Using chondroitin sulfate to improve the viability and biosynthesis of chondrocytes encapsulated in interpenetrating network (IPN) hydrogels of agarose and poly(ethylene glycol) diacrylate

2.1 Materials

2-hydroxyethyl agarose (Type VII) and high purity poly(ethylene glycol) diacrylate (PEGDA) (molecular weight 2,000 Da) were obtained from Sunbio Inc. (Anyang City, South Korea). The photoinitiator, 2-hydroxy-1-[4-(2-hydroxyethoxy)phenyl] 2-methyl-1-propanone (Irgacure 2959), was purchased from Ciba Specialty Chemicals Corp. (New York, NY). Chondroitin sulfate A sodium salt (CS) (Type A 70%, balanced with Type C, from bovine trachea; Sigma, St. Louis, MO) was used as received. Glycidyl methacrylate (97.0% purity) was purchased from Sigma-Aldrich Chemical Co. (Milwaukee, WI) and also used as received.

2.2 Bio-macromer synthesis

Chondroitin sulfate A sodium salt (CS) was methacrylated using methods adapted from Li et al. [29]. Briefly, 5 g of CS was dissolved in 50 ml of phosphate buffered saline (PBS). 5 ml of glycidyl methacrylate (GMA) was added to the CS solution and the reaction was stirred at room temperature for 15 days. Next, methacrylated chondroitin sulfate (MCS) was precipitated in acetone (1:20 v/v), then filtered and re-dissolved in 100 ml of Milli-Q water followed by chloroform extraction (1:1 v/v). The aqueous polymer solution was separated from the chloroform via a separatory funnel and concentrated back to about 50 ml by rotavap, followed by a second acetone precipitation (1:20 v/v), centrifugation, and acetone removal. Acetone was removed by drying in air at room temperature. The resulting precipitant was collected and dried at room temperature for 48 h. The MCS was re-dissolved in water and lyophilized (−46°C, 0.0140 mBar).

¹H nuclear magnetic resonance (NMR) was performed on the dried sample to quantify degree of methacrylation. High-resolution, H NMR spectra were taken on a AV500 NMR spectrometer operating at 500 MHz. Deuterated water (90% DI + 10% D₂O) was used as the solvent, and the macromer concentrations were varied between 2.5 and 3 wt%. The ratio of the average integrated region of the double bond proton to the integrated region of the anomeric proton of CS was used to calculate the degree of substitution (DS) of the methacrylate (MA) per repeating unit on MCS.

2.3 Synthesis of the agarose network

Briefly, 0.3 g agarose powder was added to 10 ml of PBS (pH = 7.4, 0.01 M) and autoclaved for 30 min. Once the agarose cooled to 39°C, it was combined in a 2:1 ratio with PBS to yield a 2% w/v agarose solution, and then pipetted into disk-shaped molds formed from a 1.9 mm thick silicone rubber gasket with 5 mm diameter circles punched out and sandwiched between glass plates. After cooling the molds at 4°C for 10 min, the gels were removed and added to a reservoir of PBS. Gels were allowed to equilibrate in PBS for at least 24 h before use.

2.4 Synthesis of a bio-functionalized MCS-PEGDA network and the IPN

A solution of 0.1% w/v Irgacure-2959 photoinitiator in deionized water (DI) was dissolved in a 15% w/v solution of MCS and PEGDA (1:10 mass ratio) in PBS at room temperature. For each ml of monomer solution, one agarose gel disk, prepared as described above, was added and soaked under constant agitation using a rocker for 2.5 h to achieve equilibrium. Afterward, the equilibrated gels were placed in rectangular silicon molds between optical glass microscope slides (1 mm thick), and the surrounding volume was filled with excess MCS-PEGDA/PBS solution from the soak vials. The gels were exposed to ultraviolet light for 5 min on each side using 312 nm light, 3.0 mW/ cm² (Spectrolinker XL-100; Spectronics Corp.). Similarly, pure PEG-DA and IPN networks in the absence of CS were synthesized using our previously described procedure [35]. Using a 2 mm biopsy punch, gels were then cut from both the pure CS-PEGDA area and the CS-IPN area and added to excess PBS. Gels having 2 mm diameter and 2 mm height were allowed to equilibrate in PBS and deionized (DI) water for at least 24 h before use. A two-step process was selected for synthesis of IPN gel because agarose and PEGDA will otherwise phase separate if blended prior to polymerization.

2.5 Swelling degree and chemical composition of gels

Solid content was analyzed to quantify the final polymer content in each of the hydrogel groups. Gels were placed in excess DI water for at least 24 h to remove extractable materials from polymer networks and to swell to equilibrium. Equilibrated gel samples were weighed and placed into a desiccation chamber over calcium sulfate at room temperature. After at least 48 h, the dried gel samples were removed and weighed again. The solid fraction is simply the ratio of the dry mass to wet mass, and its mathematical inverse is the mass swelling degree, Q. The final content of agarose and PEGDA in the IPN were determined using solid content analysis, while incorporated CS content was calculated on the basis of biochemical (GAG) analysis.

2.6 Chondrocyte isolation and culture

Articular cartilage samples were dissected aseptically from ankles of 8–9 month old male Duroc hogs, which were obtained from a local butcher. Cells were harvested within 36 h after slaughter following aseptic procedures in our recent reports [36, 37]. The articular cartilage samples were diced into approximate 1 mm³ pieces using a scalpel in autoclaved PBS. After rinsing with PBS three times, the cartilage samples were digested in a sterile filtered solution of 30 ml of 2 mg/ml type II collagenase (305 U/mg; Worthington Biochemical) for 18 h on an orbital shaker in a humidified 37°C, 5% CO2 incubator. The digested cartilage solution was filtered through a 100 µm filter cell strainer to remove undigested cartilage lumps. The filtrate was then centrifuged at 1,500 rpm for 5 min, and then the cell pellet was resuspended in chondrocyte culture medium in a humidified incubator (37°C, 5% CO2). The culture medium consisted of Dulbecco’s modified Eagle’s medium (DMEM) with 4.5 g/l d-glucose supplemented with 10% fetal bovine serum (FBS), 1% nonessential amino acids, 1% sodium pyruvate, 50 ug/ml ascorbic acid and 0.25 mg/ ml penicillin-streptomycin fungicide. The DMEM and supplements were obtained from Invitrogen (Grass Islands, NY). Cell number was determined using a hemocytometer. The freshly isolated cells were grown to 90–95% confluence in T75 flasks and then retrieved by trypsin-ethylene-diaminetetraacetic acid (EDTA) digestion. First passage chondrocytes were used in this study. Culture medium was changed every 2 or 3 days.

2.7 Encapsulation of chondrocytes into CS-incorporated IPN

Chondrocytes were encapsulated in CS-incorporated IPNs using our previously published methods [35]. The chondrocytes were detached from T75 flasks using trypsin-EDTA and resuspended in PBS at a high concentration, while an agarose solution was prepared by adding 0.3 g agarose powder (cell culture grade) to 10 ml PBS and autoclaving for 30 min. The agarose solution temperature was monitored under an aseptic environment until 39°C was reached. Cells from the PBS suspension were counted using a hemocytometer. Once the agarose solution temperature reached 39°C, the chondrocyte suspension was added to the liquid agarose in a 1:2 ratio for a final concentration of 25 million cells/ml in 2% agarose. The cell suspension was mixed thoroughly with the agarose solution and pipetted into sterilized disk-shaped molds formed from a 1.9 mm thick silicone rubber gasket with 5 mm diameter circles punched out. The molds were clamped and cooled at 4°C for 9–10 min, and then chondrocyte encapsulated agarose gel constructs were transferred to non-tissue-cultured 24-well plates. Each well was supplied with 1.5 ml of fresh growth medium, and placed in an incubator overnight.

After 24 h, chondrocyte-encapsulated agarose constructs were soaked in a sterile-filtered 15% w/v solution of photopolymerizable MCS and PEGDA macromers (1:10 mass ratio) for 2.5 h, previously shown to be adequate for equilibration with PEGDA [35]. Cellular IPN constructs were made aseptically using the same procedure as described in Sect. 2.4. Using a 3 mm biopsy punch, gel samples were then cut from the center of the IPN area. The IPN constructs with encapsulated cells were then cultured for 6 weeks at 37°C with 5% CO2 in 1.5 ml chondrocyte growth medium (Dulbecco’s modified Eagle’s medium [DMEM] with 4.5 g/l d-glucose supplemented with 10% fetal bovine serum (FBS), 1% nonessential amino acids, 1% sodium pyruvate, 50 µg/ml ascorbic acid and 0.25 mg/ ml penicillin-streptomycin fungicide), which was replaced every 2–3 days.

2.8 Mechanical testing (compression to failure)

The compressive modulus of the IPN hydrogels was determined at room temperature on a RSA-III dynamic mechanical analyzer (TA Instruments) and tested under unconfined uniaxial compression with a 35 N loading cell (n =5). The gel diameter was measured with calipers under a stereomicroscope (20× magnification) and the height was measured directly using the RSA-III. All measurements and mechanical testing were performed on IPN gels swollen to equilibrium in PBS, and compression plates were lubricated with mineral oil both to minimize any gelplate adhesion and to prevent gel drying during testing. Hydrogels were then compressed in the direction normal to the circular face of the IPN gel at a rate of 0.0005 mm/s (1.7%/min) until mechanical failure occurred (indicated by a sudden drop in the compressive force). The compressive elastic modulus, defined as the slope of the initially linear region of the stress-strain curve of a material under compression, was calculated from the initial linear portion of the curve (<20% strain). Fracture points were identified at the peak stress after which a significant (>10%) decrease in stress occurred. Shear modulus values were calculated using the neo-Hookean model for ideal elastomers. The plot of stress versus the strain function, λ – 1/λ², where λ = L/L₀ (as suggested by neo-Hookean model), yields a straight line under ideal conditions, where the slope of the line is equal to the gel shear modulus [38]. The values were used in the calculations for a set of at least 5 samples.

2.9 Live/dead assay

To compare the viability of cells encapsulated in the CS-IPN and the pure IPN, a live/dead assay was performed immediately after 0, 3, and 6 weeks of culture (n = 3) using a live/dead viability cytotoxicity kit (Molecular Probes). This kit contains 2 mM Calcein-AM to stain the living cells and 4 mM ethidium homodimer-1 to stain the dead cells. Cylindrical hydrogel constructs (n = 3) were sectioned horizontally into two equal halves and incubated in Live/Dead reagents for 30 min before imaging to promote thorough staining. Fluorescence spinning-disk confocal microscopy was used to visualize the green living cells and the red dead cells, using a Yokugawa CSU10 spinning-disk attached to an Olympus IX 81 microscope with 488 nm excitation/515–540 nm emission and 561 nm excitation/585 LP emission filters with a charge-coupled device camera (CoolSnap HQ2; Photometrics) controlled by Slidebook software (version 4.2; Intelligent Imaging Innovations, Inc.). Z-scans were performed to 350–500 µm resolution depth in areas representative of the overall IPN gels. Images were acquired in 2 × 2 binning mode. The three-dimensional images were deconvoluted using a constrained iterative algorithm (Slidebook). The percentage of total viable cells was calculated using the SlideBook (version 5.0) Mask Statistic module.

2.10 Biochemical assays

Biochemical assays were performed on the CS-incorporated IPN (CS-IPN) as well as pure IPN constructs (n = 4). At 0, 3, and 6 weeks, four samples were removed aseptically from culture and placed in microcentrifuge tubes. Samples were mechanically crushed and then homogenized with a papain mixture [125 mg/ml papain (from papaya latex), 5 mM N-acetyl cysteine, 5 mM EDTA, and 100 mM potassium phosphate buffered saline (PBS)] and allowed to digest overnight in a 60°C water bath. All reagents were obtained from Sigma. The digested constructs were then stored at −20°C. The digested constructs were thawed in a 37°C water bath and centrifuged at 10,000 rpm for 10 min to pellet fragments of polymers before conducting assays. Supernatant was used to determine GAG, DNA, and collagen content. DNA content was quantified using the Picogreen assay (Molecular probes) according to the manufacturer’s instructions. Total glycosaminoglycan (GAG) content was determined using the Dimethylmethylene Blue (DMMB) (Biocolor, Carrickfergus, UK) spectrophotometric assay, with chondroitin sulfate as a standard [39]. The GAG content associated with the incorporated CS was measured using samples (n = 4) collected from acellular CS-IPN hydrogels (5.9 ± 3.8 ug/construct, 5.8 ±1.7 µg/construct and 3.9 ± 2.9 µg/construct at 0, 3 and 6 weeks time periods, respectively). These values were subtracted from the GAG values of cellular CS-IPN samples to get the reported GAG contents for the experimental groups. As an indicator of collagen content, a hydroxyproline assay was assessed as described in our previous publication [40]. Both GAG and collagen contents were normalized to DNA content.

2.11 Histology

Histological analysis was performed using safranin O and hematoxylin & eosin (H & E) staining on cell encapsulated constructs at 0, 3, and 6 weeks (n = 2). Constructs were fixed in 10% formalin for 2–3 h, and soaked in optimal cutting temperature (OCT) medium [Ped Tella Inc., Redding, CA] overnight at 44°C. Constructs were then placed in a −20°C freezer until further processing. Sections were cut at −20°C using a cryostat (MICROM HM 550) to 8 µm thickness, mounted on a microscope slide and allowed to dry for 1 h at room temperature. Following standard histological procedures, the sections were stained with hematoxylin and safranin O, which stains nuclei purple and negatively charged GAGs orange, respectively. Eosin stains cytoplasm, connective tissues and other extracellular substance red or pink.

2.12 Statistical analysis

All quantitative data were expressed as mean ± standard deviations and were verified by single-factor analysis of variance (ANOVA) followed by Tukey’s post hoc analysis. Analysis was performed using the SPSS 17.0 statistical software package. P values of less than 0.05 were considered statistically significant.

3.1 CS modification and IPN synthesis

The synthesis of MCS macromer was illustrated in Fig. 1, where the epoxy rings in GMA are opened by transesterification and ring opening reactions to functionalize hydroxyl, carboxyl, and sulfate groups of chondroitin sulfate with methacrylate groups [41]. Figure 2 shows the proton NMR of the methacrylated CS with an inset of the chemical structure, which confirmed the successful grafting of the methacrylate group on CS. Two distinctive peaks, at 5.65 and 6.10 ppm, were attributed to methacrylate protons and at 5.1 ppm to anomeric protons from CS. The bioactive IPN was synthesized in this study using MCS, which was 6.3% methacrylated. The vinyl double bond on MCS reacted with PEGDA during photopolymerization in the presence of Irgacure 2959 photoinitiator and formed a poly(ethylene glycol)-CS (PEG-CS) copolymer network (Fig. 3). The synthesis of the bioactive interpenetrating network using agarose, MCS macromer, and PEGDA is shown schematically in Fig. 4.

3.2 Swelling degree and gel composition

Chondroitin sulfate (methacrylated) was incorporated into the agarose/PEG-DA IPN network by copolymerizing MCS with PEGDA after soaking agarose hydrogels in the macromer solutions, as outlined in Fig. 3 and Fig. 4. Acellular hydrogels as well as IPN constructs were characterized for their swelling behavior and solids content. The swelling degrees of various equilibrium-swollen networks are tabulated in Table 1. The final PEG-DA and PEG-CS contents in pure PEG-DA and PEG-DA-CS gels were 7.3 and 6.9%, respectively (wet basis). The solids content data in Table 1 show that agarose gels were composed of 2.4% agarose, hence by difference the PEG contents of IPN gels were estimated as 7.1% without CS and 6.4% with CS. These values were equal within experimental error to those of pure PEG and PEG-CS gels. Total CS content of the acellular CS-IPN gels, measured by the GAG assay, was 5.2 ± 2.8 µg/CS-IPN construct (n = 4) or about 0.5 wt% of the total CS-IPN solid content.

The PEG and PEG-CS contents in the pure IPN and CS-IPN hydrogels were calculated by subtracting the solid contents of the pure agarose gel. CS-PEG showed an 11% greater swelling degree than the pure PEG network, but this increase in swelling degree was not statistically significant. Analogous increases were observed in the CS-incorporated IPN when compared with the normal IPN network, although these differences were not statistically significant (Table 1). Macroscopic and microscopic images of gels are displayed in Fig. 5. Equilibrium-swelled unmodified IPN and CS-incorporated IPN with encapsulated chondrocytes are shown in Fig. 5a. Figure 5b displays the homogeneous distribution of encapsulated chondrocytes within the IPN gel, while Fig. 5c displays their round morphology.

3.3 Compressive modulus

The representative stress–strain curves for the various hydrogel groups are displayed in Fig. 6. All hydrogels except agarose (because of early fracture) displayed a concave upward curve characteristic of elastomeric materials with large deformation before fracture. The Young’s modulus (E) and shear modulus (G) were calculated from the slope of the initial linear neo-Hookean region of the stress–strain (<20% strain) curves (Fig. 7). Mechanical property data of all gels are summarized in Table 1. Ideal elastomers have E/G ratios of 3, although compression tests on hydrogels typically give values somewhat above 3 [42]. In Table 1, the E/G ratios average 3.6 and do not vary significantly among networks, indicating that all gels and IPNs behave as ideal elastomers and in agreement with our previously reported work on agarose-PEG IPNs [35]. The compressive shear modulus for the equilibrium swelled CS-IPN gel was 45.4 ± 7.5 kPa, while for the unmodified IPN it was 37.9 ± 9.3 kPa (Table 1). Shear moduli for acellular IPN gels were 3.7 times higher than for pure PEGDA and 4.8 times higher than pure agarose gels (P< 0.05), while the shear modulus of the CS-IPN group was 4.4 times higher than for the pure PEG-DA gels and 5.8 times higher (P <0.05) than for the pure agarose gels (Fig. 8). Similar increases in the Young’s modulus were observed (see Table 1). As far as failure properties were concerned, IPN networks were 2.5 times stronger and 3.2 times tougher than pure PEG networks, while CS-IPN networks were 2.8 times stronger and 3.6 times tougher than pure PEG networks (P < 0.05) (Table 1). Failure of pure CS-PEG copolymer networks occurred at lower strain than pure PEG networks and they shattered into small fragments. Similarly, CS-incorporated IPN networks were shattered into extremely small fragments (~85% strain) before unmodified IPN networks.

3.4 Cell viability

A live/dead assay was used to determine the cell viability at different time intervals after the encapsulation (Fig. 9). When compared with the incubation time periods of 0, 3 and 6 weeks, live–dead (Calcein-ethidium dye) cell staining indicated that the majority of cells encapsulated in CS-incorporated IPN hydrogels remained viable (stained fluorescent green with Calcein) over the culture period of 6 weeks (Fig. 9). The viability of chondrocytes in unmodified Agarose/PEGDA IPN hydrogels is high, but less than desirable. Mask statistics analysis showed that, in the absence of any biomolecules, viability drops to 45.3 ± 2.7 and 35.7 ± 4.5% at 3 and 6 weeks post-encapsulation in culture, respectively, while, by incorporating chondroitin sulfate within the IPN, the viability was 70.2 ± 10.3 and 55.6 ± 14.8% at 3 and 6 weeks, respectively (Fig. 10). These values were statistically significant from 0 week time point values (P < 0.05).

3.5 Total GAG and collagen production

DNA content analysis indicated that the resulting number of encapsulated cells in the IPNs with CS was similar to that of the unmodified IPN control at week 0, suggesting comparable cell seeding efficiencies in both unmodified as well as CS-IPN. A significant increase in DNA content was observed in both the IPN control group as well as in CS-IPN group between 0 and 6 weeks (P < 0.05), while it did not change significantly between 3 and 6 weeks (Fig. 11). DNA content decreased in CS-IPN as compared to the unmodified IPN group at 6 weeks.

Values from quantitative GAG analysis were determined and presented in Fig. 12. Both CS-IPN as well as unmodified IPN constructs accumulated sulfated GAGs over the 6-week culture period. GAG content per construct (gel) was noted to increase significantly over time(P < 0.05) and no significant differences were observed between unmodified IPN and CS-IPN at 3 and 6 weeks (Fig. 12a). Total accumulated GAG content normalized to total DNA content is presented in Fig. 12b. For GAG content calculation, the GAG value of acellular Ag/CS-PEGDA IPN hydrogels (GAG produced by incorporated CS) was subtracted from the total value. Normalized GAG content at week 0 was different between the unmodified IPN and CS-IPN gel groups, however the difference was not statistically significant. The normalized GAG contents for all groups increased over time (P < 0.05, n = 4) and significant differences were observed between the unmodified IPN and CS-IPNat the week 6 culture period. At week 3 and week 6, the CS-IPN had significantly higher normalized GAG contents (242 ± 129 and 347 ± 34 µg/ lg) than control IPNs (86 ± 13 and 158 ± 27 µg/µg). Results from the quantitative analysis of collagen (by hydroxylproline assay) are presented in Fig. 13. Similar to GAG, collagen per construct (gel) increased significantly (P < 0.05) in both of the groups from 0 to 6 weeks (Fig. 13a). Normalized collagen content increased significantly in unmodified IPN and CS-IPN hydrogels from week 0 (7.6 ± 3.6 and 9.8 ± 7.0 µg/µg) to week 6 (18 ± 3 and 22 ± 13 µg/µg) (P < 0.05, n = 4), while no significant differences were observed between unmodified IPN and CS-IPN groups at 0, 3, and 6 weeks (Fig. 13b).

3.6 Histological evaluation

Histological images of chondrocytes photoencapsulated in unmodified IPN and CS-IPN gels are shown in Fig. 14. GAG production by chondrocytes in the IPN and CS-incorporated IPN hydrogels was examined using safranin O. Intracellular staining was observed in IPN gels without CS (Fig. 14a), while the encapsulated chondrocytes in CS-IPN hydrogels exhibited dense staining in pericellular and interterritorial region (Fig. 14b). H & E staining exhibited homogeneous cell distribution and size in both the unmodified IPN group (Fig. 14c) and the CS-incorporated IPN group (Fig. 14d). Encapsulated chondrocytes in IPN gels without CS displayed the normal rounded morphology (Fig. 14c), while they showed larger diameter pericellular matrices (Fig. 14d) in CS-IPNs at week 6.