Precisely printable and biocompatible silk fibroin bioink for digital light processing 3D printing

Abstract

Although three-dimensional (3D) bioprinting technology has gained much attention in the field of tissue engineering, there are still several significant engineering challenges to overcome, including lack of bioink with biocompatibility and printability. Here, we show a bioink created from silk fibroin (SF) for digital light processing (DLP) 3D bioprinting in tissue engineering applications. The SF-based bioink (Sil-MA) was produced by a methacrylation process using glycidyl methacrylate (GMA) during the fabrication of SF solution. The mechanical and rheological properties of Sil-MA hydrogel proved to be outstanding in experimental testing and can be modulated by varying the Sil-MA contents. This Sil-MA bioink allowed us to build highly complex organ structures, including the heart, vessel, brain, trachea and ear with excellent structural stability and reliable biocompatibility. Sil-MA bioink is well-suited for use in DLP printing process and could be applied to tissue and organ engineering depending on the specific biological requirements.

Fig. 1

Fabrication of chemically modified silk fibroin (SF) by glycidyl methacrylate (GMA) (Sil-MA) as pre-hydrogel. a Modification of SF molecule with GMA. SF is covalently immobilized with GMA, which is a donor of vinyl double bond as a UV-crosslinking site. b Schematic representation for methacrylation of SF. Degummed silk was dissolved in 9.5 M LiBr LiBr and GMA was dropped into the solution with stirring for 3 h at 60 °C. It was dialyzed to remove salts in distilled water at room temperature for 4 days and then freeze-dried. Lithium phenyl(2,4,6-trimethylbenzoyl) phosphinate (LAP) was added to GMA-modified SF solution

Fig. 2

Characterization of Sil-MA pre-hydrogel and Sil-MA hydrogel depending on methacrylation degrees. a FT-IR spectra and b ¹H-NMR spectra of unsubstituted SF and Sil-MA. In the FT-IR spectra, amide I (1639 cm⁻¹), amide II (1512 cm⁻¹), and amide III (1234 cm⁻¹) shown in β-sheet of SF and spectra related with GMA such as CHOH, RR′C=CH₂ were found. The modification of lysine residues in SF with the increase of GMA was confirmed by the gradual decrease in the lysine signal at δ = 2.9 ppm and the increase in the methacrylate vinyl group signal at δ = 6.2–6 and 5.8–5.6 ppm, and the methyl group signal at δ = 1.8 ppm. c FE-SEM images of Sil-MA hydrogels, presenting the effect of the degree of methacrylation and Sil-MA contents on the pore sizes of Sil-MA hydrogels. Scale bar represents 100 μm. d Reaction process for the photopolymerization of Sil-MA in the presence of LAP photoinitiator and 3D structure of Sil-MA hydrogel formed via digital light process (DLP) printing. With the presence of photoinitiator, the vinyl double bonds on GMA could react with each other intra-chain or between chains. SF chains themselves could entangle each other during photopolymerization. Green and magenta colors indicate alpha helix and beta sheet motif in the secondary structure of SF, respectively

Fig. 3

Physical properties of Sil-MA hydrogel. a Schematic diagram of DLP bioprinting procedure using Sil-MA. Sil-MA-added LAP was printed in a layer-by-layer style with DLP printer. Herein, the desired pattern was designed through CAD software and was sliced to a layer file before transferring it to the DLP system. DMD digital micromirror device. b–d Compression test for Sil-MA hydrogel with varied Sil-MA contents. b Compressive elastic modulus of Sil-MA hydrogel at different percentage strains. Significant differences are presented. *p < 0.05 and **p < 0.005 (two-sample t-test). c Representative compressive stress–strain curve and d the Sil-MA hydrogel was compressed by a kettle bell (7 kg) for 3 min and recovered hydrogel’s initial shape after it was removed. e–h Tensile stress for Sil-MA hydrogel. e Tensile elastic modulus of Sil-MA hydrogel at 50% strain and f representative tensile stress–strain curve of Sil-MA hydrogel. g Suturing of folded lumen structure of 3D-printed Sil-MA membrane and h trachea end-to-end anastomosis of dog’s larynx and trachea using Sil-MA 3D DLP-printed hydrogel. Scale bar indicates 1 cm. i–l Swelling properties. i Water uptake of Sil-MA powder. Volume expansion rate of DLP product in j water, k medium, and l PBS (pH 7.4). Data are presented as mean ± s.d. The black line (/bar), blue line (/bar), and orange line (/bar) indicate 10%, 20%, and 30% of Sil-MA hydrogel, respectively. Each assay was conducted in triplicate

Fig. 4

Rheological analysis for Sil-MA hydrogel. a, b Strain dependency and c, d frequency dependency of a, c loss modulus (G″) and b, d storage modulus G′ for Sil-MA hydrogels at different Sil-MA contents varied from 10 to 30%. e, f In situ rheology during UV exposure. The effect of e LAP contents and f Sil-MA contents on the G′ during UV exposure for 250 s except for the line with orange squares (30%Sil-MA) exposed for only 4 s. The black line, blue line, and orange line indicate 10%, 20%, and 30% of Sil-MA hydrogel, respectively

Fig. 5

Printability of 30%Sil-MA using DLP printer. a Porous scaffold and Eiffel Tower imitation; (l) CAD images depicting scaffolds and Eiffel Tower and (r) printed images. Printed scaffolds had small pores around ~700 μm and Eiffel Tower had small holes and grid on the surface. b Ear and brain mimicked shape; (l) CAD images depicting the ear and brain and (r) printed images. Printed products were not damaged when they were compressed by fingers tightly and they were back to their original shape when relaxed his fingers. c Trachea, heart, lung, and vessel mimicked shape; (l) CAD images depicting the trachea, heart, lung, and vessel and (r) printed images at various angles. Printed products by DLP using Sil-MA showed complex structure reflecting their CAD images, including veins, arteries, folds, and holes. Scale bar indicates 1 cm

Fig. 6

Cytocompatibility of Sil-MA. a Live and dead assay; cell viability of encapsulated NIH/3T3 for 14 days (live cells in green and dead cells in red). Scale bar indicates 500 μm. b CCK-8 assay; cell proliferation in the hydrogel for 14 days. The 30%Sil-MA hydrogel showed the similar cell proliferation to that of commercial GelMA (10%). * and ** refer to statistically significant proliferation p < 0.05 and p < 0.005 (two-sample t-test), respectively, compared to 30%Sil-MA hydrogels. The black line, blue line, orange line, and gray line indicate 10%, 20%, and 30% of Sil-MA hydrogel and 10%GelMA hydrogel, respectively. c–f Cell distribution inside the Sil-MA hydrogel. Cells were distributed evenly in the hydrogels. The Sil-MA hydrogels with c a design of the letter HL (the logo of Hallym University) and d a shape of human brain were printed out with PKH67-labeled cells only. Also, the Sil-MA hydrogels with e the letter HL and f a shape of winding trachea were printed out with PKH67-labeled cells (green) and PKH26-labeled cells (red); (from left to right) CAD images, printed images, fluorescence images by c confocal or d–f single plane illumination microscopy (SPIM) microscope, and merged images of fluorescence and CAD images. Scale bar indicates 1 mm on CAD images and 5 mm on printed images. Data are presented as mean ± s.d. Each assay was conducted in triplicate

Fig. 7

In vitro histological evaluation of human chondrocytes loaded Sil-MA hydrogel for cartilage tissue engineering. Scale bars represent 1 mm (×4) and 500 μm (×10). Sil-MA hydrogel showed superior histological characteristics (cell organization and extracellular matrix distribution, including proteoglycan and collagen) of cartilage-like tissue with time in vitro. HE H&E, MT Masson’s trichrome, SFO Safranin-O, PKH PKH26 (red: chondrocytes, green Sil-MA hydrogel)