Biomacromolecules 2009, 10, 1575–1583
1575
Alginate/Hydroxyapatite Biocomposite For Bone Ingrowth: A Trabecular Structure With High And Isotropic Connectivity Gianluca Turco,*,† Eleonora Marsich,† Francesca Bellomo,†,‡ Sabrina Semeraro,† Ivan Donati,† Francesco Brun,§ Micaela Grandolfo,| Agostino Accardo,§ and Sergio Paoletti† Department of Life Sciences, University of Trieste, Via Giorgieri 1, Trieste I-34127, Italy, Department of Electrotechnics, Electronics and Informatics, University of Trieste, Via A. Valerio 10, Trieste I-34127, Italy, and Sector of Neurobiology, International School for Advanced Studies (SISSA), 34014 Trieste, Italy Received February 4, 2009; Revised Manuscript Received March 16, 2009
Alginate/hydroxyapatite composite scaffolds were developed using a novel production design. Hydroxyapatite (HAp) was incorporated into an alginate solution and internal gelling was induced by addition of slowly acid hydrolyzing D-gluconic acid δ-lactone (GDL) for the direct release of calcium ions from HAp. Hydrogels were then freeze-casted to produce a three-dimensional isotropic porous network. Scanning electron microscopy (SEM) observations, confocal laser scanning microscopy (CLSM) and microcomputed tomography (µ-CT) analysis of the scaffolds showed an optimal interconnected porous structure with pore sizes ranging between 100 and 300 µm and over 88% porosity. Proliferation assay and SEM observations demonstrated that human osteosarcoma cell lines were able to proliferate, maintain osteoblast-like phenotype and massively colonize the scaffold structure. Overall, these combined results indicate that the novel alginate based composites efficiently support the adhesion and proliferation of cells showing at the same time adequate structural and physical-chemical properties for being used as scaffolds in bone tissue engineering strategies.
Introduction Tissue engineering is widely recognized as one of the most promising approaches for bone repair and reconstruction. It is mandatory for any optimal scaffold material to act as a temporary three-dimensional support for cell adhesion, growth and mineral matrix deposition. Moreover, ideal scaffolds should be able to integrate into surrounding tissue and mimic the structure and morphology of the natural bone tissue. Strict requirements for scaffolds are biocompatibility, a design closely resembling the natural extracellular structure, an appropriate surface chemistry to promote cellular attachment, differentiation and proliferation, and a sufficient mechanical strength to withstand in vivo stresses and physiological loading. Finally, the degradation of the ideal scaffold should proceed in a controlled way, still keeping a sufficient structural integrity until the newly grown tissue has replaced the scaffold’s supporting functions. Both bioactive ceramics and polymers have been developed for use as bone composite scaffolds.1-4 Polymers have some advantages over ceramic materials. Their biodegradation rates and mechanical properties can be tailored for specific applications. They are particularly amenable for implantation and can be easily manufactured into desired shapes. The major concern associated with polymer scaffolds deals with their low mechanical strength and shape retention failure. In addition, synthetic polymers demonstrate insufficient cell adhesion and their hydrophobic surfaces hinder cell growth.5,6 They also lack * To whom correspondence should be addressed. Tel.: +39 040 5583692. Fax: +39 040 5583691. E-mail:
[email protected]. † Department of Life Sciences, University of Trieste. ‡ Present address: Clinical Department of Biomedicine, University of Trieste, Via Manzoni 16, Trieste I-34138. § Department of Electrotechnics, University of Trieste. | SISSA.
functional groups available for further surface modifications.7 When implanted in vivo, some synthetic polymers release acidic degradation products and invoke a chronic immune reaction.8 Composite scaffolds made of biodegradable natural polymers are very promising constructs: they are endowed with excellent biocompatibility and suitable mechanical properties, and they can be loaded with growth factors involved in bone formation. Natural polymers offer the advantage of being very similar, often identical, to the natural macromolecular environment of cells. This similarity introduces the interesting capability of designing biomaterials with a true molecular biological functionality, rather than a mere morphological similarity. Among the natural polymers, polysaccharides are very versatile, enabling to be decorated with signal molecules (oligosaccharides, peptides) and to interact with inorganic components. Alginate is the name of a family of linear copolymers (produced by brown algae and bacteria), containing (1f4)-linked β-D-mannuronic acid (M) and R-L-guluronic acid (G) arranged in a blockwise pattern along the chain with homopolymeric regions of M (M blocks) and G (G blocks) residues interspersed with regions of alternating structure (MG blocks). Alginate has the ability to form stable gels in the presence of millimolar concentrations of calcium or other divalent cations.9 Cell encapsulation in calcium alginate beads represents a well established method for cell protection from the host immune system,10-12 but the biological inertness of alginate has largely hampered its use in all those applications where cell adhesion is mandatory for survival and proliferation. Moreover, while calcium cross-linked gels make use of a simple chemistry and can be introduced into the body in a minimally invasive surgery, they are generally associated over longer time intervals with poor shape definition and volume instability in vivo.13,14 One possible approach to overcome the biological and mechanical limits of alginate is the use of HAp as inorganic reinforcing and osteoconductive component of alginate HAp
10.1021/bm900154b CCC: $40.75 2009 American Chemical Society Published on Web 04/06/2009
1576
Biomacromolecules, Vol. 10, No. 6, 2009
composite scaffolds.15-17 Hydroxyapatite reinforced polymer biocomposites offer a robust system to engineer synthetic bone substitutes with tailored mechanical, biological, and surgical functions.18-23 Numerous studies have consistently shown that HAp typically exhibits excellent biocompatibility, bioactivity, and if porous, osteoconduction in vivo.15,16,24 Therefore, the basic design rationale for preparing HAp-reinforced polymer composites is to further reinforce a solid biocompatible polymer matrix with a bioactive inorganic filler, mimicking the role of HAp in bone. In recent years, various alginate-based constructs for the use in bone engineering have been proposed.21,25-28 In the present paper we describe a 3D biodegradable porous scaffold prepared from a binary mixture composed of alginate and HAp. A new alginate gelification method was set up exploiting the partial release of calcium ions from hydroxyapatite following slowly acid hydrolysis of GDL; next, a highly porous structure was obtained through a freeze-casting process. The scaffolds have been extensively characterized from a structural, chemical and mechanical point of view. An approach based on micro-CT analysis has been developed to assess the 3D morphological organization of the synthesized scaffolds. This approach made it possible to obtain quantitative data on porosity, interconnectivity, and other structural parameters. The values obtained are in good agreement with the parameters required for optimal osteoconductivity and osteointegration of the implant with the surrounding tissue.29,30 Biological tests have been performed using two different cell lines, showing a good level of viability and cell proliferation.
Materials and Methods Materials. Sodium alginate samples isolated from Laminaria hyperborea stipe were provided by FMC Biopolymer (Norway; MW ) 1.3 × 105, FG ) 0.69; FGG ) 0.56). HAp powder was from Fluka (U.S.A.). Dulbecco’s modified Eagle’s medium (DMEM), McCoy’s 5a Medium Modified, fetal bovine serum (FBS), penicillin, streptomycin, trypsin/ EDTA solutions, phosphate-buffered saline (PBS), glutamine, and GDL were purchased from Sigma (U.S.A.). CellTiter Aqueous One Solution cell proliferation assay kit (MTS assay) was from Promega (U.S.A.). All other chemicals were of analytical grade. Preparation of Alginate and HAp Composites (Alg/HAp Scaffolds). Alg/HAp composite scaffolds were prepared by mixing alginate 2% (w/v) and HAp at different concentrations in water using calcium release method. HAp powder was homogenously dispersed into a stirred solution of alginate in water, followed by the addition of GDL 60 mM to release calcium ions from HAp. Aliquots of this gelling solution were then cured in 24-well tissue culture plates (h )18 mm, Ø ) 16 mm, Costar, Cambridge, MA) for 24 h at room temperature to allow complete gelification. The hydrogels in the tissue-culture plate were then stepwise cooled by immersion in a liquid cryostat. Ethylene glycol in water (3:1) was used as refrigerant fluid. Temperature was decreased stepwise from 20 to -20 °C by 5 °C steps with 30 min intervals for equilibration; the samples were then freeze-dried for 24 h to obtain porous scaffolds. For control experiments, pure alginate gels (HAp-free) were prepared by replacing HAp with CaCO3 (corresponding to 30 mM of Ca2+); pure alginate gels were then processed as HAp composite gels. Characterization of Commercial HAp. TEM images were used to identify the average dimensions of the commercial HAp particles used for the preparation of the composite Alg/HAp scaffolds. Image analysis after gray-levels segmentation has been performed by means of Image Pro Plus 6.2 software on the TEM micrographs. Several images (between 101 and 102) were processed leading to the result of an average dimension of the particles of 150 nm. Raman Spectroscopy and Microscopy. Raman spectra of Alg/HAp hydrogels were recorded with a Renishaw “inVia” Raman system at
Turco et al. 514.5 nm laser excitation coupled to a Leica DMLM microscope using a 20× objective. A thermoelectrically cooled charge coupled device (CCD) camera was used for detection. XRD Analysis. Samples of both alginate and Alg/HAp hydrogels were studied by means of X-ray diffraction technique (XRD) using a diffractometer (STOE D500, Siemens, Munich, Germany) with Cu KR radiation (λ ) 0.1541 nm), monochromatized by a secondary flat graphite crystal. The scanning angle ranges from 10 to 60° of 2θ, the steps were of 0.02 of 2θ, and the counting time was of 2 s/step. The current used was 20 mA and the voltage 40 kV. X-ray Microcomputed Tomography (µ-CT). X-ray microcomputed tomography of samples was obtained by means of a cone-beam system called TOMOLAB (www.elettra.trieste.it/Labs/TOMOLAB).31 The device is equipped with a sealed microfocus X-ray tube, which guaranteed a focal spot size of 5 µm in an energy range from 40 up to 130 kV, and a maximum current of 300 µA. As a detector, a CCD digital camera was used with a 49.9 × 33.2 mm2 field of view and a pixel size of 12.5 × 12.5 µm2. The samples were positioned onto the turn-table of the instrument and acquisitions were performed with the following parameters: distance source-sample (FOD), 100 mm; distance source-detector (FDD), 400 mm; magnification, 4×; binning, 2 × 2; resolution, 6.25 µm; tomographies dimensions (pixels), 1984 × 1024; slices dimensions (pixels), 1984 × 1984; number of tomographies, 1440; number of slices, 864; E ) 40 kV, I ) 200 µA; exposure time, from 2 to 5 s. The slices reconstruction process achieved by means of commercial software (Cobra Exxim) started once the tomographic scan was completed and all the projections were transferred to the workstation. Input projections and output slices are represented by files (one file per projection and one file per slice) using arrays of 16-bit integers. Custom produced MatLab code has been used to get a proper segmentation of the slices using Otsu’s32 method and to obtain numerical values of structural features like porosity, interconnection, pore, and trabecular size by means of parallel plate model.33 Scanning Electron Microscopy (SEM). Scaffolds structure was analyzed using a Leica-Stereoscan 430i Scanning Electron microscope. Freeze-casted samples were sectioned at various planes and directly visualized by electron microscopy after sputter-coating with an ultrathin layer of gold. Scaffolds seeded with cells were rinsed with 10 mM HEPES, pH ) 7.4, containing 10 mM CaCl2, 100 mM NaCl, and 5 mM glucose and fixed with 10% glutaraldehyde in PBS for 1 h at room temperature. Samples were then washed three times with water, dehydrated by stepwise treatment with ethanol, and finally dried with a critical point dryer, sputter-coated with gold, and visualized by electron microscopy. Mechanical Testing: Compression Tests. Scaffold samples were cut in regular cylinders (h ) 10 mm, Ø ) 16 mm). Compression tests have been performed according to ASTM D 3574-95 standard by means of a Lloyd LRX testing equipment. The device was coupled with a 100 N load cell. The compression speed was set on 1 mm/min and no preload was applied. Samples have been compressed until the 50% of strain. The same methodology has been applied to the reswollen scaffolds and to the hydrogels. A total of 10 replicates were averaged for each sample. Swelling and Dissolution Behavior. The swelling behavior of the scaffolds was investigated by exposure to milliQ (mQ) water and to simulated body fluid (SBF). The specimens were cylindrical in shape with an average diameter of 16 mm and a thickness of 4 mm. The swelling behavior was quantified by measuring the changes in sample weight as a function of sample immersion time in water and SBF. Wet weighs were determined after blotting with a filter paper to remove the surface water and the swelling ratio was calculated using the equation:
Esr(%) )
(
)
Ws - Wd × 100 Wd
where Esr is the amount of absorbed water (weight percent) by the polymer matrix, and Wd and Ws are the weights of the samples in the
Alginate/Hydroxyapatite Biocomposite For Bone Ingrowth dry and the swollen state, respectively. The results were taken as the mean values of three measurements. Structural stability and integrity in water was evaluated for 4 weeks at 37 °C in agitation. The samples were immersed in 8 mL of mQ water. Wet weight was measured after 10 min equilibration at 7, 14, 21, and 31 days of immersion after blotting on filter paper, respectively. Soaking water was changed after each measurement. Weight variation was calculated using the formula
weight variation(%) ) [1 - (Wtn/W10min)] × 100 where Wtn and W10min are the wet weights of the samples at the defined time and after 10 min of swelling, respectively. Cell Culture and Seeding. Osteosarcoma MG-63 (ATCC number: CRL-1427) human cell line was cultured in DMEM supplemented with 10% FBS, 1% penicillin-streptomycin/1% L-glutamine at 37 °C, and 5% pCO2. Saos-2 (ATCC number: HTB-85) human osteosarcoma cell line was maintained in McCoy’s Medium with 15% FBS, 1% penicillinstreptomycin/1% L-glutamine at 37 °C, and 5% pCO2. For cell seeding onto scaffolds, porous freeze-casted scaffolds produced under sterile conditions were reswollen in 5 mM CaCl2 for 30 min under agitation and immersed in complete cell culture medium for 24 h in 24-well culture plates to ensure chemical equilibration. Osteosarcoma cells, suspended in 50 µL of medium, were loaded with a micropipet over the whole upper surface of the scaffold. After 4 h, the scaffolds were placed into fresh, sterile 24-well culture plates and 1 mL of complete medium was added. Confocal Laser Scanning Microscopy (CLSM). Fluorescencelabeled scaffolds were obtained by coupling alginate with rhodamine 12334 for visualization in the CLSM. Individual constructs were placed on a coverslip and mounted on the stage of an inverted microscope LEICA TCS SP2 associated with a confocal argon-ion laser scanning microscope. Confocal data have been processed by means of Image Pro Plus 6.2 software extended with the 3D Constructor package. Laser excitation light was provided at a wavelength of 488 nm, and fluorescent emissions were collected at wavelengths between 510 and 580 nm. For image acquisition, an exposure time of 0.8 s was adopted with a binning of 2 × 2 on the charge-coupled device camera, yielding a pixel size of 1.46 µm. Cell Proliferation and Viability on Alg/HAp Composites. The viability and growth rate of MG63 and Saos-2 osteosarcoma human cell lines on Alg/HAp composites were assessed as a function of time using the MTS assay according to the protocol provided by the manufacturer (CellTiter Aqueous One Solution cell proliferation assay kit from Promega). A cell suspension of 40 × 103 was seeded on sterilized scaffolds, incubated at 37 °C in a humidified air atmosphere of containing 5% CO2 and MTS assays were performed in quadruplicate 1, 7, 14, and 21 days from cell seeding. Briefly, after 4 h of incubation with the MTS reagent in a humidified 5% PCO2 atmosphere, the medium was collected from the scaffolds and absorbance was measured on an ELISA plate reader at a wavelength of 490 nm. The background absorbance obtained from an empty scaffold (blank) was subtracted from the sample values. Phosphatase Alkaline Activity (ALP). A 100 × 103 suspension of MG63 and Saos-2 cells was seeded on the scaffolds and maintained in culture in complete medium at 37 °C in a 5% pCO2 atmosphere. At different days, scaffolds were washed at RT for 30 min in a buffer 10 mM HEPES, pH 7.4, containing 10 mM CaCl2, 100 mM NaCl, 5 mM glucose, and finally dissolved in a sodium citrate solution (50 mM sodium citrate, 100 mM NaCl, 10 mM glucose, pH ) 7.4). Cells were collected by centrifugation at 400 × g and lysed in a TritonX-100 solution (0.2% w/w TritonX-100 in 100 mM Tris/HCl buffer, pH ) 9.8). Enzymatic activity was measured in a solution of 6 mM para-nitro-phenyl-phosphate and 1 mM MgCl2 in Tris-HCl, 100 mM, pH ) 9.8, after 60 min of incubation at 37 °C. Absorbance was measured at 410 nm. The results were normalized for the amount of protein content in the cellular extract calculated by means of BCA method according to the manufacturer’s protocol (Sigma). All tests were performed in quadruplicate.
Biomacromolecules, Vol. 10, No. 6, 2009
1577
Results and Discussion Alg/HAp Composite Scaffolds. The main target of this work was to produce an isotropically very homogeneous type of porous scaffold based on a highly biocompatible biopolymer, namely, alginate. The strategy devised to achieve this goal was based on the combination of two physical chemical processes, sequentially applied in different steps. In the first one, a homogeneous calcium-alginate gel is produced in water; in the latter one, the solvent is removed, allowing the formation of pores as a result of solvent/solute phase separation. Alginate is known to produce strong and stable gels with calcium; however, the polymer/ion affinity is so high that special precautions are to be taken to avoid formation of randomly distributed zones of inhomogeneity. A homogeneous gel are prepared only if calcium ions can be uniformly released in a controlled way throughout the polysaccharide solution, instead of being directly added as external cross-linking agents (i.e., CaCl2). Lin et al.21 reported the preparation and characterization of an Alg/HAp scaffold by blend mixing of the polysaccharide and CaCl2 solution with the homogeneity granted by the use of a homogenizer. However, the main limitations of those constructs were a very low structural and morphological stability associated with the inability of the different scaffolds formulations to sustain cell viability and growth. Our gels were prepared by means of a release of calcium ions by the inorganic component due to a slight environment acidification with GDL. Various calcium salts can be used as reservoirs of Ca2+ ions for acid displacement:35 we resorted to use HAp for this purpose owing to its fundamental role of inorganic component of bone and its ability to interact with living tissues. The advantage of this “internal” gelation process resides in the slow hydrolysis of GDL that provides for a delay time between the suspension of the lactone and the gel formation, allowing homogeneous gel formation and casting of different gel shapes. The porous interconnected isotropic structure of scaffolds (Figure 1) was then achieved through a process of thermally induced phase separation and subsequent sublimation of the solvent by freeze-casting. Different freezing conditions and concentrations of HAp have been preliminarily investigated, to obtain a good balance between the structural features and the polymer/mineral ratio. The most common procedure followed in freeze-casting involves unidirectional freezing conditions, which always lead to the formation of anisotropic lamellar structures. Using this method, we obtained long parallel pores aligned in the movement direction on the ice front, in agreement with what reported by Qi et al.36 (see Figure 1 of the Supporting Information). On the contrary, isotropic porous interconnected structures were obtained by a nondirectional freezing by immersion in a cryostat. Suitable pore size (100-300 µm) was achieved by setting the cooling rate: a temperature decrease from 20 to -20 °C by 5 °C steps with 30 min equilibration intervals resulted to be the best in terms of structural features (Figure 1A and B). For lower cooling rates the final microstructure was characterized by larger pores (>500 µm) and, conversely, for faster cooling rate smaller pores were obtained (
