- Email: [email protected]
Perichondrium directed cartilage formation in silk fibroin and chitosan blend scaffolds for tracheal transplantation
Acta Biomaterialia 7 (2011) 3422–3431
Contents lists available at ScienceDirect
Acta Biomaterialia journal homepage: www.elsevier.com/locate/actabiomat
Perichondrium directed cartilage formation in silk fibroin and chitosan blend scaffolds for tracheal transplantation Mengqing Zang a,b, Qixu Zhang a, Greg Davis a, George Huang a, Mona Jaffari a, Carmen N. Ríos a, Vishal Gupta a, Peirong Yu a, Anshu B. Mathur a,⇑ a b
Tissue Regeneration and Molecular Cell Engineering Laboratories, Department of Plastic Surgery, The University of Texas M.D. Anderson Cancer Center, Houston, TX 77030, USA Department of Craniofacial Surgery, Plastic Surgery Hospital, Peking Union Medical School, Tsinghua University, Beijing 100144, People’s Republic of China
a r t i c l e
i n f o
Article history: Received 17 November 2010 Received in revised form 9 May 2011 Accepted 11 May 2011 Available online 20 May 2011 Keywords: Biodegradable biological scaffold Silk fibroin Chitosan Cartilage tissue engineering Tracheal regeneration
a b s t r a c t The purpose of this study was to investigate the potential of silk fibroin and chitosan blend (SFCS) biological scaffolds for the purpose of cartilage tissue engineering with applications in tracheal tissue reconstruction. The capability of these scaffolds as cell carrier systems for chondrocytes was determined in vitro and cartilage generation in vivo on engineered chondrocyte–scaffold constructs with and without a perichondrium wrapping was tested in an in vivo nude mouse model. SFCS scaffolds supported chondrocyte adhesion, proliferation, and differentiation, determined as features of the cells based on the spherical cell morphology, increased accumulation of glycosaminoglycans, and increased collagen type II deposition with time within the scaffold framework. Perichondrium wrapping significantly (P < 0.001) improved chondrogenesis within the cell–scaffold constructs in vivo. In vivo implantation for 6 weeks did not generate cartilage structures resembling native trachea, although cartilage-like structures were present. The mechanical properties of the regenerated tissue increased due to the deposition of chondrogenic matrix within the SFCS scaffold structural framework of the trachea. The support of chondrogenesis by the SFCS tubular scaffold construct resulted in a mechanically sound structure and thus is a step towards an engineered trachea that could potentially support the growth of an epithelial lining resulting in a tracheal transplant with properties resembling those of the fully functional native trachea. Ó 2011 Published by Elsevier Ltd. on behalf of Acta Materialia Inc.
1. Introduction Excessive tracheal resection may be required due to cancer influx into host tissues and/or benign diseases of the trachea and is treated by tracheal reconstruction [1]. While defects of up to 50% of the tracheal length in adults and 30% in infants may be reconstructed by primary closure, longer circumferential tracheal defects will require more elaborate surgical reconstructions with grafts and/or flaps [2]. Due to the unique rigid structure and function of the trachea and its air interface (airway) traditional grafting techniques, employing autografts, allografts, and/or prosthetic materials generally do not provide a stable airway, which results in failure [3,4]. Different strategies have been attempted to combine living cells and biocompatible scaffold materials to produce tracheal substitutes. Several in vivo studies have shown the importance of both stable cartilaginous support and a functional epithelium in the long-term success of tracheal reconstruction [5,6]. Therefore, the scaffold used for tracheal engineering will need to
⇑ Corresponding author. Tel.: +1 713 563 7568; fax: +1 713 563 0231. E-mail address: [email protected] (A.B. Mathur).
support regeneration of both tissue components, i.e. cartilage and epithelial lining, in a biological environment that supports the regenerative process, rather than inhibiting it. Silk fibroin (SF) and chitosan (CS) are both naturally occurring biological materials that have been extensively used for cell and tissue applications due to their excellent biocompatibility, degradability, ease of processing, and uniquely engineered structures and mechanical properties [7–9]. We have developed a blended biomaterial to mimic the biochemical composition, architecture, and biomechanical properties of the extracellular matrix (ECM), particularly the structural protein and compressive glycosaminoglycan components [10]. Employing different mold geometries and varying the blending ratios we were able to obtain three-dimensional (3-D) matrices with optimal mechanical properties [10]. SFCS blend scaffolds serve as an effective carrier for stem cells, supporting cell adhesion and migration in vitro and in vivo [11,12]. It also facilitates host cell infiltration and vascular in-growth, and has been deployed successfully for abdominal wall reconstruction [13], skin wound healing [11], and bone regeneration [14]. Recently this hybrid matrix has been proven to support adipose tissue-derived stem cell differentiation into vascular and epithelial phenotypes [11]. Based on these findings we hypothesized that SFCS is a scaffold that
1742-7061/$ - see front matter Ó 2011 Published by Elsevier Ltd. on behalf of Acta Materialia Inc. doi:10.1016/j.actbio.2011.05.012
M. Zang et al. / Acta Biomaterialia 7 (2011) 3422–3431
would support the growth of cartilage for tracheal tissue engineering. In this study we have examined the potential of SFCS in combination with precursor cells and perichondrium for the fabrication of neo-trachea using in vitro and in vivo methods. Rabbit auricular chondrocytes were seeded on SFCS scaffold, as they have been shown to be an efficient cell source for cartilage repair [15,16]. Perichondrium is a physiologically thin barrier covering cartilage and protects it from external influences and is capable of generating cartilage in vivo [17–19]. In an attempt to stabilize the immature in vitro engineered tissue (cells + scaffold) and also provide a source of cartilage precursor cells we wrapped the engineered SFCS constructs with perichondrium before in vivo implantation. The aim of this study was, first, to evaluate whether SFCS could serve as a supportive carrier for chondrocyte delivery and, second, to determine the feasibility of SFCS combined with chondrocytes and perichondrium to construct tracheal cartilage. 2. Materials and methods 2.1. Tubular SFCS scaffold fabrication Raw silk (Sau Paulo, Brazil via the Korean Sericulture Institute) was graciously donated by Dr. Samuel M. Hudson (North Carolina State University, Raleigh, NC) and degummed according to previously described procedures [10]. In brief, 0.25% (w/v) sodium dodecylsulfate (SDS) (Sigma, St. Louis, MO) and 0.25% (w/v) sodium carbonate (Sigma) were dissolved as the temperature was raised to 100 °C. Raw silk was added at 1:100 w/v and heated for 1 h. The alkaline soap solution was then drained and the degummed silk was heated to 100 °C in distilled water for an additional 1 h. Finally, any remaining sericin and surfactants were removed completely by rinsing the silk in running distilled water. The washed silk was then air dried. The degummed silk was then dissolved by adding it to a calcium nitrate tetrahydrate–methanol solution
3423
(molar ratio 1:4:2 Ca(NO3)24H2O:MeOH at 65 °C). The silk fibroin was dissolved at 10% (w/v) concentration over a 3 h period with continuous stirring. The final protein concentration of silk fibroin was calculated to be 4.08% (w/v). High molecular weight chitosan (82.7% deacetylation, Sigma) was dissolved in an aqueous solution of 2% acetic acid to create a solution of 4.08% (w/v), equal to that of the silk fibroin. Silk fibroin and chitosan were blended at 75:25 % (v/v) to prepare silk fibroin– chitosan, as this ratio showed superior mechanical properties compared with the other blends [10]. The mixture was then dialyzed (molecular weight cut-off 6–8 kDa) against deionized water for 4 days and strained through a 100 lm filter. The final aqueous solution was clear and homogeneous and was kept at room temperature. A mold to resemble the shape of the trachea was created from two concentric tubes with 5 mm separation between the outer polystyrene tube and the inner silastic tube. Tubular sfcaffolds were prepared by adding 5 ml of SFCS solution between the polystyrene tube and the silastic tube, with their ends sealed with paraffin wax. The solution was frozen in the mold at 80 °C overnight, followed by lyophilization for 72 h. The dry scaffolds were then treated in a 50:50 % (v/v) methanol:sodium hydroxide (1 N) solution for 15 min to crystallize the SF content and neutralize the CS content. The scaffolds were rinsed in distilled water and phosphate-buffered saline (PBS) (Lonza, Walkersville, MD). Before cell seeding the tabular scaffolds were sterilized by immersion in 70% ethanol overnight and then washed in PBS three times on the day of cell seeding. The final dimensions of the SFCS tubular scaffolds prior to implantation were 30 mm in length, 7 mm in outer diameter, and 2 mm in thickness (Fig. 1A). 2.2. Chondrocyte isolation and culture All animal care and surgical procedures were in compliance with the Institutional Animal Care and Use Committee (IACUC) of The University of Texas M.D. Anderson Cancer Center.
Fig. 1. Preparation of SFCS scaffold–cell–perichondrium constructs. (A) The tubular SFCS scaffold after aqueous crystallization shows an intact scaffold around a siloxane tube. (B) Phase contrast photomicrograph of passage 3 chondrocytes in monolayer culture after 24 h culture (magnification 200). (C) At the time of implantation the perichondrium was elevated from the rabbit ear cartilage and straightened in place with a needle. (D) The cell–scaffold construct (CS) was wrapped with perichondrium (CSP) before being implanted.
3424
M. Zang et al. / Acta Biomaterialia 7 (2011) 3422–3431
Ear cartilage was harvested from six adult New Zealand white rabbits (3.5–4.0 kg). The cartilage sheets were minced to approximately 1 mm3 pieces and digested in 0.1% collagenase type II (Worthington Biochemical Corp., Lakewood, NJ) in Dulbecco’s modified Eagle’s medium (DMEM) with 4.0 g l 1 glucose (Hyclone) on a shaker at 37 °C for 21 h. The resulted cell suspension was centrifuged at 200g for 8 min and the cell pellet was resuspended in growth medium consisting of DMEM with 4.0 g l 1 glucose supplemented with 10% fetal bovine serum (Atlanta Biologicals, Lawrenceville, GA) and 1% antibiotic blend (100 U ml 1 penicillin G, 100 lg ml 1 streptomycin sulfate and 250 ng ml 1 amphotericin B) (Hyclone, Waltham, MA). Then the cells were plated at a density of 10,000 cells cm 2 and incubated at 37 °C with medium changes every 3 days. At 80% confluency the cells were detached with 0.05% trypsin–EDTA (Gibco, Carlsbad, CA) and frozen in cell freezing solution (Sigma) at densities of 2–4 106 cells vial 1 and stored in liquid nitrogen. Prior to seeding of cells in the SFCS scaffold the chondrocytes were thawed and plated at a density of 8000 cells cm 2 in two-dimensional culture. The third passage cells (Fig. 1B) were used for cell seeding on SFCS scaffolds. 2.3. Characterization of the SFCS scaffold using polarized light microscopy Tubular SFCS scaffolds were imaged using polarizer, analyzer, and a red retardation plate attached to an Olympus IX70 (Olympus, Center Valley, PA) microscope. Polarized light microscopy was used to assess the alignment and orientation of the fibrils within the 3-D assembled scaffold. The diameters of the fibrils were measured using the image analysis tools in ImageJ (n = 29). 2.4. Scanning electron microscopy (SEM) examination of cell–scaffold interaction Tubular SFCS scaffolds were cut into disks (6 mm diameter) to fit in the wells of a 96-well plate. Chondrocytes were seeded on scaffolds at a density of 1.25 106 cells disk 1 and incubated in differentiation medium (DMEM with 4.0 g l 1 glucose (Hyclone) supplemented with 1% ITS + Premix™ (BD Biosciences, Franklin Lakes, NJ), 37.5 lg ml 1 ascorbate-2-phosphate (Sigma), 10 7 M dexamethasone (Sigma) and 1% antibiotic) at 37 °C overnight. The cell–scaffold constructs were then placed in a 24-well plate for further incubation and the differentiation medium was changed every 48 h. On days 1, 3, 7, 14, and 21 cell-seeded SFCS scaffolds were washed with PBS and fixed with 3% glutaraldehyde and 2% paraformaldehyde, sequentially treated with buffered tannic acid, aqueous uranyl acetate, and dehydrated ethanol, transferred to an increasing graded series of hexamethyl disilazane, and finally air dried overnight. Samples were then coated with platinum alloy and examined using a JSM-590 scanning electron microscope (JEOL Inc., Peabody, MA). Adherent cell morphology, patterns of cellular adherence with respect to structural features of SFCS, cell–cell interactions, and adherent cell density were assessed. Tubular scaffolds without cells were assessed as controls. 2.5. Cell seeding for in vitro and in vivo studies Chondrocytes (2.5 107 cells ml 1) were statically loaded onto the tubular scaffold and then resuspended in 0.5 ml of differentiation medium by constricting the volume of medium around the scaffolding within a 2 ml microtube. Following incubation for 2 h the cell–scaffold constructs were transferred to 35 mm Petri dishes. For the in vitro study the constructs were continuously cultured in differentiation medium, harvested after 1, 2, 3, 5 and 6 weeks
(n = 3), and fixed for histological analysis to determine cell distribution, accumulation of glycosaminoglycans (GAG), and deposition of collagen type II. For in vivo studies the composites were cultured for up to 1 week and the differentiation medium changed daily. These cell–scaffold constructs were then used for implantation. 2.6. Implantation of tissue engineered constructs All animal care and surgical procedures were in compliance with the Institutional Animal Care and Use Committee (IACUC) of The University of Texas M.D. Anderson Cancer Center. Under general anesthesia with isoflurane (0.5–2%, 3–5 l min 1) and oxygen, perichondria were harvested aseptically from the ears of four New Zealand white rabbits under loupe magnification (Fig. 1c). The nu/nu strain nude mice used in this study were obtained from Charles River Laboratories (Wilmington, MA). Under general anesthesia with isoflurane (0.5–2%, 3–5 l min 1) and oxygen subcutaneous pockets were created in the right and left dorsum of six nude mice. Six cell–scaffold constructs (CS) were implanted into the subcutaneous pockets in the right dorsum. On the left side of the six nude mice the cell-scaffold constructs were wrapped with strips of rabbit perichondria (CSP) (Fig. 1D) with the cambium layer in contact with the constructs. Six weeks following implantation the constructs were explanted and cut into three pieces each. One piece was fixed in 10% neutral buffered formalin for histology, one was wet weighed and snap frozen for subsequent biochemical assay of GAG content, and the other was kept on wet gauze in PBS for mechanical testing. 2.7. Histological analysis Serial sections of the specimens were stained with hematoxylin and eosin (H&E) and MOVAT pentachrome and immunostained with anti-collagen type II. The sections were examined by light microscopy and photographed using a digital camera. Histomorphometric analysis was performed with ImageJ 1.40 software. As the region of cartilaginous tissue selection by the software is based on a color threshold, MOVAT pentachrome stained sections/images were used to obtain the best color contrast between the region of interest (ROI) and the background for image analysis. 2.8. In vitro glycosaminoglycan (GAG) deposition GAG deposition analysis was performed to evaluate cartilagespecific ECM synthesis by chondrocytes in the SFCS scaffold. Regions stained with alcian blue in the MOVAT stain were considered to be the area of GAG deposition. The total region was outlined and the ROI automatically discriminated by ImageJ. The percentage GAG deposition was calculated. 2.9. In vivo chondrogenesis analysis Chondrogenesis analysis was performed to assess cartilage tissue formation in vivo. Only the regions stained blue in MOVAT containing chondrocytes surrounded by lacunae were characterized as cartilage tissue. These regions were differentiated and the percentage ROI calculated. 2.10. Biochemical assay of GAG content CS explants (n = 6), CSP explants (n = 6) and native rabbit trachea (n = 4) underwent biochemical assay to quantify GAG content. Samples were wet weighed and then digested in 1 ml papain solution (125 lg ml papain 1 (Sigma) in sterile PBS, pH 6.0, with 5 mM cysteineHCl and 5 mM EDTA) at 60 °C for 24 h. GAG content was
M. Zang et al. / Acta Biomaterialia 7 (2011) 3422–3431
measured using an alcian blue colorimetric assay (GAG Dye Binding Assay, ALPCO, Salem, NH). 2.11. Tensile testing Uniaxial tensile tests were performed on SFCS scaffolds (n = 4), CS (n = 4), and CSP explants (n = 6). Native rabbit trachea (n = 2) were also examined and served as a control. Samples were cut into rectangular strips and subjected to tensile testing using an EnduraTEC mechanical tester (Bose, Eden Prairie, MN). A stress–strain curve was obtained and analysis of the data was performed using the stress–strain curves to determine values for ultimate tensile strength (UTS), elastic modulus, and strain at failure. 2.12. Statistical analysis All quantitative data are presented as means ± standard errors of the mean (s.e.m). Differences between groups were assessed using one-way analysis of variance (ANOVA) with a post hoc Tukey or Dunn’s test using the SigmaStat 3.5 statistical program. The level of statistically significant difference was chosen as P < 0.05. 3. Results 3.1. Characterization of the tubular SFCS scaffold Polarized light microscopy, without the red retardation plate, shows that the tubular SFCS scaffold is composed of layered stacked sheets wrapped circumferentially around the cylindrical tube, with polymer chain alignment (white) evident at the end of each sheet (black areas) (Fig. 2A and B). The small SF fibrils extending from the ends of the sheets have a diameter of 0.85 ± 0.05 lm. The SF fibrillar extensions are comprised of aligned SF polymer chains within each fibril, as imaged by polarized light microscopy with the red retardation plate and under bright field microscopy (Fig. 2C and D). Silk fibroin polymer chain alignment is visible in
3425
blue for polymer chains that are at 45° to the polarizer–analyzer and perpendicular to the red retardation plate, whereas fibrils interspersed within the magenta-pink background that are parallel to the red retardation plate appear orange-pink due to the strong magenta background (Fig. 2C). SEM imaging shows similar overlapped sheets with fibrillar extensions at the ends of the sheets (Fig. 3A and B). 3.2. Cell and scaffold interaction Within 24 h of cell seeding chondrocytes covered the entire SFCS scaffold surface, over the layered morphological structures (Fig. 3C). The cells formed numerous aggregates that fused together and covered the entire scaffold, although single cells could be identified between the seamlessly fused cellular layers (Fig. 3D). Chondrocytes flattened after another 2 days in vitro culture and the construct showed a smooth surface paved with overlapping polygonal cells with a uniform morphology (Fig. 3E and F). By 7 days, while the cells became more crowded and the intercellular space blurred the cellular morphology became bulged and rounded (Fig. 3G and H). At 14 days, the cell outline was more obscure as intercellular gaps were filled with ECM deposition (Fig. 3I). Some multicellular spheroids formed on the surface and the cells around their periphery became stretched and radially oriented towards the center of the aggregate (Fig. 3J). Following another week of culture, during which ECM deposition was visible, the cells were buried by the newly synthesized ECM (Fig. 3K and L). 3.3. ECM deposition GAG and collagen type II deposition could be detected after 1 week of chondrocyte culture on SFCS scaffolds, which increased with time and was deposited by the cellular aggregates (Fig. 4). At week 6, chondrocytes appear to increase the GAG and type II collagen deposition in the top layers of the scaffold as compared to the inner areas of the scaffold where little GAG and type II
Fig. 2. Polarized light micrographs of a tubular SFCS scaffold with and without the red retardation plates. (A) Under the polarizer–analyzer (P–A) with no red retardation plate (R) the tubular SFCS scaffold shows stacked sheets with aligned sheet ends along the length of the analyzer in white and the short fibrillar extensions perpendicular to the analyzer also in white, whereas the black background exhibits unaligned smooth sheet regions of the 3-D SFCS scaffolding. (B) Rotation of the sheets perpendicular to the analyzer from a parallel configuration and at higher magnification shows stacking of the sheets, polymer chain alignment within the sheets, and heterogeneity of the system. (C) Placement of the tubular SFCS scaffold within the constraints of the polarizer–analyzer and a red retardation plate shows the longitudinal sheet ends as blue and the fibrillar ends as yellow, both perpendicular to each other as the analyzer is at 45° to the red retardation plate, turning the unaligned regions red. (D) The image of the sample under bright field microscopy shows the sheets and the fibrils with black outlines against a gray transparent background.
3426
M. Zang et al. / Acta Biomaterialia 7 (2011) 3422–3431
collagen was synthesized and deposited. The cartilaginous tissue near the surface is distinguished from that in the deeper layers by intensive staining for GAG and collagen type II. Most of the cells in the scaffold remained rounded and only cells in the outer areas formed a pericellular matrix structure similar to the lacunae of native cartilage tissue. The histomorphometric analysis of GAG deposition showed that GAG accumulated continuously over the 6 weeks in vitro cultivation. GAG content increased significantly from week 3 to 6 and reached 42.5% by week 6. 3.4. In vitro cell distribution Qualitative observation of cells seeded on the scaffold showed a distribution throughout the scaffold. At week 1, the majority of chondrocytes aggregated in the top layers of the scaffold. Cells infiltrated deeply and formed multicellular spheroids that were distributed in the core areas of the scaffold (Fig. 5A). Over the following 5 weeks, the cell aggregates became enlarged and infused throughout the scaffold, but some still remained separated in the deeper areas below the surface (Fig. 5B). Meanwhile, the scaffolds appeared to be replaced by the new matrix synthesized by the chondrocytes. 3.5. Macroscopic observation of the in vivo implants No infections occurred during the period of in vivo implantation. All mice survived. The CS explants were soft and shrank significantly after removal of the supportive tube (Fig. 6A and B). The
CSP explants (Fig. 6C and D) showed stiffness in some sites when evaluated manually by compressing the sample. Thicker walls and more apparent white patches were observed in CSP explants as compared to the CS explants. 3.6. Characterization of cartilage formation in the implants Mature cartilage tissue consisted of chondrocytes within the lacunae and rich matrix substance (collagen and GAG) in both the CS and CSP explants (Fig. 7A and B). In the CS explants, mature cartilage islands formed mainly in the peripheral areas of the scaffold with some collagen deposition between partially degraded scaffold sheets in other areas (Fig. 7A). Meanwhile in the CSP explants two distinct areas were divided by a clear boundary composed of multiple layers of collagen fibrils (Fig. 7B). In the CSP explants small areas of immature cartilaginous tissue were sporadically apparent in the interior, which contained compacted chondrocyte aggregates and weakly stained matrix (Fig. 7C). Cartilage sheets were formed in the peripheral area with an appearance similar to normal cartilage tissue. There was no scaffold found in this area. In the interior, under the collagen layers, there was a large area of immature cartilaginous tissue composed of lightly stained matrix and buried scaffold (Fig. 7D). In all explants collagen type II was only detected in the mature cartilaginous tissue (Fig. 7E and F). The histomorphometric analysis showed that the CSP explants had a significantly (P < 0.001) higher percentage neo-chondrogenesis, 13.1 ± 1.8% compared with 3.1 ± 0.5%, than the CS explants (Fig. 7G).
Fig. 3. SEM images of the surface of the tubular SFCS scaffold before and after chondrocyte seeding for 24 h and 3 and 7 days at low and high magnification. (A) The surface has stacked sheets (see arrows) similar to the polarized light microscopy images. (B) The sheets consist of embedded fibrillar extensions (see arrows). (C) By 24 h the entire scaffold surface was covered by chondrocytes. (D) While some chondrocytes maintained a spherical morphology upon initial seeding, others had fused and spread on the stacked sheet tubular SFCS scaffold surface by 24 h. (E) At low magnification it was evident that the entire surface was covered by an even layer of well-spread cells by day 3. (F) A pentagonal surface morphology of the cells was visible at higher magnification for day 3 samples. (G and H) By day 7, at both low and high magnification, more layers of cells were visible above the initial chondrocytic cell layer formed by day 3 due to cellular proliferation. (I and J) Multi-cellular spheroids were visible at low and high magnification by day 14. (K and L) These spheroids appeared to have stretched evenly as an additional layer of cells by 21 days over the existing pentagonally differentiated chondrocytes, as imaged at both low and high magnification.
M. Zang et al. / Acta Biomaterialia 7 (2011) 3422–3431
3427
Fig. 4. ECM deposition by chondrocytes in the SFCS scaffold in vitro. (Left) Representative images of MOVAT pentachrome staining of cell-scaffold constructs at different time points: (A) 1 week; (B) 3 weeks; (C) 6 weeks (the scaffold was stained red, the GAG blue). Magnification 40; scale bar: 500 lm. (Right) Immunostaining with anti-collagen type II of cell–scaffold constructs: (E) 1 week; (F) 3 weeks; (G) 6 weeks (collagen type II was detected as brown). Magnification 40; scale bar: 500 lm. At 6 weeks the difference in ECM deposition between the outer and inner areas was distinguished by staining intensity variations (C and G: outer area, white asterisk; inner area; black triangle). The majority of chondrocytes maintained a rounded morphology in SFCS scaffolds and the cells formed lacunae-like structures in the outer area (D and H: residual scaffold, white arrow). Magnification 200; scale bar: 100 lm. (I) Percent GAG deposition by chondrocytes in the SFCS scaffolds was quantified in vitro after 1, 2, 3, 5 and 6 weeks, based on histomorphometric analysis. aP < 0.05, aaP < 0.01 compared with 5 weeks and ⁄⁄P < 0.01, ⁄⁄⁄P < 0.001 compared with 6 weeks.
Fig. 5. (A) Cell distribution in the SFCS scaffold: chondrocytes formed multicellular spheroids sparsely distributed in the deeper areas of the scaffold. Magnification 100. (B) After 6 weeks the cell aggregates were enlarged and fused, but there was still unoccupied space remaining in the core area. Magnification 100.
3.7. Quantification of GAG content The GAG content assay was used to detect chondrogenic ECM formation in the CS and CSP explants after 6 weeks of in vivo implantation. GAG contents were similar in the CSP explants (1.72 ± 0.38 lg mg wet wt 1) and CS explants (1.71 ± 0.68 lg mg wet wt 1). Both explants showed significantly lower GAG contents compared with native rabbit trachea (8.93 ± 1.50 lg mg wet wt 1) (Fig. 8). 3.8. Comparison of mechanical properties The stress–strain curves for the CS and CSP explants showed a steeper increase in stress with strain compared with native
trachea. Five of six samples in the CSP group failed by abrupt fracture rather than the slow tearing observed in native trachea and the CS group. Different groups did not differ significantly in UTS, elastic modulus or strain at failure (Fig. 9). Both the CS and CSP explants showed comparable ultimate tensile strengths to that of native trachea (5.70 102 ± 1.24 102 kPa for CS, 4.60 102 ± 0.59 102 kPa for CSP and 4.08 102 ± 0.96 102 kPa for native). The CS and CSP explants appeared to be stiffer than the native trachea, as exhibited by the higher elastic modulus (1.28 103 ± 5.10 102 kPa for CS, 1.10 103 ± 2.23 102 kPa for CSP, 3.46 102 ± 1.09 102 kPa for native). Moreover, when the strain at failure was compared (0.59 ± 0.10 m m 1 for CS, 0.49 ± 0.09 m m 1 for CSP, and 1.65 ± 0.53 m m 1 for native trachea), CS- and CSP-regenerated cartilage had significantly
3428
M. Zang et al. / Acta Biomaterialia 7 (2011) 3422–3431
Fig. 6. Engineered constructs harvested at 6 weeks after implantation. Gross morphology: (A) CS explants; (C) CSP explants. Cross-section view: (B) CS explants; (D) CSP explants.
Fig. 7. The histology of engineered constructs harvested 6 weeks after implantation after MOVAT pentachrome staining: (A and C) CS explants; (B and D) CSP explants. The scaffold was stained red, GAG blue and collagen fibers yellow. Ex, exterior area; In, interior area; black arrow, immature cartilage tissue; white arrow, mature cartilage tissue; black asterisk, collagen fiber boundary. Magnification: (A and B) 40; (C and D) 200. Immunostaining of collagen type II: (E) CS explants; (F) CSP explants. Collagen type II was stained brown. Magnification 200. (G) New cartilage tissue formation in CS and CSP explants based on the histomorphometric analysis. ⁄P < 0.001 vs. CS, t-test.
M. Zang et al. / Acta Biomaterialia 7 (2011) 3422–3431
Fig. 8. Biochemical assay of the GAG content of native trachea (Native) and CSP and CS explants. ⁄P = 0.01 vs. native; §P = 0.001 vs. native, one-way ANOVA.
(P < 0.01 for both) lower strains at failure as compared with native trachea. The mechanical properties of the pre-implant SFCS tubular scaffolds were 1.88 102 ± 0.18 102 kPa for UTS, 2.47 102 ± 0.23 102 kPa for elastic modulus, and 0.81 ± 0.01 m m 1 for strain at failure. The UTS of the CS-regenerated cartilage was significantly (P < 0.05) higher than the pre-implant SFCS scaffold. The CS- and CSP-regenerated cartilage was significantly (P < 0.01 for CS and P < 0.05 for CSP) stiffer than the pre-implant scaffold. Only the native trachea had a significantly (P < 0.05) higher strain at failure than pre-implant SFCS. 4. Discussion In this study, we fabricated a unique tubular scaffold geometry with the SFCS blend polymers that has smooth layered sheets
3429
intermingled with wrapped around the tubular structure. It is evident that the SFCS composition and this new structural configuration is a good conductor of chondrogenic cells and supports neochondrogenesis in vitro and in vivo. Chondrocytes proliferate and form aggregates on the scaffold surface in vitro. The presence of perichondrium results in increased production of chondrogenic matrix in vivo, although there is no effect of the perichondrium on the relative mechanical properties of the CS- or CSP-regenerated cartilage, except for relative strain at failure. The static seeding and culture technique failed to achieve homogeneous cell and ECM distribution and consequently may have affected the fate of cells within the scaffold framework under in vivo conditions. The new SFCS tubular/tracheal framework scaffolds (third generation SFCS scaffolds) resulted in a similar sheet assembly with fibrillar extensions as previously fabricated first generation SFCS scaffolds from our group [10,20], and exhibited combined features with second generation SFCS scaffolds [20] that have aligned SF polymer chains within the fibrils, in the direction of the fibril, and within the new tubular geometry, which has not been observed before with SFCS biomaterial systems. The tubular scaffolds also have significantly (P < 0.0001) higher tensile strengths, elastic moduli, and strains at failure compared with the first generation SFCS scaffolds. The better mechanical properties could be attributed to the SF polymer chain aligned along the fibril length throughout the scaffold and the tubular geometry itself. While the combination of SF and CS has been studied before for cartilage tissue engineering, tubular structures as clinical alternatives are being studied here for the first time [21,22]. The structural configuration of the matrix/scaffold is an important parameter for cell adhesion, migration, differentiation, and morphogenesis to the cartilage phenotype. SEM examination showed that the SFCS scaffolds support high cell seeding densities, since a high density of chondrocytes attached, proliferated, and differentiated on the scaffold over time. While previous studies have utilized pure SF sponges to support chondrocyte proliferation and maintain chondrogenic phenotype expression in vitro and in vivo [21,23–25], SFCS has an additional element CS, which has structural and physical
Fig. 9. Tensile testing results: (A) representative stress and strain curves of native trachea (Native) and CSP and CS explants; (B) tensile properties comparison between native trachea (Native) and CSP and CS explants.
3430
M. Zang et al. / Acta Biomaterialia 7 (2011) 3422–3431
properties similar to the GAG found in hyaline cartilage, which also supports chondrogenesis [8,26]. Cartilage-specific ECM in pure SF, polyglycolic acid (PGA), and hyalograft scaffolds have been reported to be directly related to increased cell seeding densities in vitro and in vivo [24,27,28]. Our findings suggest that a blend of SF and CS provides an excellent 3-D matrix for in vitro and in vivo chondrogenesis for up to 6 weeks, similar to other biological materials. According to Benya and Shaffer [29] monolayer culture was used to promote rapid proliferation of cells and delay differentiation, as chondrocytes can recover their differentiated phenotype when they are relocated from a 2-D to a 3-D environment. In other studies chondrocytes have been shown to lose their chondrogenic phenotype during monolayer expansion [30,31]. This study showed that chondrocytes recovered their differentiation potential shortly after they were cultured on a 3-D SFCS scaffold. At 24 h in vitro culture most of the cells acquired a spherical morphology on the SFCS surface, usually an initial visible differentiation marker for the chondrocytic phenotype [32], accompanied by early GAG and collagen type II deposition detected at 1 week, which increased with time, thereby maintaining the cellular phenotype on the SFCS surface. Aggregates of chondrocytes were observed both on the surface and within the stacked sheet architecture of the SFCS scaffold. During prenatal development and de novo tissue formation a 3-D scaffold is required to provide an environment to facilitate cell condensation and cell–cell interactions that mimic physiological processes [33]. Cell aggregation improves cell–cell contact and to some extent mimics the mesenchymal condensation step, where mesenchymal stem cells spontaneously form aggregates, produce more GAG, and show higher transcription levels of collagen type II [34]. Comparatively, this study also suggests that cell aggregation in response to the 3-D SFCS scaffold partially contributes to early chondrocytic differentiation and continuous ECM accumulation. The static cell seeding technique may have caused an uneven cell distribution through the various dimensions of the scaffold. Initially cells concentrated near the surface of the scaffold and only a few cells infiltrated into the deeper areas. Although the core zone was eventually filled with cells as cells proliferated with time, migrated inward, and only empty spaces remained between the cell aggregates. Static cultures have limitations in terms of mass transfer [35], which could explain the differences in ECM disposition between the outer and inner areas of the scaffold. Mass transfer is important for chondrocyte culture because the tissue is avascular and nutrient exchange is mainly dependent on diffusion between the matrix and the surrounding environment [7]. Diffusion of nutrients, metabolic waste, and soluble molecules might be further limited by the accumulation of ECM components in the outer areas, resulting in substantially lower cell activity in the inner areas of the SFCS scaffold [24]. The limitations of the static cell seeding technique may also explain why we were unable to produce tubular cartilaginous structures in our in vivo studies. Because of the lack of a uniform distribution of cells throughout the scaffold, newly formed cartilage and ECM are distributed more in the surface areas. Our previous in vivo studies showed that the SFSC scaffold favored host tissue in-growth and rapid remodeling in vivo[11,13]. We observed a similar remodeling phenomenon in this study, exhibited by collagen deposition around the partially degraded remnants of the SFCS scaffold. Among the remodeling tissue, some chondrocyte aggregates with immature ECM could be recognized. It is highly plausible that the unoccupied spaces in the constructs were rapidly infiltrated by remodeling cells, which left limited room for the seeded chondrocytes to fabricate new ECM. The perichondrium was successfully used to protect and stabilize the implanted tissue engineered cartilage from initial invasion
by inflammatory cells and reduce the effects of fibrous capsule formation observed in other studies [36–38]. Additionally, perichondrium is rich in progenitor cells that are capable of differentiating as cartilage tissue [39]. We have shown in the present study that wrapping in perichondrium significantly improved new cartilage formation in the cell–scaffold constructs. There was a large area of immature cartilaginous tissue undergoing remodeling with buried non-degraded scaffold found in the internal areas. Mature cartilage-like tissue encapsulated by dense collagen fibers formed in the peripheral areas of the implanted constructs, especially where the scaffold was completely degraded. An analysis of the internal and external areas indicated that gradual degradation of the scaffold resulted in the deposition of the new matrix at the same location, thereby strengthening the existing tracheal support areas and minimizing or nullifying any loss of mechanical integrity where the scaffold remained, with or without cells. Furthermore, while perichondrium acts as a reservoir for cartilage forming precursor cells, it might also act as a barrier to isolate and protect immature cartilaginous tissue in the constructs from an early host response and allow implanted chondrocytes to deposit new ECM with local colonization. We have also shown in this study that the mechanical properties of cell–scaffold constructs were similar to native trachea, although they appeared to be stiffer and less extensible. Clinically, however, reduced extensibility is not a major concern. The rigidity of a tubular structure in maintaining an open airway is the key mechanical property of tissue engineered trachea, which comparatively was equivalent to native trachea, thus showing the potential of the SFCS for tracheal tissue reconstructive applications. It was reported that the development of tissue engineered cartilage could be enhanced by extending the precultivation time and employing advanced bioreactors [28]. Prolonged in vitro culture would allow extensive ECM formation which will protect the seeded cells and also decrease post-implantation inflammation [28,40], although the in vivo remodeling time by host derived cells after invasion of the construct may be prolonged due to the increased deposition of ECM by in vitro seeded cells. Bioreactors represent attractive tools to accelerate the development of the required biochemical and mechanical properties of engineered tissues, which also influence the type of ECM molecules excreted into the cellular microenvironment due to local mass transfer and physical stimuli effects [41]. In our future studies we will combine these factors to improve homogeneous cell seeding and extensive ECM formation in the scaffold. 5. Conclusions Cartilage formation is key to the mechanical integrity of the native trachea. In this study we focused on the reconstruction of tracheal defects in cancer patients who needed a mechanically sound tracheal graft for reconstruction of the tumor-resected defect site. Thus an approach was devised that would translate to the clinic, such that a tracheal graft could be constructed by subcutaneously implanting a biomaterial in the patient and then using the patients own cells to regenerate the tracheal graft, which thus consists of native cartilage, as opposed to tracheal allotransplantation which involves immunosuppressive therapies [42]. An SFCS scaffold provided an excellent surface for cell adhesion at high densities and a 3-D environment for chondrocyte proliferation, cell–cell contact and differentiation under in vitro conditions prior to in vivo implantation and cartilage regeneration. While wrapping with perichondrium improved neochondrogenesis in vivo, the static seeding and culture technique resulted in a heterogeneous cell and ECM distribution in vitro and in vivo. While this study shows that the in vivo tracheal reconstruction animal model and the SFCS construct works well to promote neochondrogenesis improvements
M. Zang et al. / Acta Biomaterialia 7 (2011) 3422–3431
are needed to produce a more functional and dynamic in vitro cell seeding and culture system. Acknowledgements This study was supported by the Kyte Research Fund from the Department of Plastic Surgery at the University of Texas M.D. Anderson Cancer Center. We thank Kenneth Dunner Jr. of the High Resolution Electron Microscopy Facility at M.D. Anderson Cancer Center for his assistance with the SEM, and Arnie Jimenez of VelLabs for assistance in preparing the histological sections. M.Z. was financially supported by the China Scholarship Council. References [1] Matloub HS, Yu P. Engineering a composite neotrachea in a rat model. Plast Reconstr Surg 2006;117(1):123–8. [2] Grillo HC. Tracheal replacement: a critical review. Ann Thorac Surg 2002;73(6):1995–2004. [3] Steger V, Hampel M, Trick I, Muller M, Walles T. Clinical tracheal replacement: transplantation, bioprostheses and artificial grafts. Expert Rev Med Devices 2008;5(5):605–12. [4] Baiguera S, Birchall MA, Macchiarini P. Tissue-engineered tracheal transplantation. Transplantation 2010;89(5):485–91. [5] Go T, Jungebluth P, Baiguero S, Asnaghi A, Martorell J, Ostertag H, et al. Both epithelial cells and mesenchymal stem cell-derived chondrocytes contribute to the survival of tissue-engineered airway transplants in pigs. J Thorac Cardiovasc Surg 2010;139(2):437–43. [6] Kojima K, Bonassar LJ, Roy AK, Mizuno H, Cortiella J, Vacanti CA. A composite tissue-engineered trachea using sheep nasal chondrocyte and epithelial cells. FASEB J 2003;17(8):823–8. [7] MacIntosh AC, Kearns VR, Crawford A, Hatton PV. Skeletal tissue engineering using silk biomaterials. J Tissue Eng Regen Med 2008;2(2/3):71–80. [8] Suh JK, Matthew HW. Application of chitosan-based polysaccharide biomaterials in cartilage tissue engineering: a review. Biomaterials 2000;21(24):2589–98. [9] Wang Y, Kim HJ, Vunjak-Novakovic G, Kaplan DL. Stem cell-based tissue engineering with silk biomaterials. Biomaterials 2006;27(36):6064–82. [10] Gobin AS, Froude VE, Mathur AB. Structural and mechanical characteristics of silk fibroin and chitosan blend scaffolds for tissue regeneration. J Biomed Mater Res 2005;74(3):465–73. [11] Altman AM, Yan Y, Matthias N, Bai X, Rios C, Mathur AB, et al. IFATS collection: human adipose-derived stem cells seeded on a silk fibroin–chitosan scaffold enhance wound repair in a murine soft tissue injury model. Stem Cells 2009;27(1):250–8. [12] Altman AM, Gupta V, Rios CN, Alt EU, Mathur AB. Adhesion, migration and mechanics of human adipose-tissue-derived stem cells on silk fibroin– chitosan matrix. Acta Biomater 2010;6(4):1388–97. [13] Gobin AS, Butler CE, Mathur AB. Repair and regeneration of the abdominal wall musculofascial defect using silk fibroin–chitosan blend. Tissue Eng 2006;12(12):3383–94. [14] Rios CN, Skoracki RJ, Miller MJ, Satterfield WC, Mathur AB. In vivo bone formation in silk fibroin and chitosan blend scaffolds via ectopically grafted periosteum as a cell source: a pilot study. Tissue Eng Part A 2009;15(9):2717–25. [15] Henderson JH, Welter JF, Mansour JM, Niyibizi C, Caplan AI, Dennis JE. Cartilage tissue engineering for laryngotracheal reconstruction: comparison of chondrocytes from three anatomic locations in the rabbit. Tissue Eng 2007;13(4):843–53. [16] Kamil SH, Eavey RD, Vacanti MP, Vacanti CA, Hartnick CJ. Tissue-engineered cartilage as a graft source for laryngotracheal reconstruction: a pig model. Arch Otolaryngol Head Neck Surg 2004;130(9):1048–51. [17] Ohlsen L, Nordin U. Tracheal reconstruction with perichondrial grafts. Scand J Plast Reconstr Surg Hand Surg 1976;10(2):135–45. [18] Clevens RA, Esclamado R, Naficy S. Vascularized auricular perichondrium in airway reconstruction: the effects of stenting with and without a mucosal graft. The Laryngoscope 1995;105(10):1043–8.
3431
[19] Naficy S, Esclamado RM, Clevens RA. Reconstruction of the rabbit trachea with vascularized auricular perichondrium. Ann Otol Rhinol Laryngol 1996;105(5):356–62. [20] Gupta V, Davis G, Gordon A, Altman AM, Reece GP, Gascoyne PR, et al. Endothelial and stem cell interactions on dielectrophoretically aligned fibrous silk fibroin–chitosan scaffolds. J Biomed Mater Res A 2010;94A:515–23. [21] Seda Tigli R, Ghosh S, Laha MM, Shevde NK, Daheron L, Gimble J, et al. Comparative chondrogenesis of human cell sources in 3D scaffolds. J Tissue Eng Regen Med 2009;3(5):348–60. [22] Silva SS, Motta A, Rodrigues MT, Pinheiro AF, Gomes ME, Mano JF, et al. Novel genipin-cross-linked chitosan/silk fibroin sponges for cartilage engineering strategies. Biomacromolecules 2008;10(2):764–74. [23] Aoki H, Tomita N, Morita Y, Hattori K, Harada Y, Sonobe M, et al. Culture of chondrocytes in fibroin–hydrogel sponge. Biomed Mater Eng 2003;13(4):309–16. [24] Wang Y, Blasioli DJ, Kim HJ, Kim HS, Kaplan DL. Cartilage tissue engineering with silk scaffolds and human articular chondrocytes. Biomaterials 2006;27(25):4434–42. [25] Shangkai C, Naohide T, Koji Y, Yasuji H, Masaaki N, Tomohiro T, et al. Transplantation of allogeneic chondrocytes cultured in fibroin sponge and stirring chamber to promote cartilage regeneration. Tissue Eng 2007;13(3):483–92. [26] Kim IY, Seo SJ, Moon HS, Yoo MK, Park IY, Kim BC, et al. Chitosan and its derivatives for tissue engineering applications. Biotechnol Adv 2008;26(1):1–21. [27] Mahmoudifar N, Doran PM. Effect of seeding and bioreactor culture conditions on the development of human tissue-engineered cartilage. Tissue Eng 2006;12(6):1675–85. [28] Moretti M, Wendt D, Dickinson SC, Sims TJ, Hollander AP, Kelly DJ, et al. Effects of in vitro preculture on in vivo development of human engineered cartilage in an ectopic model. Tissue Eng 2005;11(9/10):1421–8. [29] Benya PD, Shaffer JD. Dedifferentiated chondrocytes reexpress the differentiated collagen phenotype when cultured in agarose gels. Cell 1982;30(1):215–24. [30] von der Mark K, Gauss V, von der Mark H, Muller P. Relationship between cell shape and type of collagen synthesised as chondrocytes lose their cartilage phenotype in culture. Nature 1977;267(5611):531–2. [31] Holtzer H, Abbott J, Lash J, Holtzer S. The loss of phenotypic traits by differentiated cells in vitro, I. Dedifferentiation of cartilage cells. Proc Natl Acad Sci USA 1960;46(12):1533–42. [32] Archer CW, Rooney P, Wolpert L. Cell shape and cartilage differentiation of early chick limb bud cells in culture. Cell Differ 1982;11(4):245–51. [33] Wescoe KE, Schugar RC, Chu CR, Deasy BM. The role of the biochemical and biophysical environment in chondrogenic stem cell differentiation assays and cartilage tissue engineering. Cell Biochem Biophys 2008;52(2): 85–102. [34] Ghosh S, Laha M, Mondal S, Sengupta S, Kaplan DL. In vitro model of mesenchymal condensation during chondrogenic development. Biomaterials 2009;30(33):6530–40. [35] Thevenot P, Nair A, Dey J, Yang J, Tang L. Method to analyze three-dimensional cell distribution and infiltration in degradable scaffolds. Tissue Eng Part A 2008;14(4):319–31. [36] Vinatier C, Mrugala D, Jorgensen C, Guicheux J, Noel D. Cartilage engineering: a crucial combination of cells, biomaterials and biofactors. Trends Biotechnol 2009;27(5):307–14. [37] Anderson JM, Rodriguez A, Chang DT. Foreign body reaction to biomaterials. Semin Immunol 2008;20(2):86–100. [38] Haisch A, Groger A, Gebert C, Leder K, Ebmeyer J, Sudhoff H, et al. Creating artificial perichondrium by polymer complex membrane macroencapsulation: immune protection and stabilization of subcutaneously transplanted tissueengineered cartilage. Eur Arch Otorhinolaryngol 2005;262(4):338–44. [39] Togo T, Utani A, Naitoh M, Ohta M, Tsuji Y, Morikawa N, et al. Identification of cartilage progenitor cells in the adult ear perichondrium: utilization for cartilage reconstruction. Lab Invest 2006;86(5):445–57. [40] Luo X, Zhou G, Liu W, Zhang WJ, Cen L, Cui L, et al. In vitro precultivation alleviates post-implantation inflammation and enhances development of tissue-engineered tubular cartilage. Biomed Mater 2009;4(2):025006. [41] Concaro S, Gustavson F, Gatenholm P. Bioreactors for tissue engineering of cartilage. Adv Biochem Eng Biotechnol 2008;112:125–43. [42] Delaere P, Vranckx J, Verleden G, De Leyn P, van Raemdonck D. Tracheal allotransplantation after withdrawal of immunosuppressive therapy. N Engl J Med 2010;362(2):138–45.
Contents lists available at ScienceDirect
Acta Biomaterialia journal homepage: www.elsevier.com/locate/actabiomat
Perichondrium directed cartilage formation in silk fibroin and chitosan blend scaffolds for tracheal transplantation Mengqing Zang a,b, Qixu Zhang a, Greg Davis a, George Huang a, Mona Jaffari a, Carmen N. Ríos a, Vishal Gupta a, Peirong Yu a, Anshu B. Mathur a,⇑ a b
Tissue Regeneration and Molecular Cell Engineering Laboratories, Department of Plastic Surgery, The University of Texas M.D. Anderson Cancer Center, Houston, TX 77030, USA Department of Craniofacial Surgery, Plastic Surgery Hospital, Peking Union Medical School, Tsinghua University, Beijing 100144, People’s Republic of China
a r t i c l e
i n f o
Article history: Received 17 November 2010 Received in revised form 9 May 2011 Accepted 11 May 2011 Available online 20 May 2011 Keywords: Biodegradable biological scaffold Silk fibroin Chitosan Cartilage tissue engineering Tracheal regeneration
a b s t r a c t The purpose of this study was to investigate the potential of silk fibroin and chitosan blend (SFCS) biological scaffolds for the purpose of cartilage tissue engineering with applications in tracheal tissue reconstruction. The capability of these scaffolds as cell carrier systems for chondrocytes was determined in vitro and cartilage generation in vivo on engineered chondrocyte–scaffold constructs with and without a perichondrium wrapping was tested in an in vivo nude mouse model. SFCS scaffolds supported chondrocyte adhesion, proliferation, and differentiation, determined as features of the cells based on the spherical cell morphology, increased accumulation of glycosaminoglycans, and increased collagen type II deposition with time within the scaffold framework. Perichondrium wrapping significantly (P < 0.001) improved chondrogenesis within the cell–scaffold constructs in vivo. In vivo implantation for 6 weeks did not generate cartilage structures resembling native trachea, although cartilage-like structures were present. The mechanical properties of the regenerated tissue increased due to the deposition of chondrogenic matrix within the SFCS scaffold structural framework of the trachea. The support of chondrogenesis by the SFCS tubular scaffold construct resulted in a mechanically sound structure and thus is a step towards an engineered trachea that could potentially support the growth of an epithelial lining resulting in a tracheal transplant with properties resembling those of the fully functional native trachea. Ó 2011 Published by Elsevier Ltd. on behalf of Acta Materialia Inc.
1. Introduction Excessive tracheal resection may be required due to cancer influx into host tissues and/or benign diseases of the trachea and is treated by tracheal reconstruction [1]. While defects of up to 50% of the tracheal length in adults and 30% in infants may be reconstructed by primary closure, longer circumferential tracheal defects will require more elaborate surgical reconstructions with grafts and/or flaps [2]. Due to the unique rigid structure and function of the trachea and its air interface (airway) traditional grafting techniques, employing autografts, allografts, and/or prosthetic materials generally do not provide a stable airway, which results in failure [3,4]. Different strategies have been attempted to combine living cells and biocompatible scaffold materials to produce tracheal substitutes. Several in vivo studies have shown the importance of both stable cartilaginous support and a functional epithelium in the long-term success of tracheal reconstruction [5,6]. Therefore, the scaffold used for tracheal engineering will need to
⇑ Corresponding author. Tel.: +1 713 563 7568; fax: +1 713 563 0231. E-mail address: [email protected] (A.B. Mathur).
support regeneration of both tissue components, i.e. cartilage and epithelial lining, in a biological environment that supports the regenerative process, rather than inhibiting it. Silk fibroin (SF) and chitosan (CS) are both naturally occurring biological materials that have been extensively used for cell and tissue applications due to their excellent biocompatibility, degradability, ease of processing, and uniquely engineered structures and mechanical properties [7–9]. We have developed a blended biomaterial to mimic the biochemical composition, architecture, and biomechanical properties of the extracellular matrix (ECM), particularly the structural protein and compressive glycosaminoglycan components [10]. Employing different mold geometries and varying the blending ratios we were able to obtain three-dimensional (3-D) matrices with optimal mechanical properties [10]. SFCS blend scaffolds serve as an effective carrier for stem cells, supporting cell adhesion and migration in vitro and in vivo [11,12]. It also facilitates host cell infiltration and vascular in-growth, and has been deployed successfully for abdominal wall reconstruction [13], skin wound healing [11], and bone regeneration [14]. Recently this hybrid matrix has been proven to support adipose tissue-derived stem cell differentiation into vascular and epithelial phenotypes [11]. Based on these findings we hypothesized that SFCS is a scaffold that
1742-7061/$ - see front matter Ó 2011 Published by Elsevier Ltd. on behalf of Acta Materialia Inc. doi:10.1016/j.actbio.2011.05.012
M. Zang et al. / Acta Biomaterialia 7 (2011) 3422–3431
would support the growth of cartilage for tracheal tissue engineering. In this study we have examined the potential of SFCS in combination with precursor cells and perichondrium for the fabrication of neo-trachea using in vitro and in vivo methods. Rabbit auricular chondrocytes were seeded on SFCS scaffold, as they have been shown to be an efficient cell source for cartilage repair [15,16]. Perichondrium is a physiologically thin barrier covering cartilage and protects it from external influences and is capable of generating cartilage in vivo [17–19]. In an attempt to stabilize the immature in vitro engineered tissue (cells + scaffold) and also provide a source of cartilage precursor cells we wrapped the engineered SFCS constructs with perichondrium before in vivo implantation. The aim of this study was, first, to evaluate whether SFCS could serve as a supportive carrier for chondrocyte delivery and, second, to determine the feasibility of SFCS combined with chondrocytes and perichondrium to construct tracheal cartilage. 2. Materials and methods 2.1. Tubular SFCS scaffold fabrication Raw silk (Sau Paulo, Brazil via the Korean Sericulture Institute) was graciously donated by Dr. Samuel M. Hudson (North Carolina State University, Raleigh, NC) and degummed according to previously described procedures [10]. In brief, 0.25% (w/v) sodium dodecylsulfate (SDS) (Sigma, St. Louis, MO) and 0.25% (w/v) sodium carbonate (Sigma) were dissolved as the temperature was raised to 100 °C. Raw silk was added at 1:100 w/v and heated for 1 h. The alkaline soap solution was then drained and the degummed silk was heated to 100 °C in distilled water for an additional 1 h. Finally, any remaining sericin and surfactants were removed completely by rinsing the silk in running distilled water. The washed silk was then air dried. The degummed silk was then dissolved by adding it to a calcium nitrate tetrahydrate–methanol solution
3423
(molar ratio 1:4:2 Ca(NO3)24H2O:MeOH at 65 °C). The silk fibroin was dissolved at 10% (w/v) concentration over a 3 h period with continuous stirring. The final protein concentration of silk fibroin was calculated to be 4.08% (w/v). High molecular weight chitosan (82.7% deacetylation, Sigma) was dissolved in an aqueous solution of 2% acetic acid to create a solution of 4.08% (w/v), equal to that of the silk fibroin. Silk fibroin and chitosan were blended at 75:25 % (v/v) to prepare silk fibroin– chitosan, as this ratio showed superior mechanical properties compared with the other blends [10]. The mixture was then dialyzed (molecular weight cut-off 6–8 kDa) against deionized water for 4 days and strained through a 100 lm filter. The final aqueous solution was clear and homogeneous and was kept at room temperature. A mold to resemble the shape of the trachea was created from two concentric tubes with 5 mm separation between the outer polystyrene tube and the inner silastic tube. Tubular sfcaffolds were prepared by adding 5 ml of SFCS solution between the polystyrene tube and the silastic tube, with their ends sealed with paraffin wax. The solution was frozen in the mold at 80 °C overnight, followed by lyophilization for 72 h. The dry scaffolds were then treated in a 50:50 % (v/v) methanol:sodium hydroxide (1 N) solution for 15 min to crystallize the SF content and neutralize the CS content. The scaffolds were rinsed in distilled water and phosphate-buffered saline (PBS) (Lonza, Walkersville, MD). Before cell seeding the tabular scaffolds were sterilized by immersion in 70% ethanol overnight and then washed in PBS three times on the day of cell seeding. The final dimensions of the SFCS tubular scaffolds prior to implantation were 30 mm in length, 7 mm in outer diameter, and 2 mm in thickness (Fig. 1A). 2.2. Chondrocyte isolation and culture All animal care and surgical procedures were in compliance with the Institutional Animal Care and Use Committee (IACUC) of The University of Texas M.D. Anderson Cancer Center.
Fig. 1. Preparation of SFCS scaffold–cell–perichondrium constructs. (A) The tubular SFCS scaffold after aqueous crystallization shows an intact scaffold around a siloxane tube. (B) Phase contrast photomicrograph of passage 3 chondrocytes in monolayer culture after 24 h culture (magnification 200). (C) At the time of implantation the perichondrium was elevated from the rabbit ear cartilage and straightened in place with a needle. (D) The cell–scaffold construct (CS) was wrapped with perichondrium (CSP) before being implanted.
3424
M. Zang et al. / Acta Biomaterialia 7 (2011) 3422–3431
Ear cartilage was harvested from six adult New Zealand white rabbits (3.5–4.0 kg). The cartilage sheets were minced to approximately 1 mm3 pieces and digested in 0.1% collagenase type II (Worthington Biochemical Corp., Lakewood, NJ) in Dulbecco’s modified Eagle’s medium (DMEM) with 4.0 g l 1 glucose (Hyclone) on a shaker at 37 °C for 21 h. The resulted cell suspension was centrifuged at 200g for 8 min and the cell pellet was resuspended in growth medium consisting of DMEM with 4.0 g l 1 glucose supplemented with 10% fetal bovine serum (Atlanta Biologicals, Lawrenceville, GA) and 1% antibiotic blend (100 U ml 1 penicillin G, 100 lg ml 1 streptomycin sulfate and 250 ng ml 1 amphotericin B) (Hyclone, Waltham, MA). Then the cells were plated at a density of 10,000 cells cm 2 and incubated at 37 °C with medium changes every 3 days. At 80% confluency the cells were detached with 0.05% trypsin–EDTA (Gibco, Carlsbad, CA) and frozen in cell freezing solution (Sigma) at densities of 2–4 106 cells vial 1 and stored in liquid nitrogen. Prior to seeding of cells in the SFCS scaffold the chondrocytes were thawed and plated at a density of 8000 cells cm 2 in two-dimensional culture. The third passage cells (Fig. 1B) were used for cell seeding on SFCS scaffolds. 2.3. Characterization of the SFCS scaffold using polarized light microscopy Tubular SFCS scaffolds were imaged using polarizer, analyzer, and a red retardation plate attached to an Olympus IX70 (Olympus, Center Valley, PA) microscope. Polarized light microscopy was used to assess the alignment and orientation of the fibrils within the 3-D assembled scaffold. The diameters of the fibrils were measured using the image analysis tools in ImageJ (n = 29). 2.4. Scanning electron microscopy (SEM) examination of cell–scaffold interaction Tubular SFCS scaffolds were cut into disks (6 mm diameter) to fit in the wells of a 96-well plate. Chondrocytes were seeded on scaffolds at a density of 1.25 106 cells disk 1 and incubated in differentiation medium (DMEM with 4.0 g l 1 glucose (Hyclone) supplemented with 1% ITS + Premix™ (BD Biosciences, Franklin Lakes, NJ), 37.5 lg ml 1 ascorbate-2-phosphate (Sigma), 10 7 M dexamethasone (Sigma) and 1% antibiotic) at 37 °C overnight. The cell–scaffold constructs were then placed in a 24-well plate for further incubation and the differentiation medium was changed every 48 h. On days 1, 3, 7, 14, and 21 cell-seeded SFCS scaffolds were washed with PBS and fixed with 3% glutaraldehyde and 2% paraformaldehyde, sequentially treated with buffered tannic acid, aqueous uranyl acetate, and dehydrated ethanol, transferred to an increasing graded series of hexamethyl disilazane, and finally air dried overnight. Samples were then coated with platinum alloy and examined using a JSM-590 scanning electron microscope (JEOL Inc., Peabody, MA). Adherent cell morphology, patterns of cellular adherence with respect to structural features of SFCS, cell–cell interactions, and adherent cell density were assessed. Tubular scaffolds without cells were assessed as controls. 2.5. Cell seeding for in vitro and in vivo studies Chondrocytes (2.5 107 cells ml 1) were statically loaded onto the tubular scaffold and then resuspended in 0.5 ml of differentiation medium by constricting the volume of medium around the scaffolding within a 2 ml microtube. Following incubation for 2 h the cell–scaffold constructs were transferred to 35 mm Petri dishes. For the in vitro study the constructs were continuously cultured in differentiation medium, harvested after 1, 2, 3, 5 and 6 weeks
(n = 3), and fixed for histological analysis to determine cell distribution, accumulation of glycosaminoglycans (GAG), and deposition of collagen type II. For in vivo studies the composites were cultured for up to 1 week and the differentiation medium changed daily. These cell–scaffold constructs were then used for implantation. 2.6. Implantation of tissue engineered constructs All animal care and surgical procedures were in compliance with the Institutional Animal Care and Use Committee (IACUC) of The University of Texas M.D. Anderson Cancer Center. Under general anesthesia with isoflurane (0.5–2%, 3–5 l min 1) and oxygen, perichondria were harvested aseptically from the ears of four New Zealand white rabbits under loupe magnification (Fig. 1c). The nu/nu strain nude mice used in this study were obtained from Charles River Laboratories (Wilmington, MA). Under general anesthesia with isoflurane (0.5–2%, 3–5 l min 1) and oxygen subcutaneous pockets were created in the right and left dorsum of six nude mice. Six cell–scaffold constructs (CS) were implanted into the subcutaneous pockets in the right dorsum. On the left side of the six nude mice the cell-scaffold constructs were wrapped with strips of rabbit perichondria (CSP) (Fig. 1D) with the cambium layer in contact with the constructs. Six weeks following implantation the constructs were explanted and cut into three pieces each. One piece was fixed in 10% neutral buffered formalin for histology, one was wet weighed and snap frozen for subsequent biochemical assay of GAG content, and the other was kept on wet gauze in PBS for mechanical testing. 2.7. Histological analysis Serial sections of the specimens were stained with hematoxylin and eosin (H&E) and MOVAT pentachrome and immunostained with anti-collagen type II. The sections were examined by light microscopy and photographed using a digital camera. Histomorphometric analysis was performed with ImageJ 1.40 software. As the region of cartilaginous tissue selection by the software is based on a color threshold, MOVAT pentachrome stained sections/images were used to obtain the best color contrast between the region of interest (ROI) and the background for image analysis. 2.8. In vitro glycosaminoglycan (GAG) deposition GAG deposition analysis was performed to evaluate cartilagespecific ECM synthesis by chondrocytes in the SFCS scaffold. Regions stained with alcian blue in the MOVAT stain were considered to be the area of GAG deposition. The total region was outlined and the ROI automatically discriminated by ImageJ. The percentage GAG deposition was calculated. 2.9. In vivo chondrogenesis analysis Chondrogenesis analysis was performed to assess cartilage tissue formation in vivo. Only the regions stained blue in MOVAT containing chondrocytes surrounded by lacunae were characterized as cartilage tissue. These regions were differentiated and the percentage ROI calculated. 2.10. Biochemical assay of GAG content CS explants (n = 6), CSP explants (n = 6) and native rabbit trachea (n = 4) underwent biochemical assay to quantify GAG content. Samples were wet weighed and then digested in 1 ml papain solution (125 lg ml papain 1 (Sigma) in sterile PBS, pH 6.0, with 5 mM cysteineHCl and 5 mM EDTA) at 60 °C for 24 h. GAG content was
M. Zang et al. / Acta Biomaterialia 7 (2011) 3422–3431
measured using an alcian blue colorimetric assay (GAG Dye Binding Assay, ALPCO, Salem, NH). 2.11. Tensile testing Uniaxial tensile tests were performed on SFCS scaffolds (n = 4), CS (n = 4), and CSP explants (n = 6). Native rabbit trachea (n = 2) were also examined and served as a control. Samples were cut into rectangular strips and subjected to tensile testing using an EnduraTEC mechanical tester (Bose, Eden Prairie, MN). A stress–strain curve was obtained and analysis of the data was performed using the stress–strain curves to determine values for ultimate tensile strength (UTS), elastic modulus, and strain at failure. 2.12. Statistical analysis All quantitative data are presented as means ± standard errors of the mean (s.e.m). Differences between groups were assessed using one-way analysis of variance (ANOVA) with a post hoc Tukey or Dunn’s test using the SigmaStat 3.5 statistical program. The level of statistically significant difference was chosen as P < 0.05. 3. Results 3.1. Characterization of the tubular SFCS scaffold Polarized light microscopy, without the red retardation plate, shows that the tubular SFCS scaffold is composed of layered stacked sheets wrapped circumferentially around the cylindrical tube, with polymer chain alignment (white) evident at the end of each sheet (black areas) (Fig. 2A and B). The small SF fibrils extending from the ends of the sheets have a diameter of 0.85 ± 0.05 lm. The SF fibrillar extensions are comprised of aligned SF polymer chains within each fibril, as imaged by polarized light microscopy with the red retardation plate and under bright field microscopy (Fig. 2C and D). Silk fibroin polymer chain alignment is visible in
3425
blue for polymer chains that are at 45° to the polarizer–analyzer and perpendicular to the red retardation plate, whereas fibrils interspersed within the magenta-pink background that are parallel to the red retardation plate appear orange-pink due to the strong magenta background (Fig. 2C). SEM imaging shows similar overlapped sheets with fibrillar extensions at the ends of the sheets (Fig. 3A and B). 3.2. Cell and scaffold interaction Within 24 h of cell seeding chondrocytes covered the entire SFCS scaffold surface, over the layered morphological structures (Fig. 3C). The cells formed numerous aggregates that fused together and covered the entire scaffold, although single cells could be identified between the seamlessly fused cellular layers (Fig. 3D). Chondrocytes flattened after another 2 days in vitro culture and the construct showed a smooth surface paved with overlapping polygonal cells with a uniform morphology (Fig. 3E and F). By 7 days, while the cells became more crowded and the intercellular space blurred the cellular morphology became bulged and rounded (Fig. 3G and H). At 14 days, the cell outline was more obscure as intercellular gaps were filled with ECM deposition (Fig. 3I). Some multicellular spheroids formed on the surface and the cells around their periphery became stretched and radially oriented towards the center of the aggregate (Fig. 3J). Following another week of culture, during which ECM deposition was visible, the cells were buried by the newly synthesized ECM (Fig. 3K and L). 3.3. ECM deposition GAG and collagen type II deposition could be detected after 1 week of chondrocyte culture on SFCS scaffolds, which increased with time and was deposited by the cellular aggregates (Fig. 4). At week 6, chondrocytes appear to increase the GAG and type II collagen deposition in the top layers of the scaffold as compared to the inner areas of the scaffold where little GAG and type II
Fig. 2. Polarized light micrographs of a tubular SFCS scaffold with and without the red retardation plates. (A) Under the polarizer–analyzer (P–A) with no red retardation plate (R) the tubular SFCS scaffold shows stacked sheets with aligned sheet ends along the length of the analyzer in white and the short fibrillar extensions perpendicular to the analyzer also in white, whereas the black background exhibits unaligned smooth sheet regions of the 3-D SFCS scaffolding. (B) Rotation of the sheets perpendicular to the analyzer from a parallel configuration and at higher magnification shows stacking of the sheets, polymer chain alignment within the sheets, and heterogeneity of the system. (C) Placement of the tubular SFCS scaffold within the constraints of the polarizer–analyzer and a red retardation plate shows the longitudinal sheet ends as blue and the fibrillar ends as yellow, both perpendicular to each other as the analyzer is at 45° to the red retardation plate, turning the unaligned regions red. (D) The image of the sample under bright field microscopy shows the sheets and the fibrils with black outlines against a gray transparent background.
3426
M. Zang et al. / Acta Biomaterialia 7 (2011) 3422–3431
collagen was synthesized and deposited. The cartilaginous tissue near the surface is distinguished from that in the deeper layers by intensive staining for GAG and collagen type II. Most of the cells in the scaffold remained rounded and only cells in the outer areas formed a pericellular matrix structure similar to the lacunae of native cartilage tissue. The histomorphometric analysis of GAG deposition showed that GAG accumulated continuously over the 6 weeks in vitro cultivation. GAG content increased significantly from week 3 to 6 and reached 42.5% by week 6. 3.4. In vitro cell distribution Qualitative observation of cells seeded on the scaffold showed a distribution throughout the scaffold. At week 1, the majority of chondrocytes aggregated in the top layers of the scaffold. Cells infiltrated deeply and formed multicellular spheroids that were distributed in the core areas of the scaffold (Fig. 5A). Over the following 5 weeks, the cell aggregates became enlarged and infused throughout the scaffold, but some still remained separated in the deeper areas below the surface (Fig. 5B). Meanwhile, the scaffolds appeared to be replaced by the new matrix synthesized by the chondrocytes. 3.5. Macroscopic observation of the in vivo implants No infections occurred during the period of in vivo implantation. All mice survived. The CS explants were soft and shrank significantly after removal of the supportive tube (Fig. 6A and B). The
CSP explants (Fig. 6C and D) showed stiffness in some sites when evaluated manually by compressing the sample. Thicker walls and more apparent white patches were observed in CSP explants as compared to the CS explants. 3.6. Characterization of cartilage formation in the implants Mature cartilage tissue consisted of chondrocytes within the lacunae and rich matrix substance (collagen and GAG) in both the CS and CSP explants (Fig. 7A and B). In the CS explants, mature cartilage islands formed mainly in the peripheral areas of the scaffold with some collagen deposition between partially degraded scaffold sheets in other areas (Fig. 7A). Meanwhile in the CSP explants two distinct areas were divided by a clear boundary composed of multiple layers of collagen fibrils (Fig. 7B). In the CSP explants small areas of immature cartilaginous tissue were sporadically apparent in the interior, which contained compacted chondrocyte aggregates and weakly stained matrix (Fig. 7C). Cartilage sheets were formed in the peripheral area with an appearance similar to normal cartilage tissue. There was no scaffold found in this area. In the interior, under the collagen layers, there was a large area of immature cartilaginous tissue composed of lightly stained matrix and buried scaffold (Fig. 7D). In all explants collagen type II was only detected in the mature cartilaginous tissue (Fig. 7E and F). The histomorphometric analysis showed that the CSP explants had a significantly (P < 0.001) higher percentage neo-chondrogenesis, 13.1 ± 1.8% compared with 3.1 ± 0.5%, than the CS explants (Fig. 7G).
Fig. 3. SEM images of the surface of the tubular SFCS scaffold before and after chondrocyte seeding for 24 h and 3 and 7 days at low and high magnification. (A) The surface has stacked sheets (see arrows) similar to the polarized light microscopy images. (B) The sheets consist of embedded fibrillar extensions (see arrows). (C) By 24 h the entire scaffold surface was covered by chondrocytes. (D) While some chondrocytes maintained a spherical morphology upon initial seeding, others had fused and spread on the stacked sheet tubular SFCS scaffold surface by 24 h. (E) At low magnification it was evident that the entire surface was covered by an even layer of well-spread cells by day 3. (F) A pentagonal surface morphology of the cells was visible at higher magnification for day 3 samples. (G and H) By day 7, at both low and high magnification, more layers of cells were visible above the initial chondrocytic cell layer formed by day 3 due to cellular proliferation. (I and J) Multi-cellular spheroids were visible at low and high magnification by day 14. (K and L) These spheroids appeared to have stretched evenly as an additional layer of cells by 21 days over the existing pentagonally differentiated chondrocytes, as imaged at both low and high magnification.
M. Zang et al. / Acta Biomaterialia 7 (2011) 3422–3431
3427
Fig. 4. ECM deposition by chondrocytes in the SFCS scaffold in vitro. (Left) Representative images of MOVAT pentachrome staining of cell-scaffold constructs at different time points: (A) 1 week; (B) 3 weeks; (C) 6 weeks (the scaffold was stained red, the GAG blue). Magnification 40; scale bar: 500 lm. (Right) Immunostaining with anti-collagen type II of cell–scaffold constructs: (E) 1 week; (F) 3 weeks; (G) 6 weeks (collagen type II was detected as brown). Magnification 40; scale bar: 500 lm. At 6 weeks the difference in ECM deposition between the outer and inner areas was distinguished by staining intensity variations (C and G: outer area, white asterisk; inner area; black triangle). The majority of chondrocytes maintained a rounded morphology in SFCS scaffolds and the cells formed lacunae-like structures in the outer area (D and H: residual scaffold, white arrow). Magnification 200; scale bar: 100 lm. (I) Percent GAG deposition by chondrocytes in the SFCS scaffolds was quantified in vitro after 1, 2, 3, 5 and 6 weeks, based on histomorphometric analysis. aP < 0.05, aaP < 0.01 compared with 5 weeks and ⁄⁄P < 0.01, ⁄⁄⁄P < 0.001 compared with 6 weeks.
Fig. 5. (A) Cell distribution in the SFCS scaffold: chondrocytes formed multicellular spheroids sparsely distributed in the deeper areas of the scaffold. Magnification 100. (B) After 6 weeks the cell aggregates were enlarged and fused, but there was still unoccupied space remaining in the core area. Magnification 100.
3.7. Quantification of GAG content The GAG content assay was used to detect chondrogenic ECM formation in the CS and CSP explants after 6 weeks of in vivo implantation. GAG contents were similar in the CSP explants (1.72 ± 0.38 lg mg wet wt 1) and CS explants (1.71 ± 0.68 lg mg wet wt 1). Both explants showed significantly lower GAG contents compared with native rabbit trachea (8.93 ± 1.50 lg mg wet wt 1) (Fig. 8). 3.8. Comparison of mechanical properties The stress–strain curves for the CS and CSP explants showed a steeper increase in stress with strain compared with native
trachea. Five of six samples in the CSP group failed by abrupt fracture rather than the slow tearing observed in native trachea and the CS group. Different groups did not differ significantly in UTS, elastic modulus or strain at failure (Fig. 9). Both the CS and CSP explants showed comparable ultimate tensile strengths to that of native trachea (5.70 102 ± 1.24 102 kPa for CS, 4.60 102 ± 0.59 102 kPa for CSP and 4.08 102 ± 0.96 102 kPa for native). The CS and CSP explants appeared to be stiffer than the native trachea, as exhibited by the higher elastic modulus (1.28 103 ± 5.10 102 kPa for CS, 1.10 103 ± 2.23 102 kPa for CSP, 3.46 102 ± 1.09 102 kPa for native). Moreover, when the strain at failure was compared (0.59 ± 0.10 m m 1 for CS, 0.49 ± 0.09 m m 1 for CSP, and 1.65 ± 0.53 m m 1 for native trachea), CS- and CSP-regenerated cartilage had significantly
3428
M. Zang et al. / Acta Biomaterialia 7 (2011) 3422–3431
Fig. 6. Engineered constructs harvested at 6 weeks after implantation. Gross morphology: (A) CS explants; (C) CSP explants. Cross-section view: (B) CS explants; (D) CSP explants.
Fig. 7. The histology of engineered constructs harvested 6 weeks after implantation after MOVAT pentachrome staining: (A and C) CS explants; (B and D) CSP explants. The scaffold was stained red, GAG blue and collagen fibers yellow. Ex, exterior area; In, interior area; black arrow, immature cartilage tissue; white arrow, mature cartilage tissue; black asterisk, collagen fiber boundary. Magnification: (A and B) 40; (C and D) 200. Immunostaining of collagen type II: (E) CS explants; (F) CSP explants. Collagen type II was stained brown. Magnification 200. (G) New cartilage tissue formation in CS and CSP explants based on the histomorphometric analysis. ⁄P < 0.001 vs. CS, t-test.
M. Zang et al. / Acta Biomaterialia 7 (2011) 3422–3431
Fig. 8. Biochemical assay of the GAG content of native trachea (Native) and CSP and CS explants. ⁄P = 0.01 vs. native; §P = 0.001 vs. native, one-way ANOVA.
(P < 0.01 for both) lower strains at failure as compared with native trachea. The mechanical properties of the pre-implant SFCS tubular scaffolds were 1.88 102 ± 0.18 102 kPa for UTS, 2.47 102 ± 0.23 102 kPa for elastic modulus, and 0.81 ± 0.01 m m 1 for strain at failure. The UTS of the CS-regenerated cartilage was significantly (P < 0.05) higher than the pre-implant SFCS scaffold. The CS- and CSP-regenerated cartilage was significantly (P < 0.01 for CS and P < 0.05 for CSP) stiffer than the pre-implant scaffold. Only the native trachea had a significantly (P < 0.05) higher strain at failure than pre-implant SFCS. 4. Discussion In this study, we fabricated a unique tubular scaffold geometry with the SFCS blend polymers that has smooth layered sheets
3429
intermingled with wrapped around the tubular structure. It is evident that the SFCS composition and this new structural configuration is a good conductor of chondrogenic cells and supports neochondrogenesis in vitro and in vivo. Chondrocytes proliferate and form aggregates on the scaffold surface in vitro. The presence of perichondrium results in increased production of chondrogenic matrix in vivo, although there is no effect of the perichondrium on the relative mechanical properties of the CS- or CSP-regenerated cartilage, except for relative strain at failure. The static seeding and culture technique failed to achieve homogeneous cell and ECM distribution and consequently may have affected the fate of cells within the scaffold framework under in vivo conditions. The new SFCS tubular/tracheal framework scaffolds (third generation SFCS scaffolds) resulted in a similar sheet assembly with fibrillar extensions as previously fabricated first generation SFCS scaffolds from our group [10,20], and exhibited combined features with second generation SFCS scaffolds [20] that have aligned SF polymer chains within the fibrils, in the direction of the fibril, and within the new tubular geometry, which has not been observed before with SFCS biomaterial systems. The tubular scaffolds also have significantly (P < 0.0001) higher tensile strengths, elastic moduli, and strains at failure compared with the first generation SFCS scaffolds. The better mechanical properties could be attributed to the SF polymer chain aligned along the fibril length throughout the scaffold and the tubular geometry itself. While the combination of SF and CS has been studied before for cartilage tissue engineering, tubular structures as clinical alternatives are being studied here for the first time [21,22]. The structural configuration of the matrix/scaffold is an important parameter for cell adhesion, migration, differentiation, and morphogenesis to the cartilage phenotype. SEM examination showed that the SFCS scaffolds support high cell seeding densities, since a high density of chondrocytes attached, proliferated, and differentiated on the scaffold over time. While previous studies have utilized pure SF sponges to support chondrocyte proliferation and maintain chondrogenic phenotype expression in vitro and in vivo [21,23–25], SFCS has an additional element CS, which has structural and physical
Fig. 9. Tensile testing results: (A) representative stress and strain curves of native trachea (Native) and CSP and CS explants; (B) tensile properties comparison between native trachea (Native) and CSP and CS explants.
3430
M. Zang et al. / Acta Biomaterialia 7 (2011) 3422–3431
properties similar to the GAG found in hyaline cartilage, which also supports chondrogenesis [8,26]. Cartilage-specific ECM in pure SF, polyglycolic acid (PGA), and hyalograft scaffolds have been reported to be directly related to increased cell seeding densities in vitro and in vivo [24,27,28]. Our findings suggest that a blend of SF and CS provides an excellent 3-D matrix for in vitro and in vivo chondrogenesis for up to 6 weeks, similar to other biological materials. According to Benya and Shaffer [29] monolayer culture was used to promote rapid proliferation of cells and delay differentiation, as chondrocytes can recover their differentiated phenotype when they are relocated from a 2-D to a 3-D environment. In other studies chondrocytes have been shown to lose their chondrogenic phenotype during monolayer expansion [30,31]. This study showed that chondrocytes recovered their differentiation potential shortly after they were cultured on a 3-D SFCS scaffold. At 24 h in vitro culture most of the cells acquired a spherical morphology on the SFCS surface, usually an initial visible differentiation marker for the chondrocytic phenotype [32], accompanied by early GAG and collagen type II deposition detected at 1 week, which increased with time, thereby maintaining the cellular phenotype on the SFCS surface. Aggregates of chondrocytes were observed both on the surface and within the stacked sheet architecture of the SFCS scaffold. During prenatal development and de novo tissue formation a 3-D scaffold is required to provide an environment to facilitate cell condensation and cell–cell interactions that mimic physiological processes [33]. Cell aggregation improves cell–cell contact and to some extent mimics the mesenchymal condensation step, where mesenchymal stem cells spontaneously form aggregates, produce more GAG, and show higher transcription levels of collagen type II [34]. Comparatively, this study also suggests that cell aggregation in response to the 3-D SFCS scaffold partially contributes to early chondrocytic differentiation and continuous ECM accumulation. The static cell seeding technique may have caused an uneven cell distribution through the various dimensions of the scaffold. Initially cells concentrated near the surface of the scaffold and only a few cells infiltrated into the deeper areas. Although the core zone was eventually filled with cells as cells proliferated with time, migrated inward, and only empty spaces remained between the cell aggregates. Static cultures have limitations in terms of mass transfer [35], which could explain the differences in ECM disposition between the outer and inner areas of the scaffold. Mass transfer is important for chondrocyte culture because the tissue is avascular and nutrient exchange is mainly dependent on diffusion between the matrix and the surrounding environment [7]. Diffusion of nutrients, metabolic waste, and soluble molecules might be further limited by the accumulation of ECM components in the outer areas, resulting in substantially lower cell activity in the inner areas of the SFCS scaffold [24]. The limitations of the static cell seeding technique may also explain why we were unable to produce tubular cartilaginous structures in our in vivo studies. Because of the lack of a uniform distribution of cells throughout the scaffold, newly formed cartilage and ECM are distributed more in the surface areas. Our previous in vivo studies showed that the SFSC scaffold favored host tissue in-growth and rapid remodeling in vivo[11,13]. We observed a similar remodeling phenomenon in this study, exhibited by collagen deposition around the partially degraded remnants of the SFCS scaffold. Among the remodeling tissue, some chondrocyte aggregates with immature ECM could be recognized. It is highly plausible that the unoccupied spaces in the constructs were rapidly infiltrated by remodeling cells, which left limited room for the seeded chondrocytes to fabricate new ECM. The perichondrium was successfully used to protect and stabilize the implanted tissue engineered cartilage from initial invasion
by inflammatory cells and reduce the effects of fibrous capsule formation observed in other studies [36–38]. Additionally, perichondrium is rich in progenitor cells that are capable of differentiating as cartilage tissue [39]. We have shown in the present study that wrapping in perichondrium significantly improved new cartilage formation in the cell–scaffold constructs. There was a large area of immature cartilaginous tissue undergoing remodeling with buried non-degraded scaffold found in the internal areas. Mature cartilage-like tissue encapsulated by dense collagen fibers formed in the peripheral areas of the implanted constructs, especially where the scaffold was completely degraded. An analysis of the internal and external areas indicated that gradual degradation of the scaffold resulted in the deposition of the new matrix at the same location, thereby strengthening the existing tracheal support areas and minimizing or nullifying any loss of mechanical integrity where the scaffold remained, with or without cells. Furthermore, while perichondrium acts as a reservoir for cartilage forming precursor cells, it might also act as a barrier to isolate and protect immature cartilaginous tissue in the constructs from an early host response and allow implanted chondrocytes to deposit new ECM with local colonization. We have also shown in this study that the mechanical properties of cell–scaffold constructs were similar to native trachea, although they appeared to be stiffer and less extensible. Clinically, however, reduced extensibility is not a major concern. The rigidity of a tubular structure in maintaining an open airway is the key mechanical property of tissue engineered trachea, which comparatively was equivalent to native trachea, thus showing the potential of the SFCS for tracheal tissue reconstructive applications. It was reported that the development of tissue engineered cartilage could be enhanced by extending the precultivation time and employing advanced bioreactors [28]. Prolonged in vitro culture would allow extensive ECM formation which will protect the seeded cells and also decrease post-implantation inflammation [28,40], although the in vivo remodeling time by host derived cells after invasion of the construct may be prolonged due to the increased deposition of ECM by in vitro seeded cells. Bioreactors represent attractive tools to accelerate the development of the required biochemical and mechanical properties of engineered tissues, which also influence the type of ECM molecules excreted into the cellular microenvironment due to local mass transfer and physical stimuli effects [41]. In our future studies we will combine these factors to improve homogeneous cell seeding and extensive ECM formation in the scaffold. 5. Conclusions Cartilage formation is key to the mechanical integrity of the native trachea. In this study we focused on the reconstruction of tracheal defects in cancer patients who needed a mechanically sound tracheal graft for reconstruction of the tumor-resected defect site. Thus an approach was devised that would translate to the clinic, such that a tracheal graft could be constructed by subcutaneously implanting a biomaterial in the patient and then using the patients own cells to regenerate the tracheal graft, which thus consists of native cartilage, as opposed to tracheal allotransplantation which involves immunosuppressive therapies [42]. An SFCS scaffold provided an excellent surface for cell adhesion at high densities and a 3-D environment for chondrocyte proliferation, cell–cell contact and differentiation under in vitro conditions prior to in vivo implantation and cartilage regeneration. While wrapping with perichondrium improved neochondrogenesis in vivo, the static seeding and culture technique resulted in a heterogeneous cell and ECM distribution in vitro and in vivo. While this study shows that the in vivo tracheal reconstruction animal model and the SFCS construct works well to promote neochondrogenesis improvements
M. Zang et al. / Acta Biomaterialia 7 (2011) 3422–3431
are needed to produce a more functional and dynamic in vitro cell seeding and culture system. Acknowledgements This study was supported by the Kyte Research Fund from the Department of Plastic Surgery at the University of Texas M.D. Anderson Cancer Center. We thank Kenneth Dunner Jr. of the High Resolution Electron Microscopy Facility at M.D. Anderson Cancer Center for his assistance with the SEM, and Arnie Jimenez of VelLabs for assistance in preparing the histological sections. M.Z. was financially supported by the China Scholarship Council. References [1] Matloub HS, Yu P. Engineering a composite neotrachea in a rat model. Plast Reconstr Surg 2006;117(1):123–8. [2] Grillo HC. Tracheal replacement: a critical review. Ann Thorac Surg 2002;73(6):1995–2004. [3] Steger V, Hampel M, Trick I, Muller M, Walles T. Clinical tracheal replacement: transplantation, bioprostheses and artificial grafts. Expert Rev Med Devices 2008;5(5):605–12. [4] Baiguera S, Birchall MA, Macchiarini P. Tissue-engineered tracheal transplantation. Transplantation 2010;89(5):485–91. [5] Go T, Jungebluth P, Baiguero S, Asnaghi A, Martorell J, Ostertag H, et al. Both epithelial cells and mesenchymal stem cell-derived chondrocytes contribute to the survival of tissue-engineered airway transplants in pigs. J Thorac Cardiovasc Surg 2010;139(2):437–43. [6] Kojima K, Bonassar LJ, Roy AK, Mizuno H, Cortiella J, Vacanti CA. A composite tissue-engineered trachea using sheep nasal chondrocyte and epithelial cells. FASEB J 2003;17(8):823–8. [7] MacIntosh AC, Kearns VR, Crawford A, Hatton PV. Skeletal tissue engineering using silk biomaterials. J Tissue Eng Regen Med 2008;2(2/3):71–80. [8] Suh JK, Matthew HW. Application of chitosan-based polysaccharide biomaterials in cartilage tissue engineering: a review. Biomaterials 2000;21(24):2589–98. [9] Wang Y, Kim HJ, Vunjak-Novakovic G, Kaplan DL. Stem cell-based tissue engineering with silk biomaterials. Biomaterials 2006;27(36):6064–82. [10] Gobin AS, Froude VE, Mathur AB. Structural and mechanical characteristics of silk fibroin and chitosan blend scaffolds for tissue regeneration. J Biomed Mater Res 2005;74(3):465–73. [11] Altman AM, Yan Y, Matthias N, Bai X, Rios C, Mathur AB, et al. IFATS collection: human adipose-derived stem cells seeded on a silk fibroin–chitosan scaffold enhance wound repair in a murine soft tissue injury model. Stem Cells 2009;27(1):250–8. [12] Altman AM, Gupta V, Rios CN, Alt EU, Mathur AB. Adhesion, migration and mechanics of human adipose-tissue-derived stem cells on silk fibroin– chitosan matrix. Acta Biomater 2010;6(4):1388–97. [13] Gobin AS, Butler CE, Mathur AB. Repair and regeneration of the abdominal wall musculofascial defect using silk fibroin–chitosan blend. Tissue Eng 2006;12(12):3383–94. [14] Rios CN, Skoracki RJ, Miller MJ, Satterfield WC, Mathur AB. In vivo bone formation in silk fibroin and chitosan blend scaffolds via ectopically grafted periosteum as a cell source: a pilot study. Tissue Eng Part A 2009;15(9):2717–25. [15] Henderson JH, Welter JF, Mansour JM, Niyibizi C, Caplan AI, Dennis JE. Cartilage tissue engineering for laryngotracheal reconstruction: comparison of chondrocytes from three anatomic locations in the rabbit. Tissue Eng 2007;13(4):843–53. [16] Kamil SH, Eavey RD, Vacanti MP, Vacanti CA, Hartnick CJ. Tissue-engineered cartilage as a graft source for laryngotracheal reconstruction: a pig model. Arch Otolaryngol Head Neck Surg 2004;130(9):1048–51. [17] Ohlsen L, Nordin U. Tracheal reconstruction with perichondrial grafts. Scand J Plast Reconstr Surg Hand Surg 1976;10(2):135–45. [18] Clevens RA, Esclamado R, Naficy S. Vascularized auricular perichondrium in airway reconstruction: the effects of stenting with and without a mucosal graft. The Laryngoscope 1995;105(10):1043–8.
3431
[19] Naficy S, Esclamado RM, Clevens RA. Reconstruction of the rabbit trachea with vascularized auricular perichondrium. Ann Otol Rhinol Laryngol 1996;105(5):356–62. [20] Gupta V, Davis G, Gordon A, Altman AM, Reece GP, Gascoyne PR, et al. Endothelial and stem cell interactions on dielectrophoretically aligned fibrous silk fibroin–chitosan scaffolds. J Biomed Mater Res A 2010;94A:515–23. [21] Seda Tigli R, Ghosh S, Laha MM, Shevde NK, Daheron L, Gimble J, et al. Comparative chondrogenesis of human cell sources in 3D scaffolds. J Tissue Eng Regen Med 2009;3(5):348–60. [22] Silva SS, Motta A, Rodrigues MT, Pinheiro AF, Gomes ME, Mano JF, et al. Novel genipin-cross-linked chitosan/silk fibroin sponges for cartilage engineering strategies. Biomacromolecules 2008;10(2):764–74. [23] Aoki H, Tomita N, Morita Y, Hattori K, Harada Y, Sonobe M, et al. Culture of chondrocytes in fibroin–hydrogel sponge. Biomed Mater Eng 2003;13(4):309–16. [24] Wang Y, Blasioli DJ, Kim HJ, Kim HS, Kaplan DL. Cartilage tissue engineering with silk scaffolds and human articular chondrocytes. Biomaterials 2006;27(25):4434–42. [25] Shangkai C, Naohide T, Koji Y, Yasuji H, Masaaki N, Tomohiro T, et al. Transplantation of allogeneic chondrocytes cultured in fibroin sponge and stirring chamber to promote cartilage regeneration. Tissue Eng 2007;13(3):483–92. [26] Kim IY, Seo SJ, Moon HS, Yoo MK, Park IY, Kim BC, et al. Chitosan and its derivatives for tissue engineering applications. Biotechnol Adv 2008;26(1):1–21. [27] Mahmoudifar N, Doran PM. Effect of seeding and bioreactor culture conditions on the development of human tissue-engineered cartilage. Tissue Eng 2006;12(6):1675–85. [28] Moretti M, Wendt D, Dickinson SC, Sims TJ, Hollander AP, Kelly DJ, et al. Effects of in vitro preculture on in vivo development of human engineered cartilage in an ectopic model. Tissue Eng 2005;11(9/10):1421–8. [29] Benya PD, Shaffer JD. Dedifferentiated chondrocytes reexpress the differentiated collagen phenotype when cultured in agarose gels. Cell 1982;30(1):215–24. [30] von der Mark K, Gauss V, von der Mark H, Muller P. Relationship between cell shape and type of collagen synthesised as chondrocytes lose their cartilage phenotype in culture. Nature 1977;267(5611):531–2. [31] Holtzer H, Abbott J, Lash J, Holtzer S. The loss of phenotypic traits by differentiated cells in vitro, I. Dedifferentiation of cartilage cells. Proc Natl Acad Sci USA 1960;46(12):1533–42. [32] Archer CW, Rooney P, Wolpert L. Cell shape and cartilage differentiation of early chick limb bud cells in culture. Cell Differ 1982;11(4):245–51. [33] Wescoe KE, Schugar RC, Chu CR, Deasy BM. The role of the biochemical and biophysical environment in chondrogenic stem cell differentiation assays and cartilage tissue engineering. Cell Biochem Biophys 2008;52(2): 85–102. [34] Ghosh S, Laha M, Mondal S, Sengupta S, Kaplan DL. In vitro model of mesenchymal condensation during chondrogenic development. Biomaterials 2009;30(33):6530–40. [35] Thevenot P, Nair A, Dey J, Yang J, Tang L. Method to analyze three-dimensional cell distribution and infiltration in degradable scaffolds. Tissue Eng Part A 2008;14(4):319–31. [36] Vinatier C, Mrugala D, Jorgensen C, Guicheux J, Noel D. Cartilage engineering: a crucial combination of cells, biomaterials and biofactors. Trends Biotechnol 2009;27(5):307–14. [37] Anderson JM, Rodriguez A, Chang DT. Foreign body reaction to biomaterials. Semin Immunol 2008;20(2):86–100. [38] Haisch A, Groger A, Gebert C, Leder K, Ebmeyer J, Sudhoff H, et al. Creating artificial perichondrium by polymer complex membrane macroencapsulation: immune protection and stabilization of subcutaneously transplanted tissueengineered cartilage. Eur Arch Otorhinolaryngol 2005;262(4):338–44. [39] Togo T, Utani A, Naitoh M, Ohta M, Tsuji Y, Morikawa N, et al. Identification of cartilage progenitor cells in the adult ear perichondrium: utilization for cartilage reconstruction. Lab Invest 2006;86(5):445–57. [40] Luo X, Zhou G, Liu W, Zhang WJ, Cen L, Cui L, et al. In vitro precultivation alleviates post-implantation inflammation and enhances development of tissue-engineered tubular cartilage. Biomed Mater 2009;4(2):025006. [41] Concaro S, Gustavson F, Gatenholm P. Bioreactors for tissue engineering of cartilage. Adv Biochem Eng Biotechnol 2008;112:125–43. [42] Delaere P, Vranckx J, Verleden G, De Leyn P, van Raemdonck D. Tracheal allotransplantation after withdrawal of immunosuppressive therapy. N Engl J Med 2010;362(2):138–45.