bs_bs_banner

© 2013 Wiley Periodicals, Inc. and International Center for Artificial Organs and Transplantation

Feasibility of Autologous Bone Marrow Mesenchymal Stem Cell–Derived Extracellular Matrix Scaffold for Cartilage Tissue Engineering *†‡Cheng Tang, *†‡Yan Xu, *†‡Chengzhe Jin, **††‡‡Byoung-Hyun Min, §Zhiyong Li, §Xuan Pei, and *†‡Liming Wang *Department of Orthopaedic Surgery; †Cartilage Regeneration Center; ‡China-Korea United Cell Therapy Center, Nanjing First Hospital, Nanjing Medical University; §School of Biological Science & Medical Engineering, Southeast University, Nanjing, China; **Cell Therapy Center, Ajou University Hospital; ††Department of Orthopaedic Surgery, School of Medicine; and ‡‡Department of Molecular Science and Technology, Ajou University, Suwon, South Korea

Abstract: Extracellular matrix (ECM) materials are widely used in cartilage tissue engineering. However, the current ECM materials are unsatisfactory for clinical practice as most of them are derived from allogenous or xenogenous tissue. This study was designed to develop a novel autologous ECM scaffold for cartilage tissue engineering. The autologous bone marrow mesenchymal stem cell–derived ECM (aBMSC-dECM) membrane was collected and fabricated into a three-dimensional porous scaffold via cross-linking and freeze-drying techniques. Articular chondrocytes were seeded into the aBMSCdECM scaffold and atelocollagen scaffold, respectively. An in vitro culture and an in vivo implantation in nude mice model were performed to evaluate the influence on engineered cartilage. The current results showed that the

aBMSC-dECM scaffold had a good microstructure and biocompatibility. After 4 weeks in vitro culture, the engineered cartilage in the aBMSC-dECM scaffold group formed thicker cartilage tissue with more homogeneous structure and higher expressions of cartilaginous gene and protein compared with the atelocollagen scaffold group. Furthermore, the engineered cartilage based on the aBMSC-dECM scaffold showed better cartilage formation in terms of volume and homogeneity, cartilage matrix content, and compressive modulus after 3 weeks in vivo implantation. These results indicated that the aBMSCdECM scaffold could be a successful novel candidate scaffold for cartilage tissue engineering. Key Words: Autologous bone marrow—Mesenchymal stem cell— Extracellular matrix scaffold—Cartilage tissue engineering.

Articular cartilage plays a key role in absorbing repetitive load-bearing strain and various mechanical shocks. Unfortunately, injuries of articular cartilage have been found in more than 50% of patients undergoing knee arthroscopy (1). The repair of cartilage defect is of great clinical concern, because injured articular cartilage has a limited intrinsic ability to

self-heal and is frequently a precursor of degenerative joint disease. Among the various therapeutic strategies at hand for the treatment of articular cartilage, autologous chondrocyte implantation (ACI) is a commercially available cell-based technique and was first reported by Brittberg in 1994 (2). However, ACI presents some limitations including uneven distribution of implanted chondrocytes and possible cell leakage. As an advanced method of ACI, matrix-associated chondrocyte implantation (MACI) can effectively solve these problems. In MACI technique, autologous chondrocytes are seeded directly into a biodegradable type I/III collagen scaffold, and then the cell–scaffold construct is inserted directly into the defect site (3). Utilizing scaffold material could not only maintain cells in situ but also provide a favorable environment for cell proliferation and matrix

doi:10.1111/aor.12130 Received January 2013; revised April 2013. Address correspondence and reprint requests to Professor Liming Wang, Department of Orthopaedic Surgery, Nanjing First Hospital, Nanjing Medical University, Nanjing 210006, China. E-mail: [email protected] and Professor Chengzhe Jin, Department of Orthopaedic Surgery, Nanjing First Hospital, Nanjing Medical University, Nanjing 210006, China. E-mail: [email protected].

Artificial Organs 2013, 37(12):E179–E190

E180

C. TANG ET AL.

production and guide the engineered cartilage toward regenerating a more hyaline cartilage-like tissue. Although there are some drawbacks of this technique, such as the need for two surgeries, the relatively long recovery time, and the slow tissue maturation time of the implanted chondrocytes (4), it is theoretically a promising technology to improve ACI for cartilage repair. Many studies nowadays are making efforts to optimize MACI through fabricating a structural and functional scaffold (5,6). Natural extracellular matrix (ECM) scaffolds have received significant attention for their potential therapeutic application. Many ECM scaffolds, such as porcine small intestinal submucosa, fetal bovine skin, and bovine pericardium, have already been commercially used for the reconstruction of urethra, cranial, rotator cuff, and spinal dura (7,8). However, most of the current ECM scaffolds are derived from allogenous or xenogenous tissue. The implantation of these scaffolds adds risks of pathogen transmission, undesirable inflammation, and immunological reaction in clinical practice (9). In our previous studies, a porcine articular chondrocyte-derived ECM scaffold provided a better biological environment for cartilage tissue engineering than polyglycolide (PGA) scaffold (10,11).Although the xenogenous ECM scaffold had been fabricated in those studies, the use of cell-derived ECM scaffold enabled us to create a novel autologous cell-derived ECM scaffold for cartilage tissue engineering. Bone marrow mesenchymal stem cells (BMSCs) not only have the capacity to differentiate into a variety of mesenchymal tissue cells but also have a high potential of proliferation and ECM secretion. Furthermore, the cells can be easily obtained from several areas, such as the femur, tibia, and iliac crest (12,13). In this study, the autologous bone marrow mesenchymal stem cell–derived ECM (aBMSCdECM) membrane was collected and fabricated into a three-dimensional porous scaffold via cross-linking and freeze-drying techniques (10,11). We aimed to evaluate whether the aBMSC-dECM scaffold could provide a favorable environment for chondrocytes and promote hyaline cartilage-like tissue regeneration by investigating the morphology and viability of chondrocytes, matrix production, cartilaginous gene and protein expression in vitro and cartilage regeneration in vivo. MATERIALS AND METHODS Isolation and identification of BMSCs The use of animals in this experiment was approved by the Institutional Animal Experiment Artif Organs, Vol. 37, No. 12, 2013

Committee of Nanjing Medical University. The experimental protocol met the National Institutes of Health guidelines. Five New Zealand white rabbits, 2 weeks old, were euthanized by overdose injection of pentobarbital. Isolation and culture of BMSCs were performed as reported previously (14). Briefly, the bone marrow was flushed out from tibias and femurs with phosphate-buffered saline (PBS). The mononuclear cells (MNCs) were obtained by density-gradient centrifugation with Lymphoprep (AXIS-SHIELD, Oslo, Norway) at 800 × g for 20 min and resuspended with a culture medium containing Dulbecco’s modified Eagle’s medium (DMEM; Gibco, Grand Island, NY, USA), 10% fetal bovine serum (FBS; Gibco), 100 U/mL penicillin, and 100 μg/mL streptomycin (Gibco). The MNCs were seeded at a density of 1.0 × 105 cells/cm2 and were cultured at 37°C, 5% CO2, and 95% humidity. Two days later, the nonadherent cells were removed while the adherent BMSCs were allowed to culture further in order to grow into the aBMSCdECM membrane. During the culture, some BMSCs (passage 1) were collected for cell identification. A trilineage-induced differentiation experiment was performed as reported previously, to identify the multiple differentiation potential of BMSCs, which included chondrogenesis, osteogenesis, and adipogenesis (13). Preparation of aBMSC-dECM scaffold Once the primary BMSCs reached 70–80% confluence, 50 μg/mL L-ascorbic acid (Sigma, St. Louis, MO, USA) was added into the culture medium to stimulate ECM deposition. The BMSCs were cultured under the same conditions as above. The culture medium was changed three times a week up to 4 weeks to allow the formation of the aBMSCdECM membrane. After 4 weeks, the membrane was separated carefully from the bottom of the culture plate with a cell spatula and was dissected with scissors. The membrane was added to a polyethylene mold and was stored at −80°C for 3 days. The solidified membrane (in the mold) was then freeze-dried (Sihuan, Beijing, China) at −70°C under 1 Pa for 48 h. In order to improve mechanical strength, the freezedried specimen was cross-linked by 50 mM 1-ethyl-3[3-dimethylaminopropyl] carbodiimide hydrochloride and 50 mM N-hydroxysuccinimide for 24 h, washed with sodium phosphate dibasic and deionized water, and freeze-dried again. The cylindrical form of aBMSC-dECM scaffold was obtained by cutting with a biopsy punch (6 mm in diameter) and trimming off the surface layer by approximately 2 mm in thickness with a clean razor blade.

CARTILAGE TISSUE ENGINEERING Characterization of aBMSC-dECM scaffold The color and size of aBMSC-dECM scaffold was observed by gross view. The microstructure was assayed by scanning electron microscope (SEM; Hitachi, Tokyo, Japan) as described previously (10). The pore size of the aBMSC-dECM scaffold was determined according to the SEM images by taking the average of 50 measurements (15). The density and porosity of the scaffold were determined by ethanol inhalation, as reported previously (15). Briefly, all samples (n = 5) were cut into quadrate pieces with about 1 cm of height, length, and width. All sizes were precisely measured by a vernier caliper, and the volume (Vw) of each sample was calculated. The scaffold sample (weight Ws) was immersed into a density bottle, which was filled with ethanol (density ρe) and weighed (W1). After 30 min, all of the displaced ethanol was cleaned away carefully. The density bottle was again weighed (W2), and the volume of the scaffold skeleton (Vs), the density (ρs), and porosity (ε) of scaffold were calculated according to the following formulas: Vs = (W1 − W2 + Ws)/ρe, ρs = Ws/Vw, ε = (1 − Vs/Vw) ×100%. The compressive modulus of aBMSC-dECM scaffold (n = 5, 6 mm in diameter, 2 mm in thickness) was calculated based on the slope of a stress–strain curve with a universal testing machine (Instron 5943, Norwood, MA, USA) as reported previously (16). In vitro study Isolation and culture of articular chondrocytes The isolation and culture of articular chondrocytes was described elsewhere (11). Articular cartilage (from the same rabbit as above) was collected from the femoral condyles. The cartilage pieces were carefully isolated and minced to 1 × 1 × 1 mm3, washed with PBS, and digested with 0.2% type II collagenase (Gibco) for 4 h. The isolated chondrocytes were centrifuged at 600 × g for 10 min. The cell pellet was washed twice with PBS and plated (1.0 × 104 cells/ cm2) in DMEM supplemented with 10% FBS, 100 U/mL penicillin, and 100 μg/mL streptomycin. When the cells reached 70–80% confluence, they were passaged for subcultures. The passage 1 chondrocytes were pooled, aliquoted, and frozen in freezing medium (DMEM with 20% FBS and 10% dimethyl sulfoxide). After the preparation of aBMSC-dECM scaffolds, autologous chondrocyte aliquots were thawed and subcultured. The scaffolds were sterilized in 75% ethanol for 10 h and were then washed with PBS three times and immersed in culture media overnight to remove all traces of ethanol solution.The passage 3

E181

chondrocyte suspension was seeded dynamically on scaffolds at a cell density of 3 × 106/mL for 90 min on a nutating mixer, as described previously (11). The cell–scaffold constructs were then transferred to 6-well plates and cultured for 1, 2, and 4 weeks for analysis. Atelocollagen scaffold (Koraku, Tokyo, Japan) was used as a control. The atelocollagen scaffold was composed of bovine Achilles tendon fibers with 96–97% porosity. It was punched into a disc shape 6 mm in diameter, and then cut 2 mm in thickness. Morphology and viability of chondrocytes in cell–scaffold constructs The morphology of chondrocytes in cell–scaffold constructs (n = 5) was observed by SEM at 0 and 48 h after seeding. The viability of chondrocytes was evaluated by a Live/Dead cell staining kit (Biovision, Mountain View, CA, USA) at 48 h after seeding. Briefly, the cell–scaffold constructs (n = 5) were incubated in the assay reagents (2 μM calcein AM and 4 μM ethidium homodimer-1) for 15 min. A fluorescence microscope with a band-pass filter (FITC and rhodamine) was then used for the detection of stained live and dead cells. Live cells were shown in green, and dead cells in red. Histological and immunohistochemical analysis The engineered cartilage (n = 6) harvested from each group at different time points was evaluated by histological staining with Safranin O and immunohistochemical analysis of type II collagen. The samples were fixed with 4% formaldehyde for 24 h, dehydrated and embedded in paraffin, and then sectioned to 4 μm. The sections were stained with Safranin O. For immunohistochemical analysis, the section was treated sequentially with 3% H2O2 and proteinase K and was incubated with mouse monoclonal antirabbit type II collagen antibody (1:100, Acris, Herford, Germany) for 90 min at room temperature.The section was then incubated sequentially with biotinylated secondary antibody against mouse IgG (Maixin-Bio, Fuzhou, China) for 10 min and peroxidase-conjugated streptavidin solution. The section was finally counterstained with hematoxylin and mounted for microscopic observation (Olympus, Tokyo, Japan). Real-time polymerase chain reaction (PCR) The transcriptional level of engineered cartilage (n = 6) was analyzed by real-time PCR. Total RNA was extracted with Trizol reagent (Gibco). Quantitative PCR reaction was then performed with two cartilaginous gene primers (COL2A1 and ACAN) Artif Organs, Vol. 37, No. 12, 2013

E182

C. TANG ET AL.

using a one-step quantitative PCR kit (Toyobo, Osaka, Japan) and an ABI 7500 fast real-time PCR system (Applied Biosystems, Foster City, CA, USA). Glyceraldehyde-3-phephatase dehydrogenase (GAPDH) was used as a housekeeping gene. The sequences of primers used in this analysis are as follows: COL2A1 forward, 5′-CAGGCAGAGGCA GGAAACTAAC-3′, COL2A1 reverse, 5′-CAGAG GTGTTTGACACGGAGTAG-3′; ACAN forward, 5′-ATGGCTTCCACCAGTGCG-3′, ACAN reverse, 5′-CGGATGCCGTAGGTTCTCA-3′; GAPDH forward, 5′-CGTCTGCCCTATCAACTTTCG-3′, GAPDH reverse, 5′-CGTTTCTCAGGCTCCCTCT3′. The fluorescence intensity was recorded for 40 cycles under the conditions of 15 s at 90°C and 60 s at 60°C. Finally, the gene expression value of individual samples was calculated in terms of the GAPDH expression by the 2−éCt method. Each reaction was performed in triplicate (17). Western blotting Synthesis of type II collagen in engineered cartilage (n = 4) was analyzed by Western blotting. Total proteins were extracted with lysis buffer (40 mM Tris-HCl, 120 mM NaCl, 0.5% Nonidet P-40 (Sigma-Aldrich, St Louis, MO), 2 mg/mL Aprotinin (Sigma-Aldrich, St Louis, MO), 100 mg/mL phenylmethylsulfonyl fluoride) and measured by the BCA method. The 40 μg of total proteins were loaded and separated by 8% sodium dodecyl sulfatepolyacrylamide gel electrophoresis. The separated proteins were transferred to a polyacrylamide difluoride membrane (Millipore, Billerica, MA, USA) with a semidry blot apparatus (Bio-Rad, Hercules, CA, USA). The blotting membrane was blocked with 5% milk powder in Tris-buffered saline for 1 h, incubated overnight with mouse antirabbit type II collagen antibody (1:500, Acris) and then incubated with peroxidase-conjugated goat antimouse secondary antibody (1:6000, Icllab, Newberg, OR, USA). The membrane was visualized with an ECL kit (Thermo Scientific, Rockford, IL, USA) and a Chemidoc XRS system (Bio-Rad). The intensity of the immunoreactive band was quantitatively assessed by using an Image Gauge (V 3.11; Fuji Film, Tokyo, Japan). In vivo study Implantation of the cell–scaffold constructs into nude mice After being cultured for 1 week in vitro, an additional 48 cell–scaffold constructs of the aBMSCdECM scaffold group and the atelocollagen scaffold Artif Organs, Vol. 37, No. 12, 2013

group were prepared for in vivo implantation. Sixweek-old male nude mice (n = 12/group) were anesthetized with chloral hydrate solution (0.4 mg/g). The back skin of nude mice was incised under sterile conditions. Four constructs were subcutaneously implanted into each mouse. Four mice of each group were sacrificed at 1, 2, and 3 weeks after implantation. Therefore, 16 engineered cartilages could be harvested every week in each group for further analysis. Gross observation, volume measurement, and histological analysis The gross morphology of engineered cartilage (n = 6) was observed in terms of shape and color. The firmness of engineered cartilage was evaluated by the pinch test (10). The volume was measured using three-dimensional images obtained by charge coupled device camera and a computer vision system that was described previously (11). The engineered cartilage was then paraffin embedded, sectioned, and stained with Safranin O and immunohistochemical analysis of type II collagen. Chemical assay The DNA and glycosaminoglycan (GAG) content of the engineered cartilage (n = 6) was measured by chemical assay as described previously (10). Briefly, the engineered cartilage was dried at 37°C for 48 h and was then digested with a papain solution (5 mM L-cysteine, 100 mM Na2HPO4, 5 mM ethylenediaminetetraacetic acid, 125 μg/mL papain, pH 6.4) at 60°C for 24 h, followed by centrifugation at 12 000 × g for 10 min. The supernatant was used for chemical assay. Total DNA content was determined by Quit-iT dsDNA kit (Invitrogen, Eugene, OR, USA) using salmon testes DNA for a standard curve (Sigma). The supernatant was reacted with the Hoechst dye 33 258 for 30 min in the dark. The intensity of fluorescence was screened with a 96-well plate reader (excitation at 360 nm and emission at 460 nm, Perkin-Elmer LS-55,Waltham, MA, USA).The GAG content was measured by the dimethylmethylene blue (DMB) colorimetric assay. The supernatant was mixed with DMB solution to bind GAG. The GAG– dye complexes were collected by centrifugation at 12 000 × g for 10 min. The GAG content was calculated based on a standard curve of sulfate chondroitin from shark cartilage (Sigma) at 530 nm on a Benchmark plus microplate spectrophotometer (Bio-Rad, Tokyo, Japan). Native cartilage of a 2-week-old rabbit was used for the positive control. Compressive modulus testing A universal testing machine (Instron 5943) was used for compressive modulus testing, as reported

CARTILAGE TISSUE ENGINEERING previously (16). The engineered cartilage (n = 4) was cut into a uniform disk shape using a biopsy punch. A constant compressive strain rate of 0.01 mm/s at a free load of 0.01 N was applied, and a stress–strain curve was generated. The slope of the linear stress–strain curve from 10 to 16% strain was defined as the compressive modulus. Native cartilage of a 2-week-old rabbit was used for the positive control.

Statistical analysis All the data were expressed as the mean ± standard deviation (SD). The effect of time on volume, DNA content, GAG content, GAG/DNA ratios, and compressive modulus of the in vivo engineered cartilage at the three time points (1, 2, and 3 weeks) in the aBMSC-dECM scaffold group and the atelocollagen scaffold group, respectively, was evaluated by one-way analysis of variance (ANOVA). The differences in DNA content, GAG content, GAG/DNA ratios, and compressive modulus among the aBMSCdECM scaffold group, the atelocollagen scaffold group, and normal cartilage group at each time point were also analyzed by one-way ANOVA followed by post hoc test (Student–Newman–Keuls). The volume differences of the in vivo engineered cartilage between the aBMSC-dECM scaffold group and the atelocollagen scaffold group at each time point, respectively, were compared by Student’s t-test. The P value was two sided, and values of less than 0.05 were considered statistically significant. All the statistical analysis was carried out using SPSS 13.0 (SPSS Inc., Chicago, IL, USA).

E183 RESULTS

Identification of BMSCs The results showed that the BMSCs were successfully differentiated into chondrocytes as indicated by immunohistochemical analysis of type II collagen (Fig. 1A), osteocytes as indicated by Von Kossa staining (Fig. 1B), and adipocytes as indicated by Oil Red O staining (Fig. 1C). Characterization of aBMSC-dECM scaffold The aBMSC-dECM scaffold was whitish in color, sponge-like and disk-shaped, with a diameter of 6 mm and a thickness of 2 mm (Fig. 2A). SEM revealed that the scaffold demonstrated uniform porosity and a highly interconnected structure, with a pore size of 304.4 ± 108.2 μm (Fig. 2B,C).The scaffold possessed a density of 28.7 ± 1.4 mg/mL, porosity of 93.3 ± 4.5%, and a compressive modulus of 6.8 ± 1.5 KPa. In vitro study Morphology and viability of chondrocytes in the aBMSC-dECM scaffold As shown in Fig. 2E, lots of round-shaped chondrocytes were evenly distributed in the aBMSCdECM scaffold after seeding. Forty-eight hours later, the cells maintained their round shape and presented efficient adhesion to the wall of the scaffold (Fig. 2F). Live/Dead cytotoxicity assay showed that most chondrocytes in the aBMSC-dECM scaffold were alive in the fluorescence image (Fig. 2D).

FIG. 1. Characteristics of BMSCs were identified by multiple differentiation experiments. (A) Immunohistochemical analysis of type II collagen demonstrated the expression of type II collagen in the pericellular region (black arrows) of induced BMSC pellets (×400). (B) Von Kossa staining showed calcium deposition (white arrows) in induced BMSCs (×40). (C) Oil Red O staining showed the accumulation of neutral lipid vacuoles (black arrows) in induced BMSCs (×100). Artif Organs, Vol. 37, No. 12, 2013

E184

C. TANG ET AL.

FIG. 2. Structural characterization and biocompatibility of aBMSC-dECM scaffold. (A) The gross observation showed that the aBMSCdECM scaffold was white in color, sponge-type and disk-shaped, with a diameter of 6 mm and a thickness of 2 mm (the scale bar unit is millimeters). The SEM image revealed that a uniformly porous and highly interconnected structure was observed in the central (B) and peripheral (C) region of the aBMSC-dECM scaffold (×100). (D) Live/Dead cytotoxicity assay was performed to examine viability of chondrocytes in the aBMSC-dECM scaffold at 48 h after seeding (×400). The morphology of chondrocytes in the aBMSC-dECM scaffold was taken by SEM at 0 (E) and 48 h (F) after seeding (×1000).

Histological and immunohistochemical analysis The engineered cartilage formed a relatively homogeneous structure in the aBMSC-dECM scaffold group, in which hyaline cartilage-like tissue gradually increased in both outer and inner regions. Increasing lacuna structure, strong expression of sulfate proteoglycan (Fig. 3C,G,K), and type II collagen (Fig. 3D,H,L) were observed in relatively compact hyaline cartilage-like tissue. However, in the atelocollagen scaffold group, the engineered cartilage showed obviously heterogeneous structure (Fig. 3A,B,E,F,I,J). The hyaline cartilage-like tissue was only distributed in the outer region throughout all time points. The engineered cartilage of the aBMSC-dECM scaffold group formed a much thicker hyaline cartilage-like tissue layer than that of the atelocollagen scaffold group. The wall of atelocollagen scaffold was not changed during in vitro culture (Fig. 3A,E,I), whereas only a little debris could be observed in the aBMSC-dECM scaffold group after 4 weeks (Fig. 3K).

aBMSC-dECM scaffold group than those in the atelocollagen scaffold group. It is worth mentioning that the COL2A1 gene expression contiguously increased in the aBMSC-dECM scaffold group, whereas it was sharply downregulated in the atelocollagen scaffold group at 4 weeks (Fig. 4A). The expression of ACAN, the gene encoding for aggrecan, increased over time in the aBMSC-dECM scaffold group but decreased in the atelocollagen scaffold at 4 weeks (Fig. 4B). Western blot Western blot analysis of the engineered cartilage (Fig. 4C,D) demonstrated that the expression of type II collagen, a special protein abundant in articular cartilage, was constantly increased in the aBMSCdECM scaffold group during in vitro culture. Notably, the intensity of the immunoreactive band sharply decreased in the atelocollagen scaffold group at 4 weeks. In vivo study

Real-time PCR During in vitro culture, the expression of COL2A1, the gene encoding for type II collagen, were upregulated to a twofold higher degree in the Artif Organs, Vol. 37, No. 12, 2013

In vivo fate of engineered cartilage During in vivo implantation, the engineered cartilage in the aBMSC-dECM scaffold group presented a

CARTILAGE TISSUE ENGINEERING

E185

FIG. 3. Expression of sulfate proteoglycan and type II collagen of the in vitro engineered cartilage is shown. The expression of sulfate proteoglycan and type II collagen was examined in engineered cartilage by histological staining with Safranin O (A, C, E, G, I, and K) and immunohistochemical analysis (B, D, F, H, J, L). The engineered cartilage showed typical and homogeneous cartilage structure with abundant sulfate proteoglycan (C, G, and K) and type II collagen (D, H, and L) deposition in the aBMSC-dECM scaffold group, while that in the atelocollagen scaffold group showed relatively heterogeneous cartilage structure with a relatively thinner morphology (A, B, E, F, I, and J). A–L: ×100.

smooth, white-colored and hyaline cartilage-like tissue (Fig. 5B,D,F) with homogeneous and mature cartilage microstructure (Fig. 5J,K,N,O,R,S), as well as cartilage-specific ECM deposition. The relatively mature hyaline cartilage-like tissue was also

observed in the atelocollagen scaffold group by gross view (Fig. 5A,C,E) and histological analysis (Fig. 5H,I,L,M,P,Q). The volume of engineered cartilage was 24.9 ± 1.9, 30.1 ± 1.5, and 44.1 ± 2.8 mm3 in the aBMSC-dECM

FIG. 4. Gene and protein expression of the in vitro engineered cartilage are shown. The expression level value of cartilaginous genes, including COL2A1 (A) and ACAN (B), was calculated relative to GAPDH expression. The expression of type II collagen was evaluated by Western blot (C) and was then normalized to the relative expression level of β-actin (D). Results are presented as means ± SD.

Artif Organs, Vol. 37, No. 12, 2013

E186

C. TANG ET AL.

FIG. 5. Gross observation and histological analysis of the in vivo engineered cartilage is shown. The morphology of engineered cartilage (A–F) was presented at 1, 2, and 3 weeks after implantation (the scale bar unit is millimeters). The volume of engineered cartilage was significantly higher in the aBMSC-dECM scaffold group than in the atelocollagen scaffold group at each time point (G). The engineered cartilage was stained with Safranin O (H, J, L, N, P, and R) and immunohistochemical analysis of type II collagen (I, K, M, O, Q, and S) in the two groups. Results are presented as means ± SD. **P < 0.01; ***P < 0.001. H–S: ×100.

scaffold group and 19.6 ± 1.8, 15.5 ± 1.6, and 10.5 ± 2.2 mm3 in the atelocollagen scaffold group at 1, 2, and 3 weeks after in vivo implantation, respectively. Statistically, the volume of engineered cartilage markedly increased with time in the aBMSC-dECM scaffold group (P = 0.000) while it significantly decreased in the atelocollagen scaffold group (P = 0.000). Besides, the volume of engineered Artif Organs, Vol. 37, No. 12, 2013

cartilage in the aBMSC-dECM scaffold group at each time point was larger than that of the atelocollagen scaffold group (Fig. 5G). Unlike the in vitro culture, the wall of the in vivo atelocollagen scaffold was completely degraded within 2 weeks after implantation (Fig. 5L,P), whereas that of the aBMSC-dECM scaffold was degraded at a relatively slower rate (Fig. 5J,N,R).

CARTILAGE TISSUE ENGINEERING Chemical assay The DNA content of the engineered cartilage gradually decreased with time in both the aBMSCdECM scaffold group (P = 0.001) and the atelocollagen scaffold group (P = 0.001) during in vivo implantation. The DNA contents at 1, 2, and 3 weeks in the aBMSC-dECM scaffold group (2.2 ± 0.5, 1.6 ± 0.2, and 1.2 ± 0.2 ng/μg) were lower than those of the atelocollagen scaffold group (2.9 ± 0.7, 2.4 ± 0.3, and 1.7 ± 0.2 ng/μg) (Fig. 6A). Besides, the DNA contents of the engineered cartilage in the aBMSC-dECM scaffold group at 2 and 3 weeks were similar to that of the normal cartilage (1.3 ± 0.1 ng/ μg) with no statistical difference (Fig. 6A). The GAG content increased with time during the 3 weeks in vivo implantation in both the aBMSCdECM scaffold group (P = 0.000) and atelocollagen scaffold group (P = 0.000). The amounts of GAG contents of the engineered cartilage were 7.6 ± 1.8, 13.2 ± 2.3, and 16.1 ± 2.3% at 1, 2, and 3 weeks, respectively, in the aBMSC-dECM scaffold group, all of which were significantly higher than those of the atelocollagen scaffold group (1.7 ± 1.2, 5.6 ± 1.9, and 8.5 ± 2.3%) (Fig. 6B). Three weeks after implantation, the GAG content in the aBMSC-dECM scaffold group reached 52.6% of the normal cartilage and was two-times higher than that of the atelocollagen scaffold group. The GAG/DNA ratios also gradually increased in both the aBMSC-dECM scaffold group (P = 0.001) and atelocollagen scaffold group (P = 0.001). The GAG/DNA ratios of the aBMSC-dECM scaffold group were 38.0 ± 4.5, 85.7 ± 9.0, and 162.2 ± 54.4; sig-

E187

nificantly higher than those of the atelocollagen scaffold group (5.9 ± 3.0, 26.1 ± 6.2, and 50.1 ± 11.2) at each time point (Fig. 6C). The GAG/DNA ratio of the aBMSC-dECM scaffold group at 3 weeks after implantation reached 69.0% of the normal cartilage (235.1 ± 17.5). However, it was still lower than that of the normal cartilage (Fig. 6C). Compressive modulus The compressive moduli of the in vivo engineered cartilage were 90.2 ± 35.2, 292.7 ± 77.0, and 599.2 ± 169.7 KPa in the aBMSC-dECM scaffold group and 3.2 ± 0.9, 4.3 ± 1.3, and 5.3 ± 1.9 KPa in the atelocollagen scaffold group at 1, 2, and 3 weeks after implantation, respectively. The compressive moduli of the aBMSC-dECM scaffold group rapidly increased with time (P = 0.004) and were significantly higher than those of the atelocollagen scaffold group at 2 and 3 weeks after implantation (Fig. 6D). The compressive modulus in the aBMSC-dECM scaffold group reached 40.6% of the normal cartilage at 3 weeks after implantation (Fig. 6D). DISCUSSION In this study, autologous articular chondrocytes in the aBMSC-dECM scaffold formed thicker cartilage with relatively homogeneous and mature cartilage structure, and a higher expression of cartilaginous gene and protein than the atelocollagen scaffold. Furthermore, the engineered cartilage based on the aBMSC-dECM scaffold showed better cartilage formation than the atelocollagen scaffold in terms of

FIG. 6. Measurement of DNA content, GAG content, GAG/DNA ratio, and compressive modulus of the in vivo engineered cartilage. (A) The DNA content in the aBMSC-dECM scaffold group was similar to that of normal cartilage at 2 and 3 weeks after implantation. The GAG content (B), GAG/DNA ratio (C), and compressive modulus (D) were significantly higher in the aBMSC-dECM scaffold group than those in the atelocollagen scaffold group and more closely resembled normal cartilage at 3 weeks after implantation. Results are presented as means ± SD. *P < 0.05.

Artif Organs, Vol. 37, No. 12, 2013

E188

C. TANG ET AL.

volume and homogeneity, cartilage matrix content, and compressive modulus after in vivo implantation. These results indicated that the aBMSC-dECM scaffold could provide a favorable environment for chondrocyte and promote hyaline cartilage-like tissue regeneration. Our previous studies have reported the method of using porcine articular chondrocyte-derived ECM scaffold to produce the engineered cartilage (10,11). However, the technique is still unsatisfactory for clinical practice, mainly because the material was derived from xenogenous chondrocytes. One of the major barriers to xenogenous materials is that they contain galactosyl 1, 3 galactose epitopes, which are expressed in the cell membranes of all mammals except humans. In humans, the antibody stimulated by galactosyl 1, 3 galactose epitopes is present in a high concentration and triggers antibody-dependent cell-mediated cytotoxicity (18). In addition, pathogen transmission through the use of allogenous or xenogenous materials is thought to be another potentially major problem. Kim et al. (19) reported a case in which the patient received an allogenous dura mater graft (Lyodura). The patient developed iatrogenic Creutzfeldt-Jakob disease 23 years after implantation, which was pathologically confirmed as a complication after graft implantation. Furthermore, the use of xenogenous materials may provoke a number of ethical issues. In the present study, the aBMSC-dECM scaffold was originated from autologous BMSCs. It is widely reported that autologous materials can avoid potential risks of host immune response, inflammation, and epidemic spread (9,20). Therefore, the aBMSC-dECM scaffold may be a better choice for clinical application compared with xenogenous ECM scaffolds. The available autologous materials derived from donor tissues and organs are highly limited. It is almost impossible to use such materials for tissue engineering. The ECM materials secreted by autologous cells have been used to solve this problem (9,20). In previous studies, Lu et al. (9) manufactured autologous cell-derived ECM scaffolds by combining cultured autologous cells in a three-dimensional template, decellularization, and template removal. No adverse inflammatory or immune responses were found in their in vivo study. Zeitouni et al. (20) also used human autologous BMSCs as a cell source and a temporary template for preparing scaffolds. The report indicated that autologous cell-derived ECM scaffolds showed excellent biocompatibility and substantially and reproducibly improved bone repair in a mouse model. However, some complex treatments such as decellularization and template removal were Artif Organs, Vol. 37, No. 12, 2013

essential for scaffold preparation in their studies (9,20). More than 70% of collagen, elastin, glycosaminoglycan, and many growth factors were lost, and the risk of inflammation was also increased during those treatments (21). The concept of the current protocol originated from a “template-free” ECM material. The three-dimensional porous scaffold was fabricated via cross-linking and freezedrying techniques. Decellularization and template removal were not performed in our protocol. Therefore, the aBMSC-dECM scaffold could efficiently maintain the biological activity and structural integrity of ECM components. As shown in the results of the in vitro study, most of the chondrocytes in the atelocollagen scaffold group failed to migrate into the interior of the engineered cartilage. Relatively compact hyaline cartilage-like tissues were rapidly accumulated in the outer part and only slightly enlarged over time. A frothy structure was presented in the inner part due to a lack of chondrocytes and limited mass transportation. All these factors resulted in a heterogeneous structure. On the contrary, the aBMSC-dECM scaffold enhanced the thickness and homogeneity of the engineered cartilage (Fig. 3). The suggested reasons for this include: (i) Uniform distribution and highly interconnected porous structures were obtained in aBMSC-dECM scaffold, which was beneficial to cell seeding and proliferation, matrix production, and mass transportation (22). (ii) The aBMSC-dECM scaffold was made from autologous BMSCs and their ECM components. The scaffold could provide a biocompatible environment for chondrocytes (8), which was evident from the efficient adhesion, round shape, and viability of chondrocytes in the cellscaffold construct at the early stage (Fig. 2). (iii) Rapid and complete degradation of the scaffold not only promoted cell migration that further contributed to a relatively uniform distribution but also enhanced the transportation of nutrients and metabolites to the interior of the engineered cartilage (Fig. 3). However, the wall of the atelocollagen scaffold did not noticeably change during in vitro culture over 4 weeks. The slow rate of degradation may negatively influence the distribution and viability of chondrocytes in the atelocollagen scaffold, as well as the pattern of cartilage formation in vitro. (iv) Homogeneous cell distribution and efficient mass transportation might play an important role in cartilage formation in both outer and inner parts, thus contributing to the thickness and homogeneity of the engineered cartilage. Besides thickness and homogeneity, the aBMSCdECM scaffold also influenced the cartilaginous gene

CARTILAGE TISSUE ENGINEERING and protein expression pattern in the engineered cartilage. The expressions of cartilaginous genes (COL2A1 and ACAN) were analyzed by real-time PCR. The current results demonstrated that the expression of COL2A1 and ACAN were contiguously upregulated in the aBMSC-dECM scaffold group during 4 weeks in vitro culture and were markedly higher than the atelocollagen scaffold group.The results of protein expression in the two groups were similar to those of the gene expression. The expression of type II collagen was constantly and significantly increased in the aBMSC-dECM scaffold group during in vitro culture, whereas it was sharply decreased in the atelocollagen scaffold group at the end stage. The in vivo fate is the most important criterion for evaluating engineered cartilage (23). The current results demonstrated that the aBMSC-dECM scaffold showed better cartilage formation than the atelocollagen scaffold after 3 weeks implantation. We considered two reasons for the results. First, the aBMSC-dECM scaffold could provide functional and structural environment. The chondrocytes in the aBMSC-dECM scaffold maintained higher biological function than those in the atelocollagen scaffold during in vivo implantation. Second, the aBMSCdECM scaffold could efficiently regulate the in vivo cartilage formation. A very rapid infiltration and deposition of cartilage matrix was associated with aBMSC-dECM scaffold degradation at the remodeling site. However, the atelocollagen scaffold was completely degraded within 2 weeks after in vivo implantation. We think that the degradation is too rapid to provide a structural environment for chondrocytes, which is not good for cell proliferation and matrix production. It is worthwhile to mention that the cartilage matrix content and compressive modulus of engineered cartilage in the aBMSC-dECM scaffold group were lower than those in the normal cartilage group. The likely cause was that we used passage 3 chondrocytes in this study. Meretoja et al. (17) reported the engineered cartilage with passaged chondrocytes had lower cellularities and cartilagelike matrix formation in comparison to that of primary chondrocytes. Besides, the in vivo implantation of engineered cartilage in a nude mouse model cannot completely simulate the biomechanical environment of normal cartilage. Nonetheless, our results are still promising compared with previous reports. The cartilage matrix content and compressive modulus of engineered cartilage using aBMSCdECM scaffold more closely resembled hyaline cartilage compared to those using other scaffolds

E189

previously, such as fibrin/hyaluronic acid composite gel (24), hybrid scaffold (16), and PGA scaffold (10).

CONCLUSIONS The current study demonstrated that the aBMSCdECM scaffold, which had a good microstructure and biocompatibility, could efficiently enhance the thickness and homogeneity of in vitro engineered cartilage and the expression of cartilaginous genes and proteins. Most importantly, the engineered cartilage based on the aBMSC-dECM scaffold showed more homogeneous and mature cartilage structure, with a higher cartilage matrix content and a stronger compressive modulus after in vivo implantation. These results indicated that the aBMSC-dECM scaffold could provide a favorable environment for chondrocytes and promote hyaline cartilage-like tissue regeneration. It could be a successful novel candidate scaffold for cartilage tissue engineering. Acknowledgments: We thank Professor Rongbin Yu from Nanjing Medical University for his valuable statistical work.This study was funded by the Chinese National Nature Sciences Foundation (No. 31070861, 81171745).

REFERENCES 1. Chang CH, Kuo TF, Lin FH, et al. Tissue engineering-based cartilage repair with mesenchymal stem cells in a porcine model. J Orthop Res 2011;29:1874–80. 2. Brittberg M, Lindahl A, Nilsson A, Ohlsson C, Isaksson O, Peterson L. Treatment of deep cartilage defects in the knee with autologous chondrocyte transplantation. N Engl J Med 1994;331:889–95. 3. Zeifang F, Oberle D, Nierhoff C, Richter W, Moradi B, Schmitt H. Autologous chondrocyte implantation using the original periosteum-cover technique versus matrix-associated autologous chondrocyte implantation: a randomized clinical trial. Am J Sports Med 2010;38:924–33. 4. Nukavarapu SP, Dorcemus DL. Osteochondral tissue engineering: current strategies and challenges. Biotechnol Adv 2013;31:706–21. 5. Nehrer S, Chiari C, Domayer S, Barkay H, Yayon A. Results of chondrocyte implantation with a fibrin-hyaluronan matrix: a preliminary study. Clin Orthop Relat Res 2008;466:1849–55. 6. Chiang H, Liao CJ, Hsieh CH, Shen CY, Huang YY, Jiang CC. Clinical feasibility of a novel biphasic osteochondral composite for matrix-associated autologous chondrocyte implantation. Osteoarthritis Cartilage 2013;21:589–98. 7. Badylak SF. Xenogeneic extracellular matrix as a scaffold for tissue reconstruction. Transpl Immunol 2004;12:367–77. 8. Badylak SF. The extracellular matrix as a biologic scaffold material. Biomaterials 2007;28:3587–93. 9. Lu H, Hoshiba T, Kawazoe N, Chen G. Autologous extracellular matrix scaffolds for tissue engineering. Biomaterials 2011;32:2489–99. 10. Jin CZ, Park SR, Choi BH, Park K, Min BH. In vivo cartilage tissue engineering using a cell-derived extracellular matrix scaffold. Artif Organs 2007;31:183–92. Artif Organs, Vol. 37, No. 12, 2013

E190

C. TANG ET AL.

11. Jin CZ, Choi BH, Park SR, Min BH. Cartilage engineering using cell-derived extracellular matrix scaffold in vitro. J Biomed Mater Res A 2010;92:1567–77. 12. Lee HS, Huang GT, Chiang H, et al. Multipotential mesenchymal stem cells from femoral bone marrow near the site of osteonecrosis. Stem Cells 2003;21:190–9. 13. Pittenger MF, Mackay AM, Beck SC, et al. Multilineage potential of adult human mesenchymal stem cells. Science 1999;284:143–7. 14. Choi KH, Choi BH, Park SR, Kim BJ, Min BH. The chondrogenic differentiation of mesenchymal stem cells on an extracellular matrix scaffold derived from porcine chondrocytes. Biomaterials 2010;31:5355–65. 15. Yang F, Qu X, Cui W, et al. Manufacturing and morphology structure of polylactide-type microtubules orientationstructured scaffolds. Biomaterials 2006;27:4923–33. 16. Spiller KL, Liu Y, Holloway JL, et al. A novel method for the direct fabrication of growth factor-loaded microspheres within porous nondegradable hydrogels: controlled release for cartilage tissue engineering. J Control Release 2012;157:39–45. 17. Meretoja VV, Dahlin RL, Kasper FK, Mikos AG. Enhanced chondrogenesis in co-cultures with articular chondrocytes and mesenchymal stem cells. Biomaterials 2012;33:6362–9.

Artif Organs, Vol. 37, No. 12, 2013

18. Schussler O, Shen M, Shen L, Carpentier SM, Kaveri S, Carpentier A. Effect of human immunoglobulins on the immunogenicity of porcine bioprostheses. Ann Thorac Surg 2001;71(Suppl. 5):S396–400. 19. Kim HL, Do JY, Cho HJ, et al. Dura mater graft-associated Creutzfeldt-Jakob disease: the first case in Korea. J Korean Med Sci 2011;26:1515–7. 20. Zeitouni S, Krause U, Clough BH, et al. Human mesenchymal stem cell-derived matrices for enhanced osteoregeneration. Sci Transl Med 2012;4:132ra55. 21. Gilbert TW, Sellaro TL, Badylak SF. Decellularization of tissues and organs. Biomaterials 2006;27:3675–83. 22. Lee CH, Cook JL, Mendelson A, Moioli EK, Yao H, Mao JJ. Regeneration of the articular surface of the rabbit synovial joint by cell homing: a proof of concept study. Lancet 2010;376:440–8. 23. Zhang Y, Yang F, Liu K, et al. The impact of PLGA scaffold orientation on in vitro cartilage regeneration. Biomaterials 2012;33:2926–35. 24. Park SH, Cui JH, Park SR, Min BH. Potential of fortified fibrin/hyaluronic acid composite gel as a cell delivery vehicle for chondrocytes. Artif Organs 2009;33:439–47.