Efficient regeneration of rat calvarial defect with gelatin-hydroxyapatite composite cryogel

Gelatin from porcine skin (type A, ∼ 300 g Bloom) was obtained from Sigma-Aldrich (USA). Reagents including methacrylic anhydride, oleic acid, calcium chloride (CaCl₂), sodium hydroxide (NaOH), sodium dihydrogen phosphate dihydrate (NaH₂PO₄ · 2H₂O), ethanol and photo initiator I2959 were supplied by Shanghai Aladdin Biochemical Technology Co., Ltd. (China).

HANWs were synthesized via the hydrothermal reaction as previously reported [26]. Briefly, calcium oleate was synthesized by reacting oleic acid with CaCl₂ in mixed ethanol/water, followed by the addition of NaH₂PO₄ · 2H₂O, subsequently, the mixture was transferred to a Telfon-lined stainless steel reactor and kept at 180 °C for 23 h. Before use, the obtained HANWs were sintered 4 h at 700 °C to remove any residual oleic acid. GelMA, with ∼ 80% acrylate substitution basing on the amount of amino group, was synthesized by reacting gelatin with methacrylic anhydride and characterized as reported [30].

HANW-reinforced GelMA cryogel was prepared by choosing an optimized composition from our previous work [26]. Briefly, GelMA was dissolved in warm (40 °C) PBS to obtain a 10% w/v solution, to which, the photoinitiator I2959 was added (0.2% w/v). Then HANWs were ultrasonically dispersed into the solution by controlling the weight fraction of HANWs to GelMA at 1: 2. The suspension was cooled to 4 °C (1 h) and further frozen at -20 °C (24 h). Crosslinking was conducted by exposing the frozen system to ultraviolet light (365 nm, 40 mW cm⁻²) for 10 min, and then lyophilization was applied to create the macropores resulting from ice crystal sublimation. HANW-reinforced GelMA hydrogel was prepared similarly whereas without the cooling, the freezing and the following lyophilization steps. Briefly, the suspension was directly exposed to ultraviolet light (365 nm, 40 mW cm⁻²) for 10 min to complete the gelation. In both the preparations, cylinder Telflon molds were used to shape the cryogel and the hydrogel in desired shape and size for in vitro cell culture and in vivo implantation.

The prepared composite hydrogel was freeze-dried for characterization. Both the composite hydrogel and cryogel were fractured in liquid nitrogen to expose cross-sections, with scanning electron microscope (SEM, S-4700, Hitachi, Japan) being used to examine their porous structures. Both the samples were sputter-coated with Au-Pd alloy before the observation, and their pore size distributions were measured by ImageJ. The practical loading amounts of the HANWs in both the composite cryogel and hydrogel were measured by Q50 thermal gravimetric analyzer (TGA, USA) in air from room temperature to 800 °C at a heating rate of 10 °C/min. Before the TGA test, both the cryogel and hydrogel samples were lyophilized to exclude the influence of water on the measurement. Cyclic loading test was conducted using the compression mode on an electronic universal testing machine (EZ-LX, Shimadzu, Japan). The shape of all the composite cryogel and hydrogel samples submitted to the compression test was cylinder-like with a height of 4 mm and a diameter of 5 mm. The cyclic compression was repeated ten times at the compressive rate of 1 mm min⁻¹ to reach the 60% compression strain, in the meantime, stress-strain curves were recorded and videos were taken.

Bone marrow-derived MSCs (BMSCs) extracted from Sprague Dawley (SD) rats were purchased from Cyagen Biosciences (China) and cultured following an established protocol. The cells were cultured in α-Minimum Eagle's medium (α-MEM, Hyclone, USA) supplemented with 10% fetal bovine serum (FBS, Ausbain, Australia) and 1% penicillin/streptomycin (Hyclone, USA) at 37^oC in a humid incubator with 5% CO₂ supply. The cells at passage 3–4 were digested by trypsin (0.25%) for further use. GelMA/HANW cryogel discs (D = 10 mm, H = 2 mm) were sterilized by being immersed in 75% alcohol for 2 h, and ready for cell seeding after being rinsed thoroughly with PBS. The cryogel discs were fitted into the wells of 24-well cell culture plates, onto each piece of the cryogel disc, 3 × 10⁴ BMSCs were seeded. To prepare the cell-laden GelMA/HANW hydrogel discs, the GelMA solution was sterilized by filtration (filter pore size = 0.2 μm), the HANWs were sterilized by Cobalt-60 irradiation. Then the GelMA and the photoinitiator I2959 were dissolved in α-MEM, to which, HANWs were introduced and dispersed, followed by 3 × 10⁴ BMSCs being mixed into. In the system, the concentration of GelMA solution was 10% w/v and the fraction of HANWs to gelatin was 1: 2 as above mentioned in preparing the GelMA/HANW hydrogel. Subsequently, the cell-laden mixture was poured into a Teflon mold and exposed to ultraviolet light, and cell-laden hydrogel discs (D = 10 mm, H = 2 mm) were obtained. Cell culture for both the cryogel group and the hydrogel group were continued for 1, 3 and 5 d, with the culture media being refreshed every other day. At predetermined intervals, the cell/material complexes were retrieved and stained with calcein-AM/PI (Aladdin, China) for cell live/dead assay. The samples were observed using a laser scanning confocal microscope (LSCM, TCS SP-8, Leica, USA) at the excitation wavelengths of 488 nm and 561 nm, and the emission wavelengths of 510–540 nm and 580–630 nm. Fluorescent images were recorded using the 3D imaging mode, based on which, quantitative data of cell proliferation and migration were obtained by lmageJ processing.

The animal experiment was approved by the Animal Care and Use Committee of Tianjin Medical University (China). SD rats (six-week-old, 250 ± 50 g) (bought from SPF (Beijing) Biotechnology Co., Ltd., China) were randomly divided into three groups, with their calvarial defects being left unfilled (the blank control group), being filled with the GelMA/HANW cryogel (the cryogel group) or the GelMA/HANW hydrogel (the hydrogel group). Before creating the calvarial defects, the rats were prepared by steps of anesthetizing, shaving, disinfecting, incising and lifting the full-thickness skin flaps to expose the skulls. Two circular defects (5 mm in diameter) were drilled at each side of the sagittal suture using an electric trephine following a well-established protocol [31, 32]. The GelMA/HANW cryogel discs (5 mm in diameter, 1 mm in height) were pre-wetted in PBS and fitted into the skull defects. In parallel, GelMA/HANW hydrogel discs in the same size were implanted similarly. After the periosteum and the skin were sutured, the animals were allowed to move and access food freely. At six and 12 weeks post-surgery, three rats from each group were euthanized by CO₂ asphyxiation to collect the skulls for characterizations. The skulls were fixed in neutral formalin solution (10%, Sigma-Aldrich) for 24 h and subsequently evaluated by radiographic, histologic and immunohistologic analyses.

2.5.1. Micro-CT analysis

2.5.2. Histological and immunohistological analysis

All quantitative data were indicated as average ± SD (n = 4). Statistical analysis was performed using SPSS statistic 17.0 and data were analyzed using one-way analysis of variance (ANOVA) and Tukey's multiple-comparison test. The p < 0.05 and p < 0.01 were accepted as statistically significant and highly significant.

The HANWs and the GelMA applied in fabricating the composite cryogel and hydrogel were prepared following our previous works [26, 30], whose details are not shown in the present study. The GelMA had a ∼ 80% acrylate substitution based on its amount of amino group [30]. The HANWs were in one-dimensional morphology with length in microns and width in nanometers [26]. Both the composite cryogel and hydrogel were prepared by dispersing the HANWs in the GelMA aqueous solution at the fixed HANW/gelatin weight ratio (1: 2). The anhydrous GelMA/HANW cryogel sample for SEM observation was obtained by steps of freezing, crosslinking and lyophilization, while the anhydrous GelMA/HANW hydrogel sample for SEM observation was obtained by steps of crosslinking and lyophilization. As the results shown in figures 1(a) and (b), both the cross-sections of the lyophilized cryogel and hydrogel displayed porous structure, while showing different pore sizes. As the insets show, the average pore size of the composite cryogel reached 143.7 ± 32.1 μm, which was significantly larger than that in the composite hydrogel (36.4 ± 8.1 μm). In both the cases, however, the well dispersion of HANWs on pore walls was observed (figures 1(c) and (d)), and the practical loading amounts in them were similar as determined by TGA (figures 1(e) and (f)), being approximately 32 wt.% and close to the initial feeding dose.

Zoom In Zoom Out Reset image size

In carrying out the cyclic compression experiment, the GelMA/HANW cryogel and hydrogel behaved differently in every aspect. As shown in figures 2(a), (c) and Movie S1, the composite cryogel could stand the compression deformation to 60% strain and maintain its integrity throughout the ten cyclic loading, displaying excellent shape recovery performance. The GelMA/HANW composite cryogel presented an obvious hysteresis loop alongside the repeated compression and unloading, and the loops began to overlap after several loading cycles. However, the GelMA/HANW composite hydrogel was crushed at its first compression (Movie S1), barely reaching a compression strain of 50% (figures 2(b), (d)). From figures 2(c) and (d), the highest compression stress determined for the composite cryogel reached above 0.2 MPa, which was about two-fold higher than the failure stress measured for the composite hydrogel (< 0.1 MPa).

Zoom In Zoom Out Reset image size

The BMSCs seeded on the GelMA/HANW composite cryogel or embedded in the GelMA/HANW composite hydrogel, were stained with calcein-AM and PI to carry out the live/dead assay. The green fluorescence observed under LSCM represents live cell, while the red fluorescence displayed will represent dead cells. From figure 3, the BMSCs in the two cases were mainly stained in green fluorescence, revealing the robust cell viability and confirming the non-cytotoxicity of the materials. However, the cryogel and the hydrogel demonstrated totally different influences on cell behaviors. In the cryogel case, continuous growth was identified for the BMSCs from 1 d to 5 d of culture, as indicated by the ever-growing fluorescence intensity and area. The BMSCs initially seeded on the cryogel surface could attach to the substrate firmly, spread in a normal spindle-like morphology and proliferate into confluence gradually. Cell ingrowth into the cryogel was identified along with the incubation, and the cells migrated downward over 200 μm after 5 d (figure 3(d)). In turn, the migration of cells into the cryogel revealed that the pores in the cryogel were interconnected. On the contrary, the BMSCs embedded in the composite hydrogel grew at a quite slow rate, and cell spreading barely occurred. The cells seemed to be entrapped at their initial locations during the cell incubation. By quantitatively analyzing the fluorescence intensities of the images, the relative cell growth rates were significantly higher in the cryogel case than in the hydrogel case.

Zoom In Zoom Out Reset image size

The 5 mm defect created in rat calvarium presented poor self-regeneration ability if the defect was left blank. From figure 4, the defect area remained a hole under micro-CT examination even at 12 weeks post-surgery, presenting quite low BV/TV (18.0%) and BMD (0.27 g cc⁻¹) values. Though the hole in the skull was filled with tissues from the gross appearance (figure 4(b)), the tissue was identified as fibrous connective tissue by both the H&E and the Masson staining (figure 5). When the defects were filled with the GelMA/HANW composite cryogel or hydrogel, osteogenesis was detected for both the groups as shown by the increasing BV/TV and BMD values alongside the implantation time, being significantly higher than those in the blank control group (figures 4(c) and (d)). In the reconstructed micro-CT images, the grayscale values of the neo-bone and the residual materials could be distinguished by using the Mimics Medical (v 17.0) software. Therefore, the fractions of the neo-bone and the residual material were plotted separately in figure 4(c) to easily compare the extents of osteogenesis in different cases. It was noted that the amount of the residual hydrogel was more than that of the residual cryogel at the same timepoint post-operation, though they had similar initial composition. On the contrary, the neo-bone formation achieved higher efficiency in the cryogel group than in the hydrogel group. In the false-color images, the green color illustrates the location of the residual material in the defect site, from which, the cryogel could be seen well integrated with the surrounding bone tissue, while an obvious gap was observed between the implanted hydrogel and the surrounding bone tissue. To show the results more clearly, one group of 2D grayscale micro-CT images are provided as figure S3 to show the cross-sections of the defects in the cryogel, hydrogel and control groups at both six and 12 weeks post-operation. The implanted cryogel displayed a porous structure clearly, while the implanted hydrogel looked like a nonporous block, which seemed to be inconsistent with the SEM observation on the microstructure of the composite hydrogel. This was possibly due to the resolution of the micro-CT scanning was not high enough to show the fine structure. Those light gray spots scattering inside the porous cryogel were ascribed to the newly formed bone tissues.

Zoom In Zoom Out Reset image size

These findings were further proved by histological and immunohistological staining results (figure 5). Firstly, no hint of inflammation could be detected from all the H&E staining images, indicating the excellent biocompatibility of both the GelMA/HANW composite cryogel and hydrogel. From the H&E and Masson staining images, the porous structure of the implanted cryogel was confirmed. The porous structure of the composite cryogel did not collapse during the 12 weeks of implantation, providing an effective support to allow tissue ingrowth. The tissues growing into the pores of the GelMA/HANW composite cryogel expressed rich collagen components and were identified as the newly formed bone tissues in combination with the OPN and the OCN staining results. At six weeks post-operation, the immature neo-bone within the cryogel expressed rich OPN (figure 5(a)), while rich OCN was expressed at the sites to indicate their maturity at 12 weeks post-operation (figure 5(b)). However, the staining images presented for the hydrogel group provided totally different information. The majority of the implanted composite hydrogel maintained integrity without showing obvious porous structure until 12 weeks post-operation, but cracks were observed. Accordingly, limited neo-bone tissue was detected growing into these cracks instead of penetrating into the hydrogel, as revealed by the blue stained collagen elongating into the cracks and corresponding OPN/OCN expressions.

Zoom In Zoom Out Reset image size