Abstract

Multiple sclerosis (MS) is an autoimmune inflammatory disease of the central nervous system. The cause of the disease is unknown but both genetic and environmental factors are strongly involved in its pathogenesis. We derived cerebral and spinal cord organoids from induced pluripotent stem cells (iPSC) from healthy controls as well as from primary progressive MS (PPMS), secondary progressive MS (SPMS) and relapsing–remitting MS (RRMS) patients to investigate and compare oligodendrocyte differentiation and myelination capacity. In MS organoids, particularly in PPMS, we observed a decrease in p21 expression associated with a dysregulation of PAK1 and E2F1 expression. In parallel, a decrease in oligodendrocyte maturation was detected in long-term cultured cerebral and spinal cord organoids, especially in PPMS, leading to a reduced myelination capacity. Disruption of astrocyte and neuronal populations was also observed. Our findings demonstrate that in MS, inherent deficits in the p21 pathway may alter glial and neuronal cell populations and may contribute to the disease pathogenesis by reducing the capacity for myelin repair.

Introduction

Multiple sclerosis (MS) is an auto-immune inflammatory disorder that may lead to irreversible neurological disability and cognitive decline (Filippi et al., 2018). It is characterized by widespread focal lesions of primary demyelination in the brain and spinal cord, with variable axonal, neuronal, and astroglial injury. The disease presents primarily as two clinical subtypes: relapsing–remitting MS (RRMS), which accounts for 85–90% of cases which may evolve to secondary progressive MS (SPMS), and primary progressive MS (PPMS) which affects about 10% of cases and is marked by a steady functional decline from disease onset (Lublin and Reingold, 1996; Andersson et al., 1999). The origin, evolution, and physiological basis of MS’s varied phenotypic expressions remain poorly understood in part due to the relative inaccessibility of human brain tissues and limitations of animal models (Ransohoff, 2012). The development of MS is influenced by genetic factors, with familial relatives of patients, especially first-degree relatives, being more susceptible to developing MS compared to the general population. There is also epidemiological evidence that environmental triggers possibly infections are important in disease causation (Gouider et al., 2024). This interplay of genetic susceptibility and environmental factors contributes to autoimmunity and the multifaceted nature of MS (Willer et al., 2003).

We previously described cerebral organoids (c-organoids) as an innovative model to study MS. Using c-organoids derived from induced pluripotent stem cells (iPSCs) of healthy control subjects, and PPMS, SPMS and RRMS patients, we showed a decrease of proliferative capacity, notably in progressive forms of MS. This was associated with a reduction of the progenitor pool and an increase of neurogenesis possibly due to a symmetric shift of the cell division mode. We linked these effects to a strong decrease of p21 expression in PPMS organoids, unrelated to the DNA damage and apoptosis pathway (Daviaud et al., 2023). Interestingly, it has been reported that p21 is required for the differentiation of oligodendrocytes and that animal knockdown for p21 exhibited hypomyelinated brains (Zezula et al., 2001).

Here we further investigated neural and glial cell differentiation and myelination capacity in MS using our cerebral and spinal cord organoid (SCO) models. We detected a significant decrease in oligodendrocyte maturation and myelination capacity in MS organoids, especially in PPMS. Additionally, a defect of astrocyte population and an imbalance of inhibitory/excitatory neurons were found, which are key factors in MS onset and possibly in causing cognitive impairment, physical disability, and fatigue in MS patients. We confirmed that the p21 pathway dysfunction was a critical abnormality in our organoid model of MS, while expression of other Cyclin Dependent Kinase inhibitors (CDKi), such as p16, p27 and p57, was unaltered. Furthermore, we focused on the p21 regulators E2F1 and PAK1. In the oligodendroglial lineage, the transcription factor E2F1 contributes to the transition of oligodendrocyte precursor cells (OPCs) from proliferation to differentiation (Magri et al., 2014), highlighting its role at the interface between cell cycle control and lineage progression. PAK1, a serine/threonine kinase, also participates in oligodendrocyte development. Its activity has been associated with oligodendrocyte differentiation and myelination (Brown et al., 2021), as well as with proper myelin sheath formation in the central nervous system (Thurnherr et al., 2006). Moreover, PAK1 can negatively regulate p21 expression (Wang et al., 2018), thereby promoting cell growth, while p21 itself is transcriptionally regulated by E2F1 (Hiyama et al., 1998). Together, these findings suggest a functional interplay between E2F1, PAK1, and p21 during oligodendrocyte lineage progression but also in regulation of the excitatory/inhibitory network.

In conclusion, this work demonstrates that cortical and SCO derived from patients with MS can be used as an innovative tool to better understand the genetic basis of phenotypic differences seen in MS. This work provides a better understanding of the genetic/epigenetic changes in patients with MS and sheds new light on the developmental aspect of MS, which is of particular interest for understanding the pathogenesis of MS.

Materials and methodsPatient selection

Peripheral blood mononuclear cell (PBMCs) samples were collected from healthy control subjects and clinically definite MS patients diagnosed according to the revised 2017 McDonald Criteria. All MS patients underwent neurological examination and MRI imaging and were classified as having RRMS, SPMS or PPMS by board-certified neurologists specializing in MS care. All protocols were IRB approved, and all donors provided written informed consent for participation. Human iPS cells were generated from PBMCs from donor blood samples. Other cell lines were obtained from the New York Stem Cell Foundation. Donor information is summarized in Tables 1, 2.

SexAgeDiagnosisSourceM28ControlNYSCFM23ControlTisch MSRCNYM36ControlTisch MSRCNYM67ControlTisch MSRCNYF52ControlTisch MSRCNYF40ControlTisch MSRCNYF62PPMSNYSCFM61PPMSNYSCFF58PPMSTisch MSRCNYM40PPMSTisch MSRCNYF57PPMSTisch MSRCNYM38RRMSTisch MSRCNYF67RRMSTisch MSRCNYM24RRMSNYSCFF50RRMSNYSCFM60SPMSTisch MSRCNYM48SPMSTisch MSRCNYF54SPMSTisch MSRCNYF34SPMSTisch MSRCNYF67SPMSTisch MSRCNY

Table summarizing patients’ characteristics such as sex, gender, age, diagnosis, and cell source.

Statistical significanceHealthy controlsPPMSRRMSSPMSNumber of patients6545Age at collection, mean (SD), years41 (16)55.6 (8.9)44.7 (18)53 (13)n.s.Gender distribution2 F; 4 M3 F; 2 M2 F; 2 M3 F; 2 Mn.s.

Table summarizing the age and gender distribution in our cohort. No significant difference was observed in age (One-way ANOVA, p = 0.3450) or gender distribution among the different conditions (Chi-square, p = 0.7851).

Reprogramming CD34+ progenitor cells

Human PBMCs were isolated according to the “STEMCELL Integrated Workflow for the Isolation, Expansion, and Reprogramming of CD34+ Progenitor Cells”. Briefly, blood samples were collected in heparin-vacutainer tubes from donors in amounts ranging from 8 to 20 mL. CD34+ hematopoietic stem and progenitor cells were isolated from peripheral blood using the EasySep RosetteSep Kit (STEMCELL) and expanded in vitro in CD34+ expansion media consisting of StemSpan SFEM II and CD34+ expansion supplements (STEMCELL). After 7–10 days of culture, 1×106 cells were collected for reprogramming by electroporation using the Epi5 Episomal iPSC Reprogramming Kit (ThermoFisher) and Human CD34+ Cell Nucleofector Kit (Lonza) using a Nucleofector 2b Device (Lonza). After electroporation cells were cultured on Cultrex UltiMatrix coated 6-well plates (100 μg/mL, R&D systems) in CD34+ expansion media. After 3 days, ReproTeSR (STEMCELL) was added to culture media for 2 more days. On day 7 cells were cultured in ReproTeSR media only. Media was changed daily. After 2–3 weeks, iPSC colonies were isolated manually and transferred to Cultrex coated 6-well plates containing mTeSR Plus media (STEMCELL). Three subclones were created for each iPS cell line.

Human induced pluripotent stem cells

Human iPSCs were cultured as previously described (Lancaster and Knoblich, 2014; Daviaud et al., 2023). iPSCs were plated in 6-well tissue culture plates coated with diluted Cultrex UltiMatrix (100 μg/mL, R&D Systems) and maintained in mTeSR Plus Culture Media (STEMCELL), supplemented with rock inhibitor Thiazovivin (2 μM, Millipore). Media was changed daily without rock inhibitor until ready to passage at approximately 70–80% confluence or harvested.

All human pluripotent stem cells were maintained below passage 30 and confirmed negative for mycoplasma using the MycoFluor Mycoplasma Detection Kit (ThermoFisher). iPSCs were regularly evaluated for pluripotency using OCT4, NANOG and SOX2 markers (Figure 1) and were confirmed to be karyotypically normal by G-band testing by a qualified service provider (Cell Line Genetics, Madison, WI, USA) (Figure 1).

Patient cell lines pluripotency quality control. The pluripotent cell lines established for this work, derived from healthy controls and patients with MS, were tested for pluripotency and underwent quality control analysis. Additional cell lines were previously tested and characterized (Daviaud et al., 2023). Brightfield microscopy was used to assess iPSCs morphology and general health. Oct4 and Nanog immunofluorescence counterstained with DAPI was performed to verify iPSCs pluripotency. Karyotype analysis was performed by G-band testing to confirm genetic stability after reprogramming.

Generation of neural precursor cells

Neural precursor cells (NPCs) were produced following a protocol previously described (Gunhanlar et al., 2018) with minor modification. Briefly, iPSCs were dissociated with EDTA (0.5 mM, Millipore) for 5–6 min at 37 °C. Embryonic bodies (EBs) were generated by transferring 4,000 cells per well of an ultra-low attachment 96 well plates in mTeSR Plus supplemented with Thiazovivin (2 μM, Millipore). After 2 days, the medium was changed to neural induction media (StemDiff, STEMCELL) and EBs were cultured for another 4 days. On day 7, EBs were slightly dissociated by mechanical trituration and cultured on Cultrex UltiMatrix (100 μg/mL, R&D systems) coated plates in neural induction medium (StemDiff, STEMCELL) for 7 days. At d15 the medium was switched to NPC medium consisting of DMEM/F12, 1% N2 supplement (ThermoFisher), 2% B27 supplement without RA (ThermoFisher), 20 ng/mL epithelial growth factor (Peprotech), 10 ng/mL basic fibroblast growth factor (Peprotech) and 1% penicillin/streptomycin (ThermoFisher). At d15, cells were considered pre-NPCs and could be passaged and cryopreserved at confluence. From passage 3, cells were considered NPCs and were used for histologic analysis and for neural differentiation.

Neural differentiation was achieved by mitogen withdrawal. NPCs were cultured for 10 days on Cultrex UltiMatrix coated plates with differentiation media consisting of DMEM/F12, 1% N2 supplement and 2% B27 with RA supplement (ThermoFisher) and were then fixed and analyzed by immunostaining.

Generation of human cerebral organoids

C-organoids were generated from human iPSCs and processed for analysis as described (Lancaster and Knoblich, 2014; Daviaud et al., 2023) with minor modifications. iPSCs were washed with Dulbecco’s phosphate-buffered saline (DPBS, ThermoFisher) and dissociated with EDTA 0.5 mM (Millipore). A total of 8 × 103 cells were seeded into each well of an ultra-low attachment 96-well plate (Corning) to form embryoid bodies (EBs) in mTeSR Plus medium supplemented with 4 μM of Thiazovivin (Millipore) for the first 2 days. The medium was changed every other day to the same medium without Thiazovivin for another 2–3 days. After 4–5 days of culture or when EBs reached ~500–600 μm in diameter and the surface tissue began to brighten, EBs were cultured in neural induction medium (StemDiff, STEMCELL). After neuroepithelium emergence (typically at ~ day 9–10), embryoid bodies were embedded in 15 μL Cultrex UltiMatrix droplets and cultured in 6 well plates containing c-organoid differentiation medium consisting of 1:1 DMEM-F12 and Neurobasal medium (Gibco), with addition of 0.5% N2 supplement (Life Technologies), 0.5% ml MEM-NEAA (Gibco), 1% Glutamax (Gibco), 1% B27 supplement without vitamin A (Life Technologies), 0.1 μM of 2-mercaptoethanol (Millipore), 2.6 μg/mL insulin (Sigma Aldrich) in static culture for 4 days. Organoids were cultured in c-organoid differentiation medium supplemented with vitamin A on an orbital shaker (CO2 Resistant Shakers, ThermoFisher) at 80 rpm. Organoids were cultured up to 150 days. Analyses were performed at D42 and D120. D42 corresponds to an early stage when cortical structures, neural stem cells, NPCs, and neuroblasts are well-formed, but mature astrocytes and myelinating oligodendrocytes are not yet present. D120 represents a later stage when mature neurons and glial cells, including myelinating oligodendrocytes, are detectable.

Generation of human spinal cord organoids

Spinal cord organoids were generated from human iPSCs and processed for analysis as described elsewhere (Lee et al., 2022; Xue et al., 2023) with minor modifications. iPSCs were washed with DPBS (ThermoFisher) and dissociated with EDTA 0.5 mM (Millipore). A total of 8 × 103 cells were seeded into each well of an ultra-low attachment 96-well plate (Corning) to form embryoid bodies (EBs) in mTeSR Plus medium supplemented with 4 μM of Thiazovivin (Millipore) for the first 2 days. On day 3, the totality of the culture media was removed and replaced with a differentiation media consisting of 1:1 DMEM-F12 and Neurobasal medium (Gibco), with addition of 0.5% N2 supplement (Life Technologies), 0.5% ml MEM-NEAA (Gibco), 1% Glutamax (Gibco), 1% B27 supplement without vitamin A (Life Technologies), 0.1 μM of 2-mercaptoethanol (Millipore), 2.6 μg/mL insulin (Sigma Aldrich), 10 μM of SB431542 (STEMCELL), 2 μM of CHIR99021 (STEMCELL) and 0.5 μM of LDN-193189 (STEMCELL). Organoids were kept in culture for 8 days with half media change every other day. On day 11, organoids were embedded in 15 μL Cultrex UltiMatrix droplets and cultured in 6 well plates containing differentiation medium supplemented with vitamin A and 15 ng/mL of BMP4 (PEPROTECH) on an orbital shaker (CO2 Resistant Shakers, ThermoFisher) at 80 rpm. On day 16, media was changed to a differentiation media supplemented with 10 ng/mL BDNF and 10 ng/mL GDNF (PEPROTECH). Organoids were cultured for 64 days. At this stage, neurons, astrocytes, and motor neurons are clearly detectable, consistent with established spinal cord organoid differentiation protocols.

Histological analysis

At D42 and D120-150 days of culture, c-organoids were washed in DPBS and fixed in 4% PFA (ThermoFisher) for 20 min at 4 °C. After three washes with DPBS, organoids were cryoprotected in 30% sucrose overnight at 4 °C followed by snap freezing in OCT compound (ThermoFisher) and stored at −20 °C. Cryosections of organoids were cut at 15 μm thickness using a cryostat (Leica CM 1950) and mounted on microscope slides (Histobond+, VWR).

For immunofluorescence, slides were thawed to room temperature before being outlined with a PAP pen (Millipore) to create a hydrophobic barrier. Slides were washed and permeabilized with DPBS supplemented with 0.1% Triton X-100 (Millipore Sigma). Non-specific binding sites were blocked with DPBS supplemented with 0.1% Tween 20, 4% Bovine Serum Albumin (ThermoFisher) and 10% Normal Goat Serum (ThermoFisher) for 1 h at RT. Slides were then incubated overnight at 4 °C with the following primary antibodies diluted in blocking solution: mouse anti-APC (1:100, Calbiochem), mouse anti-ChAT (1:200, ThermoFisher), rat anti-CTIP2 (1:500, Abcam), guinea pig anti-DCX (1:500, Millipore), Rabbit anti-E2F1 (1:400, ThermoFisher), mouse anti-EOMES (1:100, ThermoFisher), mouse anti-GAD67 (1:200, Abcam), mouse anti-GFAP (1:500, Novus Biologicals), mouse anti-Ki67 (1:400, Millipore), chicken anti-MBP (1:500, ThermoFisher), mouse anti-Nanog (1:400, Abcam), mouse anti-O4 (1:200, R&D systems), rabbit anti-Oct4 (1:400, Abcam), rabbit anti-Olig2 (1:200, Abcam), rabbit anti-p16 (1:200, ThermoFisher), rabbit anti-p21 (1:200, ThermoFisher), rabbit anti-p27 (1:400, ThermoFisher), rabbit anti-p57 (1,400, ThermoFisher), mouse anti-Pax6 (1,100, Abcam), rabbit anti-PAK1 (1,100, ThermoFisher), mouse anti-SOX2 (1:400, Abcam), rabbit anti-TBR1 (1,400, Abcam), rabbit anti-vGluT1 (1,200, Abcam). After washing, slides were incubated with appropriate Alexa-coupled secondary antibodies (ThermoFisher) diluted in blocking solution for 1 h at RT and counterstained with DAPI, before mounting with Fluoromount Aqueous Mounting Medium (Millipore).

Immunofluorescence images were collected using a fluorescence microscope (Zeiss Imager M2) or a confocal fluorescence microscope (Zeiss LSM 510) and processed using Zen software and Fiji software (Schindelin et al., 2012).

Experimental design and statistical analysis

Organoids were derived from 20 patient iPSC lines, with a minimum of three lines per MS subtype. Three independent subclones were generated for each line. For each clone, 2–4 independent differentiation experiments (batches) were performed on separate days using independently prepared media. Each differentiation experiment generated 1–3 organoids. Multiple images (2–4) were acquired per organoid and quantified using Fiji software (Schindelin et al., 2012) and averaged.

For quantification, regions of interest (ROIs) were defined as radial columns spanning all cortical layers near the organoid surface. ROIs were positioned according to consistent anatomical criteria across all samples. For cell number analyses, marker-positive cells within each ROI were manually counted. Only cells showing clear co-localization of marker signal with a DAPI-positive nucleus were considered positive. For percentage analyses, the number of marker-positive cells was normalized to the total number of DAPI-positive nuclei within the same ROI and expressed as a percentage. For fluorescence intensity analyses, absolute fluorescence intensity (arbitrary units, A.U.) was measured within the same ROIs using identical acquisition and analysis parameters across all samples. Fluorescence intensity values were normalized to the total DAPI fluorescence intensity within each ROI to account for differences in cell density. No background subtraction or post hoc exclusion of artefactual staining was performed.

Statistical analyses were performed using GraphPad Prism 11 (GraphPad Software). Data with a hierarchical structure (technical replicates nested within patient-derived iPSC lines and grouped by clinical condition) were analyzed using a nested one-way ANOVA, with condition treated as a fixed effect and patient as a random effect nested within condition. When a significant overall effect was detected, Tukey’s multiple comparison test was applied. For comparisons involving single measurements per subject, differences between groups were analyzed using one-way ANOVA. Residual normality was assessed using the Shapiro–Wilk test. A chi-squared test was used to analyze frequency distributions where appropriate. Data are presented as mean ± SEM unless otherwise specified. Results were considered statistically significant when p < 0.05.

ResultsCerebral organoid derived from patients with MS develop and mature over time

MS is a heterogeneous disease whose course and severity can vary greatly from patient to patient. To best capture the diversity/variability of the disease, we decided to generate iPS cell lines derived from 10 male and 10 female subjects (Table 1), including 6 healthy controls and 14 MS patients. The age of the patients ranged from 23 to 67 years old with no significant differences between healthy controls and the different MS groups (One-Way ANOVA, p = 0.3450) (Table 2).

Each patient iPS cell line exhibited a normal phenotype in culture. Expression of pluripotency markers, such as Oct4 and Nanog were confirmed by immunofluorescence, and each cell line had a normal Karyotype (Figure 1).

C-organoids were generated using a previously described protocol (Lancaster and Knoblich, 2014; Daviaud et al., 2023) (Figure 2A). After 40–50 days in vitro, c-organoids exhibited immature cortical structures consisting of the ventricle aligned with proliferating cells, surrounded by the ventricular zone (VZ) containing the SOX2+ stem cell pool, the subventricular zone (SVZ) containing TBR2+ intermediate progenitors and DCX+ neuroblasts, and the cortical plate (CP) containing mature neurons (Figures 2B,D). After 120–150 days in vitro, c-organoids continued to mature, and exhibited more developed neurons, such as GABAergic and Glutamatergic neurons, but also myelinating oligodendrocytes and astrocytes (Figures 2C,E).

Cerebral organoids derived from patients with MS develop and mature over time. (A) Pictures of IPS cells derivation into cerebral organoid using brightfield microscopy. (B) Immunofluorescence of human cerebral organoids at D40-50 for the major cell types detected: Ki67+ proliferating cells, SOX2+ neural precursors, TBR2+ intermediate progenitors, DCX+ neuroblasts, TBR1+ and TBR2+ mature cortical neurons. V: Ventricles. Insets show higher magnification views of representative ZOIs (200 × 300 μm). (C) Immunofluorescence of human cerebral organoids at D120-150 for SOX2+ neural precursors, GFAP+ astrocytes, Olig2+ oligodendrocytic cells, DCX+ neuroblasts, GAD67+ and vGlut1+ GABAergic and glutamatergic neurons respectively and Olig2+O4+MBP+ myelinating oligodendrocytes. Insets show higher magnification views of representative ZOIs (200 × 300 μm). (D) Schematic representation of the cortical structures found in cerebral organoids at D40-50 including the ventricular zone containing the stem cell pool and most proliferating cells, the subventricular zone (SVZ) containing mostly iPCs, the outer SVZ (oSVZ) and the cortical plate (CP) containing neuroblasts and mature neurons. (E) Schematic representation of the structures found in cerebral organoids at D120-150. At this point there is no clear distinction between cortical layers. The rim of the organoids contains stem cells but mostly mature neurons, astrocytes, and myelinating oligodendrocytes.

CDKi involved in MS organoids appears to be restricted to p21

CDKi are proteins that bind to and inhibit the activity of CDKs, thereby controlling the cell cycle. Two major classes of CDKi have been identified. The INKs family includes p15, p16, p18 and p19 which bind to and inhibit the activities of CDK4 and CDK6. The CIP/Kips family includes p21, p27, p28 and p57 which can bind to and inhibit the activities of a wide range of CDK-cyclin complexes (Besson et al., 2008) (Figure 3A).

CDKi involved in MS organoids appears to be restricted to p21. (A) Schematic representation of the cell cycle and the involvement of the various CDK/cyclin complexes and the different CDK inhibitors. (B) Representative images of an immunofluorescence against the CDKi p16 in c-organoids at D42 in the different conditions. No ectopic location or important change in p16 expression were observed. (C) Representative images of an immunofluorescence against the CDKi p57 in c-organoids at D42 in the different conditions. p57 was evenly distributed among the layers of c-organoids. No ectopic location or important change in p57 expression were observed. (D) Representative images of an immunofluorescence against the CDKi p21 in c-organoids at D42 in the different conditions. p21 was mostly expressed in the lower layers (SVZ and VZ) of c-organoids. No ectopic location was detected, however, an important decrease of p21 expression was observed, especially in PPMS. (E) Representative images of an immunofluorescence against the CDKi p27 in c-organoids at D42 in the different conditions. p27 was mostly expressed in the outer layers of c-organoids (CP). No ectopic location or important change in p27 expression were observed. (F) Quantification of p16, p27, p57, and p21 expression. Statistical analysis was performed using a nested one-way ANOVA (p16: p = 0.7402; p27: p = 0.6003; p57: p = 0.1491; p21: p = 0.0188). Cortical regions were cropped from the original images to include all cortical layers (VZ, SVZ, and CP). Data are presented as superplots: small dots represent individual technical replicates, while large dots represent the mean value for each patient. Horizontal lines indicate the grand mean of the conditions.

We previously highlighted the involvement of p21 in a c-organoid model of MS (Daviaud et al., 2023). However, we did not evaluate other CDKi expressions in MS organoids compared to controls. Interestingly, p16 expression is increased in PPMS neural stem cells leading to senescence which could contribute to limited remyelination (Nicaise et al., 2019). p27 ensures cell cycle arrest and subsequent differentiation of oligodendrocyte precursor cells (OPCs) in mature oligodendrocytes (Casaccia-Bonnefil et al., 1997) and has also been described as a positive regulator of Schwann cell differentiation in vitro (Li et al., 2011). p57 regulates the number of divisions in OPCs before the onset of differentiation and can inhibit oligodendrocyte and Schwann cell differentiation (Nicaise et al., 2019). Thus, abnormalities in any of these CDKi may impact myelin function.

We performed an immunofluorescence at D42 for p16, p21, p27 and p57 in organoids from patients with MS and healthy controls (Figures 3B–E). No ectopic location or significant expression difference was observed for p16, p27 and p57, while a decreased expression of p21 was found in MS organoids, especially in PPMS (Figures 3B–E). Quantification revealed no significant difference for p16 (Nested one-Way ANOVA, p = 0.7402), p27 (Nested one-Way ANOVA, p = 0.6003), p57 (Nested one-Way ANOVA, p = 0.1491). In contrast, a significant overall group effect was observed for p21 expression (Nested one-way ANOVA, p = 0.0188) with reduced p21 expression in PPMS (p = 0.0197) and RRMS (p = 0.0486) compared with controls (Figure 3F).

The CDKi involved in MS NPC in vitro appears to be limited to the cell cycle inhibitor p16

To confirm the results observed in c-organoids in a simpler model, we decided to perform immunofluorescence for the CDKi p16, p21, p27 and p57 on iPS-derived NPCs. We performed these analyses on 2 conditions: NPC in expansion and NPC after 10 days of neural differentiation (Figure 4). In proliferating PPMS NPCs, p21 and p27 expression were detected in both PPMS and control conditions with no apparent differences based on qualitative assessment. In contrast, there appeared to be a visual difference in p16 expression and associated cellular morphology in PPMS NPCs compared to controls, consistent with previous findings (Nicaise et al., 2019) (Figure 4A). A potential translocation of p57 from the nucleus to the cytoplasm was noted in PPMS NPCs compared to controls under both proliferative and differentiation conditions (Figure 4A). Similarly in differentiation conditions, p27 expression was detected in PPMS and control cells with no apparent change (Figure 4B). Quantitative analyses would be required to formally assess differences in expression levels or subcellular localization.

iPSC derived NPC and neural cell from PPMS exhibit p16 induced senescence. (A) Representative images of an immunofluorescence performed on iPS-derived NPC in expansion for p57, p21, p27, and p16, in control and PPMS samples. Only p16 expression was different between the two conditions. (B) Representative images of an immunofluorescence performed on iPS-derived NPC, 8 days after mitogens withdrawal for p57 and p27 in control and PPMS samples. Insets show higher magnification views of representative ZOIs (150 × 150 μm).

P21 regulators E2F1 and PAK1 are involved in MS pathogenesis

We found that CDKi p21 is significantly decreased in MS organoids, especially in PPMS. As p21 is necessary for oligodendrocyte differentiation, we analyzed PAK1 and E2F1 expressions. PAK1 is likewise involved at multiple stages: it promotes OPC differentiation into pre-oligodendrocytes and supports maturation into myelinating oligodendrocytes (Brown et al., 2021), while also regulating cortical development by enhancing neural progenitor proliferation and neuronal migration (Pan et al., 2015). Notably, PAK1 silencing has been shown to increase p21 expression, suggesting a functional interaction between these pathways (Wang et al., 2018). In parallel, E2F1 negatively regulates oligodendrocyte maturation, and its transcript levels decrease as OPCs differentiate into oligodendrocytes (Magri et al., 2014). E2F1 is also required for Ras-mediated activation of the p21 promoter (Gartel et al., 2000), highlighting the complex interplay between cell cycle regulators and oligodendrocyte differentiation programs (Figure 5A).

P21 regulators E2F1 and PAK1 are involved in MS pathogenesis. (A) Schematic representation of the involvement of PAK1 and E2F1 in p21 regulation, but also in oligodendrocyte differentiation, maturation, and myelination capacity. (B) Immunofluorescence for PAK1 co-stained with neuroblast marker DCX and E2F1 co-stained with stem cell marker SOX2. (C) Representative images of an immunofluorescence for PAK1 and E2F1 in organoids at D42 and E2F1 at D120 in control, PPMS, RRMS, and SPMS samples. (D) Quantifications for PAK1 at D42 (Nested one-way ANOVA, p = 0.0004), E2F1 at D42 (Nested one-way ANOVA, p = 0.5999), and E2F1 in c-organoids at D120 (Nested one-way ANOVA, p < 0.0001). Insets show higher magnification views of representative ZOIs (200 × 300 μm). Data are presented as Superplots: small dots represent individual technical replicates, while large dots represent the mean value for each patient. Horizontal lines indicate the grand mean of the conditions.

We first determined the localization of E2F1 and PAK1 expression in c-organoids at D42 by immunofluorescence. PAK1 was mainly expressed in the outer layers of the organoid cortical structure and the cortical plate, suggesting that PAK1 was mainly expressed in mature cells, such as neuroblasts, astrocytes or oligodendrocytes (Figure 5B). E2F1 was mainly expressed in the inner layer of the cortical structure, especially in the ventricular zone, which indicates that E2F1 was mainly expressed in the progenitor cells, such as NPCs (Figure 5B).

We then performed an immunofluorescence for PAK1 and E2F1 in c-organoids from patients with MS compared to organoids from healthy controls (Figure 5C). At D42, very low to no E2F1 expression was detected, and quantification revealed no significant differences between groups (Nested one-way ANOVA, p = 0.5999). In contrast, at D120, strong E2F1 staining was observed, consistent with its reported role in more mature neural contexts, including neuronal and glial differentiation. A significant overall group effect was observed for E2F1 expression (Nested one-way ANOVA, p < 0.0001) with reduced E2F1 expression in PPMS organoids compared with controls (p < 0.0001) and RRMS (p = 0.0091), while no significant difference was observed compared with SPMS (p = 0.3450).

At D42, a significant overall difference in PAK1 expression was also detected (Nested one-way ANOVA, p = 0.0004) with increased PAK1 expression in PPMS (p = 0.0250) and RRMS (p = 0.0011) compared with controls (Figure 5D).

These results suggest that the dysregulation of p21 may be induced or associated with changes in E2F1 and PAK1 expression in the c-organoid model of MS.

Oligodendrocyte maturation and myelination capacity is reduced in MS organoids

The differentiation of NPCs into mature oligodendrocytes is a multi-step process. NPCs first differentiate into NG2+ OPCs, then into O4+ pre-oligodendrocytes which mature into GALC+/MBP+ myelinating oligodendrocytes. During each of these steps, oligodendroglial cells express Olig2 (Figure 6A).