(1) Industrie Biomediche Insubri S.A, Mezzovico-Vira, Switzerland
(2) Department of Innovative Technologies, University for Applied Science and Art of Southern Switzerland, Switzerland
(3) Department of Chemistry, Materials and Chemical Engineering “Giulio Natta”, Politecnico di Milano, Milan, Italy
(4) Swiss Institute for Regenerative Medicine, Taverne, Switzerland
* Corresponding author Email: email@example.com
This study discusses composite polymer-coated mineral grafts for bone regeneration.
Materials and Methods
Bone xenografts are coated with degradable synthetic [poly(L-lactide-co-ε-caprolactone)] and natural (polysaccharides) polymers in order to increase their mechanical properties, on one side, and to improve cell adhesion, on the other, with the purpose of developing a novel composite material for bone tissue engineering.
The fabricated bone graft exhibited regular microstructure similar to healthy iliac bones with an average of 27% open porosity and an adequately rigid structure, which ensures a better osteointegration once implanted.
This approach avoids the use of trial-and-error methods and consents a better
A critical-sized defect is a large disruption in a bony tissue, either a fracture or a hole that cannot spontaneously heal, and the actual size varies depending on the species and anatomical site[1,2,3]. In this framework, particularly, the restoration of mandibular defects caused by ablative surgery for oral and maxillofacial tumour, trauma, infection and congenital deformity remains a challenge for surgeons[4,5]. Hence, the need of adequate bone substitutes for remodelling of native bone tissue is evident and requires a wide spectrum of proposed solutions, belonging to academia, clinics and industry[6,7,8]. Nowadays, autologous grafting and distraction osteogenesis are most common techniques used to restore mandibular defects[9,10]. Generally, autologous grafts have been considered the ‘gold standard’ because of the advantages of osteogenesis, osteoinduction and osteoconduction. However, major drawbacks of these methods are severe morbidity, as well as donor-site availability, which call for new methods and solutions for bone reconstruction[12,13,14,15]. Naturally derived materials, most commonly of animal origin (xenografts), provide structures similar to living tissues that might induce specific cellular responses and sometimes also supersede the advantages of synthetic polymers, which indeed represent another viable alternative to autografts[16,17,18]. Xenografts may also reduce the stimulation of chronic inflammation or immunological reactions and toxicity, which, on the other hand, are often detected with synthetic polymers[16,19,20]. Nowadays, materials science, in conjunction with bio- and nanotechnologies, provides interesting solutions for the design of better performing bone[7,21,22,23,24]. In particular, biodegradable scaffolds are widely used as key artificial medical devices in tissue engineering, which aim to provide a desirable microenvironment that allows neo-tissue to be generated to repair and replace damaged or missing bone districts[25,26,27]. Indeed, innovative synthetic polymers can be tuned in terms of composition, rate of degradation, mechanical and chemical properties[16,19]. As a fundamental premise in tissue engineering, polymeric scaffold should provide (a) host tissue-like mechanical support for promoting neo-tissue growth and functioning; (b) adequate porosity and permeability for nutrient delivery and metabolite removal and (c) a controllable degradation rate of the matrix. But, even further progresses in tissue engineering, combining natural and synthetic components, could yield more favourable outcomes than current range of approaches used to repair, e.g. maxillo-cranial, facial, palate and calvarial defects[2,9,31]. Hence, the goal of the proposed approach was to combine the biocompatibility and tissue integration capability of natural materials with the possibility to tune mechanical and physical properties, typical of synthetic ones. Our bone graft, known as SmartBone[®] (SB), is hence a composite material that was obtained using bovine bone-derived matrix as starting material, reinforced with poly(L-lactide-co-ε-caprolactone) (PLCL) and polysaccharides. The bovine-derived matrix is a mineral matrix made of calcium hydroxyapatite [HA, Ca5(PO4)3(OH)] and very minor collagen residuals that present a chemical microstructure and a micromorphology that resembles the human bone[33,34,35]. However, as said above, its physical properties evidence a rigid, but not elastic, structure and thus too fragile for any reconstructive surgery applications. In addition, the due cleaning and sterilisation processes destroy its biochemical structure, increase structural porosity and hence allow neither easy graft handling nor cell adhesion. It is in this framework that our approach finds its core aims to reinforce a bovine-derived bone structure with the addition of an elastic component in terms of polymer coating, thus loosing fragility and reducing porosity, and to resemble healthy human bone. Finally, the addition of polysaccharides, even in extremely low quantities, increases the hydrophilicity of the scaffold with consequent higher blood affinity and favouring cell attachment, thus enhancing biocompatibility and osteointegration[6,36]. The grafts were characterised in terms of chemical composition, microstructure, physical properties and polymer-coating degradation. The resulting composite material has been proved to be able to mimic human bone microstructure and to ensure macro-scale properties, which include an adequate-sized open porosity with combined rigid–elastic behaviour, together with surface properties ensuring cell viability and fast tissue integration.
In addition, the proper design of coating was performed using a mathematical model that can help material design avoid classic ‘trial-and-error’ approaches[37,38]. The aim of this study was to perform a model study and characterise the materials used in composite polymer-coated mineral grafts for bone regeneration.
The bovine-derived cancellous bone was sourced as described in a study conducted by Pertici et al., being certified for human uses. The choice of polymers was based on their potential applicability within the pharmaceutical and biomedical industries and the necessity to obtain scaffolds with suitable mechanical properties. A commercially available copolymer of poly(L-lactic acid) and poly(ε-caprolactone), already used in medical applications, was chosen (Purac Biomaterials, Gorinchem, The Netherlands). Polysaccharides (Merck, Darmstadt, Germany) were also chosen in order to improve the hydrophilicity of the matrix. All materials were used as received. SB grafts were then obtained by reinforcing bovine bone-derived matrix with the mixture of PLCL and polysaccharides through a proprietary process[39,40]. Ethylene oxide sterilisation (at BioSter SpA, Bergamo, Italy) was applied after final double layer packing.
Samples were laminated with potassium bromide and then recorded using Thermo Nexus 6700 spectrometer coupled to Thermo Nicolet Continuum microscope equipped with a 15× Reflachromat Cassegrain objective.
Environmental scanning electron microscopy (ESEM) analysis and energy-dispersive analysis (EDS) were performed at 10 kV with Evo 50 EP Instrumentation (Zeiss, Jena, Germany). Each sample was first analysed on the side surfaces and then halved with a sharp scalpel and the two inner exposed surfaces were finally analysed.
Five cubic samples of SB were cut to about 8 mm in side length, for the sake of instrumental cell dimension, mounted and scanned with micro-computed tomography (micro-CT) system (VTomEx-s, Phoenix|x-ray, a GE Healthcare Company, Little Chalfont, Buckinghamshire, UK. The images consisted of 936 slices with a voxel size of 21 μm in all three axes. The scaffolds were imaged and reconstructed in three dimensions. Regions of interest (ROI) adjacent to implants of the same size were reconstructed and analysed using micro-CT with the same thresholds. The database was analysed and three-dimensional pictures of each ROI were obtained. From acquired data, graft free volume (ε) and surface/volume ratio were estimated.
For the sake of simplicity, the system is represented as a cubic lattice containing spherical pores. The number and the size of the pores can be consistently determined starting from the surface area and the void ratio determined with micro-CT analysis:
System dynamics is described through mass conservation equations, applying the method of the moments. This approach, already proposed by Perale et al., was chosen for its validated results[41,42,43,44]. Polymer degradation is then represented with following system of partial differential equations:
Equation details have been widely explained elsewhere[38,41]. Diffusion coefficients are effective values, as they are related to the degradation process through a functional dependence on average numeral molecular weight[41,44]:
Derivative terms are expressed in spherical coordinates, taking into account only radial terms, in which radius is the characteristic degradation length. Equation (5f) describes the coating thickness reduction because of mass transfer between the polymer phase and the surrounding environment. Such formalism is allowed because, for such system, the mass transfer (
Average weight, number, molecular weight as well as the polydispersity are estimated from producer data, while monomer molecular weight and polymer density are weighted average values with respect to polymer composition. Polymerisation kinetic constant was estimated as explained in Lemmouchi et al. who studied degradation behaviour of various aliphatic polyesters formulations. In particular, model was adapted in order to reproduce experimental data of molecular weight, which indicates decrease in cylindrical devices of PLCL 80:20. Monomer and water diffusivity represent a reasonable estimation for aliphatic polyester formulations (Table 1).
Simulation input data
Macroscopic view of a 10×10×10 mm SB graft is presented in Figure 1(a), which is a macroscopic view that confirms its porous structure with small and interconnected pores. Its chemical structure was studied using Fourier transform infrared (FT-IR) analysis, as presented in Figure 1(b). The peak corresponding to 3450 cm[-1] belongs to −OH groups, and 2960 cm[-1] represents stretching vibration of CH3. The presence of PLCL polymer coating is evidenced not only by peak at 2960 cm[-1] (CH3), but also by peaks at 1460 and 1100 cm[-1], carbonyl (1640 cm[-1]) and C–O–C groups, respectively. The presence of polysaccharide is visible from C–O–C groups, overlapped with polymer signal. The mineral nature of the matrix is underlined by the presence of phosphate groups typical of HA and revealed by 1050 and 600 cm[-1] peaks.
(a) Macroscopic view of SB graft (scale bar = 2.5 mm) and (b) FT-IR spectra of a cross-section of the SB graft. FT-IR, Fourier transform infrared; SB, SmartBone[®].
SB grafts were scanned to study its porosity at high resolution using a micro-CT scanner. 2D and 3D reconstruction confirms interconnected open porous structure through the whole thickness of the scaffold (Figure 2(a) and (b)). Average tested scaffolds resulted in an equivalent volume of about 515 mm3, a free volume of 140 mm3 and a surface of 2,300 mm2. Moreover, the average porosity of samples (
(a) 3D reconstruction of porous structure by micro-CT scanning and (b) representative image of one slice of 2D micro-CT scanning (scale bar = 2.5 mm). 2D, two dimensional; 3D, three dimensional; micro-CT, micro-computed tomography.
Native bovine-derived bone matrix was analysed from a microstructural point of view using ESEM and its structure is shown in Figure 3(a).
ESEM images at the same magnitude (bar scales = 100 μm) of bovine bone matrix (a) and polymer-coated SB graft (b). (c, d) ESEM image of SB graft (scale bar = 30 μm) and corresponding EDS spectra. Spot 2 is characterised only by strong signals of C and O, typical of the presence of polymer film; and spot 1 is characterised by the presence of Ca and P, confirming the mineral nature of the matrix. EDS, energy-dispersive analysis; ESEM, environmental scanning electron microscopy; SB, SmartBone[®].
Coating degradation has been modelled through the system of partial differential equations constituted by eqs. 5a-f. The numerical integration, performed by means of ode15s algorithm as implemented in MATLAB, allowed obtaining the time evolution of coating thickness (Figure 4d) for four different initial thickness values. The robustness and the consistency of the adopted approach have been evaluated through a characteristic times analysis.
(a) ESEM image of polymer coated on to mineral matrix (yellow arrow, scale bar = 20 μm) and (b) polymeric PLCL coating degradation from SB grafts with different coating thicknesses (blue = 2 μm, red = 3 μm, green = 5 μm, black = 10 μm). ESEM, environmental scanning electron microscopy; PLCL, poly(L-lactide-co-ε-caprolactone); SB, SmartBone[®].
As mentioned, a microstructural analysis of native bovine-derived bone matrix was carried out using ESEM, as shown in Figure 3(a). Pores are well distributed along the sample, but they are extremely larger with respect to the ones present in healthy human iliac crest bones. As is well-known, porosity and pore size of biomaterials play a key role in osteointegration and new bone formation; indeed, relatively larger pores favour direct osteogenesis, as they allow vascularisation and high oxygenation, while smaller ones result in osteochondral ossification. There is, however, an upper limit for porosity and pore size set by constraints associated with mechanical properties. An increase in the empty volume results in a reduction in mechanical strength of the graft, which can be critical for regeneration of load-bearing sites. Differences in bone tissues in morphological (pore size and porosity) and mechanical properties have set challenges for fabricating biomaterial scaffolds that can meet the requirements set by the specific site of application. So, following these statements, polymer reinforcement, present in SB sample, seems to be essential to reduce the pore sizes presented in Figure 3(b). The analysis of the inner surface of the sample (Figure 3c) reveals that the porous mineral structure and polymer are easily distinguishable. Owing to their properties, mineral part appears as brighter, as in Figure 3(c), while polymer coating appears darker. EDS of zone 1 is characterised by strong signals of C and O, confirming the presence of the polymeric film; while spectrum 2, by the presence of Ca and P, shows the mineral part of the sample corresponding to HA (Figure 3d).
As recently reviewed, mathematical modelling is extremely powerful in describing the polymer degradation behaviour[37,38,41], particularly when assuming that
The consistency and robustness of the adopted approach can be verified, as mentioned, through a characteristic time analysis. The diffusion characteristic time can be computed as follows:
Degradation time can be defined as follows:
Coating thickness is clearly visible in Figure 4(a), obtained again thanks to a very precise and sharp cut that evidences the polymer coated on to the mineral matrix. Hence, as mentioned, depolymerisation kinetics is tripled in order to best adhere to the reality. The practical need, here, was to obtain a reliable description of thickness reduction in time once
In this study, a composite graft made of natural mineral matrix (calcium HA) and synthetic polymer coating (PLCL) with the addition of polysaccharides was developed. The purpose behind coating mineral matrix with PLCL and polysaccharides comes from the necessity to have higher mechanical properties together with suitable microstructure, on one side, and to improve hydrophilicity for good cell adhesion, on the other. The fabricated bone graft exhibited regular microstructure similar to healthy iliac bones with an average 27% porosity. The chemical structure of the graft was verified with FT-IR and EDS, underlining the good adhesion between polymer coating and HA. PLCL coating degradation was studied and related to its thickness using a well-known mathematical model. This approach avoids the use of trial-and-error methods and consents a better
EDS, energy-dispersive analysis; ESEM, Environmental scanning electron microscopy; FT-IR, Fourier transform infrared; HA, hydroxyapatite; micro-CT, micro-computed tomography; PLCL, poly(L-lactide-co-ε-caprolactone); ROI, regions of interest; SB, SmartBone[®].
The authors thank Phoenix|x-ray (a GE company) for the support with the micro-CT analysis.
All authors contributed to the conception, design, and preparation of the manuscript, as well as read and approved the final manuscript.
All authors abide by the Association for Medical Ethics (AME) ethical rules of disclosure.
Simulation input data
|Monomer molecular weight||MWmon = 97.3 mg/mmol|
|Polymer Density||ρ = 1200 mg/cm
|Average number molecular weight||MWn = 136000 mg/mmol|
|Average weight molecular weight||MWw = 204000 mg/mmol|
|Monomer effecitve diffusvity||DM = 3.6 10-5 mm2/h|
|Water effective diffusvity||Dw = 3.6 10-3 mm2/h|
|Polymerisation kinetic constant||kp = 6.7135 .10-7 mm2/mmol/h|
|Equilibrium constant||KEQ = 10-3|
|Polydispersity||PD = 1.5|