Abstract
In the present study, polycaprolactone–tricalcium phosphate (PCL/TCP) scaffolds with two different fibre laydown patterns, which were coated with hydroxyapatite and gelatine, were used as an approach for optimizing bone regeneration in a critical-sized calvarial defect. After 12 weeks, bone regeneration was quantified using microcomputed tomography (micro-CT) analysis, biomechanical testing, and histological evaluation. Notably, the experimental groups with coated scaffolds showed lower bone formation and lower biomechanical properties within the defect compared to the uncoated scaffolds. Surprisingly, the different laydown pattern of the fibres resulted in different bone formation and biomechanical properties: the 0°/60°/120° scaffolds revealed lower bone formation and biomechanical properties compared to the 0°/90° scaffolds in all the experimental groups. Therefore, future bone regeneration strategies utilizing scaffolds should consider scaffold architecture as an important factor during the scaffold optimization stages in order to move closer to a clinical application.
Introduction
In general, bone is a dynamic and multifunctional organ, capable of good healing and remodelling capacities. However, in certain cases, surgical therapeutic intervention is required due to a limited intrinsic regeneration potential. Besides the conventional surgical procedures, the concept of tissue engineering has emerged as an important approach to bone regeneration research. There are two major bone tissue engineering approaches to develop novel treatment concepts involving scaffolds: cell-based and cell-free. Scaffolds serve as space holders for cells and allow ingrowth of host tissues into the reconstruction site after transplantation. Thus, they provide structures that facilitate the three-dimensional proliferation, differentiation, and orientation of cells in order to enable tissue-like growth in vivo. Scaffolds facilitate the transfer of loads to surrounding tissues and preferably allow the reconstruction site to be mechanically competent directly after insertion. Scaffolds also provide a space in which tissue development and maturation towards complex multi-cellular systems can occur.
The properties of a material’s surface can directly influence single cell behaviour, in the same way the three-dimensional (3D) structure plays a critical role in the orchestration of tissue formation in vivo. Surface properties and microstructure of a material refers to the material at the nanoscale or microscale level, whereas scaffold architecture defines the structure of the biomaterial in space at a tissue-length scale. Scaffolds not only provide the structural basis for cells to form a 3D tissue-like construct in vivo, but they also influence the vascularity.
To improve the mechanical and biochemical properties of the scaffolds, calcium phosphate ceramic particles have been mixed into the polymer phase directly. However, an important aspect that has been neglected in the context of bone tissue engineering is the interfacial properties between the ceramic and matrix phases, and therefore, limited improvements have been seen regarding the mechanical properties of polymer/ceramic composite scaffolds compared to polymeric scaffolds. Furthermore, the masking of the ceramic particles by a very thin polymer layer on the scaffold surface by the so called ‘skinning effect’ may diminish the proliferative and osteoconductive properties offered by some bioactive ceramic particles.
Hence, to improve the proliferative and osteoconductive properties of polymer/ceramic composite scaffolds, coating the scaffolds with a layer of mineralized apatite deposit is considered an efficient approach. In this context, to improve the mechanical properties of the polymer/ceramic composite scaffolds, we have developed silanized polycaprolactone/tricalcium phosphate (PCL/TCP(si)) scaffolds, which have significantly improved mechanical properties compared to standard PCL/TCP scaffolds. Moreover, to improve the osteoconductive properties of the PCL/TCP(si) scaffolds, biomimetic coatings were applied. The developed biomimetic apatite-coated PCL/TCP(si) showed excellent mechanical properties and promising proliferative and osteoconductive properties in vitro. Although, it is well known that the internal pore size and the architecture of the scaffolds influence the capacity for bone regeneration, the optimal properties are still much debated. The effect of different laydown patterns and hence pore architecture has been conducted in terms of mechanical and in vitro characterization. However, the effect of different laydown patterns on bone regeneration in an in vivo model has not yet been tested.
To address these issues, the use of PCL/TCP(si) scaffolds and carbonated hydroxyapatite-coated scaffolds (PCL/TCP(si)–CHA) were investigated in the present study as an approach for optimizing the bone regenerative capabilities of the respective scaffolds in a critical-sized calvarial defect model. Furthermore, in conjunction with the apatite coating, whilst keeping the overall porosities the same, the effects of two different laydown patterns of the scaffolds, namely 0–60° and 0–90°, on bone regeneration capabilities were analysed.
Materials and methods
Materials
Poly( ɛ -caprolactone) (PCL; number average molecular weight ( M n ): 80,000), 3-glycidoxypropyl trimethoxysilane (GPTMS), acetic acid (CH 3 COOH), calcium chloride (CaCl 2 ), potassium hydrophosphate (K 2 HPO 4 ), phosphoric acid (H 3 PO 4 , 85% solution in water), sodium carbonate (Na 2 CO 3 ), and sodium hydroxide (NaOH) were purchased from Sigma–Aldrich, Singapore. Tricalcium phosphate (TCP) was purchased from Progentix, the Netherlands.
Scaffold fabrication
Medical grade PCL/TCP(si) composites were fabricated as previously reported. In brief, surface activation of TCP was achieved using phosphoric acid at room temperature for 2 h. The surface-activated TCP was then washed with distilled water, and GPTMS (4 wt% of TCP) was added into the TCP solution and refluxed at 75 °C for 24 h. The GPTMS-modified TCP (TCP(si)) was collected by filtration and the washed TCP(si) was incorporated into PCL solution through homogenization. The homogenized composite was dried and finally annealed to give mPCL/TCP(si) composite. mPCL/TCP(si) scaffolds were fabricated using an in-house screw extrusion system (SES) with screw rotational speed of 25 rpm and nozzle diameter of 0.3 mm at a processing temperature of 85 °C. The scaffolds were fabricated with a size of 5 mm in diameter and 2 mm in thickness and two different laydown patterns, namely 0°/90° and 0°/60°/120° ( Fig. 1 A and B ). The final composition for the scaffold fabrication was PCL (80 wt%) and TCP (20 wt%). Scaffolds used in this study showed similar characteristics with a porosity of 67–71%, a pore size of 420–500 μm, and a scaffold surface area of 65–73 mm 2 .
Surface coating
The fabricated mPCL/TCP scaffolds were first treated in 10 ml of 5 M NaOH at room temperature for 12 h, followed by thorough washing with deionized water to remove residual NaOH. The NaOH-treated scaffolds were then dipped alternately into calcium chloride solution and potassium hydrophosphate solution to obtain a CaHPO 4 coating as a nucleation site for the next CHA coating. In brief, the NaOH-treated scaffolds were dipped in 20 ml of 0.2 M aqueous CaCl 2 solution for 10 min and then dipped in deionized water for 5 Y s, followed by air drying for 3 min. The sample was subsequently dipped in 20 ml of 0.2 M aqueous K 2 HPO 4 solution for 10 min and then dipped in deionized water for 5 Y s, followed by air drying for 3 min. The whole process was repeated three times. The CaHPO 4 -coated scaffolds were immersed in 20 ml of 0.1 M CH 3 COOH, and then 10 ml of 0.1 M CaCl 2 and 6 ml of 0.1 M H 3 PO 4 (Ca/P = 1.66) were dropped slowly through separate syringe pumps under stirring. The pumps were adjusted to keep the ratio of Ca/P at 1.66. After further stirring for 30 min, 18 ml of 0.1 M Na 2 CO 3 with the molar ratio of CO 3 /PO 4 = 3 was gradually added. The mixture was stirred for a further 30 min and then the pH of the mixture was adjusted to 9 using 1 M NaOH. The CHA-coated mPCL/TCP scaffolds (PCL/TCP–CHA) were collected after ageing the solution for 3 h. Finally, the scaffolds were thoroughly washed with deionized water and freeze-dried ( Fig. 1 C and D).
Animal surgery
Fifteen skeletally mature male Lewis rats were obtained from the Animal Resources Centre, Canning Vale, Western Australia. The rats were housed at the QUT Medical Engineering Research Facility (MERF) at the Prince Charles Hospital, Chermside. The animals received water and pelleted rations ad libitum throughout the experiment. The animal ethics committee of the Queensland University of Technology approved all experiments (Ethics-No. 09-1136). The rats were subjected to critical-sized bone defect creation in their skull and implantation of the PCL/TCP(si) scaffolds. The rats were assigned to five experimental groups. Six defects per group were allocated, with an empty control group and four groups with the different scaffold coatings and laydown patterns.
Surgical procedure
All rats were operated under general anaesthesia. Buprenorphine (0.01–0.05 mg/kg subcutaneously) was used preoperatively for pre-emptive analgesia and postoperatively every 6–12 h as painkiller. General anaesthesia was provided using a mixture of ketamine and xylazine (75–100 mg/kg ketamine + 5–10 mg/kg xylazine intraperitoneal in the same syringe). Rats were handled briefly by hand for intraperitoneal injection of anaesthetic and then released into a separate cage until ready for surgery. The frontoparietal region was prepared by clipping hair with a delicate clipper and vigorous disinfection was achieved by application of chlorhexidine in alcohol solution. One dose of broad-spectrum antibiotic was given to the rats before surgery as prophylaxis. In order to produce critical-sized bone defects, a sagittal incision of approximately 20 mm was performed over the scalp of the animal. A full-thickness bone defect (5 mm in diameter) was then trephined in the centre of each parietal bone (two defects per calvaria) using a slow-speed dental drill with irrigation to prevent heat damage of the host bone. Caution was taken when drilling down the bone not to damage the underlying exposed dura mater ( Fig. 2 ). A 2-mm wide strip of the marginal periosteum surrounding the defect was then removed. According to the implantation plan, both bone defects in each rat were implanted with one of the treatment modalities described above. During anaesthesia, surgery, and the immediate postoperative period, the rats were kept warm on a heating pad and after that were transferred to a clean warm cage for recovery.
Animals were kept for 12 weeks after surgery and then sacrificed by CO 2 inhalation. To collect the implants, the skin was dissected and the entire skull containing the defects/scaffolds was removed for further analysis.
Microcomputed tomography (micro-CT)
Twelve weeks after surgery, mineralization within the constructs was quantified using a Micro-CT 40 scanner (SCANCO Medical, Brüttisellen, Switzerland). Samples were scanned at an energy of 55 kVp and intensity of 145 μA with 226 ms integration time, resulting in an isotropic voxel size of 36 μm. From the scanned volume, a cylindrical region of interest (ROI), corresponding to the defect size of 5 mm diameter and at the location of the original defect, was selected for analysis. After segmentation of the mineralized tissue with a threshold of 220 (equivalent to 312 mg hydroxyapatite/cm 3 (mgHA/cm 3 )), a Gauss filter width of 0.8, and filter support of 1.0, the mineralized matrix volume was quantified throughout the entire construct and presented as bone volume in mm 3 .
Mechanical testing
After micro-CT analysis, the samples were wrapped in wet gauze and stored at −20 °C until further analysis. Upon thawing, the rat skulls were potted into Petri dishes with polymethylmethacrylate bone cement (Meliodent Rapid Repair, Heraeus Kulzer) to enable stable fixation for the mechanical testing. Non-destructive micro-compression on the calvaria defects was performed using a Micro Tester 5848 (Instron) with a 10-N load cell. An indenter probe was micro-fabricated for the test. Micro-compressions of up to 50% strain were conducted at an average of eight different locations on each defect site, and the load-displacement and stiffness (compression modulus) were determined. The probe locations were identified to be the pore spaces (between the scaffold struts) of the constructs to measure the modulus of regenerated tissue rather than scaffold material. Intact calvarial bone, soft tissue, and the struts of the scaffolds were used as controls. Push-out tests were then conducted to evaluate the functional mechanical integration of the tissue-engineered constructs into the host calvaria, and were performed on the Micro Tester 5848 (Instron) with a 1-kN load cell. An indenter probe of 4.5 mm diameter, slightly smaller than the scaffold diameter of 5 mm, was fabricated for the test. Six specimens were exploited for each group.
Histology/immunohistochemistry
For processing of decalcified samples into paraffin, parietal bone was fixed in 10% neutral buffered formalin for 24 h and decalcified in 15% ethylenediaminetetraacetic acid (EDTA) for 3 weeks at 4 °C. The samples were then serially dehydrated in ethanol in a tissue processor (Excelsior ES, Thermo Scientific, Franklin, MA, USA) and embedded in paraffin. Sections (5-μm) were taken using a microtome (Leica RM 2265). The slides were then deparaffinized with xylene and rehydrated with serial concentrations of ethanol, before being stained with haematoxylin and eosin (Sigma–Aldrich) and mounted with Eukitt mounting medium (Fluka Biochemika, Milwaukee, WI, USA).
For immunohistochemistry, sections were deparaffinized with xylene and rehydrated with serial concentrations of ethanol. Subsequently, sections were rinsed in distilled water and placed in 0.2 M Tris–HCl buffer (pH 7.4). Endogenous peroxidase activity was blocked by incubating the sections in 3% H 2 O 2 in Tris–HCl for 20 min. This was followed by three washes with Tris buffer (pH 7.4) for 2 min each. Sections were incubated with proteinase K (DAKO, Botany, Australia) for 20 min and subsequently incubated with 2% bovine serum albumin (BSA) (Sigma, Sydney, Australia) in DAKO antibody diluent (DAKO) in a humidified chamber at room temperature for 20 min to block non-specific binding sites. Afterwards, immunohistochemical staining was performed using a primary mouse antibody specific to the osteogenic marker type I collagen (provided by Larry Fischer, National Institutes of Health, Bethesda, MD, USA).
Non-immunized rabbit IgG (DAKO) was used as an isotype control to rule out non-specific reactions of rabbit IgG with rat tissues as well as non-specific binding of the secondary antibodies and/or peroxidase-labelled polymer to rat tissues. The sections were incubated with the specific antibody or negative control in humidified chambers at 4 °C overnight. Sections were then washed three times for 2 min with Tris buffer (pH 7.4) and incubated with peroxidase-labelled dextran polymer conjugated to goat anti-mouse and anti-rabbit immunoglobulins (DAKO EnVision+ Dual Link System Peroxidase, DAKO) at room temperature in humidified chambers for 60 min. Colour was developed using a liquid 3,3-diaminobenzidine (DAB)-based system (DAKO). Kaiser’s glycerol gelatine (DAKO) was used for coverslip mounting.
Statistics
Statistical analysis was performed for all the quantitative results using the Student’s t -test for comparing means from two independent sample groups. A confidence level of 95% was used; statistical significance was set at P < 0.05.
Materials and methods
Materials
Poly( ɛ -caprolactone) (PCL; number average molecular weight ( M n ): 80,000), 3-glycidoxypropyl trimethoxysilane (GPTMS), acetic acid (CH 3 COOH), calcium chloride (CaCl 2 ), potassium hydrophosphate (K 2 HPO 4 ), phosphoric acid (H 3 PO 4 , 85% solution in water), sodium carbonate (Na 2 CO 3 ), and sodium hydroxide (NaOH) were purchased from Sigma–Aldrich, Singapore. Tricalcium phosphate (TCP) was purchased from Progentix, the Netherlands.
Scaffold fabrication
Medical grade PCL/TCP(si) composites were fabricated as previously reported. In brief, surface activation of TCP was achieved using phosphoric acid at room temperature for 2 h. The surface-activated TCP was then washed with distilled water, and GPTMS (4 wt% of TCP) was added into the TCP solution and refluxed at 75 °C for 24 h. The GPTMS-modified TCP (TCP(si)) was collected by filtration and the washed TCP(si) was incorporated into PCL solution through homogenization. The homogenized composite was dried and finally annealed to give mPCL/TCP(si) composite. mPCL/TCP(si) scaffolds were fabricated using an in-house screw extrusion system (SES) with screw rotational speed of 25 rpm and nozzle diameter of 0.3 mm at a processing temperature of 85 °C. The scaffolds were fabricated with a size of 5 mm in diameter and 2 mm in thickness and two different laydown patterns, namely 0°/90° and 0°/60°/120° ( Fig. 1 A and B ). The final composition for the scaffold fabrication was PCL (80 wt%) and TCP (20 wt%). Scaffolds used in this study showed similar characteristics with a porosity of 67–71%, a pore size of 420–500 μm, and a scaffold surface area of 65–73 mm 2 .
