Abstract Background Despite the high prevalence of cerebral ischemic stroke, effective clinical treatments remain limited. With the development of regenerative medicine, induced neural progenitor cells (iNPCs) demonstrate ideal potential and good availability for autologous transplantation therapy. However, current differentiation protocols for iNPCs still have room for improvement in terms of purity, reproducibility, scalability and differentiation potential. Methods We aimed to develop a scalable, stable, and efficient 3D aggregate-based method for iNPC production in suspension culture, avoiding detrimental cell dissociation and replating processes. We evaluated the therapeutic potential of iNPCs in the chronic phase of a transient middle cerebral artery occlusion (tMCAO) mouse model and explored iNPC subpopulations via single-cell RNA sequencing to elucidate their pleiotropic therapeutic potentials. Results iNPCs generated from three iPSC lines displayed high NPC marker expression and an average 176-fold cell expansion over the 12-day culture period. These iNPCs could spontaneously differentiate into both neurons and glial cells in vitro. In the tMCAO model, transplanted iNPCs remodeled the microenvironment by alleviating neuroinflammation, inhibiting chronic microgliosis and astrogliosis, promoting M2 polarization of microglia, and preserving astrocytic morphology in the ischemic penumbra. Mechanistically, iNPCs can be divided into four subpopulations, with neuroepithelia being the most abundant and capable of rapidly replenishing damaged cells and mitigating microenvironmental deterioration. Conclusions We developed a simple and efficient 3D aggregate-based method for iNPC differentiation. These iNPCs showed excellent potential for post-stroke recovery and represent a valuable tool for clinical translation. Supplementary Information The online version contains supplementary material available at 10.1186/s13287-025-04433-z. Keywords: Induced neural progenitor cells, 3D aggregates, Stem cell transplantation, Chronic ischemic stroke, Post-stroke recovery Background Chronic inflammation, excessive fibrosis, and loss of progenitor cells are major hurdles that impede regenerative processes, especially in brain disorders [[44]1]. Neural progenitor cells (NPCs) exhibit self-renewal ability and multipotency to differentiate into both neuronal and glial cells of the central nervous system (CNS) [[45]2]. Therefore, significant therapeutic potential has been granted to NPCs for treating degenerative [[46]3], ischemic [[47]4] and traumatic [[48]5] brain diseases. Among these, cerebral ischemic stroke is the second leading cause of death and the most common cause of disability worldwide [[49]6]. Despite its high incidence, effective treatments to improve functional recovery remained limited. Thrombolytic therapy, the most commonly used treatment, has a narrow therapeutic window of only 4.5 h after the initial stroke [[50]7]. Thus, there is an urgent need for new therapeutic strategies targeting the subacute and chronic phases of ischemic stroke. The pathophysiological process of ischemic stroke includes blood-brain barrier dysfunction, neuroinflammation, neuronal loss, and glial activation [[51]8]. Many preclinical studies suggested that NPCs displayed pleiotropic therapeutic potentials to alleviate neuroinflammation, rescue neuronal loss, and promote angiogenesis [[52]9]. Since the development of induced pluripotent stem cell (iPSC) technology, iPSC-derived cell therapies have attracted tremendous attention in regenerative medicine. To facilitate clinical translation, an ideal NPC differentiation protocol should be easily manipulable and scalable. However, existing protocols either require specialized training for intricate steps or rely on adherent cultures, which are challenging to scale up [[53]10]. For instance, the neural rosette method involves complex steps, including embryoid body (EB) formation and neural rosette isolation [[54]11]. Additionally, prolonged cultivation of iNPCs leads to biased differentiation fate and even chromosomal abnormalities [[55]12, [56]13]. Therefore, our study aims to develop a total 3D aggregate-based iNPC differentiation method with good reproducibility, ease of manipulation, and scalability for therapeutic applications. In this study, we developed an efficient and reproducible suspension platform for the direct generation of iNPCs from human iPSCs without passaging until cell collection. The generated NPCs exhibited stable marker expressions and the capability to spontaneously differentiate into neuronal and glial cells in vitro. Co-culture experiments with homologous peripheral blood mononuclear cells (PBMCs) and microglia further revealed their immunosuppressive function. To assess the therapeutic potential in vivo, we employed a transient middle cerebral artery occlusion (tMCAO) ischemic stroke model and tested iNPCs transplantation during the chronic phase. Our results demonstrated that transplanted iNPCs survived, migrated to the lesion site, and alleviated neuroinflammation, microgliosis, and astrogliosis. Finally, we explored iNPC subpopulations with different differentiation potentials via single-cell RNA-sequencing (scRNA-seq) to elucidate their pleiotropic therapeutic effects. Based on our analysis, we demonstrated the reliable generation of iNPCs with broad therapeutic potential for in vivo application. Methods and materials All the chemicals and reagents used in this study were listed in Table [57]S1, [58]S2, [59]S5. Generation and maintenance of human iPSC Three hiPSC lines were established from healthy adult volunteers for this study: iPSC-BKM-001 (a 27-year-old man), iPSC-BKM-002 (a 33-year-old man), and iPSC-BKF-001 (a 30-year-old woman). Peripheral blood samples were collected with consent, and PBMCs were isolated using a Ficoll-based density gradient centrifugation method (TBD, China). CytoTune-iPS Sendai Reprogramming Kit (Thermo) was used to reprogram PBMCs into hiPSCs according to the manufacturer’s instructions. hiPSCs were maintained on hESC-qualified Matrigel (Corning)-coated culture plates in Essential 8 medium (Thermo) with daily medium exchange. hiPSCs were dissociated with Accutase (StemCell, Canada) every four days and seeded at a density of 1.5 × 10^4/cm^2. For the first 24 h after seeding, cells were supplemented with a cytoprotective cocktail containing Chroman 1, Emricasan, Polyamines and Trans-ISRIB (CEPT) [[60]14]. Karyotype analysis The iPSC clones were analyzed for chromosomal abnormalities by G-banding karyotyping at 400-band resolution, with twenty metaphases examined. The experiments were conducted by Kenuo Medical Lab (Shenzhen, China). In vitro differentiation into three germ layers The differentiation potential of iPSCs was analyzed according to previous research with modifications [[61]15]. Briefly, dissociated iPSCs were transferred to low-attachment plates and maintained in EB suspension medium (knockout DMEM supplemented with 20% Knockout SR, 1×NEAA, 1×GlutaMAX, and 0.1 mM β-mercaptoethanol) for 7 days with agitation. The EBs were then collected and proceeded to adherent self-differentiation medium (knockout DMEM supplemented with 10% Knockout SR, 5% FBS, 1×NEAA, 1×GlutaMAX and 0.1 mM β-mercaptoethanol) for another 7 days. Immunostaining of three germ layer markers (ectodermal: TUJ1; mesodermal: αSMA; endodermal: AFP) was performed to confirm the differentiation potential of iPSCs. Teratoma formation and analysis Cells were dissociated, and 5 × 10^6 iPSCs or 2 × 10^7 iNPCs were subcutaneously injected into the flank of nonobese diabetic/severe combined immunodeficient (NOD/SCID) mice to assess teratogenic potential. After 8–10 weeks, tumors were collected, fixed in 4% paraformaldehyde, and processed for paraffin embedding and sectioning. Histological analysis of the three germ layers was performed using hematoxylin/eosin staining and examined under a light microscope. The teratoma size was measured by the ellipsoid formula (1/2 × Length × Width × Height) as previously reported [[62]16]. The maximum teratoma size allowed before euthanasia was 2000 mm^3. All animals were euthanized by cervical dislocation when necessary. Generation of iPSC-derived NPCs through a 3D-aggregate based method For neural differentiation, the three established hiPSC lines were used and underwent a 6-day induction and a 6-day expansion period. Briefly, dissociated hiPSCs were seeded at 2.5 × 10^5/mL in Neural Induction medium, which consisted of 50% DMEM/F12, 50% Neurobasal, 1×B27 supplement with Vitamin A, 1×N2 supplement, 1×GlutaMAX, 0.1 mM β-mercaptoethanol, 10 µM SB431542 and 3 µM CHIR99021, on low-attachment plates with agitation at 80 rpm on day 0. On day 6, aggregates were transferred to Neural Expansion medium, which consisted of 50% DMEM/F12, 50% Neurobasal, 1×B27 supplement with Vitamin A, 1×N2 supplement, 1×GlutaMAX, 0.1 mM β-mercaptoethanol, 20 ng/mL bFGF and EGF. The cytoprotective cocktail CEPT [[63]14] was added to the culture on day 0. Half of the medium was replaced daily to eliminate excess floating dead cells. On day 4, day 6 and day 9, the aggregate culture was split at 1:2 or 1:3 ratios without dissociation. On day 12, aggregates were collected by gravity, rinsed with PBS solution, and dissociated using Accutase (StemCell) in a 37 °C water bath with agitation for 5 min if necessary. Cell viability analysis and LIVE/DEAD stain Cell number and viability of dissociated iPSCs or iNPCs were assessed using a dual fluorescence staining method [[64]17]. Briefly, dissociated cells were stained with acridine orange/propidium iodide (AO/PI) dye (Nexcelom, USA), and measured using a Cellometer Auto 2000 fluorescent viability cell counter (Nexcelom, USA). Cell expansion ratio on each day was calculated as: viable cell number yield on day A/viable cell number seeded on day 0. Neural aggregates were incubated with AO/PI staining solution at room temperature for 20 min, and cell viability inside the aggregates was observed under a fluorescence microscope. Neuronal differentiation The differentiation potential of iNPCs was analyzed according to previous research with modifications [[65]18, [66]19]. Briefly, dissociated iNPCs were seeded at 1 × 10^5/mL on Matrigel-coated coverslips and cultured in differentiation medium, which consisted of 50% DMEM/F12, 50% Neurobasal, 1×B27 supplement with Vitamin A, 1×N2 supplement, 1×GlutaMAX, 1×NEAA, 200 µM ascorbic acid, 500 µM dibutyryl cyclic-AMP, and 0.1 mM β-mercaptoethanol, 10ng/mL BDNF, GDNF, NGF and NT3. The cytoprotective cocktail CEPT [[67]14] was added to the medium on day 0, and the medium was refreshed every 3 days. Samples were collected on day 14 and analyzed by immunostaining for mature neuronal and glial markers and real-time PCR (RT-PCR) analysis. Immunofluorescence staining Perfused brains or neural aggregates were fixed in 4% paraformaldehyde at 4℃ overnight, dehydrated in 30% sucrose/PBS solution for 2 days at 4℃, and embedded in OCT compound (SAKURA, Japan). Sections were cut at 25–10 μm thickness using a cryostat (Leica, Germany). Cells on glass coverslips were fixed in 4% paraformaldehyde for 15 min at room temperature. Cells, aggregates and brain sections were permeabilized with 0.3% Triton X-100/PBS solution for 20 min and blocked with 0.3% Triton X-100 and 10% normal horse serum for 30 min. They were then incubated with primary antibodies (Table [68]S1) at 4℃ overnight, followed by secondary antibodies (Table [69]S2) for 2 h at room temperature. Nuclei were counterstained with Hoechst 33,342 (Invitrogen) for 15 min. Images were captured using an inverted fluorescence microscope (Olympus IX83 or CKX53) or a confocal microscope (Zeiss LSM980). Imaging analysis Quantitative analysis of fluorescent images was conducted using ImageJ Software. For GFAP and IBA1 signal analysis, the images were subjected to threshold processing and measured using the Area Integrated Density tool [[70]20]. Specifically, two fields per section from two randomly selected sections of three mice per group were analyzed, resulting in a total of 12 fields per group. To quantify iNOS and CD206-positive cells, five randomly selected fields per mouse (15 fields per group) were counted, using Hoechst counterstaining. For Sholl analysis, individual astrocytes labeled with GFAP staining were digitally reconstructed using ImageJ, and a series of concentric circles were placed 5 μm apart, centered on the astrocyte soma [[71]21]. The number of intersections between astrocyte processes and each concentric circle from two randomly selected fields per mouse were manually counted, with 11 astrocytes in the control group and 15 in iNPC-treated group. Quantitative real-time PCR Total RNA was extracted from brains or cells using RNA Easy Fast Tissue/Cell Kit (TIANGEN, China) and total cDNA was prepared with the PrimeScript™ RT reagent Kit (TaKaRA, Japan) according to the manufacturer’s instructions. RT-PCR was performed using SYBR Green Realtime PCR Master Mix (TOYOBO, Japan) on a Roche LC480II real-time system (Roche, Germany). Relative gene expression levels were calculated using the 2^−ΔCT method with GAPDH (human) or Gapdh (mouse) as internal controls. Primer sequences are listed in Table [72]S3 and [73]S4. Flow cytometry For iPSC surface marker staining, dissociated cells were incubated with antibody-fluorophore conjugates (SSEA4-APC, TRA1-60-PE, BD Biosciences) for 20 min at room temperature. For iNPCs intracellular staining, dissociated cells were fixed in 4% paraformaldehyde for 20 min at room temperature, permeabilized with Perm III buffer (BD Biosciences) for 30 min on ice, and then incubated with antibody-fluorophore conjugates (PAX6-PE, SOX1-PerCP-Cy5.5, Nestin-APC, BD Biosciences) for 30 min at room temperature. After staining, cells were rinsed, resuspended in 300 µL PBS, and analyzed on a BD Aria II. Data was analyzed by FlowJo software. Coculture of autologous PBMCs and iNPCs Coculture of activated autologous PBMCs and iNPCs was performed as described with modifications to assess the immunomodulatory effect of iNPCs [[74]22]. Briefly, PBMCs were cultured in RPMI1640 medium with 10% FBS and activated using 1 µg/mL CD3/CD28 antibodies and 50 U/mL IL2 (Seafrom biotechnology, China). Activated PBMCs were cultured alone or with autologous iNPCs at effector to target (E: T) ratios of 1:5, 1:3, 1:2, 1:1, 1.3:1 and 2:1 for 72 h. Proliferating PBMCs were labeled with 5-ethynyl-2-deoxyuridine (EdU), which was added to culture 24 h prior to analysis. EdU signal was detected using the Click-iT EdU assay kit (Thermo), and proliferating PBMCs were identified by CD45 and EdU expression via flow cytometry. Growth inhibition percentage was calculated as (%proliferating PBMCs in positive control-%proliferating PBMCs in coculture) / %proliferating PBMCs in positive control. Supernatant cytokine/chemokine expression was measured using the ProcartalPlex Human Cytokine/Chemokine/Growth Factor Panel 1 45plex kit through Luminex xMAP technology (Thermo). Transwell coculture system of HMC3 cell line and iNPCs To evaluate the immunomodulatory effect of iNPCs on microglia, a transwell coculture system was performed as described with modifications [[75]23]. Briefly, the human microglial cell line (HMC3) was obtained from Pricella Biotechnology (China) and maintained in MEM supplemented with 10% FBS and 1×NEAA. HMC3 cells were seeded in the bottom 6-well plate at 2 × 10^5 cells/well, and iNPCs were seeded in the upper culture insert (0.4 μm, PET, Nest Biotechnology) at 5 × 10^5 cells/well before transwell assembly. 24 h after cell seeding, HMC3 cells were randomly divided into CTRL, M1, and M1-iNPC groups. A mix of 20 ng/mL IFNγ and 20 ng/mL LPS proinflammatory cytokines was added to the medium in the M1 and M1-iNPC groups to induce M1 polarization. Transwells containing iNPCs were then assembled with the M1- iNPC group HMC3 wells to establish the coculture system. After an additional 24 h, cells were collected for RT-PCR analysis of M1 and M2 markers. Animals The study has been reported in line with the ARRIVE guidelines 2.0. All animal procedures were approved by the Animal Experimental Ethics Committee of the Institutional Review Boards at Shenzhen Beike Biotechnology Co., Ltd. C57BL/6 mice (males, 8–10 weeks) were maintained at 21–24℃ in the individually ventilated cages with a 12/12-hour light/dark cycle. The maximum caging density was five, and all mice had free access to food and water. Tests were conducted between 14:00 to 17:00 with randomized testing order, ensuring each mouse was tested at different times across test days. Animal health was monitored by weekly weigh measurements, food and water intake, and general observations of activity and fur condition. All animals were euthanized by cervical dislocation when necessary. tMCAO model A total of 45 adult C57BL/6 mice were obtained from BesTest Biotech (China). Cerebral ischemia was established by a 40 min tMCAO surgery on male C57BL/6 mice (8–10 weeks) as previously described with modifications [[76]24]. Briefly, mice were anesthetized with isoflurane (RWD, China) and maintained at 37℃ throughout the procedure. For tMCAO, a silicon-coated filament was inserted into the left middle cerebral artery to block the blood flow and induce cerebral ischemia. After 40 min, the filament was removed to allow reperfusion. Neurological deficits were assessed 1 day and 3 days post-surgery, with modified Neurological Severity Score (mNSS) scored above 9 indicating the successful establishment of cerebral ischemia. Body weight was measured before iNPCs transplantation, and mice weighting less than 15 g were excluded. Among the 45 mice, 15 were excluded due to death or weakness before transplantation (n = 6) or mild neurological deficits (n = 9). The remaining tMCAO mice were randomized into iNPC treatment and control groups (15 mice each) using the RAND function in Microsoft Excel. iNPC transplantation At day 5 post-tMCAO surgery, mice were anesthetized and placed in a stereoscopic apparatus (RWD, China). The skull was exposed, and a hole was drilled above the ipsilateral hemisphere for iNPCs microinjection. Recovered iNPCs (derived from line iPSC-BKM-001) were resuspended in artificial cerebrospinal fluid (aCSF) at 1 × 10^5 cells/µL, and 1.5 µL (0.5µL/min) for each point was injected into the left striatum and cortex using an infusion pump (RWD, China). After each injection, the needle was held for another 5 min to prevent leakage and achieve good absorption. The control group received an equivalent volume of aCSF. Injection coordinates were as follows: anterior-posterior (AP) + 0.8 mm, medial-lateral (ML)-1.5 mm, dorsal-ventral (DV)-1.5 mm and AP + 0.8 mm, ML-1.5 mm, DV-3 mm from the bregma point. To enhance graft survival, all animals were immunosuppressed with cyclosporine A (10 mg/kg, AbMole, China) via subcutaneous injection, starting 1 day before transplantation and continued every 3 days throughout the experiment [[77]4]. Measurement of neurological deficits and behavioral tests To ensure blinded behavioral assessments, all measurements were conducted by two investigators independent of group allocation. Two behavioral tests were performed to monitor changes post-iNPC transplantation (8 mice for each group). Tests were conducted pre-ischemic, and at 1-, 3-, 6-, 8-, 11-, and 14-day post-ischemia. Neurological function was assessed by mNSS, as previously reported [[78]25]. The mNSS scored from 0 to 18 (normal score: 0; maximum score: 18), evaluated balance, motor, and reflex functions, with points assigned for test failures or absent reflexes. The cylinder test accessed forepaw motor asymmetry by videotaping rodents exploring the cylinder wall for 5 min. Forepaw motor deficit was calculated as the ratio of paw-dragging behavior to paw touching (% paw drag/paw touch), as previously reported [[79]26]. TTC staining and analysis of relative brain atrophy Mouse brains were cut into 2-mm thick coronal sections and stained with prewarmed 2% 2,3,5-triphenyltetrazolium chloride (TTC) solution (Solarbio, China) for 15 min at 37℃. Excess dye was washed off with PBS, and sections were photographed with a digital camera. The lesion area on each slice was measured against the contralateral hemisphere, and relative brain atrophy was calculated as the average lesioned percentage per slice. Single-cell library Preparation and scRNA-seq data analysis The scRNA-seq libraries were generated using the 10x Genomics Chromium Platform following the manufacturer’s protocol. Library quality was assessed with a Qubit (Thermo Fisher Scientific) and an Agilent 2100 Bioanalyzer (Agilent Technologies, California). Barcoded libraries were sequenced on an Illumina NovaSeq 6000 instrument to a depth of at least 100,000 reads per cell in 150 bp paired-end (PE150) mode. The raw sequencing data was processed, using Cell Ranger (version 3.1.0) for base calling, adapter trimming, and demultiplexing [[80]27]. Downstream analysis was conducted with R (version 4.1.3) and Seurat (version 4.0.0) [[81]28]. Cells expressing fewer than 300 genes and genes not detected in any cells were filtered out. The final dataset included 9,000 unique genes with mitochondrial transcripts counting below 100,000 per cell. Seurat’s standard pipeline was utilized for cell clustering and differential expression analyses. Differentially expressed genes (DEGs) between samples/clusters were identified using the Seurat’s FindAllMarkers function with the Wilcoxon test. DEGs were subjected to GO and KEGG pathway enrichment analysis using clusterProfiler [[82]29] to explore associated biological processes and signaling pathways. Statistical analysis Statistical analysis for was performed using GraphPad Prism 8.0. Data was presented as mean ± standard error of the mean (SEM). Significance was determined by two-tailed unpaired Student’s t-test, one-way analysis of variance (ANOVA), or two-way ANOVA. For multiple comparisons, Tukey’s post hoc test was used. A two-tailed p < 0.05 was considered statistically significant. Results Generation and characterization of human PBMC-derived HiPSCs Human PBMCs were collected from healthy donors with informed consent and reprogrammed using commercially available Sendai viral vectors (Thermo). Reprogrammed cells exhibiting characteristic iPSC morphology were observed approximately 20 days post-transfection. Potential iPSC clones were identified through the expression of specific markers, including the transcription factors NANOG, OCT4, and SOX2, as determined by immunofluorescent staining (Fig. [83]S1a), as well as cell surface markers SSEA4 and Tra-1-60, assessed via flow cytometry (Fig. [84]S1b). The multipotency of the iPSC clones was further evaluated through in vitro differentiation into the three germ layers (Fig. [85]S1c) and teratoma formation in NOD-SCID mice (Fig. [86]S1d). Ultimately, all established iPSC clones exhibited a normal karyotype (Fig. [87]S1e). The derived human iPSC (hiPSC) lines were subsequently utilized in the following NPC differentiation experiments. Generation and characterization of iPSC-derived NPCs via a 3D aggregate-based method The timeline and corresponding figures for the generation of iPSC-derived iNPCs are presented in Fig. [88]1a. In summary, small aggregates exhibiting a uniform structure were observable following an overnight shaking period on day 1 (Fig. [89]1a). By day 4 or day 5, the characteristic apicobasal polarized morphology became evident within each aggregate (Fig. [90]1a). During the neural expansion period, buds emerged on the surface of the aggregates, which subsequently detached and underwent spontaneous growth into new aggregates (Fig. [91]1a). On day 12, the neural aggregates were collected and dissociated for subsequent analysis and animal studies. Fig. 1. [92]Fig. 1 [93]Open in a new tab Efficient generation and characterization of iPSC-derived NPCs via a 3D aggregate-based method. (a) Schematic overview and representative bright-field images of iPSC-derived NPC aggregates. The enlarged figure on day 1 showed aggregates with uniform structure, and the enlarged figure on day 4, day 6 showed a typical apicobasal polarized morphology. Red arrowheads on day 8, day 12 indicated newly formed aggregates shedding from the original ones. (b) Viability of a day 12 iNPC aggregate. Green: AO; Red: PI. (c) Cell viability on day 0 and day 12 during iNPC generation. (d) Cell expansion ratio on day 0, day 6, day 9 and day 12 during iNPC generation. (e) Representative flow cytometry results on day 12 iNPCs. (f) Average percentage of SOX1/PAX6 + and SOX1/NESTIN + cells in day 12 iNPCs. (g) Representative immunostaining images of NPC markers (NESTIN, SOX1, SOX2 and PAX6), proliferation marker (Ki67), mature neuronal marker (MAP2) and apical surface marker (ZO-1) from day 12 iNPC aggregates. White arrows indicated the centers of inner rosettes in a neural aggregate. (h) Gene expression of pluripotency marker (NANOG, OCT4), NPC marker (NESTIN, SOX1, PAX6), gliogenic marker (SOX9), proneuronal marker (DCX) and mature neuronal marker (NEUN) in day 4, day 6 and day 12 samples during iNPC generation, measured by RT-PCR. (i-j) Representative immunostaining images of the (i) neurogenic markers (MAP2, GAD65&67, vGLUT1) and (j) gliogenic markers (S100β, GFAP) from iNPCs subjected to a 14-day in vitro differentiation culture. Dotted rectangles marked the position of enlarged area below. White arrowheads demonstrated potential spine structures. Scale bar in (a): 500 μm; scale bar in (b, i-j): 100 μm; scale bar in (g): 50 μm. n = 3 for (c-d, f). n.s. represented no significance by Student’s t test in (c). * represented p < 0.05, ** represented p < 0.01, **** represented p < 0.0001, and n.s. represented no significance by one-way ANOVA followed by Tukey’s post hoc test in (h) Despite the long expansion period without cell dissociation, no necrotic zones were identified within the aggregates as determined by LIVE/DEAD staining (Fig. [94]1b). To conduct a more comprehensive assessment of iNPCs, neural aggregates were dissociated using Accutase and subsequently quantified using an automated cell counter. The visual representations depicting the neural aggregates before and after cell dissociation indicated that the dissociation process achieved optimal efficiency, as evidenced by the absence of residual aggregates after dissociation (Fig. [95]S2a). Additionally, a representative bright field image, along with the associated live (AO), dead (PI), and composite views generated by the automated cell counter, were also presented (Fig. [96]S2b). A comparative analysis of cell viability of the iNPCs derived from three distinct iPSC lines revealed no statistically significant differences (D0: 90.33 ± 0.61%, D12: 89.43 ± 0.97%, p = 0.4749; Fig. [97]1c). Furthermore, the average expansion ratio in cell number demonstrated a 17.33 ± 0.55-fold increase following the neural induction period on day 6, and a 176.10 ± 28.29-fold increase on day 12 after the neural expansion period (Fig. [98]1d). To characterize iNPCs generated through this 3D-aggregate based differentiation method, we examined the expression of NPC markers, including SOX1, PAX6 and NESTIN, by flow cytometry (Fig. [99]1e). The iNPCs derived from the three iPSC lines exhibited an average of 94.13 ± 3.63% SOX1/NESTIN double-positive cells and an average of 93.03 ± 4.58% SOX1/PAX6 double-positive cells (Fig. [100]1f). The consistent expression of these markers across the three distinct donors suggested that iNPCs were robustly and efficiently induced via the 3D aggregate-based method. Furthermore, the aggregate structure was analyzed through immunofluorescence staining. The staining results indicated that the nuclei were situated in the outer layer and the inner rosettes of the hollow aggregates, while neurites (identified by NESTIN expression) were predominantly found between these two regions (Fig. [101]1g). Additionally, the colocalization of NPC nuclear markers, specifically SOX1, PAX6 and SOX2, was visualized, and proliferating cells (denoted by Ki67 expression) were detected in both the outer layer and the inner rosette (Fig. [102]1g). The immunostaining results also demonstrated minimal expression of the mature neuronal marker MAP2 (Fig. [103]1g). The apicobasal polarity of the neuroepithelial sheets within the aggregates was evidenced by the expression of the apical marker ZO-1, which was localized on the outer surface of the aggregates and at the center of the inner rosettes (Fig. [104]1g). Gene expression analysis also demonstrated a consistent downregulation of pluripotent markers (NANOG, OCT4) and a stable upregulation of NPC markers (NESTIN, SOX1, PAX6) in day 4, 6 and 12 iNPCs, when compared to the iPSC control (Fig. [105]1h). In contrast, the expression levels of proneuronal and mature neuronal markers (DCX, NEUN) remained consistently low throughout the entire iNPC generation process (Fig. [106]1h). These findings demonstrated that iNPCs generated through the 3D aggregate-based method exhibited high purity with minimal spontaneous differentiation. To further elucidate the differentiation potential, iNPCs derived from day 12 aggregates were subjected to spontaneous differentiation for 14 days. The expression of mature neuronal marker (MAP2), GABAergic neuronal marker (GAD65/67), and glutamatergic neuronal marker (vGLUT1) were observed in the differentiated neuronal cells (Fig. [107]1i). Additionally, immature spine structures were also visualized within the differentiated GABAergic neurons (Fig. [108]1i). Gene expression analysis revealed a significant upregulation of forebrain (TBR1), midbrain (FOXA2), and hindbrain (GBX2) markers in the differentiated neuronal cells when compared to the day 12 iNPC control (Fig. S3). In addition to the neurogenic potential, the upregulation of the gliogenic marker SOX9 was observed during the iNPC generation process (Fig. [109]1h). Consistent with expectations, astrocytes exhibiting an immature phenotype were identified within the differentiated culture, as evidenced by the expression of astrocytic markers (GFAP, S100β) (Fig. [110]1j). Collectively, our findings demonstrated that this 3D aggregate-based method reliably and effectively generates iNPCs with high purity and extensive differentiation potential. iNPCs with the highest Sox1Pax6 positivity were used in the following functional assessments. To ensure the safety of the iPSC-derived iNPCs, the remaining pluripotent cells were analyzed using teratoma formation assay in NOD-SCID mice, with isogenic iPSCs as positive control. No teratomas were observed among the 6 mice injected with iNPCs over a 2-month period (Fig. [111]S4). HiNPCs exhibited dose-dependent immunosuppressive functions in vitro Given the identified gliogenic population in iNPCs, we undertook an investigation to ascertain whether these cells possess immunoregulatory functions, as previously documented in the CNS [[112]30]. Briefly, we conducted in vitro coculture experiment involving iNPCs and activated homologous PBMCs at various E: T ratios. A dose-dependent inhibitory effect on the proliferation of activated PBMCs was observed, with E: T ratios ranging from 1:5 to 2:1 (Fig. [113]S5a, b). Notably, a growth inhibition rate of 36.38% in PBMCs was detected at a relatively low E: T ratio of 1:3 (Fig. [114]S5b). To further elucidate the underlying molecular mechanisms, the expression levels of cytokines and chemokines in the coculture supernatant were analyzed using a multiplex bead-based assay (Luminex xMAP, Thermo). A significant reduction in the expression of proinflammatory cytokines (IFN-γ, TNF-α), as well as the proliferation stimulator (GM-CSF), and the chemokines (MIP1α, MIP1β) was observed in the coculture system when compared to the positive control (Fig. [115]S5c). Besides, an elevated expression of the immunosuppressive factor IL1RA was also detected (Fig. [116]S5c). Consequently, these 3D aggregate derived iNPCs demonstrated a dose-dependent immunosuppressive function, characterized by the inhibition of proinflammatory cytokines, proliferation stimulator, and chemokines, alongside an increased expression of the immunosuppressive factor IL1RA. HiNPCs transplantation facilitated postischemic recovery in tMCAO mice We next evaluated the therapeutic effects of iNPC treatment after cerebral ischemia and reperfusion. Briefly, mice subjected to tMCAO were treated with ipsilateral transplantation of hiNPCs or an equivalent volume of aCSF as a control (Fig. [117]2a). Cerebral damage was evaluated through the examination of TTC stained brain Sect. 14 days post-tMCAO surgery (Fig. [118]2b). In comparison to the control group (34.55 ± 6.13% necrosis), the iNPC-treated mice demonstrated a significantly reduced atrophy area (15.64 ± 1.30%, p < 0.05) (Fig. [119]2c). Furthermore, preservation of brain structure was maintained at 30 days post-surgery (Fig. S6a), with the iNPC-treated group displaying a diminished atrophy percentage (13.07 ± 3.69%) in contrast to the control group (34.92 ± 4.64%) (Fig. [120]S6b). Fig. 2. [121]Fig. 2 [122]Open in a new tab Effects of transplanted iNPCs on post-stroke recovery in tMCAO mice. (a) Summary of the experimental timeframes. (b) TTC staining of coronal brain sections from iNPC and control group on day 14 after tMCAO surgery. Brain infarcts were marked by dotted lines. (c) Quantification of relative brain atrophy. (d) mNSS of iNPC and the control group. n = 8. (e) Cylinder test result of the ipsilateral forepaw in iNPC and the control group. n = 3 for (b-c), n = 8 for (d-e). * represented p < 0.05 by Student’s t test in (c), * represented p < 0.05, ** represented p < 0.01 and *** represented p < 0.001 by two-way ANOVA followed by Tukey’s post hoc test in (d-e) Behavioral deficits were assessed by the mNSS and cylinder tests prior to treatment and at 1 day (Day 6), 3 days (Day 8), 6 days (Day 11), and 9 days (Day 14) post-treatment (Fig. [123]2a). Neurological assessments indicated that iNPCs exhibited therapeutic effects shortly after transplantation, as evidenced by significant reductions in mNSS at both 1 day (iNPCs: 7.75 ± 1.10, control: 11.25 ± 0.75, p < 0.01) and 3 days (iNPCs: 6.00 ± 0.93, control: 9.25 ± 0.95, p < 0.05) post-treatment (Fig. [124]2d). The lower mNSS were sustained in the iNPC group throughout the 14-day observation period (Fig. [125]2d). Furthermore, the cylinder test demonstrated a significant decrease in paw-dragging behavior on the ipsilateral forepaw of mice treated with iNPCs at both 1- and 3-day post-treatment (Fig. [126]2e), while no significant differences were observed on the contralateral side (Fig. [127]S6c). These results indicated that the transplantation of iNPCs effectively alleviated motor dysfunction and brain atrophy in tMCAO mice. Transplanted HiNPCs migrated into ischemic site and committed differentiation in tMCAO mice To evaluate the survival and fate of transplanted hiNPCs four weeks post-treatment, cerebral sections were immunolabeled by a human-specific Nestin (hNestin) antibody alongside the proliferation marker Ki67. The locations of the injection sites and the ischemic lesion area are illustrated in Fig. [128]3a. Notably, hNestin signals were observed at the periphery of the ischemic core, situated at a distance from the injection site (Fig. [129]3a, b), which suggests the migration of transplanted iNPCs towards the lesion. The colocalization of hNestin and Ki67 further substantiates that the transplanted iNPCs underwent proliferation within the ischemic site by week 4 (Fig. [130]3c). Fig. 3. [131]Fig. 3 [132]Open in a new tab Transplanted hiNPCs migrated and differentiated in the ischemic injured site. (a) Horizontal and coronal brain diagrams. The injection site was indicated by the red arrows and green dots in M1 cortex and striatum, green color indicated the transplanted iNPC disseminated areas, the grey area indicated the ischemic lesion site. (b) hiNPCs identified with hNestin in a coronal brain section. (c) Proliferating hiNPCs identified with hNestin and Ki67. (d-f) Differentiated hiNPCs identified with human cytoplasmic protein specific antibody (STEM121) and the glutamatergic marker (d) vGlut1, dopaminergic marker (e) DARPP32 or GABAergic marker (f) GAD65&67. The white arrowheads in (c-f) indicated signal colocalization. The small figures on the bottom left corner in (c-f) were the magnification of the dashed-line squares. CPu: caudate putamen. Scale bar in (b): 1 mm; scale bar in (c-f): 100 μm; scale bar in the enlarged figures of (c-f): 10 μm Furthermore, cerebral sections were stained with specific antibodies targeting human cytoplasmic protein (STEM121) and neuronal markers, including vGlut1 (glutamatergic), GAD65/67(GABAergic), and Darpp32 (dopaminergic). The signals from STEM121 indicated that the transplanted iNPCs extended extensive processes into regions exhibiting expression of GAD65/67, Darpp32, or vGlut1 (Fig. [133]3d-f), thereby suggesting their differentiation into various neuronal types. Moreover, no evidence of tumorigenesis was observed throughout the entire experimental period. HiNPCs transplantation alleviated neuroinflammation and promoted M2 polarization of microglia in the ipsilateral hemisphere Activated microglia rapidly engages in the immune cascade following an acute ischemic stroke, undergoing proliferation and differentiation into either the classic proinflammatory M1 subtype or the alternative M2 subtype [[134]31]. To determine whether transplanted iNPCs inhibit microgliosis, we assessed Iba1 expression in the ipsilateral hemisphere. As anticipated, Iba1 expression was significantly diminished in the iNPC-treated group on both day 14 and day 30 post-tMCAO when compared to the control group (p < 0.05; Fig. [135]S7a, 4a). Furthermore, immunostaining demonstrated significantly elevated Iba1 signals in the ipsilateral striatum, located just beneath the damaged cortex (ischemic penumbra) of the control group in contrast to the iNPC-treated group (Fig. [136]4c-d). Fig. 4. [137]Fig. 4 [138]Open in a new tab Transplanted hiNPCs alleviated microgliosis and promoted M2 polarization on day 30 after cerebral ischemia. (a, e-f) Quantitative RT-PCR analysis of (a) microglia marker (Iba1), (d) M1 microglia markers (Il1b, Cd86), and (e) M2 microglia markers (Arg1, Cd163) mRNA expression level relative to Gapdh in the ipsilateral or contralateral hemispheres from control and iNPC-treated mice. (b) Coronal brain diagrams. Red boxes indicated the sites of immunostaining figures taken in (c) and for statistical analysis, the grey area indicated the ischemic lesion site. (c) Representative figures of Iba1 (Green) immunostaining in the ipsilateral and contralateral side of control and iNPC treated group. White dotted lines marked the border of the ventricle. (d) Statistical analysis on normalized integrated density of Iba1 signals surrounding the ipsilateral or contralateral ventricles of control or iNPC-treated mice in (c). (g) Representative figures of immunostaining with antibodies against iNOS (green) and CD206 (red) in the penumbra region, nuclei were counterstained with Hoechst (blue). (h) Statistical analysis of CD206+/iNOS + ratio in (g). Scale bar in (c): 100 μm; scale bar in (g): 50 μm. n = 3 for (a, e-f), n = 12 for (d), and n = 15 from 3 mice for (h). * represented p < 0.05, ** represented p < 0.01, and n.s. represented no significance by one-way ANOVA followed by Tukey’s post hoc test in (e-f). * represented p < 0.05 and *** represented p < 0.001 by Student’s t test in (a, h). Con: contralateral, Ipsi: ipsilateral To investigate the impact of transplanted iNPCs on microglia phenotypes, we evaluated the expression of M1 and M2 subtype markers for 30 days post-stroke. The control group exhibited a significant upregulation of the M1 marker Cd86 (p < 0.01) and the proinflammatory cytokine Il1b (p < 0.01) in the ipsilateral hemisphere when compared to the contralateral side (Fig. [139]4e). In contrast, the iNPC-treated group showed a significant decrease in the expression of Cd86 (p < 0.01) and Il1b (p < 0.05) in the ipsilateral hemisphere compared to the controls (Fig. [140]4e). Furthermore, we analyzed the expression of M2 markers, specifically Cd163 and Arg1. The iNPC-treated group exhibited significantly elevated expressions of Cd163 (p < 0.05) and Arg1 (p < 0.01) in the ipsilateral hemisphere compared to both the controls and the contralateral side (p < 0.01 for both markers) (Fig. [141]4f). To further substantiate that transplanted iNPCs facilitated M2 polarization of microglia, immunolabeling for the M1 subtype marker iNOS and the M2 marker CD206 was performed in the penumbra regions 30 days post-stroke (Fig. [142]4g). Statistical analysis indicated a significantly elevated M2/M1 ratio (CD206/iNOS) in the iNPC-treated group (7.36 ± 1.09) in comparison to the controls (0.35 ± 0.05, p < 0.001) (Fig. [143]4h). Given that microglia are activated shortly after a stroke, with expression levels peaking on day 3 following MCAO and subsequently declining [[144]32], we investigated the effects of transplanted iNPCs at an earlier time point on day 14. The results indicated a significant reduction in proinflammatory cytokines within the iNPC-treated group, evidenced by lower expression of Il1b (p < 0.01) and Tnf (p < 0.01) in the ipsilateral hemisphere when compared to the controls (Fig. [145]S7b). Additionally, the iNPCs-treated group exhibited a decrease in markers associated with M1 microglia (p < 0.05 in Cd86, Nos2, and Fcgr3) (Fig. [146]S7c), alongside an increase in markers indicative of M2 subtype (p < 0.01 in Mrc; p < 0.05 in Cd163) (Fig. [147]S7d). A transwell coculture experiment was conducted to further evaluate the immunomodulatory effects of iNPCs on microglia. The RT-PCR results indicated that iNPC treatment significantly reduced the expression of M1 markers (IL1B, TNF) while exhibiting a modest increase in the expression of M2 markers (ARG1, TGFB1), in HMC3 cells subjected to M1 stimulation (Fig. [148]S8a-b). In summary, the transplanted iNPCs effectively alleviated neuroinflammation and facilitated the polarization of microglia towards M2 subtype in the ipsilateral hemisphere. HiNPCs transplantation inhibited chronic reactive astrogliosis and preserved astrocytic morphology in the ischemic penumbra region Inflammatory cytokines are known to activate astrocytes following cerebral ischemia and reperfusion, resulting in severe reactive astrogliosis. This condition disrupts astrocytic processes and leads to the formation of glial scars, which impede axonal regeneration during the chronic phase [[149]33]. To assess the potential modulatory effects of transplanted iNPCs on astrocytes in the chronic phase, we analyzed the expression of the reactive astrocyte marker Gfap. On both day 14 and day 30 post-ischemia, iNPC-treated mice exhibited a reduction in Gfap expression within the ipsilateral hemisphere (Fig. [150]5a). On day 30, Gfap immunostaining revealed no significant glial scars surrounding the lesioned area; however, it demonstrated chronic astrogliosis localized around the ipsilateral ventricle in controls, suggesting the persistence of chronic astrogliosis in the subventricular zone (SVZ) (Fig. [151]5b, c). This chronic astrogliosis in the ipsilateral SVZ was significantly ameliorated by iNPC treatment (Fig. [152]5d). Fig. 5. [153]Fig. 5 [154]Open in a new tab Transplanted iNPCs inhibited reactive astrogliosis but preserved astrocytic processes in the ischemic penumbra zone. (a) Quantitative RT-PCR analysis of reactive astrocytic marker Gfap in control and iNPC-treated mice. (b) Coronal brain diagrams. Red boxes indicated the sites of immunostaining figures taken in (c) and for statistical analysis, the grey area indicated the ischemic lesion site. (c) Representative figures of Gfap (Green) immunostaining in the ipsilateral and contralateral side of control and iNPC-treated group. White dotted lines marked the border of the ventricle. (d) Statistical analysis on normalized integrated density of Gfap signals surrounding the ipsilateral or contralateral ventricles of control or iNPC-treated mice in (c). (e) Representative immunostaining of Gfap in the penumbra zone of ipsilateral hemisphere. (f) Sholl analysis for the measurement of astrocytic branching. Represented astrocyte morphologies from control and iNPC-treated group were outlined from Gfap staining. Concentric rings were placed 5 μm apart around the soma. (g) Astrocytic branches shown by intersection numbers per radius against radial distance from soma. (h-i) Summary of (h) total intersection number and (i) primary intersection number in the reactive astrocytes from control and iNPC-treated group. Scale bar in (c): 100 μm, scale bar in (e): 50 μm. n = 3 for (a), n = 11 for control and n = 15 for iNPC from 3 mice for (g-i). * represented p < 0.05, and **** represented p < 0.0001 by one-way ANOVA followed by Tukey’s post hoc test in (a, d). ** represented p < 0.01, and **** represented p < 0.0001 by two-way ANOVA followed by Tukey’s post hoc test in (g). * represented p < 0.05 by Student’s t test in (h-i) Astrocytes play multiple important functions in both healthy and injured brains, engaging with synapses and contributing to the integrity of the blood brain barrier through their extensive fine processes and endfeet [[155]34]. Therefore, we evaluated the morphology and process branching of reactive astrocytes in the chronic phase of ischemic penumbra using Sholl analysis (Fig. [156]5e-f) [[157]21]. Statistical analysis demonstrated that reactive astrocytes from the iNPC-treated group exhibited a significantly greater number of short processes (5–25 μm in length), total intersections and primary branches in comparison to the controls (Fig. [158]5g-i). This finding indicated that control astrocytes displayed a shrunk morphology, whereas iNPC treatment effectively rescued this shrinkage (Fig. [159]5f). Collectively, these results indicate that iNPC treatment plays a vital role in preserving the morphology of reactive astrocytes in the remote penumbra region. Analysis of differential changes and functional enrichment in HiNPCs To further understand the cellular subpopulations and transcriptomic profiles of iNPCs, we collected samples on day 0 (iPSCs) and day 12 (iNPCs) for scRNA-seq analysis. The dimensional reduction analysis identified four distinct subpopulations on day 12: neuroepithelia (87.26%), immature neurons (7.64%), glial progenitors (2.65%), and neural crest-like progenitors (2.45%) (Fig. [160]6a). Fig. 6. [161]Fig. 6 [162]Open in a new tab Differential changes and functional enrichment in hiNPCs. (a) t-SNE plots of the scRNA-seq data of iNPCs (D12) and sector graph illustrating the percentage of each cluster in iNPCs. (b) Heatmap of 10 DEGs in each cluster. (c) t-SNE plots showing the expression levels of marker genes in each cluster of iNPCs. (d) Differential expression gene analysis showing iPSCs and iNPCs. (e-f) GO and KEGG enrichment analyses of iNPCs Specifically, the neuroepithelia, characterized by HES5 expression, exhibited the expression of genes such as LRP2, CUX2, and PRTG, which are implicated in early neural development and the maintenance of multipotency [[163]35–[164]38]. Immature neurons, identified by NEUROD4 expression, demonstrated the expression of DCC and STMN2, which are recognized markers for synaptogenesis and axonal outgrowth [[165]39, [166]40]. Glial progenitors, marked by APOE expression, displayed the expression of TPM1 and NEAT1, which are characteristic of glial populations [[167]41, [168]42]. Lastly, neural crest-like progenitors, identified by CDH6 expression, exhibited upregulation of EDNRA, ERBB3, and NPR3, all of which are essential for neural crest development [[169]43–[170]45] (Fig. [171]6b, c). The analysis of differential gene expression between day 12 iNPCs and day 0 iPSCs demonstrated a downregulation of the pluripotency marker POU5F1, alongside an upregulation of the neural progenitor marker HES5, thereby confirming effective differentiation (Fig. [172]6d). Furthermore, pathway enrichment analysis using KEGG and GO indicated that day 12 iNPCs were associated with the activation of signaling pathways and biological functions related to the development and functionality of the neural system (Fig. [173]6e, f). Discussion In this study, we reported a 3D aggregate-based differentiation method aimed at generating iPSC-derived iNPCs for therapeutic applications. To improve clinical translatability, the protocol has been designed to ensure broad applicability, easy of manipulation, and scalability. Our differentiation protocol was distinguished by the presence of double-layered neural aggregates that spontaneously bud and detach from the original ones, thereby obviating the necessity for rosette selection or cell passaging. The proliferation and cell fate of NPCs are influenced by extracellular paracrine signaling and cell-cell interactions, with a low initial cell density skewing cell fate towards the glial lineage [[174]46]. Therefore, numerous existing NPC expansion protocols, including both adherent and neurosphere methodologies, require frequent cell dissociation and replating [[175]11, [176]47]. However, it has been reported that enzymatic dissociation can induce apoptosis in various cell types, including iPSCs and iNPCs [[177]48]. The absence of cell passaging in our method has significantly enhanced both cell viability and the overall stability of this protocol. Furthermore, recent studies highlighted the therapeutic potential of NPC-derived exosomes in the treatment of brain disorders, thereby amplifying the demand for large-scale and consistent production of iNPC [[178]49, [179]50]. The scaling up of adherent cultures necessitates an increase in surface area, which in turn results in the requirement for complex equipment and labor-intensive operations [[180]51]. In contrast, 3D suspension culture facilitates a more straightforward scaling process, as evidenced by enhanced expansion and maintenance of pluripotency of iPSCs in bioreactors [[181]52, [182]53]. Moreover, prolonged cultivation is not suitable for neural stem cell (NSC) production, as it has been observed to result in the loss of glial cell fate after 15 passages, along with the acquisition of chromosomal abnormalities that increase cell proliferation [[183]12, [184]13]. Our 12-day protocol for iNPC generation utilized a chemically defined medium and an orbital shaker without the need for cell passaging, thereby rendering it readily scalable for clinical-grade applications. Furthermore, the characteristic apicobasal polarized morphology of the neural aggregates exhibited similarities to embryonic neural development. During neural development, the apicobasal polarity pathway regulates neural tube closure by modulating apical junction compartments and altering neural plate shape [[185]54]. In 2D culture, neural rosette formation was initiated by cell polarity acquisition through the redistribution of apical markers such as N-cadherin and ZO-1 to the center of the rosettes [[186]55]. In contrast, an inverted apicobasal polarity was observed in hiPSC-derived brain organoids, where N-cadherin and atypical protein kinase C were densely expressed in the outer structure, with differentiated neural cells located internally [[187]56]. In our 3D iNPC aggregates, the expression of the apical tight junction protein ZO-1 was detected both on the aggregate surface and at the center of the internal neural rosettes. This unique double-layered apicobasal polarized structure facilitated structural changes and may explain the detachment of new aggregates. To elucidate the therapeutic potential of iNPCs, a tMCAO stroke model and in vitro co-culture experiments, were employed. Despite that other proinflammatory cytokines exhibited a decline during the chronic phase, the elevation of Il1β in the control group underscored its critical role within our tMCAO model. The upregulation of Il1b is a well-documented indicator of various brain insults, including ischemic stroke [[188]57]. Given that IL1RA is the antagonist for the proinflammatory effect of IL1 signaling, its beneficial effects in post-stroke recovery have been reported, particularly through the modulation of microglia activity [[189]58]. Besides, cell-mediated delivery of IL1RA has been shown to inhibit IL1-dependent inflammation in various mouse models [[190]59]. IL1RA has also been extensively investigated as a therapeutic agent for traumatic brain injury [[191]60]. The observed increase in IL1RA expression within the coculture system suggested that iNPC treatment may exert its effects through an IL1RA-mediated immunosuppressive pathway, thereby counteracting IL1b signaling. Furthermore, the protective role of M2-polarized microglia and the adverse effects of M1-polarized microglia during the acute phase of ischemia have been extensively documented. The depletion of iNOS-positive microglia early post-stroke has been shown to reduce inflammation and alleviated cerebral ischemic damage [[192]61]. Conversely, Arg1-positive microglia have been reported to facilitate ischemic recovery, but the prevalence of these cells diminishes 5 days post-ischemia [[193]62]. Our intervention was conducted on day 5, coinciding with the decline of stroke-induced protective M2-polarized microglia. The significant increase in Arg1 expression and the presence of Arg1-positive cells in the iNPC-treated group suggested that the iNPC treatment operates through an Arg1-related pathway. Damage-induced reactive astrocytes and microglia have been extensively investigated for their protective and detrimental effects after ischemic stroke [[194]63]. In our chronic phase stroke model, it was observed that activated astrocytes and microglia were predominantly localized in the ventricular region. The processes of post-stroke reactive astrogliosis and neurogenesis in the SVZ contributed to astrocytic scar formation and neuronal replacement at the lesioned site [[195]64]. These stroke-induced reactive astrocytes in the SVZ exhibited detrimental effects by extending tortuous processes, thereby disrupting the neuroblast migratory scaffold [[196]65]. Additionally, stroke-associated microglia in the SVZ phagocytosed dying neuroblasts and diminished the neurogenic response in the affected area [[197]66]. Considering the detrimental effects associated with chronic astrogliosis and microgliosis, the significant decrease in activated astrocytes and microglia in the vicinity of the ventricle and hippocampus during the chronic phase in iNPC-treated mice indicated the therapeutic potential of these transplanted iNPCs. Since the established correlation between astrocytic morphology and their functional efficacy, a detailed analysis of astrocytic processes was conducted. The shrunken morphology of astrocytes observed in the control group resembled alterations typically associated with aging brains or neurodegenerative disorders [[198]34]. Previous research has documented an increase in the number of primary processes and the density of astrocytes in the surrounding tissue of the damaged cortex and striatum of a severe permanent MCAO stroke model [[199]67]. Conversely, in our mild tMCAO model, we observed an opposing phenomenon, characterized by the absence of a severe glial scar, yet persistent chronic astrogliosis in SVZ. iNPC treatment effectively mitigated astrocytic atrophy surrounding the ipsilateral ventricle to preserve their normal function. The tMCAO modeled mice in this study demonstrated a pronounced self-recovery ability, evidenced by a reduction in the neuroinflammatory environment by day 14, with no further necrosis developed thereafter. In contrast, other investigations have documented significant behavioral deficits and persistent neuroinflammation lasting for at least 1 month in various stroke models [[200]68, [201]69]. This mild stroke model presented a limited therapeutic window and exhibited minimal behavioral variations, as cell transplantation was conducted on day 5, thereby precluding the establishment of additional treatment groups for comparing transplantation timepoints and cell dosages. Future research endeavors should focus on enhancing the quality of the stroke model to more accurately simulate the chronic phase of ischemic stroke. We further conducted scRNA-seq to better understand the biological characteristics of iNPCs. As the primary constituents of iNPCs, neuroepithelia and immature neurons possess the capacity to rapidly replenish neuronal loss and damage, stabilize the microenvironment, and facilitate the rapid recovery of neurological function [[202]70]. Although the proportions of other cell types (neural crest and glial progenitors) were lower than those of neural progenitors, their functional contributions should not be underestimated. The paracrine-regulatory neural crest cells may collaborate with neural progenitors to modulate the inflammatory balance, thereby providing the essential environmental conditions necessary for the optimal functioning of neural progenitors [[203]71]. Furthermore, glial progenitors may play a critical role in maintaining the microenvironment, by clearing excessive neurotransmitters and metabolites resulting from post-stroke neuronal excitation [[204]72]. Consequently, the heterogeneity observed within our iNPC population enhances their adaptability to the post-stroke microenvironment and may account for the improved therapeutic effects in the tMCAO model. Although iPSC-derived cellular products exhibit considerable therapeutic potential, significant challenges persist in their clinical translation, particularly concerning immune rejection, regulatory hurdles, and the scalability of expansion. Firstly, therapies using autologous cells are expected to obviate the necessity for immunosuppression. While one report indicated the survival of syngeneic iPSC-derived NPCs following spinal grafting without immunosuppression [[205]73], another study identified abnormal gene expression and residual iPSCs that elicited unexpected immune responses in syngeneic mice [[206]74]. Consequently, it is imperative that iPSC-derived cells undergo thorough evaluation for both tumorigenicity and immunogenicity prior to their clinical application. Secondly, due to the inherent complexity and associated risks of iPSC-derived products, regulatory agencies impose stringent quality control measures and mandate comprehensive long-term safety evaluations. Variability in cell fate decisions throughout the differentiation process can lead to batch-to-batch heterogeneity, while the absence of a consensus regarding the definitions of biological function and standards for efficacy evaluation further complicates the establishment of standardized production and quality control [[207]75]. Consequently, efforts are needed to implement measures for tracking the fate of transplanted cells and to establish reliable quality standards for cell products. Lastly, the large-scale production often results in functional loss, as evidenced by the diminished immunomodulatory capacity observed in expanded mesenchymal stem cells [[208]76]. Additionally, factors such as mechanical stress regulation and adequate oxygenation within 3D cultures in bioreactors significantly impact cultivation outcomes [[209]77, [210]78]. To achieve proper differentiation and characterization of iNPCs, it is essential that all culturing parameters within the bioreactor are carefully determined. Conclusions In summary, we have developed a scalable, stable, and efficient 3D aggregate-based method for the generation of iNPCs. These aggregate-derived iNPCs exhibited high viability, robust NPC marker expression, and diverse differentiation potential. Therapeutic assessments conducted in a tMCAO model indicated that the transplanted iNPCs exhibited a rapid respond to the ischemic site and effectively remodeled the microenvironment during the chronic phase. scRNA-seq analysis further categorized the iNPCs into four subpopulations, namely neuroepithelia, immature neurons, glial progenitors, and neural crest-like progenitors, which elucidated their pleiotropic therapeutic effects. Overall, our study successfully generated iNPCs with strong therapeutic potential using a 3D aggregate-based suspension system. Electronic supplementary material Below is the link to the electronic supplementary material. [211]Supplementary Material 1^ (4MB, docx) Acknowledgements