Spatial organization of magnetic nanoparticles in chitosan-gelatin hydrogels enables enhanced magnetic actuation and cell response under dynamic stimulation

Abstract

Magnetically responsive hydrogels have emerged as promising platforms for remotely modulating HFF-1 fibroblasts behavior through dynamic mechanical stimulation; however, the influence of magnetic nanoparticle spatial organization on magnetic actuation and cell response remains unexplored. Herein, we report the development of chitosan-gelatin magnetic hydrogels with controlled Fe₃O₄ nanoparticle distribution including homogeneously dispersed, and spatially patterned configurations, as a strategy to engineer localized mechanotransduction and magnetic actuation. The role of nanoparticle organization on the physicochemical, magnetic, and biological properties of the hydrogels has been analyzed. Structural characterization confirmed the successful incorporation of Fe₃O₄ nanoparticles while preserving their crystalline inverse spinel structure. Morphological and topographical analyses revealed that patterned hydrogels exhibit well-defined nanoparticle-rich and nanoparticle-free regions, generating pronounced microscale heterogeneity throughout the material. The incorporation and spatial confinement of Fe₃O₄ nanoparticles significantly influenced the hydrogel properties, increasing stiffness, reducing swelling capacity, and enhancing magnetic responsiveness. Notably, patterned hydrogels displayed higher saturation magnetization and magnetic susceptibility than homogeneously dispersed systems, indicating that nanoparticle confinement promotes cooperative magnetic interactions and more efficient magnetic actuation. Biological studies demonstrated excellent cytocompatibility for all hydrogel formulations. While static magnetic stimulation slightly reduced fibroblast proliferation, dynamic magnetic stimulation significantly enhanced cell proliferation, spreading, and cytoskeletal organization, particularly in patterned hydrogels. These effects are attributed to cyclic mechanical cues generated by amplified magnetic-to-mechanical transduction within the patterned magnetic domains. Overall, this study demonstrates that nanoparticle spatial organization is a key design parameter for engineering magnetically responsive hydrogels with tunable magnetic actuation and controllable cellular responses for tissue engineering applications.

Graphical abstract

Spatial confinement of Fe₃O₄ nanoparticles within chitosan-gelatin hydrogels generates localized magnetic domains with enhanced magnetic susceptibility and actuation under dynamic magnetic stimulation. The amplified magnetic-to-mechanical transduction promotes fibroblast proliferation, spreading, and cytoskeletal organization, demonstrating nanoparticle patterning as an effective strategy for engineering magnetically responsive biomaterials for tissue engineering.

Highlights

• •Spatial Fe₃O₄ confinement enhanced magnetic susceptibility and responsiveness.
• •Patterned films generated amplified deformation under magnetic stimulation.
• •Dynamic actuation enhanced fibroblast proliferation, spreading and organization.
• •Nanoparticle spatial organization is a key design parameter for magnetic scaffolds.

1. Introduction

Hydrogels have emerged as one of the most promising classes of biomaterials for tissue engineering and regenerative medicine due to their high water content, soft nature, and structural similarity to the extracellular matrix (ECM). Their highly hydrated three-dimensional polymeric networks provide a favorable microenvironment for cell adhesion, proliferation, migration, and nutrient transport, making them attractive platforms for the development of biomimetic scaffolds and bioactive systems [1], [2]. In recent years, increasing attention has been directed toward the design of dynamic and stimuli-responsive hydrogels capable of actively interacting with cells and tissues rather than serving solely as passive structural supports [3], [4]. Among these systems, magnetically responsive hydrogels have attracted particular interest because magnetic fields can remotely and non-invasively modulate material properties and cellular behavior with high spatial and temporal control [5], [6].

Magnetic hydrogels are typically fabricated through the incorporation of magnetic nanoparticles (NPs), most commonly iron oxide NPs such as magnetite (Fe₃O₄), into polymeric matrices [7], [8], [9]. Fe₃O₄ NPs are especially attractive for biomedical applications owing to their biocompatibility, magnetic responsiveness, chemical stability, and ability to generate mechanical, thermal, or structural changes under external magnetic fields [10], [11]. When exposed to magnetic stimulation, these nanocomposite systems can undergo deformation, displacement, or local mechanical actuation, enabling dynamic modulation of the cellular microenvironment [12], [13]. Such magnetically induced effects have been explored for a broad range of applications including drug delivery, soft robotics, biosensing, hyperthermia, and tissue engineering [14], [15].

In the context of tissue engineering, magnetic stimulation has emerged as a powerful strategy to regulate cell behaviour through mechanotransduction pathways [16], [17]. Cells are highly sensitive to mechanical cues from their surrounding microenvironment, including substrate stiffness, topography, cyclic strain, and dynamic deformation [18], [19]. Dynamic mechanical stimulation has been shown to influence cytoskeletal organization, cell spreading, migration, proliferation, and lineage commitment [20], [21]. Magnetic hydrogels provide a unique platform to deliver such mechanical stimuli remotely, enabling contactless activation of the scaffold through externally applied magnetic fields [22]. In particular, dynamic magnetic stimulation can generate cyclic microscale deformations within the hydrogel matrix, which may mimic aspects of the dynamic mechanical environment present in native tissues [23]. Although magnetically responsive hydrogels have been investigated for tissue engineering, most studies have focused primarily on a homogeneous distribution of magnetic nanoparticles throughout the polymeric network where the main variables were nanoparticle concentration, matrix composition, or the application of external magnetic fields [24], [25], [26], [27], [28], [29]. However, biological tissues are intrinsically heterogeneous and frequently exhibit spatial variations in composition, structure, stiffness, and biochemical signalling. Consequently, homogeneous NPs distributions may limit the ability of magnetic hydrogels to reproduce the complex physicochemical gradients and localized mechanical cues characteristic of native tissue microenvironments. Thus, the fabrication of hydrogels with an ordered/anisotropic distribution of the magnetic NPs and/or magnetic gradients within the matrix have been more recently explored to mimic native tissues and demonstrating the potential of magnetic organization to introduce directional functionality [25], [26], [27], [28], [29]. For example, Antman-Passig et al. incorporated magnetic nanoparticles into collagen hydrogels and applied a magnetic field during gelation, inducing NPs assembly into chain-like structures that promoted collagen-fiber alignment and subsequently guided neuronal growth [26]. Araújo-Custódio et al. developed injectable gelatin hydrogels containing magnetically responsive cellulose nanocrystals, in which magnetic-field-induced NPs alignment generated directional microstructures and anisotropic mechanical properties that promoted preferential cell orientation [27]. Similarly, magnetically responsive short fibers have been aligned within fibrin hydrogels to create anisotropic matrices capable of directing fibroblast and neuronal growth [28], while magnetic NPs-mediated alignment of collagen has been exploited to generate tendon-mimetic constructs that guide stem-cell organization along the oriented collagen architecture [29].

Collectively, previous studies have demonstrated the potential of magnetic fields to generate ordered and anisotropic hydrogel microenvironments by inducing the alignment of magnetic NPs, rods, fibers, or extracellular-matrix components, thereby creating directional structural and mechanical cues for cell guidance and tissue engineering. However, considerably less attention has been devoted to the deliberate spatial confinement of the magnetic phase into discrete NPs-rich domains and, particularly, to understanding how such heterogeneous magnetic organization influences material properties and cellular responses during subsequent magnetic stimulation. In this context, the distinctive feature of the present work is not the incorporation of Fe₃O₄ NPs per se, nor the generation of structural anisotropy, but rather the investigation of NPs spatial distribution as a material-design variable. To this end, chitosan-gelatin hydrogels containing spatially localized Fe₃O₄-rich domains are directly compared with hydrogels containing homogeneously dispersed Fe₃O₄, while maintaining the same nominal nanoparticle loading and polymer composition. This controlled design enables the influence of magnetic-phase organization on hydrogel morphology, topography, bulk mechanical properties, swelling behavior, magnetic responsiveness, and cellular behavior to be systematically evaluated. We hypothesize that concentrating Fe₃O₄ NPs within defined regions generates local physicochemical and magnetic heterogeneity, potentially modifying interparticle magnetic interactions and the response of the material to an externally applied magnetic field. Importantly, whereas previous ordered magnetic hydrogel approaches have predominantly exploited magnetic fields during fabrication to establish and preserve structural anisotropy for directional cell guidance, here the patterned magnetic domains are subsequently subjected to both static and dynamic magnetic stimulation to investigate how their spatial organization is associated with the cellular response. Thus, rather than using magnetic organization solely to generate a permanent anisotropic architecture, this work explores spatial patterning of the magnetic phase as a strategy for tuning the functional responsiveness of the hydrogel during remote magnetic stimulation, establishing a framework for the design of magnetically active and spatially heterogeneous biomaterials for tissue-engineering applications.

Herein, we report magnetically responsive chitosan-gelatin hydrogels in which the spatial distribution of Fe₃O₄ NPs is controlled to generate either homogeneously dispersed or spatially patterned magnetic phases, together with a non-magnetic hydrogel as reference. Chitosan and gelatin were selected as complementary natural polymers because of their favorable physicochemical and biological properties, including film-forming capability, biocompatibility, and the presence of cell-adhesive motifs [30], [31]. Although Fe₃O₄-containing chitosan-gelatin systems have previously been investigated as magnetically responsive biomaterials [32], the consequences of deliberately redistributing the magnetic phase into discrete NPs-rich and NPs-poor domains remain comparatively less understood. Here, a magnetic template was employed during fabrication to induce localized NPs confinement while maintaining the same nominal Fe₃O₄ loading and polymer composition as the homogeneously dispersed magnetic hydrogel. This design allowed us to investigate how NPs spatial distribution affects hydrogel morphology, topography, mechanical and swelling properties, magnetic behavior, and cell response under static and dynamic magnetic stimulation. We hypothesized that spatial confinement of Fe₃O₄ could modify the local material microenvironment and magnetic interactions between NPs, thereby influencing the response of the hydrogel to an applied magnetic field. The results demonstrate that spatial organization of the magnetic phase is associated with distinct physicochemical and magnetic properties and, under dynamic magnetic stimulation, with enhanced fibroblast metabolic activity, spreading, and cytoskeletal organization. These findings identify magnetic NPs spatial organization as a relevant design parameter for developing magnetically responsive hydrogels and provide a basis for future studies directly linking localized magnetic actuation to cellular mechanotransduction.

2. Experimental section

2.1. Materials

Chitosan (molecular weight: 6 × 10⁵ - 8 × 10⁵ g mol⁻¹) was sourced from Acros Organics. Gelatin from porcine skin was obtained from Fluka Analytical. Glacial acetic acid (99+%) was purchased from Alfa Aesar. Fe₃O₄ NPs were acquired from Sigma-Aldrich. Sodium hydroxide (NaOH, pellets) and nitric acid (HNO₃, 65%) were supplied by PanReac AppliChem. For biological assays, the human foreskin fibroblast cell line HFF-1 (ATCC® SCRC-1041™, American Type Culture Collection) was used for all in vitro experiments. Bovine serum albumin (BSA) and Triton X-100 were provided by Sigma-Aldrich. Phosphate-buffered saline (PBS), glycine, Dulbecco's Modified Eagle Medium (DMEM) supplemented with 10% (v/v) fetal bovine serum (FBS), 1% (v/v) penicillin/streptomycin, and 1% (v/v) L-glutamine were acquired from Thermo Fisher Scientific. Live/Dead™ (calcein AM/ethidium homodimer-1), PrestoBlue™ HS Cell Viability Reagent, phalloidin-Alexa Fluor™ 546, 4′,6-Diamidino-2-Phenylindole (DAPI), and trypsin-EDTA (Invitrogen and Gibco brands) were also obtained from the same supplier. All products were used as received without further purification. All solutions were prepared using distilled water.

2.2. Synthesis of hydrogels

Three different hydrogel systems were fabricated: (i) Chitosan-gelatin hydrogels without NPs (blank), (ii) Chitosan-gelatin hydrogels containing homogeneously dispersed Fe₃O₄ NPs, and (iii) Chitosan-gelatin hydrogels containing spatially patterned Fe₃O₄ NPs. Hereafter, the hydrogels will be denoted as Chit/Gel, Chit/Gel/Fe₃O₄(D), and Chit/Gel/Fe₃O₄(P), respectively. The fabrication procedure is schematically illustrated in Fig. 1. For the preparation of Chit/Gel hydrogels, 0.1 g of chitosan was dissolved in 5 mL of a 0.5% (v/v) acetic acid (CH₃COOH) solution under magnetic stirring (250 rpm) at 45 °C for 45 min. In parallel, 0.135 g of gelatin was dissolved in 5 mL of distilled water under identical conditions. Both polymer solutions were then mixed and further stirred at 45 °C and 250 rpm for 45 min until a homogeneous polymeric solution was obtained. Subsequently, 5 mL of distilled water were added, and the resulting mixture was degassed using a speed mixer (3500 rpm, 2 min) to remove entrapped air bubbles. The solution was then cast into a Petri dish (10 cm diameter). To induce gelatin gelation, the samples were first maintained at 4 °C for 15 min and subsequently transferred to an incubator at 37 °C for 48 h until completely dried films were obtained. The dried hydrogels were then alkalinized by immersion in a 0.25 M NaOH solution under orbital agitation (50 rpm) for 30 min to promote chitosan physical crosslinking. Finally, the samples were thoroughly washed with distilled water until a neutral pH (∼7) was reached. Chit/Gel/Fe₃O₄(D) hydrogels were prepared following the same procedure, except that Fe₃O₄ NPs (100 mg) were first dispersed in the additional 5 mL of distilled water by sonication for 5 min prior to incorporation into the homogeneous chitosan-gelatin solution. For the fabrication of Chit/Gel/Fe₃O₄(P) hydrogels, Fe₃O₄ NPs (100 mg) were similarly dispersed in 5 mL of distilled water by sonication for 5 min and subsequently casted into a Petri dish positioned over a magnetic template composed of 90 cylindrical magnets (Diameter: 2 mm, thickness: 1 mm, magnetic field strength: 5 mT, Supermagnete) (Fig. S1). The magnetic template induced the spatial organization of the Fe₃O₄ NPs by attracting them toward the positions of the magnets, thereby generating a predefined particle pattern. After that, the homogeneous chitosan-gelatin solution was carefully poured onto the NPs-patterned substrate. As a result, two distinct regions were generated within the hydrogel: (i) a matrix region, corresponding to NPs-free areas, and (ii) a particle region, enriched with Fe₃O₄ NPs. The final Fe₃O₄ NP content in both Chit/Gel/Fe₃O₄(D) and Chit/Gel/Fe₃O₄(P) hydrogels was 30 wt% with respect to the total solid content.

Fig. 1.

Schematic representation of the fabrication of magnetic hydrogels with different NPs distributions. Three hydrogel configurations are generated: (i) Chitosan-gelatin hydrogels without NPs (blank) (Chit/Gel), (ii) Chitosan-gelatin hydrogels containing homogeneously dispersed Fe₃O₄ NPs (Chit/Gel/Fe₃O₄(D)), and (iii) Chitosan-gelatin hydrogels containing spatially patterned Fe₃O₄ NPs using a magnetic template (Chit/Gel/Fe₃O₄(P)).

2.3. Characterization

The morphology and microstructure of the hydrogels were analyzed by field-emission scanning electron microscopy (FE-SEM) using a Zeiss Neon40 Crossbeam™ workstation. Prior to imaging, samples were lyophilized and mounted onto aluminum pin stubs to ensure stability during analysis. A thin iridium coating (thickness ≈ 6.3 nm) was applied using a Leica EM ACE60 sputter coater to improve electrical conductivity and image quality. Imaging was performed at an accelerating voltage of 20 kV and a working distance of 7.5 mm, using different magnifications to resolve both surface and cross-sectional structural features of the hydrogel matrix. Elemental composition and NPs distribution were evaluated by energy-dispersive X-ray spectroscopy (EDX) using the same instrument. Elemental mapping was performed to analyze the spatial distribution of Fe and C within the hydrogels, allowing identification of NPs-rich and polymer-rich regions. Measurements were acquired with a resolution of 1024 pixels, a process time of 5, and a pixel dwell time of 5 μs.

The Fe₃O₄ content within the hydrogels was gravimetrically evaluated after extraction of the NPs from the hydrogel. To do so, chitosan was dissolved in acetic acid while gelatin was fused by heating the sample to 50 °C. At this point, Fe₃O₄ NPs were magnetically collected, thoroughly washed to remove residual polymer chains, dried and weighted using an analytical balance (Shimadzu, Precision 0.1 mg). The contents of Fe₃O₄ NPs in the Chit/Gel/Fe₃O₄(D) and Chit/Gel/Fe₃O₄(P) hydrogels were approximately the same as the content used to prepare the magnetic hydrogels (100 mg).

Surface topography and pattern definition of hydrated hydrogels were analyzed by optical profilometry using a confocal KLA ZETA-20 system (KLA-Tencor). Measurements were performed with a 5× objective and a vertical resolution of 1.5 μm per step. Optical micrographs, three-dimensional surface reconstructions, and height profiles were obtained using Zeta 3D software.

Fourier-transform infrared spectroscopy (FTIR), Raman spectroscopy, and X-ray diffraction (XRD) analyses were performed to evaluate the structure of the hydrogels and the Fe₃O₄ NPs. FTIR spectra were acquired using a JASCO FT/IR-6X spectrometer equipped with an ATR PRO ONE X single-reflection diamond ATR accessory. Measurements were performed at room temperature under N₂ purge, with a resolution of 2 cm⁻¹ and 64 accumulations in the range of 400-4000 cm⁻¹. A background spectrum was collected prior to each measurement to minimize atmospheric interference. Spectra were recorded using Spectra Manager software (version 2.5). Raman spectra were obtained using a Renishaw inVia Raman microscope (Qontor) equipped with a 785 nm laser and a 1200 l mm⁻¹ grating, coupled to a Renishaw Centrus 2957T2 detector. Spectra were acquired in the range of 200-2000 cm⁻¹ using WiRE software (version 5.6). For hydrogel samples, measurements were performed using a 20× objective, with an exposure time of 10 s, laser power of 0.5%, and 16 accumulations. For Fe₃O₄ NPs, a 50× objective was used with an exposure time of 10 s, laser power of 0.05%, and 16 accumulations. XRD patterns were recorded using a Bruker D8 Advance diffractometer operating in Bragg-Brentano geometry with Cu Kα radiation (λ = 1.5406 Å), at 40 kV and 40 mA. Data were collected over a 2θ range of 25-65° with a step size of 0.02°. The crystallite size (D) was estimated using the Scherrer equation (Eq. (1)):

$$D=\frac{k\mathrm{\lambda }}{\mathrm{\beta }\cos \mathrm{\theta }}$$ (1)

where k is the Scherrer constant (0.94), λ is the X-ray wavelength, β is the full width at half maximum (FWHM) of the diffraction peaks obtained by Gaussian fitting, and θ is the Bragg angle.

Mechanical properties of the hydrogels were evaluated by tensile testing under hydrated conditions using a Discovery HR-2 rheometer (TA Instruments) equipped with clamps for rectangular solid samples. Measurements were performed at room temperature using TRIOS software (version 9.54). A constant displacement rate of 10 μm s⁻¹ was applied during the tests. Prior to testing, samples were cut into rectangular specimens (≈19 mm in length and ≈12 mm in width) and subsequently immersed in PBS for 24 h to ensure full hydration. To improve grip and avoid slippage during testing, P240 sanding paper was used at the sample-clamp interface. All measurements were performed in triplicate. Mechanical parameters were obtained from stress-strain curves, including Young's modulus, tensile strength, and elongation at break. Young's modulus was calculated from the slope of the linear region of the curve (10-20% strain).

The swelling behavior of the hydrogels was determined by measuring the equilibrium swelling ratio (SR) using a gravimetric method. Samples (Ø ≈ 16 mm) were first freeze-dried overnight and weighed to obtain the dry weight (W_d). The dried samples were then immersed in distilled water, PBS, and DMEM for 24 h to reach equilibrium swelling. After incubation, samples were gently blotted with filter paper to remove excess surface liquid and immediately weighed to obtain the swollen weight (W_w). The SR was calculated according to Equation (2):

$$SR(\%)=\frac{\left(w_w-w_d\right)}{w_d}x100$$ (2)

All measurements were performed in triplicate.

Magnetic properties of the hydrogels were evaluated using a superconducting quantum interference device (SQUID) magnetometer (Quantum Design MPMS XL) under a helium atmosphere at room temperature. Field-dependent magnetization (M-H) curves were recorded at 300 K with a maximum applied field of 20 kOe. Magnetization values were expressed in emu g⁻¹, where g corresponds to the mass of magnetite within the hydrogel, allowing direct comparison between samples. The magnetic field was expressed in Oe.

Magnetic responsiveness was assessed by measuring the distance at which the hydrogels responded to an external magnetic field generated by neodymium-iron-boron (NdFeB) permanent magnets (diameter 20 mm, thickness 5 mm, Supermagnete) with field strengths of 280 (1 magnet) and 420 mT (3 stacked magnets). Fully hydrated hydrogel samples (Ø = 16.5 mm) were immersed in a water-filled container, while the magnet was fixed to one arm of a caliper and positioned above the sample. The magnet was gradually lowered until a visible response of the hydrogel was observed, defined as its upward displacement toward the magnet. The distance between the magnet and the sample at this point was recorded. Measurements were performed under both static and dynamic conditions. In static conditions, the system remained at rest (0 rpm), whereas in dynamic conditions the container was placed on an orbital shaker operating at 50 rpm to simulate fluid movement.

Cell viability, proliferation, and morphology were evaluated using human foreskin fibroblast (HFF-1) cells. Cells were cultured in DMEM supplemented with 10% (v/v) FBS, 1% (v/v) penicillin/streptomycin, and 1% (v/v) L-glutamine, and maintained at 37 °C in a humidified atmosphere with 5% CO₂. Upon reaching approximately 90% confluence, cells were detached using trypsin-EDTA, centrifuged, and resuspended in fresh culture medium. Prior to cell seeding, hydrogel samples were sterilized by immersion in 70% ethanol for 30 min, followed by washing with PBS and preconditioning in DMEM for 30 min. A cell suspension containing 2000 cells per sample was seeded onto the hydrogels.

Samples were subjected to three different conditions: (i) without magnetic stimulation, (ii) static magnetic stimulation (150 mT, 0 Hz, 30 min day⁻¹), and (iii) dynamic magnetic stimulation (150 mT, 10 Hz, 30 min day⁻¹). The magnetic stimulation system consisted of a rotating Teflon disk embedding sixteen cylindrical NdFeB permanent magnets (diameter 10 mm, thickness 5 mm; Supermagnete) arranged symmetrically. The culture plate was positioned above the rotating disk at a fixed distance of 2 mm, generating a magnetic field of 150 mT at the sample location. The magnetic field intensity was measured using a teslameter (FM 302, Projekt Elektronik GmbH), and the rotation frequency was controlled using a digital tachometer (DT-2234C, Topway).

Cell viability was assessed at days 1, 3, 7, and 14 using a Live/Dead fluorescence staining kit (calcein AM and ethidium homodimer-1) (Invitrogen). At each time point, the culture medium was removed and samples were incubated with 500 μL of staining solution (prepared according to the manufacturer's protocol) for 30 min at room temperature in the dark. Samples were then washed with PBS and imaged using a confocal microscope (Carl Zeiss LSM 800) with Zen software (version 2.6, Blue edition). A 5× objective was used to evaluate cell distribution, and a 10× objective to assess cell spreading.

Cell proliferation was quantified using the PrestoBlue™ assay (Invitrogen) at days 1, 3, 7, 10, and 14. Prior to measurement, samples were transferred to a new well plate to avoid interference from residual cells. Hydrogels without cells were used as blank controls. Samples were incubated with PrestoBlue reagent diluted in fresh DMEM (following the manufacturer's protocol) for 1 h at 37 °C. Subsequently, 100 μL of the supernatant was transferred to a black 96-well plate, and fluorescence was measured using a Tecan Infinite 200Pro microplate reader operated with Tecan i-control software (version 2.0.10.0) at excitation and emission wavelengths of 560 and 590 nm, respectively. Cell number was determined using a calibration curve obtained by correlating fluorescence intensity with known cell numbers. All measurements were performed in triplicate.

Immunofluorescence staining was performed to evaluate cell adhesion and morphology at days 1, 3, 7, and 14. Cells were permeabilized with 0.05% (w/v) Triton X-100 in PBS for 20 min, followed by blocking with 1% (w/v) BSA in PBS for 1 h. F-actin filaments were stained with phalloidin-Alexa Fluor™ 546 (1:400, in 0.05% Triton X-100) for 1 h, and nuclei were stained with DAPI (1:1000, in PBS–glycine 20 mM) for 2 min. All staining steps were performed in the dark, with intermediate washing steps using PBS-glycine (three times, 5 min each). Samples were imaged using a confocal microscope (Carl Zeiss LSM 800) with a 25× objective. Quantitative image analysis (nuclear projected area, nuclear aspect ratio, nuclear circularity, nuclear orientation) was performed using ImageJ software (version 1.54g). For each experimental condition, 3-5 representative confocal images were analyzed, ensuring a sufficient number of cells for the assessment and obtaining representative measurements of the cell population. All biological characterizations were performed in order to isolate the effects of NPs distribution and magnetic stimulation under controlled static in vitro conditions. Introducing fluid flow would add another independent mechanical variable and would make attribution of the observed effects more difficult in this initial study.

Iron released was quantified by inductively coupled plasma mass spectrometry (ICP-MS; Agilent 7800, Agilent Technologies) after incubation in PBS for 14 days. Aliquots of the solution were collected, diluted, and acidified with 2% HNO₃. Before measurement, the samples were passed through 0.22 μm syringe filters. Instrument calibration was performed using certified single-element iron standards with a concentration of 1000 mg/L, prepared with ultrapure water. The ICP-MS was operated at a radiofrequency plasma power of 1550 W. Argon flow rates were maintained at 15 L/min for the coolant gas and 1 L/min for the auxiliary gas. Sample introduction was carried out using a concentric nebulizer operated at a gas flow rate of 1.05 L/min, while the peristaltic pump was set to 6 rpm. All measurements were performed in triplicate.

Statistical analysis was performed using GraphPad Prism (version 9.3.1). Data normality was assessed using the Shapiro-Wilk test. Statistical comparisons were conducted using one-way or two-way analysis of variance (ANOVA), as appropriate, followed by Tukey's multiple comparisons test. A significance level of α = 0.05 was applied. Data are presented as mean ± standard deviation.

3. Results and discussion

3.1. Design and concept of magnetically responsive hydrogels

The design of the magnetic hydrogels was conceived to investigate the combined influence of Fe₃O₄ NPs spatial distribution and magnetic field stimulation on the physicochemical, magnetic and mechanical properties of the hydrogels and the resulting cellular response. To this end, three hydrogel configurations were developed: (i) Chitosan/Gelatin hydrogels without Fe₃O₄ NPs (blank) (Chit/Gel), (ii) Chitosan/Gelatin hydrogels containing homogeneously dispersed Fe₃O₄ NPs (Chit/Gel/Fe₃O₄(D)), and (iii) Chitosan/Gelatin hydrogels containing spatially patterned Fe₃O₄ NPs (Chit/Gel/Fe₃O₄(P)). This design strategy enables deciphering the effect of NPs spatial distribution on the nanocomposite hydrogel properties and the effect of the magnetic field on the fibroblasts response. While dispersed hydrogels provide a continuous and isotropic distribution of magnetic domains throughout the matrix, patterned hydrogels introduce spatially defined NPs-rich and NPs-free regions, thereby generating compositional anisotropy and local heterogeneity within the material. Such structural organization is expected to produce localized variations in the properties (e.g., magnetic susceptibility, stiffness) ultimately resulting in a more complex and biomimetic microenvironment. From a magnetic perspective, patterned hydrogels are anticipated to exhibit enhanced local magnetic gradients compared to homogeneously dispersed systems. In uniformly dispersed hydrogels, magnetic interactions are spatially averaged across the entire matrix, which may reduce the magnitude of localized magnetic forces and mechanical transduction. In contrast, the confinement of Fe₃O₄ NPs into discrete domains can promote cooperative magnetic interactions and localized magnetic amplification effects under external stimulation. This spatial concentration of magnetic material is expected to increase local magnetization and improve the efficiency of magnetic actuation, particularly under dynamic magnetic fields, while maintaining NPs-free regions with higher hydrogel flexibility and permeability. Moreover, the presence of NPs-enriched domains may generate localized mechanical reinforcement within the hydrogel, leading to microscale stiffness gradients between particle-rich and particle-free regions. Such gradients are highly relevant in cell-material interactions, as cells are known to respond to local mechanical heterogeneity through mechanotransduction pathways. Patterned hydrogels may therefore provide spatially controlled mechanical cues capable of directing cell adhesion, spreading, migration, and cytoskeletal organization more effectively than homogeneous systems, where the mechanical properties remain comparatively uniform. The patterned organization may also improve magnetic-to-mechanical energy conversion under dynamic stimulation. Under time-dependent magnetic fields, concentrated magnetic domains can undergo amplified local deformation, displacement, or vibrational motion due to stronger magnetic torque and dipole-dipole interactions between neighbouring NPs. These localized deformations can generate microscale mechanical stimuli within the surrounding polymeric matrix, potentially enhancing cell stimulation through cyclic strain or dynamic topographical changes. In contrast, uniformly dispersed NPs are expected to produce more attenuated and spatially diluted mechanical responses. Overall, this approach enables a systematic investigation of how NPs spatial organization and external magnetic actuation interact to modulate the magnetic, mechanical, structural, and morphological properties of the hydrogels, ultimately influencing cellular behaviour and functionality.

3.2. Morphological, topographical and structural characterization

The morphology and spatial distribution of Fe₃O₄ NPs within the hydrogels were investigated by FE-SEM coupled with EDX analysis. Surface FE-SEM images (Fig. 2) reveal distinct morphological features depending on NPs incorporation and distribution. Chit/Gel hydrogels exhibit a relatively smooth and homogeneous surface, consistent with a polymeric matrix without inorganic inclusions. In contrast, Chit/Gel/Fe₃O₄(D) hydrogels display a rougher surface with uniformly distributed granular structures, corresponding to the Fe₃O₄ NPs embedded within the matrix. Meanwhile, patterned Chit/Gel/Fe₃O₄(P) hydrogels show a clear spatial differentiation between two regions: a NPs-free matrix region and a NPs-enriched region. The matrix region presents a smooth surface morphology similar to the blank hydrogel, whereas the NPs-enriched region exhibits a highly textured surface with dense NPs accumulation. These observations are further confirmed by EDX mapping, where Fe signals are homogeneously distributed in Chit/Gel/Fe₃O₄(D) samples, while in patterned hydrogels they are localized exclusively within the defined NPs regions. Carbon mapping remains uniformly distributed in Chit/Gel and Chit/Gel/Fe₃O₄(D) hydrogels confirming the continuous polymeric network in those hydrogels. However, carbon signal is higher in the matrix-region of the patterned hydrogels, as expected, but it is also detected in the NPs-enriched region indicating the presence of the biopolymers also in those regions. Cross-sectional FE-SEM images (Fig. 3) provide further insight into the internal structure of the hydrogels. Blank samples display a compact and uniform structure across the thickness, while Chit/Gel/Fe₃O₄(D) hydrogels show a rougher surface and a homogeneous distribution of Fe₃O₄ NPs across the hydrogel section. In contrast, Chit/Gel/Fe₃O₄(P) hydrogels exhibit a well-defined interface between the matrix and the NPs regions, confirming that the spatial organization induced during fabrication is preserved throughout the bulk material. EDX mapping of cross-sections further validates this distribution, with Fe signals confined to particle-rich domains and absent in the surrounding matrix.

Fig. 2.

FE-SEM images showing the surface structure of Chit/Gel, Chit/Gel/Fe₃O₄(D), and Chit/Gel/Fe₃O₄(P) hydrogels. EDX elemental maps display the spatial distribution of Fe and C.

Fig. 3.

FE-SEM images of cross-sections for Chit/Gel, Chit/Gel/Fe₃O₄(D), and Chit/Gel/Fe₃O₄(P) hydrogels. EDX elemental maps display the spatial distribution of Fe and C.

Surface topography and pattern definition were further evaluated by optical profilometry (Fig. 4). Optical micrographs confirm the presence of distinct surface features associated with NPs distribution. While Chit/Gel and Chit/Gel/Fe₃O₄(D) hydrogels show uniform surfaces, Chit/Gel/Fe₃O₄(P) hydrogels display pronounced topographical variations due to the presence of the NPs-enriched regions. 3D surface reconstructions reveal a higher height variation in the patterned hydrogels compared to the other hydrogels. Quantitative height profiles support these observations, showing that Chit/Gel and Chit/Gel/Fe₃O₄(D) hydrogels present significant lower height differences (41.93 ± 18.94 μm and 84.88 ± 17.62 μm, respectively) than Chit/Gel/Fe₃O₄(P) hydrogels (327.08 ± 31.88 μm). These observations were further quantified by evaluating the surface roughness parameters R_a and R_q for the different hydrogel configurations (Table 1). R_a represents the arithmetic average of the absolute deviations of the surface profile from the mean line, whereas R_q corresponds to the root mean square of these deviations. Consistent with the topographical differences observed in the profilometric profiles, the Chit/Gel exhibited the lowest roughness values, whereas Chit/Gel/Fe₃O₄(D) showed significantly higher R_a and R_q values. Within the patterned hydrogel, the matrix region remained relatively smooth and did not differ significantly from the Chit/Gel, and the NPs-enriched region exhibited significantly higher roughness. These results support the successful generation of spatially organized topographical features.

Fig. 4.

Surface topography and height analysis of hydrogels. Optical micrographs, 3D surface reconstructions, and height profiles of blank, dispersed, and patterned hydrogels. Different letters (a,b) indicate statistically significant differences (p < 0.05).

Table 1.

Surface roughness parameters (R_a, R_q) of the different hydrogels. Different letters within each parameter indicate statistically significant differences between groups (p < 0.05).

Hydrogel	Ra (μm)	Rq (μm)
Chit/Gel	5.46 ± 2.14 (a)	7.62 ± 2.34 (a)
Chit/Gel/Fe3O4(D)	13.61 ± 1.44 (b)	18.43 ± 2.69 (b)
Chit/Gel/Fe3O4(P) (Matrix region)	6.08 ± 3.23 (a)	9.02 ± 3.24 (a)
Chit/Gel/Fe3O4(P) (NPs-enriched region)	12.60 ± 2.71 (b)	17.29 ± 3.66 (b)

The structural properties of the Fe₃O₄ NPs (powder) and (nano)composite hydrogels were investigated by FTIR, Raman spectroscopy, and XRD (Fig. 5). FTIR spectra (Fig. 5A) confirm the successful formation of the chitosan-gelatin hydrogel network, with characteristic bands associated with both biopolymers. The broad band in the range of 3200-3400 cm⁻¹ corresponds to O–H and N–H stretching vibrations, indicating extensive hydrogen bonding within the polymeric matrix. The presence of amide I (∼1630-1640 cm⁻¹), amide II (∼1540-1550 cm⁻¹), and amide III (∼1330-1340 cm⁻¹) bands further confirms the incorporation of gelatin. Additionally, bands corresponding to C–O–C stretching of the polysaccharide backbone (∼1020-1030 cm⁻¹) and C–O stretching of secondary alcohol groups (∼1060-1080 cm⁻¹) are attributed to the chitosan structure. The characteristic spectra of the individual components (chitosan and gelatin) are shown in Fig. S2. In Fe₃O₄ NPs-containing hydrogels, the appearance of a band around ∼530 cm⁻¹ is assigned to Fe–O stretching vibrations, confirming the presence of magnetite NPs within the polymeric network. Slight shifts in characteristic bands with respect to raw materials (chitosan, gelatin, Fe₃O₄ NPs) were observed, suggesting interactions between the biopolymers (blank) and between the Fe₃O₄ NPs and the polymer chains (nanocomposite hydrogels), likely mediated by hydrogen bonding or electrostatic interactions.

Fig. 5.

Structural characterization of the magnetite powder, Chit/Gel, Chit/Gel/Fe₃O₄(D) and Chit/Gel/Fe₃O₄(P) hydrogels. Both matrix and NPs regions were analyzed in Chit/Gel/Fe₃O₄(P) hydrogels. (A) FTIR spectra. (B) Raman spectra. (C) XRD patterns.

Raman spectra further support these findings. The characteristic spectra of the individual components (chitosan and gelatin) are shown in Fig. S3 for comparison. Bands located at ∼770-780 cm⁻¹ and ∼1320-1335 cm⁻¹ are attributed to C–H bending/skeletal vibrations and CH₂/CH₃ deformation (amide III region), respectively, confirming the contribution of the biopolymers (Fig. 5B). On the other hand, an additional peak around ∼674 cm⁻¹ is also observed in Chit/Gel/Fe₃O₄(D) and Chit/Gel/Fe₃O₄(P) hydrogels (in the NPs region, not in the matrix region). This peak corresponds to the A1g mode of magnetite confirming its successful incorporation. The presence and intensity of this band confirm the incorporation of Fe₃O₄ NPs. Minor shifts in Raman peak positions were also observed compared to the raw materials. In particular, the magnetite-related band appears at ∼674 cm⁻¹ for pristine NPs powder and shows slight shifts in Chit/Gel/Fe₃O₄(D) and Chit/Gel/Fe₃O₄(P) (∼672-673 cm⁻¹), which may indicate interactions between Fe₃O₄ NPs and the surrounding polymer network.

XRD patterns (Fig. 5C) exhibit the characteristic diffraction peaks of magnetite, corresponding to the (220), (311), (400), (422), (511), and (440) planes, confirming the crystalline structure of Fe₃O₄ NPs: a face-centered cubic lattice with an inverse spinel structure (PDF#JCPDS 19-0629). These peaks are preserved in both Chit/Gel/Fe₃O₄(D) and Chit/Gel/Fe₃O₄(P) hydrogels, indicating that the crystalline structure of the NPs remains intact after incorporation into the polymer matrix. Detailed peak parameters obtained from Gaussian fitting (Fig. S4) are summarized in Table S1. Crystallite size (D), estimated using the Scherrer equation (Eq. (1)), decreased from 50.63 ± 7.09 nm for magnetite powder to 37.60 ± 3.05 nm and 39.37 ± 4.76 nm for Chit/Gel/Fe₃O₄(D) and Chit/Gel/Fe₃O₄(P) hydrogels, respectively. This reduction suggests a confinement effect and/or modifications in NPs aggregation induced by incorporation into the hydrogel matrix.

3.3. Mechanical and swelling properties

The tensile properties of the hydrogels were obtained from stress-strain curves (Fig. S5) to evaluate the effect of Fe₃O₄ NPs incorporation and distribution. As shown in Table 2, the Young's modulus increased significantly from 4.1 ± 0.5 MPa for Chit/Gel hydrogels to 9.8 ± 0.8 MPa and 14.4 ± 0.4 MPa for Chit/Gel/Fe₃O₄(P) and Chit/Gel/Fe₃O₄(D) hydrogels, respectively. A similar trend was observed for the tensile strength, indicating that the presence of Fe₃O₄ NPs reinforces the polymeric network. In contrast, elongation at break decreased from 36.5 ± 0.5% (Chit/Gel) to 31.2 ± 0.8% (Chit/Gel/Fe₃O₄(P)) and 27.2 ± 0.5% (Chit/Gel/Fe₃O₄(D)), reflecting a reduction in flexibility due to NPs incorporation. The higher stiffness observed in dispersed hydrogels can be attributed to the homogeneous distribution of Fe₃O₄ NPs, which promotes more efficient load transfer throughout the matrix. In patterned hydrogels, the presence of distinct NPs-rich and NPs-free regions results in intermediate mechanical properties, consistent with the heterogeneous structure observed in SEM analysis. These findings highlight that NPs spatial distribution directly influences the macroscopic mechanical behavior of the hydrogels.

Table 2.

Mechanical properties of Chit/Gel, Chit/Gel/Fe₃O₄(P), and Chit/Gel/Fe₃O₄(D) hydrogels obtained from tensile testing under hydrated conditions. Different letters indicate statistically significant differences (p < 0.05) between hydrogel types within each parameter.

Mechanical properties	Chit/Gel	Chit/Gel/Fe3O4(P)	Chit/Gel/Fe3O4(D)
Young's Modulus (MPa)	4.1 ± 0.5 (a)	9.8 ± 0.8 (b)	14.4 ± 0.4 (c)
Tensile stress (MPa)	2.4 ± 0.3 (a)	3.2 ± 0.1 (b)	3.8 ± 0.2 (c)
Elongation at Break (%)	36.5 ± 0.5 (a)	31.2 ± 0.8 (b)	27.2 ± 0.5 (c)

The swelling behavior of the samples was investigated in distilled water, PBS, and DMEM to simulate different physiological environments (Fig. 6). The use of these three media allows evaluation of the influence of ionic strength and biomolecular content on hydrogel hydration. In all cases, Chit/Gel hydrogel exhibited the highest SR, followed by Chit/Gel/Fe₃O₄(P) and Chit/Gel/Fe₃O₄(D) hydrogels. This trend can be explained by the increasing presence of Fe₃O₄ NPs, which reduces the free volume within the polymeric network and limits water uptake. Additionally, the reduced chain mobility in NPs-loaded hydrogels contributes to the lower swelling capacity. The observed swelling behavior is consistent with the mechanical results, where increased stiffness correlates with reduced swelling ability. It is worth mentioning that the presence of ions in PBS and DMEM further affects the swelling due to osmotic balance and polymer-ion interactions, leading to slightly lower SR compared to distilled water.

Fig. 6.

Swelling ratios of Chit/Gel, Chit/Gel/Fe₃O₄(D), and Chit/Gel/Fe₃O₄(P) hydrogels in distilled water, PBS, and DMEM. Different letters indicate statistically significant differences within the same hydrogel type across different media, while different numbers denote significant differences between hydrogel types under the same medium (p < 0.05).

3.4. Magnetic properties and magnetic responsiveness

The magnetic properties of the hydrogels were investigated by SQUID magnetometry through field-dependent magnetization (M-H) measurements (Fig. 7A). As expected, the Chit/Gel hydrogels exhibited negligible magnetization, confirming the absence of magnetic components within the polymeric network. In contrast, both Chit/Gel/Fe₃O₄(D) and Chit/Gel/Fe₃O₄(P) hydrogels displayed typical soft ferromagnetic behavior, characterized by narrow hysteresis loops, low coercivity (H_c), and rapid saturation of the magnetization (M_s) under the applied magnetic field. Notably, the Chit/Gel/Fe₃O₄(P) hydrogels exhibited a higher M_s than the Chit/Gel/Fe₃O₄(D) counterparts. This behaviour can be attributed to the spatial confinement of Fe₃O₄ NPs within localized regions of the patterned hydrogels, which reduces the average interparticle distance and thereby promotes stronger magnetic dipole-dipole interactions. It should be noted that the magnetization values were normalized to the magnetite content of the hydrogels, as determined gravimetrically. Thus, the enhanced magnetization cannot be attributed to differences in Fe₃O₄ loading. Instead, the spatially confined NPs assemblies may behave as larger effective magnetic domains, favouring the collective alignment of magnetic moments under an applied magnetic field and, consequently, increasing the M_s. This interpretation is consistent with previous studies reporting an increase in M_s with increasing particle size or effective magnetic domain size [33]. Despite the differences observed between the two nanocomposite hydrogels, their M_s values remained close to the theoretical bulk saturation magnetization reported for magnetite (92-100 emu g⁻¹) [34]. These results suggest that incorporation of Fe₃O₄ NPs into the polymeric matrix largely preserves their intrinsic magnetic properties, while their macroscopic magnetic response can be modulated through control of the spatial organization of the NPs within the hydrogel. In agreement with this interpretation, the magnetic susceptibility (the initial slope of the M-H curves) was also higher for the Chit/Gel/Fe₃O₄(P) hydrogels, further supporting the presence of larger effective ferromagnetic domains in the patterned samples. A closer inspection of the low-field region of the hysteresis curves (Fig. 7B) revealed low H_c values of 165.4 Oe and 110.6 Oe for Chit/Gel/Fe₃O₄(D) and Chit/Gel/Fe₃O₄(P) hydrogels, respectively. These low H_c values confirm the soft ferromagnetic character of the nanocomposite hydrogels and their ability to respond efficiently to external magnetic fields. The differences in coercivity between both systems can again be associated with variations in effective magnetic domain size arising from the distinct NPs spatial distributions, in agreement with previous studies [33].

Fig. 7.

Magnetic characterization and responsiveness of the hydrogels. (A) Magnetization versus magnetic field (M-H) curves obtained by SQUID magnetometry. (B) Enlarged view of the central region of the hysteresis loops, showing a narrow hysteresis with low coercivity and minimal remanence. (C) Magnetic responsiveness evaluated as the attraction distance of the hydrogels toward NdFeB permanent magnets (150, 280 and 420 mT) under static (0 rpm) and dynamic (50 rpm) conditions. Different letters indicate statistically significant differences within the same hydrogel type across magnetic field strengths, while different numbers denote significant differences between hydrogel types under the same magnetic field (p < 0.05).

To further assess the functional magnetic response of the hydrogels, magnetic responsiveness was evaluated by measuring the distance at which the samples were attracted toward an external magnetic field (Fig. S6). As shown in Fig. 7C, both Chit/Gel/Fe₃O₄(D) and Chit/Gel/Fe₃O₄(P) hydrogels exhibited clear magnetic responsiveness under all tested conditions. Under static conditions, patterned hydrogels exhibited significantly greater attraction distances compared to dispersed hydrogels at the different magnetic field strengths evaluated (150, 280 and 420 mT), indicating enhanced responsiveness. This behavior is again associated with the presence of spatially concentrated magnetic domains, which can generate stronger localized magnetic forces. This result agrees with the higher M_s and M_s observed in the hysteresis loops for Chit/Gel/Fe₃O₄(P) hydrogels. Under dynamic conditions, an overall increase in responsiveness was observed for both hydrogel types. Notably, patterned hydrogels again demonstrated the highest attraction distances which may be related to the improved magnetic properties. These results demonstrate that NPs distribution plays a key role in determining the magnetic performance of the hydrogels. In particular, the superior responsiveness of Chit/Gel/Fe₃O₄(P) hydrogels indicates that the spatial organization of NPs can enhance the ability of the material to undergo magnetically induced motion or deformation, which may be relevant for subsequent biological interactions.

3.5. In vitro cell response with and without magnetic stimulation

The effects of Fe₃O₄ NPs distribution and magnetic stimulation on HFF-1 fibroblast behavior were determined by assessing cell viability, proliferation, and morphology within the hydrogels. Cell proliferation in Chit/Gel, Chit/Gel/Fe₃O₄(D), and Chit/Gel/Fe₃O₄(P) hydrogels under non-stimulated conditions was first quantified using the PrestoBlue metabolic assay (Fig. 8). Cell number was estimated from fluorescence intensity using the calibration curve shown in Fig. S7. All hydrogel formulations exhibited a progressive increase in metabolic activity over time, confirming their cytocompatibility and ability to support sustained cell growth. During the initial culture period (days 1-3), no significant differences were detected among the different hydrogels, indicating that neither the incorporation of Fe₃O₄ NPs nor their spatial distribution adversely affected early cell attachment or short-term viability. However, from day 7 onwards, distinct proliferation profiles emerged depending on NPs organization within the hydrogel matrix. By day 10, Chit/Gel/Fe₃O₄(P) hydrogels displayed higher metabolic activity than both Chit/Gel/Fe₃O₄(D) and Chit/Gel hydrogels, and this trend became more pronounced at day 14. These results indicate that the spatial organization of magnetic NPs plays an important role in regulating metabolic activity and viable-cell density. The enhanced cellular response observed in the patterned hydrogels may be associated with local variations in mechanical properties or microscale topography generated by NPs confinement, as previously characterized, which could promote stronger cell-matrix interactions and facilitate cell spreading and proliferation.

Fig. 8.

Quantification of HFF-1 proliferation on blank, dispersed, and patterned hydrogels using PrestoBlue assay. Different letters indicate statistically significant differences within the same hydrogel type across different days, while different numbers denote significant differences between hydrogel types on the same day (p < 0.05).

The influence of magnetic stimulation was further evaluated under static and dynamic magnetic field conditions. Under static magnetic stimulation, cell proliferation was slightly reduced compared with the non-stimulated condition, particularly in the Chit/Gel/Fe₃O₄(D) hydrogels. This decrease suggests that continuous exposure to a static magnetic field may generate persistent mechanical constraints within the magnetic matrix, limiting cell proliferation. The effect was more pronounced in hydrogels with a homogeneous NPs distribution, likely because the uniformly dispersed magnetic NPs transmit mechanical stresses more homogeneously throughout the matrix. In contrast, dynamic magnetic stimulation produced the highest metabolic activity across all conditions, exceeding those obtained under both non-stimulated and static magnetic field exposure. Notably, Chit/Gel/Fe₃O₄(P) hydrogels exhibited the greatest increase in metabolic activity and viable-cell density suggesting a more efficient mechanotransductive response with the oscillating magnetic field. Indeed, Video S1 provides direct evidence that Chit/Gel/Fe₃O₄(D) and Chit/Gel/Fe₃O₄(P) hydrogels exhibit visibly different actuation behaviours when subjected to the same dynamic magnetic stimulation (150 mT, 10 kHz). This observation confirms that redistributing the magnetic nanomaterial into spatially confined NPs-rich domains alters the way in which magnetic forces are translated into hydrogel response. This behavior agrees with the higher magnetic susceptibility and M_s resulting in an enhanced magnetic actuation of the patterned hydrogels, which likely amplify local cyclic deformations under alternating magnetic fields, thereby promoting cellular activity and proliferation.

Live/Dead fluorescence staining was performed to further evaluate the cytocompatibility of the hydrogels and to investigate the influence of Fe₃O₄ NPs distribution and magnetic stimulation on cell viability and spatial organization. In the absence of magnetic stimulation (Fig. S8), fibroblasts remained predominantly viable throughout the 14-day culture period, as evidenced by the strong green fluorescence signal and the negligible presence of red-stained dead cells. Cell density progressively increased with culture time in all hydrogel formulations, in agreement with the metabolic activity results obtained from the PrestoBlue assay. No apparent cytotoxic effects associated with either Fe₃O₄ NP incorporation or their spatial organization were observed, confirming the excellent cytocompatibility of both the dispersed and patterned magnetic hydrogels. To further assess the stability of the incorporated NPs, iron release from the hydrogels was quantified by ICP-MS. Only trace amounts of iron were detected for both Chit/Gel/Fe₃O₄(D) (∼0.03 ppm) and Chit/Gel/Fe₃O₄(P) (∼0.04 ppm), with no substantial differences between the two configurations. These low iron concentrations indicate minimal Fe₃O₄ leaching and demonstrate the effective immobilization and retention of the NPs within the hydrogel matrix. Collectively, these findings indicate that spatial patterning of the magnetic NPs does not adversely affect cytocompatibility or compromise their retention within the hydrogel network. Under static magnetic stimulation (Fig. S9), cell viability also remained consistently high for all hydrogel types, demonstrating that exposure to a constant magnetic field does not induce detectable cytotoxic effects. Nevertheless, compared with the non-stimulated condition, a lower cell density was observed at all time points, particularly at days 7 and 14. This observation is consistent with the reduced metabolic activity previously shown and suggests that static magnetic stimulation may restrict cell expansion. Such behavior may arise from persistent magnetically induced mechanical constraints within the hydrogel network hindering proliferation. In contrast, dynamic magnetic stimulation (Fig. 9) produced the most pronounced effects on cellular behavior. Although cell viability remained high under all conditions, a substantial increase in cell density was observed over time, especially in the Chit/Gel/Fe₃O₄(P) hydrogels. By day 14, fibroblasts formed dense and highly interconnected cellular layers, indicating enhanced proliferation and colonization of the patterned hydrogel surface. In addition, cells cultured on these patterned hydrogels appeared more uniformly distributed and exhibited increased local organization compared with the dispersed systems. These results suggest that cyclic magnetic actuation promotes more active cell-matrix interactions and creates a mechanically dynamic microenvironment that stimulates cell proliferation and tissue-like organization.

Fig. 9.

Live/Dead fluorescence staining of fibroblast cells cultured on blank, dispersed, and patterned hydrogels at days 1, 3, 7, and 14 under dynamic magnetic stimulation.

To further investigate the effect of magnetic stimulation on cell morphology and cytoskeletal organization, fluorescence immunostaining of actin filaments and cell nuclei was performed (Fig. 10, Figs. S10 and S11). In the absence of magnetic stimulation (Fig. S10), fibroblasts progressively spread over time and developed interconnected cellular networks across all hydrogel formulations. However, the orientation of the cells remained largely random, and no evident large-scale cytoskeletal organization was observed. Cells displayed the typical elongated fibroblastic morphology, confirming favorable adhesion and biocompatibility with the hydrogel surfaces. Under static magnetic stimulation (Fig. S11), fibroblasts exhibited morphologies comparable to those observed in the non-stimulated samples, with elongated cell bodies and well-defined actin filaments. However, cell density was visibly lower, particularly at later culture times, consistent with the Live/Dead and proliferation analyses. Although individual cells remained well spread, the overall cytoskeletal organization did not appear enhanced under static magnetic fields. These findings suggest that constant magnetic stimulation is insufficient to promote active mechanotransductive responses and may instead impose continuous mechanical stresses within the hydrogel matrix that limit collective cell expansion. In Chit/Gel/Fe₃O₄(P) hydrogels, only a slight tendency toward local alignment was observed, without the development of a clearly defined preferential orientation. Dynamic magnetic stimulation (Fig. 10) induced changes in cell morphology and cytoskeletal arrangement, that were particularly evident in the Chit/Gel/Fe₃O₄(P) hydrogels. Fibroblasts cultured under these conditions exhibited more elongated morphologies, increased spreading, and enhanced actin filament organization compared with the non-stimulated and static conditions. Moreover, cells formed densely packed and highly interconnected structures by day 14. The actin cytoskeleton appeared more organized and continuous, suggesting enhanced cell-matrix interactions. The local organization and alignment of cells were notably improved in the patterned hydrogels subjected to dynamic stimulation, which is likely associated with cyclic magnetically induced deformations within the hydrogel matrix providing repetitive mechanical cues capable of activating mechanotransduction pathways. The higher magnetic susceptibility and magnetization of the Chit/Gel/Fe₃O₄(P) hydrogels amplify these local mechanical stimuli (Video S1), thereby promoting cell spreading and proliferation.

Fig. 10.

Representative fluorescence images of HFF-1 cells cultured on blank, dispersed, and patterned hydrogels at days 1, 3, 7, and 14 under dynamic magnetic stimulation.

Quantitative analysis of nuclear morphology provided additional support for the time-dependent changes observed in the fluorescence images (Figs. S12–S14). Across the different hydrogel formulations and stimulation conditions, nuclear projected area generally increased during culture, particularly at the later time points, consistent with progressive cell adhesion and spreading (Fig. S12). At the same time, the nuclei evolved toward a more elongated morphology, as indicated by an increase in nuclear aspect ratio together with a decrease in circularity (Figs. S13 and S14). These changes were more evident at later culture times and under dynamic stimulation, indicating progressive cytoskeletal tension and morphological polarization of the fibroblasts rather than a simple increase in cell number. Importantly, the morphological response was most evident in the magnetic hydrogels under dynamic stimulation, supporting the qualitative observation of more elongated and organized fibroblast morphologies in the corresponding phalloidin images. On the other hand, nuclear orientation analysis further demonstrated that the cellular response depended on both magnetic stimulation and the type of hydrogel (Fig. S15). As expected, the non-magnetic Chit/Gel hydrogel showed broadly similar orientation distributions under non-stimulated, static, and dynamic conditions, indicating that the external magnetic field alone did not influence the orientation of the fibroblasts. In Chit/Gel/Fe₃O₄(D), the application of magnetic stimulation produced only modest changes in the angular distribution of the nuclei, with a slight tendency toward a more preferential orientation compared with the non-stimulated condition. This limited response is consistent with the homogeneous distribution of the nanoparticles, which is expected to generate a more spatially uniform and less localized mechanical stimulus. In contrast, the patterned Chit/Gel/Fe₃O₄(P) hydrogel exhibited more pronounced changes in nuclear orientation upon magnetic stimulation, particularly under dynamic conditions. The effect was spatially dependent, with the NPs-enriched, matrix, and interface regions displaying distinct angular distributions. A higher orientation was observed in the matrix region of the patterned hydrogel. These differences suggest that cyclic actuation of the spatially confined magnetic domains generates localized mechanical cues capable of promoting cell reorientation in the matrix region. Overall, these results demonstrate that both the spatial organization of magnetic NPs and the mode of magnetic stimulation play a synergistic role in modulating fibroblast behaviour. While all hydrogel formulations supported high cell viability and cell growth, the patterned Chit/Gel/Fe₃O₄(P) magnetic hydrogels subjected to dynamic magnetic stimulation provided the most favorable microenvironment for enhancing cell proliferation, cytoskeletal organization, and cellular assembly. Future studies should focus on systematically tuning the geometrical parameters of the magnetic patterns, including magnetic domain size, inter-domain spacing, or shape, to establish relationships between NPs organization, materials properties, and cellular responses. Moreover, the performance of these materials should be evaluated using flow-based magnetic bioreactor systems that more closely reproduce the dynamic mechanical and fluidic environment encountered in native tissues. Finally, extending these studies to additional cell types, mechanotransduction markers, and tissue-specific assays would help evaluate the broader applicability of the spatially patterned magnetic hydrogels in the tissue engineering field.

4. Conclusions

In this work, three different chitosan-gelatin-based hydrogels with controlled NPs spatial organization (blank, homogeneous, patterned) were successfully developed to investigate the combined effect of material design and magnetic stimulation on physicochemical, magnetic and mechanical properties and cellular response. The results demonstrate that NPs organization plays a key role in defining the structural and functional properties of the hydrogels. Chit/Gel/Fe₃O₄(P) hydrogels exhibited clear spatial heterogeneity, leading to variations in topography, and mechanical properties, which in turn influenced swelling behavior. Magnetic characterization confirmed that both dispersed and patterned hydrogels preserve the ferromagnetic behavior of Fe₃O₄ NPs, with Chit/Gel/Fe₃O₄(P) hydrogels showing enhanced magnetic responsiveness compared to Chit/Gel/Fe₃O₄(D). This behavior highlights the relevance of spatially defined magnetic domains in amplifying magnetically induced effects at the macroscopic level. From a biological perspective, all hydrogels exhibited high cytocompatibility, supporting cell viability and proliferation under all tested conditions. However, significant differences emerged depending on both NPs distribution and magnetic stimulation. While static magnetic fields led to reduced cell proliferation, likely due to sustained mechanical constraints, dynamic magnetic stimulation promoted a marked increase in cell number, particularly in Chit/Gel/Fe₃O₄(P) hydrogels. This enhancement is attributed to time-dependent mechanical stimulation, where cyclic deformation of the matrix is converted into biological signals that promote cell proliferation. This behavior was accompanied by increased cell elongation and more defined cytoskeletal organization, with a preferential orientation. Altogether, these findings demonstrate that the combination of NPs spatial organization and dynamic magnetic stimulation provides an effective strategy to modulate hydrogel properties and regulate cellular response. Chit/Gel/Fe₃O₄(P) hydrogels under dynamic magnetic stimulation exhibited the most favorable biological performance, highlighting their potential as tunable platforms for advanced biomaterials applications. This approach establishes a foundation for the development of magnetically responsive systems capable of delivering controlled mechanical cues with potential relevance to tissue engineering and related applications.

CRediT authorship contribution statement

Angel Viteri: Writing – review & editing, Writing – original draft, Methodology, Investigation, Formal analysis, Data curation. Maria-Pau Ginebra: Writing – review & editing, Validation, Supervision, Project administration, Funding acquisition. Jose García-Torres: Writing – review & editing, Visualization, Validation, Supervision, Resources, Project administration, Methodology, Funding acquisition, Formal analysis, Data curation, Conceptualization.

Data statement

Data will be made available on request.

Ethics approval and consent to participate

The present study did not involve human participants, human tissues, human data, or live animals.

All biological experiments were performed exclusively using the commercially available HFF-1 fibroblasts cell line. Therefore, institutional ethical approval and informed consent to participate were not required in accordance with institutional and national regulations.

The authors confirm that all experimental procedures were conducted in accordance with relevant institutional guidelines and good laboratory practice.

Declaration of generative AI and AI-assisted technologies in the manuscript preparation process

During the preparation of this work, the authors used ChatGPT for English language revision. The authors reviewed and edited the output as needed and take full responsibility for the content of the published article.

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 are the Supplementary data to this article: