3D Printing of Bioactive Gel-like Double Emulsion into a Biocompatible Hierarchical Macroporous Self-Lubricating Scaffold for 3D Cell Culture

The interconnected hierarchically porous structures are of key importance for potential applications as substrates for drug delivery, cell culture, and bioscaffolds, ensuring cell adhesion and sufficient diffusion of metabolites and nutrients. Here, encapsulation of a vitamin C-loaded gel-like double emulsion using a hydrophobic emulsifier and soy particles was performed to develop a bioactive bioink for 3D printing of highly porous scaffolds with enhanced cell biocompatibility. The produced double emulsions suggested a mechanical strength with the range of elastic moduli of soft tissues possessing a thixotropic feature and recoverable matrix. The outstanding flow behavior and viscoelasticity broaden the potential of gel-like double emulsion to engineer 3D scaffolds, in which 3D constructs showed a high level of porosity and excellent shape fidelity with antiwearing and self-lubricating properties. Investigation of cell viability and proliferation using fibroblasts (NIH-3T3) within vitamin C-loaded gel-like bioinks revealed that printed 3D scaffolds offered brilliant biocompatibility and cell adhesion. Compared to scaffolds without encapsulated vitamin C, 3D scaffolds containing vitamin C showed higher cell viability after 1 week of cell proliferation. This work represented a systematic investigation of hierarchical self-assembly in double emulsions and offered insights into mechanisms that control microstructure within supramolecular structures, which could be instructive for the design of advanced functional tissues.


S3. Effect of pH on colloidal stability of double emulsions
Further, in the current project, when the pH was decreased to <6, the W 1 /O/W 1 emulsion (containing soy protein particles) showed a characteristic phase separation (Figure S3).As Figure S3 shows the double emulsion presented a typical phase separation when the dispersion pH was in the range of 3.5-6.This phenomenon might be resulted from the precipitation of soy protein near to its isoelectric point (pI of approximately 5), which would induce protein aggregation.Then, this colloidal-based dispersion including the precipitated soy particles with higher particle sizes could not effectively contribute to the Pickering stylization process.But these aggregates were redispersible when the dispersion pH was out of this range (<3 or > 6) (Figure S3), which the double emulsion was physically stable in these pH ranges.However due to the inhabitation of cell growth (presented in 3.3.5 in the revised manuscript), we didn't use the pH < 3.

Methods
The encapsulation efficiency of vitamin C was determined in fresh double emulsions and after storage (3, 24 and 48 h) at 40 °C.The double emulsions were diluted (1:1) in 0.001 M phosphate buffer (pH = 6.8) containing 0.1 M NaCl and centrifuged at 500 × g for 15 min at 23 °C.The subnatant was collected and centrifuged again at 7500 × g for 30 min at 23 °C.Proteins in the subnatant obtained after the second centrifugation were precipitated by addition of 20% trichloroacetic acid solution (1:1) and removed by centrifugation at 3000 × g for 15 min at 23 °C.The vitamin C concentration in the supernatant was determined by measuring the absorbance at 361 nm according to the method of O' Regan and Mulvihill (2009) and by taking into account the dilution with trichloroacetic solution.Encapsulation efficiency [E (%)] of vitamin C in double emulsions was calculated using Eq. ( S1): (S1) where C W1 is the initial vitamin C concentration in the internal aqueous phase of the emulsion (0.2%) and C S , the vitamin C concentration in the subphase collected after centrifugation of diluted emulsion; X W1 and X W2 are, respectively, the mass fractions of the internal (0.07) and external (0.65) aqueous phase of the emulsion and D is the dilution volume (D = buffer volume/emulsion volume).

Results and discussion
A coalescence index can be calculated from the rate of change in droplet diameter over the 48-h storage period (Figure S4).Compared with non-sonicated sample (PU-0), HIU sonication notably decreased the coalescence index of the emulsions of PU-2, PU-4, and PU-6.The stability of primary emulsions was monitored during storage using a vertical optical scan analyzer; no signs of phase separation were observed (data not shown).Considering the average droplet size (Figure 1 in the main body of paper), the viscosity of oil and the density difference between the two phases, Stokes law predicts a sedimentation rate of 0.5-2 μm h −1 , which will not give a detectable phase separation after 48 h storage.
The vitamin C concentration in the external aqueous phase was measured to calculate the encapsulation efficiency.An encapsulation efficiency higher than 90% was obtained for the sonicated emulsions, indicating that the vitamin C remained entrapped within the internal aqueous phase.Storage time and external aqueous phase had no significant effect (p > 0.05) on encapsulation efficiency (data not shown).However, efficiency was significantly affected (p ≤ 0.05) by the emulsification conditions (Table S1).The difference is likely associated with aggregation/coalescence of water phase droplets inside the primarily developed W 1 /O emulsion.Encapsulation efficiency was higher (98.4%) in the case of emulsification with 4-and 6-min sonication.The sonication produced small droplets (Figure 1 in the main body of paper) with a smaller interfacial area, which minimized close contact between the internal and external aqueous phases.These emulsions showed lower phase separation than double emulsions produced by 2-and 8-min sonication (Figure S4).

S5. Printing setup
Prusa i3 is named after the third repetition of the design by Josef Průša.All parts of this 3D printing system were open-source and were part of the 3D Soft-Gel Printer project.Table S2 resumes the main characteristics depicts the original Prusa i3 Printer assembly kit and an assembled printer.

Heated platform Yes
Minimum layer thickness (mm) 0.1

Open Source
Hardware and software The component of the syringe unit was designed through SolidWorks™ (Dassault Systèmes, SolidWorks Co., Vélizy-Villacoublay, France) and was developed by Peyman Asghartabar Kashi and was 3D printed from polylactic acid (PLA, eSUN, China) using a homemade extrusion-based 3D printer.We decided to fabricate the components of the syringe unit with another 3D printer than the device we aimed to modify, as this facilitated design optimization and circumvented the need for multiple assembly-disassembly cycles, accelerating the prototyping procedure.
The machine architecture is very simple, where the extrusion head transfers in the XZ plane construction while the platform translates along the Y-axis.The horizontal translation of the head is controlled by the Xaxis and the vertical translation allows the increment along the Z-axis.The mechanical structure is minimal

Figure S1 .
Figure S1.Continuous phase test of the emulsion by the method of dilution test.

Figure S3 .
Figure S3.Visual observation of double emulsions in different pH values (48 h after preparation).(To better show a phase separation in double emulsions in different pH values, we used a Black & White mode)

Figure S4 .
Figure S4.Coalescence index (CI) of W 1 /O primary emulsion droplets for different samples.CI was calculated from the increase of droplet diameter (d) over the 48-h storage period (t).CI = Δd/Δt.Error bars represent the standard errors obtained from the statistical model.

Table S1 .
Effect of emulsification conditions on encapsulation efficiency of vitamin C in W 1 /O/W 2 double emulsions.Values given have a standard error of 0.3; values with different superscript letters within the column are significantly different at p ≤ 0.05.