Methacrylic anhydride

Polyvinyl alcohol with two molecular weights (PVA, Mw 89 kDa and Mw 30 kDa, fully hydrolyzed), polyvinylpyrrolidone (PVP, Mw 40 kDa, fully hydrolyzed), sodium hyaluronate (HA, Mw 300 kDa), d-glucose, agarose (low EEO), glutaraldehyde (grade II, 25% in water), methacrylic anhydride (MA, purity 94%), glucose oxidase (GO) assay kit, 2-hydroxy-4′-(2-hydroxyethoxy)-2-methylpropiophenone (Irgacure-2959), Dulbecco’s phosphate-buffered saline (PBS, pH 7.1–7.5) were purchased from Sigma-Aldrich (St. Louis, MO, USA). Sylgard 184 silicone elastomer was purchased from Dow Corning (Midland, MI, USA). Artificial interstitial fluid (pH 7.4) was purchased from Biochemazone (Alberta, Canada). All electronic components were purchased from Core Electronics (NSW, Australia).
All materials were obtained of the highest quality and used without further purification. Porcine cadaver ears were freshly obtained from a local butcher shop (Brisbane, Australia).

GelMA was synthesized according to protocols established previously by us and others [50 (link), 51 (link)]. A 4% (w/v) gelatin solution composed of gelatin derived from cold water fish skin (Sigma-Aldrich) in Dulbecco’s phosphate buffer solution (DPBS; Sigma-Aldrich) was homogenized over 1 h at a rate of 450 rpm at 60 °C. 16% (v/v) volume of methacrylic anhydride (Sigma-Aldrich) was added to the gelatin solution with a syringe pump at a rate of 0.5 ml min⁻¹. The solution was mixed for 3 h and then diluted with a 3x volume of DPBS (60 °C). The solution was passed through a 0.22 µm filter to remove any unreacted gelatin and poured into membrane tubing (12–14 kDa MWC; Spectrum Laboratories, Inc.). The GelMA solution was dialyzed for a minimum of five days in distilled water (60 °C). The GelMA solution was then frozen for 24 h (−80 °C) and lyophilized for at least 5 d. The resulting soft, white polymer was stored in the dark at −20 °C for use on demand.
The IL, ChoA, was generated according to previous protocols [21 (link)–23 (link)]. In brief, to synthesize ChoA, acrylic acid (Sigma-Aldrich) was added to choline bicarbonate (Sigma-Aldrich) at a 1:1 mole ratio. The solution was allowed to react at 50 °C for 5 h under vacuum and purified overnight under vacuum at room temperature.
GelMA and ChoA structures were confirmed with proton nuclear magnetic resonance (¹H NMR; 500 MHz, Varian Inova). For GelMA, the proton spectrum was generated by dissolving 1 mg of GelMA in deuterium oxide (D²O; Sigma-Aldrich). For ChoA, 50 µl of ChoA was mixed into D₂O before reading the proton spectrum.
To synthesize hydrogels, a precursor solution was made composed of a photoinitiator, liquid solvent (either Hank’s balanced salt solution (HBSS; Gibco) for material characterization or cell culture media for in vitro experiments), GelMA, and for Gel-Amin hydrogels, ChoA. Any differences in preparation are noted in the following sections. 0.5% (w/v) of lithium phenyl-2,4,6-trimethylbenzoylphosphinate (LAP; Allevi) was added to the liquid solvent and sonicated at 40 kHz with a Branson 2510 Ultrasonic Cleaner for five min to completely dissolve the LAP. GelMA and ChoA were then added to the precursor solution.
In our previous study, two different Gel-Amin formulations 7.5% (w/v) + 2.5% (w/v) and were found to support DRG outgrowth and SC viability over seven days as compared to a 10% GelMA Neuman et al [23 (link)]. Here, we opted to increase the amount of ChoA to increase the conductivity of the material further with an 8% (w/v) GelMA + 3.5% (v/v) ChoA Gel-Amin hydrogel. The percentage of GelMA was optimized to prevent degradation during the experimental timeline and to match the elastic moduli of the 9.75% (w/v) GelMA hydrogel (supplemental figure 2). The solutions were stored for 1 h at room temperature in the dark to allow all the GelMA to dissolve. Then, the homogeneous precursor solutions were photo-crosslinked with blue light (λ = 405 nm, 10 W). The exposure time varied as a function of the hydrogel height (0.25 s of exposure time per µm) [52 (link)].
The mechanical properties of GelMA and Gel-Amin hydrogels were analyzed with a TA instruments electroforce 3200 universal mechanical platform according to previous methods [23 (link)]. In brief, cylindrical hydrogels (Ø = 8 mm, H = 4 mm, n = 5, Crosslinking time = 16’ 40”) were fabricated and incubated in HBSS at 37 °C for 2 h prior to testing. A dynamic mechanical analysis (DMA) program (WinTest® 7) was applied in compression with the samples submerged in HBSS during testing. The DMA program applied a sinusoidal frequency sweep between 0.5 and 5 Hz with a mean 10% strain. The software automatically applied a Fourier transform, calculating the difference in phase (δ) between the dynamic peak-to-peak force function and the dynamic peak-to-peak displacement amplitude. This was used to calculate the elastic modulus (E’; equation (1)) and the viscous modulus (E”; equation (2)),
$$\begin{array}{ll}E{}^{\prime }& =\left(\frac{S_o}{e_o}\right)\ast \text{cos}\left(\delta \right)\end{array}$$
$$\begin{array}{ll}E{}^{\prime }& =\left(\frac{S_o}{e_o}\right)\ast \text{sin}\left(\delta \right)\end{array}$$ where $S_o$ is equal to the stress amplitude and $e_o$ is equal to the maximum strain amplitude.
The electrical properties of GelMA and Gel-Amin hydrogels were measured using electrochemical impedance spectroscopy (EIS). For this experiment, deionized water was used as a precursor solvent. Cylindrical hydrogels (Ø = 6 mm, H = 7 mm, n = 5, Crosslinking time = 29’ 10”) were fabricated and placed between two magnesium stick electrodes (Lincoln® Electric). EIS was recorded between 1 MHz and 100 mHz with a sinusoidal amplitude of ±10 mV. Data were analyzed with EC-Lab® Software. The resultant impedance was fit to an equivalent circuit model to obtain the bulk resistance (imaginary impedance = 0) according to previous work [23 (link)]. The calculated bulk resistance was used to calculate the conductivity (C) according to the following formulas:
$$\begin{array}{ll}\mathit{\rho }& =\frac{\mathit{R}\mathbf{\ast }\mathit{A}}{\mathit{L}}\end{array}$$
$$\begin{array}{ll}\mathit{C}& =\frac{\mathbf{1}}{\mathit{\rho }}\end{array}$$ where ρ is the resistivity, A is the cross-sectional area, and L is the length of the sample.
The rate and degree of hydrogel swelling following crosslinking can significantly influence cell viability. To assess the difference in swelling between our GelMA and Gel-Amin hydrogel formulations, cylindrical hydrogels (Ø = 8 mm, height = 4 mm, n = 5) were fabricated. Each hydrogel was weighted and then submerged in HBSS. The samples were incubated at 37 °C and reweighed at different time points (2, 4, 6 and 24 h). The swelling ratio calculated at each time point as the swollen hydrogel weight ( $W_{\text{S}}$ ) divided by the original hydrogel weight ( $W_0$ ).

As biopolymers, 50–190 kDa molecular weight chitosan (Chi) with 75–85% deacetylation degree (DD) and 30–100 kDa molecular weight sodium alginate (Alg) derived from Brown algae with 65:35 G:M ratio (viscosity of 4–12 cP at 1% concentration in water) were used (both from Sigma Aldrich, St. Louis, MO, USA). Methacrylation reaction of both biopolymers was performed using methacrylic anhydride (99% purity, Sigma-Aldrich, St. Louis, MO, USA). Sodium hydroxide (NaOH, 98% purity, PanReac, Darmstadt, Germany) was employed to adjust the pH. In order to perform crosslinking in the membrane thermal initiators azobisisobutyronitrile (AIBN, Merck, Darmstadt, Germany) and 2,2’-azobis(2-amidinopropane)dihydrochloride (V50 Wako, Osaka, Japan) were used, as well as 25% glutaraldehyde water solution vapor (GA, PanReac). To prepare the electrospinning mixture solutions, 83–124 kDa molecular weight polyvinyl alcohol (PVA) with 88% hydrolysis degree was utilized (Sigma-Aldrich) in combination with biopolymers in 8:2 proportion using 8% total polymer mass. For biocharcoal synthesis, anhydrous glucose with 96% purity (Sigma-Aldrich) was used.

Materials. Methacrylic anhydride, photoinitiator 1173 (HMPP), and sodium alginate were obtained from Sigma-Aldrich (USA). Quaternized chitosan, hyaluronic acid, AMP (eumenitin), and CCK-8 reagents were acquired from Beyotime Biotechnolog CO., LTD (China). Other reagents were all achieved from Macklin reagent (China).
Preparation and characterization of ODex. 5g of Dextran was dissolved in 75 mL water. Then, sodium periodate (NaIO₄) was applied to oxidize dextran. In brief, 1.32g of NaIO₄ was dissolved in 20 mL water. To adequately react with dextran, the NaIO₄ solution was added into reaction system drop by drop. After 24h incubation in the dark, the reaction was stopped by adding plenty of ethanol. Then, the reaction solution was transferred to a suitable dialysis bag and dialyzed in pure water for 72 h. The ODex was prepared through a freeze-drying process and further characterized.
Fabrication and characterization of HAMA/Alg inverse opal scaffold. The inverse opal particles were achieved by reversely replicating silica photonic crystal template. The silica photonic crystal template and HAMA polymer were synthesized according to previous work [23 (link),29 (link)]. To fabricate the pregel solution, 1g of HAMA and 0.1g Alg were dissolved in 10 mL water. The photoinitiator HMPP concentration adjusted to 1 %. The dried template was immersed in the pregel solution. The system was first photocured by an ultraviolet (UV) light source and then put into calcium chloride solution for 1 h. Then, the hybrid particles were isolated from the system and immersed in hydrofluoric acid solution. After that, HAMA/Alg inverse opal scaffold was ultimately obtained. SEM was utilized to study the micro/nano structure of particles, and resin was used for shooting.
Fabrication and characterization of ODex/QCS hydrogel. ODex and QCS were respectively dissolved in water to obtain the solution. Then, the ODex solution and QCS solution were mixed. The final concentration of ODex and QCS in the mixed solution were 4 % and 2 %, respectively. Under neutral conditions, the two polymers could react to form dynamic hydrogel. Then, the gelation progress and the chemical state of the dynamic hydrogel was analyzed through rheological measurement and X-ray photoelectron spectroscopy (XPS).
Fabrication of composite hydrogel particles. To better refill the nanovoids of the particles, the inverse opal scaffold was first undergone dehydration at 37 °C for 2 h. Then, such scaffold was immersed in ODex and QCS mixed solution with GOX (0.1 mg/mL) doped. Then, the mixed system was centrifuged for 20 min. After gelation, the CPs were separated from the system for further application.
Biocompatibility test. HAMA/Alg hydrogel and ODex/QCS hydrogel (without GOX) were first immersed in culture medium for 24h, respectively. Then, 8 × 10³ 3T3 cells were seeded into the 48-well plate. Various material leachates were introduced into these wells, and cell viability was detected by using CCK-8 kit on Day 1, Day 2, and Day3. Besides, the 3T3 cells in corresponding groups were stained through calcein to evaluate the cell viability. Meanwhile, we also use HUVECs to test the biocompatibility, and cell viability was measured on Day 3.
ODex/QCS hydrogel in vitro degradation test. The ODex/QCS hydrogel (with GOX) was immersed in PBS (pH = 7.4 or 5.5). 0.5 mL of hydrogel was placed in a centrifuge tube, followed by adding of the PBS buffer with different pH values. The centrifuge tubes were kept at 37 °C and shaken at a speed of 100 rpm/min. At each time point, the hydrogel was rinsed with deionized water and weighed. The remaining weight of the hydrogel was recorded by comparing its current weight to its initial weight. To evaluate the degradability of ODex/QCS (with GOX) in the presence and absence of glucose, PBS glucose solution (5 mM) with an initial pH of 7.4 was used for the experiment, and the subsequent procedure were the same as described above.
In vitro coagulation test. Whole blood was sampled and stored in anticoagulation tube. When the blood was removed from the anticoagulant tube, a 0.1 M calcium chloride solution was mixed and stirred for 10 s. HAMA/Alg hydrogel and CPs (without GOX) were placed at the 96-well plate (5 mg). Subsequently, 50 μL of the mixed blood was introduced into every well. At designated time points, each well was rinsed with saline to remove the unclotted substances.
In vitro drug release. FITC-BSA (EX: 490 nm; EM: 525 nm) and Rhodamine B-labeled AMP (EX: 555 nm; EM: 605 nm) were applied to explore drug release behavior of CPs. The fluorescent images of CPs were captured through confocal microscopy. The drug-loaded CPs (with GOX) were placed in PBS solution (pH = 5.5 or 7.4), respectively. To simulate in vivo conditions, the drug release behavior was monitored at 37 °C with shaking (100 rpm/min). At predesigned time points, 200 μL of the solution was sampled and replaced with fresh PBS. The fluorescence intensity of the released drug was detected using a microplate reader.
Tube formation test. 3 × 10⁴ HUVECs were planted into a 48-well plate pre-coated with growth factor-reduced Matrigel. Then, these wells were divided into four groups, including control group, blank CPs (without GOX), VEGF-loaded CPs (neutral, without GOX), and VEGF-loaded CPs (acidic, without GOX). Notably, to avoid the influence of pH, the cells in the experimental group were cultured with fresh medium mixed with the leaching solution. In the VEGF-loaded CPs (acidic, without GOX) group, the pH value of leaching solution is 5.5. After culturing for 6h, the treated cells were stained and captured with an inverted fluorescence microscope.
Antibacterial test. To evaluate the antibacterial efficacy of the particle system, E. coli and S. aureus were taken as representative bacteria strains. First, bacteria in the logarithmic growth phase were centrifuged and resuspended in PBS containing glucose. The resuspended solution was then treated with four different agents at 37 °C including the PBS group, HAMA/Alg hydrogel group, the pure CPs group (without GOX), the AMP-loaded CPs group (without GOX), and the AMP-loaded CPs group (with GOX), respectively. After culturing, a live/dead kit was performed to assess viability of these bacteria. Then, fluorescence pictures were captured using a fluorescence microscope.
In vivo animal test. Animal tests were approved by the Animal Ethical Committee of the Wenzhou Institute, University of Chinese Academy of Science (WIUCA23041304). The 5- to 6-week-old Sprague-Dawley rats were acquired. To establish diabetes, the rats were administered with streptozotocin (1 % (w/v), 65 mg/kg). When each rat's blood glucose level reached the desired standard and stabilized, a full-thickness circular wound (about 1.5 cm) was created on rat back, followed by the addition of bacterial suspension. Subsequently, these rats were erratically assigned into five groups and treated with different intervention therapies, including PBS-rinsing group (Control), pure GOX-doped CPs group (CP), AMP&GOX-loaded CPs group (CPA), VEGF&GOX-loaded CPs group (CPV), and AMP&VEGF&GOX-loaded CPs group (CPA&V). The wounds were observed every day and photographed on Day 0, 3, 6, 9, and 12. On the last day, all rats were euthanized, and their wound tissues were collected and subjected to further analysis.

Hydroxypropyl cellulose with a hydroxypropyl content ranging from 53.4% to 80.5% and a molecular weight of 100,000 g/mol as per supplier technical datasheet was purchased from Thermo Fisher GmbH (Kandel, Germany) [48 ]. We determined a degree of substitution (DS) of 2.4 and a molar substitution (MS) of 4.76 for this HPC (Figure S2). HPC was oven-dried at 100 °C for 3 h and stored in a desiccator with Silica gel orange (particle size: 2–5 mm) and an indicator from Carl Roth GmbH, (Karlsruhe, Germany) for an additional 24 h to ensure thorough drying. Moisture content was not specifically determined. Methacrylic anhydride (MA), with a purity of ≥94% and containing 2000 ppm of topanol A inhibitor, the catalyst 4-(Dimethylamino)-pyridine (DMAP) with a purity of ≥99.0% for the grafting reaction were obtained from Sigma-Aldrich (Taufkirchen, Germany). For the UV-initiated photocrosslinking, the radical initiator, phenyl bis(2,4,6-trimethylbenzoyl)-phosphine oxide with a purity of ≥96.0%, was procured from TCI Deutschland GmbH in Eschborn, Germany. p-Benzoquinone was selected as an inhibitor due to its oxygen-insensibility and purchased from Sigma-Aldrich (Taufkirchen, Germany). Acetone (≥99.5%), ethanol (≥99.8%), and glacial acetic acid (≥99.7%) were acquired from VWR International GmbH (Darmstadt, Germany). Sodium hydrogen Ecarbonate (NaHCO₃, ≥98.5%) from Thermo Fisher GmbH (Kandel, Germany) was utilized for the neutralization after grafting. Methacrylic acid (MAA) was purchased from Thermo Scientific Chemicals (Geel, Belgium) and used as a reference for Fourier Transform Infrared analysis. Sodium Methacrylate was acquired from TCI Deutschland GmbH in Eschborn, Germany for FTIR analysis. For NMR analyses, Chloroform-d1 and methanol-d4 were obtained from Euriso-Top (Saint-Aubin, France) and Chromium(III) acetylacetonate (Cr(acac)₃) was purchased from Sigma-Aldrich, Taufkirchen, Germany.