2.1 Electrical stimulation alters cortical spheroid expansion

To investigate tissue-level effects of electrical stimulation, we generated 3D spheroids from dissociated postnatal rat cortical cells cultured for seven days under ultra-low-attachment (ULA) conditions (Fig. 1A). These spheroids, comprising neurons, astrocytes, and microglia, were transferred to polyacrylamide (PA) gels and subjected to DC electric stimulation for 0, 15, or 60 min, followed by 72 h of live imaging to track migration dynamics (Fig. 1B and C). Field uniformity in the stimulation region was achieved through the chamber geometry and verified by COMSOL simulations. Agarose salt bridges and Steinberg’s solution were used to electrically couple the electrodes while minimizing electrochemical byproducts (e.g., pH shifts and gas formation) and standardizing ionic conductivity (Supplementary S1A–C).

Qualitative analysis of live-imaging data revealed a duration-dependent migratory response. Although initial spreading (0–10 h) was comparable across all groups, spheroids exposed to 60 min of ES showed a significant reduction in outward expansion beginning at 20 h (Fig. 1D). Kymograph analysis confirmed this inhibitory effect, showing a markedly flattened migration front in the 60-min group, in contrast to the sustained, steep trajectories observed in the control and 15-min cohorts (Fig. 1E).

Quantitative assessment of the relative spreading area (normalized to t = 0) revealed discrete growth kinetics (Fig. 1F). By 70 h, control spheroids showed a 5.1-fold increase in area, whereas the 60-min group expanded only 2.7-fold—representing a 47% reduction relative to controls. The 15-min group showed intermediate behavior (4.2-fold increase). Although its expansion was 35% greater than that of the 60-min group, it did not reach statistical significance compared to the control. These results suggest that short-duration ES is permissive for migration, whereas prolonged stimulation triggers a mechanical state that restricts collective cell migration.

Fig. 1

Prolonged electrical stimulation suppresses brain spheroid spreading. A Schematic illustration of brain spheroid generation from dissociated postnatal rat cortical cells and subsequent culture in ultra-low attachment (ULA) plates. B Experimental timeline showing spheroid culture, electrical stimulation for 0, 15, or 60 min on day 7, and subsequent live imaging for 72 h. C Schematic of the spreading assay after spheroid plating, illustrating cell migration and neurite extension from the attached spheroid. D Representative live images of spheroid spreading at 10, 30, 50, and 70 h after plating. Dashed outlines indicate the spreading boundary. E Representative kymographs showing temporal changes in the radial displacement of the spreading front. F Quantification of relative spreading area, expressed as the spreading area normalized to the spheroid body area. Prolonged electrical stimulation (60 min) reduced spheroid spreading compared with the control group, with significant differences observed from 20 h onward. Scale bar, 200 μm. Data are presented as mean ± SD. *P < 0.05 versus control at the indicated time point (one-way ANOVA performed separately at each analyzed time point). N = 8 spheroids total

2.2 Electrical stimulation modulates cortical spheroid cell migration

To determine whether the reduction in expansion was driven by altered cellular motility, we quantified migration dynamics using PIV (Fig. 2). Temporal analysis revealed distinct kinetic profiles: in the control and 15-min stimulation groups, migration speed increased rapidly and reached a stable plateau by 30 h. Conversely, the 60-min group showed a marked lag in acceleration, with a transient velocity dip at 20 h and a delayed peak that was not reached until 50 h (Fig. 2A and B).

Quantitatively, the maximum migration speeds in the control and 15-min groups were comparable (17.5 and 17.1 μm/h), whereas the 60-min group had a significantly lower maximum of 16 μm/h. Beyond velocity magnitude, the spatiotemporal organization of movement was fundamentally disrupted. Vector field and streamline analyses revealed coherent, radial trajectories in the control and 15-min groups, indicative of coordinated collective migration. In contrast, the 60-min condition exhibited disordered, curvilinear motion confined to the immediate spheroid perimeter, lacking the directional persistence observed in controls (Fig. 2C).

These findings suggest that prolonged stimulation induces a phenotypic shift that impairs outward polarization. Notably, this restriction in motility coincided with an increase in traction force (Supplementary Fig. S2), particularly at the periphery, where IBA1-positive microglia were enriched. This suggests that although these cells generate substantial mechanical force, it acts as a resistive anchoring force rather than a productive migratory drive, effectively pinning the spheroid to the substrate.

Fig. 2

Electrical stimulation alters cell migration dynamics. A Time-lapse PIV heatmaps showing migration velocity at indicated time points. B Maximum migration speed over time. While all groups showed initial acceleration, the 60-min stimulation group exhibited a prolonged time to reach peak velocity compared to controls. C Cumulative migration analysis showing summed velocity magnitude (left), vector fields (middle), and streamline trajectories (right). The 60-min condition displays reduced total movement and confined, curvilinear trajectories, contrasting with the extensive radial spreading observed in the control and 15-min groups. Scale bar: 200 μm. N = 8 spheroids

2.3 Electrical stimulation alters ECM composition and structural heterogeneity in cortical spheroids

Given the observed changes in cell motility and traction force, we examined whether electrical stimulation modulated the structural organization of the spheroid’s extracellular matrix (ECM). We evaluated the expression and spatial distribution of integrin β1 and fibronectin, two key proteins governing adhesion and mechanical integrity.

In control spheroids, Integrin β1 exhibited a characteristic annular (“donut-shaped”) distribution, with pronounced peripheral enrichment and a central void (Fig. 3A). Electrical stimulation dismantled this architecture; by 60 min, the central void was obliterated by the emergence of irregular, high-intensity protein subclusters in the spheroid core. This remodeling was associated with significant upregulation of Integrin β1 mRNA (2.45- and 2.49-fold in the 15- and 60-min groups, respectively; Fig. 3B). Protein-level responses followed a duration-dependent pattern: fluorescence intensity increased in both groups but reached statistical significance only in the 60-min group (1.14-fold).

In contrast, fibronectin underwent a distinct structural transition, shifting from an organized, fibrillar network in controls to a fragmented, cluster-like distribution after stimulation (Fig. 3C). Although fibronectin mRNA was significantly upregulated at both time points (~ 2.4-fold), protein expression peaked transiently. Specifically, fibronectin intensity increased significantly at 15 min (1.5-fold) and then settled into a non-significant upward trend at 60 min (1.3-fold; Fig. 3D), suggesting a rapid but perhaps more labile response than that of integrin β1.

To quantitatively assess this topological disorganization, we used Gray-Level Co-occurrence Matrix (GLCM) texture analysis. Electrical stimulation induced a pronounced decrease in the Angular Second Moment (ASM) and Inverse Difference Moment (IDM) for both proteins (Fig. 3C and D, bottom). These reductions confirm a significant loss of spatial uniformity (ASM) and local homogeneity (IDM), indicating a transition from a cohesive, peripherally organized ECM architecture to a heterogeneous, disordered microenvironment. This structural disorder likely serves as a mechanical anchor, facilitating the observed increase in traction forces while simultaneously arresting collective expansion.

Fig. 3

Electrical stimulation alters the expression and spatial organization of integrin β1 and fibronectin in brain spheroids. A Representative immunofluorescence images of integrin β1 in spheroid cryosections from the control, 15 min, and 60 min stimulation groups. Boxed regions are shown at higher magnification in the right panels. B Quantification of integrin β1, including relative mRNA expression, mean fluorescence intensity, and Haralick texture features representing uniformity (angular second moment, ASM) and local homogeneity (inverse difference moment, IDM). C Representative immunofluorescence images of fibronectin in spheroid cryosections from the control, 15 min, and 60 min stimulation groups, with magnified views of the boxed regions shown on the right. D Quantification of fibronectin, including relative mRNA expression, mean fluorescence intensity, and texture features (ASM and IDM). Electrical stimulation induced marker- and condition-dependent changes in expression and spatial distribution, with significant alterations in selected comparisons as indicated. Scale bars, 200 μm (main images) and 100 μm (magnified images). Data are presented as mean ± SD. *P < 0.05; N = 7 for mRNA expression and N = 17 for image-based quantification; one-way ANOVA with pairwise comparisons as indicated

2.4 Electrical stimulation enhances neurite extension and neural activation in cortical spheroids

To determine whether the stimulation-induced restriction of cell migration reflected a shift toward neuronal maturation, we assessed neurite outgrowth and neural activation in cortical spheroids. Previous studies have linked electrical stimulation to enhanced neuroplasticity in vivo; consistent with this, we observed distinct morphological remodeling after culturing on poly-D-lysine (PDL)-coated glass.

Immunostaining for βIII-tubulin and GFAP showed that spheroids subjected to 60 min of stimulation developed extensive, high-density neurite networks that extended well beyond the dense cellular core (demarcated by yellow contours in Fig. 4A). Notably, although GFAP-positive astrocytes remained largely sequestered within the consolidated core, neurites exhibited robust radial extension. In contrast, neurite projections in the control and 15-min groups remained largely confined to the immediate vicinity of the spheroid body.

Sholl analysis quantitatively confirmed this increase in morphological complexity (Fig. 4B). Although all groups showed peak neurite density within 100 μm of the spheroid edge, significant divergence emerged at greater radii. Beyond 200 μm, the 60-min group maintained high arborization, with 94 intersections, nearly double the complexity of the control (N = 56) and 15-min (N = 48) groups. This enhancement persisted up to 625 μm, indicating that prolonged stimulation promotes both axonal elongation and branching.

To assess the functional correlates of this structural remodeling, we profiled expression of immediate-early genes (IEGs) linked to synaptic plasticity (Fig. 4C). Electrical stimulation elicited distinct temporal dynamics across the tested genes. c-Fos mRNA showed rapid, high-magnitude induction, peaking at a 5.5-fold increase in the 15-min group and declining to a 3.6-fold elevation in the 60-min group. Conversely, Arc expression increased steadily and cumulatively, reaching significant elevations in both the 15-min (1.8-fold) and 60-min (2.2-fold) conditions. Zif268 expression trended upward (1.6-fold at 60 min) but did not reach statistical significance. Collectively, these results suggest that 60-min stimulation drives a phenotypic switch, suppressing macroscopic migration and favoring robust neuritogenesis and activation of plasticity-related transcriptional programs.

Fig. 4

Electrical stimulation modulates neurite outgrowth and immediate early gene expression in brain spheroids. A Representative immunofluorescence images of spheroids cultured on PDL-coated glass for 3 days after electrical stimulation. βIII-tubulin is shown in green and GFAP in magenta. Yellow dashed outlines indicate the dense spheroid core, and boxed regions are shown at higher magnification in the right panels. B Sholl analysis of neurite outgrowth as a function of distance from the spheroid core. The 60 min stimulation group showed greater neurite complexity and extended neurite distribution compared with the control and 15 min groups. C qPCR analysis of immediate early genes (IEGs), including Arc, c-fos, and Zif268. Arc and c-fos expression were significantly altered in selected pairwise comparisons, whereas Zif268 did not show a significant difference among groups, as indicated. Scale bars, 200 μm. Data are presented as mean ± SD in B and as box plots with individual data points in C. *P < 0.05; N = 8; one-way ANOVA with pairwise comparisons as indicated