The studies on human and animals should indicate that procedures were in accordance with institutional guidelines (i.e., Institutional Review Board and Institutional Animal Care and Use Committee). Wild-type (WTc11) hiPSCs were cryopreserved in STEMCELL Technologies and stored in liquid nitrogen. For routine culture, hiPSCs were maintained in Essential 8^TM medium (Life Technologies) on Matrigel-coated plates (Corning) and supplemented with the Rho-associated protein kinase inhibitor Y27632 (Tocris Bioscience) as needed. The cells were incubated at 37℃ in a humidified environment containing 5% CO₂. Passaging was performed weekly once the cells reached 65%∼70% confluence. For cell dissociation, 1X TrypLE Select^TM (Gibco-BRL) was applied at 37℃ for 5 minutes. Subsequently, cells were gently transferred onto fresh Matrigel-coated plates to promote attachment and proliferation. The culture medium was refreshed daily to maintain optimal growth conditions and to preserve pluripotency.

Motor cortex-like organoids were generated from cultured WTc11 iPSCs using a three-step differentiation protocol involving sequential patterning and lineage-specific differentiation. The first step was neuroepithelial differentiation. To initiate neuroepithelial differentiation, WTc11 iPSCs (3×10⁶ cells/well) were seeded into AggreWell^TM plates (STEMCELL Technologies) and cultured in Neurobasal^TM (Gibco^TM) medium supplemented with 10 μM SB431542 (Tocris Bioscience), 100 nM LDN193189 (STEMCELL Technologies), and 1 μM dorsomorphin (Sigma-Aldrich). Within 24 hours, cell aggregates formed into neural rosettes, which were subsequently transferred to Petri dishes (SPL) and cultured in the same medium for an additional 7 days. On day 8, when the rosette structures became visible, neuroepithelial cells were isolated using a Neural Rosette Selection Reagent (STEMCELL Technologies) to obtain a purified population for further differentiation. The second step involved lineage-specific differentiation of progenitors. Isolated neuroepithelial cells were cultured in lineage-specific differentiation medium for 12 days. For glutamatergic progenitor differentiation, the medium was supplemented with 20 ng/mL basic fibroblast growth factor 2 (bFGF2; PeproTech), 5 μM cyclopamine (Sigma-Aldrich), and 5 μM XAV 939 (Tocris Bioscience). For GABAergic progenitor differentiation, the medium was supplemented with 20 ng/mL bFGF2, 50 nM Sonic hedgehog protein (R&D Systems) and 1.5 μM IWR (Sigma-Aldrich). The final step involved the maturation of the motor cortex-like organoids. On day 20, differentiated progenitor cells were dissociated into single cells using Accutase^Ⓡ (Gibco^TM). A total of 8×10⁶ cells (4×10⁶ each of glutamatergic and GABAergic progenitors) were mixed and seeded into a PrimeSurface^Ⓡ 3D Culture Spheroid Plate (S-BIO) in Neurobasal^TM medium supplemented with 10 ng/mL brain-derived neurotrophic factor (BDNF; PeproTech), 10 ng/mL glial cell line-derived neurotrophic factor (R&D Systems), and 10 μM Y27632 for initial organoid formation. After 24 hours, the newly formed spheroids were transferred to suspension culture dishes and maintained in the same medium without Y27632 for an additional 9 days before characterization and cytotoxicity assays.

SK-N-SH human neuroblastoma cells (ATCC) were cultured in Roswell Park Memorial Institute 1640 medium (Gibco-BRL) supplemented with 10% heat-inactivated fetal bovine serum (Gibco-BRL). Cells were incubated at 37℃ in a humidified atmosphere with 5% CO₂. For 3D sphere formation, 3×10⁵ cells per dish were cultured in suspension in Petri dishes, allowing spheroid aggregation over 5 days. This method promoted stable and reproducible 3D structures suitable for subsequent experimental applications.

Cell viability in motor cortex-like organoids was assessed using the WST-8 (WELGENE inc.), LDH (DOJINDO Laboratories), MTT (Sigma-Aldrich) assay to evaluate neuronal toxicity. Single organoids were seeded into 96-well plates and treated for 24 hours with one of the following: 100 μM aspirin (Sigma-Aldrich); amitriptyline (Sigma-Aldrich) at concentrations of 2, 10, 20, 40, 80, or 100 μM; γ-aminobutyric acid (GABA; Sigma-Aldrich) at 5.92, 29.6, 59.2, 296, 1,480, or 2,960 μM; methanol (Sigma-Aldrich) at 5.85%, 10%, 12%, or 15%; or N-methyl-D-aspartic acid (NMDA; Sigma-Aldrich) at 5.85, 10, 12, or 15 mM. Following treatment, the viability of the cells was assessed by adding the reagent following the manufacturer’s protocol. The absorbance was measured at 450 nm using a microplate reader (BioTek). Results are expressed as the percentage of viability relative to untreated control organoids.

Total RNA of the cultured cells cells and human brain tissue (Seoul National University Hospital Brain Bank) were isolated using TRIzol Reagent (Invitrogen), followed by chloroform extraction. One microgram of RNA was reverse transcribed into cDNA using a cDNA Synthesis Kit (Invitrogen), Real-time polymerase chain reaction (PCR) was performed using an Exicycler^TM (Bioneer) with SYBR Green dye (Thermo Fisher Scientific). The PCR conditions included initial denaturation performed at 95℃ for 10 minutes, followed by 40 cycles of denaturation at 95℃ for 10 seconds, annealing at 60℃ for 30 seconds, and extension at 72℃ for 30 seconds. Primer sequences are listed in Table 1.

Cells were cultured into confocal dishes and fixed with 4% paraformaldehyde (Sigma-Aldrich) for 15 minutes at room temperature. After fixation, cells were permeabilized with 0.1% Triton X-100 (Sigma-Aldrich) for 10 minutes. To block the non-specific binding, the cells were incubated with 5% bovine serum albumin (BSA; Sigma-Aldrich) in phosphate-buffered saline (PBS; HyClone) for 1 hour at room temperature. Cells were incubated overnight at 4℃ with specific primary antibodies diluted in 3% BSA-PBS. Primary antibodies included Anti-vGlut1 (1：200, 135 303C3; Synaptic Systems), Anti-Map2b (1：500, MA1-34378; Thermo Fisher Scientific), Anti-GABA (1：1,000, A2052; Sigma-Aldrich), and Anti-TuJ1 (Anti-β- Tubulin III, 1：500, T2200; Sigma-Aldrich). Staining was visualized using a confocal laser scanning microscope (Carl Zeiss). The cell nuclei were counterstained with 4’,6-diamidino-2-phenylindole (DAPI) (1：1,000, 100-43-6; Sigma-Aldrich). Representative images were captured and analyzed for marker expression investigation.

Motor cortex-like organoids were harvested, rinsed using cold PBS, and lysed in radioimmunoprecipitation assay buffer (Thermo Fisher Scientific) containing protease and phosphatase inhibitors (Thermo Fisher Scientific). Protein concentration was determined using a BCA Protein Assay Kit (Thermo Fisher Scientific). Equal amounts of protein (30 μg per sample) were separated in 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis and transferred to polyvinylidene fluoride membranes (Millipore) using a Bio-Rad wet blotting at 100 V for 90 minutes at 4℃. Membranes were blocked with 5% nonfat milk in Tris-buffered saline containing 0.1% Tween-20 (TBS-T) for 1 hour at room temperature. Primary antibodies, including cleaved caspase-3 (1：1,000; Cell Signaling Technology), and α-tubulin (1：1,000; Abcam) as a loading control, were applied overnight at 4℃. After three washes with TBS-T, the membranes were incubated with horseradish peroxidase-conjugated secondary antibodies (1：5,000; Cell Signaling Technology) for 1 hour at room temperature, followed by three additional TBS-T washes. Chemiluminescence signals were detected using an enhanced chemiluminescence system.

Data are presented as the mean±standard error of the mean from the specified number of experiments. Statistical analyses were conducted using Student’s t-test for two-group comparisons and one-way ANOVA for multiple group comparisons. A p-value<0.05 was considered statistically significant.

First, we developed a 3D motor cortex-like organoid model derived from human iPSCs, as depicted in Fig. 1A. The schematic outlines the generation of 3D neuronal spheres by combining glutamatergic and GABAergic motor neuron types in a 1：1 ratio to create mixed 3D cultures. The differentiation process commenced with the formation of neural rosettes from iPSC-derived neuroepithelial cells. Next, we directed lineage specification by applying cyclopamine to induce glutamatergic progenitors formation from the dorsal forebrain and Sonic hedgehog protein to induce GABAergic progenitors formation from the ventral forebrain. By day 30, the application of BDNF facilitated terminal differentiation, which yielded a structurally organized organoid model that reflected the cellular diversity and complexity of the human cerebral cortex. Following terminal differentiation, as shown in Fig. 1B and 1C, successful induction of the target neuron types was verified through immunostaining. Specifically, vGlut1, a marker for glutamatergic neurons, and Map2b, a general neuronal marker, were highly expressed in the glutamatergic cluster. Additionally, the robust expression of GABA, a marker specific to GABAergic neurons, and TuJ1, a pan-neuronal marker, was detected in the GABAergic neuron differentiation cluster. These results confirm that our differentiation protocol effectively and reliably generated two distinct neuronal subtypes—glutamatergic and GABAergic—derived from human iPSCs.

To determine whether the 3D motor cortex-like organoid model exhibited characteristics of the human brain, we first performed quantitative polymerase chain reaction (qPCR). Fig. 2A presents microscopic images comparing the human neuroblastoma 3D model with both 2D and 3D models of iPSC-derived motor cortex-like organoids. Fig. 2B presents gene expression profiles of human neuroblastoma cells, iPSC-derived 2D and 3D model cultures, and human brain tissue using specific primers for vGlut1, vGlut2, ARPP21, and GABA. The results indicate that iPSC-derived 2D and 3D cultures displayed gene expression patterns closely aligned with the molecular signature of the human brain, unlike the neuroblastoma cells. To further assess cellular stability and safety within the cytotoxicity model, we evaluated the markers of autophagy, cell death, and neuronal regeneration. The analysis revealed that the 3D iPSC-derived model, which closely mimics the human brain’s tissue architecture, exhibited significantly lower expression levels of ATG5, a critical autophagy gene indicative of intracellular stability, and caspase-3, a marker of apoptosis, compared to the 2D model. Moreover, lower levels of GAP43, a key marker of neuronal regeneration, were observed in the 3D model, indicating enhanced cellular stability in the 3D cultures compared to its 2D counterpart (Fig. 2C). These findings highlight the potential of 3D motor cortex-like organoids as robust in vitro models for studying neurotoxicity and regenerative processes, providing a closer approximation of the native human brain environment.

To develop organoids as a platform for toxicity assessment, we first evaluated their structural integrity by measuring organoid size and characterizing cellular differentiation. Immunofluorescence staining performed before organoid aggregation confirmed a high expression of neural progenitor markers, with PAX6 (93.0%±4.9%/Field) and Nestin (90.5%±3.8%/Field), indicating robust neural commitment of progenitor cells (Fig. 3A). By day 15 of differentiation, approximately 550±10 μm organoids were successfully generated from neuronal progenitor-stage cells (Fig. 3B). On day 15 post-organoid formation, immunofluorescence analysis confirmed the differentiation of progenitor cells into glutamatergic and GABAergic neuronal subtypes, as indicated by the expression of vGLUT1 (38.7%±5.9% of DAPI-positive cells) and GABA (44.1%±11.0% of DAPI-positive cells). Furthermore, the high expression levels of the pan-neuronal markers Map2b (62.5%±6.8% of DAPI-positive cells) and TUJ1 (72.4%±11.2% of DAPI-positive cells) provided additional evidence of robust neuronal differentiation within the organoids (Fig. 3C). To assess cellular heterogeneity, we performed qPCR to compare gene expression levels among undifferentiated cells, 3D-differentiated glutamatergic and GABAergic neurons, and fully formed organoids. We demonstrated that significantly higher expression levels of vGlut1 and GAD1 in organoids compared to individually differentiated neuronal populations, suggesting enhanced neuronal maturation and a more heterogeneous cellular composition within the organoid structure (Fig. 3D). These findings highlight the potential of motor cortex-like organoids as a physiologically relevant platform for evaluating neurotoxicity and studying region-specific neuronal interactions.

To enhance the reliability of organoid-based neurotoxicity assessments, we implemented a multi-parametric approach by integrating WST-8, LDH, and MTT assays, each probing distinct aspects of cellular metabolism, rather than relying on a single cytotoxicity measure. Additionally, intracellular protein analysis was performed using Western blotting to elucidate molecular responses to toxic exposure. To evaluate neurotoxicity, the mature organoids were exposed to various concentrations of aspirin, GABA, NMDA (a non-neurotoxic drug) and amitriptyline and methanol (a known neurotoxic compound) (Fig. 4A-4C). Cytotoxicity assay results demonstrated that aspirin, GABA and NMDA exhibited no detectable toxic effects, whereas amitriptyline and methanol induced a dose-dependent cytotoxic response, confirming its neurotoxic potential. Furthermore, western blot analysis confirmed the activation of the apoptotic pathway, as evidenced by an increase in cleaved caspase-3, a key marker for apoptosis, in amitriptyline-treated organoids compared to controls. This increase was concentration-dependent, with significantly higher expression levels at elevated concentrations (Fig. 4D). These findings suggest that the 3D motor cortex-like organoid model is a reliable platform for assessing neurotoxicity and effectively distinguishing between non-toxic and toxic compounds.