Cell-seeded biodegradable scaffolds have attracted growing attention in regenerative medicine. Hydrogels create a suitable environment for chondrocytes and provide three-dimensional immobilization of cells.1–3 Hydrogels mimic highly hydrated natural cartilage, which contains more than 90% water.4 Optimal biomechanical properties are key requirements for hydrogels; however, they are often difficult to achieve. Consequently, the development of scaffolds that fulfill biocompatibility and biomechanical requirements is a key task of modern tissue engineering.
The application of composite materials is a promising approach for the development of mechanically durable scaffolds. Naturally derived biopolymers, such as collagen, fibrin, chitosan, agarose, alginate, and hyaluronic acid (HA) and its derivatives,1,3,5 or synthetic polymers, such as polyvinyl alcohol, polyethylene oxide, polyethylene glycol, and poly (l-lactide-co-ε-caprolactone),6 are often used to modify and improve the properties of hydrogels.7,8 Fibrin hydrogels have been used to immobilize cells on solid polymer scaffolds.9,10 A composite scaffold consisting of sodium hyaluronate, type I collagen, and fibrin has been developed in our laboratory and has shown the ability to regenerate osteochondral defects in rabbits.11,12 However, there is no commonly available collagen approved for implantation to human cartilage on the market.
The aim of this study was to investigate the efficacy of osteochondral regeneration after the implantation of chondrocyte-seeded composite scaffolds composed of a mixture of fibrin with HA without any collagen present. In addition, the effect of cell number on the healing process was also examined. The histological and biomechanical parameters were examined in regenerated tissue 6 months after implantation. This study was conducted according to good laboratory practice (GLP) with the aim of enabling a future human clinical trial.
Materials and Methods
Housing and Animal Care
For this study, 16 male miniature pigs obtained from the Institute of Animal Physiology and Genetics of the ASCR (Libechov, CZ) were used. Animal care was in compliance with the Act of the Czech National convention for the protection of vertebrate animals. The animals were divided into a treatment group (group I) and a control group (group II), each with eight animals.
Pig Hyaline Cartilage Collection
Surgery was performed under general anesthesia. The animals were premedicated with an intramuscular injection of 2 mg/kg azaperonum (Stresnil, Janssen Pharmaceutica N.V.; Beerse, BE) and 1 mg atropine (Atropin Biotika, BB Harma, SK) pro toto. General anesthesia was induced by ketamine (20 mg/kg; Narketan 10, Vetoquinol; CZ) and maintained with 1.5% isoflurane at a 2 L/min constant flow. A lateral arthrotomy was performed with medial luxation of the patella, and samples were taken from the trochlea femoris and placed in Iscove's Modified Dulbecco's Medium (IMDM) (PAA, AT) supplemented with 40 μg/ml gentamicin (Lek Pharmaceuticals, SI). The lesion was sutured in the followed layers: the joint theca, muscles, and subcutis using absorbable catgut and the cutis with nonabsorbable material. All the animals received preventive doses of antibiotics (procaini benzylpenicillinum 8,000 IU/kg b.w.; Peni-Kel 300, a.u.v, KELA Laboratoria N.V., Hoogstraten, Belgium) by intramuscular administration and analgesics (metamizolum 30 mg/kg b.w., Vetalgin inj., Intervet International B.V.; Boxmeer, NL). The sutures in the cutis were removed 12 days after the operation.
After a 14-hour enzymic digestion in a collagenase solution (0.36 U/ml collagenase NB6, GMP Grade, Serva, DE), the chondrocytes (cca 1 × 106 cells) were centrifuged for 5 minutes at 270 g and cultured in IMDM (PAA, AT) supplemented with 10% fetal bovine serum (Pharma Grade, PAA, AT), 100 nM dexamethasone (Dexamed, Medochemie, CY), 40 μg/ml gentamicin (Lek Pharmaceuticals, SI), 10 μg/ml insulin (Actrapid 100 mU/ml inj., Novo Nordisk, DK), and 20 μg/ml ascorbic acid (Acidum ascorbicum, Biotika, SK) for 19 days in an incubator with a humidified atmosphere, 5% CO2 at 37°C. The culture was passaged using Accutase (PAA, AT) before confluence was reached, and autologous chondrocytes were cultivated for 19 days (∼[7 − 9] × 106 cells).
The scaffolds were prepared at about 4°C by mixing 66 μl sodium hyaluronate (10 mg/ml, Hyalgan, Fidia CH), 132 μl of cell suspension (7 × 106 chondrocytes), 0.32 ml of Tissucol in aprotinine (fibrinogen 70–110 mg/ml, aprotinine 3,000 KIU/ml), and 0.32 ml of thrombin (4 IU/ml) in CaCl2 (40 μmol/ml, Tissucol Kit, Baxter, AT). Scaffolds for biomechanical testing were prepared as described earlier in the text.11,12 The gel was formed at 37°C. The viability of the cells in the hydrogel scaffold was demonstrated previously11 and was higher than 93% after 14 days of cultivation.
The biomechanical properties of the samples were tested using a blunt impact testing method.13 Briefly, a pendulum-like apparatus permits a rapid increase of acting force resembling physiological cartilage loading. Sample deformation was detected by a piezoelectric accelerometer (Bruel Kjaer 4375) and a Laser Doppler Vibrometer (Polytec OFV-302). Measurements were obtained using a constant initial striking body velocity of v = 0.017 m/s. The Young's modulus was subsequently evaluated from the linear-elastic region of the loading curve. Samples were obtained from three pigs (not included in the GLP study, scaffolds with 8.5–9 × 106 chondrocytes/1 ml).
Artificial Cartilage Implantation
A fibrin/HA composite hydrogel containing autologous chondrocytes was implanted into the femoral trochlea of the right knee of eight miniature pigs. Anesthesia, medication, and subsequent care were as described previously (Pig Hyaline Cartilage Collection). A lateral arthrotomy of the right knee joint with medial luxation of the patella was performed. A circular defect of the joint cartilage in the trochlea femoris (diameter 6 mm) was created until reaching the subchondral bone tissue. The scaffold with autologous chondrocytes was fixed in situ with a tissue adhesive (Tissucol). The lesion was sutured in anatomical layers.
Surgery of Control Animals
A control group of eight animals was made. The same procedure as in the treatment group was used for making and suturing the defect in the right knee; however, the defects were left untreated.
The animals were sacrificed 24 weeks after the operation. The thoracic and visceral cavities were opened and examined macroscopically. The right knees were removed and the femoral condyles, including the site of hydrogel implantation, were fixed in 4% phosphate buffered formaldehyde for histological examination.
The bones were decalcified with ethylenediaminetetraacetic acid (EDTA). Histological slides were made using a common paraffin technique and stained with hematoxylin and eosin (HE), a combined staining technique using Alcian blue (AB) at a pH of 2.5 followed by periodic acid-Shiff (PAS) reaction for mucosubstances, van Gieson stain, orcein, and the Gomori method. In addition, immunohistochemical detection of type II collagen was performed to demonstrate the formation of differentiated cartilaginous tissue. A mouse monoclonal primary antibody against type II collagen, clone II-II6B3, obtained from the Developmental Studies Hybridoma Bank, followed by a horseradish peroxidase-labeled rabbit polyclonal anti-mouse secondary antibody (Sigma A 9004) and DAB visualization (SIGMAFAST DAB tablets set) were used.
Histological and Histochemical Scoring System
The repair of the osteochondral defects in all groups was compared using a scoring system for histological and histochemical results with a maximum of 24 points, modified from van Susante et al.14 The blinded analysis was performed by two investigators.
Scaffold Preparation and Biomechanical Testing
Scaffolds were produced in the absence or presence of type I collagen (0.1 mg/ml). The biomechanical properties of these in vitro prepared hydrogels were tested using a blunt impact method. The presence of collagen considerably influenced the biomechanical parameters of the chondrografts in vitro. Chondrografts prepared with collagen were significantly more rigid and were characterized by a ∼25% higher Young's modulus (Table 1).
Chondrograft Implantation and Animal Treatment
Despite the positive effect that collagen had on the biomechanical properties observed in in vitro experiments, only hydrogels without collagen were applied in this in vivo study.
The animals of the treatment group (group I) received scaffolds with an increasing cell concentration (between 8.1 × 106 and 10.3 × 106 cells per 1 ml of scaffold). During a healing period of 6 months, body weight gradually increased in both groups in correspondence with the age of the animals (Table 2). Nonetheless, the body weight of the animals in group I was slightly higher at the end of the observation period. We assume that this was a consequence of accelerated healing and, consequently, better feeding. No treatment-related clinical changes were recorded, and the healing of the surgical wounds did not show signs of secondary inflammation. The motility of the knee joints was not affected by the operation. All the animals survived to their scheduled necropsy.
Histopathology and Immunohistochemistry
Gross pathology examination of cartilage from miniature pigs with osteochondral defects, treated with hydrogel implants formed from a mixture of fibrinogen and HA (group I), was performed 6 months after the operation. In general, the defects were typically observed as whitish oval foci with a diameter between 6 and 9 mm (Figure 1A). Sometimes central depressions were seen. No signs of exudation into the joint lumen or inflammation of the surrounding soft tissues were found. Other organs in the thoracic and abdominal cavities contained no abnormalities in any of the eight pigs.
Interestingly, the results of microscopic examination differed between samples and depended on the number of cells seeded on the scaffolds. Five animals with cell concentrations ranging from 8 to 9 × 106 cells/ml of scaffold gave relatively similar results (group I[A]). Cartilage defects were lined on both sides by a thin noncellular transient zone toward the joint cartilage. This was followed by a layer of isogenous groups of chondrocytes with signs of advanced differentiation merging into fibrocartilaginous tissue at the center (Figure 2A). Only solitary fat cells and weak or solitary vascularization were observed. The PAS reaction was positive in the basis of the defect and weakly positive in the center. A similar result was obtained with AB staining (Figure 2C), whereas the covering fibrous tissue was negative with both stainings. Immunohistochemistry for type II collagen revealed positivity in the newly formed cartilage on the borders of the defects. Collagen formation was characterized by young isogenous groups in the noncellular transient zone and in the center of the overlaying fibrous tissue. Type II collagen was less present in the center of the defect (Figure 2E).
The healing process for scaffolds seeded with more than 9 × 106 cells/ml was deteriorated worse (group I[B]). The cartilage defects in a pig with an implanted scaffold containing 9 × 106 cells/1 ml were lined on both sides by a thin noncellular transient zone toward the joint cartilage. At the borders and the bottom of the defect, layers of fibrocartilage containing isogenous groups of chondrocytes with signs of advanced differentiation were seen. The center of the lesion was filled by loose connective tissue with multiple fat cells and relatively weak or solitary vascularization. The presence of multiple small foci of chondrocytes was revealed, most likely as remnants of the implanted autologous chondrocytes. The lesion was covered by a layer of dense fibrous tissue. Immunohistochemistry yielded positive staining for type II collagen in the newly formed cartilage on the border of young isogenous groups in the defect and in the area of the noncellular transient zone. Collagen was less present in the center of the defect. A similar observation was found with AB and PAS staining.
In the two remaining samples, the defect was filled with loose collagenous tissue, abundant fat cells, and vascularization. The cartilage defect was lined on both sides by a thin noncellular transient zone toward the joint cartilage and a layer of fibrocartilage. PAS and AB showed positive staining in the region of isogenous chondrocyte groups, whereas the central region revealed only weak positivity. Positive staining for type II collagen was found in the noncellular layers and adjacent to newly formed fibrocartilaginous tissue.
The control nontreated group of eight miniature pigs (group II) revealed a similar gross pathology as that of group I(B). Oval whitish foci of diameters between ∼8 and 11 mm were present in the center of the cartilage of the right-side trochlea femoris. No signs of exudation into the joint lumen or inflammation of the surrounding soft tissues were found (Figure 1B).
Histopathological examination showed artificial defects penetrating deeply into the bone (approximately 2–3 times deeper than the thickness of the cartilage). A thin noncellular transient zone was present on both sides of the defect toward the joint cartilage. On the lateral sides of the defect, multiple foci of isogenous groups of chondrocytes undergoing initial differentiation to hyaline cartilage were found. The basis of the defect was lined by a single layer of cubic cells, most likely osteoblasts, covered by a variable layer of fibrous tissue with multiple capillaries. The center of the lesion was filled with loose collagenous tissue with abundant fat cells and marked vascularization (Figure 2B). The defect was covered by dense fibrous tissue. PAS and AB showed positive staining only in the region of isogenous chondrocyte groups, whereas the central part of the defect was weakly positive (Figure 2D). Immunohistochemistry for type II collagen was weakly positive in the newly formed cartilage on the borders of the defects (Figure 2F). In most cases, impressed bone debris was present in the underlying bone lacunae.
Histological and Histochemical Scoring System
Histological images from the treatment and control groups were evaluated using a histological and histochemical scoring system (Table 3). Interestingly, the value of the histological score was dependent on the number of cells incorporated in the hydrogel scaffold (Table 4). Although scaffolds with a cell number lower than 9 × 106 cells/ml reached an average histological score of 16.0 ± 3.8, the scaffolds seeded with higher cell numbers were characterized by an average value of only 9.0 ± 0.8. This was statistically similar to the control group without treatment (9.7 ± 2.3).
Biomechanical Properties of the Remodeled Tissue
The biomechanical properties of tissue formed 6 months after fibrin/HA composite hydrogel implantation were compared with those of the native cartilage (Table 1). Engineered cartilage implants formed macroscopic tissue with mechanical characteristics reaching 73.4% of the native cartilage. The Young's modulus changed dramatically and approached that of the native cartilage. Also, the stress-strain diagram of the remodeled tissue resembled that of native cartilage (Table 1). The newly remodeled tissue was biomechanically well developed despite the absence of collagen in the implanted scaffold.
Fibrin mixed with high-molecular weight (MW) HA, in the presence or absence of collagen, was found to be a suitable hydrogel for the seeding of chondrocytes in vitro. Clearly, the presence of collagen changed the viscoelasticity and improved the chondrograft biomechanical properties in vitro. This observation agrees with previously published data.11,12,15,16 However, collagen as an approved substance for human clinical use is rather difficult to obtain. Although several human-based collagen products have been approved for human use, none of them received approval for implanting into bone, tendons or ligaments.17 Although, there is still the risk of transmissible bovine encephalopathy in bovine-derived collagen, human recombinant collagen has not found wide practicability yet.
Therefore, a suitable alternative was introduced for our in vivo study; a composite fibrin-based hydrogel in combination with high-MW HA only. This chondrograft without collagen was, due to the presence of HA, sufficiently rigid and showed appropriate biomechanical properties for implantation.
In contrast to the results of our in vitro experiments, the presence of collagen in the hydrogels was avoidable in cartilage regeneration in vivo and unnecessary for the formation of biomechanically stable tissue. The implantation of seeded scaffolds into osteochondral defects in miniature pigs was well tolerated. The healing process of the treated defects mostly penetrated deeply into the bone tissue without fat cells and vascularization in the defect center. It was characterized by a significantly larger area of type II collagen formation at the sites of early chondrocyte differentiation and proliferation and by the more advanced differentiation of the cells and intercellular substance toward hyaline cartilage in comparison with controls. The regenerated cartilage 6 months after implantation showed improved biomechanical and histological parameters with lower chondrocyte concentrations (8–9 × 106 cells/ml of scaffold), whereas scaffolds seeded with a higher chondrocyte concentration were comparable with the control group. We hypothesize that the biomechanical properties and/or viscoelasticity of our scaffolds were probably limiting factors during the healing process. A lower diffusion coefficient and the partial destruction of the fibrin scaffold could reflect this phenomenon and could be connected with a worse nutrition supply and/or waste removal. The free volume of a single cell encapsulated in the scaffolds ranged from 2–3 times smaller than cells in native tissue.18 Exceeding certain cell number limits per hydrogel volume could reduce the nutrition exchange in artificial scaffolds, when compared with native cartilage. In addition, the secretion of plasminogen or matrix metalloproteinases from encapsulated cells can undoubtedly influence the integrity of fibrin scaffolds and degradation products negatively influence the cell microenvironment.19 Theoretical calculations can be used to deal with this problem in detail. This work was performed in compliance with the principles of GLP.
Regenerated cartilage showed improved biomechanical and histological properties only 6 months after implantation. Notably, the quality of the healing process was dependent on the initial chondrocyte concentration in the scaffolds. This work was performed in compliance with the principles of GLP and represents a significant advance toward human application.
Supported by the Academy of Sciences of the Czech Republic (institutional research plans AV0Z50390703 and AV0Z50390512), the Ministry of Education, Youth, and Sports of the Czech Republic (research programs NPV II 2B06130), the Grant Agency of the Academy of Sciences (grant no. IAA500390702), the Czech Science Foundation (grant numbers GA202/09/1151 and P304/10/1307), the EU project BIOSCENT (ID number 214539), and the Grant Agency of Charles University (grant no. 119209). The authors are grateful to Sam Norris for proof reading of this manuscript.
1. Drury JL, Mooney DJ: Hydrogels for tissue engineering: Scaffold design variables and applications. Biomaterials
24: 4337–4351, 2003.
2. Eyrich D, Brandl F, Appel B, et al
: Long-term stable fibrin gels for cartilage engineering. Biomaterials
28: 55–65, 2007.
3. Ng KW, Wang CCB, Mauck RL, et al
: A layered agarose approach to fabricate depth-dependent inhomogeneity in chondrocyte-seeded constructs. J Orthop Res
23: 134–141, 2005.
4. Buckwalter JA, Mankin HJ: Articular cartilage: Degeneration and osteoarthritis, repair, regeneration, and transplantation. Instr Course Lect
47: 487–504, 1998.
5. Tan H, Chu CR, Payne KA, Marra KG: Injectable in situ forming biodegradable chitosan-hyaluronic acid based hydrogels for cartilage tissue engineering. Biomaterials
30: 2499–2506, 2009.
6. Jung Y, Kim SH, Kim YH, Kim SH: The effect of hybridization of hydrogels and poly(L-lactide-co-epsilon-caprolactone) scaffolds on cartilage tissue engineering. J Biomater Sci Polym Ed
21: 581–592, 2010.
7. Bryant SJ, Anseth KS: Hydrogel properties influence ECM production by chondrocytes photoencapsulated in poly(ethylene glycol) hydrogels. J Biomed Mater Res
59: 63–72, 2002.
8. Kobayashi M, Toguchida J, Oka M: Preliminary study of polyvinyl alcohol-hydrogel (PVA-H) artificial meniscus. Biomaterials
24: 639–647, 2003.
9. Lind M, Larsen A, Clausen C, et al
: Cartilage repair with chondrocytes in fibrin hydrogel and MPEG polylactide scaffold: An in vivo study in goats. Knee Surg Sports Traumatol Arthrosc
16: 690–698, 2008.
10. Kim DY, Pyun J, Choi JW, et al
: Tissue-engineered allograft tracheal cartilage using fibrin/hyaluronan composite gel and its in vivo implantation. Laryngoscope
11. Filová E, Jelínek F, Handl M, et al
: Novel composite hyaluronan/type I collagen/fibrin scaffold enhances repair of osteochondral defect in rabbit knee. J Biomed Mater Res B Appl Biomater
87: 415–424, 2008.
12. Filová E, Rampichová M, Handl M, et al
: Composite hyaluronate-type I collagen-fibrin scaffold in the therapy of osteochondral defects in miniature pigs. Physiol Res
56 (suppl 1): S5–S16, 2007.
13. Varga F, Drzík M, Handl M, et al
: Biomechanical characterization of cartilages by a novel approach of blunt impact testing. Physiol Res
56 (suppl 1): S61–S68, 2007.
14. van Susante JL, Buma P, Schuman L, et al
: Resurfacing potential of heterologous chondrocytes suspended in fibrin glue in large full-thickness defects of femoral articular cartilage: An experimental study in the goat. Biomaterials
20: 1167–1175, 1999.
15. Kolácná L, Bakesová J, Varga F, et al
: Biochemical and biophysical aspects of collagen nanostructure in the extracellular matrix. Physiol Res
56 (suppl 1): S51–S60, 2007.
16. Park S, Hung CT, Ateshian GA: Mechanical response of bovine articular cartilage under dynamic unconfined compression loading at physiological stress levels. Osteoarthritis Cartilage
12: 65–73, 2004.
17. Bauman L: CosmoDerm/CosmoPlast (human bioengineered collagen) for the aging face. Facial Plast Surg
20: 125–128, 2004.
18. Temenoff JS, Mikos AG: Review: Tissue engineering for regeneration of articular cartilage. Biomaterials
21: 431–440, 2000.
19. Passaretti D, Silverman RP, Huang W, et al
: Cultured chondrocytes produce injectable tissue-engineered cartilage in hydrogel polymer. Tissue Eng
7: 805–815, 2001.