Resveratrol-conjugated poly-ε-caprolactone facilitates in vitro mineralization and in vivo bone regeneration

Acta Biomaterialia 7 (2011) 751–758

Contents lists available at ScienceDirect

Acta Biomaterialia journal homepage: www.elsevier.com/locate/actabiomat

Resveratrol-conjugated poly-e-caprolactone facilitates in vitro mineralization and in vivo bone regeneration Yan Li a,⇑, Staffan Dånmark b, Ulrica Edlund b, Anna Finne-Wistrand b, Xu He a, Maria Norgård d, Eva Blomén c, Kjell Hultenby c, Göran Andersson d, Urban Lindgren a a

Department for Clinical Science, Intervention and Technology (CLINTEC), Division of Orthopedics, Karolinska Institute, Stockholm, Sweden Fibre and Polymer Technology, School of Chemical Science and Engineering, Royal Institute of Technology, Stockholm, Sweden Department of Laboratory Medicine, Division of Clinical Research Centre, Karolinska University Hospital Huddinge, Karolinska Institute, Stockholm, Sweden d Department of Laboratory Medicine, Division of Pathology, Karolinska University Hospital Huddinge, Karolinska Institute, Stockholm, Sweden b c

a r t i c l e

i n f o

Article history: Received 4 April 2010 Received in revised form 30 August 2010 Accepted 8 September 2010 Available online 16 September 2010 Keywords: Resveratrol Biodegradable scaffolds Bone regeneration Vapor phase grafting Tissue engineering

a b s t r a c t Incorporation of osteoinductive factors in a suitable scaffold is considered a promising strategy for generating osteogenic biomaterials. Resveratrol is a polyphenol found in parts of certain plants, including nuts, berries and grapes. It is known to increase DNA synthesis and alkaline phosphatase (ALP) activity in osteoblasts and to prevent femoral bone loss in ovariectomized (OVX) rats. In the present study resveratrol was coupled through a hydrolysable covalent bond with the carboxylic acid groups in porous polye-caprolactone (PCL) surface grafted with acrylic acid (AA). The osteogenic effect of this new scaffold was evaluated in mesenchymal cell culture and in the rat calvarial defect model. We found that the incorporation of resveratrol caused increased ALP activity of rat bone marrow stromal cells and enhanced mineralization of the cell–scaffold composites in vitro. After 8 weeks the calvarial defects implanted with resveratrol-conjugated PCL displayed a higher X-ray density than the defects implanted with control PCL. Bone-like structures, positively immunostained for bone sialoprotein, were shown to be more extensively formed in the resveratrol-conjugated PCL. These results show that incorporation of resveratrol into the AA-functionalized porous PCL scaffold led to a significant increase in osteogenesis. Ó 2010 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved.

1. Introduction A plethora of local growth factors have been identified with osteotrophic properties, such as bone morphogenetic proteins (BMPs), fibroblast growth factors (FGFs) and insulin-like growth factors I and II (IGF I/II) [1]. In bone tissue engineering these factors have been combined with different materials, ranging from inorganic bone graft substitutes (e.g. hydroxyapatite, calcium phosphatebased cements) and natural tissue components (e.g. collagen, hyaluronan) to different types of synthetic polymers, with the purpose of creating osteoinductive or osteoconductive scaffolds [2]. Although promising results have been noted, these factors are relatively unstable during industrial processing, making their application costly and the techniques demanding [3]. Therefore, the search for other osteotrophic substances has become an intriguing route to the development of bone regenerative scaffolds. Resveratrol is a polyphenol found in parts of certain plants, including nuts, berries and grapes [4,5]. It is known to exert a variety of health benefits in mammals, among which the cancer ⇑ Corresponding author. Tel.: +46 8 58583869; fax: +46 8 58582224. E-mail address: [email protected] (Y. Li).

chemopreventive and cardio- and neuro-protective properties have attracted most attention [6]. Its osteotrophic effects were first reported by Mizutani et al., who showed that resveratrol dosedependently increased DNA synthesis and alkaline phosphatase (ALP) activity in osteoblasts [7]. Since these effects could be blocked by tamoxifen, an anti-estrogen, they concluded that resveratrol acted via estrogen receptors (ER) [7]. This notion was substantiated by their subsequent in vivo study, which showed that resveratrol prevented femoral bone loss in ovariectomized (OVX) rats [8]. A recent study by Su et al. [9] further clarified the underlying mechanism of resveratrol-mediated bone protection. They found that resveratrol stimulated BMP-2 production by osteoblasts through Src kinase-dependent ER activation and increased the serum concentration of BMP-2 in OVX rats [9]. Besides the ER/BMP-2 pathway, our previous findings indicated that resveratrol epigenetically modified the gene expression in mesenchymal stem cells (MSCs) through activation of Sirt1, a NAD-dependent histone deacetylase, which resulted in increased osteoblast differentiation in MSC cultures [10]. Despite the extensive biological effects in lower organisms, resveratrol is known to be quickly metabolized in the human body, resulting in an extremely short plasma half-life and very low

1742-7061/$ - see front matter Ó 2010 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.actbio.2010.09.008

752

Y. Li et al. / Acta Biomaterialia 7 (2011) 751–758

bioavailability [11]. Recently, attempts at developing synthetic carrier systems for target-specific delivery of resveratrol have been reported [12–15], however, their effects have not been evaluated using in vivo models. Synthetic biodegradable polymers are often used as protein carriers since their structure, geometry, porosity, mechanical properties, surface properties and degradation kinetics can be tailored according to requirements [16–20]. We have previously invented a vapor phase grafting method for surface functionalization of synthetic polymers in a non-degrading, solvent-free fashion [21]. This method for covalent immobilization of a hydrophilic graft chain on the substrate surface has been shown to not only significantly increase hydrophilicity and biocompatibility of the material [21–23] but also facilitate the attachment of bioactive molecules [24]. In the present study we functionalized the inner and outer surfaces of a porous poly-e-caprolactone (PCL) scaffold by vapor phase grafting with acrylic acid (AA). The graft chain carboxylic acid pendant groups were then covalently bonded with resveratrol through a hydrolysable linkage. The biocompatibility and osteoconductive effect of this new scaffold were evaluated in rat bone marrow stromal cell cultures and rat calvarial defect models. 2. Materials and methods 2.1. Reagents Modified Eagle’s medium alpha (a-MEM), fetal bovine serum (FBS), L-glutamine and gentamycin were purchased from Invitrogen (Life Technologies, Paisley, UK). Resveratrol was purchased from Sigma–Aldrich. Polymer scaffolds were produced from polye-caprolactone (PCL) (Mn = 80000, Aldrich) using chloroform (Labscan) and NaCl (Fischer, ground and dried before use). The polymer structure and molecular weight was verified by 1H NMR and size exclusion chromatography (SEC) analyses. Benzophenone (BPO) (>99% pure, Acros), acrylic acid (AA) (Aldrich, distilled at reduced pressure before use), resveratrol (99% GC, Sigma), 1-ethyl-3-(3dimethylaminopropyl) carbodiimide hydrochloride (EDC) (Aldrich) and ethanol (EtOH) (95%, Kemetyl AB, Sweden) were used for surface functionalization the following chemicals. A 0.1 M MES buffer, pH 6.0, was prepared from 2-[N-morpholino]ethanesulfonic acid hemisodium salt (MES) (Sigma–Aldrich, Germany). 2.2. Preparation of porous PCL scaffolds Porous PCL scaffolds were prepared as previously described [25] in a salt leaching process using NaCl as the porogen. The porogen was ground once in a mortar prior to use. No size discrimination was carried out. After extensive drying circular pieces of 5 mm diameter were cut for in vivo rat calvarial defect implantation and samples with a diameter of 16 mm were cut for cell culture. The pieces had a thickness of 2.5 mm. They were stored under vacuum until use. The material had a broad pore size range and an open porosity of 77%, as determined by the Swedish Ceramic Institute. The polymer structure was verified by 1H NMR, recorded at 500 MHz in a Bruker DMX 500 spectrometer using Bruker software. The samples were dissolved in CHCl3 in 5 mm diameter sample tubes. Non-deuterated CHCl3 (d = 7.26 ppm) was used as an internal standard. 2.3. Covalent grafting of porous scaffolds Our previously developed vapor phase grafting method was used for covalent grafting [21–23] of AA on the inner and outer surfaces of the porous scaffolds. Grafting was performed at a molar ratio of monomer to initiator of 10:1 at 30 ± 0.2 °C for 15 min. The

grafted scaffolds were rinsed extensively in deionized water for 3– 4 h and then rinsed in ethanol (99.5%). Finally, the scaffolds were thoroughly dried under vacuum. The surface structure before and after grafting was assessed by Fourier transform infrared (FTIR) spectrometry by freezing the porous discs in liquid nitrogen and cutting them through the middle into two slices of the same thickness. The outside surface as well as the interior pore surfaces, the inside, of the discs were then analyzed. A Perkin–Elmer Spectrum 2000 FTIR, equipped with an attenuated total reflectance (ATR) crystal accessory (Golden Gate) was used, providing an analysis of the surface down to a depth of approximately 1 lm. All spectra were the means of 32 individual scans in the 4000–600 cm1 interval. 2.4. Coupling of resveratrol The acid functionalized AA grafted porous scaffolds were covalently coupled to resveratrol according to the following protocol. Scaffolds were pre-wetted in water for 2 h. Resveratrol (1% w/v) and EDC (5% w/v) and ethanol (40 vol.%) were then added and allowed to react at room temperature for 24 h under gentle shaking. A hydrolysable ester linkage formed between resveratrol and the acid grafted PCL scaffolds. The scaffolds were rinsed with plenty of fresh distilled water and ethanol and vacuum dried. Scaffolds that underwent the same procedure but without resveratrol were used as negative controls. 2.5. Cell culture Bone marrow stromal cells were obtained from 6-week-old male Sprague–Dawley rats as described previously [10]. Briefly, the rats were killed using 4% isofluorane in CO2. The femurs were aseptically excised and the bone marrow flushed out with 20 ml of a-MEM. After centrifugation at 1000 rpm for 5 min the cell pellet was collected and cultured in a-MEM medium supplemented with 10% fetal bovine serum (FBS), 1 mM L-glutamine and 100 lg ml1 gentamycin. Non-adherent cells were removed after 24 h. After reaching 80% confluence the cells were seeded onto 16 mm scaffolds, which were pre-soaked in culture medium for at least 24 h in 24-well plates at a density of 50,000 cells per well. Osteoblast differentiation was initiated the following day with culture medium supplemented with 50 lg ml1 L-ascorbic acid, 10 mM b-glycerophosphate and 0.1 mM dexamethasone. The medium was changed twice per week. Based on our experience, cells harvested by this procedure maintained their multipotent differentiation capacity [10]. 2.6. Quantification of alkaline phosphatase (ALP) activity and ALP staining A Phosphatase Substrate Kit (Pierce, Rockford, IL) containing p-nitrophenyl phosphate disodium salt (pNPP) was used to quantify the ALP activity in the cell culture medium. The medium was collected on days 3, 7, 14 and 21 and frozen at 80 °C until analysis. pNPP solution was prepared by dissolving two pNPP tablets in 8 ml of distilled water and 2 ml of diethanolamine substrate buffer. About 50 ll of medium was taken from each sample and incubated with 100 ll well1 pNPP solution for 30 min at room temperature in a 96-well plate. About 50 ll of 2 N NaOH was then added to each well to stop the reaction. Unused culture medium treated in the same way was used as a blank control. The absorbance was measured at 405 nm in a kinetic ELISA reader (Spectra MAX 250, Molecular Devices, Sunnyvale, CA). A TRACP and ALP double stain kit (Kakara Bio Inc., Otsu, Japan) was used to stain for ALP in the cell–scaffold composites. The substrate for ALP was dissolved in 10 ml of distilled water before use. After washing twice with

Y. Li et al. / Acta Biomaterialia 7 (2011) 751–758

phosphate-buffered saline (PBS) the cell–scaffold composites were fixed in fixation buffer for 5 min. After washing with distilled water three times, 1 ml of ALP substrate was added to each well. Samples were incubated at 37 °C for 30 min, followed by washing with distilled water. 2.7. Staining for mineralization of the cell–scaffold construct Rat bone marrow stromal cells were seeded on the scaffolds at 50,000 cells per well in 24-well plates and cultured in bone-inducing medium for 4 weeks. Both alizarin red and von Kossa staining were used for analysis of in vitro mineralization. For von Kossa staining the cell–scaffold constructs were washed with PBS and fixed with 4% paraformaldehyde. After rinsing with distilled water each scaffold was immersed in 2 ml of 5% silver nitrate and placed under UV light for 30 min. To remove the unreacted silver the scaffolds were subsequently immersed in 5% sodium thiosulfate for 5 min, followed by rinsing with distilled water twice. For alizarin red staining the cell–scaffold constructs were fixed in ice-cold 70% ethanol for 1 h and then incubated with 40 mM alizarin red at pH 4.2 for 10 min. Non-specific staining was removed by washing twice with PBS. 2.8. Animal experiments Ten male Sprague–Dawley rats (12 weeks, weighing 500 g, labeled randomly 1–10) were used in the experiments in accordance with Karolinska Institute Animal Care and Use Committee guidelines (ethical approval number S 123-06). Critical calvarial bone defects (5 mm in diameter) were created in the animals by the procedure described by Bosch et al. [26]. Briefly, the rats were anesthetized with 2.7 ml kg1 i.p. Hypnorm/midazolam solution (one part HypnormÒ (0.315 mg ml1 fentanyl citrate and 10 mg ml1 fluanisone) (Janssen Pharmaceuticals, Antwerp, Belgium) to two parts distilled water and one part midazolam (5 mg/ml) (DumexAlpharma, Copenhagen, Denmark). Circular calvarial defects were made on both parietal bones using a trephine burr without perforating the dura mater. For rats 1–5 the left calvarial defects were filled with porous PCL scaffolds without resveratrol incorporation (CTRL-PCL) and the right defects were filled with resveratrol-modified PCL scaffolds (RSV-PCL). For rats 6–10 the left calvarial defects were filled with RSV-PCL and the right defects were filled with CTRL-PCL. All animals were killed 8 weeks after surgery and the scaffolds together with surrounding bone were extracted from the skulls. 2.9. Digital radiography and data analysis A Siemens Opti 150/30/50 HC-100 model 4803388 instrument (München, Germany) was used to acquire a plain digital radiograph of the skull specimens. After acquisition the digital data were blind analyzed by an observer using Sectra Bildvisare computer software (Sectra Imtec AB, Stockholm, Sweden). The area and the mean X-ray density of the calvarial defects were recorded. 2.10. Histological study The scaffolds from in vitro cell cultures and the specimens from animal experiments were rinsed with saline and fixed in 10% formalin. The in vivo skull samples were decalcified in 10% formic acid for 2 weeks. To observe matrix formation inside the scaffolds in the cell cultures 9 lm frozen sections were prepared from the center of the discs and stained with toluidine blue. For histological examination of the calvarial implants the samples were embedded in paraffin and 7 lm transverse sections cut at the center of the bone defects. Three histological sections, representing the center

753

of the original surgical defect, were stained with hematoxylin and eosin (H&E) and observed by light microscopy for histological and histomorphometric analyses. Computer-assisted histomorphometry was performed to measure the amount of newly formed bone within the defect. The images of the histological sections were captured with a digital camera (Leica DFC490, Leica Microsystems AB, Stockholm, Sweden) fitted to a light microscope (Leica DMRBE) at an original magnification of 16. The digital images were saved on a computer and histomorphometric analysis was completed using Leica Qwin Plus version 3.5.1. The criteria used in this study to standardize the histomorphometric analysis of the digital images followed the work of Furlaneto et al. [27]. The percentage of newly formed bone was calculated and compared. 2.11. Immunohistochemistry (IHC) Paraffin sections from the animal experiments were treated with a 0.3% H2O2 in methanol solution for 30 min at 37 °C. Then the sections were incubated with 5% bovine serum albumin (BSA) and 2% normal goat serum (NGS) for 20 min to avoid non-specific immunoreactions. Primary rabbit anti-bone sialoprotein (BSP) polyclonal antibody (AB1854, Chemicon, Stockholm, Sweden) diluted in PBS containing 2% NGS and 5% BSA (1:2500) was applied for 24 h at 4 °C. Biotinylated goat anti-rabbit IgG (AP132B, Chemicon) was used as the secondary antibody (1:300 in PBS containing 2% NGS and 5% BSA). Finally, sections were treated with an ABC kit (Vector Laboratories, Stockholm, Sweden) and immuno-positive loci were detected by incubation with 3,30 -diaminobenzidine tetrahydrochloride (DAB) solution (Vector Laboratories). Sections were counterstained with Mayer’s hematoxylin for 30 s. 3. Results 3.1. Linkage of resveratrol to acrylic acid grafted PCL scaffolds PCL is one of the most popular resorbable candidates for biomedical applications, with many favorable properties, being tough yet flexible, with documented non-toxicity. The number of carboxylic groups immobilized on the surfaces during grafting is directly related to the grafting time and can be quantified as the mean surface number of COOH groups by titration [28]. After 15 min grafting of AA on PCL surfaces the COOH surface concentration was 2.4 lmol cm2, constituting the maximum number of conjugation points in subsequent coupling reactions. Spectra confirmed the immobilization of AA chains on the surface. The typical C@O stretching at 1721 cm1 stemming from the ester functionality of each PCL repeating unit was, after grafting, mostly overlaid by a new peak emerging at 1705 cm1 attributed to the C@O functionalitity of the carboxylic acid pendant groups in each repeating unit of AA. A hydroxyl band at 3000–3600 cm1 was also present in the grafted samples, as expected. The porous discs were cut in half to verify that the structure change mediated by grafting was present on both the inner and the outer pore surfaces (Fig. 1). The carboxylic acid groups were then used as linkage groups in the covalent coupling of resveratrol to the grafted surfaces. A water soluble carbodiimide, EDC, was used as the coupling agent. The carboxylic acid groups on the graft chains reacted with the carbodiimide forming a reactive O-acylisourea intermediate that readily reacted with hydroxyl moieties resulting in an ester bond between the graft chains and resveratrol (Fig. 2). Two hydroxyl groups were available on each resveratrol molecule, but coupling was expected to be less likely to occur at the bisubstituted phenyl group of resveratrol due to steric hindrance. The resveratrol was covalently connected to the graft chain pendant carboxyl moieties via a degradable aliphatic ester linkage, providing a means for the resve-

754

Y. Li et al. / Acta Biomaterialia 7 (2011) 751–758

Fig. 1. ATR-FTIR spectrum confirming the grafting of acrylic acid both on the surface (c) and inside (b) the porous scaffold. (a) Non-grafted PCL. The C@O stretching at 1721 cm1 from the PCL repeating unit is overlaid by the peak from the C@O functionality of the carboxylic acid groups at 1705 cm1 after grafting, seen as the shoulder in (b) and (c) on the 1721 cm1 peak. As expected, the surface has a higher graft yield compared with the inside of the porous scaffold. A hydroxyl band at 3000–3600 cm1 is also present in the grafted samples, with the same trend as at 1700 cm1.

ratrol moiety to be released by hydrolytic cleavage during in vivo use of the scaffolds, hence providing a supply of the bioactive substance in the host tissue. 3.2. In vitro study ALP activity could be quantified in the conditional medium from the in vitro cultures by pNPP analysis. Fig. 3A shows a representative result from one of the three independent measurements. For cells seeded on control PCL the medium ALP activity increased dramatically in the early culture period, reaching a peak on day 7 (about 3.35 times the value on day 3). It then decreased, but remained at a constant level in the later culture period (about 2.02 times the value on day 3 on day 14 and 1.98 times on day 21). For cells seeded on RSV-PCL a similar pattern of ALP activity was found, but at all three time points ALP activity was significantly higher compared with the control group (about 161.7% of the control on day 7, 153.0% on day 14 and 130.2% on day 21). Consistent with the quantitative analysis, direct staining of intracellular

Fig. 3. Incorporation of resveratrol in the porous PCL scaffolds significantly increased ALP activity of the co-cultured mesenchymal stem cells. The rat bone marrow stromal cells were cultured on the RSV-PCL and CTRL-PCL scaffolds for up to 3 weeks. (A) ALP activity of the culture medium was measured by PNPP analysis on days 3, 7, 14 and 21. The medium harvested from the RSV-PCL cell cultures had significantly higher ALP activity compared with the medium from the CTRL-PCL cell cultures. (B) This effect could also be seen by ALP staining of the cell–polymer composites with an ALP staining kit (Kakara Bio) on day 7. The experiment was repeated three times. Each data point represents the mean ± SEM of three samples. *P < 0.05 indicates significant difference to time matched CTRL-PCL.

ALP on day 7 showed a much stronger signal on cells seeded on the RSV-PCL scaffolds (Fig. 3B). In order to visualize the tissue structure formed inside the scaffolds, frozen sections of the polymer–cell composites were stained with toluidine blue. As indicated in Fig. 4A, more proteinaceous matrix was seen in the RSV-PCL scaffolds. Rat bone marrow stromal cells are known to form mineralized nodules after long-term culture in bone-inducing medium, which could be shown by von Kossa and alizarin red staining [10]. We found that these methods could also be applied to the PCL–cell composites. Fig. 4B shows that after culture for 4 weeks the RSV-PCL scaffolds demonstrated stronger staining. 3.3. In vivo study All animals remained in good health and did not show any wound complications post surgery. The gross dimensions of the

Fig. 2. Strategy for coupling resveratrol through a hydrolysable linkage to poly(epsilon-caprolactone) surfaces. Reaction product B is less likely due to steric hindrance of the bisubstituted phenyl group.

Y. Li et al. / Acta Biomaterialia 7 (2011) 751–758

755

of implants. However, as shown in Table 1, the defects implanted with RSV-PCL displayed a significantly higher X-ray density than the defects implanted with CTRL-PCL. This difference was further confirmed by histological analysis. H&E staining showed bone-like particles inside the scaffolds. Fig. 6A–D show the representative histological features from one of the experimental animals: the two scaffolds shared many common features, such as fibrous encapsulation, infiltration with elongated cells and collagen fibers (confirmed by type I collagen staining, data not shown), as well as bone extension from the defect edges. In the RSV-PCL scaffold a number of bone-like structures were formed in the region near the dura mater, while similar structures could not be observed in the CTRL-PCL scaffold. With the aid of computer-based histomorphormetric analysis, the percentage of bone forming area was quantified. Fig. 6E shows that RSV-PCL scaffolds had a significantly greater area of bone regeneration than CTRL-PCL scaffolds (25.37% versus 10.57%, P < 0.01, paired t-test). In order to confirm bone formation in the bone-like structures the presence of a specific bone marker, BSP, was assessed by IHC. As shown in Fig. 7, like the surrounding calvarial bone, the bone-like structures were positively stained for BSP while the fibrous tissue inside the scaffolds was negative.

4. Discussion

Fig. 4. Incorporation of resveratrol in the porous PCL scaffolds increased matrix production and mineralization of the cell–scaffold co-cultures. The rat bone marrow stromal cells were cultured on the RSV-PCL and CTRL-PCL scaffolds for 4 weeks. (A) More matrix was seen in the RSV-PCL scaffolds on toluidine blue staining of the frozen sections of the cell–scaffold composites. (B) von Kossa and alizarin red staining showing stronger mineralization of the RSV-PCL scaffolds.

original implants were still visible in the explanted calvaria. All implants were enveloped in thin tissue capsules without gross signs of inflammation or adverse tissue reaction, e.g. necrosis. The rats demonstrated individual variance in bone healing, which was noted by palpating the defect region at autopsy and confirmed by radiographic (Fig. 5) and histological (Fig. 6) analysis. Since each rat was implanted with both types of scaffold a paired t-test was used to analyze the difference in the mean X-ray density and defect sizes between RSV-PCL and CTRL-PCL implants. No significant difference was found regarding the defect size between the two types

Fig. 5. Radiography of the rat calvarial defects 8 weeks after scaffold implantation. Two critical bone defects (5 mm in diameter) were created on both sides of the parietal bone in 10 male Sprague–Dawley rats. For rats 1–5 the left calvarial defects were filled with CTRL-PCL scaffolds and the right defects with RSV-PCL scaffolds. For rats 6–10 the scaffolds were placed in the opposite manner. All animals were killed 8 weeks after surgery and the scaffolds together with the surrounding bone were extracted from the skulls.

Osteoconductive capacity is one of the major requirements for scaffolds used in bone tissue engineering. MSCs from the surrounding tissue must proliferate and differentiate into osteoblasts and participate in bone regeneration. This process is activated and stimulated by a number of systemic and local factors which generate a cascade of intracellular signals in the osteoprogenitors [29]. Resveratrol has been shown to direct MSC differentiation towards the osteoblast lineage [10] and to stimulate the proliferation and activity of pre-osteoblasts [7]. Resveratrol has a simple and stable structure [30] and can be extracted from plants or industrially synthesized [6]. It has been documented that the physiologically most active form, i.e. trans-resveratrol, was stable for at least 28 days in buffers ranging from pH 1 to 7 and no degradation was observed after evaporation with alcohol [31]. These features, together with the osteotrophic effects, suggest that resveratrol is a promising candidate for bone tissue engineering. In this study resveratrol was incorporated into AA-functionalized porous PCL. PCL is considered to be a non-toxic and tissue compatible polymer [16]. It is frequently used as a scaffold in bone tissue engineering because of its resistance to rapid hydrolysis and good mechanical strength [32]. Our in vitro study showed that ALP activity in the conditional medium increased for up to 7 days and then decreased. ALP acts as a transmembrane receptor involved in osteoprogenitor–osteoblast adhesion, migration and differentiation [33]. In vivo ALP is mainly detected in osteoprogenitors and pre-osteoblasts, well before mineralization and prior to the expression of non-collagenous matrix molecules [34,35]. The ALP expression pattern found in our study is therefore consistent with the typical expression curve of this osteoblastic marker [36]. The relatively higher ALP activity in RSV-PCL cultures indicates that under our culture conditions the resveratrol released from PCL was biologically functional. The greater matrix formation inside the RSV-PCL discs, as demonstrated by toluidine blue staining, could be the result of either more osteoblasts derived from progenitor cells or enhanced matrix production. Like most MSC cultures, the cell–scaffold composites showed mineralization after culture in bone-inducing medium for 4 weeks. In vitro mineralization is known to be caused by coprecipitation of calcium (from the culture medium) and inorganic phosphates (generated by ALP catalyzed b-glycerophosphate degradation) onto a collagen matrix [37]. The stronger von Kossa

756

Y. Li et al. / Acta Biomaterialia 7 (2011) 751–758

Fig. 6. H&E staining of the implanted polymers in rat 8. Both CTRL-PCL (A and C) and RSV-PCL (B and D) scaffolds were encapsulated by thin fibrous layers and filled with cells and matrix. In RSV-PCL a number of bone-like structures (D) were seen extending from the dura side of the scaffold (B). However, these structures were not observed in CTRLPCL. In addition, the cell number was much higher in CTRL-PCL, with many of the cells demonstrating a fibroblast phenotype. (A, B) 4 magnification; (C, D) 40 magnification. SS, subcutaneous side of the scaffold; DS, dura side of the scaffold; BLS, bone-like structures. (E) Histomorphometric analyses of bone regeneration in the calvarial bone defects (P < 0.01, paired t-test). Table 1 X-ray analysis of bone regeneration. Rat No.

1 2 3 4 5 6 7 8 9 10 Mean ± S.D. * P value *

X-ray density (units) CTRL

RSV

1033 ± 53 1080 ± 54 1114 ± 64 980 ± 64 1092 ± 68 1093 ± 61 1027 ± 100 1124 ± 50 1002 ± 58 1044 ± 50 1077 ± 66 P = 0.000251

1078 ± 77 1126 ± 97 1168 ± 69 1048 ± 67 1135 ± 59 1174 ± 99 1346 ± 115 1198 ± 87 1161 ± 97 1111 ± 88 1153 ± 82

Paired-samples t-test was used for statistical analysis.

and alizarin red staining of the RSV-PCL discs further supports the increased ALP activity and matrix production. It should be pointed

out that we immersed the cell–scaffold composites in protein lysis buffer in order to measure the expression of several intracellular or membrane-bound osteoblast markers, such as Runx2, osterix, osteocalcin and osteopontin. However, there was too little protein to perform a successful Western blot. Despite this limitation, our in vitro results provide a strong indication that incorporation of resveratrol in porous PCL by vapor phase grafting can enhance the bone-forming activity of osteoprogenitors. To study the in vivo osteoinductive effects the polymers were implanted in rat calvarial defects for 8 weeks and bone regeneration was evaluated by X-ray and histological methods. We found that although the defects implanted with RSV-PCL scaffolds had significantly higher X-ray density, their size was not significantly reduced compared with CTRL-PCL implantation. We also noticed that the main X-ray density difference between the two groups was observed in the center rather than the edge of the scaffold. This finding differs from several previous reports in which bone regeneration originated from the edge of the calvarial defects [38,39]. The surrounding structures that possibly contribute to

Y. Li et al. / Acta Biomaterialia 7 (2011) 751–758

757

structures. Taken together, these results give a reasonable indication that incorporation of resveratrol enhances the bone regenerating capacity of porous PCL scaffolds. However, it has to be borne in mind that neither the increased X-ray density nor the enhanced bone island formation directly indicate better mechanical function, which could only be shown by biomechanical assay of the implants and might be a task for future study. Since its discovery resveratrol has been shown to have a number of physiological properties that could be useful in human medicine [11]. However, its rapid metabolism leads to an extremely low plasma concentration of around 0–26 nM [44], which is several thousand fold lower than the concentrations demonstrating in vitro effects [10]. Therefore, combining resveratrol with biodegradable polymers by vapor phase grafting may provide a new route by which to deliver this agent to a local target and strengthen its potency. At present we have optimized neither the reagent concentrations nor the amount of RSV needed to produce the optimal effect. This could be one reason why bone regeneration in RSV-PCL, although significant compared with CTRL-PCL, was less than has been shown for example on scaffolds containing BMPs [45]. Therefore, further studies, including a kinetic quantification of RSV released from the scaffolds, enhancing the efficiency of resveratrol coupling and/or the addition of other osteotrophic substances, are needed in order to optimize the bone regenerating effects. Furthermore, based on the broad range of health benefits of resveratrol, we hope the advantage of this functionalization technique can be demonstrated for other disease models, such as cancers, diabetes and cardiovascular disease. 5. Conclusion We have shown for the first time that incorporation of resveratrol in vapor phase surface grafted AA-functionalized porous PCL significantly increased the osteoinductive ability of the scaffolds both in vitro and in vivo. Acknowledgements The study was supported by the program ‘‘Biomedical Functional materials” (Dnr: A3 02:139) funded by the Swedish Foundation for Strategic Research and the Ulla and Gustaf of Ugglas Foundation. We thank Peter Plikk for helpful discussions about polymer scaffolds. Appendix A. Figures with essential colour discrimination

Fig. 7. In order to confirm the new bone formation in RSV-PCL, a specific bone marker, BSP was stained by IHC in paraffin sections. Like the surrounding calvarial bone (A), the bone-like structures in the RSV-PCL scaffold were positively stained by BSP (B), while the fibrous structures inside CTRL-PCL were negative (C). BLS, bonelike structures.

the regeneration of calvarial defects include the periosteum, dura and adjacent bone [40]. Due to the much larger contact surface, the progenitor cells for skull morphogenesis and repair are considered to be mainly located in the dural tissue [41,42]. This notion was supported by our histological findings, which showed that the bone-like structures extended from the dura side. Since decalcified samples are unsuitable for the study of mineralization, we evaluated the expression of BSP, a marker of terminally differentiated osteoblasts. BSP is selectively expressed in mineralizing tissues and, especially, at sites of de novo bone formation [43]. As expected, BSP could only be detected in the bone-like islands inside all the implanted scaffolds and not in the surrounding fibrous

Certain figures in this article, particularly Figures 3, 4, 6 and 7 are difficult to interpret in black and white. The full colour images can be found in the on-line version, at doi:10.1016/j.actbio.2010.09.008. References [1] Salgado AJ, Coutinho OP, Reis RL. Bone tissue engineering: state of the art and future trends. Macromol Biosci 2004;4:743–65. [2] Rose FR, Hou Q, Oreffo RO. Delivery systems for bone growth factors – the new players in skeletal regeneration. J Pharm Pharmacol 2004;56:415–27. [3] Wozney JM, Rosen V. Bone morphogenetic protein and bone morphogenetic protein gene family in bone formation and repair. Clin Orthop Relat Res 1998:26–37. [4] Athar M, Back JH, Tang X, Kim KH, Kopelovich L, Bickers DR, et al. Resveratrol: a review of preclinical studies for human cancer prevention. Toxicol Appl Pharmacol 2007;224:274–83. [5] Bavaresco L. Role of viticultural factors on stilbene concentrations of grapes and wine. Drugs Exp Clin Res 2003;29:181–7. [6] Pirola L, Frojdo S. Resveratrol: one molecule, many targets. IUBMB Life 2008;60:323–32. [7] Mizutani K, Ikeda K, Kawai Y, Yamori Y. Resveratrol stimulates the proliferation and differentiation of osteoblastic MC3T3–E1 cells. Biochem Biophys Res Commun 1998;253:859–63.

758

Y. Li et al. / Acta Biomaterialia 7 (2011) 751–758

[8] Mizutani K, Ikeda K, Kawai Y, Yamori Y. Resveratrol attenuates ovariectomyinduced hypertension and bone loss in stroke-prone spontaneously hypertensive rats. J Nutr Sci Vitaminol (Tokyo) 2000;46:78–83. [9] Su JL, Yang CY, Zhao M, Kuo ML, Yen ML. Forkhead proteins are critical for bone morphogenetic protein-2 regulation and anti-tumor activity of resveratrol. J Biol Chem 2007;282:19385–98. [10] Backesjo CM, Li Y, Lindgren U, Haldosen LA. Activation of Sirt1 decreases adipocyte formation during osteoblast differentiation of mesenchymal stem cells. J Bone Miner Res 2006;21:993–1002. [11] Baur JA, Sinclair DA. Therapeutic potential of resveratrol: the in vivo evidence. Nat Rev Drug Discov 2006;5:493–506. [12] Huang ZM, He CL, Yang A, Zhang Y, Han XJ, Yin J, et al. Encapsulating drugs in biodegradable ultrafine fibers through co-axial electrospinning. J Biomed Mater Res A 2006;77:169–79. [13] Das S, Ng KY, Ho PC. Formulation and optimization of zinc-pectinate beads for the controlled delivery of resveratrol. AAPS Pharm Sci Tech 2010;11:729–42. [14] Das S, Ng KY. Colon-specific delivery of resveratrol: optimization of multiparticulate calcium-pectinate carrier. Int J Pharm 2010;385:20–8. [15] Das S, Ng KY. Resveratrol-loaded calcium-pectinate beads: effects of formulation parameters on drug release and bead characteristics. J Pharm Sci 2010;99:840–60. [16] Gunatillake PA, Adhikari R. Biodegradable synthetic polymers for tissue engineering. Eur Cell Mater 2003;5:1–16. [17] Andronova N, Albertsson AC. Resilient bioresorbable copolymers based on trimethylene carbonate, L-lactide, and 1,5-dioxepan-2-one. Biomacromolecules 2006;7:1489–95. [18] Hakkarainen M, Hoglund A, Odelius K, Albertsson AC. Tuning the release rate of acidic degradation products through macromolecular design of caprolactone-based copolymers. J Am Chem Soc 2007;129:6308–12. [19] Plikk P, Målberg S, Albertsson AC. Design of resorbable porous tubular copolyester scaffolds for use in nerve regeneration. Biomacromolecules 2009;10:1259–64. [20] Ryner M, Albertsson AC. Resorbable and highly elastic block copolymers from 1,5-dioxepan-2-one and L-lactide with controlled tensile properties and hydrophilicity. Biomacromolecules 2002;3:601–8. [21] Edlund U, Kallrot M, Albertsson AC. Single-step covalent functionalization of polylactide surfaces. J Am Chem Soc 2005;127:8865–71. [22] Kallrot M, Edlund U, Albertsson AC. Surface functionalization of degradable polymers by covalent grafting. Biomaterials 2006;27:1788–96. [23] Kallrot M, Edlund U, Albertsson AC. Surface functionalization of porous resorbable scaffolds by covalent grafting. Macromol Biosci 2008;8: 645–54. [24] Edlund U, Danmark S, Albertsson AC. A strategy for the covalent functionalization of resorbable polymers with heparin and osteoinductive growth factor. Biomacromolecules 2008;9:901–5. [25] Odelius K, Plikk P, Albertsson AC. Elastomeric hydrolyzable porous scaffolds: copolymers of aliphatic polyesters and a polyether-ester. Biomacromolecules 2005;6:2718–25. [26] Bosch C, Melsen B, Vargervik K. Importance of the critical-size bone defect in testing bone-regenerating materials. J Craniofac Surg 1998;9:310–6.

[27] Furlaneto FA, Nagata MJ, Fucini SE, Deliberador TM, Okamoto T, Messora MR. Bone healing in critical-size defects treated with bioactive glass/calcium sulfate: a histologic and histometric study in rat calvaria. Clin Oral Implants Res 2007;18:311–8. [28] Edlund US, Sauter T, Albertsson A-C. Covalent VEGF protein immobilization on resorbable polymeric surfaces. Polym Adv Technol 2010. doi:10.1002/pat.1772. [29] Albrektsson T, Johansson C. Osteoinduction, osteoconduction and osseointegration. Eur Spine J 2001;10(Suppl. 2):S96–S101. [30] Prokop J, Abrman P, Seligson AL, Sovak M. Resveratrol and its glycon piceid are stable polyphenols. J Med Food 2006;9:11–4. [31] Trela CB, Waterhouse AL. Resveratrol: isomeric molar absorptivities and stability. J Agric Food Chem 1996;44:1253. [32] Petrie Aronin CE, Cooper Jr JA, Sefcik LS, Tholpady SS, Ogle RC, Botchwey EA. Osteogenic differentiation of dura mater stem cells cultured in vitro on threedimensional porous scaffolds of poly(epsilon-caprolactone) fabricated via coextrusion and gas foaming. Acta Biomater 2008;4:1187–97. [33] Hui M, Hu M, Tenenbaum HC. Changes in cell adhesion and cell proliferation are associated with expression of tissue non-specific alkaline phosphatase. Cell Tissue Res 1993;274:429–37. [34] Aubin JE. Bone stem cells. J Cell Biochem 1998;30/31(Suppl.):73–82. [35] Long MW. Osteogenesis and bone-marrow-derived cells. Blood Cells Mol Dis 2001;27:677–90. [36] Lian JB, Stein GS. Development of the osteoblast phenotype: molecular mechanisms mediating osteoblast growth and differentiation. Iowa Orthop J 1995;15:118–40. [37] Chung CH, Golub EE, Forbes E, Tokuoka T, Shapiro IM. Mechanism of action of beta-glycerophosphate on bone cell mineralization. Calcif Tissue Int 1992;51:305–11. [38] Park JK, Yeom J, Oh EJ, Reddy M, Kim JY, Cho DW, et al. Guided bone regeneration by poly(lactic-co-glycolic acid) grafted hyaluronic acid bi-layer films for periodontal barrier applications. Acta Biomater 2009;5:3394–403. [39] Schneider OD, Weber F, Brunner TJ, Loher S, Ehrbar M, Schmidlin PR, et al. In vivo and in vitro evaluation of flexible, cottonwool-like nanocomposites as bone substitute material for complex defects. Acta Biomater 2009;5:1775–84. [40] Ozerdem OR, Anlatici R, Bahar T, Kayaselcuk F, Barutcu O, Tuncer I, et al. Roles of periosteum, dura, and adjacent bone on healing of cranial osteonecrosis. J Craniofac Surg 2003;14:371–9 [discussion 80–2]. [41] Hobar PC, Masson JA, Wilson R, Zerwekh J. The importance of the dura in craniofacial surgery. Plast Reconstr Surg 1996;98:217–25. [42] Ogle RC, Tholpady SS, McGlynn KA, Ogle RA. Regulation of cranial suture morphogenesis. Cells Tissues Organs 2004;176:54–66. [43] Chen J, McKee MD, Nanci A, Sodek J. Bone sialoprotein mRNA expression and ultrastructural localization in fetal porcine calvarial bone: comparisons with osteopontin. Histochem J 1994;26:67–78. [44] Vitaglione P, Sforza S, Galaverna G, Ghidini C, Caporaso N, Vescovi PP, et al. Bioavailability of trans-resveratrol from red wine in humans. Mol Nutr Food Res 2005;49:495–504. [45] Bessa PC, Casal M, Reis RL. Bone morphogenetic proteins in tissue engineering: the road from laboratory to clinic, part II (BMP delivery). J Tissue Eng Regen Med 2008;2:81–96.