Tissue-engineered cartilage constructed by a biotin-conjugated anti-CD44 avidin binding technique for the repairing of cartilage defects in the weight-bearing area of knee joints in pigs

Preparation of chondrocytes

Cartilage was collected aseptically from the knee and hip joints of five-week-old porcine (Animal Center, Zhongshan Hospital, Fudan University). First, it was rinsed with phosphate buffer solution (PBS), divided with a scalpel, and washed twice with PBS. The ECM was digested in Dulbecco’s Modified Eagle’s Medium (DMEM; Biowest, Nuaillé, France), containing 0.2% type II collagenase (Sigma-Aldrich, St Louis, Missouri), at 37°C in a water bath and shaken for two hours. The tissue debris was removed by filtration using a 40 μm sieve, and chondrocytes in the filtrate were collected by centrifugation at 300 g for ten minutes. They were re-suspended and cultured in complete DMEM with 10% foetal bovine serum (FBS; GE Healthcare, Logan, Utah), 100 U/mL penicillin, and 100 μg/mL streptomycin. The cell suspension was adjusted to a density of 2×10⁵/mL, seeded into 75 cm² flasks (Corning Inc, Corning, New York), and cultured at 37°C with 5% CO₂. The culture medium was changed every two to three days. Confluent chondrocytes were passaged at a ratio of 1:3, and passage 2 chondrocytes were used.

The construction of tissue-engineered cartilage with biotin-conjugated anti-CD44 avidin binding system

Porous chitosan scaffolds were prepared as follows: 10 mL chitosan solution was poured into a mould, frozen at -20°C for 24 hours, and immersed in -5°C ethanol for another 24 hours. After removing the mould, the well-shaped scaffold with ethanol was equilibrated at room temperature for two hours. The scaffold was then put into 5% NaHCO₃ solution to neutralise the residual acetic acid for six hours, and finally washed with de-ionised water five times. The scaffold was prepared with a diameter of 10 mm and a height of 3 mm. Its porosity was > 90%, and the pore diameter was 100 µm to 150 µm. It was sterilised with ethylene oxide.

Avidination of the porous chitosan scaffold

The avidination of the porous scaffolds was performed in accordance with the manufacturer’s recommended procedure and with previous studies.²⁸ The prepared scaffold was placed in a 12-well plate to which 3.0 mL of 0.1 mg/mL avidin solution was added, and the plates were incubated in a shaker for one hour at room temperature. The scaffold was washed twice with PBS containing 100 U/mL penicillin, 100 μg/mL streptomycin and 25 mg/mL amphotericin B, and air-dried.

Tissue-engineered cartilage constructed with biotin-conjugated anti-CD44 avidin binding system

The tissue-engineered cartilage was divided into three groups: group 1, chitosan scaffold + chondrocytes (control group); group 2, avidinised porous chitosan scaffold + biotinylated chondrocytes (BA group); and group 3, avidinised porous chitosan scaffold + chondrocytes treated with biotin-conjugated anti-CD44 (CBA system group). The tissue-engineered cartilage was constructed as described previously²⁸. Briefly, 5 × 10⁷/mL of chondrocytes were treated with biotin (Sigma-Aldrich) for 30 minutes (1 mg biotin /1 × 10⁶ chondrocytes) or biotin-conjugated anti-CD44 antibody (BioLegend, San Diego, California) (0.25 µg of CD44 monoclonal antibody biotin /1 × 10⁶ chondrocytes), and washed twice with PBS. The chondrocytes were then seeded into the scaffolds in the 24-well plate (200 μL cell suspension per scaffold). The chondrocyte-scaffold complexes were cultured in a CO₂ incubator, at 37°C with atmosphere containing 5% CO₂. Four hours after the initial seeding, the complexes were transferred to another 24-well plate. The culture medium was collected and the detached chondrocytes were counted in a blood counting chamber and the ratio of detachment was calculated. Fresh culture medium (2 mL) was added to the new 24-well plate with the chondrocyte-scaffold constructs. The 24-well plate was then placed in a CO₂ incubator and the culture medium was renewed every two days. Except for the complexes that were tested for real-time polymerase chain reaction, the remaining chondrocyte-scaffold complexes were transplanted to repair the full-thickness defects of the articular cartilage of the knees in the pigs.

Repair of the full-thickness defects of weight-bearing knee-joint articular cartilage in pigs

We obtained 28 mini-type pigs (three months old; mean weight, 15 kg, sd 2.5; gender not considered). They were checked to ensure that there were no visible wounds, infection or deformity near the knees. Anaesthesia was induced by muscular ketamine, 10 mg/kg. Using a sterile technique, a medial parapatellar incision was made in both hind limbs, followed by a lateral parapatellar arthrotomy. The patella was retracted medially, and full-thickness cartilage defects (10 mm diameter, 3 mm depth) were created by trepan on the load-bearing region of the lateral femoral condyle (Fig. 2). A total of 52 full-thickness defects were created. Two pigs died while under anaesthetic.

Fig.

a) Schematic image showing the location of the cartilage defects on the load-bearing region of the lateral femoral condyle; b) the surgical approach showing location of cartilage defects on the lateral femoral condyle.

The experimental groups included: a full-thickness defect repaired with traditionally engineered cartilage (control group, n = 12); a full-thickness defect repaired with engineered cartilage constructed by biotin-avidin binding technique (BA group, n =12); a full-thickness defect repaired with engineered cartilage constructed by CBA binding technique (CBA group, n = 12); a full-thickness defect repaired with the plug harvested when creating the defect (auto group, n = 12). The defects were randomly assigned into the four groups, with 12 defects in each group. Two pigs (four defects) were used as the blank control and received no treatment. The tissue-engineered articular cartilage (or autologous articular cartilage) was transplanted into the defects after the basal region was sprayed with medical adhesive. After thorough haemostasis, the incision was closed in layers using absorbable sutures. Each pig received 2.4 × 10⁵ units of gentamicin intramuscularly during and after the surgery to prevent infection. Movement and weight-bearing were allowed as tolerated after surgery. At 12 and 24 weeks after surgery, respectively, 12 pigs in the experimental group (three in each group) and one in the blank group were killed with an overdose of pentobarbital. The hind limbs were disarticulated at the hip. Six samples from each group were used for the testing at each time.

Evaluation and methods: cell detachment assay

Four hours after the initial seeding, the chondrocyte-scaffold complexes were transferred to another 24-well plate for further culture. The culture medium was collected and the detached chondrocytes were counted in a blood counting chamber, and the ratio of detachment was counted using a blood counting instrument (N_un). The ratio of detachment was calculated using formulae 1 and 2, below. A lower ratio meant a higher ratio of planted chondrocytes and a stronger adhering ability.

Formula (1):

Formula (2):

Evaluation and methods: Real-time PCR

The gene expressions of the constructs were examined during in vitro differentiation. One week after culture, the chondrocyte-scaffold complex was harvested for the extraction of RNA. TRIzol (Invitrogen, Carlsbad, California) was used to extract the total RNA according to the manufacturer’s protocol. After resuspending in 30 μl DNase/RNase-free water, the concentration and purity were checked by the ratio of the absorbance at 260 and 280 nm and measured using a spectrophotometer (FP-6500; JASCO Inc., Tokyo, Japan). Subsequently, 2 μg of RNA was reverse transcribed into cDNA using a reverse transcription system (Promega, Fitchburg, Wisconsin). Subsequently, the real-time PCR (RT-qPCR) was carried out on a Bio-Rad iCycler (Bio-Rad Laboratories, Hercules, California). The real-time PCR was initiated with a two-minute 50°C hold, followed by a ten-minute 95°C hold, and then continued with 40 cycles of 15 seconds at 95°C and one minute at 60°C.

Evaluation and methods: Gross view

The activity, feeding behaviour and wound healing of the pigs after surgery were recorded. After being killed by intravenous pentobarbital 12 or 24 weeks after surgery, the knee and implants were harvested. After an examination for findings suggestive of rejection or infection, such as severe inflammation or extensive fibrosis, the degree of cartilage repair was grossly evaluated. The colour, lustre, irregularity, presence of depression or bulging of the repaired tissue in the defect and the state of the border with the surrounding normal cartilage tissue, were examined carefully. The midline of the defects was dissected in order to see the longitudinal cross-section, and the interface between the repaired tissue and adjacent normal osteochondral tissue. The regenerated cartilage was scored macroscopically according to Moran’s criteria at each time point.²⁹ A higher grade represents a better repair. The scoring was performed as a blind test by three independent researchers (JZ, LC, HRW).

Evaluation and methods: biomechanics tests

The biomechanical properties of the samples used for these analyses were immediately evaluated under fresh conditions. They were trimmed along the rim of the regenerated tissue using a trephine. The corresponding normal articular cartilage samples were harvested by the same method. The biomechanical property of the engineered cartilage was tested by measuring the compressive modulus as described previously.^10,30 Briefly, the harvested parts (5 mm in diameter) were fixed onto a sample table of the biomechanical analyser TMI UTM-10T (Toyo Baldwin Co., Tokyo, Japan). A force-displacement curve was obtained by applying a constant compressive strain at a rate of 0.5 mm/min until reaching the maximal force at 400 N. Each sample underwent five indentation tests. The elastic modulus and stiffness were calculated from the curve using formulae 3 and 4.

Formula (3):

Formula (4):

As the defects in the control group at 12 weeks were filled with a small amount of fibrous tissue, which further collapsed at 24 weeks, biomechanical tests on the regenerated tissue could not be performed properly.

After the biomechanical tests, one half of each sample was used for histological examination and immunohistochemical staining, while the glycosaminoglycan (GAG) content of the other half was determined.

Evaluation and methods: Western blot assay

After the biomechanical analysis, the samples were collected. Western blotting methods were used to detect the collagen II (COL II) and aggrecan proteins. The tissue was homogenised with Pierce T-PER tissue protein extraction reagent (Thermo Fisher Scientific Inc., Waltham, Massachusetts) according to the manufacturer’s instructions. The total quantity of protein was determined using the Pierce BCA Protein Assay Kit (Thermo Fisher Scientific Inc.). Samples were loaded onto 10% to 14% SDS-PAGE (Beyotime Biotechnology, Shanghai, China) gels and transferred to a polyvinylidene fluoride membrane (Merck Millipore, Billerica, Massachusetts). The membranes were then blocked with 5% skimmed milk in 0.05% tris-buffered saline/Tween 20 (TBST), followed by incubation with the indicated antibodies at their recommended dilutions in Western blocking buffer (WBB, PBS, 5% powdered milk, 0.1% Tw-20). Samples were then incubated with horseradish peroxidase (HRP)- conjugated secondary antibody, followed by enhanced chemiluminescence detection (ImageJ, National Institute of Health, Maryland). Chemiluminescence detection values were used to quantitate the Western blot results. A ratio of each band of interest to the internal control was obtained and the statistical significance of differences between control and experimental groups determined.

Evaluation and methods: histological analysis

Specimens were fixed with buffered formalin and decalcified in a solution of ethylenediamine tetra-acetic acid (EDTA) in a shaker. The samples were dehydrated with a gradient ethanol series, cleared with xylene, and embedded in paraffin blocks. Sections, 5 mm thick, were obtained and stained with haematoxylin and eosin (H&E) and toluidine blue (TB) for histological analysis.

According to the scoring system described previously by Wakitani et al,³¹ the histological structure, matrix staining, surface regularity and thickness of the repaired cartilage, and the integration of the repaired tissue with its surrounding normal tissue were graded blindly by three observers (JZ, LC, HRW).

Evaluation and methods: immunohistochemical staining

The expression of COL II in the engineered cartilage was examined by immunohistochemical staining at 12 and 24 weeks after implantation. Briefly, paraffin sections were deparaffinised followed by hydration in ethanol solutions of decreasing concentrations (100% to 70%), and pre-treated with 0.1% of Proteinase K (Vivantis Technologies, Oceanside, California) at room temperature for 30 minutes. The sections were then treated with protein blocking agent ( UltraTech HRP Kit, Beckman Coulter France, Villepinte, France) for ten minutes and incubated with goat serum to block non-specific sites. The slides were washed in PBS three times and the sections were incubated at 37°C for one hour with mouse anti-collagen type II monoclonal antibody diluted in PBS (1:200), followed by incubation with 1:100 diluted horseradish peroxidase (HRP)-conjugated anti-mouse antibody (Dako, Carpinteria, California) for 30 minutes. Colour was developed with diaminobenzidine tetrahydrochloride (DAB).

Statistical analysis

Normality testing was carried out using SPSS 16.0 software (IBM Corp., Armonk, New York) and the values were expressed as mean sd (x±s). One-way analysis of variance (ANOVA) was also carried out. Statistical significance was set at a p-value < 0.05.

Ratio of chondrocyte detachment

After the chondrocytes were seeded in the scaffold and incubated for four hours, the ratio of detachment was significantly lower in the CBA group than in either the control or BA groups (p = 0.001 and p = 0.002, and the ratio of chondrocyte detachment in the control group was 3.82 times higher and 2.11 times higher than the corresponding values in the BA and CBA groups, respectively (Fig. 3).

Fig 3

Chart showing the ratio of chondrocyte detachment in control, BA and CBA groups after the chondrocytes were planted in the scaffold for four hours Analysis of variance and least-significant differences (LSD) test compared with control group, *p =0.001; compared with BA group, #p=0.002.

Real-time PCR

Real-time PCR was used to test the expression of mRNA of collagen II, aggrecan and SOX9 in the samples after they were cultured for one week. As illustrated in Figure 4, all three expressions were highest in the CBA group, followed by the BA group, with the lowest expression in the control group (CBA group > BA group > control group, p = 0.001 and p=0.001).

Fig 4

Chart showing the expression of mRNA of collagen II, aggrecan and SOX9 in the samples after they were cultured for one week. Analysis of variance and least-significant differences (LSD) test compared with control group, *p = 0.001; compared with BA group, #p = 0.001.

Gross view

The gross appearance of articular cartilage defects in the weight-bearing area was assessed at both 12 and 24 weeks after surgery. The pigs recovered quickly after surgery and at the time of being killed there were no signs of wound infection, limitation of movement or synovitis in the operated knees. No free tissue-engineered cartilage was found away from the cartilage defects in the knee.

All defects in the blank group were mainly occupied by fibrous tissue with little cartilage-like tissue regenerated at 12 weeks post-implantation. The defect was also wider and deeper with the collapse of its adjacent subchondral bone 24 weeks post-implantation. At 12 weeks after implantation, the defects had glossy white, well integrated, regenerated tissue in the CBA and auto groups, although they remained slightly depressed in the centre. Defects in the control group were mostly filled with rough whitish tissue and was clearly distinguishable from the normal cartilage. Defects in the BA group repaired more quickly than those in the control group and slowly than those in the CBA and auto groups. At 24 weeks post-implantation, the defects in the CBA and auto groups were covered by the smooth, consistent, white hyaline tissue almost indistinguishable from the surrounding normal cartilage. No clear signs of margin with normal cartilage could be identified on the surface of the regenerated areas. In contrast, the defects in the control group were partially repaired with fibrous tissue, leaving a small depression in the defect, and the defects in the BA group had a thin irregular surface tissue with obvious defects and cracks surrounding the normal cartilage (Fig. 5).

Fig.

Macroscopic view of articular cartilage defects in control, BA, CBA, and auto groups at six months post-implantation.

The grading of all cases according to Moran’s criteria in the experimental and control groups is shown in Table I. The auto group had the best repair, indicated by the highest grade, both at 12 and 24 weeks after implantation. Although the repair in the CBA group was slightly poorer than in the auto group, there was no statistically significant difference between them (p = 0.732). Both the CBA and auto groups were better than the BA group (p = 0.003), and the BA group was better than the control group (p = 0.012). Meanwhile, the repair improved with time, and was generally better at 24 weeks than at 12 weeks after implantation (p = 0.004).

Biomechanical test

The biomechanical properties of the engineered cartilage were determined by measuring the elastic modulus and stiffness. The value was also compared with that of normal cartilage. The biomechanical test was not performed on the samples from the blank group because no cartilage was found. The elastic modulus and stiffness in each group increased with time. At 12 weeks post-implantation, the elastic modulus in the control, BA, CBA and auto groups was 61.2%, 72.8%, 81.6% and 90.3% of the normal cartilage, respectively, and there were significant differences among groups (p = 0.012). The stiffness in the control, BA, CBA and auto groups was 68.1%, 77.2%, 88.3% and 94.9% of normal cartilage, respectively (p = 0.004). At 24 weeks post-implantation, the elastic modulus in the experimental groups respectively increased to 70.6%, 79.2%, 87.8% and 95.3% of normal cartilage, which had a modulus of 23.02 sd 2.12 MPa. The stiffness in the experimental groups respectively increased to 75.2%, 83.3%, 92.6% and 97.8% of that in normal cartilage, which was 75.76 sd 6.48 N/mm. There were significant differences in the elastic modulus and stiffness between groups (p < 0.05). (Fig. 6)

Fig.

Biomechanical properties in the four groups at each time point. Analysis of variance (ANOVA) and least-significant differences (LSD) test. Compared with control group, *p < 0.05; compared with BA group, #p < 0.05; compared with CBA group, +p < 0.05 (Western blot analysis).

Western blot analyses were performed at 12 and 24 weeks after implantation, and the levels of COL II and aggrecan were determined. The level of expression for each target protein was presented as the ratio of the grey value of the protein band versus that of the tubulin band. The results are shown in Figure 7. The COL II and aggrecan in the auto group were the highest at 12 and 24 weeks post-implantation, followed by the CBA group. In contrast, the content in the BA and control groups was much lower. Statistical analysis showed a significant difference between the CBA, BA and control groups (p < 0.05) at both 12 and 24 weeks post-implantation.

Fig.

Charts showing comparison of the a) aggrecan and b) COL II content in the control, BA, CBA and Auto groupafter surgery. Analysis of variance (ANOVA) and least-significant differences (LSD) test. Compared with control group, *p < 0.05; compared with BA group, #p < 0.05; compared with CBA group, +p < 0.05.

Histological test

The H&E stain at 12 weeks post-implantation showed similar results in the CBA and auto groups. Chondrocytes were arranged in column form with a relatively higher density, and cartilage lacunae were formed. The histological features of the engineered cartilage in these groups were similar to those in normal cartilage. In the control group, the repaired cartilage was distinctly different from that in normal cartilage, with fewer chondrocytes and lacunae. At 24 weeks post-implantation, the defects in the CBA and auto groups were repaired by newly generated hyaline cartilage (Fig. 8).

Fig.

At 24 weeks post-implantation, the defects in the c) CBA and d) auto groups were repaired by newly generated hyaline cartilage shown by the formation of lacunae and cell clusters. The new-formed cartilage integrated well with its adjacent cartilage, although the interface between engineered and normal cartilage could still be identified. However, there were few cells and lacunae in the control group a). The BA group b) showed better results in terms of healing ability compared with the control group, but worse than those in the CBA group, haematoxylin and eosin ×200.

Deposition of GAG in the engineered cartilage was confirmed by TB staining at 12 and 24 weeks post-implantation. The staining showed superiority of the 24-week engineered cartilage construct over the 12-week samples in terms of cartilage-specific ECM. At 12 weeks post-implantation, the defects in the CBA and auto groups were filled with regenerated cartilage similar to a mixture of hyaline cartilage and abundant ECM, as shown by TB staining. However, the arrangement of the cells in the new cartilage still lacked zonal organisation. The TB staining in the BA group was better than in the control group but worse in the range and density than in the CBA group. At 24 weeks, the repaired sites in the CBA and auto groups showed uniform, positive staining for cartilaginous ECM, comparable with that of normal cartilage. The 24-week TB staining showed that the engineered cartilage in the CBA group was better than that in the BA group, while the engineered cartilage was poorest in the control group (Fig. 9). The positively dyed region and the degree of staining in the control group clearly distinguished the engineered tissue from normal tissue, with some asymmetrical and blurred distribution of matrix. Part of the region still lacked matrix and stained lightly.

Fig.

At 24 weeks, the repaired sites in the c) CBA and d) auto groups displayed uniform, positive staining for cartilaginous ECM, comparable with that of normal cartilage. The results of the 24-week TB staining showed that the engineered cartilage in the CBA group c) was better than that in the BA group b), while the engineered cartilage in the control group a) was the poorest among the groups, TB×200.

Histological scoring results

The repair was evaluated using the scoring system of Wakitani et al,³¹ in which the total is 14, and a lower score signifies a better effect. The auto group had the best effect at both 12 and 24 weeks post-implantation (Table II), although the scores in the CBA group were slightly poorer than those in the auto group, and there were no significant differences (p > 0.05). The scores in the CBA and auto groups were better than in the BA group (p < 0.05) whose scores were better than in the control group (p < 0.05). Furthermore, the engineered cartilage scores in each group at 24 weeks post-implantation were significantly better than their corresponding scores at 12 weeks (p = 0.003).

Results of the immunohistochemical staining

Immunohistochemical staining was consistent with the TB staining, and showed better quality and quantity of repair in the CBA group compared with the BA group and the controls. At 12 weeks post-implantation, an even distribution and expanded darker positive immunohistochemical staining of COL II was seen in the CBA and auto groups, indicating the presence of hyaline cartilage in the regenerated tissue. In contrast, the staining of the control group and the BA group showed an abundance of collagen II in the regenerated tissue, indicating the minimal production, or complete absence, of hyaline cartilage. At 24 weeks post-implantation, with the maturation of engineered cartilage, the expression and distribution of COL II in the CBA and auto groups were similar to those of normal cartilage, with positive staining of the ECM around the evenly distributed chondrocyte-like cells at the site of the repair. Most of the scaffolds in the CBA group were degraded, endocytosed, and replaced. The BA group had better results than the control group but worse results than the CBA group (Fig. 10).

Fig.

At 24 weeks after surgery, with the maturation of engineered cartilage, the expression and distribution of COL II in the c) CBA and d) auto groups were similar to those of normal cartilage, with positive staining of the ECM matrix around the evenly distributed chondrocyte-like cells at the site of the repair tissue. Most of the scaffolds in the CBA group were degraded, endocytosed, and replaced. The BA group b) had better results than the control group a) but worse results than the CBA group. ▲: traces of the scaffold; R, regenerated tissue; N: normal tissue.