4D morphogenetic tissue engineering via gradient-crosslinked microporous hydrogel scaffolds

Abstract

Shape-morphing hydrogels offer great promise for 4D tissue engineering by enabling dynamic scaffolds that recapitulate morphogenetic transformations. However, their densely crosslinked networks often restrict mass transport, nutrient diffusion, and extracellular matrix remodeling, limiting tissue development. Here, we present a strategy to engineer microporous gradient hydrogels with programmable shape morphing for 4D tissue engineering. Gradient network densities were generated through light-attenuation–mediated photocrosslinking, while interconnected micropores were introduced using sacrificial gelatin microspheres (GMSs). The resulting internal stress mismatch induced differential swelling, enabling controlled shape transformations. By tuning GMS content, photocrosslinking time, and construct geometry, precise control over microporosity, mechanical stiffness, swelling, and deformation behavior was achieved. The constructs supported high cell viability and maintained deformability after cell encapsulation. Complex 3D shapes with varied curvature profiles were readily realized by modulating gradient direction and range. As a proof of concept, mesenchymal stem cell (MSC)-laden constructs were osteogenically differentiated for four weeks to form bone-like tissues. The gradient constructs retained stable curved configurations, and GMS incorporation markedly enhanced alkaline phosphatase (ALP) activity and calcium deposition compared to nonporous controls. This study establishes a versatile and tunable platform for creating microporous gradient hydrogels with spatiotemporal morphing capabilities, offering a new route for developing dynamic, cell-instructive scaffolds in 4D tissue engineering.

Graphical abstract

Highlights

• •Gradient hydrogels are engineered via light-attenuation photocrosslinking.
• •Microporosity is introduced by sacrificial gelatin microspheres (GMSs).
• •Programmable, stable shape morphing is achieved under physiological conditions.
• •High cell viability and deformability are maintained after encapsulation.
• •Enhanced osteogenic differentiation is achieved in microporous gradient constructs.

1. Introduction

Tissue engineering aims to restore or replace damaged biological tissues by integrating cells, biomaterials, and biochemical signals to replicate the structural and functional properties of native tissues [1]. Among various biomaterials, hydrogels have gained particular attention as cell-laden scaffolds due to their high water content, adjustable mechanical properties, and structural similarity to the extracellular matrix (ECM) [2,3]. These polymeric networks provide a three-dimensional (3D) microenvironment that supports cell adhesion, proliferation, and differentiation while allowing the transport of nutrients and signaling molecules [4,5]. However, traditional hydrogel scaffolds are typically static in geometry and lack the ability to respond dynamically to the changing physiological environments observed in natural tissue development and repair [6,7]. As a result, conventional tissue-engineered constructs often fail to fully reproduce the morphogenetic processes and mechanical adaptability inherent in living tissues [8,9].

To address this gap, the concept of four-dimensional (4D) tissue engineering has emerged, incorporating time as the fourth dimension to create constructs capable of autonomous and preprogrammed shape transformations [[10], [11], [12]]. 4D tissue engineering utilizes smart biomaterials that respond to internal or external stimuli, such as temperature, pH, light, or cell-generated forces, to achieve spatiotemporal remodeling of scaffold architectures [[13], [14], [15]]. These dynamic morphing behaviors not only enable engineered tissues to mimic developmental processes like folding, bending, and self-assembly but also facilitate minimally invasive implantation and improved integration with host tissues [[16], [17], [18]]. Consequently, developing 4D-responsive biomaterials represents a promising shift toward creating living constructs that can adapt their form and function during maturation and regeneration.

The realization of 4D tissue engineering depends on strategic material and structural design. Among these strategies, gradient network design has proven particularly effective at programming shape morphing [[19], [20], [21]]. By introducing spatial gradients in crosslinking density, polymer composition, or microstructures, a mismatch in internal stress or strain can be established across the hydrogel network [[22], [23], [24]]. This stress gradient drives controlled deformations such as bending, twisting, or coiling, depending on the magnitude and distribution of the gradient. Gradient hydrogels thus provide a robust method to achieve morphing without external actuation systems, offering high precision and biocompatibility. Moreover, these gradient structures can create microenvironments that better emulate the heterogeneous mechanical and biochemical landscapes of native tissues, guiding cell migration, differentiation, and matrix deposition in a spatially controlled manner [[25], [26], [27]].

Despite these advances, several challenges remain in translating gradient-based 4D hydrogels into functional tissue scaffolds. One key limitation is the dense polymer networks typically formed during crosslinking, which can hinder mass transport, nutrient exchange, waste diffusion, and cellular infiltration, thereby restricting long-term tissue development and remodeling [28,29]. Research has demonstrated that introducing microporosity into hydrogels is a promising strategy to address these issues [30,31]. Microporous architectures, characterized by interconnected pores in the 1 μm to 100 μm range, can significantly enhance permeability, promote cell-matrix interactions, and facilitate the exchange of metabolites. Additionally, combining microporous structures with gradient crosslinking networks has the potential to synergistically integrate mechanical adaptability with mass transport efficiency, allowing scaffolds to sustain dynamic shape changes while maintaining cellular viability and supporting tissue maturation.

This study presents the design and fabrication of gradient-crosslinked microporous hydrogels intended as dynamic scaffolds for 4D shape-morphing in tissue engineering. By utilizing a controlled photopolymerization method combined with gradient light attenuation, a spatially varying crosslinking density within the hydrogel matrix was established. Additionally, gelatin microspheres (GMSs) were incorporated as a porogen to create uniform microporous structures upon dissolution [32,33]. This design combines the necessary crosslinking gradient for 4D shape transformation with the microscopic porosity required for nutrient transport and cellular integration. The structural, mechanical, and swelling behaviors of these hydrogels were systematically characterized, and their capacity to undergo programmed deformations in response to environmental cues was evaluated. Furthermore, cell-laden hydrogels were fabricated to assess their biocompatibility and potential to support tissue morphogenesis during culture. Overall, this study establishes a straightforward yet robust platform for developing microporous, gradient-crosslinked hydrogels that can dynamically adapt in shape and function, thereby advancing the next generation of 4D tissue scaffolds for regenerative medicine.

2. Materials and methods

2.1. Synthesis of ALMA

Methacrylated alginate (ALMA) was synthesized following a modified protocol from previous studies [34]. Briefly, 1 g of sodium alginate (Sigma, derived from brown algae, viscosity ≥2000 cP, 2% w/v, 25 °C; Mw 93 kDa; M/G ratio 1.56) was dissolved in 100 mL of deionized (DI) water under stirring at room temperature. The pH was adjusted to 8.0 using 5 N NaOH. Methacrylic anhydride (10 mL) was then added dropwise (1 mL/min) into the solution under continuous stirring at 4 °C. The reaction was maintained at pH 8.0 for 24 h while allowing the temperature to gradually return to room temperature. The reaction mixture was precipitated into 1 L of chilled methanol, and the resulting white precipitate was collected, dialyzed (MWCO 10 kDa) against DI water for 5 days at room temperature, and lyophilized for 10 days to obtain dry ALMA. The degree of methacrylation was determined to be 13.2% via ¹H NMR.

2.2. RGD functionalization of ALMA

To promote cell adhesion, RGD peptides were covalently conjugated to ALMA through carbodiimide chemistry [35]. ALMA (1% w/v) was dissolved in DI water, followed by sequential addition of EDC·HCl and NHS at a molar ratio of 1:2:10 (EDC:NHS:COOH in alginate). The mixture was stirred at room temperature for 10 min before adding cysteine-L-arginyl-glycyl-L-aspartic acid (cRGD; GenScript, China) at 15 μmol per gram of alginate. The reaction was allowed to proceed overnight under gentle stirring. The product was precipitated in chilled methanol, dialyzed (MWCO 10 kDa) against DI water for 3 days, and lyophilized for 7 days to obtain RGD-functionalized ALMA.

2.3. Synthesis of GMS

Type-A gelatin (Sigma, from porcine skin; Mw 89 kDa; gel strength ∼190 g Bloom) was purified via dialysis (MWCO 14 kDa) in DI water at 60 °C for 5 days and lyophilized. The purified gelatin (5% w/v) was dissolved in DI water at 60 °C, then emulsified dropwise into 200 mL of olive oil under vigorous stirring (800 rpm) to form a water-in-oil emulsion. The mixture was maintained at 60 °C for 30 min and then cooled to 10 °C for 1 h to induce gelation. The emulsion was poured into 1 L of chilled acetone to extract and solidify GMSs. The microspheres were collected by filtration, washed with acetone three times, and vacuum-dried overnight at room temperature. The obtained GMSs were sterilized by UV irradiation (10 mW/cm²) overnight.

2.4. Hydrogel fabrication

To prepare ALMA hydrogels, RGD-functionalized ALMA (2% w/v) was dissolved in DI water containing 0.05% w/v Irgacure 2959 photoinitiator. For gradient formation, 0.03% w/v 4-hydroxy-3-methylacetophenone was added as a photoabsorber. The precursor solution was exposed to UV light (15 mW/cm², 30 s) to obtain either non-gradient (ALMA w/o G) or gradient (ALMA_G) hydrogels.

For GMS-incorporated hydrogels, GMSs were dispersed into the ALMA precursor at specified concentrations (10–50 mg/mL) using mild vortexing to ensure homogeneity before UV crosslinking. The resulting hydrogels were denoted as GMS w/o G (non-gradient) or GMS_G (gradient).

For cell-laden hydrogels, ALMA precursors (2% w/v) were prepared in low-glucose Dulbecco's Modified Eagle Medium (DMEM, Fisher Scientific, USA) containing 0.05% w/v Irgacure 2959 (Sigma, USA), with or without photoabsorber and GMSs. Cells were suspended at 1 × 10⁷ cells/mL before gentle mixing and photocrosslinking (15 mW/cm², for 30 s).

2.5. Swelling and degradation tests

For mass swelling, lyophilized hydrogel discs (D_o = 16 mm) were weighed (m_i) and immersed in DI water at 25 °C overnight. After surface water removal, swollen hydrogels were weighed (m_s), and the swelling ratio (g/g) was calculated as m_s/m_i.

For volume swelling, hydrated hydrogel discs (D_o = 8 mm) were immersed in DI water at 25 °C or 37 °C overnight, and their swollen diameter (D_s) was measured to calculate swelling percentage as (D_s/D_o) × 100%.

For degradation, dried hydrogel discs were incubated in DI water at 25 °C or 37 °C for predetermined durations. The remaining dry mass (m_d) was obtained after lyophilization, and mass loss (%) was calculated as [(m_i - m_d)/m_i] × 100%.

2.6. Scanning Electron Microscopy (SEM)

Hydrogels with different GMS concentrations were incubated at 37 °C for 48 h, flash-frozen in liquid nitrogen, and lyophilized. Samples were sputter-coated with gold (∼7 nm) and imaged using a Hitachi SU3800 SEM at 5.0 kV accelerating voltage.

2.7. Mechanical characterization

Rheological measurements were performed using a DHR-2 rheometer (TA Instruments, USA) with an 8 mm parallel plate geometry and 0.8 mm gap. Dynamic frequency sweeps were conducted from 0.1 to 100 Hz at 1% strain and 25 °C to determine storage modulus (G′). Compressive tests were performed on an MTS CMT6104 instrument (China) up to 80% strain at a rate of 1% strain/s. Young's modulus (E) was calculated from the linear region of the stress–strain curve. Nanoindentation was performed to characterize the microscale mechanical properties of the hydrogels using a nanoindenter (Piuma, Optics11, Amsterdam, The Netherlands) equipped with a cantilever-based probe featuring a spherical tip (radius: 99 μm) and a cantilever stiffness of 0.49 N m⁻¹. The effective Young's modulus was determined by fitting the resulting load–indentation curves using the Hertzian contact model [36].

2.8. Cell culture and expansion

A549 (H0-0701), 293T (H4-1401), hDFs (EZ252286), and rat bone marrow-derived MSCs (RASMX-01001) were obtained from Cyagen and EasyCellBio (China). Cells were cultured in growth medium (DMEM + 10% FBS + 1% penicillin/streptomycin) at 37 °C in a humidified 5% CO₂ incubator. Cells were harvested at 80–90% confluence for encapsulation.

2.9. Live/dead staining

Cell-laden hydrogels were incubated in 1 mL of growth medium containing 1 μL fluorescein diacetate (5 mg/mL) and 3 μL propidium iodide (2 mg/mL) for 5 min at room temperature. Live (green) and dead (red) cells were visualized using an AmScope T800 fluorescence microscope (Irvine, USA).

2.10. Osteogenic differentiation

MSC-laden hydrogels (1 × 10⁷ cells/mL) were cultured in osteogenic medium (DMEM supplemented with 100 nM dexamethasone, 100 μM ascorbic acid, 10 mM β-glycerophosphate, and 50 ng/mL BMP-2) for 28 days with medium changes every other day. Constructs cultured in growth medium served as controls.

DNA content was quantified using the Quant-iT PicoGreen dsDNA kit (P7589, Thermo Fisher, USA) following the manufacturer's protocol (excitation/emission at 480 nm/520 nm). ALP activity was assessed using a Sigma ALP assay kit (MAK447, Sigma, USA) with absorbance measured at 405 nm. Calcium deposition was quantified using a colorimetric assay (ab102505, Abcam, UK) at 575 nm.

2.11. Histology

Cell-laden hydrogel constructs cultured under different conditions were fixed in 10% neutral buffered formalin at room temperature for 6 h, followed by dehydration and paraffin embedding. Sections (6 μm) were prepared, deparaffinized, and rehydrated prior to histological analyses. The samples were subjected to hematoxylin and eosin (H&E) staining, as well as immunofluorescence staining for E-cadherin and F-actin, and alizarin red S (ARS) staining for mineral deposition.

3. Results

3.1. Gradient microporous network engineering and characterization

Hydrogel precursors consisting of photocrosslinkable polymers, GMSs, photoinitiators, photoabsorbers, and live cells were developed to construct gradient, microporous cell-laden hydrogel networks (Fig. 1A). In the formulation, methacrylated alginate (ALMA) was synthesized as the photocrosslinkable polymer (Fig. S1). Alginate-based hydrogels are well known for their cytocompatibility and are widely employed as cell-laden scaffolds; however, their lack of intrinsic cell-adhesive ligands limits effective cell-matrix interactions. To address this limitation, ALMA was further functionalized with RGD peptides to provide integrin-binding sites and enhance cell-ECM adhesion [37].

Fig. 1.

Fabrication and characterization of microporous 4D hydrogels. (A) Schematic illustration of the fabrication process using a gradient crosslinking strategy: (i) photocrosslinking of ALMA into a cell-laden hydrogel with gradient network density; (ii) shape morphing during culture in media; and (iii) GMS dissolution to generate microporous hydrogels. (B) SEM images of hydrogels containing different concentrations of GMS (mg/mL) after swelling in water at 37 °C for 24 h. (C) ¹H NMR spectra of ALMA in the presence of photoinitiator and photoabsorber after UV crosslinking for different durations. (D) Correlation between UV exposure time and the degree of conversion (p) or water uptake capacity of the hydrogel. (E) Young's modulus of hydrogels crosslinked for different UV exposure times. Scale bar: 200 μm.

GMSs were incorporated into the ALMA precursor solution as porogens to create microporous architectures within the hydrogel matrix. The average diameters of GMSs before and after swelling in PBS (pH 7.4) were measured as (18.9 ± 9.2) μm and (35.8 ± 19.9) μm, respectively (Fig. S2). The presence of GMSs did not appear to affect photopolymerization, and thus the crosslinking of ALMA polymers remained unaffected (Fig. S3). GMSs are thermally dissolvable at physiological temperature (37 °C) (Fig. S4), and their removal after crosslinking leaves voids within the hydrogel (Fig. S5). The porosity of the resulting scaffolds increased with higher GMS loading, as evidenced by SEM images showing enhanced pore density following incubation at 37 °C for 24 h (Fig. 1B).

To establish gradient crosslinking, photoabsorbers were introduced into the precursor solution to modulate the light intensity along the optical propagation path during photocrosslinking. When UV light was projected from the top surface, the light intensity decreased with depth, generating a crosslinking gradient through the hydrogel thickness, which is strongest at the top and weakest at the bottom. This attenuation phenomenon follows the Beer–Lambert law (Equation (1)) [38]:

$$I\left(d\right)=I_0{e}^{-\alpha d}$$ (1)

where I is the transmitted light intensity, I₀ is the incident light intensity, d is the light penetration distance, and α is the absorption coefficient determined by the photoabsorber concentration and light wavelength. As light intensity diminishes with increasing depth, the local photopolymerization rate decreases accordingly. The relationship between local light intensity and photoconversion can be described by (Equation (2)) [39]:

$$p\left(d\right)=q\left(1-e^{-k_{1}t}\right)I_0{e}^{-\alpha d}$$ (2)

where p(d) represents the degree of conversion, q is the quantum efficiency of absorbed photons initiating polymerization, k₁ is the reaction rate constant, and t is the UV irradiation time.

To evaluate the photoconversion efficiency of ALMA in the presence of the photoabsorber 4-hydroxy-3-methylacetophenone, ¹H NMR spectra were recorded after different UV exposure times, and the degree of conversion p was calculated using (Equation (3)):

$$p(\%)=\left[1-\frac{{\left(\raisebox{1ex}{${\oint }_{Me}$}\!\left/ \!\raisebox{-1ex}{${\oint }_{IR}$}\right.\right)}_t}{{\left(\raisebox{1ex}{${\oint }_{Me}$}\!\left/ \!\raisebox{-1ex}{${\oint }_{IR}$}\right.\right)}_{t_0}}\right]\times 100\%$$ (3)

where ∮_Me denotes the integral of methacrylate proton peaks (approximately 5.6 and 6.0 ppm, =CH₂), and ∮_IR corresponds to the integral of internal reference proton peaks. The ¹H NMR spectra demonstrated a gradual decrease in methacrylate signal intensity with prolonged UV irradiation (Fig. 1C), confirming progressive photopolymerization. The calculated degree of conversion increased with irradiation time (Fig. 1D, green curve), indicating a UV dose-dependent crosslinking process.

Because the UV dose varies spatially due to light attenuation, photocrosslinking occurs heterogeneously throughout the hydrogel, generating a gradient network structure. Given that hydrogel swelling and mechanical properties are highly dependent on crosslinking density, this spatial heterogeneity resulted in differential water uptake (Fig. 1D, yellow curve) and Young's modulus values (Fig. 1E). The top, more crosslinked regions exhibited higher stiffness and lower swelling, whereas the bottom, less crosslinked regions were softer and more hydrated. This built-in gradient in network density and mechanical strength is expected to induce internal stress mismatches during swelling, leading to predictable, directional deformation, thereby enabling controlled shape morphing of the hydrogel scaffolds during culture in media.

3.2. GMS concentration-dependent swelling, degradation, and mechanical properties

The concentration of GMSs plays a critical role in modulating the microporosity of hydrogels and consequently affects their swelling, degradation, and mechanical properties. To elucidate these effects, a series of microporous gradient hydrogels incorporating different GMS concentrations (10–50 mg/mL) were fabricated (denoted as GMS_Gx, where x represents the GMS concentration in mg/mL). For comparison, hydrogels without GMS incorporation were prepared with and without gradient formation (denoted as ALMA_G and ALMA w/o G, respectively), and a non-gradient microporous hydrogel containing 30 mg/mL GMS (denoted as GMS w/o G30) was used as an additional control.

The swelling behaviors of these hydrogels were evaluated at 25 °C by measuring volumetric expansion (Fig. 2A and B). Both the presence of a gradient and the incorporation of GMSs significantly influenced the swelling capacity. Specifically, gradient hydrogels exhibited markedly higher swelling compared with their non-gradient counterparts (ALMA_G vs ALMA w/o G; GMS_Gx vs GMS w/o G30), attributable to the locally reduced crosslinking density induced by the incorporation of the photoabsorber. This behavior arises from the differential osmotic pressure generated across the crosslinking gradient, which drives internal expansion during swelling.

Fig. 2.

Temperature-dependent swelling and degradation behavior of hydrogels in H₂O. (A) Representative images of hydrogel disks after 24 h incubation at 25 °C. Quantitative analysis of swelling ratios at (B) 25 °C and (C) 37 °C, and degradation profiles at (D) 25 °C and (E) 37 °C. ∗p < 0.05 compared to all other groups; #p < 0.05 compared to other groups except GMS_G10; @p < 0.05 compared to other groups except GMS_G30 and GMS_G50; $p < 0.05 compared to other groups except GMS_G40. Scale bar: 5 mm.

Among the gradient hydrogels, increasing the GMS concentration led to a gradual reduction in overall swelling capacity. This inverse relationship is likely due to the increased solid fraction and reduced polymer network flexibility associated with higher GMS loading. Similar swelling behaviors were observed at 37 °C (Fig. 2C), with the extent of swelling remaining comparable to that at 25 °C. These results are consistent with prior studies reporting that the incorporation of crosslinked GMSs suppresses the equilibrium swelling ratio by reinforcing the hydrogel network structure [40].

The degradation profiles of the hydrogels were examined over 4 weeks of incubation in DMEM at 25 °C and 37 °C (Fig. 2D and E). All hydrogels, regardless of gradient formation or GMS incorporation, exhibited an initial mass loss during the first few days, followed by stabilization over the remainder of the culture period. This initial loss is primarily attributed to the diffusion of unreacted or loosely bound polymer chains, a phenomenon commonly observed in photocrosslinked hydrogel systems [41]. Notably, gradient hydrogels demonstrated slightly higher initial mass loss compared with non-gradient hydrogels, likely due to their lower overall crosslinking density in certain regions. At 25 °C, all GMS_Gx hydrogels displayed similar degradation profiles. However, at 37 °C, an increase in mass loss was observed with increasing GMS content, corresponding to the thermal dissolution of GMSs. The removal of GMSs at physiological temperature not only increased porosity but also facilitated gradual hydrogel softening, which may be advantageous for supporting tissue remodeling.

The mechanical properties of the hydrogels were systematically analyzed under both as-prepared and swollen states (25 °C and 37 °C) (Fig. 3). Across all conditions, non-gradient hydrogels exhibited significantly higher storage modulus (G′) and compressive elastic modulus (E) compared with their gradient counterparts (e.g., ALMA w/o G vs ALMA_G; GMS w/o G30 vs GMS_G30). This difference reflects the higher uniform crosslinking density and reduced internal stress gradients in the non-gradient systems.

Fig. 3.

Mechanical characterization of hydrogels in different states. (A) G′ as a function of frequency, (B) G′ at 1 Hz, and (C) compressive elastic modulus of as-prepared hydrogels. ∗p < 0.05 compared to all other groups; #p < 0.05 compared to other groups except GMS_G50; &p < 0.05 compared to other groups except GMS_G20 and GMS_G30; $p < 0.05 compared to other groups except GMS_G30. (D) G′ as a function of frequency, (E) G′ at 1 Hz, and (F) compressive elastic modulus of hydrogels after swelling at 25 °C. ∗p < 0.05 compared to all other groups; #p < 0.05 compared to other groups except GMS_G30, GMS_G40, and GMS_G50. (G) G′ as a function of frequency, (H) G′ at 1 Hz, and (I) compressive elastic modulus of hydrogels after swelling at 37 °C. ∗p < 0.05 compared to all other groups; #p < 0.05 compared to other groups except GMS_G40 and GMS_G50.

As expected, hydrogels in the as-prepared state exhibited higher moduli than those in the swollen state, due to polymer chain relaxation and increased water content upon swelling. Interestingly, increasing GMS concentration led to distinct trends depending on temperature: at 25 °C, both G′ and E increased with higher GMS concentrations, indicating that the solid microspheres acted as reinforcing fillers within the polymer matrix [42]. In contrast, at 37 °C, G′ and E decreased progressively with increasing GMS content. A similar reduction in local Young's modulus (E∗) after swelling at 37 °C was also observed for GMS-laden hydrogels, such as GMS_G30 (Fig. S6). This reversal arises from the thermal liquefaction and dissolution of GMSs, which disrupts the internal structure and reduces the effective load-bearing capacity of the network [14].

These findings confirm that GMS incorporation not only enables tunable microporosity but also provides a convenient strategy to regulate the hydrogel's mechanical and swelling properties in a temperature-responsive manner. This dual functionality is particularly advantageous for 4D tissue engineering applications, where dynamic remodeling, controlled softening, and mesoporosity are required to accommodate cellular growth and tissue morphogenesis.

3.3. Shape-morphing behavior of gradient microporous hydrogels

Hydrogels with gradient network densities are anticipated to undergo shape morphing when cultured in aqueous environments, owing to differential swelling across the gradient. To investigate this behavior, gradient microporous hydrogels containing 30 mg/mL GMSs (GMS_G30) were fabricated in straight strip geometries and cultured in water at 25 °C and 37 °C. For comparison, non-gradient counterparts (ALMA w/o G and GMS w/o G30) and gradient hydrogels without GMS incorporation (ALMA_G) were also examined.

After 24 h of incubation to achieve equilibrium swelling, both ALMA_G and GMS_G30 hydrogels displayed pronounced bending and formed curled configurations, whereas their non-gradient counterparts (ALMA w/o G and GMS w/o G30) retained their original straight morphology (Fig. 4A). The formation and orientation of the crosslinking gradient were directly visualized in the photomicrographs of rhodamine B-conjugated ALMA networks, revealing a distinct intensity gradient along the hydrogel thickness (Fig. S7). This observation was further corroborated by cross-sectional mechanical profiling using nanoindentation, which showed a progressive decrease in mechanical properties with increasing depth (Fig. S8), consistent with a gradient crosslinking density [18]. Comparable bending deformations were observed at both 25 °C and 37 °C, consistent with the similar swelling and mechanical properties obtained previously under these conditions. Quantitatively, ALMA_G exhibited significantly higher bending angles than GMS_G30 at both temperatures (Fig. 4B and C), which can be attributed to the higher swelling capacity of ALMA_G relative to GMS_G30.

Fig. 4.

Temperature- and medium-dependent bending behavior of hydrogel strips. (A) Representative images of bent strips in H₂O at 25 °C and 37 °C. Rhodamine B was conjugated to ALMA networks for fluorescence visualization. Quantitative analysis of bending angles at (B) 25 °C and (C) 37 °C. (D) Time-dependent changes in bending angles of strips cultured in H₂O at 25 °C and 37 °C. (E) Representative images and (F) corresponding bending angles of GMS_G30 strips immersed in H₂O, PBS, and DMEM at 37 °C. Scale bars: 5 mm. The GMS_G30 image in H₂O at 37 °C shown in Panel E is the same representative image as that shown in Panel A, because both panels depict the identical experimental condition.

The morphing kinetics of these hydrogel strips were further characterized at both temperatures (Videos S1–S5). As shown in Fig. 4D and summarized in Table S1, non-gradient hydrogels exhibited no detectable bending throughout the entire incubation period. In contrast, the gradient hydrogels demonstrated rapid and substantial shape transformation. At 25 °C, ALMA_G showed faster and larger morphing compared to GMS_G30, reaching its maximum bending within 120 s (3.95° s⁻¹ from 20 to 120 s, total ∼540°), while GMS_G30 reached maximum bending within 240 s (3.64° s⁻¹ from 20 to 60 s, ∼160°; 0.66° s⁻¹ from 60 to 240 s, ∼320°). Increasing the temperature from 25 °C to 37 °C markedly accelerated the morphing kinetics, as evidenced by the higher rate observed for GMS_G30 (3.84° s⁻¹ from 20 to 120 s, total ∼320°) at 37 °C.

To further examine environmental responsiveness, the morphing behaviors of GMS_G30 hydrogels were assessed in different media, including water, PBS, and DMEM at 37 °C. As shown in Fig. 4E and F, hydrogel strips in H₂O exhibited the most pronounced bending, with significantly greater bending angles compared to those cultured in PBS or DMEM, while the latter two conditions yielded comparable deformation levels. These results suggest that ionic strength and medium composition can modulate the morphing behavior, likely through their effects on osmotic swelling pressure and polymer-solvent interactions.

3.4. Impact of fabrication parameters on shape morphing

The influence of GMS concentration, UV crosslinking time, and hydrogel geometry on the shape-morphing behavior of gradient microporous hydrogels was systematically examined. As shown in Fig. 5A, increasing GMS concentration from 10 to 50 mg/mL led to a progressive decrease in bending angles, which can be attributed to the reduced swelling capacity at higher GMS loadings, as previously demonstrated. Similarly, extending the UV exposure time from 10 s to 60 s resulted in a gradual decline in bending angles (Fig. 5B), owing to the enhanced crosslinking density and thus reduced differential swelling between the top and bottom layers.

Fig. 5.

Systematic evaluation of fabrication parameters on hydrogel strip bending after 24 h incubation in media at 37 °C: effects of (A) GMS concentration, (B) UV crosslinking time, (C) strip length, (D) strip thickness, and (E) strip width. Unless otherwise specified, hydrogel strips were fabricated with 30 mg/mL GMS, 30 s UV crosslinking time, and dimensions of 20 mm (length) × 3 mm (width) × 0.8 mm (thickness). GMS concentration was evaluated in H₂O, while other parameters were evaluated in DMEM. ∗p < 0.05%. Scale bars: 5 mm.

The effect of hydrogel dimensions, including strip length, thickness, and width, was further investigated (Fig. 5C–E). Increasing both the length and thickness of the hydrogel strips enhanced their overall bending, whereas variations in width exhibited negligible influence. Notably, while longer strips displayed greater total bending angles, the bending curvature (Fig. S9) remained unchanged, as the internal gradient range within each sample was constant. In contrast, thickness-dependent bending was more pronounced due to the direct modulation of the gradient range: thicker hydrogels experienced a broader crosslinking gradient, resulting in greater differential stress and thus more substantial deformation. This observation aligns well with previous studies [8]. Conversely, increasing the strip width up to 8 mm did not produce a significant effect on bending (Fig. S10).

Photoinitiator absorption and light intensity significantly influence the gradient range and, consequently, the deformability of the hydrogels. As described by Equations (1) and (2), reducing light absorption or lowering light intensity decreases photoconversion efficiency, which enhances the formation of the crosslinking gradient and, in turn, increases the deformation of gradient hydrogels. Consistently, increasing the photoabsorber concentration from 0.01% to 0.03% significantly enhanced the bending angles (Fig. S11A), as the reduced light penetration promoted a wider gradient range. Conversely, increasing the light intensity from 10 to 50 mW cm⁻² led to decreased bending angles (Fig. S11B) due to a more uniform crosslinking profile and a reduced gradient range.

These results demonstrate that tuning fabrication conditions and geometric parameters provides a simple yet effective strategy to modulate the deformation behavior of gradient microporous hydrogels, offering valuable design flexibility for constructing shape-morphing tissue scaffolds with programmable geometries and mechanical responses.

3.5. Live cell incorporation and shape morphing

The capability of the gradient microporous hydrogels to serve as cell-laden scaffolds while maintaining shape-morphing functionality was systematically investigated. Four representative cell types, including A549, 293T, hDFs, and MSCs, were encapsulated at a concentration of 1.0 × 10⁷ cells/mL within the hydrogel precursor solution. After incubation in FBS-containing DMEM at 37 °C for 24 h, all cell-laden hydrogel strips exhibited pronounced bending into “C”-shaped configurations with comparable bending angles (Fig. 6A and B). The bending angles were similar to those of the corresponding cell-free hydrogels, indicating that cell incorporation exerted a negligible influence on the shape-morphing behavior under these conditions. Live/dead staining confirmed high cell viability within the constructs (Fig. 6C–S12), with all encapsulated cell types maintaining >90% viability after 24 h (Fig. 6D), demonstrating excellent cytocompatibility of the photocrosslinked gradient-based network.

Fig. 6.

Shape morphing behavior and cell viability of cell-laden hydrogels. (A) Representative images and (B) quantified bending angles of GMS_G30 strips encapsulating different cell types. Scale bar: 5 mm. (C) Live (green)/dead (red) fluorescence images and (D) quantified viability of encapsulated cells. “ND” indicates no statistically significant difference detected. Scale bar: 100 μm. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)

To further explore the versatility of gradient encoding for creating more complex, multicurvature morphing, cell-laden hydrogels with designed gradient orientations and segmental structures were fabricated (Fig. 7). For instance, introducing two segments with opposite gradient directions but identical gradient ranges (Fig. S13A) yielded a symmetric “S”-shaped construct, whereas implementing opposite gradient directions with different gradient ranges (Fig. S13B) produced an asymmetric “S” shape. Moreover, a tri-segmental design with the same gradient orientation and range in Segments 1 and 3, and a non-gradient middle Segment 2 (Fig. S13C), generated a “heart”-like configuration.

Fig. 7.

Complex cell-laden architectures generated via 4D shape transformation of gradient microporous hydrogels. Scale bars: 5 mm.

Beyond linear constructs, complex architectures were achieved by varying the initial hydrogel geometries. Cross-shaped precursors transformed into “four-arm gripper” structures, while sheet-, circular-, and fan-based geometries underwent predictable bending into corresponding curved forms. These results collectively demonstrate that by rationally programming the initial geometry and spatial gradient orientation, diverse, self-evolving 3D tissue-like architectures can be engineered from simple 2D precursors, highlighting the broad potential of this system for complex tissue architecture fabrication.

Complex cell-laden architectures with substantially reduced feature sizes were further fabricated in a batch manner using a photomask-based spatial patterning strategy. Photomasks containing arrays of cross- and star-shaped patterns were placed over the hydrogel precursors prior to photocrosslinking, yielding multiple microscale constructs (∼200 μm in thickness) with programmed shape-morphing capability. Upon incubation in PBS at 25 °C, these constructs rapidly transformed into well-defined curved configurations (Fig. S14) and exhibited high structural stability, remaining intact under pipette flushing (Video S6).

The engineered curvatures are expected to improve conformal adaptation to curved substrates. To demonstrate this potential, a curved disc construct was generated via the 4D shape-morphing process and applied to a smooth glass tube, with a non-curved disc serving as a control. The programmed curved construct exhibited intimate, full-surface contact with the glass tube, whereas the non-curved counterpart showed only minimal attachment (Fig. S15). Consequently, the curved construct remained firmly adhered during tube rotation, while the non-curved construct readily detached (Video S7).

3.6. Cell morphology and proliferation during culture

The incorporation of pores within hydrogels is expected to facilitate cell proliferation by creating void spaces and enhancing nutrient and metabolite exchange between the hydrogel interior and the surrounding medium. Cell morphology and proliferation were continuously monitored over 21 days of culture in GM. In the nonporous ALMA_G30 constructs, cells predominantly exhibited a rounded morphology and were evenly distributed throughout the hydrogel. Although high cell viability was maintained, no noticeable increase in cell density was observed during the culture period (Fig. 8A). Consistent with this observation, DNA quantification revealed minimal change over the first 14 days and a modest increase (2.3-fold) by Day 21 compared to Day 0 (Fig. S16A and B).

Fig. 8.

Morphology and live/dead staining of MSC-laden hydrogel constructs over 21 days of culture. (A) ALMA_G30 and (B) GMS_G30. Blue arrows indicate partial GMSs, while yellow arrows mark partial cell clusters. Scale bars: 100 μm. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)

In contrast, the GMS_G30 constructs exhibited pronounced changes in cell morphology and spatial distribution over time (Fig. 8B). Initially, the presence of GMSs was clearly visible (blue arrows), and cells appeared individually dispersed in a rounded shape within the hydrogel matrix. By Day 7, the GMSs had dissolved, leaving behind observable voids. However, cells remained largely isolated with no distinct clustering. At Day 14, distinct cell clusters began to appear, likely due to cell migration and colonization within the void spaces generated by GMS dissolution. Continued culture to Day 21 led to the formation of larger and denser clusters, accompanied by a significant 2.35-fold increase in DNA content relative to Day 0 (Fig. S16C and D).

H&E, E-cadherin, and F-actin staining were performed over 21 days of culture to gain further insight into cellular behavior within the constructs. Consistent with the trends observed in live/dead staining and proliferation analyses, the histological results revealed similar patterns of cell morphology and spatial distribution over time. Specifically, cells encapsulated in nonporous ALMA_G30 constructs maintained a predominantly round morphology and were uniformly distributed throughout the constructs during the entire culture period (Fig. S17). In contrast, GMS_G30 constructs exhibited progressive cell clustering with prolonged culture (Fig. 9). Notably, the expression of E-cadherin, which plays a key role in mediating cell-cell adhesion and regulating cell migration, and F-actin, a critical cytoskeletal component governing cell morphology, motility, and contractility, was markedly enhanced in GMS_G30 constructs.

Fig. 9.

Histological analysis of cell-laden ALMA_G30 hydrogels over 21 days of culture: (A) H&E staining, (B) E-cadherin staining, and (C) F-actin staining. Scale bars: 50 μm.

These results demonstrate that introducing microporosity through GMS incorporation markedly enhances cell infiltration, aggregation, and proliferation within the hydrogel matrix. The findings underscore the pivotal role of micropores in promoting cellular growth within engineered hydrogel scaffolds.

3.7. 4D tissue engineering

4D hydrogels offer a transformative approach to tissue engineering by providing dynamic scaffolds capable of evolving their structure and mechanical environment in response to external or programmed cues [43]. Such dynamicity enables the recapitulation of tissue morphogenesis processes and facilitates the generation of functional, spatially organized tissue constructs [44,45]. To demonstrate the potential of the gradient microporous hydrogel platform for 4D tissue engineering, we conducted a proof-of-concept bone-like tissue formation study by differentiating encapsulated MSCs within osteogenic environments over a 4-week culture period.

Specifically, MSC-laden gradient microporous hydrogels (GMS_G30) were cultured in osteogenic medium [denoted as GMS_G30 (OM)], while non-porous gradient hydrogels [ALMA_G (OM)] served as controls to assess the contribution of microporosity to osteogenic outcomes. In addition, non-gradient counterparts, ALMA w/o G (OM) and GMS w/o G30 (OM), were included for comparison, along with GMS_G30 (GM) and GMS w/o G30 (GM) constructs cultured in standard growth medium to serve as non-differentiation controls.

Throughout the 4-week culture, gradient hydrogels maintained their predefined curved configurations with excellent structural integrity, while non-gradient constructs remained undeformed regardless of the medium (Figs. 10A and 9B). Consistent with the continuous live/dead staining and DNA assays described above, quantification of DNA content at Week 4 revealed significantly higher cell numbers in GMS-incorporated constructs (Fig. 10C). Moreover, comparable DNA levels were detected among all GMS-incorporated groups, independent of the culture medium or 4D shape-morphing process.

Fig. 10.

4D osteogenic tissue engineering. (A) Representative images of cell-laden strips cultured in different media for 4 weeks. Scale bar: 5 mm. (B) Bending angles of cell-laden strips over 4 weeks of culture. Quantitative analysis of (C) DNA content, (D) ALP/DNA ratio, and (E) calcium/DNA ratio of cell-laden strips at Week 4. (F) ARS-stained strips. Scale bar: 5 mm. (G) ARS staining of sectioned constructs. Scale bar: 100 μm. (i) ALMA w/o G (OM); (ii) GMS w/o G30 (GM); (iii) GMS w/o G30 (OM); (iv) ALMA_G (OM); (v) GMS_G30 (GM); (vi) GMS_ G30 (OM). ∗,#p < 0.05 compared to groups with a different symbol or unlabeled.

Osteogenic differentiation was assessed via ALP activity and calcium deposition, both normalized to DNA content (Fig. 10D and E). Constructs cultured in growth medium [GMS_G30 (GM) and GMS w/o G30 (GM)] exhibited minimal ALP activity and calcium accumulation, whereas those in osteogenic medium displayed markedly elevated levels of both biomarkers. Notably, microporous constructs [GMS_G30 (OM) and GMS w/o G30 (OM)] showed significantly higher ALP expression and calcium deposition than their non-porous counterparts [ALMA_G (OM) and ALMA w/o G (OM)], underscoring the role of microporosity in promoting nutrient transport and facilitating osteogenic matrix mineralization.

Macroscopic ARS staining corroborated these quantitative results (Fig. 10F): constructs cultured in osteogenic medium displayed intense red staining indicative of mineralized matrix formation, while those in growth medium exhibited only faint staining. Although the macroscopic ARS staining intensity appeared similar between GMS- and ALMA-based constructs, histological analysis revealed a distinct difference (Fig. 10G). GMS-based tissues exhibited more extensive and spatially enriched mineral deposition, consistent with the enhanced osteogenic marker expression observed in quantitative assays.

3.8. Cellular responses to gradient mechanics

The impact of gradient mechanics on cell behavior is of considerable interest due to its biomimetic relevance to many regenerating tissues, such as osteochondral defects, which exhibit a continuous transition from relatively soft cartilage to stiffer subchondral bone [46]. Previous studies have demonstrated that hydrogel constructs with long-range mechanical gradients spanning several orders of magnitude (e.g., from kPa to MPa) can markedly influence MSC osteogenesis through mechanotransduction pathways [47].

In the present study, however, direct investigation of gradient mechanics was challenging due to the limited hydrogel thickness (0.8 mm), within which the achievable mechanical gradient was relatively narrow. The mechanical gradient in our system arises from differential photoconversion along the thickness direction induced by controlled photocrosslinking. To evaluate whether such gradient mechanics could influence cellular behaviors, including viability, proliferation, and osteogenic differentiation, we fabricated cell-laden hydrogels subjected to photocrosslinking times of 30 s (mimicking the top region), 25 s (middle region), and 20 s (bottom region). Two construct types, ALMA w/o G and GMS w/o G30, were examined and cultured in osteogenic medium for 21 days.

Live/dead staining revealed comparable cell morphology, distribution, and clustering across constructs with different photocrosslinking times within each material group. Specifically, cells encapsulated in ALMA-based constructs maintained a rounded morphology with homogeneous distribution, whereas cells in GMS-based constructs formed pronounced multicellular clusters (Fig. S18). High cell viability (>90%) was maintained in all groups, and DNA quantification showed no significant differences among constructs receiving different photocrosslinking durations within the same material group (Fig. S19). Furthermore, osteogenic markers, including ALP activity and calcium deposition, exhibited comparable levels among constructs with varying photocrosslinking times (Fig. S20), although GMS-based constructs consistently showed enhanced osteogenesis relative to ALMA-based counterparts, in agreement with the 4D morphogenetic differentiation results described above.

4. Discussion

4D biofabrication represents a transformative advancement in tissue engineering, enabling the creation of dynamic, stimuli-responsive constructs that undergo programmed shape transformations over time. This technology allows engineered tissues to recapitulate complex morphogenetic processes of native organs, facilitating controlled folding, bending, or expansion that can mimic developmental or regenerative pathways. Such dynamic capability offers unique benefits for tissue morphogenesis and enhanced tissue regeneration, including improved cellular organization, spatially controlled differentiation, and more efficient tissue matrix deposition. Moreover, 4D biofabricated constructs have demonstrated seamless, sutureless integration with host tissues, reducing surgical complexity and promoting stable anatomical repair [[48], [49], [50]]. These attributes are particularly promising for vascular tissue engineering and the treatment of aneurysms [51], where precise geometrical conformity, robust tissue integration, and minimally invasive deployment are critical for clinical success.

Gradient design in hydrogels has emerged as an important strategy to enable 4D transformations, offering advantages such as simplicity and tunability over multilayered hydrogel systems. A variety of approaches have been developed to engineer gradient hydrogels, including grayscale printing, phase separation, and microfluidics [19]. In particular, light-attenuation–mediated photolithography provides a robust platform to generate gradient crosslinking with high reproducibility. Recent advances have highlighted the significant progress of graded hydrogels for biomedical applications. However, a critical limitation in the field remains the widespread use of bulk hydrogels with low porosity, which hinders nutrient transport, cell infiltration, and effective tissue matrix production.

In this study, we developed microporous gradient hydrogels with programmable shape morphing for 4D tissue engineering by integrating light-attenuation–mediated gradient crosslinking with sacrificial GMSs. The resulting hydrogels exhibited tunable mechanical stiffness, swelling behavior, and curvature profiles, enabling precise control over 4D deformation (Fig. 2, Fig. 3, Fig. 4, Fig. 5). Incorporation of GMSs introduced interconnected micropores, markedly enhancing cell infiltration, aggregation, and proliferation, as demonstrated by live/dead imaging, DNA quantification, and cytoskeletal organization (Fig. 8, Fig. 9). Furthermore, MSC-laden constructs exhibited improved osteogenic differentiation, with higher ALP activity and calcium deposition in GMS-containing gradient hydrogels compared to nonporous controls (Fig. 10).

When comparing 4D-curved and non-4D straight constructs, the former exhibited slightly, but not significantly, higher osteogenic marker expression and ARS staining intensity. This indicates that, under the present conditions, programmed morphing neither compromised cell differentiation nor substantially enhanced osteogenic outcomes. Nevertheless, prior studies have shown that 4D scaffolds can promote tissue regeneration through anisotropic mechanical stress distributions and enhanced cell mechanotransduction [24,52]. Thus, further tuning of hydrogel degradability, stiffness, and morphing kinetics may amplify the regenerative benefits of dynamic morphing, paving the way for advanced 4D tissue constructs that better emulate in vivo morphogenetic processes.

Investigation of gradient mechanics on cellular behavior revealed no significant differences across constructs engineered with different photocrosslinking times. This suggests that, within the relatively narrow mechanical gradient generated, differential photocrosslinking did not elicit measurable effects on cell proliferation, viability, or differentiation. The limited gradient magnitude likely explains the minimal mechanotransduction responses, indicating that future studies employing thicker constructs or steeper, multiscale gradients may be necessary to fully elucidate the role of mechanical heterogeneity in guiding cell fate.

This platform also enables high-throughput fabrication of small-sized, shape-programmable, cell-laden constructs via photomask-based spatial patterning, providing opportunities for both translational studies and fundamental mechanobiology investigations involving cell-matrix and cell-cell interactions. Advanced vat-polymerization–based 3D printing technologies, such as stereolithography (SLA) and digital light processing (DLP), have been applied to engineer shape-morphing hydrogel scaffolds [53,54]. Owing to their high microscale resolution, these techniques may further expand the fabrication of downscaled tissue constructs with fine architectural features. However, successful implementation generally requires careful reduction of resin viscosity (e.g., <5 Pa s) to ensure smooth layer replenishment and recoating, which remains a critical consideration for cell-laden systems [55].

Alginate-based hydrogels demonstrate excellent biocompatibility and have been widely reported to elicit minimal inflammatory responses following in vivo implantation [56]. Owing to these properties, alginate hydrogels are often used as wound-healing platforms, particularly when combined with anti-inflammatory or bioactive agents such as simvastatin [57] or bioactive glasses [58] to further modulate local inflammatory responses. Nevertheless, the relatively slow degradation of alginate hydrogels must be carefully matched with new tissue formation and remodeling to prevent long-term persistence or adverse host responses. The proof-of-concept demonstration highlights the potential of 4D shape-morphed constructs to improve conformal integration with anatomically curved tissue defects in vivo, such as those on organ surfaces, within the trachea, and along vascular or other tubular structures. By enabling programmed curvature and intimate surface adaptation, these dynamic constructs may reduce interfacial gaps and micromotion at the tissue–implant interface, which are known to impair integration and exacerbate foreign body responses. Collectively, these results suggest that 4D shape-morphing hydrogel platforms hold promise for promoting seamless tissue integration in regenerative applications involving complex and curved anatomical environments.

Despite the advantages demonstrated, several limitations should be acknowledged. First, although gradient crosslinking enabled programmable shape morphing, the spatial resolution of the gradient was primarily governed by light attenuation and construct thickness, which may limit the precise generation of complex or multiaxial deformation patterns. Second, microporosity was introduced using sacrificial GMSs with a relatively broad size distribution, resulting in heterogeneous pore sizes that may lead to nonuniform local mechanical and biological microenvironments; refined porogen fabrication or microfluidic strategies could improve pore uniformity and interconnectivity. Third, while proof-of-concept seamless integration with smooth curved surfaces was demonstrated, and in vitro cell viability, proliferation, and osteogenic differentiation were enhanced, long-term matrix remodeling and in vivo performance were not assessed, which are critical for translational applications. In addition, the current system relied on nondegradable or slowly degradable hydrogel networks, potentially limiting tissue maturation over extended culture periods. Finally, although microscale fabrication was demonstrated via photomask-based patterning, systematic integration with advanced high-resolution 3D printing remains to be explored. Addressing these limitations in future work will be essential to fully realize the potential of this platform for advanced 4D tissue engineering and regenerative medicine.

5. Conclusions

In summary, a facile yet powerful strategy for engineering microporous gradient hydrogels capable of programmable shape morphing was developed for 4D tissue engineering applications. By integrating light-attenuation–mediated gradient photocrosslinking with GMS incorporation, simultaneous control over crosslinking density and microporous architecture within a single construct was achieved. The resulting hydrogels exhibited tunable mechanical properties, swelling ratios, and morphing behaviors, enabling predictable and reversible deformations under physiological conditions. The system maintained high cytocompatibility across multiple cell types and supported cell-laden morphogenesis without compromising cell viability or scaffold integrity. Through rational modulation of gradient direction, range, and initial geometry, diverse 3D configurations ranging from symmetric and asymmetric “S” shapes to complex “heart” and gripper-like forms were realized, demonstrating the platform's programmability and scalability. Furthermore, osteogenic differentiation of MSC-laden constructs validated the functional utility of the system, where microporosity significantly enhanced ALP activity and calcium mineralization compared to nonporous controls. The synergistic integration of gradient crosslinking and microporous structuring offers a promising route toward creating next-generation 4D biomaterials capable of guiding tissue morphogenesis, promoting regeneration, and advancing the frontiers of biofabrication and regenerative medicine.

CRediT authorship contribution statement

Haitao Yu: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Writing – original draft, Writing – review & editing. Guodong Wu: Formal analysis, Funding acquisition, Methodology. Jian Zhang: Conceptualization, Formal analysis, Investigation, Methodology, Resources, Supervision, Writing – original draft, Writing – review & editing.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Appendix A. Supplementary data

The following is/are the supplementary data to this article:

Data availability

Data will be made available on request.