• Sonuç bulunamadı

An in vivo study on the effect of scaffold geometry and growth factor release on the healing of bone defects

N/A
N/A
Protected

Academic year: 2021

Share "An in vivo study on the effect of scaffold geometry and growth factor release on the healing of bone defects"

Copied!
10
0
0

Yükleniyor.... (view fulltext now)

Tam metin

(1)

An

in vivo study on the effect of scaffold geometry and

growth factor release on the healing of bone defects

P. Yilgor

1,8

, G. Yilmaz

2

, M. B. Onal

3

, I. Solmaz

3

, S. Gundogdu

4

, S. Keskil

5

, R. A. Sousa

6

, R. L. Reis

6

,

N. Hasirci

1,7,8

and V. Hasirci

1,8,9

*

1

Department of Biotechnology, Middle East Technical University, Ankara, Turkey

2

Department of Pathology, Gazi University Faculty of Medicine, Ankara, Turkey

3

Department of Neurosurgery, Gulhane Military Medical Academy, Ankara, Turkey

4

Department of Radiology, Ufuk University Faculty of Medicine, Ankara, Turkey

5

Department of Neurosurgery, Bayindir Medical Centre, Kavaklidere, Ankara, Turkey

6

3Bs Research Group, University of Minho, Headquarters of the European Institute of Excellence on Tissue Engineering and Regenerative Medicine, Guimarães, Portugal

7

Department of Chemistry, Middle East Technical University, Ankara, Turkey

8

BIOMATEN Centre of Excellence in Biomaterials and Tissue Engineering, Ankara, Turkey

9

Department of Biological Sciences, Biotechnology Research Unit, Middle East Technical University, Ankara, Turkey

Abstract

The hypothesis of this study was that the extent of bone regeneration could be enhanced by using scaffolds with appropriate geometry, and that such an effect could be further increased by mimicking the natural timing of appearance of bone morphogenetic proteins BMP-2 and BMP-7 after fracture. Bioplotted poly(«-caprolactone) (PCL) disks with four different fibre organizations were used to study the effect of 3D scaffold architecture on the healing of bone defects in a rat pelvis model. Moreover, one PCL construct was further modified by introducing a nanoparticulate sequential BMP-2/BMP-7 delivery system into this scaffold. Scaffolds and functionalized construct along with free nanocapsules were implanted using a rat iliac crest defect model. Six weeks post-implantation, the defects were evaluated by CT scan and histology. Analysis revealed that the basic architecture, having the highest pore volume for tissue ingrowth, presented the highest bone formation as determined by the bone mineral density (BMD) within the defect (144.2 7.1); about four-fold higher than that of the empty defect (34.9 10.7). It also showed the highest histological analysis scores with a high amount of bone forma-tion within the defect, within the scaffold pores and along the outer surfaces of the scaffold. The basic scaffold carrying the BMP-2/BMP-7 delivery system showed significantly higher bone formation than the growth factor-free basic scaffold at 6 weeks (BMD 206.8 15.7). Histological analysis also revealed new bone formation in close to or in direct contact with the construct interface. This study indicates the importance of open and interconnecting pore geometry on the better healing of bone defects, and that this effect could be further increased by supplying growth factors, as is the case in nature. Copyright © 2012 John Wiley & Sons, Ltd.

Received 8 April 2011; Revised 1 November 2011; Accepted 24 November 2011

Keywords bone regeneration; scaffold; bone morphogenetic protein; controlled release; 3D plotting

1. Introduction

Impaired healings and non-unions account for 5–10% of all fractures on an annual basis (Navarro et al., 2008)

and necessitate development of proper bonefillers that in-duce regeneration while providing structural and me-chanical support. Tissue engineering is considered to be a promisingfield for the production of such artificial bone substitutes that would eventually replace bone grafting. It involves the use of cells together with biodegradable scaf-folds that structurally and mechanically mimic the defect structure and composition. Therefore, scaffolds play a central role in tissue engineering applications because

*Correspondence to: V. Hasirci, BIOMATEN, Middle East Technical University, Department of Biological Sciences, Biotechnology Research Unit, 06800 Ankara, Turkey. E mail: vhasirci@metu.edu.tr

(2)

they support and guide regeneration and define the ulti-mate shape. Along with the need for adaptability to the defect site, architecture of the scaffolds is of utmost importance, because the structural design defines the geometry of cell adhesion sites and thus influences cell adhesion, spreading and orientation (Benya and Schaffer, 1982). Another important property in scaffold design is the extent and interconnectivity of the porosity, which facilitates the population of the scaffolds by the cells and enables nutrient and oxygen transfer, as well as aiding bone ingrowth and vascularization. Several techniques are being employed in the production of porous, three-dimensional (3D) structures (Mohajeri et al., 2010; Cao et al., 2009; Se et al., 2006; Tuzlakoglu et al., 2005; Almir-all et al., 2004); however, most lack the ability to produce structures with predefined form and properties. Rapid prototyping (RP), on the other hand, can create a scaffold, directly from computed tomography (CT) or magnetic res-onance imaging scans of the damaged region, which has a distinctly defined porosity in a specific 3D shape to fit into an irregular defect site to provide a structurally and me-chanically perfect fit (Moroni et al., 2006; Yeong et al., 2004; Lam et al., 2002). With the capability to create scaf-folds with defined composition and architecture, RP is a good method to investigate the effect of scaffold proper-ties such as geometry on cell behaviour for further optimi-zation of the scaffold design for a given application.

Poly(e-caprolactone) (PCL) is a suitable material for use in long-term, load-bearing applications due to its low degradation rate, resulting from its hydrophobicity and crystallinity (Pitt, 1990). PCL scaffolds have been created with a variety of RP techniques, including fused deposition modelling (Rohner et al., 2003), shape deposi-tion modelling (Marra et al., 1999), selective laser sinter-ing (Williams et al., 2005), low-temperature deposition (Xiong et al., 2002) and multi-nozzle free-form deposition (Sun et al., 2004), owing to its ease of processing and high decomposition temperature. In these systems, the cells were observed to adhere and start growing on the PCL scaffolds both in vitro and in vivo.

Growth factors are regulators of cellular activities and their contribution is critical in tissue engineering applica-tions and regenerative medicine. Conventionally, osteoin-ductive growth factors are introduced directly to the culture medium in vitro or injected to the defect site in in vivo applications. However, the highly labile growth fac-tors are susceptible to degradation and generally are rapidly cleared from the defect site. For this reason, their duration of action is generally limited. In order to avoid this, a new approach was employed that involved encapsulation of growth factors in carriers to protect their activity, localize and prolong their action at the defect site as well as to control their presence and maintain their concentration.

Bone morphogenetic proteins (BMPs) are the most osteo-genic members of the transforming growth factor-b (TGFb) superfamily proteins, and they promote the formation of cartilage and bone by inducing mesenchymal stem cells (MSCs) toward chondroblastic and osteoblastic differentia-tion, also causing them to proliferate in vivo (Urist, 1965).

Several of them, BMP-2–BMP-18, were identified in humans, and of these BMP-2, -4, -6, -7 and 9 are known to induce complete bone morphogenesis and are considered to be the most osteogenic ones (Bessa et al., 2008), where BMP-2 and 7 possess strong ability to induce bone forma-tion and are the only ones that have received FDA approval for use within collagen carriers in applications for spinal fu-sion and sinus lift (White et al., 2007; McKay et al., 2007).

In our earlier studies, PCL scaffolds produced by 3D plotting were characterized in situ, and further in vitro analysis revealed that 3D organization of fibres consti-tuting the scaffolds is influential on the attachment, prolif-eration and differentiation of MSCs to the osteoblastic lineage (Yilgor et al., 2008). In addition, BMP-2 and BMP-7 encapsulated in different populations of nanocap-sules were loaded onto these scaffolds and their sequential delivery (first BMP-2, then BMP-7) was achieved as in the natural bone regeneration process (Yilgor et al., 2010a). The positive effect of co-administration of BMP-2 and BMP-7 in a sequential fashion was illustrated as enhanced MSC differentiation.

Enhanced bone regeneration results obtained in in vitro encouraged us to further test the system under in vivo conditions. Here, the results of 3D fibre orientation on the healing of rat iliac crest defects, using PCL scaffolds with four different geometries, are reported. Further, bone regeneration potential of basic scaffold functiona-lized by the addition of BMP-2/BMP-7-carrying nanocap-sules was studied with CT and histology.

2. Materials and methods

2.1. Materials

PCL (MW 3.7 104) was purchased from Solvay Caprolac-tones (CAPA 6404; UK). Poly(lactic acid-co-glycolic acid) (PLGA; 50:50, i.v. 0.32–0.44 dl/g, 0.1% in chloroform, 25C; ResomerWRG503H) was purchased from Boehrin-ger-Ingelheim (Germany). Poly(3-hydroxybutyrate-co-3-hydroxyvalerate) (PHBV; HV content 8%, w/w) was bought from Sigma-Aldrich (Germany). Recombinant hu-man BMP-2 was from InductOsW(Wyeth Pharmaceuticals, USA) and recombinant human BMP-7 from Prospec Tany Technogene (Israel).

2.2. Preparation of BMP-2- and BMP-7-loaded

nanocapsules

PLGA and PHBV nanocapsules containing BMP-2 and BMP-7, respectively, were prepared by the w/o/w double emulsion technique, as reported earlier (Yilgor et al., 2010b). Briefly, an aqueous solution of BMP was emulsi-fied in a dichloromethane solution of PLGA or PHBV and this was then introduced to a larger volume of aqueous polyvinyl alcohol solution. Nanocapsules were collected by centrifugation, resuspended in distilled water and dried by lyophilization.

(3)

2.3. Production of PCL scaffolds

PCL scaffolds were fabricated using a BioplotterW(Envisiontec GmbH, Germany) (Yilgor et al., 2008). Briefly, PCL was melted at 140C in the heated cartridge unit of the Bioplotter and CO2pressure (5 bar) was applied to the syringe through a

pressurized cap. 3D scaffolds were plotted layer by layer, upto 10 layers, by extrusion of the polymerfibres (Figure 1a–d). Each layer was 20 20 mm with a thickness of 0.25 mm, yielding afinal 10-layered scaffold of 20  20  2.5 mm. Scaf-folds with different architectures were produced by changing the respective orientation of the depositedfibres at different layers, using CAD/CAM software. Four different architectures were obtained: (a) basic (B); (b) basic-offset (BO); (c) crossed (C); and (d) crossed-offset (CO) (Figures 1d–g). Disks to use in the in vivo implantation were cut using a circular die of 5 mm diameter.

2.5. Loading nanocapsules onto PCL scaffolds

Scaffolds were treated with oxygen plasma at 50 W for 1 min (Advanced Plasma Systems, USA) to increase the

hydrophilicity of thefibres and also to remove any rem-nants of extrusions formed during plotting (Yilgor et al., 2010a). Then, nanocapsules suspended in 1%, w/v alginic acid solution were introduced to both sides of the scaf-folds, yielding 40 ng of each BMP/scaffold (40 ng BMP-2 and 40 ng BMP-7/scaffold). After air drying, the scaffolds were dipped into ethanol and then kept in 5% w/v CaCl2

for 1 h to crosslink the alginic acid. The constructs were sterilized with ethylene oxide (Steri-Vac gas sterilizer 5XL, at 37C for 4 h 45 min) prior to implantation.

2.6. Surgery

The experimental protocols used in this study were ap-proved by Gazi University Animal Research Ethical Com-mittee (Ankara). Twenty-one healthy Sprague–Dawley rats, aged 4 weeks and weighing 250–300 g, were ran-domly divided into seven groups. General anaesthesia was induced with a combination of ketamine (60 mg/kg; Ketalar, Pfizer, Turkey) and xylazine (9 mg/kg; Rompun, Bayer, Turkey) intraperitoneally (i.p.). The rats were prone positioned on the operating table under sterile

Figure 1. Production and structure of 3D plotted PCL scaffolds: (a) production of thefirst layer of C scaffold; (b) production of the second layer on thefirst layer of C scaffold; (c) two-layered C scaffold; (d) 10-layered C scaffold; (e) 10-layered B scaffold; (f) 10-lay-ered BO scaffold; (g) 10-lay10-lay-ered CO scaffold

(4)

conditions. Immediately after bilateral vertical pelvic inci-sions, each superior iliac crest of pelvis was exposed (Figure 2a). Bilateral burr holes of 5 mm diameter  2.5 mm deep were created with a pneumatic drill under a surgical microscope (Figure 2b), matching the size of the scaffolds implanted. In the negative control group, the burr holes were left empty, while in all the other groups free nanocapsules or scaffolds were grafted, ensuring that all the surfaces of the samples wereflush with the surfaces of the bone (Figures 2c, d). In these groups, two samples were implanted in each animal, one on each side of the su-perior iliac crest of the pelvis. In total, each sample group was represented by six implants. The experimental groups were: (a) empty defect, E; (b) free nanocapsules, NC; (c) basic scaffold, B; (d) basic-offset scaffold, BO; (e) crossed scaffold, C; (f) crossed-offset scaffold, CO; and (g) basic scaffold loaded with BMP-2-carrying PLGA and BMP-7-carrying PHBV nanocapsules (B+NC). After appropriate haemostasis, the wound site was closed using a size 3.0 monofilament, non-absorbable polypropylene suture.

2.7. Radiological evaluation

CT analysis was used as a quantitative tool to measure the amount of bone formation within the defect 6 weeks postimplantation. All animals were scanned completely immediately after sacrifice in the right lateral decu-bitus position, using a General Electric (GE) Discovery PET/CT system (USA) (voltage 80 kV, current 120 mA, slices 0.6 mm thick). Volumetric reconstruction and anal-ysis were conducted using GE Advantage Workstation AW 4.2-06 software. A threshold of 350 was used for the scans.

To calculate the amount of bone formation within the defect, a circular area of interest of 5 mm diameter was centred over the defect. The size of the defects was con-stant and the same-sized circle (D, 5 mm) was placed in all of the measurements in order to avoid experimental variations between samples during CT analysis. Bone mineral density (BMD) was measured within this area using CT Analyser software (Figure 3).

2.8. Histopathological analysis

Immediately after CT scans, pelvic areas were dissected and fixed in 4% neutral buffered formalin. The samples were decalcified with 5% formic acid at 37C for 5 days with daily

solution replacement. All samples were dehydrated through a graded series of ethanols (70–100%), cleared in xylene and embedded in paraffin. Coronal sections (5 mm) were made with a diamond saw. Three sections were taken from each sample and stained with haematoxylin and eosin (H&E). These sections were evaluated via light microscopy (BX41, Olympus, Japan) by three separate, blinded pathol-ogists for: (a) hard tissue response at the bone–scaffold in-terface; (b) hard tissue response within the scaffold; and (c) quantity of bone formation within the defect; and scored with a quantitative grading scale (0–4) (Patel et al., 2008). The scores were averaged for each sample and then aver-aged for each group to determine the overall score for the group. For hard tissue response at the bone–scaffold inter-face, grade 0 referred to dense inflammation and poorly or-ganized tissue; 1 to unoror-ganizedfibrous tissue; 2 to fibrous tissue surrounding the implant; 3 to remodelling lacuna with osteoblasts and/or osteoclasts at the surface; and 4 to direct bone–implant contact. For hard tissue response

Figure 2. Scaffold implantation into superior iliac crest: (a) exposed iliac crest (IC); (b) bone defect created (black arrow); (c) implan-tation of free nanocapsules (black arrow); (d) implanimplan-tation of empty or BMP-carrying constructs (black arrow) in the defect

(5)

within the scaffold, grades 0–4 referred to dense inflamma-tion without bone (0), inflammation and connective tissue components (1), immature fibrous tissue (2), some bone within mature fibrous tissue (3) and mostly bone (4), re-spectively. Finally, to quantity bone formation within the defect, grades 0–4 indicated 0% (0), 1–24% (1), 25–49% (2), 50–74% (3) and 75–100% (4). The scores were aver-aged for each sample.

2.9. Statistical analysis

All quantitative results were expressed as mean standard deviation (SD; n = 6). Data were analysed with statistically significant values defined as p < 0.05, based on one-way analysis of variance (ANOVA) followed by Tukey’s test for

determination of the significance of difference between dif-ferent groups (p≤ 0.05).

3. Results

3.1. Macroscopic observation and light

microscopy

Macroscopic observation at 6 weeks after implantation showed the integration of B scaffold with the native bone, where it was clearly visible that PCL scaffold retained its structural integrity during this whole period and provided structural and mechanical support to the defect site dur-ing healdur-ing (Figure 4). In some of the samples, the

Figure 3. Determination of bone mineral density at the defect site at 6 weeks post-implantation using computed tomography (CT). A circle 5 mm in diameter (black) was placed, centering the defect site at each side of the pelvis. (a) Left iliac crest defect (LICD). (b) Right iliac crest defect (RICD) in the right lateral decubitus position. BMD was measured in this circular area by CT software

Figure 4. Macroscopic examination of B type scaffolds at 6 weeks post-implantation: (a) some scaffolds were slightly displaced (raised from the bone surface; white arrow), while (b) most retained theirfit (flush with the bone surface; white arrow)

(6)

scaffold was slightly displaced from its original position (Figure 4a), while in the rest, scaffolds completelyfilled the defect and close scaffold–bone interactions were ob-served (Figure 4b).

Micrographs of histopathological analysis for scaffold B were used to demonstrate scaffold–bone interactions. Analysis revealed that B scaffold was successfully inte-grated with the surrounding tissue and its pores were filled with connective tissue, with or without new bone formation (Figure 5a). At higher magnifications, new bone formation within the scaffold pores was apparent, along with remodelling by the activity of osteoblasts and osteoclast-type giant cells (Figure 5b).

Figure 5c shows the presence of new bone formation areas, signs of mineralization and the presence of bony spi-cules among thefibres constituting scaffold B. At a higher

magnification of the mineralization area, the presence of multinucleated osteoclast-type giant cells was observed, along with newly formed bony spicules (Figure 5d). More-over, it was also observed that inflammation was gradually being replaced by mature/immaturefibrous tissue.

Remodelling and new bone formation by the osteo-blasts and osteoclasts was apparent at 6 weeks post-implantation (Figure 5e). Osteoblasts surrounding the bone-formation area within the scaffold pores, where osteocytes were present, was observed (Figure 5f).

3.2. Quantitative histopathological analysis

All samples were evaluated for the quality of bone–scaffold interface, tissue response within scaffold pores and bone

Figure 5. Histopathological sections of the implantation site of scaffold B at 6 weeks post-implantation (H&E staining): (a) the scaf-fold in the iliac crest; the pores of the scafscaf-fold arefilled with connective tissue (20); (b) new bone formation within scaffold (40); (c) osteoclast-type giant cells and mineralization within scaffold (100); (d) higher magnification of (c) showing mineralization and bony spicules (400); (e) remodelling by osteoclasts and osteoblasts; (f) new bone formation within defect by active osteocytes and osteoblasts (1000). B, native bone; S, scaffold; BF, new bone formation; M, mineralization; black arrowhead, osteoclast-type giant cells; black arrow, active bone formation with osteoblasts; white arrow, mature osteocytes in lacunae

(7)

formation within the defect by quantitative histological analysis (Figure 6). Figure 6a shows the scoring results for the hard tissue response at the bone–scaffold interface. It was observed that the bone–scaffold interface quality was somewhat dependent on scaffold geometry and the scores were higher in the case of B and C scaffolds in comparison to their offset counterparts (p≥ 0.05). Free nanocapsules induced bone formation to some extent at the interface with the native bone; however, when they were loaded on the B scaffold, the highest scores were obtained with much lower inflammation and higher direct bone–scaffold contact than with other experimental groups (p< 0.01).

The scoring of hard tissue response within scaffold pores is shown in Figure 6b. Bone formation within scaf-fold B was better than BO, C and CO (p< 0.01). The same

trend observed at the bone–scaffold interface was again attained; the scores for offset scaffolds were lower than regular ones (the ones with higher available pore area). Also, the presence of a growth factor delivery system on scaffold B significantly improved the amount of mature fi-brous tissue within the scaffold pores and reduced the in-flammatory response (p < 0.05).

Bone formation within the defect (Figure 6c) was again higher in scaffold B than with other scaffolds with a lower available pore area (p< 0.05). The presence of BMPs on scaffold B significantly improved bone formation and heal-ing within the defect (p< 0.01). The defects left empty (E) showed almost no healing and union (p< 0.001), depict-ing the importance of scaffolddepict-ing introduced to the bone defects, and this effect was further enhanced by the se-quential growth factor delivery. Free nanocapsules (NC) also induced bone regeneration but to a lower degree than with scaffolds.

3.3. CT analysis

The samples were evaluated for bone formation, using CT as a quantitative method. Bone mineral density (BMD) cal-culations were done by the CT software on the projections where the defect was seen maximally (Figure 7). The posi-tive effect of open pore geometry on the healing of bone defects was also verified with CT analysis. The BMD of the defects implanted with B disks were significantly higher (144.2 7.1) than BO (97.6  6.8) (p < 0.05). Although not significant, B was also higher than C (120.48  32.3); on the other hand, C was significantly higher than CO (79.19.1) (p < 0.05).

The presence of BMP-2 and BMP-7 within free nano-capsules (NC) significantly improved bony healing within the defect, as was characterized by their higher BMD value (108.416.5) in comparison to defects left empty (E) (34.9 10.7) (p < 0.001). When growth factors were provided in a sequential manner while a scaffolding was also present at the defect site, the highest bony healing

Figure 6. Results of histopathological scoring at 6 weeks post-implantation: scores for (a) the hard tissue response at the bone–scaffold interface; (b) the tissue response within the scaf-fold pores; (c) the amount of bone formation within the defect. E, empty defect; NC, free nanocapsules; B, basic scaffold; BO, ba-sic-offset scaffold; C, crossed scaffold; CO, crossed-offset scaf-fold; B + NC, basic scaffold loaded with BMP-2-carrying PLGA and BMP-7-carrying PHBV nanocapsules. Error bars represent SD forn = 6; *p < 0.05, **p < 0.01, ***p < 0.001

Figure 7. Results of CT quantification of bone mineral density at 6 weeks post-implantation. E, empty defect; NC, free nanocap-sules; B, basic scaffold; BO, basic-offset scaffold; C, crossed scaf-fold; CO, crossed-offset scafscaf-fold; B+NC, basic scaffold loaded with BMP-2-carrying PLGA and BMP-7-carrying PHBV nanocap-sules. Error bars represent SD for n = 6; *p < 0.05, **p < 0.01, ***p < 0.001

(8)

and regeneration and highest BMD were obtained (206.8 15.7) (p < 0.01).

4. Discussion

This study aimed to investigate under in vivo conditions the effect of 3D scaffold architecture on the healing of bone defects, using a rat pelvis model. We had previously reported on the influence of 3D orientation and architec-ture of scaffolds on the extent of bone regeneration, mainly by studying the proliferation and differentiation of bone marrow-derived MSCs (Yilgor et al., 2008, 2010a).

Earlier in vitro and in vivo data also suggested that scaf-fold geometry and pore size are parameters that affect bone regeneration. Although there are conflicting views, the majority of researchers indicated that the requisite pore size for bone ingrowth into porous implants should be in the range 100–500 mm, with the interconnections being> 100 mm (Parikh, 2002; Karageorgiou and Kaplan, 2005). However, most of the studies under the in vitro set-tings used scaffolds with pore sizes much smaller than these and surprisingly showed successful population of the scaffolds by the cells (Huang et al., 2010; Dong et al., 2010; Thein and Misra, 2009; Moreau and Xu, 2009; Wei et al., 2007). Under in vivo conditions where osteogenesis depends on other processes such as vascularization in ad-dition to cell infiltration, porosity and especially pore size becomes even more important. The effect of pore size on bone formation was studied using hydroxyapatite (HA) scaffolds with varying pore sizes (100–600 mm) in a subcu-taneous model on rats (Kuboki et al., 2001). It was reported that pore sizes< 300 mm were not favoured by the biological system in terms of both vascularization and bone formation. Subcutaneous implantation of honey-comb-like HA scaffolds with small (100mm) and large (350mm) tunnel diameters revealed that small diameter tunnels favoured chondrogenesis, whereas in larger ones bone formation occurred (Kuboki et al., 2002). A similar result was obtained by Jin et al. (2000), indicating that scaffold geometries allowing enhanced vascularization favoured direct bone formation, whereas low oxygen conditions in the honeycomb-shaped scaffolds induced chondrogenesis. Pore geometry was also shown to be an important parameter; longer and tortuous pores delayed penetration of cells, nutrients and capillaries, and there-fore bone formation occurred mainly at the scaffold surface.

In the present study, PCL scaffolds with four different architectures, B, BO, C and CO (Figure 1), were implanted into rat iliac crest defects created to be fullyfilled by the scaffolds. The percent porosities of these scaffolds were previously assessed by m-CT analysis, indicating that overall scaffold porosity was not significantly affected by the 3Dfibre orientation (Yilgor et al., 2008). On the other hand, the 3D orientation of fibres affected the available surface area for bone ingrowth, calculated from the ratio

of superficial area constituted by the fibres and the dis-tances between them (Table 1). The calculations were done based on the maximum intensity projection images of the scaffolds, where the total area was 19.63 mm2 (r = 2.5 mm) and the pore surface area was the area which was not occupied by the fibres within this area. According to this analysis, B scaffold had highest available pore area (23% of the total superficial area), while the values for BO, C and CO were 11%, 17% and 2%, respec-tively. A higher available pore surface was expected to enhance penetration of cells from the surrounding host tissue and, therefore, to enhance bone regeneration. Enhanced bone regeneration was observed with B scaf-folds, starting with the gross observation of the pelvic area after dissection 6 weeks postimplantation (Figure 4). It was observed that B scaffold was successfully integrated with the native bone, even though no means of fixation was used for the implant at the orthotopic site. Although in a few cases the scaffold was observed to rise slightly from the defect surface, full integration with the defect was evident (Figure 4). This indicates rapid and successful bone ingrowth into the scaffold pores. At this time point, PCL fibres were clearly visible (Figure 4), as the scaffold was expected to have only minimal degradation in 6 weeks. Therefore, the scaffolds provided structural and mechanical support to the defect area during the regeneration process as well as an osteoconductive matrix.

Histopathological analysis revealed that the pores of the B scaffold werefilled with connective tissue and remodel-ling was taking place under the action of active osteo-blasts, osteocytes and osteoclast-type giant cells (Figure 5). Signs of mineralization were observed, characterized by formation of bony spicules in the scaffold pores.

Table 1. Available surface area of PCL scaffolds with different 3Dfibre orientations for bone ingrowth

Scaffold

Maximum intensity projection images

Pore surface

area (mm2) Pore surface area/total area (%)

B 4.5 23

BO 2.1 11

C 3.3 17

(9)

Microscopic analysis of histopathological sections revealed that inflammation caused by formation of the defect and implantation was gradually being replaced with newly-formedfibrous tissue where bone formation centres were apparent on the samples. Further analysis for some more weeks before sacrifice would yield better results in terms of bone formation within scaffold pores, although 6 weeks sufficiently reflects the early tissue response and gives comparative results among the experimental groups.

Figure 6 illustrates the results of histopathological scor-ing, based on microscopical examination of the sections using a quantitative scale. It was observed that the differ-ence in the tortuosity of path and available pore area among B, BO, C and CO scaffolds was not significantly in-fluential on the response at the bone–scaffold interface; however, it affected bone formation within the scaffold and within the defect by influencing the penetration of cells. Scores for bone formation within the scaffold and the defect was highest for B, followed by C, BO and CO (Figure 6), which was also the decreasing order of avail-able pore area (Tavail-able 1). Therefore, bone formation within the defect was found to be directly proportional with the available pore area.

In addition to investigating the effect of scaffold archi-tecture, another purpose of this study was to demonstrate the efficacy of scaffolds functionalized with the sequential growth factor delivery system. Growth factors are regula-tors of cellular activities and multiples of them act in coor-dination in a time- and concentration-dependent manner during fracture healing and bone regeneration processes. Although having great potential to mimic the in vivo con-ditions and to enhance bone regeneration, the literature on the investigation of multiple growth factor-loaded scaf-folds in animal models is limited. Of the available investi-gations, most were conducted using ectopic models, such as combined BMP-2 and VEGF release from polyelectro-lyte multilayer films studied in an intramuscular pocket model (Shah et al., 2011). Some recent studies showed the potential of combined growth delivery on bone regen-eration in orthotopic models. Examples include octacal-cium phosphate-coated implants delivering combined BMP-2/VEGF, investigated in a pig frontal skull model (Ramazanoglu et al., 2011), and delivery of the same cocktail of growth factors (BMP-2/VEGF) from gelatin microparticles within porous scaffolds in a rat cranial de-fect model (Patel et al., 2008; Young et al., 2009), in which the positive effect of using these growth factors in a combined fashion was illustrated. However, there is no study in the literature showing the effect of combined de-livery of BMP-2 and BMP-7 on in vivo bone regeneration. Previous in vitro results by our group illustrated a signi fi-cant increase in bone formation, when bone morphogenetic proteins BMP-2 and BMP-7 encapsulated in nanocapsules (Yilgor et al., 2010b) and loaded on the scaffolds were supplied to the medium in a sequential manner, as happens in nature (Yilgor et al., 2010a). In those earlier studies, sequential BMP-2 and BMP-7 release (first BMP-2 and then BMP-7) was verified with growth factors, both in free form and as incorporated in nanocapsules. In the present study,

the system that had provided promising results in terms of in vitro bone regeneration was evaluated under in vivo conditions. Free growth factor-loaded nanocapsules and unloaded B scaffolds were used as controls to study the ef-fect of BMP-2/BMP-7-loaded B scaffold on the healing of pelvic defects. Histopathological scoring revealed the posi-tive contribution of nanocapsules loaded with BMP-2 and BMP-7, even without a scaffold to bone formation within the defect, in comparison to empty defect (Figure 6c). How-ever, the presence of scaffolds (B, BO, C and CO) further en-hanced bone formation in comparison to empty defect and free nanocapsules, indicating the importance of the pres-ence of a material to guide and support regeneration. More-over, when the nanocapsules containing osteogenic growth factors were supplied to the defect site by loading on the scaffolds, the effect became even more profound and the highest scores were obtained for bone regeneration within scaffold pores and within the defect.

Bone mineral density within defects was evaluated 6 weeks post-implantation, using CT as a quantitative tool. CT analysis verified the results obtained with histological score analysis; scaffolds with a higher available pore area revealed higher BMD values within the defect (Figure 7). Moreover, the presence of osteogenic factor-releasing nano-capsules within the defect enhanced bone formation in comparison to empty defects. The amplifying effect of B scaffold to induce bone regeneration was further enhanced in the presence of the BMP-2 and BMP-7 delivery system.

5. Conclusion

This study aimed to investigate (a) the effect of 3Dfibre orientation and (b) the effect of sequential BMP-2/BMP-7 delivery from the scaffolds on the regeneration of bone defects in a rat pelvis model, as a next step in scaffold de-sign based on the positive results obtained in in vitro set-tings. It was observed that open, accessible pore geometry and large pore surface area enhanced bone regeneration within the scaffold and the defect, although not signi fi-cantly influencing the bone–scaffold interface response. Moreover, as verified previously with rat bone marrow MSCs, BMP-2 and BMP-7 delivered sequentially to the de-fect site, as occurs in nature, further increased bone for-mation in vivo. Finally, the 3D-plotted PCL scaffolds and growth factor-loaded multifunctional constructs en-hanced bone regeneration in the case of open and accessi-ble pores and did this better in the presence of BMPs.

Acknowledgements

This study was conducted within the scope of the EU FP6 NoE Project Expertissues (Grant No. NMP3-CT-2004-500283). We ac-knowledge the support to P.Y. through the same project in the form of an integrated PhD grant. We would also like to acknowl-edge support from Scientific and Technical Research Council of Turkey (TUBITAK) through the project METUNANOBIOMAT (Grant No. TBAG 105T508).

(10)

References

Almirall A, Larrecq G, Delgado JA, et al. 2004; Fabrication of low temperature macroporous hydroxyapatite scaffolds by foaming and hydrolysis of ana-TCP paste. Biomaterials 25: 3671–3680.

Benya PD, Schaffer JD. 1982; Detiated chondrocytes reexpress the differen-tiated collagen phenotype when cultured in agarose gels. Cell 30: 215–224. Bessa PC, Casal M, Reis RL. 2008; Bone

mor-phogenetic proteins in tissue engineering: the road from the laboratory to the clinic, part I (basic concepts), J Tissue Eng Regen Med 2: 1–13.

Cao H, Chen X, Huang L, et al. 2009; Electro-spinning of reconstituted silk fiber from aqueous silk fibroin solution. Mater Sci Eng C 29: 2270–2274.

Dong JL, Li LX, Mu WD, et al. 2010; Bone regeneration with BMP-2 gene-modified mesenchymal stem cells seeded on nano-hydroxyapatite/collagen/poly(L-lactic acid) scaffolds. J Bioact Compat Pol 25: 547–566. Huang Y, Ren J, Ren T, et al. 2010; Bone marrow stromal cells cultured on poly(lactide-co-glycolide)/nano-hydroxyapatite composites with chemical immobilization of Arg–Gly– Asp peptide and preliminary bone regenera-tion of mandibular defect thereof. J Biomed Mater Res 95A: 993–1003.

Jin QM, Takita H, Kohgo T, et al. 2000; Effects of geometry of hydroxyapatite as a cell substratum in BMP-induced ectopic bone formation. J Biomed Mater Res 51: 491–499.

Karageorgiou V, Kaplan D. 2005; Porosity of 3D biomaterial scaffolds and osteogenesis. Biomaterials 26: 5474–5491.

Kuboki Y, Jin Q, Kikuchi M, et al. 2002; Ge-ometry of artificial ECM: sizes of pores controlling phenotype expression in BMP-induced osteogenesis and chondrogenesis. Connect Tissue Res 43: 529–534.

Kuboki Y, Jin Q, Takita H. 2001; Geometry of carriers controlling phenotypic expression in BMP-induced osteogenesis and chon-drogenesis. J Bone Joint Surg Am 83A: S105–115.

Lam CXF, Mo XM, Teoh SH, et al. 2002; Scaffold development using 3D printing with a starch-based polymer. Mater Sci Eng C 20: 49–56. Marra KG, Szem JW, Kumta PN, et al. 1999;

In vitro analysis of biodegradable polymer blend/hydroxyapatite composites for bone tissue engineering. J Biomed Mater Res 47: 324–335.

McKay WF, Peckham SM, Badura JM. 2007; A comprehensive clinical review of

recombinant human bone morphogenetic protein-2 (INFUSEW Bone Graft). Int Orthop 31: 729–734.

Mohajeri S, Hosseinkhani H, Ebrahimi NG, et al. 2010; Proliferation and differentia-tion of mesenchymal stem cell on collagen sponge reinforced with polypropylene/ polyethylene terephthalate blend fibers. Tissue Eng Part A 16: 3821–3830. Moreau JL, Xu HHK. 2009; Mesenchymal

stem cell proliferation and differentia-tion on an injectable calcium phosphate– chitosan composite scaffold. Biomaterials 30: 2675–2682.

Moroni L, de Wijn JR, van Blitterswijk CA. 2006; 3D fiber-deposited scaffolds for tissue engineering: Influence of pores geometry and architecture on dynamic mechanical properties. Biomaterials 27: 974–985.

Navarro M, Michiardi A, Castano O, et al. 2008; Biomaterials in orthopaedics. R Soc Interface 5: 1137–1158.

Parikh SN. 2002; Bone graft substitutes: past, present, future. J Postgrad Med 48: 142–148.

Patel ZS, Young S, Tabata Y, et al. 2008; Dual delivery of an angiogenic and osteogenic growth factor for bone regeneration in a critical size defect model. Bone 43: 931– 940.

Pitt CG. 1990; Poly(e-caprolactone) and its copolymers. In Biodegradable Polymers as Drug Delivery Systems, Chasin M, Langer R (eds). Marcel Dekker: New York; 71–120.

Ramazanoglu M, Lutz R, Ergun C, et al. 2011; The effect of combined delivery of recombi-nant human bone morphogenetic protein-2 and recombinant human vascular endothe-lial growth factor 165 from biomimetic calcium-phosphate-coated implants on osseointegration. Clin Oral Implant Res 22: 1433–1439.

Rohner D, Hutmacher DW, Cheng TK, et al. 2003; In vivo efficacy of bone marrow coated polycaprolactone scaffolds for the reconstruction of orbital defects in the pig. J Biomed Mater Res B Appl Biomater 66: 574–580.

Se HO, Soung GK, Jin HL. 2006; Degradation behavior of hydrophilized PLGA scaffolds prepared by melt-molding particulate-leaching method: comparison with control hydrophobic one. J Mater Sci Mater Med 17: 131–137.

Shah NJ, Macdonald ML, Beben YM, et al. 2011; Tunable dual growth factor delivery

from polyelectrolyte multilayerfilms. Bio-materials 32: 6183–6193.

Sun W, Starly B, Darling A, et al. 2004; Computer-aided tissue engineering: appli-cation to biomimetic modelling and design of tissue scaffolds. J Biotech Appl Biochem 39: 49–58.

Thein HWW, Misra RDK. 2009; Biomimetic chitosan–nanohydroxyapatite composite scaffolds for bone tissue engineering. Acta Biomater 5: 1182–1197.

Tuzlakoglu K, Bolgen N, Salgado AJ, et al. 2005; Nano- and micro-fiber combined scaffolds: a new architecture for bone tis-sue engineering. J Mater Sci Mater Med 16: 1099–1104.

Urist MR. 1965; Bone: formation by autoin-duction. Science 150: 893–899.

Wei G, Jin Q, Giannobile WV, et al. 2007; The enhancement of osteogenesis by nano-fibrous scaffolds incorporating rhBMP-7 nanospheres. Biomaterials 28: 2087–2096. White AP, Vaccaro AR, Hall JA, et al. 2007; Clinical applications of BMP-7/OP-1 in fractures, non-unions and spinal fusion. Int Orthop 31: 735–741.

Williams JM, Adewunmi A, Schek RM, et al. 2005; Bone tissue engineering using polycaprolactone scaffolds fabricated via selective laser sintering. Biomaterials 26: 4817–4827.

Xiong Z, Yan Y, Wang S, et al. 2002; Fabrica-tion of porous scaffolds for bone tissue en-gineering via low-temperature deposition. Sci Mater 46: 771–776.

Yeong WY, Chua CK, Leong KF, et al. 2004; Rapid prototyping in tissue engineering: challenges and potential. Trends Biotechnol 22: 643–652.

Young S, Patel ZS, Kretlow JD, et al. 2009; Dose effect of dual delivery of vascular en-dothelial growth factor and bone morpho-genetic protein-2 on bone regeneration in a rat critical-size defect model. Tissue Eng 15A: 2347–2362.

Yilgor P, Sousa RA, Reis RL, et al. 2008; 3D plotted PCL scaffolds for stem cell based bone tissue engineering. Macromol Symp 269: 92–99.

Yilgor P, Sousa RA, Reis RL, et al. 2010a; Ef-fect of scaffold architecture and BMP2/ BMP7 delivery on in vitro bone regen-eration. J Mater Sci Mater Med 21: 2999–3008.

Yilgor P, Hasirci N, Hasirci V. 2010b; Sequen-tial BMP-2/BMP-7 delivery from polyester nanocapsules. J Biomed Mater Res 93A: 528–536.

Şekil

Figure 1. Production and structure of 3D plotted PCL scaffolds: (a) production of the first layer of C scaffold; (b) production of the second layer on the first layer of C scaffold; (c) two-layered C scaffold; (d) 10-layered C scaffold; (e) 10-layered B scaf
Figure 2. Scaffold implantation into superior iliac crest: (a) exposed iliac crest (IC); (b) bone defect created (black arrow); (c) implan- implan-tation of free nanocapsules (black arrow); (d) implanimplan-tation of empty or BMP-carrying constructs (black
Figure 4. Macroscopic examination of B type scaffolds at 6 weeks post-implantation: (a) some scaffolds were slightly displaced (raised from the bone surface; white arrow), while (b) most retained their fit (flush with the bone surface; white arrow)
Figure 5c shows the presence of new bone formation areas, signs of mineralization and the presence of bony  spi-cules among the fibres constituting scaffold B
+3

Referanslar

Benzer Belgeler

By using PARN-mutated cells from patients as well as a PARN knock-out human cell line generated by CRISPR/Cas9 and carrying an inducible complementing PARN allele, we examined

Çok daha iyi bir ya­ şantıya layıktı, ama o eşi bulunmaz al­ çakgönüllülüğüyle düzeltme şefliğiyle yetiniyor, kendine özgü şiirlerini yaz­ makla mutlu

This study aimed to increase bone formation in the inter-maxillary suture and decrease retention time with the help of grape seed extract (GS), which can stimulate bone

Sultan Baybars, Hülagü’nün Temmuz 1265’de ölümünden sonra yerine geçen oğlu Abaka Han’ın (1265-1281), Altınorda Devleti ile çatışmasından da istifade ederek 17

entelektüel kesimin ortaya çıktığı görülür. Yerli ve Rus olan her şeyi yücelten bu kesimin kültür ve kimliğe ilişkin her türlü varsayımında Batı, bir bakıma

Propolisin in vivo olarak antitümör etkilerinin belirlenmesi amacıyla genellikle Balb/c ırkı fareler kullanılmaktadır ve propolis etken maddeleri gavaj yoluyla, kas veya tümör

Tez çalışmasının ana amacı, Baltalimanı-Sarıyer Sahil Kuşaklama Kollektörleri Projesi uygulamasında kullanılan Alman Herrenknecht firmasından satın alınmış 2

Motivated by the prior investigations in data mining, machine learning, the variable selection techniques with numerous calculation measures and the several searching