Fabrication Parameter-Dependent Physico-Chemical Properties of Thiolated Gelatin/PEGDA Interpenetrating Network Hydrogels

Abstract

BACKGROUND:

The development of three-dimensional hydrogels using polymeric biomaterials is a key technology for tissue engineering and regenerative medicine. Successful tissue engineering requires the control and identification of the physicochemical properties of hydrogels.

METHODS:

Interpenetrating network (IPN) hydrogel was developed using thiolated gelatin (GSH) and poly(ethylene glycol) diacrylate (PEGDA), with the aid of ammonium persulfate (APS) and N,N,N,N'-tetramethylethylenediamine (TEMED) as radical initiators. Each component was prepared in the following concentrations, respectively: 2.5 and 5% GSH (LG and HG), 12.5 and 25% PEGDA (LP and HP), 3% APS/1.5% TEMED (LI), and 4% APS/2% TEMED (HI). IPN hydrogel was fabricated by the mixing of GSH, PEGDA, and initiators in 5:4:1 volume ratios, and incubated at 37 °C for 30 min in the following 6 experimental formulations: (1) HG–LP–LI, (2) HG–LP–HI, (3) LG–HP–LI, (4) LG–HP–HI, (5) HG–HP–HI, and (6) HG–HP–LI. Herein, the physico-chemical characteristics of IPN hydrogels, including their morphological structures, hydrolytic degradation properties, mechanical properties, embedded protein release kinetics, and biocompatibility, were investigated.

RESULTS:

The characteristics of the hydrogel were significantly manipulated by the concentration of the polymer, especially the conversion between HP and LP, rather than the concentration of the initiator, and no hydrogel formulation exhibited any toxicity to fibroblast and HaCaT cells.

CONCLUSION:

We provide structural–physical relationships of the hydrogels by which means their physical properties could be conveniently controlled through component control, which could be versatilely utilized for various organizational engineering strategies.

Introduction

In tissue engineering and regenerative medicine, the development of three-dimensional (3D) hydrogels with suitable polymeric biomaterials is a key technology that could offer the benefits of providing architectural support and mechanical rigidity [1, 2]. These implantable hydrogel-mediated scaffolding systems could also provide a suitable reservoir for cells and biomolecules to facilitate tissue regeneration. In order to fabricate functionally effective implants into human body, polymeric biomaterial-based hydrogels should provide the following momentous parameters of (1) a similar architecture to native extracellular matrix (ECM) [3] and native-tissue-like properties [4], (2) a porous structure for sufficient cell migration, attachment, and growth [5], (3) low toxicity and high biocompatibility [6], (4) guidance of infiltration for regenerated tissues [7], and (5) controllable delivery capability for cargo molecules, such as cells, drugs, and growth factors [8].

Various natural and synthetic polymers have been widely used for the fabrication of hydrogel. Gelatin (i.e., denatured biomacromolecules from collagen), is well known to contain desirable intrinsic physicochemical properties in terms of biocompatibility, low antigenicity, biodegradability, and cell attachability [9–11]. In addition, this natural polymer is listed by the food and drug administration (FDA) as generally recognized as safe (GRAS) [12].

However, the critical drawbacks of gelatin, such as low mechanical property, low thermal stability, and instantaneous degradability in water, could inhibit its use for versatile tissue engineering applications [13, 14]. These physical problems could be overcome through (1) the chemical modification of gelatin molecules to increase crosslinking density or (2) the formation of interpenetrating polymer networks (IPNs) with complementary substances. Among a series of chemically modified polymers, thiolated polymers have been extensively utilized for hydrogel design, due to the high reactivity of sulfhydryl moiety toward carbon–carbon double bond [15]. For the fabrication of IPN hydrogel, a synthetic poly(ethylene glycol) diacrylate (PEGDA) polymer with excellent controllability was incorporated with thiolated gelatin (GSH). The resulting GSH/PEGDA IPN hydrogels could exhibit high mechanical strength via chain gross polymerization, and outperform their components’ physical and chemical fragility [16, 17]. The free radical polymerization of PEGDA was initiated using a redox initiator pair of ammonium persulfate (APS) and N,N,N,N′-tetramethylethylenediamine (TEMED) [18, 19]. The addition of TEMED accelerates the cleavage of APS, causing the splitting of the persulfate ions into sulfate free radicals [20]. Hence, the formed sulfate-free radicals initiate the polymerization (free radical reaction) of PEGDA. Finally, IPN hydrogels were fabricated using GSH and PEGDA polymerized by an initiator pair, APS and TEMED (Fig. 1).

Fig. 1.

A schematic illustration of the preparation of A thiolated gelatin and B GSH/PEGDA IPN hydrogel platform

Herein, we analyze the physico-chemical characteristics of the GSH/PEGDA IPN hydrogels developed, including their surface morphology, degradation property, Young’s modulus, cytocompatibilty, and protein release profile in various hydrogel formulations. Based on this detailed analysis, we establish key structure–property relationships and provide framework material that can be utilized in a wide range of tissue engineering strategies.

Experimental section

Materials

Gelatin, ethylenediaminetetraacetic acid (EDTA), γ-thiobutyrolactone, β-mercaptoethanol, imidazole, PEGDA, APS, TEMED, FITC-BSA, and Ellman’s reagent were purchased from Sigma–Aldrich. Live and dead staining kit was obtained from ThermoScientific. Dulbecco’s modified Eagle’s medium (DMEM), penicillin–streptomycin solution, and fetal bovine serum (FBS) were obtained from Corning. EZ-Cytox was received from Daeil Lab Service.

Thiolated gelatin synthesis

To synthesize GSH, 1 g of gelatin was dissolved in 100 mL of N₂-purged DW at 40 °C. Subsequently, 1 mL of 1.85% (w/v) EDTA and 2 mL of 34% (w/v) imidazole were added into the gelatin solution and stirred for 10 min. For reaction, 840 μL of ϒ-thiobutyrolactone was injected into the mixture and stirred at 40 °C for 24 h. After 24 h, the reacted solution was transferred to the dialysis bag (molecular weight cutoff 14,000), and dialyzed against 0.2% (v/v) of β-mercaptoethanol for 20 h and degassed DW for the next 4 h. Finally, the produced thiolated gelatin solution was freeze-dried, and stored at − 80 °C, before use. To prevent the oxidization of free thiol groups, all steps were done under nitrogen atmosphere. Thiolation degree was confirmed by Ellman’s assay. Ellman’s working buffer consisting of 0.1 M dibasic sodium phosphate and 1 mM EDTA was prepared, and GSH was added to make a 1% (w/v) GSH solution. Thereafter, 25 µL of Ellman’s reagent, 125 µL of 1% GSH solution, and 1,250 µL of working buffer were mixed, and incubated at 37 °C for 15 min. The absorbance at 412 nm of sample was measured by Multiskan GO microplate spectrophotometry (Thermo Fisher Scientific, Waltham, MA, USA). The degree of thiolation was calculated as 186 μmol/g using the cysteine standard curve.

Fabrication of thiolated gelatin/PEGDA interpenetrating hydrogel

Table 1 shows the concentrations to which hydrogel was fabricated. First, lyophilized GSH sponge was dissolved in PBS at 2.5% (w/v) of low concentration GSH (LG) and 5% (5%) of high concentration GSH (HG). PEGDA solution was prepared in PBS at 12.5% (w/v) of low concentration PEGDA (LP) and 25% (w/v) of high concentration PEGDA (HP). Then, 3% and 4% (w/v) of APS and 1.5% and 2% of TEMED were made in PBS, respectively. 3% APS and 1.5% TEMED pair were used as a low-concentration initiator (LI), while the 4% APS and 2% TEMED pair were used as a high-concentration initiator (HI). Hydrogel was fabricated by the mixing of GSH, PEGDA, and initiator at 5:4:1 volume ratios. These mixtures form a hydrogel at 37 °C for 30 min after injection via syringe. For easy comparison, hydrogels were prepared in 6 concentration groups: (1) HG–LP–LI, (2) HG–LP–HI, (3) LG–HP–LI, (4) LG–HP–HI, (5) HG–HP–HI, and (6) HG–HP–LI. Through the comparison, it is possible to understand how the selection of GSH, PEGDA, and initiator concentrations (low-concentration versus high-concentration) affect the properties of the hydrogel.

Table 1.

Compositions of GSH/PEGDA hydrogels

Concentration (w/v)	GSH (%)	PEGDA (%)	APS (%)	TEMED (%)
HG-LP-LI	5	12.5	3	1.5
HG-LP-HI	5	12.5	4	2
LG-HP-LI	2.5	25	3	1.5
LG-HP-HI	2.5	25	4	2
HG-HP-LI	5	25	3	1.5
HG-HP-HI	5	25	4	2

Scanning electron microscopy

In order to observe the porous morphology of the hydrogel, the prepared hydrogels were lyophilized, until all solvents were completely removed. After that, the dried hydrogels were coated by Pt using Sputter (E-1010, HITACHI, Tokyo, Japan), and surface morphologies were visualized by scanning electron microscopy (SEM) (S-3000 N, HITACHI) at 20 kV accelerating voltage. Pore size and porosity of hydrogels were analyzed via ImageJ software (http://rsbweb.nih.gov/ij/, National Institutes of Health, Bethesda, MD, USA). Pore sizes of hydrogel were measured by randomly selected 50 pores from 3 of independent hydrogel cross section SEM images.

In vitro hydrolytic degradation properties

Mixed hydrogel precursor was injected into the polytetrafluoroethylene mold (8 mm in diameter and 1 mm of height) using a syringe, and incubated at 37 °C for 30 min. The formed hydrogels were then incubated in 5 mL of PBS at 37 °C for 14 days. At each time point (i.e., 6 h, and 1, 3, 7, and 14 d), the hydrolytic degradation properties of IPN hydrogels, including the swelling ratio, sol fraction, and mass remaining, were obtained according to the following equations:

$$\text{Swelling}\text{ratio}=\frac{\text{Ws}-\text{Wd}}{\text{Wd}}$$ 1

$$\%\text{Sol}\text{fraction}=\frac{\text{Wi}-\text{Wd}}{\text{Wi}}\times 100\text{\%}$$ 2

$$\%\text{Mass}\text{remaining}=\frac{\text{Wd}}{\text{Wi}}\times 100\text{\%}$$ 3

where Ws is the weight of swollen hydrogels at each time point, Wi is the weight of the initially dried hydrogels prior to swelling, and Wd is the weight of the dried hydrogels at each time point.

Cargo BSA release kinetics

In order to quantify the release kinetics of cargo protein molecules from IPN hydrogels, a model protein FITC-BSA was incorporated into the hydrogels. FITC-BSA powder was dissolved in PBS, and mixed with PEGDA to make LP and HP. After that, each hydrogel was constructed as previously mentioned, with a total of 10 μg of FITC-BSA per gel. The fabricated hydrogel was subsequently incubated in 1 mL of PBS on a shaker table at 50 rpm and 37 °C. At predetermined time points (0 min, and 1, 3, 5, 7, 10, 14, 21, and 28 d), supernatant was collected, and replenished with the same amount of fresh PBS solution. Collected supernatant was kept at − 20 °C, until use. The amounts of un-encapsulated FITC-BSA (i.e., loading efficiency) and released FITC-BSA were quantified using a microplate spectrophotometry at 490 nm.

Cell viability and cytotoxicity tests

In vitro biocompatibility tests were performed according to the International Organization for Standardization 10993-5 and a previous related studies [21, 22]. Since the releasates of biomaterials could flow into surrounding cells/tissues and cause toxicity, evaluation of the biocompatibility of hydrogel releasates is absolutely necessary for successful biomaterial development. The cytotoxic effects of IPN hydrogel releasates and degradation products on human cell viability were investigated. Human dermal fibroblasts (HDFs) and HeCaT keratinocytes were cultured in basal medium consisting of DMEM (89% (v/v)), penicillin–streptomycin solution (1% (v/v)), and FBS (10% (v/v)). Trypsinized cells were then seeded at 4,000 cells/well on 96-well cell culture plates, and cells were incubated at 37 °C, 5% CO₂, and 95% humidity for 24 h. Additionally, in order to obtain hydrogel releasates, 80 µL of fabricated hydrogel was incubated with 200 µL of basal medium at 37 °C for 24 h. After 24 h of cell seeding, old media was replaced by 200 µL of releasates-containing media (i.e., mixture of fresh media and releasates at a 9:1 v/v ratio), and incubated at 37 °C, 5% CO₂, and 95% humidity for 24 h. To evaluate cell viability, WST-1 assays and Live/Dead (L/D) fluorescence staining were performed. For WST-1 assay, EZ-Cytox solution was added into each experimental group after washing cells using PBS, and incubated at 37 °C for 3 h under protection by light. The optical density (OD) at 450 nm was measured by a microplate spectrophotometry. The cell viability percentage was normalized by the OD value of the control group:

$$\text{Cell}\text{viability}\text{(\% )}=\frac{\text{OD}\text{value}\text{of}\text{experimental}\text{group}}{\text{Average}\text{OD}\text{value}\text{of}\text{control}\text{group}}\times 100\text{\%}$$ 4

Moreover, 2 μM of Calcein-AM and 4 μM of Ethidium homodimer-1 were applied to each experimental group, and cells were incubated at 37 °C for 30 min under protection by light for L/D staining. Stained cells were observed by fluorescent microscopy (Nikon Ti-E, Tokyo, Japan).

Statistical analysis

All quantitative data were expressed as mean standard deviation, and analyzed by one-way analysis of variance (ANOVA) with Tukey’s post-hoc method, using Graphpad Prism V 7.0 (Graphpad Software Inc., San Diego, CA, USA). Statistical significance was assumed at p < 0.05.

Results and discussion

Phase transition of thiolated gelatin/PEGDA hydrogel

The sol–gel phase transitions of GSH/PEGDA hydrogels and subsequent gelation were evaluated by vial tilting method. Figure 2 shows a phase transition from hydrogel precursor solution to gel with different concentrations of GSH, PEGDA, and initiator. Sol was decided by vial tilting method after incubation at 37 °C for 30 min. In the previous studies, 60 kPa hydrogel exhibited the highest osteogenic differentiation of MSC as well as in vivo bone tissue regeneration that 5 and 20 kPa hydrogels [23, 24]. Consequently, to classify soft gel and hard gel, we adopted a criterion based on an elastic modulus of 60 kPa.

Fig. 2.

Sol–gel phase transition of hydrogels. A Macroscopic image observation of inverted hydrogel vial and B phase diagram of the composite

Mainly, higher concentrations of polymer precursors and initiators induced the degree of gelation. However, the HG–LP–LI did not form hydrogel until 37 °C and 30 min incubation, whereas a higher inhibitor concentration with the same polymeric composition (i.e., HG–LP–HI) exhibited gel formation, while the resulting hydrogels containing HP exhibited harder and stiffer properties. These results demonstrate that to sufficiently induce crosslinking between polymeric chains, polymers and initiators above the critical concentration are required [25].

Morphological characterization of hydrogels

The 3D microstructures of the cross-sectional morphology of dehydrated hydrogels were observed by SEM. Although the internal structure of the hydrogel after lyophilization might differ from that of the natural swollen hydrogel state, it is nevertheless still useful to appreciate the 3D design and fabrication, especially in the development of transplantable hydrogel scaffold [26]. Hydrogels with different initiator concentrations but the same GSH and PEGDA concentration exhibited similar pore sizes with porosity (Fig. 3). Moreover, a remarkable decrease in pore size was observed in the HG-HP hydrogel groups with high crosslinking density, which demonstrated that a higher polymeric composition resulted in dense pore structures [27]. Furthermore, it has been shown that these structural properties could regulate the incorporated stem cell fate and behavior [28]. Thus, a suitable pore size could be proposed according to the target tissues for specific treatments: 5 μm for vascularization, 40 μm for nervous tissue regeneration, 20–125 μm for skin regeneration, 300 μm for cartilage regeneration, and 100–350 μm for bone regeneration [29–31].

Fig. 3.

Surface morphology of hydrogels prepared in various formulations. A SEM images of hydrogel cross-section pore structure. B Pore size distribution and C porosity were obtained based on SEM image analysis

Mechanical properties

Implantable scaffold materials must have sufficient mechanical strength to withstand the loads exerted by the tissue to prevent tissue collapse, and provide adequate mechanotransductional cues for cell migration, proliferation, and differentiation [32, 33]. Our GSH/PEGDA IPN hydrogel system exhibited mechanical properties that could be manipulated from soft-gel to hard-gel by controlling the incorporated polymer concentration (Fig. 4). HG–HP hydrogel groups manipulated (regardless of the concentration of initiators) showed the highest Young’s modulus of approximately 131 and 118 kPa, as compared with LG-HP and HG-LP hydrogels. Additionally, LG–HP–HI, LG–HP–LI, and HG–LP–HI hydrogels revealed 60, 66, and 14 kPa of Young’s modulus, respectively, but there was no statistical significance between the concentrations of the initiator in the same polymer formulation. Moreover, the regulation of PEGDA content (i.e., HP and LP) tended to have a greater influence on the elastic properties of the resulting IPN hydrogels, rather than the content of thiolated gelatin (i.e., HG and LG). This observation could be associated with the pore sizes in the gel structure. The IPN hydrogels with higher polymeric contents and subsequently smaller pore sizes could exhibit more even distribution of applied stress over the entire surfaces (Fig. 3) [34]. Hence, these results demonstrated that the present IPN hydrogel systems could be utilized as versatile implantable materials for a functional reservoir with a wide range of stiffness and mechanical hardness. In addition, depending on the target tissue for a specific treatment, mechanical strength could be suggested: ~ 20 kPa for nerve, ~ 40 kPa for muscle, ~ 80 kPa for cartilage, and ~ 190 kPa for bone in our hydrogel [35]. Therefore, the GSH/PEGDA hydrogels proposed in this study might be extensively applied for transplantation from cartilage (i.e., soft tissue) to the skull, and even harder tissues.

Fig. 4.

Elastic modulus of hydrogels with different formulations. (*) indicates a significant difference as compared with between groups (p < 0.05)

Hydrolytic degradation

Swelling ratio, sol fraction, and mass remaining are considered as major hydrogel degradation parameters (Fig. 5) [18, 19, 27]. Although, the swelling ratio and equilibrium swelling ratio were higher in HG–LP–HI hydrogel (approximately 13.8% at equilibrium swelling ratio) than other groups, the swelling ratio was stably maintained for 14 d in all hydrogel formulations (Fig. 5A–C). In addition, the relatively increased sol fraction (Fig. 5D–F) and decreased mass remaining (Fig. 5G–I) and were determined in HG–LP–HI (approximately 30.2% of mass remaining and 70.9% of sol fraction at equilibrium point) hydrogel. On the other hand, LG-HP-HI, LG-HP-LI, HG-HP-HI, and HG-HG-LI hydrogels exhibited similar degradation phenomenon. The degree of decomposition of the hydrogel was not significantly affected by the concentration of the initiator, LG-HP hydrogel groups shown approximately 6.3%, 19.1%, and 88.1% of equilibrium swelling ratio, sol fraction, and mass remaining, respectively. In addition, HG-HP hydrogel groups exhibited approximately 6.3%, 16.8%, and 86.1% of equilibrium swelling ratio, sol fraction, and mass remaining, respectively. Our previous study also showed a similar PEGDA-dependent degradation trend [27].

Fig. 5.

Hydrolytic degradation properties of hydrogel at 37 °C for 14 days including swelling ratio (A–C), sol fraction (D–F), and mass remaining (G–I). Hydrogels with various compositions were fabricated with initiators of low concentration (LI) (A, D, G) and high concentration (HI) (B, E, H). (*) indicates a significant difference as compared with all groups (p < 0.05)

The degradation behavior of hydrogels is closely dependent on the crosslinking density in polymeric chains and networks [36, 37]. A variety of tissue engineering applications require hydrogels that could be degraded in a physiological environment, and have a proven duration [38]. In particular, a precise match between the rate of regeneration of the target tissue and the rate of degradation of the implanted scaffold should be emphasized. Our results demonstrate that a proper selection of the concentration of the hydrogel precursor allows customized hydrogel degradation properties for the desired application. In particular, it is well known that gelatin is a natural polymer that degrades rapidly, whereas PEGDA is a synthetic polymer that degrades remarkably slowly, due to the additional covalent formation through acrylate moieties [18, 19, 27]. A main mechanism of the in vivo degradation of PEGDA-based hydrogel is due to the hydrolysis of the acrylate esters end group [39]. More specifically, Waldeck and Kao investigated why physically entangled gelatin hydrogels lost more than 70% in weight by 48 h at 37 °C [40], while the original shape of subcutaneously implanted PEGDA hydrogel was maintained for 21 d [41]. Consequently, a single component hydrogel of excessively degradable gelatin or less-degradable PEGDA is considered unsuitable for long-term use as an implantable material [39, 40]. Proper control of the hydrogel precursor formulations could be tailored to the desired application by adjusting the hydrogel degradation properties [42–44]. Therefore, these results prove that our IPN hydrogel using both thiolated gelain and PEGDA could be widely used for long-term (i.e., 4–12 weeks) transplantable tissue substitute materials [18, 27].

BSA release profile

In this study, BSA was incorporated into the hydrogel, as a model protein drug (Fig. 6 and Table 2). High initial loading efficiency of (91.8–94.9%) was achieved in all groups. Among various formulations, the HG–LP–HI group exhibited slightly faster release, as compared with both LG–HP–HI and HG–HP–HI, and reached approximately 76% of cumulative release for 28 d. LG–HP–HI (63%) and HG–HP–HI (60%) hydrogels revealed the slowest BSA release phenomenon, compared with other groups. The release kinetics of incorporated drug or therapeutic molecules dominantly depend on the pore size, crosslinking density, and degradation property of the delivery scaffold [45, 46]. LG–HP–HI and HG–HP–HI with small pore sizes (Fig. 3), could allow easier physical entrapment of cargo BSA into polymeric matrix. In addition, since the HG–LP–HI formulation showed a faster degradation performance as compared with LG–HP–HI and HG–HP–HI (Fig. 5), infiltration of the surrounding water molecule and downstream hydrolysis could be facilitated. Hence, integrated cargo BSA could rapidly diffuse out towards the surrounding aqueous environment [46]. Interestingly, the highest burst release profiles were evaluated in LG–HP–LI (97%) and HG–HP–LI (98%). This observation has been reported, indicating that the release kinetics of BSA from PEG hydrogel partially depend on the levels of the incorporated initiators [47]. According to these results, an increase in the initiator concentration induces protein-to-polymer binding, along with a higher crosslinking density (i.e., denser polymeric matrix in the IPN gel), and leads to a decrease in the kinetic release of BSA.

Fig. 6.

% cumulative BSA release profiles from IPN hydrogels with difference formulations. (*) indicates a significant difference as compared with all groups. (#) indicates a significant difference as compared with both LG-HP-HI and HG-HP-HI hydrogel groups (p < 0.05)

Table 2.

Release kinetics (% release per day) of BSA from GSH-PEGDA IPN hydrogel

Experimental group	Phase 1 (~ 3 days)	Phase 2 (3–7 days)	Phase 3 (7–14 days)	Phase 4 (14–28 days)
HG-LP-HI	15.08 ± 0.65	2.26 ± 0.48	1.02 ± 0.28	0.57 ± 0.14
LG-HP-LI	20.04 ± 0.53	3.10 ± 0.40	1.39 ± 0.23	0.59 ± 0.11
LG-HP-HI	7.01 ± 1.21	2.63 ± 0.91	1.37 ± 0.52	0.95 ± 0.26
HG-HP-LI	21.40 ± 0.90	3.21 ± 0.68	1.16 ± 0.39	0.48 ± 0.14
HG-HP-HI	6.82 ± 0.90	2.38 ± 0.68	1.70 ± 0.39	0.91 ± 0.19

Moreover, our previous studies suggested that further immobilization of a secondary protein delivery platform containing coacervate (Coa) and gelatin microparticle (GMP) could accurately control the release kinetics of encapsulated protein drugs [27]. Coacervate is a self-assembled polyelectrolyte complex by electrostatic interaction. Polycation poly(ethylene argininylaspartate diglyceride) (PEAD) interacts with the anionic polymer heparin and cargo growth factor to form Coa [48, 49]. GMP physically absorbs growth factors, and sustainably releases them [19]. Thus, bone morphogenetic protein-2 loaded Coa or GMP-embedded GSH–PEGDA hydrogel system could precisely the control release profile of cargo molecule, and correspondingly could regulate the optimal available dose in a physiological environment for effective damaged tissue regeneration [27]. In addition, the bioactivity of the released cargo molecules from GSH–PEGDA IPN hydrogel was confirmed previously [18]. Here, insulin-like growth factor-1 (IGF-1) was embedded in the GSH–PEGDA hydrogel after encapsulation in the Coa. Subsequently, the released cargo IGF-1 amount was measured, and treated to MCF-7 breast cancer cells. The bioavailability of IGF-1 was estimated through increasing MDF-7 proliferation, compared with same amount of freshly prepared IGF-1. As a result, the activity of IGF-1 released until 7 d was maintained at the same level as that of freshly prepared IGF-1, and the activity of more than 64% was maintained until 21 d. The ability to maintain the bioactivity of the cargo molecules incorporated in the hydrogel could be further improved with a dual protection strategy supporting Coa. Therefore, it could be reasonably anticipated that the present GSH–PEGDA IPN hydrogel could be utilized as a functional reservoir for therapeutic protein molecule, as well as multiple delivery vehicles.

In Vitro biocompatibility

Since APS and TEMED are free radical initiators, it is essential to evaluate the biocompatibility of the polymerized hydrogel with the aid of these initiators, although gelatin and PEGDA are polymers that are very well-known to have excellent biocompatibility [50]. The biocompatibility of hydrogel releasates in aqueous environments was determined by L/D fluorescence staining (Fig. 7A) and WST-1 assay (Fig. 7B, C) in HaCaT cells and HDFs. L/D staining images indicated that all HaCaT cells and HDFs were stained with green fluorescence (i.e., live cell) without red staining (i.e., dead cell), as well as no morphological deformation (Fig. 7A). Furthermore, WST-1 results revealed no significant difference in all groups, regardless of the concentration of hydrogel components (Fig. 7B, C). In addition, our previous study has also shown a proper biocompatibility of GSH–PEGDA IPN hydrogels for encapsulated stem cell populations [18]. Here, adipose-derived mesenchymal stem cells (ADSCs) were laden in the GSH–PEGDA IPN hydrogel. Additionally, GSH–PEGDA hydrogels were implanted into the mouse subcutaneous pocket for 3, 7, and 15 days. After sacrifice at each time point, harvested skin tissue sections were stained with CD68 macrophage marker for inflammation analysis. The control PEGDA hydrogel group exhibited a severe inflammatory response, whereas the GSH-PEGDA hydrogel induced only a mild inflammatory response. Therefore, it is concluded that our IPN hydrogel systems could be widely used as highly biocompatible and long-term transplantable tissue engineering materials, along with suitable mechanical strength and capability of exogenous protein delivery.

Fig. 7.

In vitro biocompatibility test. A Live and dead fluorescence staining (scale bar = 200 μm) and WST-1 assay on B HaCaT cells and c human dermal fibroblasts

Conclusion

In this study, we developed an injectable GSH–PEGDA IPN hydrogels, and investigated how the hydrogel fabrication parameters, including the concentrations of polymeric contents and initiators, influence physico-chemical properties of resulted gels. Mechanical characteristics and structural characteristics, such as the morphology, elastic modulus, and degradation properties of GSH–PEGDA formulations, could be precisely controlled by modulating the composition of synthetic polymer PEGDA, rather than either the thiolated gelatin contents or the initiator (particularly the APS & TEMED thermal initiator pair used in this study) concentrations. In addition, all formulations of GSH–PEGDA IPN hydrogels exhibited high cargo BSA loading efficacy and sustained release manner. Notably, the most sustained release pattern of entrapped BSA was shown in the higher polymeric content groups (HG–HP) with the highest crosslinking density with the aid of a higher level of polymerization by HI, while LI-mediated gels had the most rapid release profile, due to the low protein–polymer interaction. Therefore, it is concluded that (1) the modulation in hydrogel fabrication parameters regulated the physico-chemical properties of the resulting GSH–PEGDA IPN hydrogels, and (2) the controllability of our hydrogel platforms, together with their high biocompatibility, mean that they could be utilized as an excellent candidate for a wide range of tissue engineering fields.

Authors’ contribution

SK: Conceptualization, Experiments, Data analysis, Illustration works, and Writing the paper; YC: Biocompatibility experiments and corresponding data analysis and Illustration works; WL: BSA loading efficacy and release experiments and corresponding data analysis; KK: Ideas, Supervision, Writing-Reviewing and Editing, Funding acquisition.

Declarations

Conflict of interest

The authors disclose no conflicts of interest in this work.

Ethical statement

There are no animal experiments carried out for this article.