Photocrosslinked gelatin hydrogel preparation
Gelatin (beMatrix, Nitta Gelatin, Inc., Japan) was dissolved in Milli-Q ultrapure water at 10 or 5 wt% at 37 °C. The visible light photoinitiator, pentamethylcyclopentadienyl triphenylphosphine ruthenium chloride, and sodium persulfate (10 mM; Advanced BioMatrix, Inc., CA) were gently mixed with gelatin solution immediately before use. LED (Thorlab, Inc., wavelength: 455 nm) was used for the light source.
Rheological properties
Rheological properties of hydrogels were analyzed as previously described55. Briefly, dynamic rheology experiments were performed using a HAAKE MARS photo-rheometer (Thermo Fisher Scientific, Inc.) with parallel-plate geometry and an LED light source at 37 °C. Time-sweep oscillatory tests of the mixture of gelatin with the photoinitiator were performed (n = 3). Strain sweeps were performed on the pre-gel solution to verify the linear response. The gelling point was determined at the time when the torsion modulus (G′) surpassed the loss modulus (G″).
Swelling ratio of hydrogels
The hydrogels (n = 8) were incubated in PBS at 37 °C for 24 h and then lightly blotted dry and weighed (Ws). Hydrogels were then freeze-dried and weighed to determine the dry weight (Wd). The swelling ratio (SR) of the swollen gel was calculated as follows56:
$$mathrm{SR}=frac{(mathrm{Ws}-mathrm{Wd})}{mathrm{Wd}}.$$
Burst pressure test
The burst pressure test was performed as previously described57. Briefly, a piece of 2 × 2 cm2 ICR mouse skin was cut. The skin was fixed to the measurement device linked to a syringe filled with PBS (pH 7.4, 37 °C). A 2-mm incision was made on the skin surface, and the surface was kept wet. Then, 500 μL of the gelatin and photoinitiator mixture was applied onto the incision of the wet surface of mouse subcutaneous tissue, after which the hydrogels formed in situ at the puncture site after visible light (wavelength: 455 nm, 30 mW/cm2) illumination. The thickness of the hydrogels was ~ 3 mm, and burst pressure was measured after gel formation. Peak pressure before pressure loss was considered the burst pressure. All measurements were repeated five times to ensure replicability. Fibrin Glue (Bolheal, Teijin Pharma, Limited, Japan) was tested using the same parameters and conditions.
Cytotoxicity test of hydrogel
The proliferation of L929 mouse fibroblast cells was assessed using a DNA assay. Briefly, L929 cells were seeded into 96-well plates at a density of 1000 cells/100 μL/well and cultured in Dulbecco’s modified Eagle’s medium (DMEM; Thermo Fisher Scientific, Waltham, MA, USA) containing 10% fetal bovine serum (FBS; Thermo Fisher Scientific) for 24 h at 37 °C in a 5% CO2 humidified incubator to obtain a monolayer of cells. Gelatin, pentamethylcyclopentadienyl triphenylphosphine ruthenium chloride, and sodium persulfate was dissolved in DMEM with FBS at final concentration of 10 wt%, 1 mM, and 10 mM respectively. The culture medium was replaced with hydrogel components in culture medium (100 μL/well), and cells were further incubated for 1, 3, 5 and 7 days. The sample solution was removed, and cells were carefully washed using PBS (pH 7.4, 37 °C) three times, and their numbers were determined using a Hoechst 33342 based DNA assay kit (Dojindo, Japan) according to the instruction manual. Fluorescence was measured using a microplate reader (SpectraMax iD5, Molecular Devices, CA) at Ex: 350 nm and Em: 461 nm. For each sample, eight independent cultures were prepared, and proliferation assays were repeated three times for each culture. Experimental data were represented by dividing each value by that of the control group (incubated with normal culture medium) for each day.
In addition, gelatin and a photoinitiator mixture containing L929 cells (100 μL) was irradiated (30 mW/cm2, 30 s) to produce cell-laden hydrogels, and cells were cultured in DMEM with 10% FBS at 37 °C with 5% CO2 for 3 and 5 days. Cell viability was determined using the live/dead cytotoxicity staining kit (Dojindo, Japan). By incubating the cells with this reagent for 60 min, the cells in the hydrogel were stained. Encapsulated cells were imaged under a fluorescence microscope (BZ-9000, Keyence, Japan).
In vivo biocompatibility of hydrogel
All animal experiments performed for this study were carried out in compliance with the ARRIVE guidelines and the protocol was approved by the Committee on the Ethics of Animal Experiments of National Defense Medical College (approved numbers: 19010 and 19065). All methods were performed in accordance with relevant guidelines and regulations of National Defense Medical College. Female ICR mice (~ 30 g) were used for in vivo biocompatibility studies. A 1 cm incision was made in the mediodorsal skin of the mice, and a lateral subcutaneous pocket was prepared. Hydrogel samples (10 × 1 mm cylinders) were implanted under sterile conditions. At designated time intervals (days 1, 3, 7, and 14), mice (n = 5 for each day) were sacrificed, and the fluorescence of the remaining hydrogel in mice subcutaneous tissue was imaged using the IVIS Lumina XR (Ex: 460 nm, Em: 620 nm), and the intensity of fluorescence was analyzed using Living Image Software (PerkinElmer Inc., MA) and were processed for histological analyses.
In vivo hydrogel degradation and sustained release of bFGF
bFGF (Fujifilm Wako pure Chemical Corp., Japan) was desalted and fluorescently labeled using the Alexa Fluor 594 Microscale Protein Labeling Kit (Thermo Fisher Scientific, Inc.) according to the manufacturer’s instructions. The absorption spectrum of Alexa Fluor 594 is completely different from that of the hydrogel, meaning that they do not interfere with each other during fluorescence imaging. Female ICR mice (~ 30 g) were used to study in vivo degradation and bFGF sustained release from a hydrogel. A 1 cm incision was made in the mediodorsal skin of mice, and a lateral subcutaneous pocket was prepared. Hydrogel with 5 μg bFGF samples (10 × 1 mm cylinders) was implanted under sterile conditions. Alexa Fluor 594-labeled bFGF aqueous solution was used as a control group (10 μg in PBS, pH 7.4) and was subcutaneously injected. At designated time intervals (days 1, 3, 7, and 14), mice (n = 5 for each day) were sacrificed. Fluorescence of the remaining hydrogel and bFGF on mouse subcutaneous tissue were imaged using the IVIS Lumina XR (excitation, 460 nm and emission, 620 nm for hydrogel imaging; excitation, 580 nm and emission, 620 nm for Alexa Fluor 594-labeled bFGF imaging), and the intensity of fluorescence was analyzed using Living Image Software (PerkinElmer Inc., MA).
Wound healing in diabetic mice through the sustained release of bFGF
The wound closure efficacy of the photocrosslinked gelatin hydrogels with bFGF was determined in vivo by creating full-thickness skin incisions on the backs of diabetic mice (db mice, C57BLKS/J Iar-+ Leprdb/ + Leprdb, 8 weeks of age, purchased from Nihon SLC). Under anesthesia, these mice were shaved and depilated. Cutaneous wounds were created as previously described58. Immediately after creating the wound, the following treatments were applied topically on the wound for different experiment groups (n = 5 per group): Group A, saline solution; Group B, conventional method of Fiblast spray (bFGF solution; 1 µg/cm2/day) without gelatin; Group C, gelatin and photoinitiator without bFGF were photocrosslinked on the wound; Group D, gelatin and photoinitiator with bFGF (7 µg/cm2) were photocrosslinked on the wound. Groups C and D were photocrosslinked through illumination with visible light (wavelength: 455 nm, 30 mW/cm2 for 30 s). Film dressing was covered to prevent drying and contamination. At designated time intervals (days 3, 5, 7, 10, and 14), mice (n = 5 for each day) were sacrificed. After peeling off the hydrogel from the wound, images of the wound were obtained, and the wound area was evaluated using ImageJ software. The area of the skin defect and the inner diameter were measured at each time point. The skin defect area on day 0 was used as the control (100%) to normalize the ratios of areas obtained from later observations on days 3, 5, 7, 10, and 1459. Based on the results of this experiment, the wounds on post-operative day 7, which had the largest difference between Group B and D, were histologically evaluated. The harvested wounds were fixed with a 5% formalin neutral buffer solution and embedded in paraffin. Further, 5 μm-thick sections were cut from the central region of the wound. H&E staining was performed following the standard protocol. Sections were imaged with a BZ-9000 microscope (Keyence, Japan). Paraffin sections were stained for CD31 for microvessel staining followed by standard immunohistochemical staining with DAB to examine the vascular density of the wound post-treatment. Five fields were randomly selected, and CD31-DAB-positively stained vessels were counted using the BZ-2 Analyzer (Keyence, Japan).
Improved flap survival in mice by sustained release of bFGF
Improvements in skin flap survival mediated by hydrogels with bFGF were determined in vivo by creating a random pattern skin flap on the back of ATP imaging mice. The ATP levels of cells in flap tissues were estimated by expressing FRET-based biosensors in transgenic mice, namely “ATeam,” to visualize ATP levels (ATeam mice were kindly gifted by Dr. Masamichi Yamamoto, Kyoto University, Japan. Papers are in preparation). A modified McFarlane flap60 template with a size of 4 × 2 cm2 (exceeding a 2:1 length-to-width ratio) was traced using a surgical marker on the dorsal surface of the ATeam mice after shaving and depilating under anesthesia. The mice were then divided into four equal groups (n = 5 per group) as follows: Group A, a saline solution was injected subcutaneously; Group B, conventional method of subcutaneous bFGF spray (1 µg/cm2/day) without gelatin; Group C, gelatin and photoinitiator without bFGF were molded subcutaneously on demand via light irradiation (wavelength: 455 nm, 30 mW/cm2 for 30 s) according to the shape of the skin flap area; Group D, gelatin and a photoinitiator with bFGF (7 µg/cm2) were molded subcutaneously on demand via light irradiation (wavelength: 455 nm, 30 mW/cm2 for 30 s) according to the shape of the skin flap area. The flaps were visually assessed 10 days after treatment after analyzing digital images of the flaps recorded using a digital camera. The ATP level in the skin flap was imaged by the IVIS Lumina XR, and the FRET ratio was analyzed with Living Image Software (PerkinElmer Inc.).
The flaps were harvested from euthanized animals on post-operative day 10. The harvested wounds were fixed with a 5% formalin neutral buffer solution and embedded in paraffin. Paraffin sections were stained for CD31 for microvessel staining followed by standard immunohistochemical staining with DAB to examine the vascular density of the wound post-treatment. Five fields were randomly selected, and CD31-DAB-positively stained vessels were counted using the BZ-2 Analyzer (Keyence, Japan).
Statistical analysis
All data are presented as the mean ± SD. Differences between the values were evaluated using one-way analysis of variance (ANOVA, Tukey’s post-hoc test), except in the in vivo biocompatibility experiment where two-way ANOVA with Tukey’s post-hoc test was used. A value of p < 0.05 was considered statistically significant.
