Page 474 - IJB-10-4
P. 474

International Journal of Bioprinting                                  3D-bioprinted peripheral nerve scaffold




            NGF release at 51.88 ± 2.12 pg/ml (Figure 7D). This   used, and no nanomaterials were introduced. Although the
            result indicated that scSHEDs had elevated levels of NGF   freeze-drying had its limitations, it was sufficient to display
            release compared to SHEDs. Taken together, most SHEDs   the porous microstructure of this hydrogel. It has been
            underwent differentiation to form scSHEDs.         reported  that porous structures can enhance cell growth
                                                                      43
                                                               and vascularization, suggesting that our hydrogel provided
            3.2. Physical and chemical characteristics of      an optimal environment for promoting cell adhesion,
            the bioinks                                        proliferation, and ultimately nerve fiber regeneration.
            According to the SEM images (Figure 3A), the hydrogel
            surface exhibited a rough and porous structure with   The 3D-printed hydrogel scaffolds exhibited a maximal
            interconnected  pores.  The  porosity  of  the  hydrogel  was   tensile force of 0.02 ± 0.01 N before fracture; the PCL
            measured at 58.32 ± 24.33%, while the pore size averaged   scaffolds demonstrated a significantly higher tensile force
            at 166.58 ± 102.72 μm. Although supercritical drying could   of 4.00 ± 0.34 N; the hydrogel-PCL composite scaffolds had
            better preserve the microstructure of dehydrated hydrogels,   a tensile force of 4.06 ± 0.32 N (Figure 3B). Consequently,
            especially the nanostructures, freeze-drying is still widely   hydrogels are favored for their biocompatibility, whereas
            used. The hydrogel composition in this study is commonly   PCL is commonly employed as a reinforcing material to
















































            Figure 3. Bioink characterization and scaffold fabrication. (A) RGD-Alg/GelMA hydrogel scaffold morphology via SEM (×320). (B) Maximum tensile
            force of hydrogel, PCL, and composite scaffolds. (C) FTIR analysis of RGD-Alg/GelMA and Alg/GelMA hydrogels. (D) Shear–viscosity rate curve. (E)
            Time–viscosity curve. (F) Temperature–viscosity curve. (G) Computer-aided design of the printed path of the conduit model. (H) Nozzle 1 to evaluate PCL
            printing conditions. (I) Structure of the PCL nerve conduit before rolling. (J) PCL and hydrogel layers of the nerve conduit (dried for observation). (K)
            PCL and hydrogel layers after crosslinking. (L) 3D structure of the conduit. Scale bar: 0.5 mm. ***p < 0.001. Abbreviations: Alg/GelMA: Alginate-gelatin
            methacrylate;  FTIR: Fourier transform infrared; PCL: Polycaprolactone; RGD-Alg/GelMA: arginine-glycine-aspartic acid-modified sodium alginate-
            gelatin methacrylate; SEM: Scanning electron microscopy.


            Volume 10 Issue 4 (2024)                       466                                doi: 10.36922/ijb.2908
   469   470   471   472   473   474   475   476   477   478   479