Gelatin (type B, 100 and 250 blooms, Aladdin Shanghai, China), sodium carbonate, sodium bicarbonate decahydrate, sodium hydroxide, acetohydroxamic acid, hydroxylamine, sodium dodecyl sulfate, and alanine were purchased from Aladdin (Shanghai, China). Methacrylic anhydride (MAA), 2-hydroxy-4'-(2-hydroxyethoxy)-2-methylpropiophenone (I2959), and collagenase (‖, 125 CDU/mg) were purchased from Sigma-Aldrich (Shanghai, China). Deuterium oxide (D₂O) and 2,2,3,3-D₄ (D, 98%) sodium-3-trimethylsilylpropionate (TMSP) were obtained from Cambridge Isotope Laboratories (Andover, United States). Human dental pulp stem cells (hDPSCs) were received from Wenzhou Medical University (Luo et al., 2021) Dulbecco’s Modified Eagle’s Medium, fetal bovine serum, Hanks’ Balanced Salt Solution (HBSS) (10X), penicillin/streptomycin, and Live/Dead^® Cell Viability/Cytotoxicity kit were purchased from Life Technologies (Shanghai, China). All the reagents were used as received.

GelMA samples were synthesized according to the literature (Shirahama et al., 2016). Briefly, type B gelatin (100 and 250 blooms) was dissolved at 10 w/v% at 50°C in .25 M carbonate-bicarbonate [CB] buffer). GelMA samples were prepared by reacting gelatin with methacrylic anhydride (MAA, 94%) at a feed ratio of MAA (60 mL) to gelatin (100 g) at 50°C for 2 h in CB with pH maintenance at 9.0 in a time-lapse loading manner. After the 2 h reaction, the solutions were readjusted to pH of 7.4, filtered, dialyzed using PALL Minimate TFF Capsule with 10 kDa MWCO at 50°C for 1 day, lyophilized, and stored at −20°C until further use. The degree of substitution (DS) of GelMA was verified by ¹H-NMR (Avance I 400MHz, Bruker) in deuterium oxide. For ¹H-NMR testing about gelatin and GelMA, 20 mg of each sample was dissolved in 800 µL of deuterium oxide with .1 w/v% TMSP as an internal reference at 40°C. The DS calculation was performed according to the previous literature (Zhu et al., 2019).

For GelMA ICC hydrogels, polystyrene beads with a diameter of 138 ± 2.0 µm (Duke Scientific Corporation) dispersed in a 70% ethanol solution were self-assembled in 6 mm-diameter polypropylene molds by shaking for 48 h and annealed at 134°C for 6 h to obtain lattices. GelMA samples were dissolved at 30 w/v% in distilled water at 50°C. Then, GelMA solutions containing .5 w/v% 2-Hydroxy-4'-(2-hydroxyethoxy)-2-methylpropiophenone (I2959) were infiltrated into the lattices by centrifugation at 15,000 rpm at 40°C for 10 min. The GelMA-infiltrated lattices were cured by ultraviolet light (UV; 365 nm at 50 mW/cm²) for 10 min, and then the polystyrene lattices were removed in tetrahydrofuran. The PS beads were soaked into THF and removed by multiple changes of a THF solution. During the THF treatment, PS beads dissolved in THF quickly; however, ICC GelMA hydrogels did not dissolve in THF and instead exchanged some part of water with THF. ICC GelMA gels were shrunk partially during the THF process but they recovered their original shape in aqueous solutions. The resultant GelMA ICC scaffolds were sterilized with a 70% ethanol solution and washed three times with PBS, finally being stored in distilled water or PBS at 4°C until further use. GelMA bulk hydrogels were prepared in the polypropylene molds with a diameter of 6 mm under the low power of photocuring conditions (UV; 365 nm at 4 mW/cm²) without the templating of polystyrene beads.

GelMA ICC and bulk hydrogels were freeze-dried for 3 days using a freeze dryer. In order to evaluate their swelling kinetics, each freeze-dried sample was immersed in a DPBS solution to measure the weight of each sample over time. The swelling ratio was calculated by the following equation: S w e l l i n g   r a t i o = W i W d   (1) Where Wi is the weight of the swollen sample and Wd is the weight of the freeze-dried sample.

The porosity of GelMA ICC hydrogels and bulk hydrogels was characterized by acquiring the weight of swollen scaffold W _i and the weight of freeze-dried macro- and microporous hydrogel scaffolds W _d . For scaffolds with large porosity, following Du’s opinion (Jiang et al., 2019) the porosity can be calculated according to the equation: P o r o s i t y = W i − W d W d (2)

For compression testing, GelMA bulk hydrogels and ICC hydrogels were prepared by the previous method using 30 w/v% GelMA containing .5 w/v% I2959 through the UV photopolymerization. GelMA bulk hydrogels and ICC hydrogels with a diameter of 4–5 mm and a thickness of 4 mm were fabricated and tested using a universal mechanical testing machine (UTM2102; Shenzhen, China). The speed of the crosshead was .25 mm s⁻¹. The compression strain and stress were recorded until the hydrogel was crushed. The compressive modulus was determined as the slope of the stress-strain curve from 0 to .15 mm/mm strain. The modulus of the porous material, E, can be related to the modulus of the non-porous hydrogel (E ₀ ) by the following equation, where the porosity is assigned the variable (Welzel et al., 2012; Welzel et al., 2014; Sedlačík et al., 2020), p: E = E 0   ∙ 1 − p 2 (3)

Mechanical properties of aqueous GelMA solutions (30 w/v%) containing I2959 (.5 w/v%) were characterized by sinusoidal shear rheometry. Frequency-sweep measurements were conducted using a rheometer (TA DHR-2). The storage moduli of GelMA ICC and bulk hydrogels were measured with an 8 mm parallel-plate geometry at .1% strain and .1 Hz at 37°C throughout the measurements. Mesh size of the GelMA bulk hydrogel was calculated according to the following equation (Lee et al., 2017). Where G’ is the storage modulus, N _A is the Avogadro constant, R is the molar gas constant, and T is the temperature. ξ = G ′ ∙ N A R T − 1 3 (4)

ICC and bulk hydrogels made of GelMA (30 w/v%) were tested for enzymatic degradation in 2 mg/mL of collagenase type ‖ (125 CDU/mg solid) in Hank’s Balanced Salt Solution (HBSS), containing 3 mM CaCl₂. Surface morphology of the GelMA ICC hydrogels and GelMA hydrogels was observed by optical microscopy. For each degradation time point, gross images were taken, and mass loss of GelMA ICC and GelMA bulk hydrogels was simultaneously measured. The initial swollen weight (W_i) of each hydrogel sample (n = 5) was measured, and then each hydrogel sample was put into a 2 mg/mL collagenase type ‖ solution and was incubated at 37°C. At each degradation time point, each sample was taken out and washed with HBSS solution 3 times, and the excess surface water was removed using Kim-wipes, and the degraded weight (W_d) of each sample was recorded again. M a s s   l o s s   % = W i − W d W i (5)

Human dental pulp stem cells (hDPSCs) were maintained in alpha-MEM Dulbecco’s Modified Eagle’s Medium (Hyclone) with 10% fetal bovine serum (Hyclone) and 1% penicillin/streptomycin (Life Technologies) in a humidified atmosphere at 37°C with 5% CO₂. The medium was changed every 2–3 days. The hDPSCs and protocols were independently reviewed and approved by the Ethics Committee of the School and Hospital of Stomatology, Wenzhou Medical University (No. WYKQ2018008SC) (Luo et al., 2021). Before cell seeding, GelMA ICC scaffolds were placed in 24-well plates, consequently washed with PBS and kept in 2 mL media for 30 min before cell loading. After media aspiration, 1×10⁵ cells in 10 µL of medium were carefully pipetted on top of each GelMA ICC scaffold. Also, for cell encapsulation of GelMA bulk hydrogels, 1×10⁵ cells were mixed with a 30 w/v% GelMA solution of 40 µL (.5 w/v%) and cured by 365 nm UV for 5 min (UV; 365 nm at 4 mW/cm²). After 4 h, the cell-encapsulated scaffolds were transferred into a new 24-well plate with 1 mL of media. For both cell-laden hydrogel systems, the media were changed every 2–3 days. The loading efficiency of each GelMA system was measured by counting the real number of loaded cells after treating cell-laden hydrogels with .25% trypsin/EDTA solution via measuring the unloaded cells in each well (N = 3).

Cell viability in GelMA ICC hydrogels and GelMA bulk hydrogels was measured by the cell counting Kit-8 (CCK-8, Donjindo Molecular Technologies) assay according to the manufacturers’ protocol. To test the O. D value, a 400 μL CCK-8 solution (1:10 dilution to culture media) was added to each sample at different culture time (1day, 7 days, and 21 days), each sample was incubated for 4 h at 37°C, and the absorbance was measured at 450 nm.

Cell viability in 3D GelMA ICC scaffolds and 3D GelMA bulk scaffolds were characterized using LIVE/DEAD^® Cell Viability/Cytotoxicity kit (Life Technologies). Briefly, 4 µM Calcein-AM and 8 µM ethidium homodimer-1 (EthD-1) in media were added to the samples, followed by incubation of 1 h at 37°C. The cytoplasm of live cells and nucleus of dead cells were stained by Calcein-AM (green) and EthD-1 (red), respectively and observed with a confocal microscope (Nikon, A1, Japan). The number of living and dead cells were counted using ImageJ.

To investigate cell morphology inside of hydrogels, F-actin of cells in each hydrogel sample was stained by Alexa Flour^® 488 labeled phalloidin (Life Technologies). The nuclei of cells were stain by 4,6-Diamidino-2-Phenylindole Dihydrochloride (DAPI; Life Technologies). The samples were washed twice with PBS, and fixed with 4% paraformaldehyde (PFA) for 5min, then permeabilized with .1% Triton X-100 for 30min, washed with PBS twice, and incubated in 3% BSA blocking buffer for 1 h. Alexa Flour^® 488 was then added to label phalloidin to each sample by reacting for 2 h at room temperature, followed by washing twice with PBS. Finally, sample stain by 10ug/ml DAPI for 10 min, before being imaged using a confocal microscope. The roundness and aspect ratio were measured using ImageJ. R o u n d n e s s = 4 π ∙ a r e a c o n v e x   p e r i m e t e r 2 (6) A s p e c t   r a t i o = L e n g t h W i d t h (7)

To investigate the microscopic morphology of cells in hydrogels, scanning electron microscopy (SEM) was utilized. Samples were fixed with 4% PFA and were treated by sequential ethanol dehydration at 25, 50, 75, 95, and 100% for 15 min each. The scaffolds were frozen at −80°C, followed by lyophilization for 3 days. The SEM samples were coated by Pt with a thickness of 10 nm using a sputter coater (Leica, EM ACE600, Germany), and their images were taken with a field emission scanning electron microscope (HITACHI, SU8010, Japan) at an acceleration voltage of 5 kV.

RGD motif is known to mediate cell attachment on the gelatin scaffolds. The RGD motif in the different GelMA hydrogel systems was calculated via Du’s method (Jiang et al., 2019). A GelMA scaffold in the size of π *3²*1 (mm³) is made of 4 mg of gelatin: the total molecular number of RGD motif in the scaffold can be estimated as follows (where N _A is defined as Avogadro constant): N R G D =   W e i g h t G e l M A M W p r o c o l l a g e n × 7 × N A (8)

Considering the porosity of 95%, the volume for gelatin inside can be calculated: V G e l M A = V S c a f f o l d × 1 − p (9)

The average volume containing one single RGD motif would be estimated as: V R G D = V g e l a t i n N R G D (10)

The average spacing between nearby RGD motifs would be estimated as: G a p R G D = V R G D 3 (11)

Statistical analysis was performed using the Microsoft Excel statistical analysis software package. Comparisons between the two treatments were made using a two-tailed pair Student’s t-test. Two-way ANOVA was used to test for differences among at least three groups. The standard deviation was calculated and presented for each treatment group (mean ± SD). Results with a p-value of below .05 were considered statistically significant. (*: p < .05; **: p < .01; ***: p < .001; ****: p < .0001).

GelMA with a high degree of substitution was successfully prepared using our previous synthesis method (Shirahama et al., 2016; Zhu et al., 2019; Luo et al., 2021). Figure 1A shows that gelatin has amino and hydroxyl groups susceptible to react with methacrylic anhydride (MAA) to form methacrylamide groups and methacrylate groups, respectively. GelMA made of gelatin with a higher bloom number of 250 shows higher viscosity below 25°C (Figure 1B) as compared to GelMA made of gelatin with a lower bloom of 100. High viscous GelMA solutions can cause a handling issue during cell encapsulation process and ICC fabrication. GelMA with a lower bloom of 100 exhibited low viscosity over a wide range of temperature from 10°C to 40°C, which is suitable for the next stage of experiments (cell encapsulation and ICC microfabrication) (Figure 1B).

In this study, a high degree of modification was achieved (GelMA DS₃₀₀). The amount of methacryloylation (AM) of GelMA was quantified by ¹H-NMR (AM_NMR). ¹H-NMR was used to quantify both methacrylamide and methacrylate groups in GelMA (Yue et al., 2017; Zheng et al., 2018; Zhu et al., 2019). The ¹H-NMR spectra of GelMA, as compared with gelatin (Figure 1C) show new proton peaks (a+b + e) at 5.4–6.1 ppm. They correspond to the acrylic protons (CH₂ = C(CH₃)CONH-) of methacrylamide groups at 5.7–5.6 and 5.5–5.4 ppm (peaks a and b) and acrylic protons (CH₂ = C(CH₃)COO-) of methacrylate groups at small peaks at 6.1 and 5.7 ppm (peak e). In addition, GelMA also showed a peak at 1.9 ppm (peak d) for methyl protons (CH₂ = C(CH₃)COO- and CH₂ = C(CH₃)CONH-) of the methacryloyl groups, compared with gelatin. When the lysine group of gelatin was completely functionalized with MAA, the specific peak of the unreacted lysine group (NH₂CH₂CH₂CH₂CH₂-) at 3.0 ppm (peak c) completely disappeared. The quantified amount of methacryloyl groups in GelMA is summarized in the table in Figure 2C. The AM_NMR value of GelMA was 1.19 ± .02 mmol/g. According to the previous literature we calculated DS (the equation shows in Supplementary Figure S1) reached to 302.80%. Both the amino groups and the hydroxyl groups are susceptible to substitution, which is why the DS can be higher than 100%. We used GelMA with a higher DS that supported its denser crosslinking and longer stability as well as facilitated ICC fabrication process owing to its lower viscosity at room temperature.