Horseradish peroxidase-catalyzed hydrogelation of fish gelatin with tunable mechanical properties and biocompatibility

Materials

Gelatin from cold-water fish, 3-(4-hydroxy-phenyl) propionic acid (HPA), 1-ethyl-3-(3-dimethylaminopropyl)-carbodiimide (EDC), and N-hydroxysuccinimide (NHS) were obtained from Sigma–Aldrich (St. Louis, MO, USA). Horseradish peroxidase (HRP, type VI, 250–330 U/mg solid), collagenase from Clostridium histolyticum (type II, 0.5–5.0 FALGPA U/mg solid), and hydrogen peroxide (H₂O₂, 30 wt% in H₂O) were also obtained from Sigma–Aldrich. Dimethylformamide (DMF) was purchased from Junsei (Tokyo, Japan). Dulbecco’s modified Eagle’s medium (DMEM), penicillin-streptomycin (P/S), and trypsin/ethylenediaminetetraacetic acid were purchased from Gibco BRL (Grand Island, NY, USA). Fetal bovine serum (FBS) and Dulbecco’s phosphate-buffered saline (PBS) were purchased from Wisent (Saint-Bruno, Quebec, Canada). Live/dead cell viability kit and Alexa Fluor™ 488 Phalloidin were obtained from Invitrogen (Carlsbad, CA, USA). EZ-Cytox enhanced cell viability assay kit (WST-1 assay reagent) was purchased from ITSBIO (Seoul, South Korea). Other chemicals and solvents were used without further purification.

Polymer synthesis and hydrogel preparation

FGH was synthesized by conjugating 3-(4-hydroxyphenyl) propionic acid onto gelatin backbone, using EDC and NHS coupling reaction. Briefly, 5 g of gelatin was dissolved in deionized water (DIW) at room temperature (RT). HPA (3.32 g) was activated by EDC (3.83 g) and NHS (3.2 g) in a co-solvent of DIW and DMF (volume ratio = 3: 2) for 1 h. Subsequently, the activated HPA solution was added to the gelatin solution. The mixture was stirred for 24 h, and then, the resulting solution was dialyzed against DIW for three days, using the dialysis bags (MWCO = 3.5 kDa). After dialysis, the solution was filtered, and lyophilized to obtain the sponge-type FGH polymer. The chemical structure of FGH conjugates was characterized by ¹H NMR spectroscopy (AS400, OXFORD Instruments, UK). The phenolic content of the conjugate was determined by UV measurement at 275 nm (V-750 UV/vis/NIR, Jasco, Japan).

The viscosity of FGH and GH polymer solutions at different concentrations (5 and 10 wt%) was measured by a rheometer (GEM-150–050, Bohlin Instruments, USA), operated in a flow mode at the shear rate of 1/s. The temperature was gradually increased from 20 to 40°C during the measurement.

The FGH hydrogels (200 µl) were fabricated in 1 ml vial. Polymer solutions (5.55–11.11 wt%, 90 µl) and HRP solution (0.05–0.15 mg/ml, 10 µl) was added into a microtube. In another microtube, the polymer solution was mixed with H₂O₂ solution (0.3 wt%, 10 µl). The hydrogels were formed after mixing the solutions of two microtubes. The final polymer concentrations were 5, 7.5, and 10 wt%, corresponding to FGH5, FGH7.5, and FGH10 hydrogels, respectively. The gelation time of hydrogels was determined by vial-tilting method as previously reported.^2,3,10,24,25 The time point at which no flow was observed by inverting the solution was considered as the gelation point.

Characterization of physical properties of hydrogels

The mechanical properties of hydrogels were measured using a rheometer GEM-150–050 (Bohlin Instruments) in an oscillatory mode as previously reported. ²⁴ For the measurement, 300 µl of hydrogels were prepared on the plate of the instrument. The elastic modulus (G’) of hydrogels was performed by dynamic time sweeps, depending on the concentrations of polymer (5–10 wt%) and H₂O₂ (0.01–0.03 wt%) at 0.1% strain and frequency of 0.1 Hz at 37°C. A solvent trap wetted with DIW was used to prevent the evaporation of hydrogels during the measurement.

To test the swelling of hydrogels, hydrogels (300 µl) were incubated in PBS solution (0.01 M, pH 7.4) for 24 h to reach their equilibrium swollen state. Then, we removed the surface excess water by using filter paper and the hydrogels were weighed (W_s). The swelling ratio (SR, %) of the hydrogels was calculated using the following formula

The scanning electron microscopy (SEM, JSM 7001 F, JEOL, Tokyo, Japan) was applied to characterize the porous structure of the hydrogels. Briefly, FGH hydrogels were prepared with different concentrations of polymer (5, 7.5, and 10 wt%). Then, the samples were frozen and freeze-dried. To observe the interior structure, the freeze-dried hydrogels were cross-sectioned and gold-coated before SEM analysis.

In vitro proteolytic degradation

FGH hydrogels (300 µl) were fabricated in the microtubes and subsequently incubated in 1 ml of PBS containing 0.0001 mg/ml collagenase at 37°C. At predetermined time points, the media was removed and the hydrogels were weighed (W_d). Then, the fresh media was added to the hydrogel samples. The percentage of the remaining hydrogels was calculated by the following equation

In vitro 2 D cell study

The FGH hydrogels were formed at the bottom of a 12-well plate. hDFBs cells were seeded at a density of 10³ cells per well and cultured with low-glucose DMEM supplemented with 10% FBS and 1% PS under standard cell culture conditions (37°C and 5% CO₂). The cells were cultured for 24 h to allow adherence and spreading onto the hydrogels. Then, media was removed and the cells in each well were fixed with 4% paraformaldehyde solution for 20 min at RT. The gels were washed with DPBS for two times and cells were permeabilized with 0.1% Triton-X (in DPBS) for 30 min and washed with DPBS for two more times. For the immunofluorescent staining, the cells were incubated with Alexa Fluor (AF) 488 phalloidin (1:40 dilution, Invitrogen) for 30 min to stain actin and diamidopyridine (DAPI) for 5 min to stain nuclei. The cell spreading was observed using a confocal microscope (LSM 710 Zeiss).

In vitro 3 D cell study

The viability and proliferation of hDFBs encapsulated in the hydrogels matrices with different mechanical properties were evaluated by WST-1 and live/dead assays. For sterilization, all solutions were dissolved in DPBS and filtered using a syringe filter (pore size of 0.2 µm). Briefly, hDFBs (10⁵ cells) were carefully suspended in 60 µl of FGH solution containing HRP and put in the 48-well plate. Then, 60 µl of FGH solution containing H₂O₂ was added to the wells. The final polymer concentrations were 5, 7.5, and 10 wt%. After gelation, the hDFBs encapsulated FGH hydrogels were cultured with 1 ml of DMEM supplemented with 10% FBS and 1% PS under standard cell culture conditions (37°C and 5% CO₂) and the media were refreshed every two days. The cell viability after seven days was examined using a live/dead assay kit (Invitrogen). The cell-hydrogel matrices were incubated with 500 µl of mixture consisting of 2 mM of acetomethoxy derivate of calcein (calcein AM) and 4 mM of ethidium homodimer-1(EthD-1) at 37°C for 30 min. The cell morphologies were observed by a fluorescence microscope (TE2000, Nikon, Japan). The cell proliferation after three and seven days was quantified by WST-1 assay. Each hydrogel well was incubated with the media containing 10% (v/v) WST-1 reagent for 2 h. Subsequently, the optical density of solution was measured at 450 nm using a microplate reader (VersaMax Tunable Microplate reader, Molecular Devices, USA). The cellular behaviors cultured on/in GH hydrogels (porcine gelatin) were evaluated as the control group.

Gene expression of hDFBs encapsulated in FGH hydrogels

The relative expression levels of collagen I (Col I), collagen III (Col III), and fibronectin (FN) of hDFBs encapsulated in FGH hydrogels were analyzed by reverse transcription-polymerase chain reaction (RT-PCR). In brief, hDFBs (5 × 10⁵ cells/ml) were encapsulated in hydrogels in a 48-well plate as mentioned above and incubated under the standard cell culture conditions (37°C and 5% CO₂). After 10 days of incubation, the hydrogels were degraded with collagenase solution and the encapsulated cells were obtained by centrifugation. Then, the hDFBs were treated with TRIzol reagent (Invitrogen, Grand Island, NY, USA) for the total RNA extraction, and the cDNA synthesis was processed by a Superscript first-strand synthesis system (Invitrogen). Total RNA of 1 μg was reverse-transcribed into cDNA and 1 μg of cDNA was amplified using a first strand cDNA Synthesis Kit for RT-PCR (AMV, Roche, Mannheim, Germany) in the presence of specific primers, as described in Table S2 (Supporting Information). ⁸ The RT-PCR was performed by a GenePro Thermal Cycler (Bioer, Hercules, China). The results were obtained by normalization with GAPDH (housekeeping gene) as a reference gene, and the relative gene expression levels of Col I, Col III, and FN were expressed as a fold change to control (cells cultured on the tissue culture plate, TCPS).

Statistical analyses

We performed the experiments for hydrogel characterization and in vitro cell studies (WST-1 assay for toxicity of polymer solution and proliferation of encapsulated cells, gene expression) using triplicate samples for each data point. The images for Phalloidin/DAPI and Live/Dead staining were captured from randomly chosen fields for each samples. The experimental data from the studies were analyzed using Student’s t-test. Statistical significance is considered as having *P < 0.05. All the experiments were performed in triplicate, and data were presented as the mean ± SD.

Polymer synthesis and hydrogel preparation

FGH polymer was synthesized as previously reported. ²⁴ By using the EDC/NHS-mediated coupling reaction, HPA readily bonds to available primary amine groups on the gelatin backbone. The chemical structure of FGH conjugate was fully characterized by proton nuclear magnetic resonance spectrometry (¹H NMR) (Figure 1(a)). The successful conjugation was confirmed by the presence of peaks at 6.8 and 7.1 ppm, corresponding to the aromatic protons of the phenol groups of HPA. To determine the degree of substitution of HPA (termed “phenol content”), we measured the UV spectroscopy of polymer solutions (1 mg/ml) at 275 nm (Figure 1(b)). It was found that the phenolic contents of FGH were 174.6 μmol/1 g of FGH, comparable with that of GH (gelatin from porcine skin, 157.8 μmol/1 g).

OPEN IN VIEWER

Figure 1.

Characterization of FGH conjugate: (a) 1 H NMR spectra, (b) UV-Vis spectra of fish gelatin and FGH, (c) water solubility and thermal gelation of fish (FGH) and porcine (GH) gelatin derivatives at 37°C and room temperature (25°C), and (d) viscosity of FGH and GH polymer solutions at different temperatures.

The thermoreversible gelation of FGH and GH solutions at RT (25°C) and body temperature (37°C) are illustrated in Figure 1(c). Both FGH and GH were completely dissolved at 37°C. However, as the temperature decreased to 25°C, only FGH maintained in a liquid form, whereas GH became a gel form. It has been well known that fish gelatin has a lower melting point than porcine gelatin because it has less proline and hydroxyproline amino acids. ²³ Therefore, fish gelatin is easily dissolved in aqueous solution at RT. The dynamic viscosity of gelatin derivative solutions at different temperatures was also measured (Figure 1(d)). Interestingly, FGH solutions showed low and no change in viscosity at both 5 and 10 wt% (FGH5 and FGH10) for a wide range of temperature (20–40°C). In contrast, the variation in dynamic viscosity of the GH solution was strongly dependent on temperature, which increased as the temperature decreased from 40 to 20°C. Particularly, the viscosity of GH10 was not measurable when the temperature was below 25°C due to the thermal gelation. It has been noted that the viscosity of an injectable hydrogel should be sufficiently low before gelation, in order to allow a homogeneous dispersion with encapsulated therapeutic agents. ²⁶ Therefore, the fish gelatin solution with low and temperature-stable viscosity behavior offers great benefits in biofabrication applications as well as generation of uniform gel structures. ²⁷

Hydrogels were formed by enzymatic cross-linking reaction of HRP, in the presence of H₂O₂ (Figure 2(a)). In this system, the phenoxy radicals produced from phenol moieties via HRP/H₂O₂-mediated reaction react with each other, resulting in C-C or C-O bonds for a stable cross-link network. ¹⁵ To assess the gelation kinetics, we varied the concentrations of polymer (5–10 wt%) and HRP (0.005–0.015 mg/ml). We found that the higher concentrations of FGH and HRP induced faster gelation rate, ranging from 40 to 7 s (Figure 2(b)). It could be explained by the increase of phenoxy radicals at higher polymer and HRP concentrations, which accelerates the gel formation. These results also suggest that it would be easy to control the gelation rate for intended use, by simply changing the concentrations of polymer and HRP. Previously, there are few studies for enzymatic cross-linking of fish gelatin hydrogels, using transglutaminase (TGase)^28,29 or laccase and tyrosinase. ³⁰ Their results also indicated that the higher concentration of gelatin and enzyme decreased the gelation time. In this study, we systematically examined the enzymatic curing of fish gelatin in the presence of HRP/H₂O₂ for the first time. Compared to the slow gelation cured by TGase (over 10 minutes for 10% gelatin solution), ²⁹ HRP/H₂O₂ induced a faster gelation rate (within seconds for 10% polymer solution). This such rapid hydrogelation offers advantages in therapeutic delivery (e.g., drugs, cells, proteins, growth factors), because it prevents the diffusion of encapsulated agents and localizes the implants at the injection sites.

OPEN IN VIEWER

Figure 2.

(a) Schematic presentation for in situ formation of FGH hydrogels using HRP/H₂O₂-mediated cross-linking reaction and (b) gelation time of FGH hydrogels with varying concentrations of HRP and polymers (n = 3).

Characterization of physical properties of hydrogels

It is essential to characterize the mechanical properties of ECM mimicking hydrogels since they have significant effects on cellular behaviors, including cell attachment, proliferation, and differentiation. ²¹ Herein, we investigated the elastic modulus (G’) of FGH hydrogels with varying the polymer and H₂O₂ concentrations, using a rheological analysis. As shown in Figure 3(a), the G’ value of hydrogels increased as the polymer and H₂O₂ concentrations increased, similar to the results of gelation kinetic studies. It is comprehensible because of the increase of cross-linking density with greater generation of phenoxy radicals during the HRP-catalyzed reaction. The elastic modulus of FGH hydrogels was ranged from 570 to 4100 Pa, which can be used for a wide range of biomedical applications. Importantly, the mechanical strength of hydrogels can be independently adjusted with gelation time by altering the concentrations of H₂O₂ and HRP, respectively, whereas the change in other enzyme concentrations (e.g. TGase, tyrosinase) affects both gelation kinetics and mechanical properties of hydrogels.

OPEN IN VIEWER

Figure 3.

Physical properties of FGH hydrogels: (a) effect of H2O2 concentration (left) and polymer concentration (right) on elastic modulus, (b) swelling ratio, and (c) pore size of hydrogels with varying polymer concentration. The scale bar is 100 μm. **P < 0.005 versus FGH5 hydrogel (n = 3).

The swelling ratio is another important physical parameter that affects solute diffusion, mechanical properties, and surface properties of hydrogels.^17,31 It is also an indicator for the cross-linking density of the gel network. In this study, the effect of polymer content on swelling ratio of hydrogels was investigated. We found that when the polymer concentration increased from 5 to 10 wt%, the swelling ratio of FGH hydrogels decreased from 98 to 85% after incubating in PBS solution for 24 h, which correlated with the increase of cross-linking network density (Figure 3(b)). This result is also attributed to the increased pore size of the hydrogel matrices (Figure 3(c)), which decreased the water uptake capacity. ³²

In vitro proteolytic degradation

The injectable hydrogels with controllable degradation rate have been demonstrated to facilitate various biomedical applications, including drug delivery, cell-based therapy and tissue engineering. ²⁶ In particular, for cell encapsulation applications, they created a 3 D microenvironment that allows enough space for cell growth, proliferation, and subsequent new tissue formation. ² Gelatin derivatives can be cleaved by various proteases, such as matrix metalloproteinase 2 (MMP-2) and MMP-9 in the body. ⁴ Thus, we tested the proteolytic degradation behaviors of FGH hydrogels using collagenase type II to mimic the in vivo degradation profile. Figure 4 shows the in vitro degradation profile of FGH hydrogels formed by different polymer concentrations. All the hydrogels samples were completely degraded after 24 days, and the degradation rate decreased as increasing the concentration of FGH from 5 to 10 wt%. For instance, the degradation time was 14, 19, and 24 days for FGH5, FGH7.5, and FGH10 hydrogels, respectively. This result may be explained by the higher gel content and cross-linking density, which contributes to the slower cleavage of gelatin molecules by collagenase.

OPEN IN VIEWER

Figure 4.

In vitro proteolytic degradation of FGH hydrogels with different elastic modulus (n = 3).

In vitro 2 D cell adhesion on FGH hydrogel

In tissue regenerative applications, the construction of a scaffold that allows the biological cells to grow or interact within their 3D environment is crucial. ³³ Gelatin has the RGD cell adhesive motif, which promotes the cell attachment for subsequent cellular processes, such as proliferation, migration, and differentiation. ⁵ Herein, we investigated the effect of gel stiffness on the cell attachment and spreading behaviors of hDFBs. As shown in Figure 5, the cells adhered onto the hydrogels and exhibited the cell spreading behaviors that depended on the elastic modulus. On FGH hydrogels with low mechanical moduli (FGH5-0.7 kPa), cells tended to spread polygonally, whereas cells on hydrogels with high mechanical moduli (FGH7.5–3.5 kPa and FGH10-4.1 kPa) were elongated. At the similar mechanical moduli, the cell spread on FGH hydrogels (FGH-0.7 kPa) more than on GH hydrogels (GH-0.65 kPa). The difference of cellular morphologies between FGH and GH hydrogels may result from different amino acid components of the gelatin sources, which influences the surface properties of the respective hydrogels. The effect of the matrix stiffness on the cell behaviors has been thoroughly studied. ³ Similar results were reported by Bauer et al., who assessed the effect of elastic moduli on the spreading of mouse myoblast (C2C12). ³⁴ In another study, Yoon et al. reported cold water fish gelatin hydrogels formed in situ via UV polymerization after introducing a methacrylamide group to fish gelatin (GelMA) ²¹ and investigated the morphology of cells attached onto different matrix stiffness. They found that NIH3T3 fibroblasts on a softer hydrogel were elongated, whereas those on a hard hydrogel were polygonal. The inverse results from their study and ours may be attributed to the different cell lines and the density of cells seeded onto the hydrogels. In our study, we used low cell density to observe separate cells and prevent interconnection between cells, which may affect the morphology of cells.

OPEN IN VIEWER

Figure 5.

The morphology of hDFBs adhered on FGH hydrogels with different elastic moduli, in comparison with porcine gelatin hydrogel (GH). Actin is represented in green (phalloidin) and the nucleus is presented in blue (DAPI). The scale bar is 100 μm (n = 3).

In vitro 3 D cell encapsulation in FGH hydrogels

To further study the potential of FGH hydrogels as bioactive matrices for cell therapy and tissue engineering applications, we encapsulated hDFBs in the hydrogels. The viability and morphology of hDFBs cultured within hydrogel matrices for seven days were observed using the live/dead assay kit. As the results in Figure 6(a), all hydrogels maintained the high levels of viability of cells during the culture time, regardless of matrix stiffness. The cells in FGH10 hydrogels (4.1 kPa) were round, while those were spread into a spindle shape in FGH5 hydrogels (0.7 kPa), indicating that the softer matrix was more favorable for cell growth and migration. These results were in the good agreement with previous studies that investigated the cellular responses according to matrix stiffnesses. ⁵ Meanwhile, the cell proliferation was quantitative analyzed by WST-1 assay for three and seven days of culture periods (Figure 6(b)). By increasing the gelatin content, cell proliferation was decreased because of the suppression of nutrient transport and cell migration as the result of high cross-linking density and slow degradation rate, respectively. In addition, there was no significant difference in cellular morphology and proliferation between fish (FGH5, 0.7 kPa) and porcine (GH5, 0.65 kPa) gelatin hydrogels. This HRP/H₂O₂-cross-linked hydrogel system easily modulated the cellular adhesion, spreading and proliferation by guiding the hydrogel stiffness, comparable to fish gelatin hydrogels formed in situ by photopolymerization. ²¹ However, the HRP-catalyzed cross-linking reaction is favorable to fabrication of hydrogels for internal organs because it does not require the exposure of polymer precursors to light. The combination of cell morphology and proliferation clearly demonstrates that FGH hydrogels have excellent biocompatibility for cellular activities, which can be useful as an injectable scaffold for cells/drug delivery in tissue regenerative medicine.

OPEN IN VIEWER

Figure 6.

In vitro 3 D cell encapsulation in hydrogels with different matrix stiffness. (a) Live/dead staining of hDFBs cultured in hydrogel matrices after 7 d (scale bar represent 100 μm), (b) quantification of cell proliferation in hydrogel matrices after 3 and 7 d, using WST-1 assay. *P < 0.05 versus FGH 5 hydrogel (3 days); ^#P < 0.05; ^##P < 0.005 versus FGH hydrogel (7 days) (n = 3).

Gene expression of hDFBs encapsulated in FGH hydrogels

Next, we investigated the effect of hydrogel stiffness on collagen and fibronectin production level of encapsulated hDFBs. Besides FN, Col I and Col III were reported as the major collagenous gene products of human skin fibroblasts. ³⁵ As shown in Figure 7, the cells produced different concentrations of ECM components (Col I, Col III, FN), depending on the matrix stiffness. Specifically, cells in FGH5 hydrogels (0.7 kPa) showed the higher relative expression of Col I, Col III and FN, which is similar to that of cells cultured on TCPS. When the hydrogel stiffness increased (FGH10; 4.1 kPa), these gene expression levels were decreased. This result could be explained by the decrease in cellular proliferation capacity in harder stiffness matrix that was discussed in 3 D cell encapsulation assay. The modulation of gene expression from cells by tuning the hydrogel stiffness has been widely reported.^8,36–38 Taken together, our results demonstrated that FGH hydrogels are potential scaffolds to support the cellular proliferative capacities and modulate the production of collagen and fibronectin from hDFBs, to form the new ECM.

OPEN IN VIEWER

Figure 7.

(a) The mRNA expression levels of Col I, Col III and FN of hDFBs encapsulated in FGH hydrogels and (b) the normalization of relative gene expressions by GAPDH (housekeeping gene), as compared to the cells cultured on TCPS. *P < 0.05; **P < 0.01 (n = 3).