium iodide was used for staining of cell nuclei (red). Figure 5 was obtained by the Zeiss LSM Image Browser Program, and using these images, the location, interaction, and density of the cells were determined. Confocal microscopy photos demonstrated that MC3T3-E1 cells migrated into the scaffolds, and cells were observed even at a depth of 100 lm from the surface. It can be clearly seen that CHC scaffolds induced cellular proliferation, and the
609
highest cell densities were obtained for the cells on these scaffolds (Fig. 5C). ECM has different functions such as being a reservoir for endogenous growth factors, localizing growth factors, and preventing their degradation.4,36 The ECM of osteoblasts is composed of collagenous, non-collagenous proteins, growth factors, and BMPs.37 Shibata et al.38 found that soluble heparin stimulates collagen synthesis in MC3T3-E1 cell cultures. In preosteoblasts, BMP-containing ECM can induce alkaline phosphatase activity, which is an early differentiation marker of osteoblast phenotype.37,39 Because differentiation is the third major step of regeneration in vitro, we evaluated cells also in terms of differentiation markers such as ALP activity and OCN secretion. ALP activity of the MC3T3-E1 cells was evaluated by measuring absorbances of p-nitrophenol at 405 nm ( Fig. 6A). The absorbance values increased with time for all scaffolds except those with free heparin, and the values were significantly different for all the samples compared to the ones for the CH scaffolds. Similar to proliferation studies, the highest values were obtained for heparin-immobilized scaffolds. Although MTT values were low for the cells on the EHC scaffolds, these scaffolds induced the differentiation of MC3T3-E1 preosteoblasts. Similarly, the highest OCN values were obtained when EHC and CHC scaffolds were used (Fig. 6B).
Figure 4. SEM images of cells on different chitosan scaffolds at the 4th and 11th days of culture: (A) EHC, 4th day, 3000; (B) CHC, 4th day, 1530; (C) CHF, 4th day, 1500; (D) EHC, 11th day, 5000; (E) CHC, 11th day, 2500; (F) CHF 11th day 5000. Scale bars indicate 10, 20, 10, 2, 10 and 20 lm, respectively.
610
M. Gümüsßdereliog˘lu, S. Aday / Carbohydrate Research 346 (2011) 606–613
Figure 5. 3D CLSM images of MC3T3-E1 cells obtained by the Zeiss LSM Image Browser Program on the 8th day of culture. Left: unmodified chitosan scaffold (CH); middle: unmodified chitosan scaffold + free heparin (CHF); right: covalently heparin-immobilized chitosan scaffolds (CHC). Alexa Fluor 488 phalloidin was used to stain F-actin (green) and propidium iodide was used for staining of cell nuclei (red). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
Figure 6. ALP activities (A) and osteocalcin values (B) of cells cultured on different chitosan scaffolds (statistically significant differences, n = 3,⁄: p <0.001, ⁄⁄: p <0.01. Unmodified chitosan scaffold is control group).
In order to activate BMP-dependent signal transduction, which is important for osteogenic differentiation, the ligand should bind to the Type II BMP receptor, which then associates with and phosphorylates the Type I receptor.37 It is possible to sequester and localize BMPs produced by MC3T3-E1 preosteoblasts by introducing the heparin functionality to the scaffolds. Heparin binds cellsecreted BMP, increases its local availability, and by this way helps to sustain autocrine signaling effects.4 For our study, there are two
possible theories to explain the effects of different immobilization methods on cellular differentiation. According to the first theory, heparin-functionalized scaffolds sequestered BMP produced by cells and facilitated BMP availability and BMP receptor activation.4,37 Because of this effect of heparin, higher ALP and OCN values were obtained for the cells on heparin-modified scaffolds. It is known that there is an important relation between ECM and BMP activity, and cells should be in contact with a collagen-containing
M. Gümüsßdereliog˘lu, S. Aday / Carbohydrate Research 346 (2011) 606–613
ECM before they differentiate.40 Xiao et al.37 found that in the absence of ECM, very little BMP activity is observed for MC3T3-E1 preosteoblast cells. However, in the presence of ECM, addition of BMP stimulates OCN, and bone sialoprotein gene expression.37 The well-organized ECM secreted by cells on CHC scaffolds (Fig. 4 E) and higher cell numbers ( Fig. 3) can be the reason of higher ALP activities for these scaffolds. However, lower OCN values were obtained for the cells on the CHC scaffolds compared to those on the EHC scaffolds. As a second theory, we may think that the adsorption of matrix proteins (e.g., collagen and fibronectin) that were secreted onto CHC scaffolds during culture caused a feedback inhibition.1 In the literature, better hemocompatibility due to the progressive release of heparin was shown where ionic bonds were used to immobilize heparin onto surfaces.41 In our study, there was a continuous release from EHC surfaces. Another reason for the stimulation of osteogenic differentiation on EHC scaffolds can be the fact that heparin enhances the BMP activities by forming complexes with BMPs in the culture media, and in this way the ligands are continuously made available to their receptors.6 However, long-term delivery of heparin can be also important when osteogenic efficacy of in vivo studies is considered.5 Since CHC scaffolds also stimulate cellular proliferation, it is possible to have a fast and accurate regeneration when these scaffolds are used. 3. Conclusion Heparin, which has long been used as an antithrombogenic agent, has a great impact on designing novel scaffolds for tissue engineering due to its unique characteristics and its involvement in key biological processes. In the present study, porous chitosan scaffolds prepared by a freeze-drying method were further modified with heparin by covalent immobilization or electrostatic interactions. The results showed that these immobilization methods affected cellular activity through different mechanisms of action. Cell-culture studies demonstrated that, while covalently immobilized heparin-containing scaffolds were inducing both cell proliferation and differentiation, electrostatically-immobilized heparin-containing scaffolds supported cellular differentiation. In conclusion, these novel scaffolds can stimulate osteogenic regeneration by mimicking the natural ligand–receptor mechanisms more closely, and thus they should have great advantages when used in the bone tissue engineering field. 4. Experimental 4.1. Materials Chitosan and non-fractionated heparin sodium salt (202 USP units mg 1) were purchased from Sigma–Aldrich (Germany). The degree of deacetylation (DD) was >85% for chitosan, and the molecular weight interval of deacetylated chitin was 190–375 kDa. Acetic acid (HPLC grade) and ethanol (96% v/v) were obtained from Riedel-de Haën (Germany). 1-Ethyl-3-(dimethylaminopropyl)carbodiimide (EDAC), Toluidine Blue and 2-N-(morpholinoethanesulfonic acid) (MES) were obtained from Sigma–Aldrich (Germany). N-Hydroxysuccinimide (NHS) was obtained from Fluka (ABD). Glutaraldehyde, p-nitrophenyl phosphate liquid substrate system (pNPP) and hexamethyldisilazane were purchased from Sigma– Aldrich (USA). For cell culture studies, Minimum Essential Medium Alpha Modification (a-MEM), fetal bovine serum (FBS) and trypsin/ EDTA solution were obtained from Hyclone (USA). 1-(4,5-Dimethylthiazol-2-yl)-3,5-diphenylformazan (MTT), b-glycerol phosphate and ascorbic acid were obtained from Sigma–Aldrich (USA). Triton X-100 from Sigma–Aldrich (USA) and 2-amino 2-methyl1,3-propanol from Sigma–Aldrich (Germany) were used in the alkaline phosphatase (ALP) assay. For the osteocalcin (OCN) assay,
611
formic acid was purchased from E. Merck (Germany) and the OSTEASIA kit was obtained from BioSource (Belgium). 4.2. Preparation of chitosan scaffolds Porous chitosan scaffolds were prepared using a freeze-drying method developed in our group’s previous study.28 Briefly, chitosan solutions with concentrations of 2% (w/v) were prepared by dissolving of chitosan in 0.2 M acetic acid with filtration to eliminate the impurities. The solution was poured into 24-well tissueculture polystyrene dishes (TCPS, Nunc, Denmark) and frozen at 20 °C for 24 h. The frozen samples were then transferred into the freeze-dryer (Christ, Germany) and lyophilized at 80 °C until they were completely dried. Freshly lyophilized scaffolds were soaked in ethanol solutions (96% v/v ethanol for overnight and 70% v/v ethanol for 1 h) for stabilization and then lyophilized at 80 °C. The cylindrical porous scaffolds were then cut into 2-mm thick sections with a diameter of 10 mm. 4.3. Heparin immobilization Heparin-immobilized chitosan scaffolds were obtained by the use of two different methods, that is, covalent immobilization and electrostatic interaction. In the first method, heparin immobilization to scaffolds was achieved using the amino groups of chitosan and the carboxyl groups of heparin. For this purpose, stabilized chitosan scaffolds were immersed in 50 mL of MES buffer containing 0.5% (w/v) heparin (202 USP units/mg), 0.40 g of EDAC and 0.097 g of NHS and incubated overnight at +4 °C. After heparin immobilization, the scaffolds were washed with ultrapure water for 4–5 times and subsequently freeze-dried. On the other hand, electrostatic modification was performed using the ionic interactions between chitosan and heparin. In this approach, stabilized scaffolds were incubated overnight at +4 °C in phosphate buffer saline (PBS, pH 7.4) containing 0.5% (w/v) heparin. The effects of buffers, immobilization, and stabilization methods were evaluated prior to selection of the final immobilization procedures. In these experiments, besides MES buffer, sodium citrate buffer (0.2325 g citric acid. H2O and 0.4075 g sodium citrate in 250 mL pure water) containing 0.40 g EDAC and 0.097 g NHS was used. To determine the effects of the stabilization method, the scaffolds stabilized by NaOH were used as well as the ones stabilized by ethanol. For NaOH stabilization, the freshly lyophilized scaffolds were immersed in 0.4 M NaOH solution (50% ethanol and 50% water) for 24 h and washed with pure water for 4–5 times. Then heparin was immobilized to these scaffolds electrostatically and covalently by using the same methods described above. 4.4. Characterization studies The amount of heparin immobilized to chitosan scaffolds was determined by the Toluidine Blue assay.42 Toluidine Blue (25 mg) was dissolved in 0.01 N hydrochloric acid containing 0.2 wt % NaCl. Heparin solutions with different concentrations (2 mL) were added to the Toluidine Blue solution (3 mL) and vortexed for 30 s. To this solution, 3 mL of n-hexane was added, and the mixture was shaken well to extract the Toluidine Blue–heparin complex. The absorbance of the aqueous phase containing the unextracted Toluidine Blue was measured by a UV spectrophotometer at 631 nm. A series of standard heparin solutions (0–20 lg mL 1) were used to prepare a calibration curve. In order to determine the amount of heparin immobilized onto the scaffolds, the scaffolds were immersed in a mixture of Toluidine Blue solution (3 mL) and water (2 mL) for 30 min. Then n-hexane (3 mL) was added to ensure uniformity of the treatment.
612
M. Gümüsßdereliog˘lu, S. Aday / Carbohydrate Research 346 (2011) 606–613
After shaking well, scaffolds were removed from the solution, and the absorbance was measured at 631 nm. The amount of immobilized heparin was calculated using a calibration curve. In vitro release profile and stability of heparin were investigated at pH 7.4 using PBS solution. Scaffolds were incubated in tubes containing 1 mL of release medium (PBS), and the tubes were placed in a horizontal shaker with a speed of 20 rpm at 37 °C. At specific time intervals, the release medium was replaced, and the amount of heparin remaining on scaffolds was determined by the Toluidine Blue test as described previously. To indicate the heparin integrated to the structure of chitosan scaffolds, elemental analysis of unmodified and heparin-immobilized chitosan scaffolds was performed at Tübitak Atal (Ankara Test and Analysis Laboratory, Turkey). The scaffolds were minced into pieces using a lancet, and elemental analyses (for carbon, hydrogen, nitrogen and sulfur) were performed using a vario MICRO CHNS instrument (Elementar Analysensysteme GmbH, Germany). Compression tests were performed at constant crosshead speed of 5 mm/min with a load of 500 N cell (Lloyd Ins., UK). Test samples of chitosan scaffolds having a 2 mm thickness were fully rehydrated in PBS (pH 7.4) at 37 °C for 1 h before material testing. The initial compressive modulus was calculated from the slope of the linear region in the stress–strain curves for both the untreated and heparin-immobilized scaffolds. 4.5. Cell culture studies The MC3T3-E1 cell line originating from fetal mouse calvaria was obtained from RIKEN Cell Bank (Saitama, Japan). The cells were subcultured in flasks using alpha-MEM supplemented with 10% (v/v) FBS and 1% (v/v) antibiotic solution (penicillin–streptomycin, Sigma–Aldrich, USA). The cells, maintained at 37 °C in a humidified CO2 (5%) atmosphere (Heraus Instruments, Germany), were dissociated with 0.25% trypsin–EDTA, centrifuged and resuspended in the medium prior to cell seeding. The cell-culture studies were carried out in Parafilm-lined 24well TCPS in stationary conditions. For sterilization, the Parafilmlined wells and chitosan scaffolds were soaked into 70% ethanol for 30 min and then placed under UV light for 45 min. Then the scaffolds were equilibrated in sterile Dulbecco’s PBS (pH 7.4, 24 h) and immersed in conditioning medium for 1 h prior to cell seeding. The MC3T3-E1 cells were trypsinized from the flasks, and a cell suspension was obtained by the addition of fresh differentiation medium similar to the culture medium stated above supplemented with 2 mM L-glutamine (Sigma–Aldrich, USA), 10 mM b-glycerol phosphate and 50 lg mL 1 of L-ascorbic acid. A cell suspension (1.5 mL) containing 1 105 cells mL 1 was seeded onto the scaffolds. Cells were cultivated in 5% CO2 at 37 °C in 95% relative atmospheric humidity for 21 days. The differentiation medium (0.5 mL) was replenished every 3 days. Four types of chitosan scaffolds, which were stabilized with ethanol, were used for cell culture experiments: (1) Unmodified chitosan scaffolds (CH). (2) Electrostatically bound heparin containing chitosan scaffolds (EHC). (3) Covalently bound heparin containing chitosan scaffolds (CHC). (4) Unmodified chitosan scaffold + free heparin in culture medium (CHF). To determine the effect of free heparin on cellular behavior, cells were seeded onto untreated chitosan scaffolds, and cell culture studies were performed using the same differentiation medium enriched with 10 USP units mL 1 heparin. 4.5.1. MTT assays The mitochondrial activity of the MC3T3 preosteoblasts on the scaffolds was quantitatively assessed with the 3-[4,5-dimethylthiazol-2-yl]-diphenyltetrazolium bromide (MTT) test at different culture periods up to 16 days. At selected time intervals, the culture medium was aspirated, and scaffolds were washed with
900 lL of prewarmed PBS (pH 7.4). Serum-free culture medium (900 lL) supplemented with 90 lL MTT solution (2.5 mg MTT dissolved in 1 mL PBS) was added to each sample. After incubating at 37 °C for 3 h, the medium was replaced with 400 lL of 0.04 M HCl in 2-propanol to dissolve the formazan crystals. The solution was transferred into a 96-well micro-ELISA plate, and the absorbance was read at 540 nm with a plate reader (ASYS Hitech UVM 340 Plate Reader, Eugendorf, Austria). 4.5.2. Cellular morphology The cytoskeleton organization of the cells within the chitosan scaffolds was assessed by confocal laser scanning microscopy (CLSM, Zeiss LSM 510, Germany) on the 8th day of the culture. For this purpose, the scaffolds were transferred to a new 24-well TCPS and washed with PBS (pH 7.4) three times at room temperature. Samples were fixed with 2.5% (v/v) glutaraldehyde in 0.1 M PBS for 10 min at 4 °C. After fixation, scaffolds were stored at 4 °C in PBS solution until observation by CLSM. Prior to observation, fixed cells were permeated by 0.1% Triton X-100 solution for 5 min. Cell cytoskeletal filamentous actin (F-actin) was visualized by treating the cells with 2.5% (v/v) Alexa Fluor 488 phalloidin (Invitrogen Inc., USA) for 20 min in darkness. Nuclei were stained with 10 lg mL 1 propidium iodide for 5 min. The attachment, spreading, and morphology of the MC3T3 cells cultured on heparin-immobilized and untreated chitosan scaffolds were observed by a scanning electron microscope (SEM) (Zeiss Evo 50, Germany). After the culture medium was removed, the scaffolds were gently washed with PBS, and the cells were fixed with 2.5% (v/v) glutaraldehyde in 0.1 M PBS (pH 7.4) for 30 min at 4 °C. Then the scaffolds were dehydrated in ethanol series (30%, 50%, 70%, 90%, and 100% (v/v)) and substituted with hexamethyldisilazane (HMDS) before being dried. After the scaffolds were completely dry, samples were coated with a gold–palladium layer prior to SEM analysis. 4.5.3. Cell differentiation Alkaline phosphatase (ALP) activities of the cells were evaluated on the 15th and 21st days of culture. At the end of each culture period, scaffolds were rinsed twice with PBS and stored at 80 °C. After freeze-drying, samples were cut down with a scissors and homogenized by brief sonication at 0 °C in the lysis buffer (1% Triton X-100). Cell lysates were centrifuged at 12,000 rpm for 10 min at 4 °C. The supernatant was assayed for ALP activity using p-nitrophenyl phosphate (pNPP) as substrate. About 50 lL of Triton lysate was added to 125 lL of ALP solution (56 mM 2-amino-2methyl-1,3-propanediol (pH 9.8) and 1 mM MgCl2 containing 10 mL pNPP), and the mixture was incubated at 37 °C for 30 min. The reaction was stopped by adding 50 lL of 2.5 M sodium hydroxide, and the ALP activity was determined by the measurement of absorbance of p-nitrophenol (pNP) at 405 nm using a microplate reader (Asys UVM 340, Australia). To determine the osteocalcin (OCN) content of cells cultured on scaffolds at the 15th and 21th days of culture, the scaffolds were washed three times with PBS and then treated with 1 mL of 40% (v/v) formic acid for 10 days at 4 °C using vortex mixing for decalcification. The resulting solution was freeze-dried and stored at 80 °C. Osteocalcin content was determined using the commercially available solid-phase enzyme-amplified sensitivity immunoassay OST-EASIA kit (BioSource, Belgium) according to the supplier’s instructions. 4.6. Statistical analysis All experiments were done in triplicates and results were presented as means ± standard deviation (SD) for n = 3. Statistical comparisons were performed using GraphPad Software Instant.
M. Gümüsßdereliog˘lu, S. Aday / Carbohydrate Research 346 (2011) 606–613
Acknowledgments This study was financially supported by the Turkish Scientific _ and Research Council (TÜBITAK) Project No. 108M346. S.A. was _ _ supported by the TÜBITAK-BIDEB 2210 National Scholarship Program for MSc Students. References 1. Benoit, D. S. W.; Anseth, K. S. Acta Biomater. 2005, 1, 461–470. 2. Nie, T.; Baldwin, A.; Yanaguchi, N.; Kiick, K. L. J. Controlled Release 2007, 122, 287–296. 3. Wissink, M. J. B.; Beernink, R.; Scharenborg, N. M.; Poot, A. A.; Engbers, G. H. M.; Beugeling, T.; van Aken, W. G.; Feijen, J. J. Controlled Release 2000, 67, 141–155. 4. Benoit, D. S. W.; Colins, S. D.; Anseth, K. S. Adv. Funct. Mater. 2007, 17, 2085– 2093. 5. Jeon, O.; Song, S. J.; Yang, H. S.; Bhang, S. H.; Kang, S. W.; Sung, M. A.; Lee, J. H.; Kim, B. S. Biochem. Biophys. Res. Commun. 2008, 369, 774–780. 6. Takada, T.; Katagiri, T.; Ifuku, M.; Morimura, N.; Kobayashi, M.; Hasegawa, K.; Ogamo, A.; Kamijo, R. J. Biol. Chem. 2003, 278, 43229–43235. 7. Zhao, B.; Katagiri, T.; Toyoda, H.; Takada, T.; Yanai, T.; Fukuda, T.; Chung, U.; Koike, T.; Takaoka, K.; Kamijo, R. J. Biol. Chem. 2006, 281, 23246–23253. 8. Spivak-Kroizman, T.; Lemmon, M. A.; Dikic, I.; Ladbury, J. E.; Pinchasi, D.; Huang, J.; Jaye, M.; Crumley, G.; Schlessinger, J.; Lax, I. Cell 1994, 79, 1015– 1024. 9. Thorton, S.; Mueller, S. N.; Levine, E. M. Science 1983, 222, 623–625. 10. Gospodarowicz, D.; Cheng, J. J. Cell. Physiol. 1986, 128, 475–484. 11. Giraux, J. L.; Matou, S.; Bros, A.; Tapon-Bretaudiere, J.; Letourneur, D.; Fischer, A. M. Eur. J. Cell Biol. 1998, 77, 352–359. 12. Rosenbaum, J.; Tobelem, G.; Molho, P.; Barzu, T.; Caen, J. P. Cell Biol. Int. Rep. 1986, 10, 437–446. 13. Foster, A.; Matthew, H. W. T. Proc. NOCChe. 1998, 25, 127–129. 14. Chupa, J. M.; Foster, A. M.; Sumner, S. R.; Madihally, S. V.; Matthew, H. W. T. Biomaterials 2000, 21, 2315–2322. 15. Massia, S. P.; Hubbell, J. A. Cytotechnology 1992, 10, 189–204. 16. Schoen, F. J.; Castellot, J. J., Jr. J. Vasc. Surg. 1991, 13, 758–761.
613
17. Türkog˘lu-S ßasßmazel, H.; Aday, S.; Gümüsßdereliog˘lu, M. Int. J. Biol. Macromol. 2007, 41, 338–345. 18. Gümüsßdereliog˘lu, M.; Agi, P. React. Funct. Polym. 2004, 61, 211–220. 19. Karakeçili, A.; Satriano, C.; Gümüsßdereliog˘lu, M.; Marletta, G. Acta Biomater. 2008, 4, 989–996. 20. Lin, W. C.; Liu, T. Y.; Yang, M. C. Biomaterials 2004, 25, 1947–1957. 21. Kweon, D. K.; Song, S. B.; Park, Y. Y. Biomaterials 2003, 24, 1595–1601. 22. Zhu, A.; Zhang, M.; Wu, J.; Shen, J. Biomaterials 2002, 23, 4657–4665. 23. Zhu, A. P.; Ming, Z.; Jian, S. Appl. Surf. Sci. 2005, 241, 485–492. 24. Fujimura, K.; Bessho, K.; Okubo, Y.; Kusumoto, K.; Segami, N.; Iizuka, T. Arch. Oral Biol. 2002, 47, 577–584. 25. Suck, K.; Behr, L.; Fischer, M.; Hoffmeister, H.; van Griensven, M.; Stahl, F.; Scheper, T.; Kasper, C. J. Biomed. Mater. Res. Part A 2007, 80, 268–275. 26. Madihally, S. V.; Matthew, H. W. T. Biomaterials 1999, 20, 1133–1142. 27. Zeltinger, J.; Sherwood, J. K.; Graham, D. A.; Mueller, R.; Griffith, L. G. Tissue Eng. 2001, 7, 557–571. 28. Tıg˘lı, R. S.; Karakeçili, A.; Gümüsßdereliog˘lu, M. J. Mater. Sci.: Mater. Med. 2007, 18, 1665–1674. 29. Aday, S.; Gümüsßdereliog˘lu, M. Polym. Compos. 2010, 31, 1418–1426. 30. Gupta, P.; McCarthy, J. B.; Verfaillie, C. M. Blood 1996, 87, 3229–3236. 31. Tillman, J.; Ullm, A.; Madihally, S. V. Biomaterials 2006, 27, 5618–5626. 32. Iozzo, R. V. Annu. Rev. Biochem. 1998, 67, 609–652. 33. Wang, X. H.; Li, D. P.; Wang, W. J.; Feng, Q. L.; Cui, F. Z.; Xu, Y. X.; Song, X. H. Int. J. Biol. Macromol. 2003, 33, 95–100. 34. Dekker, A.; Reitsma, K.; Beugeling, T.; Bantjes, A.; Feijien, J.; Van Aken, W. G. Biomaterials 1991, 12, 130–138. 35. Sargeant, T. D.; Rao, M. S.; Koh, C. Y.; Stupp, S. I. Biomaterials 2008, 29, 1085– 1098. 36. Schonherr, E.; Hausser, H. J. Dev. Immunol. 2000, 7, 89–101. 37. Xiao, G.; Gopalakrishnan, R.; Jiang, D.; Reith, E.; Benson, M. D.; Franceschi, R. T. J. Bone Miner. Res. 2002, 17, 101–110. 38. Shibata, Y.; Abiko, Y.; Goto, K.; Moriya, Y.; Takiguchi, H. Biochem. Int. 1992, 28, 335–344. 39. Sun, J. S.; Chang, W. H.; Chen, L. T.; Huang, Y. C.; Juang, L. W.; Lin, F. H. Biomaterials 2004, 25, 607–616. 40. Franceschi, R. T.; Iyer, B. S. J. Bone Miner. Res. 1992, 7, 235–246. 41. Marconi, W.; Benvenuti, F.; Piozzi, A. Biomaterials 1997, 18, 885–890. 42. Kang, I. K.; Kwon, O. H.; Lee, Y. M.; Sung, Y. K. Biomaterials 1996, 17, 841–847.
Contents lists available at ScienceDirect
Carbohydrate Research journal homepage: www.elsevier.com/locate/carres
Heparin-functionalized chitosan scaffolds for bone tissue engineering Menemsße Gümüsßdereliog˘lu ⇑, Sezin Aday Hacettepe University, Chemical Engineering and Bioengineering Departments, 06800 Beytepe, Ankara, Turkey
a r t i c l e
i n f o
Article history: Received 16 September 2010 Received in revised form 6 December 2010 Accepted 8 December 2010 Available online 14 December 2010 Keywords: Heparin Chitosan Osteogenic activity Electrostatic interactions Covalent immobilization
a b s t r a c t The aim of this study is to investigate the effects of heparin-functionalized chitosan scaffolds on the activity of preosteoblasts. The chitosan scaffolds having the pore size of 100 lm were prepared by a freeze-drying method. Two different methods for immobilization of heparin to chitosan scaffolds were successfully performed. In the first method, functionalization of the scaffolds was achieved by means of electrostatic interactions between negatively charged heparin and positively charged chitosan. The covalent immobilization of heparin to chitosan scaffolds by 1-(3-dimethylaminopropyl)-3-ethylcarbodiimide (EDAC) and N-hydroxysuccinimide (NHS) was used as a second immobilization method. Morphology, proliferation, and differentiation of MC3T3-E1 preosteoblasts on heparin-functionalized chitosan scaffolds were investigated in vitro. The results indicate that covalently bound heparin containing chitosan scaffolds (CHC) stimulate osteoblast proliferation compared to other scaffolds, that is, unmodified chitosan scaffolds (CH), electrostatically bound heparin containing chitosan scaffolds (EHC), and CH+free heparin (CHF). SEM images also proved the stimulative effect of covalently bound heparin on the proliferation of preosteoblasts. Alkaline phosphatase (ALP) and osteocalcin (OCN) levels of cells proliferated on CHC and EHC were also higher than those for CH and CHF. In vitro studies have demonstrated that chitosan scaffolds increase viability and differentiation of MC3T3-E1 cells especially in the presence of immobilized heparin. Ó 2010 Elsevier Ltd. All rights reserved.
1. Introduction Although heparin has been used as an anticoagulant for a great many years, much more attention has recently been given to its unique role in other biological processes. It is known that heparin can interact with a great number of proteins related to adhesion (e.g., fibronectin and vitronectin), proliferation (e.g., fibroblast growth factor, FGF)1–3 or osteogenic differentiation (e.g., bone morphogenic proteins, BMPs).4–7 It induces oligomerization of FGF molecules, which is an important step for receptor activation8 and helps BMPs to remain in the extracellular space for longer periods while preventing ligand inactivation as well.7 It is also suggested that heparin enhances BMP activities by making complexes.6 Because of its special interactions with proteins, heparin can be used as a delivery system for growth factors.1 In the literature, there are conflicting results related to the effect of heparin on cellular proliferation. Some studies have indicated that heparin increases endothelial cell proliferation in the presence of heparin-binding growth factors,9,10 which others claimed that it decreases cell proliferation or it has no effect at all.11,12 In addition, it
⇑ Corresponding author. Tel.: +90 312 2977447; fax: +90 312 2992124. E-mail address: [email protected] (M. Gümüsßdereliog˘lu). Present address: Biomaterials and Stem Cell-Based Therapeutics Group, BiocantCenter of Innovation and Biotechnology, Nucleo 4, Lote 3, 3060-197 Cantanhede, Portugal.
0008-6215/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.carres.2010.12.007
has been shown that glycosaminoglycans (GAGs) inhibit the proliferation of smooth muscle cells.13,14 However, it is known that chitosan membranes can stimulate endothelial regeneration by stimulating cell proliferation.14 Since incomplete endothelization and smooth muscle cell hyperplasia are the major problems for small vascular grafts,15,16 heparin–chitosan complexes can be used in the bone regeneration process that requires neo-vascularization. In addition, it has been indicated that heparin-modified polyethylene glycol (PEG) gels promote human mesenchymal stem cell (hMSC) adhesion, spreading, and proliferation.1 In our previous study, it was shown that heparin-immobilized non-woven polyester fabric (NWPF) discs supported proliferation of L929 fibroblasts in lowserum media.17 Chitosan is a hydrophilic, non-toxic, biodegradable, antibacterial, and biocompatible polymer that can be widely used in the biomedical area to develop tissue substitutes, delivery vehicles, and membranes for hemodialysis.18–20 In the literature, water-soluble chitosan/heparin ointments were studied as wound healing accelerators.21 Jeon et al.5 prepared heparin-conjugated nanospheres to enhance long-term delivery of BMP-2. Zhu et al.22,23 used expanded polytetrafluoroethylene (ePTFE)/chitosan and polylactic acid (PLA)/chitosan surfaces in order to immobilize heparin using ionic interactions between chitosan and heparin. Taking the advantage of the natural binding mechanism of heparin to design bioactive materials is promising due to the fact that this method prevents growth factor damage or degradation.1 Additionally, since different growth factors can bind to heparin, it is possible
M. Gümüsßdereliog˘lu, S. Aday / Carbohydrate Research 346 (2011) 606–613
to obtain dual or multiple growth factor-releasing systems by using heparin.24 In the present study, the aim was to develop heparin-immobilized chitosan scaffolds for bone tissue engineering applications. Our results are important in order to understand the effects of different forms of heparin, that is, electrostatically or covalently immobilized and the free form, on cell proliferation and differentiation. In addition, since no additional growth factor was used in our study, we were able to mimic the natural environment of the cells more precisely by using just the interactions between heparin and different growth factors released during proliferation and differentiation of MC3T3-E1 cells.
2. Results and discussion 2.1. Preparation and characterization of chitosan scaffolds Scaffolds, which are widely used in tissue engineering, are required to improve cell proliferation and differentiation, to prompt cells to obtain 3D tissues and to reinforce the regenerating tissue.25,26 It is known that scaffolds having a pore diameter greater than 100 lm can enhance osteoblast growth.27 In this study, porous chitosan scaffolds that have >85% DD with 2% (w/v; weight of chitosan in volume of acetic acid) composition were prepared by a freeze-drying method. SEM examination of these scaffolds indicated a highly interconnected morphological structure with a pore size of 100 lm. Since osteoblasts are 10–30 lm long, this pore size allows cells to migrate into the chitosan ‘sponges’. The porosity of the scaffolds calculated as 82 ± 1.7%. Further details of these scaffolds can be found in our previous studies.28,29 2.2. Heparin immobilization The scaffolds described above were modified with heparin to take the advantage of the unique properties of this molecule. Heparin is a member of the glycosaminoglycan family of polysaccharides (GAGs) that are responsible for regulating a great number of functions such as localization of mitogenic factors,30 bioactivity of growth factors,10 and cellular proliferation.14,31 Additionally, it is known that extracellular matrices (ECMs) are composed of different GAGs such as chondroitin sulfate, heparan sulfate, and hyaluronate. Different structural characteristics of GAGs (e.g., arrangement and degree of sulfation) or their size and location can effect the interactions of these molecules with other molecules, and, additionally, their activity.31,32 When all these important points are taken into consideration, it can be concluded that it is important to use immobilized heparin to mimic the cellular environment more closely. In our study, to find the best method for heparin immobilization, different buffers, stabilization, and immobilization methods were compared. Modified scaffolds were evaluated using the Toluidine Blue test which depends on the metachromatic interactions between heparin and the dye. In this test, higher concentrations of heparin give lower absorbances because of the extraction of heparin–dye complexes into the organic phase (n-hexane). The best test results were obtained when the scaffolds were stabilized by ethanol and covalently modified with heparin in MES buffer containing EDAC and NHS. These scaffolds did not lose their shapes and swelling characteristics in neutral aqueous solutions in contrast to scaffolds stabilized with NaOH. After these preliminary tests, ethanol-stabilized and heparin-functionalized scaffolds were further evaluated. Two different immobilization methods, that is, electrostatic interaction and covalent immobilization, were chosen to determine the effect of different modification methods on heparin release and cellular activity. Although heparin solutions
607
having concentrations of more than 1% (w/v) were used in the literature,3,33 we chose the initial heparin concentration for immobilization as 0.5% (w/v) according to the results of our preliminary studies. The mechanisms of immobilization methods are given in Figure 1. In the covalent method, heparin was immobilized onto scaffolds through the interaction between the carboxyl groups on heparin activated by NHS/EDAC and the amino groups of chitosan (Fig. 1A). Chitosan’s cationic nature and high charge density make it a potential polymer for biomaterial applications. Due to its charge density, it is possible to obtain insoluble ionic complexes of chitosan with anionic polysaccharides.34 As it can be seen in Figure 1B, the electrostatic immobilization of heparin was achieved using negative groups on heparin and positive groups on chitosan. The amount of heparin immobilized on the scaffolds was determined by the Toluidine Blue test, and the results are given as milligram heparin per gram of chitosan scaffold. These values were 3.8 ± 0.8 mg and 4.7 ± 0.4 mg for electrostatically (EHC) and covalently modified scaffolds (CHC), respectively. Heparin modification was further proved by elemental analysis. It is known that chitosan does not contain sulfur in its structure, while heparin is a highly sulfated polysaccharide. Elemental analysis results indicated that after modification of scaffolds, the percentage of sulfur in the complex was increased about 7–10-fold (1.478% and 2.044% for EHC and CHC, respectively) compared to unmodified CH scaffolds (0.210% from contamination). In addition, the nitrogen percentage of the scaffolds modified covalently with heparin (7.41%) was higher than that for unmodified scaffolds (6.60%). In general, it is known that covalent bonding has several advantages compared to other functionalization techniques (e.g., control of molecular orientation, minimization of nonspecific interactions, and prevention of the dissolution, desorption and degradation of the molecule).35 In our study, it was shown that covalent bonding supported the long-term stability of heparin, and even after 20 days, the amount of heparin released from the scaffolds was not significant enough to be determined using the Toluidine Blue test. However, it was reported that heparin can be easily released from the surface when it is immobilized by physical or ionic bonding.20 Since the charge density of chitosan is pH dependent, it is possible to use chitosan as a delivery vehicle for polyanions by taking advantage of the dissociation of the chitosan–polyanion complexes under physiological conditions.22,23 Similarly, in our study, when electrostatic interactions were used, 45% of immobilized heparin was released in the first three days, and during 20 days 71% of heparin was released (Fig. 2). The integration of heparin into the structure affected the mechanical properties of the scaffolds in a positive way. The compression modulus of scaffolds increased significantly, and the values of 1.993 ± 0.035 N/mm2 and 2.853 ± 0.429 N/mm2 were obtained for the CH and CHC scaffolds, respectively. 2.3. Cell culture studies In tissue engineering approaches three major steps, namely cell attachment, proliferation, and differentiations are crucial for regeneration of tissues. The MTT assay, which is a colorimetric test based on the reduction of a yellow tetrazolium salt into a purple formazan product, is widely used to determine cellular toxicity, proliferation or viability. Figure 3 shows the MTT assay results of the different scaffolds examined. As can be seen from this figure, CHC scaffolds showed relatively higher absorbances during the culture period. There was no significant difference among the MTT values for CH and EHC scaffolds and free heparin up to the 12th day of the culture. The early stimulative effect of heparin-immobilized scaffolds can be explained by initial attachment of the cells. It is known that serum
608
M. Gümüsßdereliog˘lu, S. Aday / Carbohydrate Research 346 (2011) 606–613
Figure 1. Schematic diagram showing the immobilization of heparin to chitosan scaffolds: (A) covalent immobilization, (B) electrostatic immobilization.
Figure 2. Heparin release graph for the scaffolds prepared using electrostatic interactions (in PBS, pH 7.4, 37 °C).
contains some attachment factors such as fibronectin and vitronectin that have important roles in cellular attachment onto surfaces. Since heparin can interact with these proteins,1 it can be expected that heparin-immobilized scaffolds will adsorb higher amounts of these proteins compared to unmodified scaffolds. Although this stimulative effect was maintained during the culture period for CHC scaffolds, EHC scaffolds did not induce the cell proliferation during the culture period. The low cell numbers on the EHC scaffolds can be related to desorption of electrostatically bound heparin from the chitosan scaffolds. Heparin release studies indicated that more than 50% of the immobilized heparin was released from EHC during the first 10 days of culture. However, no significant release profile was obtained for covalently immobilized heparin. Chupa et al.14 claimed that the desorption of GAGs may prevent the adsorption of serum components such as fibronectin and collagen. In addition, depending on the concentrations or ionic environment, the binding interactions with heparin can enhance or inhibit growth factor activity.14 High
Figure 3. MTT results for different chitosan scaffolds (statistically significant differences, n = 3, ⁄: p <0.05, ⁄⁄: p <0.001. Unmodified chitosan scaffold is control group).
M. Gümüsßdereliog˘lu, S. Aday / Carbohydrate Research 346 (2011) 606–613
levels of soluble GAGs present in the culture may bind and sequester growth factors that are either supplied by the medium or secreted endogenously. Although it is claimed that cells on chitosan–GAG membranes (2D structures) exhibit rounded shapes,14,31 SEM characterization of the scaffolds at the 4th and 11th days of culture indicated that cells on chitosan–heparin scaffolds attached well and showed a spread-out morphology (Fig. 4). These results were consistent with the ones obtained from 3D GAG–chitosan matrice studies in the literature.31 Similarly, Zhu et al.22 also showed that the attachment of fibroblastic cells increases significantly on chitosan and chitosan/ heparin surfaces. The cell numbers on the electrostatically modified scaffolds were lower as expected from the MTT results (Fig. 4A and D). The distribution of the cells in the 3D structure of the chitosan scaffolds was examined by CLSM on the 8th day of culture (Fig. 5). Alexa Fluor 488 phalloidin was used to stain F-actin (green), and propidium iodide was used for staining of cell nuclei (red). Figure 5 was obtained by the Zeiss LSM Image Browser Program, and using these images, the location, interaction, and density of the cells were determined. Confocal microscopy photos demonstrated that MC3T3-E1 cells migrated into the scaffolds, and cells were observed even at a depth of 100 lm from the surface. It can be clearly seen that CHC scaffolds induced cellular proliferation, and the
609
highest cell densities were obtained for the cells on these scaffolds (Fig. 5C). ECM has different functions such as being a reservoir for endogenous growth factors, localizing growth factors, and preventing their degradation.4,36 The ECM of osteoblasts is composed of collagenous, non-collagenous proteins, growth factors, and BMPs.37 Shibata et al.38 found that soluble heparin stimulates collagen synthesis in MC3T3-E1 cell cultures. In preosteoblasts, BMP-containing ECM can induce alkaline phosphatase activity, which is an early differentiation marker of osteoblast phenotype.37,39 Because differentiation is the third major step of regeneration in vitro, we evaluated cells also in terms of differentiation markers such as ALP activity and OCN secretion. ALP activity of the MC3T3-E1 cells was evaluated by measuring absorbances of p-nitrophenol at 405 nm ( Fig. 6A). The absorbance values increased with time for all scaffolds except those with free heparin, and the values were significantly different for all the samples compared to the ones for the CH scaffolds. Similar to proliferation studies, the highest values were obtained for heparin-immobilized scaffolds. Although MTT values were low for the cells on the EHC scaffolds, these scaffolds induced the differentiation of MC3T3-E1 preosteoblasts. Similarly, the highest OCN values were obtained when EHC and CHC scaffolds were used (Fig. 6B).
Figure 4. SEM images of cells on different chitosan scaffolds at the 4th and 11th days of culture: (A) EHC, 4th day, 3000; (B) CHC, 4th day, 1530; (C) CHF, 4th day, 1500; (D) EHC, 11th day, 5000; (E) CHC, 11th day, 2500; (F) CHF 11th day 5000. Scale bars indicate 10, 20, 10, 2, 10 and 20 lm, respectively.
610
M. Gümüsßdereliog˘lu, S. Aday / Carbohydrate Research 346 (2011) 606–613
Figure 5. 3D CLSM images of MC3T3-E1 cells obtained by the Zeiss LSM Image Browser Program on the 8th day of culture. Left: unmodified chitosan scaffold (CH); middle: unmodified chitosan scaffold + free heparin (CHF); right: covalently heparin-immobilized chitosan scaffolds (CHC). Alexa Fluor 488 phalloidin was used to stain F-actin (green) and propidium iodide was used for staining of cell nuclei (red). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
Figure 6. ALP activities (A) and osteocalcin values (B) of cells cultured on different chitosan scaffolds (statistically significant differences, n = 3,⁄: p <0.001, ⁄⁄: p <0.01. Unmodified chitosan scaffold is control group).
In order to activate BMP-dependent signal transduction, which is important for osteogenic differentiation, the ligand should bind to the Type II BMP receptor, which then associates with and phosphorylates the Type I receptor.37 It is possible to sequester and localize BMPs produced by MC3T3-E1 preosteoblasts by introducing the heparin functionality to the scaffolds. Heparin binds cellsecreted BMP, increases its local availability, and by this way helps to sustain autocrine signaling effects.4 For our study, there are two
possible theories to explain the effects of different immobilization methods on cellular differentiation. According to the first theory, heparin-functionalized scaffolds sequestered BMP produced by cells and facilitated BMP availability and BMP receptor activation.4,37 Because of this effect of heparin, higher ALP and OCN values were obtained for the cells on heparin-modified scaffolds. It is known that there is an important relation between ECM and BMP activity, and cells should be in contact with a collagen-containing
M. Gümüsßdereliog˘lu, S. Aday / Carbohydrate Research 346 (2011) 606–613
ECM before they differentiate.40 Xiao et al.37 found that in the absence of ECM, very little BMP activity is observed for MC3T3-E1 preosteoblast cells. However, in the presence of ECM, addition of BMP stimulates OCN, and bone sialoprotein gene expression.37 The well-organized ECM secreted by cells on CHC scaffolds (Fig. 4 E) and higher cell numbers ( Fig. 3) can be the reason of higher ALP activities for these scaffolds. However, lower OCN values were obtained for the cells on the CHC scaffolds compared to those on the EHC scaffolds. As a second theory, we may think that the adsorption of matrix proteins (e.g., collagen and fibronectin) that were secreted onto CHC scaffolds during culture caused a feedback inhibition.1 In the literature, better hemocompatibility due to the progressive release of heparin was shown where ionic bonds were used to immobilize heparin onto surfaces.41 In our study, there was a continuous release from EHC surfaces. Another reason for the stimulation of osteogenic differentiation on EHC scaffolds can be the fact that heparin enhances the BMP activities by forming complexes with BMPs in the culture media, and in this way the ligands are continuously made available to their receptors.6 However, long-term delivery of heparin can be also important when osteogenic efficacy of in vivo studies is considered.5 Since CHC scaffolds also stimulate cellular proliferation, it is possible to have a fast and accurate regeneration when these scaffolds are used. 3. Conclusion Heparin, which has long been used as an antithrombogenic agent, has a great impact on designing novel scaffolds for tissue engineering due to its unique characteristics and its involvement in key biological processes. In the present study, porous chitosan scaffolds prepared by a freeze-drying method were further modified with heparin by covalent immobilization or electrostatic interactions. The results showed that these immobilization methods affected cellular activity through different mechanisms of action. Cell-culture studies demonstrated that, while covalently immobilized heparin-containing scaffolds were inducing both cell proliferation and differentiation, electrostatically-immobilized heparin-containing scaffolds supported cellular differentiation. In conclusion, these novel scaffolds can stimulate osteogenic regeneration by mimicking the natural ligand–receptor mechanisms more closely, and thus they should have great advantages when used in the bone tissue engineering field. 4. Experimental 4.1. Materials Chitosan and non-fractionated heparin sodium salt (202 USP units mg 1) were purchased from Sigma–Aldrich (Germany). The degree of deacetylation (DD) was >85% for chitosan, and the molecular weight interval of deacetylated chitin was 190–375 kDa. Acetic acid (HPLC grade) and ethanol (96% v/v) were obtained from Riedel-de Haën (Germany). 1-Ethyl-3-(dimethylaminopropyl)carbodiimide (EDAC), Toluidine Blue and 2-N-(morpholinoethanesulfonic acid) (MES) were obtained from Sigma–Aldrich (Germany). N-Hydroxysuccinimide (NHS) was obtained from Fluka (ABD). Glutaraldehyde, p-nitrophenyl phosphate liquid substrate system (pNPP) and hexamethyldisilazane were purchased from Sigma– Aldrich (USA). For cell culture studies, Minimum Essential Medium Alpha Modification (a-MEM), fetal bovine serum (FBS) and trypsin/ EDTA solution were obtained from Hyclone (USA). 1-(4,5-Dimethylthiazol-2-yl)-3,5-diphenylformazan (MTT), b-glycerol phosphate and ascorbic acid were obtained from Sigma–Aldrich (USA). Triton X-100 from Sigma–Aldrich (USA) and 2-amino 2-methyl1,3-propanol from Sigma–Aldrich (Germany) were used in the alkaline phosphatase (ALP) assay. For the osteocalcin (OCN) assay,
611
formic acid was purchased from E. Merck (Germany) and the OSTEASIA kit was obtained from BioSource (Belgium). 4.2. Preparation of chitosan scaffolds Porous chitosan scaffolds were prepared using a freeze-drying method developed in our group’s previous study.28 Briefly, chitosan solutions with concentrations of 2% (w/v) were prepared by dissolving of chitosan in 0.2 M acetic acid with filtration to eliminate the impurities. The solution was poured into 24-well tissueculture polystyrene dishes (TCPS, Nunc, Denmark) and frozen at 20 °C for 24 h. The frozen samples were then transferred into the freeze-dryer (Christ, Germany) and lyophilized at 80 °C until they were completely dried. Freshly lyophilized scaffolds were soaked in ethanol solutions (96% v/v ethanol for overnight and 70% v/v ethanol for 1 h) for stabilization and then lyophilized at 80 °C. The cylindrical porous scaffolds were then cut into 2-mm thick sections with a diameter of 10 mm. 4.3. Heparin immobilization Heparin-immobilized chitosan scaffolds were obtained by the use of two different methods, that is, covalent immobilization and electrostatic interaction. In the first method, heparin immobilization to scaffolds was achieved using the amino groups of chitosan and the carboxyl groups of heparin. For this purpose, stabilized chitosan scaffolds were immersed in 50 mL of MES buffer containing 0.5% (w/v) heparin (202 USP units/mg), 0.40 g of EDAC and 0.097 g of NHS and incubated overnight at +4 °C. After heparin immobilization, the scaffolds were washed with ultrapure water for 4–5 times and subsequently freeze-dried. On the other hand, electrostatic modification was performed using the ionic interactions between chitosan and heparin. In this approach, stabilized scaffolds were incubated overnight at +4 °C in phosphate buffer saline (PBS, pH 7.4) containing 0.5% (w/v) heparin. The effects of buffers, immobilization, and stabilization methods were evaluated prior to selection of the final immobilization procedures. In these experiments, besides MES buffer, sodium citrate buffer (0.2325 g citric acid. H2O and 0.4075 g sodium citrate in 250 mL pure water) containing 0.40 g EDAC and 0.097 g NHS was used. To determine the effects of the stabilization method, the scaffolds stabilized by NaOH were used as well as the ones stabilized by ethanol. For NaOH stabilization, the freshly lyophilized scaffolds were immersed in 0.4 M NaOH solution (50% ethanol and 50% water) for 24 h and washed with pure water for 4–5 times. Then heparin was immobilized to these scaffolds electrostatically and covalently by using the same methods described above. 4.4. Characterization studies The amount of heparin immobilized to chitosan scaffolds was determined by the Toluidine Blue assay.42 Toluidine Blue (25 mg) was dissolved in 0.01 N hydrochloric acid containing 0.2 wt % NaCl. Heparin solutions with different concentrations (2 mL) were added to the Toluidine Blue solution (3 mL) and vortexed for 30 s. To this solution, 3 mL of n-hexane was added, and the mixture was shaken well to extract the Toluidine Blue–heparin complex. The absorbance of the aqueous phase containing the unextracted Toluidine Blue was measured by a UV spectrophotometer at 631 nm. A series of standard heparin solutions (0–20 lg mL 1) were used to prepare a calibration curve. In order to determine the amount of heparin immobilized onto the scaffolds, the scaffolds were immersed in a mixture of Toluidine Blue solution (3 mL) and water (2 mL) for 30 min. Then n-hexane (3 mL) was added to ensure uniformity of the treatment.
612
M. Gümüsßdereliog˘lu, S. Aday / Carbohydrate Research 346 (2011) 606–613
After shaking well, scaffolds were removed from the solution, and the absorbance was measured at 631 nm. The amount of immobilized heparin was calculated using a calibration curve. In vitro release profile and stability of heparin were investigated at pH 7.4 using PBS solution. Scaffolds were incubated in tubes containing 1 mL of release medium (PBS), and the tubes were placed in a horizontal shaker with a speed of 20 rpm at 37 °C. At specific time intervals, the release medium was replaced, and the amount of heparin remaining on scaffolds was determined by the Toluidine Blue test as described previously. To indicate the heparin integrated to the structure of chitosan scaffolds, elemental analysis of unmodified and heparin-immobilized chitosan scaffolds was performed at Tübitak Atal (Ankara Test and Analysis Laboratory, Turkey). The scaffolds were minced into pieces using a lancet, and elemental analyses (for carbon, hydrogen, nitrogen and sulfur) were performed using a vario MICRO CHNS instrument (Elementar Analysensysteme GmbH, Germany). Compression tests were performed at constant crosshead speed of 5 mm/min with a load of 500 N cell (Lloyd Ins., UK). Test samples of chitosan scaffolds having a 2 mm thickness were fully rehydrated in PBS (pH 7.4) at 37 °C for 1 h before material testing. The initial compressive modulus was calculated from the slope of the linear region in the stress–strain curves for both the untreated and heparin-immobilized scaffolds. 4.5. Cell culture studies The MC3T3-E1 cell line originating from fetal mouse calvaria was obtained from RIKEN Cell Bank (Saitama, Japan). The cells were subcultured in flasks using alpha-MEM supplemented with 10% (v/v) FBS and 1% (v/v) antibiotic solution (penicillin–streptomycin, Sigma–Aldrich, USA). The cells, maintained at 37 °C in a humidified CO2 (5%) atmosphere (Heraus Instruments, Germany), were dissociated with 0.25% trypsin–EDTA, centrifuged and resuspended in the medium prior to cell seeding. The cell-culture studies were carried out in Parafilm-lined 24well TCPS in stationary conditions. For sterilization, the Parafilmlined wells and chitosan scaffolds were soaked into 70% ethanol for 30 min and then placed under UV light for 45 min. Then the scaffolds were equilibrated in sterile Dulbecco’s PBS (pH 7.4, 24 h) and immersed in conditioning medium for 1 h prior to cell seeding. The MC3T3-E1 cells were trypsinized from the flasks, and a cell suspension was obtained by the addition of fresh differentiation medium similar to the culture medium stated above supplemented with 2 mM L-glutamine (Sigma–Aldrich, USA), 10 mM b-glycerol phosphate and 50 lg mL 1 of L-ascorbic acid. A cell suspension (1.5 mL) containing 1 105 cells mL 1 was seeded onto the scaffolds. Cells were cultivated in 5% CO2 at 37 °C in 95% relative atmospheric humidity for 21 days. The differentiation medium (0.5 mL) was replenished every 3 days. Four types of chitosan scaffolds, which were stabilized with ethanol, were used for cell culture experiments: (1) Unmodified chitosan scaffolds (CH). (2) Electrostatically bound heparin containing chitosan scaffolds (EHC). (3) Covalently bound heparin containing chitosan scaffolds (CHC). (4) Unmodified chitosan scaffold + free heparin in culture medium (CHF). To determine the effect of free heparin on cellular behavior, cells were seeded onto untreated chitosan scaffolds, and cell culture studies were performed using the same differentiation medium enriched with 10 USP units mL 1 heparin. 4.5.1. MTT assays The mitochondrial activity of the MC3T3 preosteoblasts on the scaffolds was quantitatively assessed with the 3-[4,5-dimethylthiazol-2-yl]-diphenyltetrazolium bromide (MTT) test at different culture periods up to 16 days. At selected time intervals, the culture medium was aspirated, and scaffolds were washed with
900 lL of prewarmed PBS (pH 7.4). Serum-free culture medium (900 lL) supplemented with 90 lL MTT solution (2.5 mg MTT dissolved in 1 mL PBS) was added to each sample. After incubating at 37 °C for 3 h, the medium was replaced with 400 lL of 0.04 M HCl in 2-propanol to dissolve the formazan crystals. The solution was transferred into a 96-well micro-ELISA plate, and the absorbance was read at 540 nm with a plate reader (ASYS Hitech UVM 340 Plate Reader, Eugendorf, Austria). 4.5.2. Cellular morphology The cytoskeleton organization of the cells within the chitosan scaffolds was assessed by confocal laser scanning microscopy (CLSM, Zeiss LSM 510, Germany) on the 8th day of the culture. For this purpose, the scaffolds were transferred to a new 24-well TCPS and washed with PBS (pH 7.4) three times at room temperature. Samples were fixed with 2.5% (v/v) glutaraldehyde in 0.1 M PBS for 10 min at 4 °C. After fixation, scaffolds were stored at 4 °C in PBS solution until observation by CLSM. Prior to observation, fixed cells were permeated by 0.1% Triton X-100 solution for 5 min. Cell cytoskeletal filamentous actin (F-actin) was visualized by treating the cells with 2.5% (v/v) Alexa Fluor 488 phalloidin (Invitrogen Inc., USA) for 20 min in darkness. Nuclei were stained with 10 lg mL 1 propidium iodide for 5 min. The attachment, spreading, and morphology of the MC3T3 cells cultured on heparin-immobilized and untreated chitosan scaffolds were observed by a scanning electron microscope (SEM) (Zeiss Evo 50, Germany). After the culture medium was removed, the scaffolds were gently washed with PBS, and the cells were fixed with 2.5% (v/v) glutaraldehyde in 0.1 M PBS (pH 7.4) for 30 min at 4 °C. Then the scaffolds were dehydrated in ethanol series (30%, 50%, 70%, 90%, and 100% (v/v)) and substituted with hexamethyldisilazane (HMDS) before being dried. After the scaffolds were completely dry, samples were coated with a gold–palladium layer prior to SEM analysis. 4.5.3. Cell differentiation Alkaline phosphatase (ALP) activities of the cells were evaluated on the 15th and 21st days of culture. At the end of each culture period, scaffolds were rinsed twice with PBS and stored at 80 °C. After freeze-drying, samples were cut down with a scissors and homogenized by brief sonication at 0 °C in the lysis buffer (1% Triton X-100). Cell lysates were centrifuged at 12,000 rpm for 10 min at 4 °C. The supernatant was assayed for ALP activity using p-nitrophenyl phosphate (pNPP) as substrate. About 50 lL of Triton lysate was added to 125 lL of ALP solution (56 mM 2-amino-2methyl-1,3-propanediol (pH 9.8) and 1 mM MgCl2 containing 10 mL pNPP), and the mixture was incubated at 37 °C for 30 min. The reaction was stopped by adding 50 lL of 2.5 M sodium hydroxide, and the ALP activity was determined by the measurement of absorbance of p-nitrophenol (pNP) at 405 nm using a microplate reader (Asys UVM 340, Australia). To determine the osteocalcin (OCN) content of cells cultured on scaffolds at the 15th and 21th days of culture, the scaffolds were washed three times with PBS and then treated with 1 mL of 40% (v/v) formic acid for 10 days at 4 °C using vortex mixing for decalcification. The resulting solution was freeze-dried and stored at 80 °C. Osteocalcin content was determined using the commercially available solid-phase enzyme-amplified sensitivity immunoassay OST-EASIA kit (BioSource, Belgium) according to the supplier’s instructions. 4.6. Statistical analysis All experiments were done in triplicates and results were presented as means ± standard deviation (SD) for n = 3. Statistical comparisons were performed using GraphPad Software Instant.
M. Gümüsßdereliog˘lu, S. Aday / Carbohydrate Research 346 (2011) 606–613
Acknowledgments This study was financially supported by the Turkish Scientific _ and Research Council (TÜBITAK) Project No. 108M346. S.A. was _ _ supported by the TÜBITAK-BIDEB 2210 National Scholarship Program for MSc Students. References 1. Benoit, D. S. W.; Anseth, K. S. Acta Biomater. 2005, 1, 461–470. 2. Nie, T.; Baldwin, A.; Yanaguchi, N.; Kiick, K. L. J. Controlled Release 2007, 122, 287–296. 3. Wissink, M. J. B.; Beernink, R.; Scharenborg, N. M.; Poot, A. A.; Engbers, G. H. M.; Beugeling, T.; van Aken, W. G.; Feijen, J. J. Controlled Release 2000, 67, 141–155. 4. Benoit, D. S. W.; Colins, S. D.; Anseth, K. S. Adv. Funct. Mater. 2007, 17, 2085– 2093. 5. Jeon, O.; Song, S. J.; Yang, H. S.; Bhang, S. H.; Kang, S. W.; Sung, M. A.; Lee, J. H.; Kim, B. S. Biochem. Biophys. Res. Commun. 2008, 369, 774–780. 6. Takada, T.; Katagiri, T.; Ifuku, M.; Morimura, N.; Kobayashi, M.; Hasegawa, K.; Ogamo, A.; Kamijo, R. J. Biol. Chem. 2003, 278, 43229–43235. 7. Zhao, B.; Katagiri, T.; Toyoda, H.; Takada, T.; Yanai, T.; Fukuda, T.; Chung, U.; Koike, T.; Takaoka, K.; Kamijo, R. J. Biol. Chem. 2006, 281, 23246–23253. 8. Spivak-Kroizman, T.; Lemmon, M. A.; Dikic, I.; Ladbury, J. E.; Pinchasi, D.; Huang, J.; Jaye, M.; Crumley, G.; Schlessinger, J.; Lax, I. Cell 1994, 79, 1015– 1024. 9. Thorton, S.; Mueller, S. N.; Levine, E. M. Science 1983, 222, 623–625. 10. Gospodarowicz, D.; Cheng, J. J. Cell. Physiol. 1986, 128, 475–484. 11. Giraux, J. L.; Matou, S.; Bros, A.; Tapon-Bretaudiere, J.; Letourneur, D.; Fischer, A. M. Eur. J. Cell Biol. 1998, 77, 352–359. 12. Rosenbaum, J.; Tobelem, G.; Molho, P.; Barzu, T.; Caen, J. P. Cell Biol. Int. Rep. 1986, 10, 437–446. 13. Foster, A.; Matthew, H. W. T. Proc. NOCChe. 1998, 25, 127–129. 14. Chupa, J. M.; Foster, A. M.; Sumner, S. R.; Madihally, S. V.; Matthew, H. W. T. Biomaterials 2000, 21, 2315–2322. 15. Massia, S. P.; Hubbell, J. A. Cytotechnology 1992, 10, 189–204. 16. Schoen, F. J.; Castellot, J. J., Jr. J. Vasc. Surg. 1991, 13, 758–761.
613
17. Türkog˘lu-S ßasßmazel, H.; Aday, S.; Gümüsßdereliog˘lu, M. Int. J. Biol. Macromol. 2007, 41, 338–345. 18. Gümüsßdereliog˘lu, M.; Agi, P. React. Funct. Polym. 2004, 61, 211–220. 19. Karakeçili, A.; Satriano, C.; Gümüsßdereliog˘lu, M.; Marletta, G. Acta Biomater. 2008, 4, 989–996. 20. Lin, W. C.; Liu, T. Y.; Yang, M. C. Biomaterials 2004, 25, 1947–1957. 21. Kweon, D. K.; Song, S. B.; Park, Y. Y. Biomaterials 2003, 24, 1595–1601. 22. Zhu, A.; Zhang, M.; Wu, J.; Shen, J. Biomaterials 2002, 23, 4657–4665. 23. Zhu, A. P.; Ming, Z.; Jian, S. Appl. Surf. Sci. 2005, 241, 485–492. 24. Fujimura, K.; Bessho, K.; Okubo, Y.; Kusumoto, K.; Segami, N.; Iizuka, T. Arch. Oral Biol. 2002, 47, 577–584. 25. Suck, K.; Behr, L.; Fischer, M.; Hoffmeister, H.; van Griensven, M.; Stahl, F.; Scheper, T.; Kasper, C. J. Biomed. Mater. Res. Part A 2007, 80, 268–275. 26. Madihally, S. V.; Matthew, H. W. T. Biomaterials 1999, 20, 1133–1142. 27. Zeltinger, J.; Sherwood, J. K.; Graham, D. A.; Mueller, R.; Griffith, L. G. Tissue Eng. 2001, 7, 557–571. 28. Tıg˘lı, R. S.; Karakeçili, A.; Gümüsßdereliog˘lu, M. J. Mater. Sci.: Mater. Med. 2007, 18, 1665–1674. 29. Aday, S.; Gümüsßdereliog˘lu, M. Polym. Compos. 2010, 31, 1418–1426. 30. Gupta, P.; McCarthy, J. B.; Verfaillie, C. M. Blood 1996, 87, 3229–3236. 31. Tillman, J.; Ullm, A.; Madihally, S. V. Biomaterials 2006, 27, 5618–5626. 32. Iozzo, R. V. Annu. Rev. Biochem. 1998, 67, 609–652. 33. Wang, X. H.; Li, D. P.; Wang, W. J.; Feng, Q. L.; Cui, F. Z.; Xu, Y. X.; Song, X. H. Int. J. Biol. Macromol. 2003, 33, 95–100. 34. Dekker, A.; Reitsma, K.; Beugeling, T.; Bantjes, A.; Feijien, J.; Van Aken, W. G. Biomaterials 1991, 12, 130–138. 35. Sargeant, T. D.; Rao, M. S.; Koh, C. Y.; Stupp, S. I. Biomaterials 2008, 29, 1085– 1098. 36. Schonherr, E.; Hausser, H. J. Dev. Immunol. 2000, 7, 89–101. 37. Xiao, G.; Gopalakrishnan, R.; Jiang, D.; Reith, E.; Benson, M. D.; Franceschi, R. T. J. Bone Miner. Res. 2002, 17, 101–110. 38. Shibata, Y.; Abiko, Y.; Goto, K.; Moriya, Y.; Takiguchi, H. Biochem. Int. 1992, 28, 335–344. 39. Sun, J. S.; Chang, W. H.; Chen, L. T.; Huang, Y. C.; Juang, L. W.; Lin, F. H. Biomaterials 2004, 25, 607–616. 40. Franceschi, R. T.; Iyer, B. S. J. Bone Miner. Res. 1992, 7, 235–246. 41. Marconi, W.; Benvenuti, F.; Piozzi, A. Biomaterials 1997, 18, 885–890. 42. Kang, I. K.; Kwon, O. H.; Lee, Y. M.; Sung, Y. K. Biomaterials 1996, 17, 841–847.