- Email: [email protected]
Role of scaffold internal structure on in vivo bone formation in macroporous calcium phosphate bioceramics
ARTICLE IN PRESS
Biomaterials 27 (2006) 3230–3237 www.elsevier.com/locate/biomaterials
Role of scaffold internal structure on in vivo bone formation in macroporous calcium phosphate bioceramics Maddalena Mastrogiacomoa,b,1, Silvia Scaglionec,d,1, Roberta Martinettie, Laura Dolcinie, Francesco Beltramec,d, Ranieri Canceddaa,b, Rodolfo Quartod,f, a
Dip. di Oncologia, Biologia e Genetica, Universita` di Genova, Italy b Istituto Nazionale per la Ricerca sul Cancro, Genova, Italy c Dip. di Informatica, Sistemistica, Telematica, Universita` di Genova, Italy d Centro Biotecnologie Avanzate, Genova, Italy e FIN-CERAMICA FAENZA s.r.l., Faenza, Italy f Dip. di Chimica e Tecnologie Farmaceutiche ed Alimentari, Universita` di Genova, Italy Received 29 August 2005; accepted 27 January 2006 Available online 20 February 2006
Abstract Purpose of this study was the analysis of the role of density and pore interconnection pathway in scaffolds to be used as bone substitutes. We have considered 2 hydroxyapatite bioceramics with identical microstructure and different macro-porosity, pore size distribution and pore interconnection pathway. The scaffolds were obtained with two different procedures: (a) sponge matrix embedding (scaffold A), and (b) foaming (scaffold B). Bone ingrowth within the two bioceramics was obtained using an established model of in vivo bone formation by exogenously added osteoprogenitor cells. The histological analysis of specimens at different time after in vivo implantation revealed in both materials a significant extent of bone matrix deposition. Interestingly enough, scaffold B allowed a faster occurrence of bone tissue, reaching a steady state as soon as 4 weeks. Scaffold A on the other hand reached a comparable level of bone formation only after 8 weeks of in vivo implantation. Both scaffolds were well vascularised, but larger blood vessels were observed in scaffold A. Here we show that porosity and pore interconnection of osteoconductive scaffolds can influence the overall amount of bone deposition, the pattern of blood vessels invasion and finally the kinetics of the bone neoformation process. r 2006 Elsevier Ltd. All rights reserved. Keywords: Bone tissue engineering; Porous bioceramics; Interconnection pathway; Vascularisation
1. Introduction Bone regeneration in presence of synthetic bone graft can take a rather long time for the complete anatomical and functional recovery. With this in mind, scientists are focusing on higher performance biomaterials in bone tissue engineering to decrease the bone healing time. Many synthetic materials as bone substitute are easily available, including metals, natural and synthetic polymers, ceramics and glasses [1–22]. Calcium phosphate and hydroCorresponding author. Laboratorio Cellule Staminali, Centro Biotecnologie Avanzate, Largo Rosanna Benzi no. 10, 16132 Genova, Italy. Tel.: +39 010 5737 240; fax: +39 010 5737 505. E-mail address: [email protected] (R. Quarto). 1 Equally contributed to the experimental work.
0142-9612/$ - see front matter r 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.biomaterials.2006.01.031
xyapatite ceramics are considered among the most promising bone substitutes [1–4,6,8,10–12,19–22], because of their bonelike chemical composition and mechanical properties. Hydroxyapatite ceramics, in particular, are used in orthopaedics, maxillofacial and dental implant surgery, either alone or in combination with other substances or materials. Particulate form of hydroxyapatite are also used to fill bone defects or as a coating on metal implants [5,16,21,24–27]. Besides chemical composition, the other critical parameter to improve the efficiency of biomaterials to be used in bone tissue engineering is the overall structure: density, pore shape, pore size and pore interconnection pathway [23,28–33]. Porosity is necessary for the in vivo bone tissue ingrowth since it allow migration and proliferation of osteoblasts
ARTICLE IN PRESS M. Mastrogiacomo et al. / Biomaterials 27 (2006) 3230–3237
and mesenchymal cells, and matrix deposition in the empty spaces. Several studies investigated the minimum pore size required to regenerate mineralized bone [29,30,32]. Although macroporosity has a strong impact on osteogenic outcomes, interconnection pathway plays an important role as well. An incomplete pore interconnection could represent a constraint of the overall biological system, limiting blood vessels invasion. A proper blood supply represents the base for bone tissue growth in porous biomaterials [1,34]. Bone vascularisation, besides providing nutrients essential for tissue survival, plays also a crucial role in coordinating the activity of bone cells and their migration for bone remodelling [34]. Therefore, bioceramics with high porosity and appropriate interconnection pathway should allow the tissue to infiltrate and fill the scaffold. In the present work two different technologies were used to achieve hydroxyapatite porous devices with different total porosity and interconnection pathways. Both bioceramics were compared using an established model of in vivo bone formation by exogenously added osteoprogenitor cells [35]. The osteo-inductivity of the two ceramics scaffolds, the blood vessels invasion and the extent of bone matrix deposition was assessed through the histological analysis of specimens at different time after in vivo implantation. 2. Materials and methods 2.1. Biomaterial production and characterization Samples of porous bioceramics were prepared starting from the same batch of hydroxyapatite powder but employing two different production technologies. The first bioceramic (A) was prepared by sponge matrix embedding: a cellulose matrix with controlled pore morphology and distribution was embedded in a hydroxyapatite slurry to obtain the bimodal porous structures. After drying the green sample at room temperature, a pyrolysis of the organic phase was carried out and maintained in air atmosphere at 600 1C for 2 h. The sample was then sintered at 1280 1C for 1 h. The second bioceramic (B) was prepared with an innovative process (Patent pending) changing the surface energy of the slurry during processing (foaming method): a slurry with surfactants and high powder concentration (20 wt%) was used and expanded in a known volume (40–60 vol% of the total) to achieve a controlled morphology and an inner porosity close to 80 vol%; the bioceramic in this case was sintered at 1250 1C for 1 h. Both bioceramics were produced by FinCeramica Faenza (Italy). Physico-chemical and morphological characterisation of the final components were carried out for both sets of samples. Special attention was given to the characterisation of the total porosity and of micro-macro porosity distribution. The porosity study was carried out in conformity to the standard ASTM E562 using a image analyser (Leica Imaging System Ltd. Q500MW by Qphase Application, Cambridge, UK). The morphology of the scaffolds was analysed by scanning electron microscopy (SEM) to evaluate micro and macro structure (Leica, Cambridge, UK): for SEM observation the samples were dehydrated, firmly mounted, and coated with a thin layer of a conductive material. For coating we took advantage of a small device called sputter coater which uses argon gas and a small electric field. The sample was placed in a small chamber under vacuum. Argon gas was then introduced and an electric field was used to cause an electron to be removed from the argon atoms making the atoms ions positively charged. Phases investigation and chemical analyses were carried out, respectively by X-ray diffraction
3231
analysis (CuKa radiation, Rigaku Miniflex, Tokyo, Japan) and by inductively coupled plasma-ICP (Spectroflame Modula, Spectro, equipped with Ultrasonic Nebulizer Cetac U-6000 AT+, Kleve, Germany). The surface area was carried out following B.E.T. method and measured on a Quantasorb instrument (Quantachrone Instrument Greenvale, Ny, USA) while the density was determine as mass/volume [36–39].
2.2. Osteoprogenitor cell culture and ectopic bone formation assay in ID mice Four 10-months-old female sheep of Italian Biellese strain, as bone marrow donors, were involved in this study with approval of competent ethical committee and legal authorities. Twenty millilitre of bone marrow aspirate were harvested from the posterior iliac crest of each animal. Bone marrow samples were washed twice with PBS (phosphate buffered saline). Marrow specimens were stained with a nuclear stain (0.1% methyl violet in 0.1 M citric acid) and the nucleated fraction was counted. Cells were suspended in Coon’s modified Ham’s F12 medium supplemented with 10% FCS, 100 I.U./ml penicillin and 100 mg/ml streptomycin and plated at a density of 1 106 cells/cm2. Human recombinant fibroblast growth factor 2 (FGF2) (1 ng/ml) was added since the beginning of the culture [40]. Medium was changed 2 days after the original plating and then twice a week. When culture dishes were nearly confluent (passage 0), bone marrow stromal cells (BMSC) were detached with 0.05% trypsin-0.01% EDTA and 5 105 cells were replated in 100-mm dishes (passage 1) until the next confluence. A BMSC pool obtained from four sheeps was used for all the experiments described. 2.5 106 cells were resuspended into 40 ml of 20 mg/ml fibrinogen solution in PBS. 40 ml of cell suspension in fibrinogen were combined with each cube of biomaterial (3 3 3 mm) and 25 NIH U of thrombin were added. The polymerization reaction was allowed to proceed at 37 1C for 10 min, then bioceramic/BMSC composites were subcutaneously implanted in immuno-deficient (ID) (CD-1 nu/nu) mice by using an established model of ectopic bone formation [35]. In these conditions, BMSC generate bone tissue in a fashion proportional to the number of cells implanted [41–42]. In agreement and with the approval of the competent ethical committee and legal authorities, recipient ID mice of 1 month of age, purchased from Charles River Italia, were kept in a controlled environment and given free access to food and water. Mice were anesthetized by intramuscular injection of Xilazine (20 mg/ml) and Ketamine (30 mg/ml). Bioceramic/ BMSC composites were implanted subcutaneously on the back of the mice (up to 4 implants for each animal, for an amount of 6 animal). Animals were sacrificed 2, 4, and 8 weeks after implantation. Grafts were harvested and processed for histological analysis.
2.3. Optical microscope analysis Samples were decalcified, paraffin embedded, sectioned, stained with hematoxylin/eosin and analyzed for the bone tissue quantification. The amount of newly formed bone was assessed using a Zeiss Axiophot optical fluorescence microscope (Oberkochen, Germany) and quantified using the free image analysis software Scion Image and an image processing macro previously developed [43]. For each ceramics type, twelve samples (4 at each time point) were harvested after 2, 4 and 8 weeks of implant. For each construct, ten sections of 5 mm thick cut at different levels (100 mm gap level) were HE stained for newly formed bone quantification. All images were acquired using both transmitted and fluorescence light. For each construct, the amount of bone was assessed as percentage of the total bone tissue versus total tissue formed [43]. The pores interconnection size of the two different scaffolds was measured using both SEM images and histological sections. In scaffold A interconnection path was mostly represented by ‘‘tunnels’’ (T) with a low percentage of smaller interconnections, generated by contacts (C) of adjacent tunnel walls and pores. Therefore, the limiting size of the tunnels interconnecting adjacent pores and mean diameters of smaller inter-
ARTICLE IN PRESS 3232
M. Mastrogiacomo et al. / Biomaterials 27 (2006) 3230–3237
connections were measured. For scaffold B, mean diameters of contact points of adjacent pores (C), the only interconnections of this material, were measured. For both biomaterials, eight images were analysed, all interconnection values ordered according to their size, and their distribution was plotted (average values7std dev). For each scaffolds, mean blood vessels size was obtained by using four representative histological images at 2.5 magnification. For each blood vessel section only the minor axis was considered. Average values7std dev were derived.
2.4. Statistical analysis Mean values and standard deviations were obtained analysing raw data with Microsoft Excel software. Correlation analysis was performed using the parametric Student t-test. Statistical differences were considered significant when Po0:05.
3. Results 3.1. Bioceramic characterisation The bioceramics obtained with 2 different production methods were characterised. Both porous hydroxyapatite
bioceramics were complying with the ISO 13779 standard. According to X-ray diffraction analysis they resulted to be single phase crystalline hydroxyapatite, with purity X95% according to ICCD card no. 9-432; Ca/P ratio had a value of 1.6570.02 and trace elements (total heavy metals) resulted to be under the maximum allowable limit (50 mg/kg). Both scaffolds do not display any significant resorption within 6 months of in vivo implantation, as previously shown [1]. The density determined as mass/volume of the samples analysed was found to be 1.26 g/cm3 (70.16) for bioceramic A and 0.63 g/cm3 (70.10) for bioceramic B. These values were compatible with a total porosity of 6575% (A) and 8073% (B). Surface area was measured for both ceramics: in scaffold A an average of 0.8770.04 m2/g was observed, despite the 1.6370.08 m2/g of surface area for the scaffold B. The two bioceramic structures are shown in Fig. 1. At low magnification (panels A and B) were evident for the differences in pore size, shape and interconnection path-
Fig. 1. Scanning electron microscope (SEM) analysis at low magnification shows the macro structure of the two different materials used as osteoconductive grafts. The morphology of sponge matrix (scaffold type A) and of the foam sample (scaffold type B) are shown in panels A and B, respectively. Biomaterials show a different porosity, in terms of pores shape, size and interconnection. Black spots represent interconnections between neighbour pores. SEM analysis at higher magnification of scaffold A and scaffold B is also shown (C and D, respectively). Biomaterials display a very similar microstructure and micro-porosity, according to the same size of grains used. Bars: A ¼ 1000 mm, B ¼ 200 mm; C ¼ 5 mm; D ¼ 2 mm. A schematic view of the two scaffold’s structures is shown: scaffold type A is obtained starting from a cellulose matrix (black network in E), whose trabeculae are covered by the HA slurry (grey part) thus generating a porous structure (white area representing the final pores). On the other hand, structure of scaffold type B (F) may be seen as multiple interconnected gas bubbles (white area) within a HA matrix (grey area).
ARTICLE IN PRESS M. Mastrogiacomo et al. / Biomaterials 27 (2006) 3230–3237
ways between the two scaffolds. At higher magnification (panels C and D) the microstructures of the scaffolds were displayed, in particular the hydroxyapatite grain shape and size, the intergranular porosity and the micro-interconnections. For both scaffolds the porosity distribution was found to be bi-modal and both scaffolds were formed by interconnected micro-macropores, where micropores represent the empty inter-grain space, while macropores are the macro-cavities obtained during the ceramic processing (Fig. 1). Grain size resulted to be in the range 0.5–2.5 mm and the size of micro/intergranular pores was in the range of 0.2–10 mm for both materials (Fig. 1(C–D)). An important difference between the two scaffolds was observed. Given the two production methods used, the pore interconnection pathway in bioceramic A was mostly represented by ‘‘tunnels’’ and therefore corresponded to the macro-pore size itself with a low percentage of smaller interconnections between adjacent tunnel walls (Fig. 1(E)), whereas in bioceramic B the interconnection pathway was mainly due to confluence of adjacent pore walls (Fig. 1(F)).
3233
Therefore, based on the analysis performed on SEM images of the raw materials (Fig. 2(C and E)), in scaffold A most of the interconnections were found to be above 200 mm, whereas in scaffold B about 70% of the interconnections displayed a size below 100 mm with only a bare 3% above 200 mm (Fig. 2(A)). The analysis performed on histological sections (Fig. 2(D and F)) confirmed with minor variations, possibly due to different treatments of the samples, the results obtained on SEM images (Fig. 2(B)). 3.2. Ectopic bone formation in ID mice Equal numbers of BMSC were loaded onto scaffold cubes and subcutaneously implanted in ID mice. Samples were harvested after 2, 4 and 8 weeks of in vivo implant and histologically analysed. Neo-bone formation was already present after 2 weeks of implant in the peripheral area of scaffold type B, whereas no trace of bone tissue was detected at same time in the less porous bioceramic A (data not shown).
Fig. 2. The distribution of interconnection pore size of the two different scaffolds was plotted, either using SEM images of scaffolds before their use (A) or using histological sections of samples after 8 weeks in vivo (B). In scaffold A, most of the interconnections, represented by ‘‘tunnels’’ (T) and corresponding to the macro-pore, were found having a size above 200 mm, whereas in scaffold B about 70% of the interconnections displayed a size below 100 mm, corresponding to the contact (C) between adjacent pores, with almost no pore interconnections above 200 mm. Arrows indicate the measurement procedure, which was carried out using both SEM images of the scaffolds type A and B (C and E, respectively) and histological sections (D and F, respectively).
ARTICLE IN PRESS 3234
M. Mastrogiacomo et al. / Biomaterials 27 (2006) 3230–3237
After 4 weeks of in vivo implantation, scaffold A displayed a low amount of bone tissue deposition (Fig. 3(A)), whereas in scaffold B, a significant amount of bone tissue was evenly distributed within pores (Fig. 3(B)). After 8 weeks, a significant improvement of bone formation was observed in scaffold A (Fig. 3(C)), whereas no major changes were observed in the extent of bone formed in scaffold B (Fig. 3(D)). At higher magnification, representative histological sections of the two ceramics displayed a good bone matrix deposition filling the pores from the periphery to the centre, while the interconnections of adjacent pores were filled by undifferentiated vascularised mesenchymal tissue (Fig. 3(E and F)). The amount of bone matrix deposition within the implant was quantified. Histological analysis and computer-assisted quantification of the bone/total tissue formed revealed that the bone formation within scaffold B pores had already reached after 4 weeks the 35% of the total
tissue neo-formed, value close to that found in the twin set of samples analysed after 8 weeks (36%). In bioceramic A, on the other hand, by 4 weeks newly formed bone was only 22% of the total tissue and a 37% of bone tissue formation was obtained only after 8 weeks of in vivo implantation. In both materials, a significant amount of blood vessel was observed within the bioceramic pores, as well as between contiguous pores running through the interconnections. In scaffold A though a higher number of large blood vessels were found (mean diameter: 91738 mm) whereas in scaffold B, although the vascularisation was well represented, we never observed large blood vessel running through several consecutive pores (mean diameter: 61718 mm). Fig. 4 displays a representative histological field of a scaffold A tunnel run by a large blood vessel and a comparable field of scaffold B where a smaller blood vessel runs through an interconnection joining two adjacent pores (Fig. 4(A) and (B), respectively).
Fig. 3. Histology of tissues formed in ectopic transplants of sheep BMSC loaded on two different biomaterials (scaffold A: A–C–E; scaffold B: B–D–F) in immunodeficient mice after 4 and 8 weeks (A–B and C–D, respectively). A deposition of bone matrix from the edge of ceramic pore toward the centre is observed. After 4 weeks of implantation, scaffold A displayed a low amount of bone tissue deposition (A), whereas in scaffold B a relevant amount of neobone tissue deposition, stained in pink, was evenly distributed within pores (B). After 8 weeks, a significant enhancement of bone formation was observed in scaffold A (C), whereas no major changes were evident in the scaffold B (D). A good interconnection of pores and vascularisation is also shown in both ceramics. Panel E and F: higher magnification of comparable histological sections. Staining: hematoxylin-eosin. Bars: A2D ¼ 1000 mm, E2F ¼ 250 mm.
ARTICLE IN PRESS M. Mastrogiacomo et al. / Biomaterials 27 (2006) 3230–3237
3235
Fig. 4. Blood vessel running through pore interconnections: (A) (scaffold A) a representative large blood vessel is displayed; (B) a smaller blood vessel running through an interconnection joining two adjacent pores of the scaffold B. Bar: 100 mm. B ¼ bone tissue; L ¼ blood vessel lumen; RBC ¼ red blood cells.
4. Discussion A number of bone substitute biomaterials are easily available. Among modern biomaterials, bioceramics (calcium phosphate and hydroxyapatite) are very promising candidates in bone substitution, because of their bone-like chemical composition and mechanical properties [1–4,6,8,10–12,19–22]. A modern bioceramic to be used as a bone substitute should be highly efficient in terms of bone ingrowth to allow a faster regeneration and a better functional recovery. Besides its chemical composition, one of the most important parameters in a three-dimensional scaffold is its internal structure. Several reports have been aimed at enlightening relationships among different parameters (density, pore shape, pore size and pore interconnection pathway) of bone substitutes biomaterials and bone ingrowth [23,28–33]. Although pore size, extent of interconnection and interconnection pathway are critical factors for bone formation, it is still not completely clear how these parameters are able to influence the efficiency of the bone neo-formation process [30]. In the present study we investigated how different internal architecture of porous bioceramics would affect the total amount of bone neo-formation. The two materials we have considered had the same chemical composition (stoichiometric hydroxyapatite) and micro-structure, but different fabrication procedures and therefore different architectures in terms of macro-porosity, pore size distribution and interconnection pathway. To verify bone formation occurrence in the biomaterials under study, we used an established model of ectopic bone formation [35]. In such an experimental model, BMSC were loaded onto the biomaterials and subcutaneously implanted in immunodeficient mice. In these conditions, BMSC generate bone tissue in a fashion proportional to the number of cells implanted. In our experiments, bioceramic samples were loaded with the same amount of sheep BMSC, which, considering the densities and the overall porosity of the 2 materials, was an advantage for bioceramic A with a higher cell density for surface unit. Still, the histological analysis of bone tissue formed after 2
weeks already revealed that bone had been evidently deposited only in the less dense bioceramic (scaffold B), whereas undifferentiated mesenchymal tissue was found in the corresponding samples of the more dense material (scaffold A) despite the estimated higher number of BMSC/pore seeded. The analysis of the 4-week samples revealed that within scaffold B pores bone formation level had already reached a significant 35%, value which was very close to that found in the twin samples of foaming HA after 8 weeks in vivo. Type A bioceramic by 4 weeks was also colonised by newly formed bone but at a lower extent (22%) and had gradual increase the level of bone formation reaching a 37% only after 8 weeks of in vivo implantation. These data indicate that bone formation had occurred more rapidly and more efficiently in the less dense material (scaffold B) despite the estimated lower number of cells loaded for surface unit. It would be nice to speculate that the higher in vivo bone forming efficiency of scaffold B at earlier time points could be due to a ‘‘filter effect’’ of this scaffold, according to its pore interconnection size distribution. Scaffold B is less dense, but with smaller pore and smaller interconnection size, therefore it could generate a higher cell density in the outer zone of the scaffold, which in turn would possibly promote an earlier and more vigorous bone formation. This speculation is in agreement with evidences reported by Wendt and coworkers who have shown a non-homogeneous cell distribution, leading to the formation of a cell density gradient, in comparable scaffolds when cells are loaded under similar conditions [44]. On the other hand, bone formation rate in scaffold B, already high after 4 weeks, remained almost unchanged up to 8 weeks. This evidence suggests that bone neo-formation is proportional to the surface available for tissue growth and may also suggest that bone neo-formation is somehow limited by factors intrinsic to the material itself (i.e. resorbability) and possibly to the physiological equilibrium reached by the ‘‘bone organoid’’ (bone matrix, blood vessels, marrow and undifferentiated mesenchymal cells which all together constitute the bone organoid) [45]. Pores filled with newly formed bone in fact, always present a peripheral area of bone tissue and a central area of undifferentiated vascularised mesenchymal tissue. It would
ARTICLE IN PRESS 3236
M. Mastrogiacomo et al. / Biomaterials 27 (2006) 3230–3237
be nice to postulate that these four components (bone matrix, blood vessels, marrow and undifferentiated mesenchymal cells) coexist in a kind of steric but dynamic equilibrium, which would mean, in terms of bone formation, reaching a saturation level. Such level could be overpassed only using more resorbable bioceramics (i.e. tricalcium phosphate, carbonate hydroxyapatite, magnesium-doped hydroxyapatite, etc.), which would allow bone to grow in a fashion proportional to biomaterial resorption. From the histological analysis of the two scaffolds it was also evident that the level of vascularisation was abundant in both materials and virtually all pores were reached by blood vessels of various size. An interesting observation was that, according to the pore interconnection size distribution, in scaffold B, the interconnections appeared to be limiting the size of blood vessels running through adjacent pores. In other words, the interconnection pathway could limit the vascularisation (at least in terms of vessel maximal diameter) of bioceramics representing a bottleneck for blood vessel invasion. This physical constraint could have important implications with larger ceramic specimens where larger vessel are indeed required.
[2]
[3]
[4]
[5]
[6]
[7]
[8]
5. Conclusion [9]
All together our data indicate that in osteoconductive bioceramics, both pore size and pore interconnection pathway are critical factors. Pore size would be directly related to bone formation, since it provides surface and space for cell adhesion and bone ingrowth; pore interconnection on the other hand would provide the way for cell distribution/migration and allow an efficient in vivo blood vessel formation suitable for sustaining bone tissue neo-formation and possibly remodelling. The results obtained in the present study show that a significant improvement in bone repair can be achieved using novel technologies of bioceramic production. Our observations also suggest that scaffolds characterised by an engineered structure with larger interconnections and higher total available surface area seem to be the right direction to investigate for future bone scaffold improvements. Acknowledgements We are grateful to Mr. G. Martignani and L. Pilotti (ENEA, Faenza) for their contribution to the Image Analyzer study and SEM investigation. This work supported with financial support from the European Commission under the Growth Programme GRD1-1999-10590.
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17] [18]
References [19] [1] Boyde A, Corsi A, Quarto R, Cancedda R, Bianco P. Osteoconduction in large macroporous hydroxyapatite ceramic implants: evidence
for a complementary integration and disintegration mechanism. Bone 1999;24(6):579–89. Dong J, Kojima H, Uemura T, Kikuchi M, Tateishi T, Tanaka J. In vivo evaluation of a novel porous hydroxyapatite to sustain osteogenesis of transplanted bone marrow-derived osteoblastic cells. J Biomed Mater Res 2001;57(2):208–16. Dong J, Uemura T, Shirasaki Y, Tateishi T. Promotion of bone formation using highly pure porous beta-TCP combined with bone marrow-derived osteoprogenitor cells. Biomaterials 2002;23(23): 4493–502. Flautre B, Anselme K, Delecourt C, Lu J, Hardouin P, Descamps M. Histological aspects in bone regeneration of an association with porous hydroxyapatite and bone marrow cells. J Mater Sci Mater Med 1999;10(12):811–4. Fujimura K, Bessho K, Okubo Y, Segami N, Iizuka T. A bioactive bone cement containing Bis-GMA resin and A-W glass-ceramic as an augmentation graft material on mandibular bone. Clin Oral Implants Res 2003;14(5):659–67. Gauthier O, Muller R, von Stechow D, Lamy B, Weiss P, Bouler JM, et al. In vivo bone regeneration with injectable calcium phosphate biomaterial: a three-dimensional micro-computed tomographic, biomechanical and SEM study. Biomaterials 2005;26(27):5444–53. Kim H, Suh H, Jo SA, Kim HW, Lee JM, Kim EH, et al. In vivo bone formation by human marrow stromal cells in biodegradable scaffolds that release dexamethasone and ascorbate-2-phosphate. Biochem Biophys Res Commun 2005;332(4):1053–60. Kon E, Muraglia A, Corsi A, Bianco P, Marcacci M, Martin I, et al. Autologous bone marrow stromal cells loaded onto porous hydroxyapatite ceramic accelerate bone repair in critical-size defects of sheep long bones. J Biomed Mater Res 2000;49(3):328–37. Livingston T, Ducheyne P, Garino J. In vivo evaluation of a bioactive scaffold for bone tissue engineering. J Biomed Mater Res 2002;62(1): 1–13. Lu JX, Gallur A, Flautre B, Anselme K, Descamps M, Thierry B, et al. Comparative study of tissue reactions to calcium phosphate ceramics among cancellous, cortical, and medullar bone sites in rabbits. J Biomed Mater Res 1998;42(3):357–67. Livingston TL, Gordon S, Archambault M, Kadiyala S, McIntosh K, Smith A, et al. Mesenchymal stem cells combined with biphasic calcium phosphate ceramics promote bone regeneration. J Mater Sci Mater Med 2003;14(3):211–8. Marcacci M, Kon E, Zaffagnini S, Vascellari A, Neri MP, Iacono F. New cell-based technologies in bone and cartilage tissue engineering. I. Bone reconstruction. Chir Organi Mov 2003;88(1):33–42. Mauney JR, Blumberg J, Pirun M, Volloch V, Vunjak-Novakovic G, Kaplan DL. Osteogenic differentiation of human bone marrow stromal cells on partially demineralized bone scaffolds in vitro. Tissue Eng 2004;10(1-2):81–92. Mauney JR, Jaquiery C, Volloch V, Heberer M, Martin I, Kaplan DL. In vitro and in vivo evaluation of differentially demineralized cancellous bone scaffolds combined with human bone marrow stromal cells for tissue engineering. Biomaterials 2005;26(16): 3173–85. Meinel L, Karageorgiou V, Hofmann S, Fajardo R, Snyder B, Li C, et al. Engineering bone-like tissue in vitro using human bone marrow stem cells and silk scaffolds. J Biomed Mater Res A 2004;71(1):25–34. Moreira-Gonzalez A, Lobocki C, Barakat K, Andrus L, Bradford M, Gilsdorf M, et al. Evaluation of 45S5 bioactive glass combined as a bone substitute in the reconstruction of critical size calvarial defects in rabbits. J Craniofac Surg 2005;16(1):63–70. Nazarov R, Jin HJ, Kaplan DL. Porous 3-D scaffolds from regenerated silk fibroin. Biomacromolecules 2004;5(3):718–26. Sul YT, Johansson C, Byon E, Albrektsson T. The bone response of oxidized bioactive and non-bioactive titanium implants. Biomaterials 2005;26(33):6720–30. Tampieri A, Celotti G, Sprio S, Delcogliano A, Franzese S. Porositygraded hydroxyapatite ceramics to replace natural bone. Biomaterials 2001;22(11):1365–70.
ARTICLE IN PRESS M. Mastrogiacomo et al. / Biomaterials 27 (2006) 3230–3237 [20] Turhani D, Cvikl B, Watzinger E, Weissenbock M, Yerit K, Thurnher D, et al. In vitro growth and differentiation of osteoblast-like cells on hydroxyapatite ceramic granule calcified from red algae. J Oral Maxillofac Surg 2005;63(6):793–9. [21] Marcacci M, Kon E, Zaffagnini S, Giardino R, Rocca M, Corsi A, et al. Reconstruction of extensive long-bone defects in sheep using porous hydroxyapatite sponges. Calcif Tissue Int 1999;64(1):83–90. [22] Quarto R, Mastrogiacomo M, Cancedda R, Kutepov SM, Mukhachev V, Lavroukov A, et al. Repair of large bone defects with the use of autologous bone marrow stromal cells. N Engl J Med 2001;344(5):385–6. [23] Chang BS, Lee CK, Hong KS, Youn HJ, Ryu HS, Chung SS, et al. Osteoconduction at porous hydroxyapatite with various pore configurations. Biomaterials 2000;21(12):1291–8. [24] Smartt Jr JM, Karmacharya J, Gannon FH, Ong G, Jackson O, Bartlett SP, et al. Repair of the immature and mature craniofacial skeleton with a carbonated calcium phosphate cement: assessment of biocompatibility, osteoconductivity, and remodeling capacity. Plast Reconstr Surg 2005;115(6):1642–50. [25] Schiroli G. Single-tooth implant restorations in the esthetic zone with PureForm ceramic crowns: 3 case reports. J Oral Implantol 2004; 30(6):358–63. [26] Govindaraj S, Costantino PD, Friedman CD. Current use of bone substitutes in maxillofacial surgery. Facial Plast Surg 1999;15(1): 73–81. [27] Li SH, De Wijn JR, Layrolle P, de Groot K. Synthesis of macroporous hydroxyapatite scaffolds for bone tissue engineering. J Biomed Mater Res 2002;61(1):109–20. [28] De Oliveira JF, De Aguiar PF, Rossi AM, Soares GA. Effect of process parameters on the characteristics of porous calcium phosphate ceramics for bone tissue scaffolds. Artif Organs 2003; 27(5):406–11. [29] Gauthier O, Bouler JM, Aguado E, Pilet P, Daculsi G. Macroporous biphasic calcium phosphate ceramics: influence of macropore diameter and macroporosity percentage on bone ingrowth. Biomaterials 1998;19(1–3):133–9. [30] Karageorgiou V, Kaplan D. Porosity of 3D biomaterial scaffolds and osteogenesis. Biomaterials 2005;26(27):5474–91. [31] Lu JX, Flautre B, Anselme K, Hardouin P, Gallur A, Descamps M, et al. Role of interconnections in porous bioceramics on bone recolonization in vitro and in vivo. J Mater Sci Mater Med 1999; 10(2):111–20. [32] Mankani MH, Kuznetsov SA, Fowler B, Kingman A, Robey PG. In vivo bone formation by human bone marrow stromal cells: effect of carrier particle size and shape. Biotechnol Bioeng 2001;72(1):96–107.
3237
[33] Navarro M, del Valle S, Martinez S, Zeppetelli S, Ambrosio L, Planell JA, et al. New macroporous calcium phosphate glass ceramic for guided bone regeneration. Biomaterials 2004;25(18): 4233–41. [34] Barou O, Mekraldi S, Vico L, Boivin G, Alexandre C, Lafage-Proust MH. Relationships between trabecular bone remodeling and bone vascularization: a quantitative study. Bone 2002;30(4):604–12. [35] Muraglia A, Martin I, Cancedda R, Quarto R. A nude mouse model for human bone formation in unloaded conditions. Bone 1998;22(5 Suppl):131S–4S. [36] Fabbri M, Celotti G, Ravaglioni A. Hydroxyapatite-based porous aggregates: physico-chemical nature, structure, texure and architecture. Biomaterials 1995;16:225–8. [37] Martinetti R, Belpassi A, Nataloni A, Piconi C. Porous hydroxyapatite cell carrier for tissue engineering. Key Eng Mater 2001;192–195:507–10. [38] Martinetti R, Dolcini L, Belpassi A, Quarto R, Mastrogiacomo M, Cancedda R, et al. Inspired porosity for cells and tissues. Key Eng Mater 2004;254–256:1095–8. [39] Martinetti R, Dolcini L. Mangano physico and chemical aspects of a new porous hydroxyapatite. Anal Bioanal Chem 2005;381: 634–8. [40] Martin I, Muraglia A, Campanile G, Cancedda R, Quarto R. Fibroblast growth factor-2 supports ex vivo expansion and maintenance of osteogenic precursors from human bone marrow. Endocrinology 1997;138(10):4456–62. [41] Goshima J, Goldberg VM, Caplan AI. Osteogenic potential of culture-expanded rat marrow cells as assayed in vivo with porous calcium phosphate ceramic. Biomaterials 1991;12(2):253–8. [42] Wilson CE, Dhert WJ, Van Blitterswijk CA, Verbout AJ, De Bruijn JD. Evaluating 3D bone tissue engineered constructs with different seeding densities using the alamarBlue assay and the effect on in vivo bone formation. J Mater Sci Mater Med 2002;13(12):1265–9. [43] Martin I, Mastrogiacomo M, De Leo G, Muraglia A, Beltrame F, Cancedda R, et al. Fluorescence microscopy imaging of bone for automated histomorphometry. Tissue Eng 2002;8(5):847–52. [44] Wendt D, Marsano A, Jakob M, Heberer M, Martin I. Oscillating perfusion of cell suspensions through three-dimensional scaffolds enhances cell seeding efficiency and uniformity. Biotechnol Bioeng 2003;84(2):205–14. [45] Bianco P, Kuznetsov SA, Riminucci M, Fisher LW, Spiegel AM, Robey PG. Reproduction of human fibrous dysplasia of bone in immunocompromised mice by transplanted mosaics of normal and Gsalpha-mutated skeletal progenitor cells. J Clin Invest 1998;101(8): 1737–44.
Biomaterials 27 (2006) 3230–3237 www.elsevier.com/locate/biomaterials
Role of scaffold internal structure on in vivo bone formation in macroporous calcium phosphate bioceramics Maddalena Mastrogiacomoa,b,1, Silvia Scaglionec,d,1, Roberta Martinettie, Laura Dolcinie, Francesco Beltramec,d, Ranieri Canceddaa,b, Rodolfo Quartod,f, a
Dip. di Oncologia, Biologia e Genetica, Universita` di Genova, Italy b Istituto Nazionale per la Ricerca sul Cancro, Genova, Italy c Dip. di Informatica, Sistemistica, Telematica, Universita` di Genova, Italy d Centro Biotecnologie Avanzate, Genova, Italy e FIN-CERAMICA FAENZA s.r.l., Faenza, Italy f Dip. di Chimica e Tecnologie Farmaceutiche ed Alimentari, Universita` di Genova, Italy Received 29 August 2005; accepted 27 January 2006 Available online 20 February 2006
Abstract Purpose of this study was the analysis of the role of density and pore interconnection pathway in scaffolds to be used as bone substitutes. We have considered 2 hydroxyapatite bioceramics with identical microstructure and different macro-porosity, pore size distribution and pore interconnection pathway. The scaffolds were obtained with two different procedures: (a) sponge matrix embedding (scaffold A), and (b) foaming (scaffold B). Bone ingrowth within the two bioceramics was obtained using an established model of in vivo bone formation by exogenously added osteoprogenitor cells. The histological analysis of specimens at different time after in vivo implantation revealed in both materials a significant extent of bone matrix deposition. Interestingly enough, scaffold B allowed a faster occurrence of bone tissue, reaching a steady state as soon as 4 weeks. Scaffold A on the other hand reached a comparable level of bone formation only after 8 weeks of in vivo implantation. Both scaffolds were well vascularised, but larger blood vessels were observed in scaffold A. Here we show that porosity and pore interconnection of osteoconductive scaffolds can influence the overall amount of bone deposition, the pattern of blood vessels invasion and finally the kinetics of the bone neoformation process. r 2006 Elsevier Ltd. All rights reserved. Keywords: Bone tissue engineering; Porous bioceramics; Interconnection pathway; Vascularisation
1. Introduction Bone regeneration in presence of synthetic bone graft can take a rather long time for the complete anatomical and functional recovery. With this in mind, scientists are focusing on higher performance biomaterials in bone tissue engineering to decrease the bone healing time. Many synthetic materials as bone substitute are easily available, including metals, natural and synthetic polymers, ceramics and glasses [1–22]. Calcium phosphate and hydroCorresponding author. Laboratorio Cellule Staminali, Centro Biotecnologie Avanzate, Largo Rosanna Benzi no. 10, 16132 Genova, Italy. Tel.: +39 010 5737 240; fax: +39 010 5737 505. E-mail address: [email protected] (R. Quarto). 1 Equally contributed to the experimental work.
0142-9612/$ - see front matter r 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.biomaterials.2006.01.031
xyapatite ceramics are considered among the most promising bone substitutes [1–4,6,8,10–12,19–22], because of their bonelike chemical composition and mechanical properties. Hydroxyapatite ceramics, in particular, are used in orthopaedics, maxillofacial and dental implant surgery, either alone or in combination with other substances or materials. Particulate form of hydroxyapatite are also used to fill bone defects or as a coating on metal implants [5,16,21,24–27]. Besides chemical composition, the other critical parameter to improve the efficiency of biomaterials to be used in bone tissue engineering is the overall structure: density, pore shape, pore size and pore interconnection pathway [23,28–33]. Porosity is necessary for the in vivo bone tissue ingrowth since it allow migration and proliferation of osteoblasts
ARTICLE IN PRESS M. Mastrogiacomo et al. / Biomaterials 27 (2006) 3230–3237
and mesenchymal cells, and matrix deposition in the empty spaces. Several studies investigated the minimum pore size required to regenerate mineralized bone [29,30,32]. Although macroporosity has a strong impact on osteogenic outcomes, interconnection pathway plays an important role as well. An incomplete pore interconnection could represent a constraint of the overall biological system, limiting blood vessels invasion. A proper blood supply represents the base for bone tissue growth in porous biomaterials [1,34]. Bone vascularisation, besides providing nutrients essential for tissue survival, plays also a crucial role in coordinating the activity of bone cells and their migration for bone remodelling [34]. Therefore, bioceramics with high porosity and appropriate interconnection pathway should allow the tissue to infiltrate and fill the scaffold. In the present work two different technologies were used to achieve hydroxyapatite porous devices with different total porosity and interconnection pathways. Both bioceramics were compared using an established model of in vivo bone formation by exogenously added osteoprogenitor cells [35]. The osteo-inductivity of the two ceramics scaffolds, the blood vessels invasion and the extent of bone matrix deposition was assessed through the histological analysis of specimens at different time after in vivo implantation. 2. Materials and methods 2.1. Biomaterial production and characterization Samples of porous bioceramics were prepared starting from the same batch of hydroxyapatite powder but employing two different production technologies. The first bioceramic (A) was prepared by sponge matrix embedding: a cellulose matrix with controlled pore morphology and distribution was embedded in a hydroxyapatite slurry to obtain the bimodal porous structures. After drying the green sample at room temperature, a pyrolysis of the organic phase was carried out and maintained in air atmosphere at 600 1C for 2 h. The sample was then sintered at 1280 1C for 1 h. The second bioceramic (B) was prepared with an innovative process (Patent pending) changing the surface energy of the slurry during processing (foaming method): a slurry with surfactants and high powder concentration (20 wt%) was used and expanded in a known volume (40–60 vol% of the total) to achieve a controlled morphology and an inner porosity close to 80 vol%; the bioceramic in this case was sintered at 1250 1C for 1 h. Both bioceramics were produced by FinCeramica Faenza (Italy). Physico-chemical and morphological characterisation of the final components were carried out for both sets of samples. Special attention was given to the characterisation of the total porosity and of micro-macro porosity distribution. The porosity study was carried out in conformity to the standard ASTM E562 using a image analyser (Leica Imaging System Ltd. Q500MW by Qphase Application, Cambridge, UK). The morphology of the scaffolds was analysed by scanning electron microscopy (SEM) to evaluate micro and macro structure (Leica, Cambridge, UK): for SEM observation the samples were dehydrated, firmly mounted, and coated with a thin layer of a conductive material. For coating we took advantage of a small device called sputter coater which uses argon gas and a small electric field. The sample was placed in a small chamber under vacuum. Argon gas was then introduced and an electric field was used to cause an electron to be removed from the argon atoms making the atoms ions positively charged. Phases investigation and chemical analyses were carried out, respectively by X-ray diffraction
3231
analysis (CuKa radiation, Rigaku Miniflex, Tokyo, Japan) and by inductively coupled plasma-ICP (Spectroflame Modula, Spectro, equipped with Ultrasonic Nebulizer Cetac U-6000 AT+, Kleve, Germany). The surface area was carried out following B.E.T. method and measured on a Quantasorb instrument (Quantachrone Instrument Greenvale, Ny, USA) while the density was determine as mass/volume [36–39].
2.2. Osteoprogenitor cell culture and ectopic bone formation assay in ID mice Four 10-months-old female sheep of Italian Biellese strain, as bone marrow donors, were involved in this study with approval of competent ethical committee and legal authorities. Twenty millilitre of bone marrow aspirate were harvested from the posterior iliac crest of each animal. Bone marrow samples were washed twice with PBS (phosphate buffered saline). Marrow specimens were stained with a nuclear stain (0.1% methyl violet in 0.1 M citric acid) and the nucleated fraction was counted. Cells were suspended in Coon’s modified Ham’s F12 medium supplemented with 10% FCS, 100 I.U./ml penicillin and 100 mg/ml streptomycin and plated at a density of 1 106 cells/cm2. Human recombinant fibroblast growth factor 2 (FGF2) (1 ng/ml) was added since the beginning of the culture [40]. Medium was changed 2 days after the original plating and then twice a week. When culture dishes were nearly confluent (passage 0), bone marrow stromal cells (BMSC) were detached with 0.05% trypsin-0.01% EDTA and 5 105 cells were replated in 100-mm dishes (passage 1) until the next confluence. A BMSC pool obtained from four sheeps was used for all the experiments described. 2.5 106 cells were resuspended into 40 ml of 20 mg/ml fibrinogen solution in PBS. 40 ml of cell suspension in fibrinogen were combined with each cube of biomaterial (3 3 3 mm) and 25 NIH U of thrombin were added. The polymerization reaction was allowed to proceed at 37 1C for 10 min, then bioceramic/BMSC composites were subcutaneously implanted in immuno-deficient (ID) (CD-1 nu/nu) mice by using an established model of ectopic bone formation [35]. In these conditions, BMSC generate bone tissue in a fashion proportional to the number of cells implanted [41–42]. In agreement and with the approval of the competent ethical committee and legal authorities, recipient ID mice of 1 month of age, purchased from Charles River Italia, were kept in a controlled environment and given free access to food and water. Mice were anesthetized by intramuscular injection of Xilazine (20 mg/ml) and Ketamine (30 mg/ml). Bioceramic/ BMSC composites were implanted subcutaneously on the back of the mice (up to 4 implants for each animal, for an amount of 6 animal). Animals were sacrificed 2, 4, and 8 weeks after implantation. Grafts were harvested and processed for histological analysis.
2.3. Optical microscope analysis Samples were decalcified, paraffin embedded, sectioned, stained with hematoxylin/eosin and analyzed for the bone tissue quantification. The amount of newly formed bone was assessed using a Zeiss Axiophot optical fluorescence microscope (Oberkochen, Germany) and quantified using the free image analysis software Scion Image and an image processing macro previously developed [43]. For each ceramics type, twelve samples (4 at each time point) were harvested after 2, 4 and 8 weeks of implant. For each construct, ten sections of 5 mm thick cut at different levels (100 mm gap level) were HE stained for newly formed bone quantification. All images were acquired using both transmitted and fluorescence light. For each construct, the amount of bone was assessed as percentage of the total bone tissue versus total tissue formed [43]. The pores interconnection size of the two different scaffolds was measured using both SEM images and histological sections. In scaffold A interconnection path was mostly represented by ‘‘tunnels’’ (T) with a low percentage of smaller interconnections, generated by contacts (C) of adjacent tunnel walls and pores. Therefore, the limiting size of the tunnels interconnecting adjacent pores and mean diameters of smaller inter-
ARTICLE IN PRESS 3232
M. Mastrogiacomo et al. / Biomaterials 27 (2006) 3230–3237
connections were measured. For scaffold B, mean diameters of contact points of adjacent pores (C), the only interconnections of this material, were measured. For both biomaterials, eight images were analysed, all interconnection values ordered according to their size, and their distribution was plotted (average values7std dev). For each scaffolds, mean blood vessels size was obtained by using four representative histological images at 2.5 magnification. For each blood vessel section only the minor axis was considered. Average values7std dev were derived.
2.4. Statistical analysis Mean values and standard deviations were obtained analysing raw data with Microsoft Excel software. Correlation analysis was performed using the parametric Student t-test. Statistical differences were considered significant when Po0:05.
3. Results 3.1. Bioceramic characterisation The bioceramics obtained with 2 different production methods were characterised. Both porous hydroxyapatite
bioceramics were complying with the ISO 13779 standard. According to X-ray diffraction analysis they resulted to be single phase crystalline hydroxyapatite, with purity X95% according to ICCD card no. 9-432; Ca/P ratio had a value of 1.6570.02 and trace elements (total heavy metals) resulted to be under the maximum allowable limit (50 mg/kg). Both scaffolds do not display any significant resorption within 6 months of in vivo implantation, as previously shown [1]. The density determined as mass/volume of the samples analysed was found to be 1.26 g/cm3 (70.16) for bioceramic A and 0.63 g/cm3 (70.10) for bioceramic B. These values were compatible with a total porosity of 6575% (A) and 8073% (B). Surface area was measured for both ceramics: in scaffold A an average of 0.8770.04 m2/g was observed, despite the 1.6370.08 m2/g of surface area for the scaffold B. The two bioceramic structures are shown in Fig. 1. At low magnification (panels A and B) were evident for the differences in pore size, shape and interconnection path-
Fig. 1. Scanning electron microscope (SEM) analysis at low magnification shows the macro structure of the two different materials used as osteoconductive grafts. The morphology of sponge matrix (scaffold type A) and of the foam sample (scaffold type B) are shown in panels A and B, respectively. Biomaterials show a different porosity, in terms of pores shape, size and interconnection. Black spots represent interconnections between neighbour pores. SEM analysis at higher magnification of scaffold A and scaffold B is also shown (C and D, respectively). Biomaterials display a very similar microstructure and micro-porosity, according to the same size of grains used. Bars: A ¼ 1000 mm, B ¼ 200 mm; C ¼ 5 mm; D ¼ 2 mm. A schematic view of the two scaffold’s structures is shown: scaffold type A is obtained starting from a cellulose matrix (black network in E), whose trabeculae are covered by the HA slurry (grey part) thus generating a porous structure (white area representing the final pores). On the other hand, structure of scaffold type B (F) may be seen as multiple interconnected gas bubbles (white area) within a HA matrix (grey area).
ARTICLE IN PRESS M. Mastrogiacomo et al. / Biomaterials 27 (2006) 3230–3237
ways between the two scaffolds. At higher magnification (panels C and D) the microstructures of the scaffolds were displayed, in particular the hydroxyapatite grain shape and size, the intergranular porosity and the micro-interconnections. For both scaffolds the porosity distribution was found to be bi-modal and both scaffolds were formed by interconnected micro-macropores, where micropores represent the empty inter-grain space, while macropores are the macro-cavities obtained during the ceramic processing (Fig. 1). Grain size resulted to be in the range 0.5–2.5 mm and the size of micro/intergranular pores was in the range of 0.2–10 mm for both materials (Fig. 1(C–D)). An important difference between the two scaffolds was observed. Given the two production methods used, the pore interconnection pathway in bioceramic A was mostly represented by ‘‘tunnels’’ and therefore corresponded to the macro-pore size itself with a low percentage of smaller interconnections between adjacent tunnel walls (Fig. 1(E)), whereas in bioceramic B the interconnection pathway was mainly due to confluence of adjacent pore walls (Fig. 1(F)).
3233
Therefore, based on the analysis performed on SEM images of the raw materials (Fig. 2(C and E)), in scaffold A most of the interconnections were found to be above 200 mm, whereas in scaffold B about 70% of the interconnections displayed a size below 100 mm with only a bare 3% above 200 mm (Fig. 2(A)). The analysis performed on histological sections (Fig. 2(D and F)) confirmed with minor variations, possibly due to different treatments of the samples, the results obtained on SEM images (Fig. 2(B)). 3.2. Ectopic bone formation in ID mice Equal numbers of BMSC were loaded onto scaffold cubes and subcutaneously implanted in ID mice. Samples were harvested after 2, 4 and 8 weeks of in vivo implant and histologically analysed. Neo-bone formation was already present after 2 weeks of implant in the peripheral area of scaffold type B, whereas no trace of bone tissue was detected at same time in the less porous bioceramic A (data not shown).
Fig. 2. The distribution of interconnection pore size of the two different scaffolds was plotted, either using SEM images of scaffolds before their use (A) or using histological sections of samples after 8 weeks in vivo (B). In scaffold A, most of the interconnections, represented by ‘‘tunnels’’ (T) and corresponding to the macro-pore, were found having a size above 200 mm, whereas in scaffold B about 70% of the interconnections displayed a size below 100 mm, corresponding to the contact (C) between adjacent pores, with almost no pore interconnections above 200 mm. Arrows indicate the measurement procedure, which was carried out using both SEM images of the scaffolds type A and B (C and E, respectively) and histological sections (D and F, respectively).
ARTICLE IN PRESS 3234
M. Mastrogiacomo et al. / Biomaterials 27 (2006) 3230–3237
After 4 weeks of in vivo implantation, scaffold A displayed a low amount of bone tissue deposition (Fig. 3(A)), whereas in scaffold B, a significant amount of bone tissue was evenly distributed within pores (Fig. 3(B)). After 8 weeks, a significant improvement of bone formation was observed in scaffold A (Fig. 3(C)), whereas no major changes were observed in the extent of bone formed in scaffold B (Fig. 3(D)). At higher magnification, representative histological sections of the two ceramics displayed a good bone matrix deposition filling the pores from the periphery to the centre, while the interconnections of adjacent pores were filled by undifferentiated vascularised mesenchymal tissue (Fig. 3(E and F)). The amount of bone matrix deposition within the implant was quantified. Histological analysis and computer-assisted quantification of the bone/total tissue formed revealed that the bone formation within scaffold B pores had already reached after 4 weeks the 35% of the total
tissue neo-formed, value close to that found in the twin set of samples analysed after 8 weeks (36%). In bioceramic A, on the other hand, by 4 weeks newly formed bone was only 22% of the total tissue and a 37% of bone tissue formation was obtained only after 8 weeks of in vivo implantation. In both materials, a significant amount of blood vessel was observed within the bioceramic pores, as well as between contiguous pores running through the interconnections. In scaffold A though a higher number of large blood vessels were found (mean diameter: 91738 mm) whereas in scaffold B, although the vascularisation was well represented, we never observed large blood vessel running through several consecutive pores (mean diameter: 61718 mm). Fig. 4 displays a representative histological field of a scaffold A tunnel run by a large blood vessel and a comparable field of scaffold B where a smaller blood vessel runs through an interconnection joining two adjacent pores (Fig. 4(A) and (B), respectively).
Fig. 3. Histology of tissues formed in ectopic transplants of sheep BMSC loaded on two different biomaterials (scaffold A: A–C–E; scaffold B: B–D–F) in immunodeficient mice after 4 and 8 weeks (A–B and C–D, respectively). A deposition of bone matrix from the edge of ceramic pore toward the centre is observed. After 4 weeks of implantation, scaffold A displayed a low amount of bone tissue deposition (A), whereas in scaffold B a relevant amount of neobone tissue deposition, stained in pink, was evenly distributed within pores (B). After 8 weeks, a significant enhancement of bone formation was observed in scaffold A (C), whereas no major changes were evident in the scaffold B (D). A good interconnection of pores and vascularisation is also shown in both ceramics. Panel E and F: higher magnification of comparable histological sections. Staining: hematoxylin-eosin. Bars: A2D ¼ 1000 mm, E2F ¼ 250 mm.
ARTICLE IN PRESS M. Mastrogiacomo et al. / Biomaterials 27 (2006) 3230–3237
3235
Fig. 4. Blood vessel running through pore interconnections: (A) (scaffold A) a representative large blood vessel is displayed; (B) a smaller blood vessel running through an interconnection joining two adjacent pores of the scaffold B. Bar: 100 mm. B ¼ bone tissue; L ¼ blood vessel lumen; RBC ¼ red blood cells.
4. Discussion A number of bone substitute biomaterials are easily available. Among modern biomaterials, bioceramics (calcium phosphate and hydroxyapatite) are very promising candidates in bone substitution, because of their bone-like chemical composition and mechanical properties [1–4,6,8,10–12,19–22]. A modern bioceramic to be used as a bone substitute should be highly efficient in terms of bone ingrowth to allow a faster regeneration and a better functional recovery. Besides its chemical composition, one of the most important parameters in a three-dimensional scaffold is its internal structure. Several reports have been aimed at enlightening relationships among different parameters (density, pore shape, pore size and pore interconnection pathway) of bone substitutes biomaterials and bone ingrowth [23,28–33]. Although pore size, extent of interconnection and interconnection pathway are critical factors for bone formation, it is still not completely clear how these parameters are able to influence the efficiency of the bone neo-formation process [30]. In the present study we investigated how different internal architecture of porous bioceramics would affect the total amount of bone neo-formation. The two materials we have considered had the same chemical composition (stoichiometric hydroxyapatite) and micro-structure, but different fabrication procedures and therefore different architectures in terms of macro-porosity, pore size distribution and interconnection pathway. To verify bone formation occurrence in the biomaterials under study, we used an established model of ectopic bone formation [35]. In such an experimental model, BMSC were loaded onto the biomaterials and subcutaneously implanted in immunodeficient mice. In these conditions, BMSC generate bone tissue in a fashion proportional to the number of cells implanted. In our experiments, bioceramic samples were loaded with the same amount of sheep BMSC, which, considering the densities and the overall porosity of the 2 materials, was an advantage for bioceramic A with a higher cell density for surface unit. Still, the histological analysis of bone tissue formed after 2
weeks already revealed that bone had been evidently deposited only in the less dense bioceramic (scaffold B), whereas undifferentiated mesenchymal tissue was found in the corresponding samples of the more dense material (scaffold A) despite the estimated higher number of BMSC/pore seeded. The analysis of the 4-week samples revealed that within scaffold B pores bone formation level had already reached a significant 35%, value which was very close to that found in the twin samples of foaming HA after 8 weeks in vivo. Type A bioceramic by 4 weeks was also colonised by newly formed bone but at a lower extent (22%) and had gradual increase the level of bone formation reaching a 37% only after 8 weeks of in vivo implantation. These data indicate that bone formation had occurred more rapidly and more efficiently in the less dense material (scaffold B) despite the estimated lower number of cells loaded for surface unit. It would be nice to speculate that the higher in vivo bone forming efficiency of scaffold B at earlier time points could be due to a ‘‘filter effect’’ of this scaffold, according to its pore interconnection size distribution. Scaffold B is less dense, but with smaller pore and smaller interconnection size, therefore it could generate a higher cell density in the outer zone of the scaffold, which in turn would possibly promote an earlier and more vigorous bone formation. This speculation is in agreement with evidences reported by Wendt and coworkers who have shown a non-homogeneous cell distribution, leading to the formation of a cell density gradient, in comparable scaffolds when cells are loaded under similar conditions [44]. On the other hand, bone formation rate in scaffold B, already high after 4 weeks, remained almost unchanged up to 8 weeks. This evidence suggests that bone neo-formation is proportional to the surface available for tissue growth and may also suggest that bone neo-formation is somehow limited by factors intrinsic to the material itself (i.e. resorbability) and possibly to the physiological equilibrium reached by the ‘‘bone organoid’’ (bone matrix, blood vessels, marrow and undifferentiated mesenchymal cells which all together constitute the bone organoid) [45]. Pores filled with newly formed bone in fact, always present a peripheral area of bone tissue and a central area of undifferentiated vascularised mesenchymal tissue. It would
ARTICLE IN PRESS 3236
M. Mastrogiacomo et al. / Biomaterials 27 (2006) 3230–3237
be nice to postulate that these four components (bone matrix, blood vessels, marrow and undifferentiated mesenchymal cells) coexist in a kind of steric but dynamic equilibrium, which would mean, in terms of bone formation, reaching a saturation level. Such level could be overpassed only using more resorbable bioceramics (i.e. tricalcium phosphate, carbonate hydroxyapatite, magnesium-doped hydroxyapatite, etc.), which would allow bone to grow in a fashion proportional to biomaterial resorption. From the histological analysis of the two scaffolds it was also evident that the level of vascularisation was abundant in both materials and virtually all pores were reached by blood vessels of various size. An interesting observation was that, according to the pore interconnection size distribution, in scaffold B, the interconnections appeared to be limiting the size of blood vessels running through adjacent pores. In other words, the interconnection pathway could limit the vascularisation (at least in terms of vessel maximal diameter) of bioceramics representing a bottleneck for blood vessel invasion. This physical constraint could have important implications with larger ceramic specimens where larger vessel are indeed required.
[2]
[3]
[4]
[5]
[6]
[7]
[8]
5. Conclusion [9]
All together our data indicate that in osteoconductive bioceramics, both pore size and pore interconnection pathway are critical factors. Pore size would be directly related to bone formation, since it provides surface and space for cell adhesion and bone ingrowth; pore interconnection on the other hand would provide the way for cell distribution/migration and allow an efficient in vivo blood vessel formation suitable for sustaining bone tissue neo-formation and possibly remodelling. The results obtained in the present study show that a significant improvement in bone repair can be achieved using novel technologies of bioceramic production. Our observations also suggest that scaffolds characterised by an engineered structure with larger interconnections and higher total available surface area seem to be the right direction to investigate for future bone scaffold improvements. Acknowledgements We are grateful to Mr. G. Martignani and L. Pilotti (ENEA, Faenza) for their contribution to the Image Analyzer study and SEM investigation. This work supported with financial support from the European Commission under the Growth Programme GRD1-1999-10590.
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17] [18]
References [19] [1] Boyde A, Corsi A, Quarto R, Cancedda R, Bianco P. Osteoconduction in large macroporous hydroxyapatite ceramic implants: evidence
for a complementary integration and disintegration mechanism. Bone 1999;24(6):579–89. Dong J, Kojima H, Uemura T, Kikuchi M, Tateishi T, Tanaka J. In vivo evaluation of a novel porous hydroxyapatite to sustain osteogenesis of transplanted bone marrow-derived osteoblastic cells. J Biomed Mater Res 2001;57(2):208–16. Dong J, Uemura T, Shirasaki Y, Tateishi T. Promotion of bone formation using highly pure porous beta-TCP combined with bone marrow-derived osteoprogenitor cells. Biomaterials 2002;23(23): 4493–502. Flautre B, Anselme K, Delecourt C, Lu J, Hardouin P, Descamps M. Histological aspects in bone regeneration of an association with porous hydroxyapatite and bone marrow cells. J Mater Sci Mater Med 1999;10(12):811–4. Fujimura K, Bessho K, Okubo Y, Segami N, Iizuka T. A bioactive bone cement containing Bis-GMA resin and A-W glass-ceramic as an augmentation graft material on mandibular bone. Clin Oral Implants Res 2003;14(5):659–67. Gauthier O, Muller R, von Stechow D, Lamy B, Weiss P, Bouler JM, et al. In vivo bone regeneration with injectable calcium phosphate biomaterial: a three-dimensional micro-computed tomographic, biomechanical and SEM study. Biomaterials 2005;26(27):5444–53. Kim H, Suh H, Jo SA, Kim HW, Lee JM, Kim EH, et al. In vivo bone formation by human marrow stromal cells in biodegradable scaffolds that release dexamethasone and ascorbate-2-phosphate. Biochem Biophys Res Commun 2005;332(4):1053–60. Kon E, Muraglia A, Corsi A, Bianco P, Marcacci M, Martin I, et al. Autologous bone marrow stromal cells loaded onto porous hydroxyapatite ceramic accelerate bone repair in critical-size defects of sheep long bones. J Biomed Mater Res 2000;49(3):328–37. Livingston T, Ducheyne P, Garino J. In vivo evaluation of a bioactive scaffold for bone tissue engineering. J Biomed Mater Res 2002;62(1): 1–13. Lu JX, Gallur A, Flautre B, Anselme K, Descamps M, Thierry B, et al. Comparative study of tissue reactions to calcium phosphate ceramics among cancellous, cortical, and medullar bone sites in rabbits. J Biomed Mater Res 1998;42(3):357–67. Livingston TL, Gordon S, Archambault M, Kadiyala S, McIntosh K, Smith A, et al. Mesenchymal stem cells combined with biphasic calcium phosphate ceramics promote bone regeneration. J Mater Sci Mater Med 2003;14(3):211–8. Marcacci M, Kon E, Zaffagnini S, Vascellari A, Neri MP, Iacono F. New cell-based technologies in bone and cartilage tissue engineering. I. Bone reconstruction. Chir Organi Mov 2003;88(1):33–42. Mauney JR, Blumberg J, Pirun M, Volloch V, Vunjak-Novakovic G, Kaplan DL. Osteogenic differentiation of human bone marrow stromal cells on partially demineralized bone scaffolds in vitro. Tissue Eng 2004;10(1-2):81–92. Mauney JR, Jaquiery C, Volloch V, Heberer M, Martin I, Kaplan DL. In vitro and in vivo evaluation of differentially demineralized cancellous bone scaffolds combined with human bone marrow stromal cells for tissue engineering. Biomaterials 2005;26(16): 3173–85. Meinel L, Karageorgiou V, Hofmann S, Fajardo R, Snyder B, Li C, et al. Engineering bone-like tissue in vitro using human bone marrow stem cells and silk scaffolds. J Biomed Mater Res A 2004;71(1):25–34. Moreira-Gonzalez A, Lobocki C, Barakat K, Andrus L, Bradford M, Gilsdorf M, et al. Evaluation of 45S5 bioactive glass combined as a bone substitute in the reconstruction of critical size calvarial defects in rabbits. J Craniofac Surg 2005;16(1):63–70. Nazarov R, Jin HJ, Kaplan DL. Porous 3-D scaffolds from regenerated silk fibroin. Biomacromolecules 2004;5(3):718–26. Sul YT, Johansson C, Byon E, Albrektsson T. The bone response of oxidized bioactive and non-bioactive titanium implants. Biomaterials 2005;26(33):6720–30. Tampieri A, Celotti G, Sprio S, Delcogliano A, Franzese S. Porositygraded hydroxyapatite ceramics to replace natural bone. Biomaterials 2001;22(11):1365–70.
ARTICLE IN PRESS M. Mastrogiacomo et al. / Biomaterials 27 (2006) 3230–3237 [20] Turhani D, Cvikl B, Watzinger E, Weissenbock M, Yerit K, Thurnher D, et al. In vitro growth and differentiation of osteoblast-like cells on hydroxyapatite ceramic granule calcified from red algae. J Oral Maxillofac Surg 2005;63(6):793–9. [21] Marcacci M, Kon E, Zaffagnini S, Giardino R, Rocca M, Corsi A, et al. Reconstruction of extensive long-bone defects in sheep using porous hydroxyapatite sponges. Calcif Tissue Int 1999;64(1):83–90. [22] Quarto R, Mastrogiacomo M, Cancedda R, Kutepov SM, Mukhachev V, Lavroukov A, et al. Repair of large bone defects with the use of autologous bone marrow stromal cells. N Engl J Med 2001;344(5):385–6. [23] Chang BS, Lee CK, Hong KS, Youn HJ, Ryu HS, Chung SS, et al. Osteoconduction at porous hydroxyapatite with various pore configurations. Biomaterials 2000;21(12):1291–8. [24] Smartt Jr JM, Karmacharya J, Gannon FH, Ong G, Jackson O, Bartlett SP, et al. Repair of the immature and mature craniofacial skeleton with a carbonated calcium phosphate cement: assessment of biocompatibility, osteoconductivity, and remodeling capacity. Plast Reconstr Surg 2005;115(6):1642–50. [25] Schiroli G. Single-tooth implant restorations in the esthetic zone with PureForm ceramic crowns: 3 case reports. J Oral Implantol 2004; 30(6):358–63. [26] Govindaraj S, Costantino PD, Friedman CD. Current use of bone substitutes in maxillofacial surgery. Facial Plast Surg 1999;15(1): 73–81. [27] Li SH, De Wijn JR, Layrolle P, de Groot K. Synthesis of macroporous hydroxyapatite scaffolds for bone tissue engineering. J Biomed Mater Res 2002;61(1):109–20. [28] De Oliveira JF, De Aguiar PF, Rossi AM, Soares GA. Effect of process parameters on the characteristics of porous calcium phosphate ceramics for bone tissue scaffolds. Artif Organs 2003; 27(5):406–11. [29] Gauthier O, Bouler JM, Aguado E, Pilet P, Daculsi G. Macroporous biphasic calcium phosphate ceramics: influence of macropore diameter and macroporosity percentage on bone ingrowth. Biomaterials 1998;19(1–3):133–9. [30] Karageorgiou V, Kaplan D. Porosity of 3D biomaterial scaffolds and osteogenesis. Biomaterials 2005;26(27):5474–91. [31] Lu JX, Flautre B, Anselme K, Hardouin P, Gallur A, Descamps M, et al. Role of interconnections in porous bioceramics on bone recolonization in vitro and in vivo. J Mater Sci Mater Med 1999; 10(2):111–20. [32] Mankani MH, Kuznetsov SA, Fowler B, Kingman A, Robey PG. In vivo bone formation by human bone marrow stromal cells: effect of carrier particle size and shape. Biotechnol Bioeng 2001;72(1):96–107.
3237
[33] Navarro M, del Valle S, Martinez S, Zeppetelli S, Ambrosio L, Planell JA, et al. New macroporous calcium phosphate glass ceramic for guided bone regeneration. Biomaterials 2004;25(18): 4233–41. [34] Barou O, Mekraldi S, Vico L, Boivin G, Alexandre C, Lafage-Proust MH. Relationships between trabecular bone remodeling and bone vascularization: a quantitative study. Bone 2002;30(4):604–12. [35] Muraglia A, Martin I, Cancedda R, Quarto R. A nude mouse model for human bone formation in unloaded conditions. Bone 1998;22(5 Suppl):131S–4S. [36] Fabbri M, Celotti G, Ravaglioni A. Hydroxyapatite-based porous aggregates: physico-chemical nature, structure, texure and architecture. Biomaterials 1995;16:225–8. [37] Martinetti R, Belpassi A, Nataloni A, Piconi C. Porous hydroxyapatite cell carrier for tissue engineering. Key Eng Mater 2001;192–195:507–10. [38] Martinetti R, Dolcini L, Belpassi A, Quarto R, Mastrogiacomo M, Cancedda R, et al. Inspired porosity for cells and tissues. Key Eng Mater 2004;254–256:1095–8. [39] Martinetti R, Dolcini L. Mangano physico and chemical aspects of a new porous hydroxyapatite. Anal Bioanal Chem 2005;381: 634–8. [40] Martin I, Muraglia A, Campanile G, Cancedda R, Quarto R. Fibroblast growth factor-2 supports ex vivo expansion and maintenance of osteogenic precursors from human bone marrow. Endocrinology 1997;138(10):4456–62. [41] Goshima J, Goldberg VM, Caplan AI. Osteogenic potential of culture-expanded rat marrow cells as assayed in vivo with porous calcium phosphate ceramic. Biomaterials 1991;12(2):253–8. [42] Wilson CE, Dhert WJ, Van Blitterswijk CA, Verbout AJ, De Bruijn JD. Evaluating 3D bone tissue engineered constructs with different seeding densities using the alamarBlue assay and the effect on in vivo bone formation. J Mater Sci Mater Med 2002;13(12):1265–9. [43] Martin I, Mastrogiacomo M, De Leo G, Muraglia A, Beltrame F, Cancedda R, et al. Fluorescence microscopy imaging of bone for automated histomorphometry. Tissue Eng 2002;8(5):847–52. [44] Wendt D, Marsano A, Jakob M, Heberer M, Martin I. Oscillating perfusion of cell suspensions through three-dimensional scaffolds enhances cell seeding efficiency and uniformity. Biotechnol Bioeng 2003;84(2):205–14. [45] Bianco P, Kuznetsov SA, Riminucci M, Fisher LW, Spiegel AM, Robey PG. Reproduction of human fibrous dysplasia of bone in immunocompromised mice by transplanted mosaics of normal and Gsalpha-mutated skeletal progenitor cells. J Clin Invest 1998;101(8): 1737–44.