Reconfigurable Magnetic Liquid Building Blocks for Constructing Artificial Spinal Column Tissues

Abstract

All-liquid molding can be used to transform a liquid into free-form solid constructs, while maintaining internal fluidity. Traditional biological scaffolds, such as cured pre-gels, are normally processed in solid state, sacrificing flowability and permeability. However, it is essential to maintain the fluidity of the scaffold to truly mimic the complexity and heterogeneity of natural human tissues. Here, this work molds an aqueous biomaterial ink into liquid building blocks with rigid shapes while preserving internal fluidity. The molded ink blocks for bone-like vertebrae and cartilaginous-intervertebral-disc shapes, are magnetically manipulated to assemble into hierarchical structures as a scaffold for subsequent spinal column tissue growth. It is also possible to join separate ink blocks by interfacial coalescence, different from bridging solid blocks by interfacial fixation. Generally, aqueous biomaterial inks are molded into shapes with high fidelity by the interfacial jamming of alginate surfactants. The molded liquid blocks can be reconfigured using induced magnetic dipoles, that dictated the magnetic assembly behavior of liquid blocks. The implanted spinal column tissue exhibits a biocompatibility based on in vitro seeding and in vivo cultivating results, showing potential physiological function such as bending of the spinal column.

Keywords: alginate surfactants; all-liquid molding; in vivo cultivated tissues; spinal column structures; structured magnetic droplets.

Figure 1

Fabricating spinal column tissue by means of all‐liquid molding techniques. A) All‐liquid molding process is illustrated. Morphologies of droplets can be retained with high fidelity after being released from the spinal column‐like mold cavity by levitating. The carboxylated alginate and amino‐functionalized POSS assemble at oil‐water interfaces until jamming, and lock the spinal column‐like shape in non‐equilibrium state. B) Molding, assembling and welding magnetic aqueous biomaterial ink into spinal column‐like droplets, seeding BMSCs/chondrocytes, blocks crosslinking, and in vivo cultivating with rats (BMSCs indicates bone marrow mensychmal stem cells, DBM indicates demineralized bone matrix, CAM indicates cartilage acellular matrix). C) Mammalian spinal columns are composed of repeating units, including vertebrae and intervertebral disc cartilage. D) Bright and fluorescent images of spinal column‐like tissue. Green color represents DBM, and red for CAM. E) In vivo cultivated spinal column tissues kept their initial shapes well after being implanted subcutaneously in the back of rats for 1 month (Scale bars in D, E: 1 mm).

Figure 2

Reproducible mass production of liquid droplets with arbitrary shapes and characterization of the interfacial assembly and jamming of alginate‐surfactants. A) Triangle droplet levitated from the mold cavity immersed in the heavier oil with (Jammed) or without (Un‐jammed) POSS‐NH₂ dissolved (Water solution: 10 g L⁻¹ alginate, 100 g L⁻¹ GelMA and 2.5 g L⁻¹ MNPs dissolved in DI water, pH = 5; Oil: volume ratio of toluene/silicon oil/CCl₄ is 4:3:1.). B) Liquid droplets were molded into various shapes, including cylinder, moon, doughnut and cross (Water solution: 10 g L⁻¹ alginate, 100 g L⁻¹GelMA and 2.5 g L⁻¹ MNPs dissolved in DI water, pH = 5; 0.3 g L⁻¹ POSS‐NH₂ dissolved in oil solution where volume ratio of toluene/silicon oil/CCl₄ is 4:3:1.). C) Surface coverage (SC) of liquid droplets were tuned by varying components, concentrations and pH, SC = S_J/S_I (Oil: 0.3 g L⁻¹ POSS‐NH₂ dissolved in 100 µL toluene and 5 mL silicon oil.). D) Interfacial tension of liquid droplets were tuned by varying components, concentrations and pH (Oil: 0.3 g L⁻¹ POSS‐NH₂ dissolved in 100 µL toluene and 5 mL silicon oil.). E) The morphologies of molded liquid triangles with different SCs (Scale bars: 1 mm).

Figure 3

Characterization of magnetic liquid droplets. A) Hysteresis loops of individual 10 µL of liquid droplet (2.5 g L⁻¹ MNPs in 10 g L⁻¹ alginate solution as alginate solution, 2.5 g L⁻¹ MNPs in 10 g L⁻¹ crosslinked alginate solution as alginate hydrogel, 2.5 g L⁻¹ MNPs in DI water as DI water) at pH 5, immersed in the oil containing 0.3 g L⁻¹ of POSS‐NH₂ ligands. B) Hysteresis loops of individual 10 µL liquid droplet (mixed with 10 g L⁻¹ alginate and 2.5 g L⁻¹ MNPs, pH = 5) immersed in toluene without (un‐jammed) and with POSS‐NH₂ ligands (0.3 g L⁻¹, jammed). C) Reconstructed 3D micro‐CT image of liquid cylinders with reconfigurable directional magnetic dipoles. D) A liquid cylinder was magnetized repeatedly by external uniform magnetic field. Magnetic domains and dipoles were reconfigured accordingly by applying magnetic field with specific strength and directions (Scale bars in C, D: 1 mm).

Figure 4

Magnetically‐controlled assembly and joint of structured liquid blocks. A,B) Formation of “H”‐shaped liquid blocks through all‐liquid molding, magnetizing and patterning. C) Bone‐like and cartilaginous liquid blocks were suspended in the oil, and manipulated by external magnetic field to assemble into a spinal column‐shaped structure. Then the gaps of droplets were welded and integrated into a scaffold for the following cell seeding and tissue growth (Scale bar in B: 2 mm, scale bar in C: 1 mm).

Figure 5

Growth and characterization of spinal column tissue. A) Bone‐like blocks containing DBM (white parts) and cartilage blocks containing CAM (blue). B) The assembled, welded and crosslinked spinal column‐like hydrogel scaffold was stable with a high mechanical strength. C) Cross‐section SEM images of spinal column‐like hydrogel scaffold indicated that the welded hydrogel blocks are closely packed with obvious boundaries between them. D) Spinal column‐like hydrogel scaffold composed of three layers of cell‐ECM synergies.(Green fluorescence: BMSCs, red fluorescence: chondrocytes). E) SEM images of seeded BMSCs/chondrocytes on spinal column‐like scaffold after culturing for 5 days. F) Quantitative analysis of the number of viable cells in different groups (White arrows indicate the seeded BMSCs/chondrocytes.). G) Calcein/PI staining of seeded BMSCs/chondrocytes on spinal column‐like scaffold, wherein green fluorescence indicated cytoplasm of live cells and red fluorescence indicated the nucleus of dead cells. H) H&E staining of explants indicated that the spinal column tissue kept its shapes. And the cells grew into the spinal column tissues in vivo with slight deformation after 4‐week implantation subcutaneously. I) Immunohistochemical staining of aggrecan and type 2 collagen indicated that the seeded chondrocytes were distributed in the center of the explant. J) Immunohistochemical staining of osteocalcin (OCN) and type 1 collagen indicated that the seeded BMSCs were distributed in the periphery area of the explant. (Scale bars in A–D and H: 1 mm; scale bars in E, G: 20 µm; scale bars in I,J: 100 µm).