3D Freeform Printing of Nanocomposite Hydrogels through in situ Precipitation in Reactive Viscous Fluid

Abstract

Composite hydrogels have gained great attention as three-dimensional (3D) printing biomaterials because of their enhanced intrinsic mechanical strength and bioactivity compared to pure hydrogels. In most conventional printing methods for composite hydrogels, particles are preloaded in ink before printing, which often reduces the printability of composite ink with little mechanical improvement due to poor particle-hydrogel interaction of physical mixing. In contrast, the in situ incorporation of nanoparticles into a hydrogel during 3D printing achieves uniform distribution of particles with remarkable mechanical reinforcement, while precursors dissolved in inks do not influence the printing process. Herein, we introduced a "printing in liquid" technique coupled with a hybridization process, which allows 3D freeform printing of nanoparticle-reinforced composite hydrogels. A viscoplastic matrix for this printing system provides not only support for printed hydrogel filaments but also chemical reactants to induce various reactions in printed objects for in situ modification. Nanocomposite hydrogel scaffolds were successfully fabricated through this 3D freeform printing of hyaluronic acid (HAc)-alginate (Alg) hydrogel inks through a two-step crosslinking strategy. The first ionic crosslinking of Alg provided structural stability during printing, while the secondary crosslinking of photo-curable HAc improved the mechanical and physiological stability of the nanocomposite hydrogels. For in situ precipitation during 3D printing, phosphate ions were dissolved in the hydrogel ink and calcium ions were added to the viscoplastic matrix. The composite hydrogels demonstrated a significant improvement in mechanical strength, biostability, as well as biological performance compared to pure HAc. Moreover, the multi-material printing of composites with different calcium phosphate contents was achieved by adjusting the ionic concentration of inks. Our method greatly accelerates the 3D printing of various functional or hybridized materials with complex geometries through the design and modification of printing materials coupled with in situ post-printing functionalization and hybridization in reactive viscoplastic matrices.

Keywords: Hydrogels; Multi-materials; Nanocomposites; Three-dimensional freeform printing; Viscous fluid matrix; in situ precipitation.

Figure 1

Schematic of three-dimensional freeform printing system of nanocomposite hydrogels through a two-step crosslinking process coupled with in situ precipitation.

Figure 2

Schematics and optical images of three-dimensional-printed scaffolds. (A) Hyaluronic acid-alginate (HAc-Alg) hydrogel scaffolds (perspective view, top view, and side view). (B) HAc-Alg/30 wt% calcium phosphate composite hydrogel scaffolds (perspective view, top view, and side view).

Figure 3

Surface and cross-sectional morphologies of dehydrated 3D-printed scaffolds. (A-C) Surface morphology of hyaluronic acid-alginate (HAc-Alg) hydrogel scaffolds. (D-F) Surface morphology of HAc-Alg/30 wt% calcium phosphate (CaP) composite hydrogel scaffolds. (G) Cross-sectional morphology of a printed HAc-Alg filament. (H and I) Cross-sectional morphology of a printed HAc-Alg/30 wt% CaP filament.

Figure 4

Mineral phases of hyaluronic acid-alginate (HAc-Alg)/calcium phosphate (CaP) hydrogels. (A) X-ray diffractometer patterns of (a) HAc-Alg hydrogels, (b) HAc-Alg/30 wt% CaP composite hydrogels prepared by mixing, and (c) HAc-Alg/30 wt% CaP composite hydrogels prepared by in situ precipitation. (B) Transmission electron microscope images of CaP nanoparticles in HAc-Alg/CaP composite hydrogels (prepared by in situ precipitation) and corresponding selected-area electron diffraction patterns of CaP nanoparticles.

Figure 5

In vitro enzymatic degradation behavior of three-dimensional-printed scaffolds. (A) Remaining weight over incubation time after the hydrogel scaffolds were immersed in hyaluronidase solution, and optical images of the scaffolds before and after enzymatic degradation. (B) Fourier-transform infrared spectra of (a) glycidyl methacrylate-hyaluronic acid (HAc) hydrogel, (b) alginate hydrogel, (c) HAc-alginate (HAc-Alg) after degradation, and (d) HAc-Alg/30 wt% calcium phosphate (CaP) after degradation. (C) X-ray diffractometer pattern of HAc-Alg/30 wt% CaP scaffolds after degradation.

Figure 6

Cytocompatibility of hyaluronic acid-alginate (HAc-Alg) and HAc-Alg/30 wt% calcium phosphate (CaP) hydrogels using L929 fibroblasts. (A) Scanning electron microscope images of cells attached to the surfaces of HAc-Alg and HAc-Alg/30 wt% CaP hydrogels. (B) Cell viability of HAc-Alg and HAc-Alg/30 wt% CaP hydrogels measured by AlamarBlue assay after 3 and 5 days (n > 3, **P < 0.01). (C) Confocal laser scanning microscope (CLSM) z-stack images of L929 fibroblasts adhered to HAc-Alg/30wt% CaP three-dimensional (3D) scaffolds after 7 days, indicating the selected parts for imaging. The two layers from the bottom and center regions were imaged to confirm the cell distribution throughout the scaffold. (D) CLSM z-stacked confocal images of L929 fibroblasts within the 3D printed scaffold after a 14-day incubation period.

Figure 7

Quantitative gene expression of four markers for in vitro cell differentiation. (A) Collagen type 1, (B) RunX2, (C) osteocalcin, and (D) osteopontin of MC3T3-E1 cells after 14 days of culturing on well plate (Control) or three-dimensional printed hyaluronic acid-alginate/30 wt% calcium phosphate composite hydrogel scaffolds with cell maintenance medium or osteogenic medium (n > 3, *P < 0.05 and **P < 0.01).

Figure 8

Multiphase material printing with composite hydrogel inks. (A) Three-dimensional (3D) printing strategy of multiphase materials for gradient biomaterials. (B) Schematics of proposed bi-material scaffolds varying mineral contents. (C) Optical images of 3D printed multiphase composite scaffolds composed of hyaluronic acid-alginate (HAc-Alg)/10 wt% calcium phosphate (CaP) and HAc-Alg/30wt% CaP.

Supplementary Figure 1

HAc-Alg/CaP scaffold directly printed on a glass slide. Left: one layered structure of the scaffold, Right: three-layered structure of the scaffold. During 3d printing, the HAc-Alg ink with (NH4)2HPO4 was deposited on the surface of a glass slide, followed by UV crosslinking. The printed scaffolds were treated in CaCl2 solution for physical crosslinking of alginate and mineralization. The one-layer scaffold could maintain its meshed structure. However, for a scaffold with multi-layered structure, the top layers were collapsed during printing, being merged with the bottom layers.

Supplementary Figure 2

High resolution SEM images of (A,B) HAc-Alg hydrogel scaffolds ; (C,D) HAc-Alg/ 10wt% CaP composite hydrogel scaffolds; (E,F) HAc-Alg/ 30wt% CaP composite hydrogel scaffolds. The average particle size of hydrogels was determined by analysing the SEM images. The average particle size of calcium carbonate for HAc-Alg was too small to be measured using the SEM image at x 50,000 magnification. For HAc-Alg/10wt% CaP, the average particle size is ~60 nm. In case of HAc-Alg/30wt% CaP, the average particle size was determined using both SEM and TEM images, resulting particle size is ~60 nm in diameter. Based on the particle size analysis, the increased CaP content is likely attributed to the increased number of CaP particles instead of the crystal growth of nanoparticles.

Supplementary Figure 3

SEM images and EDS spectra of (A) HAc-Alg and (B) HAc-Alg/30wt% CaP. HAc-Alg contains a significant amount of calcium due to crosslinking of alginate as well as the existence of calcium carbonate. Because of calcium in alginate, the ratio of Ca/P in HAc-Alg/30wt% CaP is found to be ~1.6. Thus, the Ca/P ratios of CaP nanoprecipitates should be determined using TEM and corresponding EDS analysis instead of SEM.

Supplementary Figure 4

Cross-section image of HAc-Alg/30wt% CaP with EDS mapping analysis. Calcium and phosphorous atoms are found throughout the thickness of printed hydrogel filaments, indicating that nanoparticles were successfully formed and uniformly distributed in the whole hydrogel matrix.

Supplementary Figure 5

(A) TEM image of CaP nanoparticles in the HAc-Alg/30wt% CaP hydrogels, (B) EDS mapping analysis of the nanoparticles from the selected region indicated in the TEM image and (C) EDS line profile analysis of the selected nanoparticle from point A to point B. From the EDS analysis, the Ca/P ratio of CaP nanoparticles was determined as ~0.6.

Supplementary Figure 6

TGA results of HAc-Alg hydrogel (black solid line), HAc-Alg/10wt% CaP (blue solid line) and HAc-Alg/ 30wt% CaP (red solid line). The remaining weight at 1000 °C for three types of specimens was used for the calculation of mineral contents for two composite hydrogels. The remaining weight for HAc-Alg should be attributed to calcium oxide and organic ashes after thermal degradation, which was used for the baseline of other two hydrogels.

Supplementary Figure 7

(A) Storage modulus over frequency of HAc and HAc-Alg hydrogels, (B) Storage modulus over frequency with different CaP amount, (C) compressive stress-strain curves of 3D printed HAc-Alg and HAc-Alg/30wt% CaP scaffolds. (inset: optical images of hydrogel scaffolds before and after deformation at ε = 0.8), and (D) swelling ratios of three hydrogel specimens: HAc-Alg, HAc-Alg/30wt% CaP by physical mixing of CaP nanoparticles and HAc-Alg and HAc-Alg/30wt% CaP by in-situ precipitation (n > 3). There are significant differences between any pair-wise comparisons, implying that HAc-Alg/30wt% CaP by in-situ precipitation shows remarkable improvement on mechanical and swelling behavior.

Supplementary Figure 8

Weight loss over incubation time after hydrogel scaffolds were immersed in hyaluronidase solution. The weight loss of hydrogels was calculated from Figure 6A of the main manuscript, with the assumption that only enzymatic degradation of hyaluronic acid causes weight reduction.

Supplementary Figure 9

SEM images of HAc-Alg/30 wt% CaP composite hydrogel scaffolds after 2 weeks of degradation. The degraded specimens were fully dehydrated using critical point dryer and coated with gold. The surface of HAc-Alg/30 wt% CaP clearly exhibits needle-like nanocrystals, which is well-known to be the morphology of apatite.

Supplementary Figure 10

(A-B) SEM images of L929 cells on the surfaces of HAc-Alg and HAc-Alg/30wt% CaP after 3 days of culturing and (C) CLSM images of L929 fibroblasts cultured on HAc-Alg and HAc-Alg/30wt% CaP after 3 days of culturing. Cells didn’t fully cover the surface of HAc-Alg/30wt% CaP after 3 days of culturing, showing significant local variation of cell density. On the other hand, cells on HAc-Alg didn’t attach to the surface well, forming cell clusters. Cell morphologies of two hydrogel surfaces are also different: globular cells on HAc-Alg vs. elongated and flattened cells on HAc-Alg/30wt% CaP.

Supplementary Figure 11

Schematics of two bi-material porous scaffolds with corresponding printing sequences: (A) Bi-layered porous scaffold and (B) core-shell porous scaffold.

Supplementary Figure 12

H1NMR spectra of GM-HAc, assigning methacrylate protons and methyl protons to chemical shifts. GMHA were dissolved in deuterium oxide (Sigma-Aldrich, Singapore) at a concentration of 5 mg/mL prior to the test. The H1NMR spectra was recorded by Bruker Avance II 300MHz (Bruker, Germany). The methacrylation of HAc can be determined by the existence of three proton chemical shifts (δ = ~1.85, ~5.65, and ~6.1). The approximate percent of methacrylation (a ratio of methacrylate protons (a) to methyl protons (d)) was calculated from the relative integrations of the methacrylate protons and methyl protons of HAc (two peaks at ~ 1.9 ppm) 2-3. The linear fitting function of MestReNova software was used for deconvolution and integration of those two peaks. Based on the calculation, the degree of methacrylation for GM-HAc in this study was found to be ~15 %.